This Item

Transcription

This Item
Sept/Oct 2009
www.slm-neurology.com
The European
Neurological Journal (enj)
Print ISSN 2041-8000
Vol 1, No 1
EPDA
tel / fax: +44 (0) 1732 457 683
email: info@epda.eu.com
EUROPEAN PARKINSON’S DISEASE ASSOCIATION
Spreading the word
www.epda.eu.com
epda charter
global declaration on parkinson’s disease
the red tulip
parkinson’s awareness campaign
epda plus
multidisciplinary conferences
world pd day
promoting international understanding and awareness
quality of life issues promoting informed choices supporting national parkinson’s organisations pddoc
patient information
promoting a constructive dialogue between science and society
european participation in life survey
supporting research initiatives
parkinson’s decision aid
news and updates
professional information
european surveys pdns core competencies real life real pd ot survey
coping strategies
parkinson’s passport providing access to best practices
encouraging interaction between scientific and patient communities
medical and surgical information
www.parkinsonsAwareness.eu.com
guide to living with parkinson’s
www.parkinsonsDecisionAid.eu.com
epnn journal
rewrite tomorrow
www.rewriteTomorrow.eu.com
The European
Neurological Journal (enj)
September 2009
The Editors and Publishers of
the European Neurological Journal (enj)
Owned and Published by San Lucas Medical Limited
John Gault, MD, Editor-in-Chief
Vol 1, No 1
Media Partners
The European Neurological Journal wishes to thank
our media partners for their assistance. This does not
constitute any official endorsement.
Editorial Board
Alan B. Ettinger, USA
Alberto J. Espay, USA
Alessandro Di Rocco, USA
Alexander Storch, Germany
Alfons Schnitzler, Germany
Alireza Minagar, USA
Andrew S.C. Rice, UK
Andy Dowson, UK
Antonie Beiske, Norway
Ashwin A Patkar, USA
Athina Vadalouca, Greece
Birgit Högl, Austria
Bruno Vellas, France
Carlo Pozzilli, Italy
Christine Van Broeckhoven,
Belgium
Clete A. Kushida, USA
Cris S Constantinescu, UK
Dag Nyholm, Sweden
Dawn Marcus, USA
Denis Vivien, France
Dennis Bourdette, USA
Didier Bouhassira, France
Dimos D. Mitsikostas, Greece
Dirk Dressler, Germany
Donald Grosset, UK
Elena Pita Calandre, Spain
Emma Reynish, UK
Ezio Giacobini, Switzerland
Francesc Valldeoriola, Spain
Frederick Marshall, USA
Giancarlo Pepeu, Italy
Giorgio Sandrini, Italy
Giorgio Zanchin, Italy
Giuliano Avanzini, Italy
Giustino Varrassi, Italy
Hans-Peter Hartung, Germany
Harald Kollegger, Austria
Heinz Reichmann, Germany
Hubert Fernandez, USA
Hugh Markus, UK
Jaime Kulisevsky, Spain
Jay Salpekar, USA
Jean-Claude Baron, UK
Jens Volkmann, Germany
Jette Frederiksen, Denmark
Joachim K. Krauss, Germany
Joseph A. Lieberman III, USA
Joseph Neimat, USA
Juan Jose Martin, Argentina
Koji Abe, Japan
Krzysztof Selmaj, Poland
Larry Tune, MD, Director, USA
Ley Sander, UK
Luigi Ferini-Strambi, Italy
Maja Relja, Croatia
Mamede de Carvalho, Portugal
Marco Onofrj, Italy
Maria M. Wysocka-Bakowska,
Poland
Mark S Freedman, Canada
Martin J Brodie, UK
Matthew Walker, UK
Mauro Manconi, Italy
Michael Brainin, Austria
Michael Ferrante, USA
Michael Hutchinson, Ireland
Michael Moskowitz, USA
Michael Schüpbach, France
Michael Thorpy, USA
Michael Zenz, Germany
Micheal P Namaka, USA
The European Neurological Journal (enj)
editor@slm-journals.com
Copyright © 2009 San Lucas Medical Limited
Michel Lantéri-Minet, France
Miguel JA Láinez, Spain
Murray Raskind, USA
Nadine Attal, France
Neil W. Kowall, USA
Norman Putzki, Switzerland
Olivier Heinzlef, France
Per Soelberg Sørensen, Denmark
Peter D. Schellinger, Germany
Peter Hagell, Sweden
Peter Rieckmann, USA
Peter Wolf, MD, Denmark
Philip Patsalos, UK
Philippe Coubesa, France
Randolph W. Evans, USA
Renzo Guerrini, Italy
Richard Dodel, Germany
Rico-Villademoros, Spain
Robert Zivadinov, USA
Roger Simon, USA
Samuel Gandy, USA
Sid Gilman, MD, USA
Solomon L. Moshé, USA
Stefan Schwab, Germany
Stephane Thobois, France
Stewart J. Tepper, USA
Sudhansu Chokroverty, USA
Susanna Lindvall, Sweden
Svein I. Johannessen, Norway
Thomas Roth, USA
Vincent Meininger, France
Walter Paulus, Germany
William O. Tatum IV, USA
William R. Markesbery, USA
Zuzana Walker, UK
www.slm-neurology.com
Print ISSN 2041-8000
The European
Neurological Journal (enj)
Publishing Information
The European Neurological Journal (ENJ) [Print ISSN
2041-8000] is published quarterly (February/March, July, August and November), by San Lucas Medical, Ltd. Copyright
2009 by San Lucas Medical, Ltd.
Scientific Review and Acceptance
The ENJ publishes original research manuscripts, reviews,
editorials, correspondences and special articles on topics such
as treatment guidelines. The focus of The European Neurological Journal is on original reports, but other high-quality work
may be considered.
The European Neurological Journal is distributed in print
and electronic form across Europe to thousands of physicians,
researchers, academics, nurses and related care practitioners
with an interest in neurology. Both subscription and access are
free and there are no author fees for publication.
Submitted manuscripts are reviewed by members of the ENJ
editorial board by a double-blind peer review process with the
understanding that manuscripts have not been submitted for
publication elsewhere. Papers are accepted for publication
based on originality, quality and relevance to the audience.
Manuscript Submission
All prospective authors must submit their manuscripts
electronically; a manuscript may be e-mailed directly to the
editorial team (editor@slm-journals.com) or submitted online
(http://www.slm-neurology.com). An acknowledgement of receipt will be sent to the author within 48 hours, and the author
will be given the assigned reference number of his/her manuscript. Prior to submitting a manuscript, authors are encouraged to subscribe to the journal, as having a subscription will
allow us to better track manuscripts. This will also allow us,
once your manuscript is published, to quickly send you a copy
of the journal and proofs.
Manuscript Lead Times
Authors should expect to receive a first decision with reviewers’ comments approximately four weeks after submission.
Instructions for Authors
Manuscripts should be prepared in concise, clear, and grammatically correct English. They should contain the following:
a title page; a short letter transferring copyright to San Lucas
Medical Limited if the article is accepted for publication; an
abstract of up to 250 words; and up to 7 keywords for indexing
purposes. Abstracts are also required for review articles. Text
should contain the following: introduction, results, discussion,
materials and methods, acknowledgments and references,
figure legends, footnotes, tables (with descriptive titles and
legends), conflicts of interest, and a disclosure of funding.
Style & Format
MS Word (version 2.0 and above), Word Perfect (version
5.1 and above) and RTF are the preferred formats to receive
submissions.
Referencing
Manuscripts must use the Vancouver (author-number) style
of referencing (i.e., a number series to indicate references). The
bibliography must then list these references in the numerical
order in which they appear in the text.
Images & Tables
Tables and charts should complement, rather than duplicate,
the text. Photographs and illustrations must be the following:
supplied in high resolution, at least 300 dpi, JPEG/TIF format,
and CMYK or grayscale.
Fees
There are no author fees.
Free Subscriptions
Access to The European Neurological Journal is free. To
obtain a free subscription, please send your contact details to
editor@slm-journals.com. Please remember to include your
full name, title, medical speciality, postal address and email
address. Alternatively, you can subscribe online at http://www.
slm-neurology.com
Correspondence
The European Neurological Journal is committed to allowing a range of opinions to be expressed and publishes appropriate reader correspondence. To comment on a published article,
please send a letter to the editor to editor@slm-journals.com
Copyright and Permissions
Any use of this journal must include the customary bibliographic citation that includes author attribution, date and article title. All articles are made available and publicly accessible
via the Internet and are free to use for non-commercial purposes. San Lucas Medical offers a reprint service for individuals
who require professional quality reproductions of articles.
Advertising
For price quotes and to purchase display advertising, please
contact Jim Jones (editor@slm-journals.com).
Artwork must be submitted as high-resolution digital media
with a minimum of 300 dpi. Preferred formats are as follows:
PDF/Acrobat 4.05 or later; Adobe Illustrator (with all fonts outlined); Adobe Photoshop EPS, TIFF or JPEG (CMYK and 300
dpi): and Adobe InDesign (packaged with all links). Please send
all artwork electronically to editor@slm-journals.com.
Disclaimer
Every precaution has been taken by the authors, editors
and the board to verify the results and data published in The
European Neurological Journal (ENJ). However, the opinions
expressed are the responsibility of the contributors, and not
necessarily those of the ENJ editorial team or board. As a
result, ENJ accepts no liability for the consequences of any inaccurate or misleading data, statement or opinion. Please write
to the editor if you identify an error.
The European Neurological Journal wishes to thank our
media partners for their assistance. This does not constitute
any official endorsement.
The European
Neurological Journal (enj)
Contents
September 2009 – Vol 1, No 1
REVIEW ARTICLES
Third-generation Antiepileptic Drugs for Partial-onset Seizures: Lacosamide,
Retigabine, and Eslicarbazepine Acetate
Steve S. Chung .................................................................................................................................................................................... 1
The Impact of the Extended Time Window seen in the ECASS III Trial
on the Guidelines for Stroke Management in Europe
Keith W. Muir ...................................................................................................................................................................................... 13
Incidence and Lifetime Risk of Parkinson’s Disease in Advanced Age:
Review and Estimates from the United States
Jane A. Driver and Tobias Kurth ...................................................................................................................................................... 19
Imaging in Familial Frontotemporal Lobar Degeneration with Mutations in MAPT or PGRN
Jennifer L. Whitwell and Keith A. Josephs ...................................................................................................................................... 25
Frontal and Periventricular Brain White Matter Lesions and Cortical Deafferentation
of Cholinergic and other Neuromodulatory Axonal Projections
N.I. Bohnen, C.W. Bogan and M.L.T.M. Müller................................................................................................................................. 33
Restless Legs Syndrome and Peripheral Neuropathy—A Critical Review
ET Hattan, C Chalk and RB Postuma .............................................................................................................................................. 51
Orthostatic Headache with and without CSF Leak
Andrea N. Leep Hunderfund and Bahram Mokri . ........................................................................................................................ 47
Obesity, Diet, and Risk of Restless Legs Syndrome
Xiang Gao and Shivani Sahni . ....................................................................................................................................................... 59
Neuroimaging of Primary Progressive Aphasia
Jonathan D Rohrer and Nick C Fox ............................................................................................................................................... 65
Subcortical Gliosis and Leukodystrophy Overlap Syndromes as a Cause of Late-Onset Dementia
Russell H Swerdlow, Bradley B Miller, H Robert Brashear and Jeffrey M Burns ............................................................................ 75
The European Neurological Journal (enj)
editor@slm-journals.com
Copyright © 2009 San Lucas Medical Limited
www.slm-neurology.com
Print ISSN 2041-8000
Data Collection & Benchmarking in Europe
The European
Neurological Journal (enj)
Subscriptions
To obtain your FREE subscription to the journal please complete and return the form below.
First Name _ ____________________________________ Last Name _ _______________________________________
Job Title ________________________________________ Primary Specialisation ____________________________
Main Hospital/Clinic ____________________________ e-Mail _____________________________________________
Postal Address Line 1 __________________________________________________________________________________
Postal Address Line 2 __________________________________________________________________________________
City _____________________________________________ County/State/Province ____________________________
Postcode/Zip ____________________________________ Country ___________________________________________
Subscription preference (print or electronic):
❏
Print
❏
Electronic
Refer a colleague (please enter their details)
First Name _ ____________________________________ Last Name _ _______________________________________
Job Title ________________________________________ Primary Specialisation ____________________________
Main Hospital/Clinic ____________________________ e-Mail _____________________________________________
Postal Address Line 1 __________________________________________________________________________________
Postal Address Line 2 __________________________________________________________________________________
City _____________________________________________ County/State/Province ____________________________
Postcode/Zip ____________________________________ Country ___________________________________________
Subscription preference (print or electronic):
❏
Print
❏
Electronic
Remove and fax this form to +44 (0) 20 7084 0330 or an electronic version, that can be completed can accessed from
www.slm-neurology.com or e-mail the completed form to editor@slm-journals.com.
All subscriptions are Free.
11th International Geneva/Springfield Symposium
on Advances in Alzheimer Therapy
As at the previous ten meetings, the 2010 Geneva meeting will continue
to focus on the Pharmacological Therapy of Alzheimer Disease, giving
particular emphasis to the discovery of new drugs. The conference will be
useful to specialists as well as non-specialists in understanding the present
pharmacological approach to the disease.
Organizers:
Ezio Giacobini, MD
Gabriel Gold, MD
Coordinators:
Ann Hamilton, (USA)
Christine Mesmer,
(Europe)
Meeting date
Location
March 24-27, 2010
CICG, International Conference Centre
Geneva, Switzerland
Sponsors are:
Southern Illinois University School of Medicine, Springfield, Illinois, USA;
University of Geneva Medical School, Geneva, Switzerland; and Geneva
University Hospitals, Department of Rehabilitation and Geriatrics, Geneva,
Switzerland.
For more information, visit our web site: www.siumed.edu/cme
European Neurological Journal
review article
Third-generation Antiepileptic Drugs
for Partial-onset Seizures: Lacosamide,
Retigabine, and Eslicarbazepine Acetate
Steve S. Chung
Affiliations: Department of Neurology, Barrow Neurological Institute, St. Joseph’s Hospital and Medical Center, Phoenix,
Arizona, USA
Submission date: 1st September 2009, Revision date: 17th September 2009, Acceptance date: 22nd September 2009
A B STRA C T
Despite the advent of new antiepileptic drugs (AEDs), more than 30% of epilepsy patients remain poorly controlled with
current AEDs. For these patients, combined administration of AEDs or the application of novel AEDs are the most appropriate therapeutic options when surgical treatment cannot be offered. Second-generation and more recently developed
AEDs tend to offer new mechanisms of action and more favorable safety profiles than the first-generation AEDs. The purpose of this article is to compare and review the information on the molecular mechanisms of action, pharmacokinetic
profiles and the preliminary results of phase II and III clinical trials of three new AEDs – lacosamide (LCM), eslicarbazepine
acetate (ESL), and retigabine (RTG).
Keywords: partial seizure, new anticonvulsant, lacosamide, eslicarbazepine acetate, retigabine, KCNQ channels, sodium
channels
Correspondence: Steve S. Chung, MD, Department of Neurology, 500 West Thomas Road Suite 300, Phoenix, Arizona 85013,
USA. Tel: +1-602-4066271; fax: +1-602-7980852; e-mail: sschung@chw.edu
INTRODUCTION
huperzine, lacosamide, losigamone, remacemide hydrochloride, retigabine, rufinamide, safinamide, soretolide,
stiripentol, talampanel, tonabersat, and valrocemide.
Some of these AEDs have already undergone preclinical
and clinical studies, and this article will focus on three
of them (lacosamide, eslicarbazepine acetate, and retigabine), which have completed phase II and III clinical
trials, and will review their pharmacokinetic profiles,
drug interactions, molecular mechanisms of action, efficacy, and tolerability.
Epilepsy is one of the most common neurological disorders affecting up to 2% of the population worldwide,
and almost 2 million people in the United States alone
[1]. Treatment of epilepsy often imposes an exposure to
various antiepileptic drugs (AEDs) and requires longterm commitment and compliance from the patient.
Excluding the small percentage of people who undergo
successful epilepsy surgery, the vast majority of patients
are maintained through chronic medical management
for appropriate seizure control. Despite the advent of
new AEDs over the past 15 years, approximately 30%
of epilepsy patients experience recurrent seizures [2, 3]
and many experience undesirable side effects. Therefore, there are still unmet needs for the treatment of
epilepsy and there remains a need to develop new AEDs
that can reduce seizure frequency and severity as well as
improve tolerability and safety.
LACOSAMIDE (LCM)
LCM, (R) - 2 - acetamido - N - benzyl - 3 - methoxypropionamide (Table 1), is a novel antiepileptic drug that
is the result of focused research on functionalized amino
acids with anticonvulsant activity [4, 5]. Based on the
efficacy and therapeutic index observed in a range of
animal models of epilepsy at the National Institutes of
Health (NIH) Anticonvulsant Screening Program, LCM
warranted further evaluation and was subsequently developed as an AED for both oral and intravenous use.
Additionally, LCM is available as an oral syrup (15 mg/
mL) in Europe. It has a novel mode of action (MOA) that
appears to be different from existing AEDs, namely the
selective enhancement of slow inactivation of voltagegated sodium channels. In August 2008, LCM was ap-
For those patients with medically refractory epilepsy,
combined administration of AEDs or the use of new
AEDs are the most appropriate therapeutic options.
Since many new novel AEDs have recently been investigated, this group of AEDs is often referred to as thirdgeneration AEDs, and includes brivaracetam, carabersat, carisbamate, eslicarbazepine acetate, ganaxolone,
ENJ 2009; 1: (1). September 2009
1
www.slm-neurology.com
European Neurological Journal
atinine clearance of ≤30 mL/min) and in patients with
end stage renal disease, a maximum dose of 250 mg/day
(EU) or 300 mg/day (USA) is recommended [13, 14].
Table 1. Chemical Structures of the Newest AEDs
AED
Chemical structure
Drug Interactions
Lacosamide
LCM has a low potential for drug–drug interactions.
The minimal binding of LCM to plasma proteins minimizes the potential for displacement of other drugs [9].
Furthermore, LCM has no interaction or minimal interaction with CYP-450 isoforms, making an effect on the
metabolism of other drugs unlikely [13, 14]. In clinical
efficacy and safety trials, LCM did not alter the drug
plasma levels of other AEDs (carbamazepine, oxcarbazepine, gabapentin, lamotrigine, levetiracetam, phenytoin, topiramate, valproic acid and zonisamide). Specific
drug-interaction studies involving carbamazepine, valproic acid, omeprazole, metformin, digoxin and an oral
contraceptive (ethinyl estradiol and levonorgestrel) also
demonstrated no relevant interaction influence on the
pharmacokinetics of these drugs or LCM [15, 16].
Eslicarbazepine
Retigabine
proved by the European Medicines Agency (EMEA) as
an adjunctive treatment for partial-onset seizures for
patients ≥16 years, and in October 2008 by the U.S. Food
and Drug Administration (FDA) for patients ≥17 years.
Mechanisms of Action
The precise mechanisms by which LCM exerts its antiepileptic effect in humans are not fully understood,
but a novel mode of action has been suggested. LCM selectively enhances slow inactivation of voltage-dependent sodium channels without affecting fast inactivation,
which may normalize neuronal firing thresholds [17].
Classical anticonvulsant drugs such as carbamazepine,
phenytoin, and lamotrigine act on fast inactivation of
voltage-dependent sodium channels [17].
Pharmacokinetics
LCM has a linear pharmacokinetic profile with high
oral bioavailability [6]. Studies in healthy volunteers
have demonstrated that LCM is rapidly and completely
absorbed [7–9]. The rate and extent of absorption are
not affected by the presence of food [7]. Peak serum concentrations occur 0.5 to 4 h after oral intake, and the
elimination half-life of LCM is about 13 h, allowing convenient twice-daily dosing [5, 6, 10]. The LCM solution
for infusion is typically administered over 15 to 60 min
and the maximum concentration (Cmax) is reached at the
end of infusion. Studies in healthy volunteers demonstrated bioequivalence for Cmax and area under the curve
(AUC) for both the 30 and 60 min infusion durations
[11]. Infusion over 15 min was near bioequivalent, with
a slightly higher Cmax and equivalence for AUC [12].
LCM demonstrated potent anticonvulsant activity
in a broad range of animal models of partial onset and
pharmacoresistant seizures, generalized tonic-clonic
seizures, as well as status epilepticus. Intraperitoneal
LCM was effective in preventing seizures in the 6 Hz
psychomotor seizure model (dose of drug that is pharmacologically effective for 50% of the population exposed
to the drug (ED50), 9.99 mg/kg) and audiogenic seizure
model (ED50 0.63 mg/kg). Intraperitoneal LCM (20 mg/
kg) completely prevented tonic convulsions and 50 mg/
kg provided partial protection against clonic convulsions
induced by N-methyl-D-aspartate (NMDA) in mice [17,
18]. LCM was also effective in amygdala and hippocampal kindling models. In hippocampal kindled rats, the
activity of LCM (25 mg/kg) was superior to that of maximally effective doses of phenytoin (150 mg/kg) and carbamazepine (50 mg/kg), valproic acid (250 mg/kg) and
ethosuximide (250 mg/kg) [17]. However, LCM was not
effective against clonic seizures induced by pentylenetetrazole (half maximal effective concentration (EC50)
~25 mg/kg), bicuculline (EC50 >50 mg/kg), or picrotoxin
(EC50 >30 mg/kg) in rodents [17, 18]. LCM was also effective in models of status epilepticus, stopping limbic
seizures induced by self-sustaining status epilepticus
in rats within 15 min of administration and preventing
their recurrence over the next 24 h [17].
LCM has low plasma protein binding (≤15%) and the
volume of distribution is approximately 0.6 L/kg, which
is similar to total body water [13]. The pharmacokinetics of both oral and intravenous LCM were dose-proportional (up to 800 mg), with low intra- and intersubject
variability. Following twice-daily administration of oral
LCM, steady-state plasma concentrations were reached
after 3 days [5, 13].
Lacosamide is primarily eliminated renally as unchanged drug (>40%) and an inactive metabolite, Odesmethyl metabolite (<30%) [5, 6, 10]. Although the
hepatic isoenzyme 2C19 is mainly responsible for the
formation of the O-desmethyl metabolite, coadministration of CYP2C19 inhibitors did not cause clinically relevant differences in the pharmacokinetics of LCM, indicating that the metabolic pathway involving CYP2C19 is
minor. For patients with severe renal impairment (creENJ 2009; 1: (1). September 2009
2
www.slm-neurology.com
Third-generation Antiepileptic Drugs for Partial-onset Seizures: Lacosamide, Retigabine, and Eslicarbazepine Acetate
Efficacy
LCM onset of action appears rapid, since there was already a significant seizure reduction compared to placebo as early as the first week when patients were receiving 100 mg per day regardless of assigned dose group in
the pooled analysis (median percent reduction in seizure
frequency: 33.0% vs. 19.4%, P<0.01) [24].
Three pivotal studies (one phase II and two phase III
studies) have been conducted to establish the efficacy
and safety of LCM [19–21]. Three doses of LCM (200,
400, and 600 mg/day) were administered as adjunctive
therapy for patients with partial epilepsy with or without secondary generalization, with a starting dosage of
50 mg twice a day (BID), followed by a weekly increase
of 100 mg/day to the target dose. The titration phase
was followed by a 12-week maintenance phase with an
option for continued open-label treatment. In total, 1294
patients were randomized in three studies with a mean
age of 38.6 years. The studies were conducted in a very
refractory population, with 84.4% of subjects taking
two or three concomitant AEDs (including substantial
numbers on newer AEDs) and 17% being additionally
treated with the vagal nerve stimulator. Approximately
half of the participants had tried 7 or more AEDs in the
past [22, 23].
Safety and Tolerability
LCM was generally well tolerated in patients with partial-onset seizures, with most treatment-emergent adverse events (TEAEs) being of mild or moderate severity
[25, 26]. The most common TEAEs of oral LCM were
dizziness, headache, nausea and diplopia. All of these
TEAEs were dose-related except for headache, with incidence typically reported during titration rather than
during the maintenance phase. Overall, discontinuation
rates due to TEAEs were 8% in LCM 200 mg/day, 17%
in 400 mg/day, and 29% in 600 mg/day, compared to 5%
of placebo recipients [22, 23, 25]. The incidence of somnolence during the treatment period was approximately
5% for placebo and 7% for the total LCM groups, and did
not appear to be dose-related [25]. The incidence of rash
was low for patients randomized to LCM similar to that
reported with placebo (3%). No rashes were serious and
all were assessed as mild to moderate in intensity.
The primary assessment of efficacy was based on
the change in partial-onset seizure frequency and was
evaluated in two ways: (1) the change in seizure frequency per 28 days from baseline to the maintenance
period, and (2) the proportion of patients who experienced a 50% or greater reduction in seizure frequency
from baseline to maintenance period (50% responder
rate). The primary efficacy analysis was conducted on
the intent-to-treat (ITT) population, which is defined as
all randomized patients who received at least one dose
of the trial medication and had at least one postbaseline efficacy assessment. In the phase II study, the 50%
responder rates were 32.7% for 200 mg/day (P=0.090),
41.1% for 400 mg/day (P=0.004), and 38.1% for 600 mg/
day (P=0.014), compared with 21.9% for the placebo
group [19]. Percent reduction in seizure frequency per
28 days over placebo was 14.6% in the 200 mg/day group
(P=0.101) and reached statistical significance for both
the LCM 400 mg/day (28.4%, P=0.002) and 600 mg/
day (21.3%, P=0.008) groups. Two subsequent phase
III studies confirmed the efficacy and safety of LCM at
doses of 200–600 mg/day [20, 21].
Results of clinical laboratory tests and vital sign measurements across treatment groups did not identify
changes of significant clinical concern that appeared to
be associated with LCM. LCM does not prolong the QTc
interval or have clinically important effects on QRS duration. A small increase in mean PR interval was seen
during treatment with the mean maximum placebo-subtracted change of 1.5 ms for 200 mg and 3.1 ms for 400
mg [13]. There were no reports of adverse events associated with PR interval prolongation, and the degree of
increase is considered to be similar to other AEDs that
may affect PR interval, such as carbamazepine (8–16 ms
increase), lamotrigine (5 ms increase), and pregabalin
(up to 5 ms increase) [27–30].
The tolerability profile of short-term intravenous LCM
was similar to oral lacosamide, and the incidence of injection site pain was low. In a 2 day, randomized, double-blind, placebo-controlled study, patients currently
undergoing treatment with oral lacosamide (n=60, aged
19–61years) were randomized to either oral LCM (plus
placebo infusion) or 30 or 60 min intravenous LCM infusions (plus oral placebo) [31]. The intravenous LCM
dosage was the same as the previous oral dosage range
(200–600 mg/day). TEAEs associated with intravenous
LCM were mild or moderate in intensity and included
dizziness (0%, 5%, 10% in placebo, 60 min, and 30 min
infusion, respectively), headache (5%, 10%, 0%), back
pain (0%, 10%, 0%), and somnolence (0%, 0%, 11%).
Infusion site-related pain was infrequent (0% in the 60
min infusion and 11% in the 30 min infusion), and did
not result in discontinuations of LCM [31]. In another open-label study (n=60) in which LCM was infused
Subsequent analysis of pooled efficacy data from these
trials further supports the overall efficacy of LCM at doses of 200–600 mg/day. For the pooled analysis, the 50%
responder rates per 28 days from baseline to the maintenance period were 22.6% for placebo, 34.1% for LCM
200 mg/day, and 39.7% for LCM 400 mg/day. The median
percent reduction in seizure frequency was 18.4% for
placebo, 33.3% for LCM 200 mg/day, and 36.8% for LCM
400 mg/day [22, 23]. Overall, the LCM 600 mg/day group
showed similar efficacy to the 400 mg/day group. For
those who completed the maintenance period, pooled
analysis demonstrates that complete seizure freedom
during the maintenance period was achieved in 2.7%,
3.3% and 4.8% of patients randomized to LCM 200, 400,
and 600 mg/day, respectively, compared with 0.9% in the
placebo group [22, 23].
www.slm-neurology.com
3
ENJ 2009; 1: (1). September 2009
European Neurological Journal
more rapidly over 10, 15, or 30 min for 2 to 5 days (200
to 800 mg/day), the incidence of adverse events was similar with headache (5%, 7%, 8%) and dizziness (5%, 6%,
8%) being most commonly reported, respectively [32].
warfarin [12]. Although glucuronidation is the major
metabolic pathway for both eslicarbazepine and lamotrigine, pharmacokinetic studies between ESL 1200 mg/
day and lamotrigine 150 mg/day in 32 healthy male volunteers showed no changes in Cmax or AUC for either
drug [33]. Population pharmacokinetics analysis of
data from phase III studies in adults with epilepsy also
showed no relevant effect of ESL on the clearance of carbamazepine, phenytoin, topiramate, clobazam, gabapentin, phenobarbital, levetiracetam and valproic acid
[41–43]. In addition, protein binding of eslicarbazepine
was not affected significantly by the presence of warfarin, diazepam, digoxin, phenytoin and tolbutamide [12].
ESLICARBAZEPINE ACETATE (ESL)
ESL is a prodrug of eslicarbazepine (S-9-(–)-10-acetoxy-10,11-dihydro-5H-dibenz/b,f/azepine-5-carboxamide) and shares with carbamazepine and oxcarbazepine
the dibenzazepine nucleus bearing the 5-carboxamide
substitute but is structurally different at the 10,11-position (Table 1). This molecular variation results in differences in metabolism and tolerability, and once daily
dosing. ESL is considered as a third-generation, single
enantiomer member of the established family of dibenz/
b,f/azepine AEDs represented by carbamazepine and
oxcarbazepine [33]. It was granted marketing authorization in April 2009 by EMEA as an adjunctive therapy
for partial seizures in patients ≥18 years, although it has
not yet been approved by the U.S. Food and Drug Administration.
A clinical study with ESL 1200 mg daily demonstrated
that the plasma concentrations of oral contraceptives,
both ethynylestradiol and levonorgestrel, were reduced,
and AUC decreased by 32% and 24%, respectively [34],
which may have clinical consequences.
Mechanisms of Action
The precise MOA of ESL is not known but in vitro
electrophysiological studies indicate that both ESL
and eslicarbazepine competitively interact with the inactivated state of a voltage-gated sodium channel, and
thereby prevent its return to the active state [44, 45].
Effects at the voltage-gated sodium channels are probably the main MOA of ESL to limit sustained repetitive firing, ictogenesis and seizure spread. The affinity
of ESL for the sodium channel in the resting state is
similar to that of carbamazepine, but the affinity for the
channel is about three times lower, possibly suggesting
the higher inhibitory selectivity of ESL for rapidly firing neurons over those displaying normal activity [45].
Earlier studies demonstrated that ESL acted similarly
to carbamazepine and oxcarbazepine in inhibition of
release of neurotransmitters or neuromodulators, such
as glutamate, gamma-aminobutyric acid (GABA), aspartate and dopamine in rat striatal slices [46, 47].
Pharmacokinetics
Eslicarbazepine is the main active metabolite of ESL
and represents about 95% of the total systemic drug
exposure. ESL is rapidly and extensively metabolized
to eslicarbazepine by a hydrolytic first-pass metabolism within 1 to 4 h [34]. Unlike carbamazepine, ESL
is not metabolized to carbamazepine-10,11-epoxide and
is not susceptible to metabolic autoinduction [35]. Unlike oxcarbazepine, which is a prodrug to both eslicarbazepine (also called S-licarbazepine or S-MHD) and
R-licarbazepine (also called R-MHD), ESL is a prodrug of eslicarbazepine [36]. In adult epilepsy patients,
the half-life of eslicarbazepine was 13 to 20 h and the
steady-state concentration was reached within 4 to 5
days of once daily dosing [34]. The pharmacokinetics
of eslicarbazepine is linear and not affected by gender
[37]. ESL is almost completely absorbed (>90%) with
or without food [34, 38]. The volume of distribution of
eslicarbazepine is about 34 liters and protein binding is
estimated to be less than 40% [34].
ESL demonstrated anticonvulsant activity in several
animal models. It blocks tonic seizures in the maximal
electroshock seizure (MES) model and limbic seizures
in the corneal kindled mouse and amygdala-kindled rat
[34]. However, ESL displays only weak effects against
clonic seizures induced by pentylenetetrazole (PTZ),
bicuculline, picrotoxin, and 4-aminopyridine [12].
The metabolites of ESL are primarily excreted through
kidney in unchanged form and as glucuronide conjugates (30%). As their clearance is dependent on renal
function, dosage adjustment may be necessary in patients with creatinine clearance below 60 mL/min [39].
However, the clearance of ESL and its metabolites was
not affected by moderate hepatic impairment [40].
Efficacy
An early phase II placebo-controlled study found that
ESL could be an efficacious and well-tolerated treatment
option for patients with refractory partial-onset seizures
[48]. The trial was conducted in Croatia, Czech Republic, Germany, Lithuania, and Poland. In this study, the
percentage of responders showed a statistically significant difference between ESL and placebo groups (54%
vs. 28%; P=0.008).
Drug Interactions
ESL does not cause an inhibitory effect on the activity of CYP1A2, CYP2A6, CYP2B6, CYP2D6, CYP2E1,
CYP3A4 and CYP2C9, but a moderate inhibitory effect
was seen on CYP2C19 [12]. ESL may have a mild inducing effect on CYP2C9 since coadministration of ESL
with warfarin showed a decrease in exposure to (S)ENJ 2009; 1: (1). September 2009
One of the three subsequent phase III trials recently
4
www.slm-neurology.com
Third-generation Antiepileptic Drugs for Partial-onset Seizures: Lacosamide, Retigabine, and Eslicarbazepine Acetate
published [49] also showed similar results. The study was
conducted in 11 European countries including Austria,
Croatia, Czech Republic, Germany, Hungary, Lithuania,
Poland, Romania, Russia, Switzerland, and Ukraine. In
this study, patients were randomized to placebo (n=102)
or once daily ESL 400 mg (n=100), 800 mg (n=98), or
1200 mg (n=102) in the double-blind treatment phase.
The starting dose of ESL was 400 mg with a weekly increase of 400 mg to the full target doses. The study found
that the 50% responder rate in the ITT population was
20% in placebo, 23% in 400 mg (not significant), 34%
in 800 mg (P=0.0359), and 43% in the 1200 mg group
(P=0.0009). The median reduction in seizure frequency
was 16% (placebo), 26% (400 mg, not significant), 36%
(800 mg, P<0.01), and 45% (1200 mg, P<0.01). The
most frequent concomitant AEDs were carbamazepine
(56–62% of patients), followed by lamotrigine and valproic acid (22–28%). Overall, similar efficacy results were
obtained in patients administered ESL with or without
carbamazepine as concomitant AED. Two other phase
III trials in 23 European countries also showed similar
results. All three phase III studies utilized a multi-center, randomized, double-blind, placebo-controlled design
and included patients with at least four partial seizures
per 4 weeks despite treatment with up to three AEDs.
Three doses of ESL (400, 800 or 1200 mg once daily)
were examined as an adjunctive therapy and consisted
of an 8 week baseline period, followed by double-blind
2 week titration and a double-blind 12 week maintenance period [12]. The most commonly used AED in any
treatment group was carbamazepine, and it was used
in approximately 60% of the study patients. Combined
analysis showed that ESL dosages of 800 and 1200 mg
once daily demonstrated a significant median seizure
frequency reduction compared to placebo (P<0.0001).
However, no significantly different responder rate was
found between the 400 mg and placebo arms in any
study [12]. Long-term open-label treatment up to 1 year
showed a reduction in seizure frequency with ESL 800
mg daily dose, and improvement in health-related quality of life and depressive symptoms [50].
ability has been reported between adults and elderly patients, and no abnormal vital signs were seen in patients
on ESL [34, 50]. Evaluation of electrocardiogram (EKG)
recordings during clinical trials showed an increase in
PR interval in ESL-treated patients, which was highest
in the 1200 mg dose group (mean increase of 5.5 ± 30.6
ms), compared to the placebo group (mean decrease of
–0.8 ± 20.6 ms) [51].
RETIGABINE (RTG)
RTG (N - [2 - amino - 4 - (4 - fluorobenzylamino) -phenyl] carbamic acid ethyl ester) is a new antiepileptic
medication with a novel mechanism of action (Table
1). It was initially identified through a drug screening
program at the National Institutes of Health in 1991,
and subsequently introduced in 1994 as a chemical compound D-23129 with a broad-spectrum activity in animal models of epilepsy [52]. More recently, it has been
developed as an adjunctive treatment for partial epilepsy. RTG’s anticonvulsant properties are primarily mediated by opening or activating neuronal voltage-gated potassium channels [53–55]. Up to the present time, RTG
has not been approved either by EMEA or FDA.
Pharmacokinetics
RTG demonstrates a linear pharmacokinetic profile
with dosages up to 1200 mg/day [56]. It is rapidly absorbed following oral administration and reaches peak
plasma concentrations within 1 to 1.5 h. The bioavailability of orally administered RTG is estimated to be
about 60%, and the total RTG absorption is not affected
when administered with food [57]. However, the peak
plasma concentration of RTG is delayed to approximately 2 h with food, and modestly increased when RTG is
taken with a high-fat meal, although AUC remains unchanged. Bioavailability is not affected by gender or age
(age range: 18–81 years). Protein binding is estimated
to be less than 80% and the volume of distribution at
steady-state is about 2–3 L/kg [58].
RTG is metabolized by N-acetylation to the monoacetylated metabolite (primary metabolite) and by glucuronidation to form an N-glucuronide structure, which
demonstrate minimal pharmacologic activity [59, 60].
Both RTG and its primary metabolite have a plasma
half-life of 8 h (7.2 to 9.4 h). After reaching steady-state,
mean values for oral clearance were 0.51–0.71 L/h/kg.
Safety and Tolerability
The pooled population of adults with epilepsy included
in placebo-controlled studies showed that the most commonly reported TEAEs with an incidence >2% (ESL
vs. placebo) were dizziness (18.8% vs. 5.7%), somnolence (11.2% vs. 7.4%), nausea (6.5% vs. 2.4%), diplopia
(6.3% vs. 1.2%), and headache (5.5% vs. 2.1%), vomiting
(4.8% vs. 1.2%), abnormal coordination (4.4% vs. 1.8%),
blurred vision (3.5% vs. 0.9%), vertigo (2.1% vs. 0%) and
fatigue (2.1% vs. 1.8%) [12]. Overall, TEAEs of ESL were
mild to moderate and appeared to be dose-dependent
[12]. The incidence of AE-related discontinuation was
low (4.5% with placebo, 8.7% with ESL 400 mg, 11.6%
with 800 mg and 19.3% with 1200 mg). The incidence
of psychiatric complications, rash, or hyponatremia was
low (<1% of patients) [12, 50]. No difference in tolerwww.slm-neurology.com
The majority of the drug and its metabolites are renally excreted without further hepatic metabolism.
Although RTG does not affect CYP2C8, CYP2C9,
CYP2C19, CYP3A4/5, and CYP4A9, it has a modest potential to inhibit the CYP2A6 isoform [12]. Renal clearance of RTG was reduced by approximately 25% in individuals with mild renal dysfunction and approximately
50% in those with moderate or severe renal disease or
those who required dialysis [12]. In the elderly population, RTG clearance is reduced by approximately 30%
when compared with younger subjects, probably related
5
ENJ 2009; 1: (1). September 2009
European Neurological Journal
channels are composed of a heteromeric or homologous
assembly of the different subunits KCNQ1, KCNQ2,
KCNQ3, KCNQ4, and KCNQ5. RTG selectively enhances M-currents through heteromeric KCNQ2/3 channels
[53, 55, 68, 69] and KCNQ3/5 channels [55] as well as
homomeric KCNQ5 channels [70, 71]. Selectivity of
RTG is important for safety since KCNQ1 subunits are
present in cardiac cells [72, 73], and KCNQ4 subunits in
the auditory system [74, 75]. Mutations of gene products
for these potassium channels can result in epilepsy syndrome known as benign familial neonatal convulsions
[76–78]. Therefore, these potassium channels may play
an integral role in controlling epileptiform discharges,
and enhancement of M-type current acts as a braking
current on action potential discharges [79].
to changes in renal function [12]. In patients with moderate or severe hepatic impairment, RTG clearance is reduced by 30% to 50% [12]. RTG displayed a mild degree
of intrasubject variability of less than 30%, and some
diurnal variation in that trough plasma concentration
was approximately 35% lower in the evening than in the
morning [58].
Drug Interactions
RTG does not induce or inhibit its own metabolism.
A number of studies in healthy volunteers and epilepsy patients have revealed no clinically significant
pharmacokinetic interactions between retigabine and
valproate or topiramate [56, 61]. However, phenytoin
and carbamazepine may increase the clearance of RTG
(N-dealkylation pathway) by approximately 30%, especially when a higher dose of retigabine (1200 mg/day) is
administered [56]. In contrast, population pharmacokinetic analyses of data from all clinical studies involving
more than 800 patients did not identify any effect of
enzyme-inducing AEDs (ie, carbamazepine, phenytoin,
or phenobarbital) and no clinically meaningful effects
of nonenzyme-inducing AEDs on RTG pharmacokinetics [62]. RTG combination therapy in patients with epilepsy did not alter the pharmacokinetics of phenytoin,
carbamazepine, valproic acid, or topiramate. In another study with healthy volunteers, lamotrigine mildly
increased the half-life of RTG, while RTG increased
lamotrigine clearance by about 20%, which is probably
because both medications are partially metabolized by
glucuronidation [12, 63]. RTG does not alter the pharmacokinetics or metabolism of the oral contraceptive
steroids ethinyl estradiol/norgestrel [64].
Efficacy
To date, there have been 3 pivotal studies (1 phase
II and 2 phase III studies) conducted to evaluate the
efficacy and safety of RTG. The phase II study was a
double-blind, placebo-controlled, randomized clinical
trial evaluating three doses of RTG (600, 900, and 1200
mg/day) administered as adjunctive therapy in adult patients with partial epilepsy with or without secondary
generalization [80]. In total, 537 patients were screened
and 399 patients were randomized into four different
arms of the study with age range from 16 to 70 years.
The starting dosage of RTG was 100 mg three times a
day (TID), followed by a weekly increase of 150 mg/day
to the target dose. The titration phase was followed by
maintenance phase (8 weeks in phase II and 12 weeks
in phase III trials) where RTG dose reduction was allowed (in phase III trials) mainly due to intolerability.
Those patients who completed the maintenance phase
then had an option to enroll in a long-term, open-label,
extension study.
Mechanisms of Action
RTG demonstrated potent anticonvulsant activity
in various animal models of epileptic seizures [65–67],
which included electrically induced (amygdala kindling, corneal kindling, maximal electroshock), chemically induced (pentylenetetrazole, picrotoxin, cobalthomocysteine thiolactone, and N-methyl-D-aspartate
(NMDA)), and genetic (audiogenic) epilepsy models.
Although clinical trials were limited to partial onset seizures, in animal studies, RTG was effective against both
partial seizure models (amygdala, corneal and hippocampal kindling models) and generalized seizure models
(maximal electroshock seizure (MES), pentylenetetrazole, picrotoxin, genetic epilepsy models) as well as the
status epilepticus model (cobalt-homocysteine thiolactone).
The primary efficacy of all three studies was measured
as the change from baseline monthly (28-day) seizure
frequency. In the phase II study, in the ITT population,
the median percent change in seizure frequency was
23.4% for 600 mg/day, 29.3% for 900 mg/day (P=0.0387),
and 35.2% for 1200 mg/day (P=0.0024), compared with
13.1% for the placebo group [80]. The difference was significant for the RTG 900 and 1200 mg/day arms when
compared to placebo, but no significant difference was
noted between retigabine 600 mg/day and placebo.
When the efficacy was measured by responder rates in
ITT analysis, a significant reduction in partial seizure
frequency vs. placebo was seen in the 900 mg/day and
1200 mg/day arms in a dose-dependent manner: 23%
for 600 mg/day (not significant), 32% for 900 mg/day
(P=0.0214), and 33% for 1200 mg/day (P=0.0214) compared with 16% for placebo [80].
Results of many different studies have indicated that
the anticonvulsant effect of RTG is primarily due to
opening of neuronal voltage-gated potassium channels,
which enhances the M-type potassium current [68]. Mtype potassium currents are a species of subthreshold
voltage-gated potassium current that control neuronal
excitability through stabilizing membrane potentials. MENJ 2009; 1: (1). September 2009
More recently, two phase III studies (RESTORE 1 and
2) confirmed the dose-dependent efficacy of 600, 900,
1200 mg/day retigabine. RESTORE 1 was conducted in
the United States and had two arms (placebo and 1200
6
www.slm-neurology.com
Third-generation Antiepileptic Drugs for Partial-onset Seizures: Lacosamide, Retigabine, and Eslicarbazepine Acetate
Table 2. Summary of the Main Properties of the Newest AEDs
Lacosamide
Eslicarbazepine Acetate
Retigabine
Adjunctive therapy for partial
seizures (≥16 years)
Adjunctive therapy for partial
seizures (≥18 years)
Adjunctive therapy for partial
seizures in adults (proposed)
Approval status
Both EMEA and FDA [26]
EMEA only
Not approved
Mode of action
Selective enhancement of slow
inactivation of sodium channels
Inhibition of voltage-gated
sodium channels
Enhancement of voltage-gated
potassium channels
Indications
Starting dose
Initial target dose
Dosing schedule
Half-life (h)
100 mg/day
400 mg/day
300 mg/day
200–400 mg/day
800–1200 mg/day
600–900 mg/day
BID
QD
TID
13
13–20 (eslicarbazepine)
8–9
Time to Cmax (h)
0.5–4
2–3
1–2
Oral bioavailability (%)
~100
>90
60
Protein binding (%)
<15
<40
<80
BID, twice daily; QD, once daily; TID, three times a day.
50% Responder Rate over Placebo
mg/day), while RESTORE 2 was conducted mainly in
Europe and Australia with three different arms (placebo, 600, and 900 mg/day). In RESTORE 1 (n=301), median seizure frequency vs. placebo was significantly reduced in ITT analysis: 44% for 1200 mg/day (n=151) vs.
18% for the placebo group (n=150) [81]. In RESTORE
2, a significant reduction in partial seizure frequency
was found in both RTG doses vs. placebo (P<0.001):
28% for 600 mg/day, 40% for 900 mg/day, and 16% for
placebo [82].
Safety and Tolerability
During the phase II trial, more frequent central nervous system (CNS)-related TEAEs were seen in all RTG
arms (46% for 600 mg/day, 60% for 900 mg/day, and
72% for 1200 mg/day) than in placebo-treated patients
(32%, P=0.004) [80]. In the same trial, the incidence of
CNS-related symptoms appeared to be dose-related, and
included somnolence (6% for placebo, 17% for 600 mg/
day, 21% for 900 mg/day), headaches (10% for placebo,
11% for 600 mg/day, 15% for 900 mg/day), dizziness (4%
for placebo, 8% for 600 mg/day, 18% for 900 mg/day),
confusion (5% for placebo, 5% for 600 mg/day, 8% for
900 mg/day), and asthenia (9% for placebo, 14% for 600
mg/day, 19% for 900 mg/day), followed by less frequent
speech disorder, vertigo, tremor, amnesia, and abnormal
gait [80]. Although there were no deaths in the study, 29
patients experienced serious treatment emergent AEs
during the double-blind treatment phase (8 in placebo,
8 in 600 mg/day, 3 in 900 mg/day, and 10 in 1200 mg/day
arms). A total of 79 (20%) patients withdrew from the
study due to TEAEs (17 from 600 mg/day, 19 from 900
mg/day, 31 from 1200 mg/day, and 12 from placebo). The
most common reasons for withdrawal were confusion,
speech disorder, dizziness, and somnolence for the RTG
arms and confusion for the placebo arm. The dropout
rate of 32% in the RTG arms is considered higher than
www.slm-neurology.com
Figure 1. C
omparison of pooled efficacy from clinical studies (intent-to-treat analysis) at their proposed optimal doses,
shown in 50% responder rate over placebo effects
LCM, lacosamide; ESL, eslicarbazepine acetate; RTG,
retigabine.
that of other clinical trials of newer AEDs such as levetiracetam, lamotrigine, oxcarbazepine, topiramate, and
zonisamide. However, 91% of overall dropouts occurred
during the titration phase of RTG, which could have
been due to the aggressive titration schedule.
Among non-CNS events, bladder-related adverse
events (eg, urinary hesitancy) were observed with RTG
in another study [12], primarily with 1200 mg. Bladder
ultrasound revealed a modest increase in mean postvoid
residual volume at the 1200 mg dose but not at lower
doses. These adverse events may reflect inhibition of
bladder contractility and urinary retention secondary to
RTG’s effects on KCNQ channels in the detrusor muscle
of the bladder [12]. Otherwise, there were no clinically
relevant findings in laboratory measurements including
urinalysis, ECG findings, neurologic examinations, or
ophthalmologic examinations related to RTG administration.
7
ENJ 2009; 1: (1). September 2009
European Neurological Journal
Table 3. Comparison and Practical Considerations of the Newest AEDs
Strengths
Lacosamide
Limitations
• Novel mechanism of action
• High incidence of dizziness
• Clean pharmacokinetics
• Required dose adjustment in renally impaired patients
• Rapid onset of action
• Potential PR prolongation on EKG
• Low drug interactions
• Unknown efficacy and safety in children
• Available iv solution
• No interaction with oral contraceptives
• Low incidence of sedation, rash or weight gain
Eslicarbazepine
Acetate
• Convenient QD dosing
• Probably narrow spectrum
• Rapid onset of action
• Old mechanism of action
• Low potential for drug interactions
• Hepatic metabolism
• No required adjustment in renally impaired patients
• Potential interaction with warfarin
• Interaction with oral contraceptives
Retigabine
• Novel mechanism of action
• Inconvenient TID dosing
• Low drug interaction potentials
• Low bioavailability
• No interaction with oral contraceptives
• Reduced absorption with food
• Potentially broad spectrum
• High protein binding
• No significant rash or weight gain
• Potential interaction with LTG, DPH, CBZ
• Possible effect on bladder function
CBZ, carbamazepine; DPH, diphenylhydantoin; EKG, electrocardiogram; LTG.
DISCUSSION
Results from multiple clinical studies have demonstrated that LCM, ESL, and RTG were well tolerated
and effective treatment options in reducing partial onset seizures as an adjunctive therapy. Although some
preclinical studies have indicated that these medications
(especially LCM and RTG) could be effective against
generalized onset seizures, clinical studies in human
are needed to determine whether they are indeed broad
spectrum AEDs. Despite the fact that LCM and RTG
display unique and novel MOAs in seizure treatment,
the question still remains whether the MOA of any
AEDs matters at all in clinical practice. However, when
combination therapy is considered, using AEDs with different MOAs may provide better efficacy and tolerability, and even possibly a synergic (supra-additive) effect.
On the other hand, using AEDs with similar MOAs may
result in simple additive or even antagonistic (infraadditive) effects. More recently, isobolographic analysis has
been used to determine whether the combination of two
medications could be synergic, additive, or antagonistic
to each other. Luszczki et al examined isobolographic
analysis of RTG in order to evaluate the pharmacodynamic interactions with carbamazepine, lamotrigine,
and valproate utilizing the mouse MES model [83]. They
found that the combination of RTG with valproate at
fixed ratios of 1:3, 1:1, and 3:1 exerted synergic interaction, while combinations with carbamazepine and lamotrigine produced additive interaction. A similar study
by Stöhr et al demonstrated the synergic effect of LCM
When a new AED is introduced, many questions are
raised by clinicians: Is it better than existing medications? What is ‘new’ about the new medication compared to the existing AEDs? How quickly does it work?
Does it work for generalized seizures as well as partial
seizures? These questions may ultimately lead to more
difficult but perhaps more important question: Does it
improve overall seizure control and quality of life for patients with epilepsy?
Despite the fact that their efficacy may be similar to
each other (Figure 1), understanding of the differences
in pharmacokinetics, MOA, potential drug-to-drug interactions, and tolerability may provide useful guidance
when choosing a new AED for epilepsy patients. Favored
AEDs should have 100% bioavailability, linear kinetics,
low or no drug-to-drug interactions, low protein binding,
renal clearance, longer half-life, and convenient dosing,
preferably coupled with novel MOAs. Table 2 shows a
comparison of the main properties of these AEDs. Although not all three compounds display favorable properties in this regard, each has some notable advantages
over existing AEDs. For example, LCM has a new MOA
with pretty clean pharmacokinetic properties. ESL offers once daily dosing. RTG also has a unique MOA and
could be the next broad spectrum AED. Table 3 further
describes the strengths and limitations of LCM, ESL,
RTG in clinical practice.
ENJ 2009; 1: (1). September 2009
8
www.slm-neurology.com
Third-generation Antiepileptic Drugs for Partial-onset Seizures: Lacosamide, Retigabine, and Eslicarbazepine Acetate
when combined with levetiracetam or carbamazepine at
fixed ratios of 1:3, 1:1, and 3:1 in a mouse 6 Hz psychomotor seizure model [84]. Nonetheless, it is not yet clear
how these combinations would translate into clinical
practice. To date, there is no clinical study that examines the pharmacodynamic interactions or clinical efficacy of these medications with other existing AEDs.
epilepsy and may provide significant benefit to some patients who remain refractory to other AEDs.
Disclosures: Steve Chung, MD, is a consultant for Medtronics, Inc., GlaxoSmithKline plc. and UCB S.A., is on the speakers’ bureau of Cyberonics, Inc., GlaxoSmithKline plc., and UCB
S.A., and receives grant and research support from Schwarz
Pharma A.G., GlaxoSmithKline plc., UCB S.A., Valeant, Eisai
Inc., Ortho-McNeil and Medtronics, Inc. This article was supported by UCB. However, the views and opinions expressed
herein do not necessarily reflect those of UCB.
In summary, LCM is a novel anticonvulsant with a favorable pharmacokinetic profile that includes absolute
bioavailability, low protein binding, renal excretion, lack
of hepatic enzyme induction or inhibition, low potential for drug-to-drug interactions, and a relatively long
half-life. Efficacy data showed rapid onset of anticonvulsant effects and a significant reduction of partial-onset
seizures at 200 and 400 mg/day even in a severely refractory population. LCM was well tolerated with the
most common adverse event being dizziness, followed
by headache, nausea, and diplopia. LCM was substantially less associated with sedation, cognitive dysfunction, rash, and mood disorders when compared to many
other existing AEDs. Although ESL may not display a
novel mechanism of action, the favorable efficacy and
long-term safety profiles of ESL at 800 mg and 1200 mg
make it a valuable addition to the current treatment of
partial seizures. ESL can be administered through convenient once daily dosing up to 1200 mg. Despite the
fact that ESL is derived from carbamazepine and oxcarbazepine, the incidence of hyponatremia or weight gain
is rare, unlike its parent drug.
REFERENCES
1. Centers for Disease Control and Prevention. Prevalence of
self-reported epilepsy – United States, 1986–1990. JAMA.
1994;272:1893.
2. Kwan P, Brodie M. Early identification of refractory epilepsy. N
Engl J Med. 2000;342:314–319.
3. Perucca E. Development of new antiepileptic drugs: Challenges,
incentives, and recent advances. Lancet Neurol. 2007;6:793–804.
4. Andurkar SV, Stables JP, Kohn H. The anticonvulsant activities of N-benzyl 3-methoxypropionamides. Bioorg Med Chem.
1999;7:2381–2389.
5. Hovinga CA. SPM-927 (Schwarz Pharma). IDrugs. 2003;6:479–
485.
6. Horstmann R, Bonn R, Cawello W, Doty P, Rudd GD. Basic clinical
pharmacological investigations of the new antiepileptic drug SPM
927. Epilepsia. 2002;43:188.
7. Cawello W, Kropeit D, Schiltmeyer B, Hammes W, Horstmann, R.
Food does not affect the pharmacokinetics of SPM 927. Epilepsia.
2004;45:307.
8. Doty P, Rudd GD, Stoehr T, Thomas D. Lacosamide. Neurotherapeutics. 2007;4:145–148.
9. Thomas D, Scharfenecker U, Nickel B, Doty P, Cawello W, Horstmann R. Low potential for drug-drug interaction of lacosamide.
Epilepsia. 2006;47:200.
10. Bialer M, Johannessen SI, Kupferberg HJ, Levy RH, Perucca
E, Tomson T. Progress report on the new antiepileptic drugs: A
summary of the Eighth Eilat Conference (EILAT VIII). Epilepsy
Res. 2007;73:1–52.
11. Kropeit D, Schiltmeyer B, Cawello W, Hammes W, Horstmann R.
Bioequivalence of short-time infusions compared to oral administration of SPM 927. Epilepsia. 2004;45:123.
12. Bialer M, Johannessen S, Levy R, Perucca E, Tomson T, White
S. Progress report on new antiepileptic drugs: A summary of the
Ninth Eilat Conference (EILAT IX). Epilepsy Res. 2009;83:1–43.
13. UCB. Vimpat® (Lacosamide) Tablets and Injection [U.S. prescribing information]. Smyrna, GA: UCB, Inc.; 2008.
14. UCB. Vimpat® Summary of Product Characteristics. Brussels,
Belgium: UCB Pharma S.A.; 2008.
15. Ben-Menachem E. Lacosamide: an investigational drug for
adjunctive treatment of partial-onset seizures. Drugs Today.
2008;44:35–40.
16. Beydoun A, D’Souza J, Hebert D, Doty P. Lacosamide: pharmacology, mechanisms of action and pooled efficacy and safety data in
partial-onset seizures. Expert Rev Neurother. 2009;9:33–42.
17. Beyreuther BK, Freitag J, Heers C, Krebsfanger N, Scharfenecker
U, Stohr T. Lacosamide: a review of preclinical properties. CNS
Drug Rev. 2007;13:21–42.
18. Stöhr T, Kupferberg HJ, Stables JP, et al. Lacosamide, a novel
anti-convulsant drug, shows efficacy with a wide safety margin in
rodent models for epilepsy. Epilepsy Res. 2007;74:147–154.
19. Ben-Menachem E, Biton V, Jatuzis D, Abou-Khalil B, Doty P,
Rudd GD. Efficacy and safety of oral lacosamide as adjunctive therapy in adults with partial-onset seizures. Epilepsia.
2007;48:1308–1317.
20. Halász P, Kälviäinen R, Mazurkiewicz-Beldzinska M, et al.
ESL is not yet approved by the FDA, but recently, clinical trials have begun in northern America. Although
both ESL 800 and 1200 mg are effective dosages, ESL
800 mg was better tolerated in previous clinical trials,
and perhaps provides a better benefit-to-risk ratio than
ESL 1200 mg. RTG is a new anticonvulsant in clinical
development, which activates neuronal M-current by
opening voltage-gated potassium channels. RTG had
demonstrated potent anticonvulsant activity in various
animal models of epileptic seizures including partial and
generalized seizure models as well as the status epilepticus model. Three pivotal clinical studies showed that
retigabine doses of 600 to 1200 mg/day (200 to 400 mg
three times daily) were associated with a significant reduction in seizure frequency when compared with placebo. Retigabine was generally well tolerated and the
most commonly reported adverse events were CNS-related (ie, dizziness and confusion) in clinical trials. RTG
is currently under review for approval by the EMEA and
the FDA.
CONCLUSION
LCM, ESL, and RTG are new generation AEDs with
favorable pharmacokinetics and potential novel MOAs.
Results from clinical studies have demonstrated that all
three AEDs are well tolerated and effective in reducing
partial onset seizures as adjunctive therapy. These new
AEDs expand the treatment options for patients with
www.slm-neurology.com
9
ENJ 2009; 1: (1). September 2009
European Neurological Journal
Adjunctive lacosamide for partial-onset seizures: efficacy and
safety results from a randomized controlled trial. Epilepsia.
2009;50:443–453.
21. Chung S, Sperling M, Biton V, Krauss G, Beaman M, Hebert D.
Lacosamide: efficacy and safety as oral adjunctive treatment for
partial-onset seizures. Epilepsia. 2007;48:321.
22. Chung SS, Rudd R, Hebert D, Doty P. Evaluation of lacosamide
efficacy in subjects with partial-onset seizures across the dose
range used in phase II/III clinical trials. Epilepsia. 2008;49:426.
23. Ben-Menachem E, Chung S, Rudd D, Hebert D, Doty P. Evaluation of lacosamide efficacy in subjects with partial-onset seizures
across the dose range used in phase II/III clinical trials. Poster
presented at: 28th International Epilepsy Congress; June 28 –
July 2, 2009; Budapest, Hungary.
24. Sperling M, Rudd D, Hebert D, Doty P. Early onset of efficacy in
the initial weeks of treatment with lacosamide: a pooled analysis
of three phase II/III trials. Epilepsia. 2008;49:457.
25. US FDA. Vimpat (Lacosamide) Tablets and Injection: US prescribing information [online]. US FDA. http://www.fda.gov/cder/
foi/label/2008/022253lbl.pdf. Accessed October 30, 2008.
26. European Medicines Agency. Lacosamide (Vimpat): Summary
of Product Characteristics [online]. EMEA. http://www.emea.
europa.eu/humandocs/PDFs/EPAR/vimpat/H-863-PI-en.pdf. Accessed November 5, 2008.
27. Kennebäck G, Bergfeldt L, Tomson T. Electrophysiological evaluation of the sodium-channel blocker carbamazepine in healthy
human subjects. Cardiovasc Drugs Ther. 1995;9:709–714.
28. Kennebäck G, Bergfeldt L, Vallin H, Tomson T, Edhag O. Electrophysiologic effects and clinical hazards of carbamazepine treatment for neurologic disorders in patients with abnormalities of
the cardiac conduction system. Am Heart J. 1991;121:1421–1429.
29. Matsuo F, Bergen D, Faught E, et al. Placebo-controlled study
of the efficacy and safety of lamotrigine in patients with partial
seizures. U.S. Lamotrigine Protocol 0.5 Clinical Trial Group.
Neurology. 1993;43:2284–2291.
30. Laville MA, de la Gastine B, Husson B, Le Boisselier R, Mosquet B, Coquerel A. Should we care about pregabalin for elderly
patients with a history of cardiac dysrhythmia? Rev Med Interne.
2008;29:152–154.
31. Biton V, Rosenfeld WE, Whitesides J, Fountain NB, Vaiciene
N, Rudd GD. Intravenous lacosamide as replacement for oral
lacosamide in patients with partial-onset seizures. Epilepsia.
2008;49:418–424.
32. Krauss G, Ben Menachem E, Mameniskiene R, Vaiciene N, Rudd
D, Brock M. Lacosamide: safety and tolerability of adjunctive
intravenous lacosamide as short-term replacement for oral
lacosamide in subjects with partial-onset seizures during 30-, 15-,
and 10-minute infusions. Eur J Neurol. 2007;14:319–320.
33. Nunes T, Maia J, Almeida L. Pharmacokinetic interaction
between eslicarbazepine acetate and lamotrigine in healthy subjects. Epilepsia. 2007;48:39–40.
33. Johannessen LC, Johannessen SI. Modifications of antiepileptic
drugs for improved tolerability and efficacy. Perspect Med Chem.
2008;2:21–39.
34. Almeida L, Soares-da-Silva P. Eslicarbazepine acetate (BIA 2-093).
Neurotherapeutics. 2007;4:88–96.
35. Bialer M. New antiepileptic drugs that are second generation
to existing antiepileptic drugs. Expert Opin Investig Drugs.
2006;15:637–647.
36. Almeida L, Minciu I, Nunes T, et al. Pharmacokinetics, efficacy,
and tolerability of eslicarbazepine acetate in children and adolescents with epilepsy. J Clin Pharmacol. 2008;48:966–977.
37. Falcão A, Maia J, Almeida L, Mazur D, Gellert M, Soares-da-Silva
P. Effect of gender on the pharmacokinetics of eslicarbazepine
acetate (BIA 2-093), a new voltage-gated sodium channel blocker.
Biopharm Drug Dispos. 2007;28:249–256.
38. Maia J, Vaz-da-Silva M, Almeida L, et al. Effect of food on the
pharmacokinetic profile of eslicarbazepine acetate (BIA 2-093).
Drugs R D. 2005;6:201–206.
39. Maia J, Almeida L, Potgieter JH, Soares-da-Silva P. Effect of renal
ENJ 2009; 1: (1). September 2009
impairment on the pharmacokinetics of eslicarbazepine acetate.
Int J Clin Pharmacol Ther. 2008;46:119–130.
40. Almeida L, Potgieter JH, Maia J, Potgieter MA, Mota F, Soaresda-Silva P. Pharmacokinetics of eslicarbazepine acetate in patients with moderate hepatic impairment. Eur J Clin Pharmacol.
2008;64:267–273.
41. Elger C, Halász P, Maia J, Almeida L, Soares-da-Silva P; BIA2093-301 Investigators Study Group. Efficacy and safety of
eslicarbazepine acetate as adjunctive treatment in adults with
refractory partial-onset seizures: a randomized, double-blind,
placebo-controlled, parallel-group phase III study. Epilepsia.
2009;50:454–463.
42. Chien S, Yao C, Mertens A, Verhaeghe T, Solanki B, Doose DR. An
interaction study between the new antiepileptic and CNS drug
carisbamate (RWJ-333369) and lamotrigine and valproic acid.
Epilepsia. 2007;48:1328–1338.
43. Chien S, Bialer M, Solanki B, et al. Pharmacokinetic interaction
study between the new antiepileptic and CNS drug RWJ-333369
and carbamazepine in healthy adults. Epilepsia. 2006;47:1830–
1840.
44. Ambrosio AF, Soares-Da-Silva P, Carvalho CM, Carvalho AP.
Mechanisms of action of carbamazepine and its derivatives,
oxcarbazepine, BIA 2-093, and BIA 2-024. Neurochem Res.
2002;27:121–130.
45. Bonifacio MJ, Sheridan RD, Parada A, Cunha RA, Patmore L,
Soares-da-Silva P. Interaction of the novel anticonvulsant, BIA
2-093, with voltage-gated sodium channels: comparison with
carbamazepine. Epilepsia. 2001;42:600–608.
46. Parada A, Soares-da-Silva P. The novel anticonvulsant BIA 2-093
inhibits transmitter release during opening of voltage-gated
sodium channels: a comparison with carbamazepine and oxcarbazepine. Neurochem Int. 2002;40:435–440.
47. Sierra-Paredes G, Oreiro-García MT, Vázquez-Illanes MD, SierraMarcuño G. Effect of eslicarbazepine acetate (BIA 2-093) on
latrunculin A-induced seizures and extracellular amino acid concentrations in the rat hippocampus. Epilepsy Res. 2007;77:36–43.
48. Elger C, Bialer M, Cramer JA, Maia J, Almeida L, Soaresda-Silva
P. Eslicarbazepine acetate: a double-blind, addon, placebo-controlled exploratory trial in adult patients with partial-onset seizures.
Epilepsia. 2007;48:497–504.
49. Elger C, Halász P, Maia J, Almeida L, Soares-da-Silva P; BIA2093-301 Investigators Study Group. Efficacy and safety of
eslicarbazepine acetate as adjunctive treatment in adults with
refractory partial-onset seizures: a randomized, double-blind,
placebo-controlled, parallel-group phase III study. Epilepsia.
2009;50:454–463.
50. McCormack PL, Robinson DM. Eslicarbazepine acetate. CNS
Drugs. 2009;23:71–79.
51. European Medicines Agency. CHMP Assessment Report for
Zibinix [online]. EMEA. http://www.emea.europa.eu/humandocs/
PDFs/EPAR/zebinix/H-988-en6.pdf. Accessed September 15,
2009.
52. Kapetanovic I, Yonekawa W, Kupferberg H. The effects of
D-23129, a new experimental anticonvulsant drug, on neurotransmitter amino acids in the rat hippocampus in vitro. Epilepsy
Res. 1995;22:167–173.
53. Rundfeldt C, Netzer R. The novel anticonvulsant retigabine activates M-currents in Chinese hamster ovary-cells transfected with
human KCNQ2/3 subunits. Neurosci Lett. 2000;282:73–76.
54. Blackburn-Munro G, Dalby-Brown W, Mirza NR, Mikkelsen JD,
Blackburn-Munro RE. Retigabine: chemical synthesis to clinical
application. CNS Drug Rev. 2005;11:1–20.
55. Wickenden AD, Yu W, Zou A, Jegla T, Wagoner PK. Retigabine, a
novel anticonvulsant, enhances activation of KCNQ2/Q3 potassium channels. Mol Pharmacol. 2000;58:591–600.
56. Ferron GM, Sachdeo R, Partiot A, Fritz T, Althouse S, Troy S.
Pharmacokinetic interaction between valproic acid, topiramate,
phenytoin or carbamazepine and retigabine in epileptic patients.
Clin Pharmacol Ther. 2001;63:18.
57. Hermann R, Ferron GM, Erb K, et al. Effects of age and sex on
10
www.slm-neurology.com
Third-generation Antiepileptic Drugs for Partial-onset Seizures: Lacosamide, Retigabine, and Eslicarbazepine Acetate
the disposition of retigabine. Clin Pharmacol Ther. 2003;73:61–
70.
58. Ferron GM, Paul J, Fruncillo R, et al. Multiple-dose, linear doseproportional pharmacokinetics of retigabine in healthy volunteers. J Clin Pharmacol. 2002;42:175–182.
59. McNeilly PJ, Torchin CD, Anderson LW, Kapetanovic IM, Kupferberg HJ, Strong JM. In vitro glucuronidation of D-23129, a
new anticonvulsant, by human liver microsomes and liver slices.
Xenobiotica. 1997;5:431–441.
60. Hempel R, Schupke H, McNeilly PJ, et al. Metabolism of retigabine (D-23129), a novel anticonvulsant. Drug Metab Dispos.
1999;27:613–622.
61. Bialer M, Johannessen SI, Kupferberg HJ, Levy RH, Loiseau P,
Perucca E. Progress report on new antiepileptic drugs: a summary of the Fourth Eilat Conference (EILAT IV). Epilepsy Res.
1999;34:1–41.
62. Ferron GM, Patat A, Parks V, Rolan P, Troy SM. Lack of pharmacokinetic interaction between retigabine and phenobarbitone at steady-state in healthy subjects. Br J Clin Pharmacol.
2003;56:39–45.
63. Hermann R, Knebel NG, Niebch G, Richards L, Borlak J, Locher
M. Pharmacokinetic interaction between retigabine and lamotrigine in healthy subjects. Eur J Clin Pharmacol. 2003;58:795–802.
64. Paul J, Ferron GM, Richards L, Getsy J, Troy S. Retigabine does
not alter the pharmacokinetics of a low-dose oral contraceptive in
women. Neurology. 200;56:335–336.
65. Tober C, Rostock A, Bartsch R. D-23129: a potent anticonvulsant
in the amygdala kindling model of complex partial seizures. Eur
J Pharmacol. 1996;303:163–169.
66. Rostock A, Tober C, Rundfeldt C, et al. D-23129: a new anticonvulsant with a broad spectrum activity in animal models of
epileptic seizures. Epilepsy Res. 1996;23:211–223.
67. Dailey JW, Cheong JH, Ko KH, Adams-Curtis LE, Jobe PC. Anticonvulsant properties of D-20443 in genetically epilepsy-prone
rats: prediction of clinical response. Neurosci Lett. 1995;195:77–
80.
68. Main MJ, Cryan JE, Dupere JRB, Cox B, Clare JJ, Burbidge SA.
Modulation of KCNQ2/3 potassium channels by the novel anticonvulsant retigabine. Mol Pharmacol. 2000;58:253–262.
69. Wuttke TV, Seebohm G, Bail S, Maljevic S, Lerche H, The new
anticonvulsant retigabine favors voltage-dependent opening of
the Kv7.2 (KCNQ2) channel by binding to its activation gate. Mol
Pharmacol. 2005;67:1009–1017.
70. Lerche C, Scherer CR, Seebohm G, et al. Molecular cloning and
functional expression of KCNQ5, a potassium channel subunit
that may contribute to neuronal M-current diversity. J Biol
Chem. 2000;275:22395–22400.
71. Schroeder BC, Hechenberger M, Weinreich F, Kubisch C, Jentsch
TJ. KCNQ5, a novel potassium channel broadly expressed in
brain, mediates M-type currents. J Biol Chem. 2000;31:24089–
24096.
72. Barhanin J, Lesage F, Guillemare E, Fink M, Lazdunski M,
Romey G. KvLQT1 and 1sK (minK) proteins associate to form
the IKS cardiac potassium current. Nature. 1996;384:78–80.
73. Sanguinetti MC, Curran ME, Zou A, et al. Coassembly of KvLQT1
and minK (IsK) proteins to form cardiac I (Ks) potassium channel. Nature. 1996;384:80–83.
74. Kubisch C, Schroeder BC, Friedrich T, et al. KCNQ4, a novel
potassium channel expressed in the sensory outer hair cells, is
mutated in dominant deafness. Cell. 1999;96:437–446.
75. Kharkovets T, Hardelin J-P, Safieddine S, et al. KCNQ4, a K+
channel mutated in a form of dominant deafness, is expressed in
the inner ear and the central auditory pathway. Proc Natl Acad
Sci USA. 2000;97:4333–4338.
76. Biervert C, Schroeder BC, Kubisch C, et al. A potassium channel
mutation in neonatal human epilepsy. Science. 1998;279:403–406.
77. Charlier C, Singh NA, Ryan SG, et al. A pore mutation in a novel
KQT-like potassium channel gene in an idiopathic epilepsy family.
Nat Genet. 1998;18:53–55.
78. Singh NA, Charlier C, Stauffer D, et al. A novel potassium
www.slm-neurology.com
channel gene, KCNQ2, is mutated in an inherited epilepsy of
newborns. Nat Genet. 1998;18:25–29.
79. Brown DA. M-currents. In: Narahashi T, ed. Ion Channels. Vol. 1.
New York: Plenum; 1988:55–99.
80. Porter RJ, Partiot A, Sachdeo R, Nohria V, Alves WM. Randomized, multicenter, dose-ranging trail of retigabine for partialonset seizures. Neurology. 2007;68:1197–1204.
81. French J, Mansbach H; RESTOR 1 investigators. 1200 mg/day
retigabine as adjunctive therapy in adults with refractory partialonset seizures. Epilepsia. 2008;49:112–113.
82. Brodie M, Mansbach H; RESTOR 2 investigators. Retigabine 600
or 900 mg/day as adjunctive therapy in adults with partial-onset
seizures. Epilepsia. 2008;49:110.
83. Luszczki JJ, Wu JZ, Raszewski G, Czuczwar SJ. Isobolographic
characterization of interactions of retigabine with carbamazepine, lamotrigine, and valproate in the mouse maximal
electroshock-induced seizure model. Naunyn Schmiedebergs Arch
Pharmacol. 2009;379:163–179.
84. Stöhr T, Shandra P, Kashenko O, Shandra A. Synergism of lacosamide and first-generation and novel antiepileptic drugs in the 6
Hz seizure model in mice. Epilepsia. 2007;48:252.
11
ENJ 2009; 1: (1). September 2009
European Neurological Journal
Original Manuscript
The Impact of the Extended Time Window
Seen in the ECASS III Trial on the Guidelines
for Stroke Management in Europe
Keith W. Muir1,2
Affiliations: 1Division of Clinical Neurosciences, University of Glasgow, Glasgow, Scotland, UK; 2Institute of Neurological
Sciences, Southern General Hospital, Glasgow G51 4TF, Scotland, UK
Submission date: 1st September 2009, Revision date: 5th September 2009, Acceptance date: 9th August 2009
A B STRA C T
Pooled analysis from the major randomized controlled trials of intravenous treatment with the recombinant tissue plasminogen activator alteplase for acute ischemic stroke showed a time-dependent benefit. Individual trials used windows
of between 3 and 6 h from symptom onset and individually had only been conclusive for treatment within 3 h. The pooled
analysis indicated that a significant improvement in the proportion of patients making an excellent recovery might extend to 4.5 h. The ECASS III trial recently confirmed benefit in the 3 to 4.5 h time window, with odds of excellent outcome
of around 1.40, in line with predictions. There was no increase in risk of bleeding. Confirmation of safety in practice came
from the Safe Implementation of Thrombolysis in Stroke–International Stroke Thrombolysis Register (SITS-ISTR) analysis of
patients treated between 3 and 4.5 h. In response to these data, European and other guidelines have been updated to
recommend treatment in patients up to 4.5 h after onset of symptoms who otherwise fulfill current European license terms.
A formal change in licensed indications is awaited.
Keywords: stroke, cerebrovascular disease, guidelines, thrombolysis, alteplase, recombinant tissue plasminogen activator,
rt-PA, acute treatment
Correspondence: Professor Keith Muir, Institute of Neurological Sciences, Southern General Hospital, Glasgow G51 4TF,
Scotland, UK. Tel.: +44-141-201-2494; fax: +44-141-201-2510; e-mail: k.muir@clinmed.gla.ac.uk
INTRODUCTION
EXISTING EVIDENCE ON THE
TIME WINDOW FOR THROMBOLYSIS
Evidence of significantly improved chances of neurological recovery with the alteplase form of recombinant
tissue plasminogen activator (rt-PA), delivered intravenously (iv) for thrombolysis within 3 h of stroke onset,
has been with us since the publication of the National
Institute of Neurological Disorders and Stroke (NINDS)
trial in 1995 [1].
The NINDS trial was conducted in two parts: in part
two, significantly better odds of favorable outcome,
defined by complete, or near-complete, recovery on a
combination of four separate scales of neurological impairment, disability and handicap, were seen with iv alteplase delivered within 3 h of symptom onset.
The statistical approach was strengthened by consistency across the different outcome scales. The large
absolute increase in the proportion with favorable outcome translates into a number needed to treat of only 8
to gain one additional independent survivor (probably
even lower at 5 on reanalysis that adjusted for baseline
imbalances in prognostic factors [2]) and perhaps 3 for
one person to improve in disability grade [3].
Treatment was licensed in the United States in the following year, in Canada in 1998, and a conditional license
was granted in Europe in 2003, subject to two further
studies being conducted: first, a register of all treated
cases to establish that alteplase achieved similar safety
to that evident in clinical trials in routine clinical use;
and second, a further randomized controlled trial (RCT)
that would seek supportive evidence of efficacy in an extended time window.
In addition to the main time window of 3 h in the
NINDS trial, the trial design required centers to recruit
an almost equal number of patients within the first 90
min of onset compared to 91 to 180 min, something that
the investigators duly (and uniquely) delivered. Of the
data from 311 patients contributing to the 0 to 90 min
Further evidence favoring an extended time window
for delivery has arisen from both components of these
requirements, and reflected in recent changes to European, US, and local guidelines.
ENJ 2009; 1: (1). September 2009
13
www.slm-neurology.com
European Neurological Journal
Figure 1. P
ooled data analysis of NINDS, ATLANTIS and ECASS I and II trials (shaded gray) showing odds ratios and 95% confidence intervals for favorable outcome in different time windows from onset, adjusted for prognostic confounders, with ECASS III outcome
superimposed (shaded black)
epoch of treatment in a pooled analysis of all major RCTs
of alteplase, 302 were from the NINDS trial.
dow was amended to be 3 to 4.5 h following the evidence
from the pooled NINDS, ATLANTIS and ECASS trials.
Other large RCTs had longer time windows and a
greater median onset-to-treatment time. The European
Cooperative Acute Stroke Study (ECASS) [4], and later
ECASS II [5], allowed treatment up to 6 h after onset
and average time to treatment was 4.5 h. Neither trial
showed a significant result for its primary end point,
although secondary analyses using different definitions
of disability were consistent with benefit, as was a subgroup analysis of patients in the ECASS trial treated
within 3 h of onset [6].
Drug dose, treatment exclusions, and other major aspects were identical to the existing European license for
alteplase. Eight hundred and twenty-one subjects were
recruited from 19 European countries, 730 of whom received treatment according to all aspects of the protocol
and formed a ‘per protocol’ analysis population.
The unadjusted odds ratio of full or nearly full recovery (defined as a modified Rankin Scale score of 0 or 1
at Day 90) was 1.34 (1.02–1.76) in favor of alteplase, and
after adjusting for some baseline imbalances in prognostic markers, the odds ratio was 1.42 (1.02–1.98), identical to what had been observed in the existing pooled
analysis [11]. When the per protocol analysis was undertaken, the unadjusted odds ratio was 1.47 (1.10–1.97)
in favor of alteplase. Using the full four outcome scale
approach taken in the NINDS trial, a secondary end
point in ECASS III, the odds ratio of favorable outcome
was 1.28 (1.00–1.65) for intention-to-treat and 1.39
(1.07–1.80) for the per-protocol populations. The results
therefore confirmed the benefit of iv alteplase in the 3 to
4.5 h window, and narrowed the confidence interval for
the estimated treatment effect that had been seen previously in the pooled analysis.
The Alteplase Thrombolysis for Acute Noninterventional Therapy in Ischemic Stroke (ATLANTIS) trial
[7–9] varied the time window according to the availability of new data through its course, but also recruited
predominantly in the 3 to 5 h interval after symptom
onset (just over 4.5 h in the main part B of the trial), and
again did not find clear evidence of benefit.
When individual patient data from all of these trials
(NINDS, ATLANTIS, ECASS I and II) were combined in
a pooled analysis, evidence of a time-dependent benefit
was present [10], with the modeled odds of full recovery
significantly greater than for placebo out to 4.5 h after
stroke onset (Figure 1 illustrates the adjusted odds ratios for favorable outcome). Thereafter, the confidence
intervals overlapped neutrality. This finding informed
the amended design of the ECASS III trial.
With respect to safety, the risk of symptomatic intracerebral hemorrhage was consistent with what had
previously been seen, being 1.9% using the Safe Implementation of Treatments for Stroke (SITS) definition
(based on clinical deterioration of 4 or more points on
the National Institutes of Health Stroke Scale (NIHSS)
and evidence of a parenchymal hematoma on brain computed tomography (CT) scan that has mass effect independent of any associated infarction). The population
recruited was broadly similar to those in previous trials, but strokes were slightly less severe (median NIHSS
THE ECASS III TRIAL
The ECASS III trial [11] was initiated as one of the
European Medicines Evaluation Agency’s (EMEA) requests linked to the granting of a conditional license for
alteplase, and was initially charged with identifying supportive evidence of efficacy, for which a 3 to 4 h time window after stroke onset was selected. The trial time winENJ 2009; 1: (1). September 2009
14
www.slm-neurology.com
The Impact of the Extended Time Window Seen in the ECASS III Trial on the Guidelines for Stroke Management in Europe
The Echoplanar Imaging Thrombolytic Evaluation
Trial (EPITHET) [15] randomized patients to iv alteplase or placebo in the 3 to 6 h time window after initial
routine CT followed by MRI including DWI and perfusion sequences, in order to test the hypothesis that the
MRI pattern of diffusion-perfusion mismatch identified
potential responders to treatment. However, while EPITHET confirmed some aspects of the hypothesis such
as the increased likelihood of reperfusion with alteplase, and attenuation of infarct growth with reperfusion,
the sample size (101 subjects) was insufficient to prove
whether the mismatch pattern was a predictor of clinical response. Further refinement of MRI definitions for
mismatch on the basis of the EPITHET results should
increase the robustness of future trials. The study itself
was too small to demonstrate the efficacy of alteplase in
the 3 to 6 h window.
score of 9 in the alteplase group and 10 in the placebo
group), in keeping with previous observations that more
severe strokes tend to present to hospital earlier [10].
OBSERVATIONAL DATA –
SITS-INTERNATIONAL STROKE
THROMBOLYSIS REGISTER (SITS-ISTR)
The Safe Implementation of Treatments for Stroke
(originally the Safe Implementation of Thrombolysis
for Stroke, SITS) registry was established to collect
data on all patients treated with alteplase in Europe
and identify whether the safety outcomes were equivalent to those seen in RCTs. The SITS-Monitoring Study
(SITS-MOST) reported data on more than 6000 patients
treated within the terms of the European license (therefore restricted to those under 80 years of age and treated
within 3 h of onset), and found both safety and efficacy
outcomes to be in line with the RCTs [12].
ONGOING CLINICAL TRIALS
All contributors to the SITS registry were also encouraged to document treatments delivered outwith the conditions of the alteplase license, and shortly before the
ECASS III trial findings were reported, analyzed data
from 664 subjects who fulfilled all license criteria except
for late delivery of alteplase in the 3 to 4.5 h time window [13].
The third International Stroke Trial (IST-3) is randomizing patients to alteplase or placebo 0 to 6 h after
stroke onset. While potentially adding information regarding efficacy in the 4.5 to 6 h window population,
the trial also intends to gather evidence in populations
arbitrarily excluded from other major trials to date or
outwith the current license, notably those over 80 years
of age, with prior stroke and history of diabetes, or with
NIHSS score >25.
The SITS-ISTR study found no significant differences
in either safety or efficacy end points for patients in the
3 to 4.5 h window compared to 11865 patients treated
within 3 h, and therefore offered support for the safety
of treatment within the extended time window. Symptomatic intracerebral hemorrhage occurred in 2.2% compared to 1.6% in the <3 h cohort.
The EXTENDS trial is an evolution of EPITHET
where the MRI mismatch hypothesis will be explored in
a randomized comparison of alteplase and placebo in patients treated 4.5 to 9 h after symptom onset who have
had diffusion and perfusion MRI.
SITS-ISTR provides an interesting mirror to ECASS
III, in that the great majority registered within the extended time window received alteplase within 20 min of
the 3 h cut-off, while in ECASS III, 90% of those randomized were treated between 3.5 and 4.5 h. SITS-ISTR
likely represents patients being treated marginally later
than the licensed window because of minor administrative delays or amendment of onset time as more information becomes available, whereas ECASS III patients were
in a later time period when clinical uncertainty persisted.
Alternative thrombolytic agents are being evaluated predominantly in later time windows than those
covered by the alteplase license. The largest volume
of data to date exists for desmoteplase. In three RCTs,
desmoteplase has been administered 3 to 9 h after
stroke onset on the basis of the presence of a diffusionperfusion mismatch on MRI. The results have been inconsistent to date, and a further trial is ongoing in the
4.5 to 9 h window.
GUIDELINES
MAGNETIC RESONANCE
IMAGING (MRI) SELECTION
The current European license for alteplase is based on
treatment within 3 h, pending regulatory review of the
ECASS III data. However, faced with new randomized
controlled trial evidence and supportive clinical observational data from SITS-ISTR, both local and regional
guideline-writing groups have amended their recommendations in advance of any licensing changes.
It has been hypothesized that the presence on MRI of
a larger perfusion deficit than the diffusion-weighted imaging (DWI) lesion corresponds to salvageable tissue (approximating the ischemic penumbra), and that this MRI
mismatch pattern may allow selection of patients with a
longer time window and possibly enhanced safety. Prospective observational case series have reported similar
clinical outcomes when MRI selection was up to 6 h after
symptom onset compared to CT selection within 3 h, but
these were nonrandomized comparisons [14]. Symptomatic intracranial hemorrhage rates were unaffected.
www.slm-neurology.com
Before presentation of the ECASS III results, the European Stroke Organization’s (ESO) 2008 guideline for
stroke and transient ischemic attack (TIA) management
[16] recommended iv alteplase within 3 h of stroke onset. The revision of these guidelines at the Karolinska
15
ENJ 2009; 1: (1). September 2009
European Neurological Journal
Table 1. Guideline Statements Following ECASS III Results
Guideline
Date
Recommendation statement
ESO [16]
2009
Intravenous rt-PA (0.9 mg/kg body weight, maximum 90 mg), with 10% of the dose given as a bolus
followed by a 60 min infusion, is recommended within 4.5 h of onset of ischemic stroke (Class I,
Level A), although treatment between 3 and 4.5 h is currently not included in the European labeling
(modified January 2009 http://www.eso-stroke.org/pdf/ESO%20Guidelines_update_Jan_2009.pdf)
AHA/ASA [17]
2009
rt-PA should be administered to eligible patients who can be treated in the time period of 3 to 4.5 h
after stroke (Class I Recommendation, Level of Evidence B).
The eligibility criteria for treatment in this time period are similar to those for persons treated at
earlier time periods, with any one of the following additional exclusion criteria: Patients older than
80 years, those taking oral anticoagulants with an international normalized ratio >1.7, those with a
baseline National Institutes of Health Stroke Scale score >25, or those with both a history of stroke
and diabetes
SIGN [23]
2008
• P
atients admitted with stroke within four and a half hours of definite onset of symptoms,
who are considered suitable, should be treated with 0.9 mg/kg (up to maximum 90 mg)
intravenous alteplase (rt-PA)
• Systems should be optimized to allow the earliest possible delivery of iv alteplase (rt-PA)
within the defined time window
benefit from the extension will depend greatly on local
circumstances. Severe strokes present earlier to hospital [10], and are more likely to come directly to hospital than to seek advice from primary care services [20].
Large delays in hospital pathways for stroke care were a
major factor in hospitals before alteplase licensing [21],
and can be substantially reduced by improved in-hospital organization [22].
Stroke Update meeting in November 2008 proposed an
amendment to reflect a longer time window, while noting that the license does not yet cover this extended time
window (Table 1).
Local guideline bodies also exist across Europe, and at
least one of these has so far been able to respond to the
new data. The Scottish Intercollegiate Guidelines Network’s (SIGN) updated recommendations on stroke published in December 2008 were able to consider ECASS
III and also recommended treatment up to 4.5 h. The
SIGN advice also emphasized the importance of early
treatment delivery (Table 1) with a recommendation
concerning organization of systems for delivery in addition to advice on individual practice.
The impact of the extended time window is likely to be
most immediately felt in services where significant time
delays remain the major barrier to treatment. For large
hospitals serving urban populations, where the service
is already well-configured for treatment under 3 h, the
additional numbers may be modest (local estimates suggest that only 10% of referrals fall in the 3 to 4.5 h window). Areas with rural populations may also find significant increases in the proportion of eligible patients with
the extended time window.
The American Heart Association/American Stroke Association recently issued an advisory statement [17] updating their 2007 guidelines on alteplase use [18], and
again this was based on ECASS III and SITS-ISTR. This
recommended treatment in eligible patients who can be
treated in the 3 to 4.5 h time window. The US guidelines
for treatment within 3 h of onset differ from the European license for alteplase in not placing restrictions
on the basis of age, advising only ‘caution’ in the use
of alteplase in severe strokes, and permitting treatment
in patients with prior stroke and diabetes mellitus, and
also in those on oral anticoagulants provided their International Normalized Ratio (INR) is <1.7. Accordingly,
the new advisory statement includes specific detail of
those features of ECASS III exclusion criteria that differ from the previous <3 h advice. It also includes advice
to avoid delays in treatment initiation.
A potential negative impact may arise if the extended
time window is taken to mean that healthcare systems
can respond at a more leisurely pace, and the numbers
of patients being treated within 3 h decline. The onus
remains on stroke physicians to emphasize the timedependent benefit of alteplase and continue to work toward optimizing systems for delivery of treatment at the
earliest possible time after symptom onset. Both US and
some local guidelines emphasize the time factor.
Perhaps the greatest advantage derives from the confirmation of significant benefit of iv thrombolysis in a
further clinical trial. Stroke physicians accustomed to
using alteplase have perhaps forgotten that ECASS III
is only the second large RCT of iv thrombolysis to report
a positive effect of treatment on its primary end point,
all other indications of benefit having derived from secondary analyses, meta-analysis or data pooling. That
there remained a sizeable body of physicians (or service
designers) prior to ECASS III who were insufficiently
convinced of efficacy to adjust their systems to deliver
POTENTIAL IMPACT – SPECULATION
While presentation outwith the present licensed time
window is the most common reason for patients to be
ineligible for thrombolysis [19], the small extension offered post-ECASS III presents only a narrow opening.
The total numbers of additional patients who might
ENJ 2009; 1: (1). September 2009
16
www.slm-neurology.com
The Impact of the Extended Time Window Seen in the ECASS III Trial on the Guidelines for Stroke Management in Europe
thrombolytic treatment is likely to represent one major reason for low treatment rates in many European
countries and in the USA. The unanimity of guidelines
since ECASS III now sends a powerful and consistent
message that iv thrombolytic treatment is an integral
component of stroke care, and not only in the 3 to 4.5
h window. Emphasis on stroke as a medical emergency
will hopefully increase the numbers of patients treated,
and allow systems to develop to enhance the earliest
possible delivery of treatment.
trial: results for patients treated within 3 hours of stroke onset.
Alteplase Thrombolysis for Acute Noninterventional Therapy in
Ischemic Stroke. Stroke. 2002;33:493–495.
8. Clark WM, Albers GW, ATLANTIS Stroke Study Investigators.
The ATLANTIS rt-PA (alteplase) acute stroke trial: final results.
Stroke. 1999;30:234.
9. Clark WM, Wissman S, Albers GW, Jhamandas JH, Madden KP,
Hamilton S. Recombinant tissue-type plasminogen activator
(Alteplase) for ischemic stroke 3 to 5 hours after symptom onset.
The ATLANTIS Study: a randomized controlled trial. Alteplase
Thrombolysis for Acute Noninterventional Therapy in Ischemic
Stroke. JAMA. 1999;282:2019–2026.
10. Hacke W, Donnan G, Fieschi C, et al; ATLANTIS Trials Investigators; ECASS Trials Investigators; NINDS rt-PA Study Group
Investigators. Association of outcome with early stroke treatment: pooled analysis of ATLANTIS, ECASS, and NINDS rt-PA
stroke trials. Lancet. 2004;363:768–774.
11. Hacke W, Kaste M, Bluhmki E, et al. Thrombolysis with alteplase 3 to 4.5 hours after acute ischemic stroke. N Engl J Med.
2008;359:1317–1329.
12. Wahlgren N, Ahmed N, Davalos A, et al. Thrombolysis with
alteplase for acute ischaemic stroke in the Safe Implementation
of Thrombolysis in Stroke-Monitoring Study (SITS-MOST): an
observational study. Lancet. 2007;369:275–282.
13. Wahlgren N, Ahmed N, Dávalos A, et al. Thrombolysis with
alteplase 3-4.5 h after acute ischaemic stroke (SITS-ISTR): an
observational study. Lancet. 2008;372(9646):1303–1309.
14. Schellinger PD, Thomalla G, Fiehler J, et al; SITS Investigators.
MRI-based and CT-based thrombolytic therapy in acute stroke
within and beyond established time windows: an analysis of 1210
patients. Stroke. 2007;38:2640–2645.
15. Davis SM, Donnan GA, Parsons MW, et al. Effects of alteplase
beyond 3 h after stroke in the Echoplanar Imaging Thrombolytic
Evaluation Trial (EPITHET): a placebo-controlled randomised
trial. Lancet Neurol. 2008;7:299–309.
16. European Stroke Organisation (ESO) Executive Committee; ESO
Writing Committee. Guidelines for management of ischaemic
stroke and transient ischaemic attack 2008. Cerebrovasc Dis.
2008;25:457–507.
17. del Zoppo GJ, Saver JL, Jauch EC, Adams HP Jr. Expansion of
the time window for treatment of acute ischemic stroke with
intravenous tissue plasminogen activator: a science advisory from
the American Heart Association/American Stroke Association.
Stroke. 2009;40:2945–2948.
18. Adams HP Jr, del Zoppo G, Alberts MJ, et al. Guidelines for the
early management of adults with ischemic stroke. A guideline
from the American Heart Association/American Stroke Association Stroke Council, Clinical Cardiology Council, Cardiovascular
Radiology and Intervention Council, and the Atherosclerotic
Peripheral Vascular Disease and Quality of Care Outcomes in Research Interdisciplinary Working Groups: the American Academy
of Neurology affirms the value of this guideline as an educational
tool for neurologists. Stroke. 2007;38:1655–1711.
19. Barber PA, Zhang J, Demchuk AM, Hill MD, Buchan AM. Why
are stroke patients excluded from TPA therapy? An analysis of
patient eligibility. Neurology. 2001;56:1015–1020.
20. McCormick MT, Reeves I, Baird T, Bone I, Muir KW. Implementation of a stroke thrombolysis service within a tertiary neurosciences centre in the United Kingdom. QJM. 2008;101:291–298.
21. Harraf F, Sharma AK, Brown MM, Lees KR, Vass RI, Kalra L. A
multicentre observational study of presentation and early assessment of acute stroke. BMJ. 2002;325:17.
22. Hill MD, Barber PA, Demchuk AM, et al. Building a “brain attack” team to administer thrombolytic therapy for acute ischemic
stroke. CMAJ. 2000;162:1589–1593.
23. Scottish Intercollegiate Guidelines Network. Management of
Patients with Stroke or TIA: Assessment, Investigation, Immediate Management and Secondary Prevention. Guideline No. 108.
Edinburgh, UK: Scottish Intercollegiate Guidelines Network;
2008.
CONCLUSIONS
Guidelines have changed to reflect the publication of
new evidence from RCTs that alteplase remains effective as a treatment for selected patients with ischemic
stroke 3 to 4.5 h after symptom onset, although the
status of the alteplase license has not moved as quickly. Guidelines updated since the publication of ECASS
III are consistent in their recommendations in favor
of treatment. It remains to be seen whether licensing
authorities will greet the new evidence from ECASS III
with the same enthusiasm, and there currently remains
some potential for conflict between guideline advice and
the more restrictive license.
Further evidence for patient groups currently not
within the license, particularly the over 80s, and those
with very severe strokes, is likely to be forthcoming in
the medium term, and other ongoing or imminent trials
will address both further extensions to the time window
and refinements of selection criteria such as the use of
imaging.
Disclosure: The author has received support from Boehringer-Ingelheim for travel to meetings. Boehringer-Ingelheim
manufacture alteplase.
REFERENCES
1. The National Institute of Neurological Disorders and Stroke rtPA Stroke Study Group. Tissue plasminogen activator for acute
ischemic stroke. N Engl J Med. 1995;333:1581–1587.
2. Ingall TJ, O’Fallon WM, Asplund K, et al. Findings from the
reanalysis of the NINDS tissue plasminogen activator for acute
ischemic stroke treatment trial. Stroke. 2004;35:2418–2424.
3. Saver JL. Number needed to treat estimates incorporating effects over the entire range of clinical outcomes: novel derivation
method and application to thrombolytic therapy for acute stroke.
Arch Neurol. 2004;61:1066–1070.
4. Hacke W, Kaste M, Fieschi C, et al. Intravenous thrombolysis
with recombinant tissue plasminogen activator for acute hemispheric stroke: The European Cooperative Acute Stroke Study
(ECASS). JAMA. 1995;274:1017–1025.
5. Hacke W, Kaste M, Fieschi C, et al. Randomised double-blind
placebo-controlled trial of thrombolytic therapy with intravenous alteplase in acute ischaemic stroke (ECASS II). Second
European-Australasian Acute Stroke Study Investigators. Lancet.
1998;352:1245–1251.
6. Steiner T, Bluhmki E, Kaste M, et al. The ECASS 3-hour cohort.
Secondary analysis of ECASS data by time stratification. ECASS
Study Group. European Cooperative Acute Stroke Study. Cerebrovasc Dis. 1998;8:198–203.
7. Albers GW, Clark WM, Madden KP, Hamilton SA. ATLANTIS
www.slm-neurology.com
17
ENJ 2009; 1: (1). September 2009
European Neurological Journal
review article
Incidence and Lifetime Risk of Parkinson’s Disease in
Advanced Age: Review and Estimates from the United States
Jane A. Driver1 and Tobias Kurth1—5
Affiliations: 1Divisions of Aging and 2Preventive Medicine, Department of Medicine, Brigham and Women’s Hospital, Harvard
Medical School, Boston, MA, USA; 3Department of Epidemiology, Harvard School of Public Health, Boston, MA, USA; 4INSERM
Unit 708—Neuroepidemiology, Paris, France; 5Faculty of Medicine, Pierre et Marie Curie University, Paris, France
Submission date: 27th June 2009, Revision date: 15th July 2009, Acceptance date: 5th August 2009
A B STRA C T
Age is the strongest known risk factor for Parkinson’s disease (PD). However, its incidence after the age of 80 years is controversial. We examined existing data on the incidence of PD in advanced age, with a focus on studies from the United
States. The evidence suggests that PD incidence in men continues to increase in an age-dependent fashion after age 80
years, at least until age 90 years. Data in women are insufficient to draw any conclusions.
Lifetime risk is a helpful way of summarizing the absolute risk of ever developing a disease during one’s remaining lifetime.
It reflects the true risk of disease in an elderly population because it adjusts for competing risks of death. We discuss the utility of the lifetime risk statistic and present data from the Physicians’ Health Study. In this cohort of initially healthy and longlived men, the lifetime risk of PD at age 45 years was 1 in 15, higher than the risk of lung, colorectal, and bladder cancer.
While the remaining lifetime risk of PD declined slightly with increasing age, it remained substantial at age 80 years, with a
risk of 1 in 21. As life expectancy continues to increase worldwide, the burden of PD will grow dramatically. More estimates
of the lifetime risk of PD in general populations are needed.
Keywords: Parkinson’s disease, incidence, lifetime risk, cohort study, review
Correspondence: Jane A. Driver, Division of Aging, Brigham and Women’s Hospital, 1620 Tremont Street, Boston, MA 02120,
USA. Tel: +1-617-525-7946; fax: +1-617-525-7739; e-mail: jdriver@partners.org
INTRODUCTION
competing risk of death, as has been elegantly shown for
Alzheimer’s disease and stroke by Seshadri et al [3, 4].
Of all known risk factors for Parkinson’s disease (PD),
age is by far the most potent. The worldwide prevalence
of PD will grow by more than 100% over the next 20
years as a factor of increasing life expectancy [1]. There
is good evidence that PD incidence increases exponentially between the ages of 55 and 79 years; whether it
continues this trajectory after age 80 years remains
unclear [2]. The answer to this question has enormous
implications for predicting future disease burden, and
might provide valuable insights into the pathophysiology of PD.
The relevant question for an older individual is “At my
age, what is the chance I will get PD before dying of
something else?” The answer is provided by the lifetime
risk statistic, which controls the risk of disease over
one’s remaining lifetime for competing risks of death.
Lifetime risk presents a simple and powerful means of
expressing the impact of a disease. The statistic that
“one in nine” women will develop breast cancer was
used very effectively in the United States to create public awareness and bolster research funding [5]. Despite
its utility for both patients and population scientists,
lifetime risk is underutilized in neuroepidemiology [6].
The goals of this article are to review the current literature on the incidence of PD in advanced age, discuss the
concept of lifetime risk, and present data from the Physicians’ Health Study (PHS), a large prospective cohort
of men with exceptional longevity.
Measuring and interpreting the risk of PD in advanced
age is challenging for a number of reasons. Most importantly, the majority of available studies of PD incidence
lack substantial follow-up in participants over the age
of 80 years, making estimates in the oldest patients less
stable. A decline in PD incidence in the very elderly
might reflect changes in medical surveillance or exposure to risk factors rather than a true change in risk.
Finally, methods routinely used to measure the longterm probability of disease in younger populations overestimate risk in the elderly by failing to control for the
ENJ 2009; 1: (1). September 2009
REVIEW OF INCIDENCE STUDIES
Estimates of PD incidence vary dramatically based on
the age distribution of the population. Our purpose was
19
www.slm-neurology.com
European Neurological Journal
not to conduct a systematic review of PD incidence. We
selected high-quality population-based studies that use
widely accepted diagnostic criteria for PD and present
incidence data in people 60 or 65 years and older, measured in person–years (yrs) [7–14]. This allowed us to
recalculate estimates in order to compare studies in a
meaningful way. The crude age-specific incidence rates
of these studies are displayed in Figure 1A. The average annual incidence for men in this age range was 214.4
cases/100 000 person–yrs. Six studies show PD incidence
continuing to increase in the oldest age category (80 or
85+), while at least three show a decline after age 75
or 79 years [11, 13, 15]. Studies that performed population screening for PD followed by in-person examination
reported incidence rates of roughly 1000 cases/100 000
person–yrs in men aged 85 years or older, while those
relying on administrative databases and/or medical
records reported substantially lower rates (95–343 cases/100 000 person–yrs). In the PHS, in which new cases
of PD were self-reported by the physician participants,
men aged 85 years and older had an incidence of 570
cases/100 000 person–yrs, an estimate that falls between
studies of direct ascertainment and medical records.
This wide range of estimates illustrates the importance
of case ascertainment in incidence studies. Communitybased studies that perform direct population screening
and in-person examinations likely offer the most accurate data, and suggest that PD is much more common in
advanced age than previously suspected.
Figure 1. A
ge-specific incidence rates of PD in men (A) and
women (B) aged 60 years and older. Rates from refs [11]
and [12] were recalculated to include all races. The oldest group in ref. 14 is 80+ years.
over time. Long-term estimates of absolute risk may be
more meaningful to the lay public than short-term or
relative risks, and are invaluable in the prediction of
population disease burden [17].
As most previous studies have combined all patients
aged 85 years and older into one group, it is difficult to
know whether PD incidence continues to increase indefinitely with age. Owing to the long follow-up and exceptional longevity of the PHS cohort, we were able to
calculate incidence rates in men through age 100 years
(Table 1). Our data provide good evidence that PD incidence in men continues to increase at least to age 90
years.
The most commonly used means of estimating longterm disease risk is cumulative incidence. This method
is problematic in elderly populations because it makes
the assumption that those who die have the same risk as
survivors. Because those who are dead have no possibility of developing disease, cumulative incidence overestimates risk in a population with a substantial mortality
rate [18]. The appropriate measure of long-term-risk in
this setting is lifetime risk. By adjusting cumulative incidence for the risk of death from other causes, lifetime
risk summarizes the absolute risk of developing a disease during the rest of one’s life. Figure 2 compares
the cumulative incidence and lifetime risk curves of PD
in the PHS cohort. The curves are identical until age
75 years, then begin to diverge when death from other
causes becomes common. The remaining lifetime risk of
PD for a 45-year-old man in our cohort is 6.7% (1 in 15).
Cumulative incidence overestimates this risk (9.9% or 1
in 10) by failing to adjust for mortality.
As expected, incidence rates in women (Figure 1B)
were substantially lower than in men, ranging from 69
to 419 cases/100 000 in patients aged 80 or 85 years and
older. In contrast to the data in men, there was no predominant pattern of PD incidence in elderly women. A
substantial increase in the oldest group was seen in only
one study [7], whereas in three, there was a slight increase or plateau [9, 11, 12], and in three a decline [8, 12,
14]. As PD is a much rarer disease in women, more data
from large population studies with a long follow-up are
needed to better define its incidence after age 80 years.
Elbaz et al [11] reported a lifetime risk for PD of 2% (1
in 50) for men and 1.3% (1 in 77) for women in Olmstead
County, Minnesota (USA). Lifetime risk estimates from
the PHS are substantially higher, in part due to the longevity of our health-conscious participants. Longer life
expectancy means a longer period at risk of developing
PD. In addition, the PHS has substantially fewer smokers than a general population, and the lifetime risk of PD
is profoundly influenced by smoking status (Figure 3).
ESTIMATING PD RISK IN ADVANCED AGE
An incidence rate is calculated by dividing the number
of people who develop a disease by the sum of the time
contribution for all people followed [16]. Thus, it accounts for important factors in elderly populations such
as competing risks of death and loss to follow-up. However, this statistic does not provide information on risk
ENJ 2009; 1: (1). September 2009
20
www.slm-neurology.com
Incidence and Lifetime Risk of Parkinson’s Disease in Advanced Age: Review and Estimates from the United States
Table 1. Age-specific and Overall Annual Incidence Rates of PD in the Physicians’ Health Study per 100 000 person–years
Age group (years)
No. of PD cases
Person–years
Incidence rate
95% CI
40–44
0
12 553.0
0.0
0.0
45–49
2
34 324.0
5.83
1.0–19.3
50–54
9
54 448.5
16.53
8.1–30.3
55–59
28
72 172.0
38.80
26.3–55.3
60–64
57
84 889.0
67.15
51.3–86.4
65–69
101
75 356.5
134.03
109.7–162.2
70–74
111
58 443.0
189.93
157.0–227.8
75–79
101
39 660.0
254.67
208.5–308.1
80–84
80
22 512.5
355.36
283.6–439.9
85–89
59
9 597.5
614.74
472.2–784.4
90–99
15
3 361.0
446.30
259.3–719.6
Total
563
467 316.5
120.48
110.8–130.7
65 and older
467
208 932
223.52
203.9–244.5
Adapted from Driver JA, et al. Neurology. 2009;72:432–438 with permission.
The cumulative incidence of PD from age 45 years declined from 10.8% in never smokers to 3.5% in heavy
smokers, illustrating the well-known “protective” association between smoking and PD [19]. A similar pattern
was seen for lifetime risk, which decreased from 7.8%
to 2.3% with increasing smoking exposure. In this case,
the decreased risk was due to smoking-related death,
primarily from cardiovascular disease, pulmonary disease, and cancer. In a cohort with a higher prevalence of
smoking, the incidence and lifetime risks of PD would
be even lower.
Figure 2. C
umulative incidence (CI) vs lifetime risk (LTR) of PD in
the Physicians’ Health Study Cohort. For a 45-year-old
man who was free of PD at age 45 years, the CI predicted a risk of 9.9% for developing PD from age 45 years,
whereas mortality-adjusted LTR was only 6.7%. Adapted
from Driver JA, et al. Neurology. 2009;72:432–438 with
permission.
Because lifetime risks reflect the mortality rate of a
given population, they cannot be compared easily across
studies. However, they are very useful in comparing the
risks of disease within the same population. In the PHS,
a healthy cohort with high rates of disease prevention,
screening, and medical surveillance, the lifetime risk of
PD for a 45-year-old man was 1 in 15. This was higher
than that of all major cancer types excluding prostate
cancer (Table 2).
IS PD AN AGE-DEPENDENT DISEASE?
Over 20 years ago, Brody and Schneider [20] defined
two classes of age-associated diseases that continue to
serve as a model for understanding of the epidemiology
and pathophysiology of chronic illness. “Age-dependent”
diseases, such as congestive heart failure, are closely
linked to the normal aging of the host, and their incidence increases indefinitely, whereas “age-related” diseases, such as multiple sclerosis, are associated with a
particular age range and then decline. The remaining
lifetime risk of congestive heart failure remains stable
in the face of the increasing mortality rate in later life
[21]. In contrast, the risk of many cancers is outpaced
by the risk of death, leading to a substantial decline in
remaining lifetime risk at older ages [22].
www.slm-neurology.com
Figure 3. C
umulative incidence and lifetime risk of PD in the
Physician’s Health Study by baseline smoking status. The
cumulative incidence decreased as smoking exposure
increased, suggesting that smoking protects against
PD. Lifetime risk also decreased with increasing smoking exposure, likely due to increased mortality among
smokers. Adapted from Driver JA, et al. Neurology.
2009;72:432–438 with permission.
21
ENJ 2009; 1: (1). September 2009
European Neurological Journal
Table 2. C
omparison of Lifetime Risks of Various Conditions
in the Physicians’ Health Study
PD and Alzheimer’s disease have always been considered age dependent, as their mortality rate increases exponentially with age [20]. In addition, key pathological
features of these conditions, such as Lewy bodies and
neurofibrillary tangles, can also be seen in normal aging. PD is characterized by loss of dopaminergic neurons
in the substantia nigra. This tissue seems particularly
vulnerable to age-related changes, and loss of these cells
can be seen in older individuals without disease [23]. An
age-dependent decline in dopamine levels and receptors
has also been documented [24]. Thus, PD seems to be intimately related to the process of brain aging. Our finding that PD incidence increases in an exponential fashion at least until age 90 years suggests an age-dependent
disease pattern. However, incidence declines after age
90 years, and remaining lifetime risk declines modestly
by age 80 years. This decline must be interpreted with
caution, as estimates in men aged 90 years and older
are based on only 15 cases of PD and the confidence intervals are quite wide. A decline might simply reflect a
decrease in diagnosis or the difficulty of distinguishing
idiopathic PD from other forms of parkinsonism and
other comorbidities [25]. In this case, we would expect
carefully conducted future studies of PD in elderly populations to show that its incidence continues to increase
indefinitely. However, if future work finds that risk truly declines after the age of 90 years, this would suggest
that survivors to very old ages have increased resistance
to PD. Understanding the factors that promote such resistance might uncover new avenues for treatment and
prevention, even for PD occurring at earlier ages. As our
populations continue to age, there will be increasing opportunities to measure the incidence and lifetime risk of
PD in the ninth and tenth decades in order to address
this intriguing question.
Condition
Prostate cancer
Lifetime risk
20.9 (20.00–21.72)
1/5
15.6 (14.6–16.5)
1/6
Stroke
Myocardial infarction
14.9 (14.07–15.73)
1/7
Parkinson’s disease
6.72 (6.01–7.43)
1/15
Colorectal cancer
5.70 (5.12–6.28)
1/18
Lymphoma
3.55 (3.09–4.02)
1/28
Lung cancer
3.53 (3.05–4.01)
1/28
Bladder cancer
1.93 (1.56–2.30)
1/52
years, at least until age 90 years. More studies of the
incidence of PD in women of advanced age are needed,
as current data are insufficient to draw any conclusions.
In the PHS, a cohort of relatively healthy and long-lived
men, the lifetime risk of PD for a 45-year-old man was 1
in 15, higher than the lifetime risks of most major cancers. As effective treatment and prevention continue to
decrease mortality rates from heart disease, stroke, and
cancer in developed nations, the lifetime risk of PD will
increase substantially unless preventive strategies are
found. Additional information on the lifetime risk of PD
in general populations is needed to help with patient
education and public health planning. More data are
also needed to determine whether the incidence of PD
continues to increase indefinitely with age, or whether
those with exceptional longevity find a means of “escaping” it.
Funding: This research is funded by a grant from the Parkinson’s Disease Foundation. The Physicians’ Health Study
is supported by grants CA-34944, CA-40360, and CA-097193
from the National Cancer Institute and grants HL-26490 and
HL-34595 from the National Heart, Lung, and Blood Institute,
Bethesda, MD, USA.
SUMMARY
The weight of existing evidence suggests that the incidence of PD in men continues to increase after age 80
Table 3. L ifetime Risk Estimates for the Development of Parkinson’s Disease in Men in the Physicians’ Health Study and Olmstead County,
Minnesota, and Congestive Heart Failure in Men in the Framingham Heart Study
Index age*
Lifetime risk PD
PHS [10]
Lifetime risk PD Minnesota [11]
Lifetime risk CHF Framingham [21]
Men only
Men
Women
Men
Women
40
6.7
(1 in 15)
2.1
(1 in 50)
1.3
(1 in 77)
21.0
(1 in 5)
20.3
(1 in 5)
50
6.7
(1 in 15)
2.1
(1 in 50)
1.3
(1 in 77)
20.9
(1 in 5)
20.5
(1 in 5)
60
6.6
(1 in 15)
2.0
(1 in 50)
1.3
(1 in 77)
20.5
(1 in 5)
20.5
(1 in 5)
70
6.1
(1 in 17)
1.6
(1 in 63)
1.1
(1 in 100)
20.6
(1 in 5)
20.2
(1 in 5)
80
4.8
(1 in 21)
0.7
(1 in 143)
0.7
(1 in 143)
20.2
(1 in 5)
19.3
(1 in 5)
*Age reached free of PD.
ENJ 2009; 1: (1). September 2009
22
www.slm-neurology.com
Incidence and Lifetime Risk of Parkinson’s Disease in Advanced Age: Review and Estimates from the United States
Disclosures: Dr. Driver has received research grants from the
Parkinson’s Disease Foundation, the Eleanor and Miles Shore/
Harvard Medical School Scholars in Medicine Program, and
the Hartford Foundation’s Center of Excellence in Geriatric
Medicine at Harvard Medical School. Dr. Kurth has received
investigator-initiated research funding as Principal or CoInvestigator from the National Institutes of Health, McNeil
Consumer & Specialty Pharmaceuticals, Merck, and Wyeth
Consumer Healthcare; he is a consultant to i3 Drug Safety
and World Health Information Science Consultants, LLC, and
received honoraria from Genzyme, Merck, and Pfizer for educational lectures.
16. Rothman KJ. Epidemiology: an introduction. Oxford: Oxford
University Press, 2002.
17. Feuer EJ, Wun LM, Boring CC, Flanders WD, Timmel MJ, Tong
T. The lifetime risk of developing breast cancer. J Natl Cancer
Inst. 1993;85:892–897.
18. Schouten LJ, Straatman H, Kiemeney LA, Verbeek AL. Cancer
incidence: life table risk versus cumulative risk. J Epidemiol
Community Health. 1994;48:596–600.
19. Hernan MA, Takkouche B, Caamano-Isorna F, Gestal-Otero JJ. A
meta-analysis of coffee drinking, cigarette smoking, and the risk
of Parkinson’s disease. Ann Neurol. 2002;52:276–284.
20. Brody JA, Schneider EL. Diseases and disorders of aging: a hypothesis. J Chronic Dis. 1986;39:871–876.
21. Lloyd-Jones DM, Larson MG, Leip EP, et al. Lifetime risk for developing congestive heart failure: the Framingham Heart Study.
Circulation. 2002;106:3068–3072.
22. Driver JA, Djousse L, Logroscino G, Gaziano JM, Kurth T.
Incidence of cardiovascular disease and cancer in advanced age:
prospective cohort study. BMJ. 2008;337:a2467.
23. McGeer PL, McGeer EG, Suzuki JS. Aging and extrapyramidal
function. Arch Neurol. 1977;34:33–35.
24. Rinne JO. Muscarinic and dopaminergic receptors in the aging
human brain. Brain Res. 1987;404:162–168.
25. Bower JH, Maraganore DM, McDonnell SK, Rocca WA. Influence
of strict, intermediate, and broad diagnostic criteria on the ageand sex-specific incidence of Parkinson’s disease. Mov Disord.
2000;15:819–825.
ACKNOWLEDGMENTS
We are grateful to the staff of the Physicians’ Health
Study and to the 22 071 dedicated physicians who have
made this project possible.
REFERENCES
1. Dorsey ER, Constantinescu R, Thompson JP, et al. Projected
number of people with Parkinson disease in the most populous
nations, 2005 through 2030. Neurology. 2007;68:384–386.
2. Hirtz D, Thurman DJ, Gwinn-Hardy K, Mohamed M, Chaudhuri
AR, Zalutsky R. How common are the “common” neurologic
disorders? Neurology. 2007;68:326–337.
3. Seshadri S, Beiser A, Kelly-Hayes M, et al. The lifetime risk
of stroke: estimates from the Framingham Study. Stroke.
2006;37:345–350.
4. Seshadri S, Wolf PA, Beiser A, et al. Lifetime risk of dementia and
Alzheimer’s disease. The impact of mortality on risk estimates in
the Framingham Study. Neurology. 1997;49:1498–1504.
5. One in Nine. Washington Post. March 17, 1992;Sect. A:16.
6. Seshadri S, Wolf PA. Lifetime risk of stroke and dementia: current concepts, and estimates from the Framingham Study. Lancet
Neurol. 2007;6:1106–1114.
7. Baldereschi M, Di Carlo A, Rocca WA, et al. Parkinson’s disease
and parkinsonism in a longitudinal study: two-fold higher incidence in men. ILSA Working Group. Italian Longitudinal Study
on Aging. Neurology. 2000;55:1358–1363.
8. Benito-Leon J, Bermejo-Pareja F, Morales-Gonzalez JM, et al.
Incidence of Parkinson disease and parkinsonism in three elderly
populations of central Spain. Neurology. 2004;62:734–741.
9. de Lau LM, Giesbergen PC, de Rijk MC, Hofman A, Koudstaal
PJ, Breteler MM. Incidence of parkinsonism and Parkinson
disease in a general population: the Rotterdam Study. Neurology.
2004;63:1240–1244.
10. Driver JA, Logroscino G, Gaziano JM, Kurth T. Incidence and
remaining lifetime risk of Parkinson disease in advanced age.
Neurology. 2009;72:432–438.
11. Elbaz A, Bower JH, Maraganore DM, et al. Risk tables for parkinsonism and Parkinson’s disease. J Clin Epidemiol. 2002;55:25–
31.
12. Mayeux R, Marder K, Cote LJ, et al. The frequency of idiopathic
Parkinson’s disease by age, ethnic group, and sex in northern
Manhattan, 1988–1993. Am J Epidemiol. 1995;142:820–827.
13. Morens DM, Davis JW, Grandinetti A, Ross GW, Popper JS, White
LR. Epidemiologic observations on Parkinson’s disease: incidence
and mortality in a prospective study of middle-aged men. Neurology. 1996;46:1044–1050.
14. Van Den Eeden SK, Tanner CM, Bernstein AL, et al. Incidence of
Parkinson’s disease: variation by age, gender, and race/ethnicity.
Am J Epidemiol. 2003;157:1015–1022.
15. Kuopio AM, Marttila RJ, Helenius H, Rinne UK. Changing
epidemiology of Parkinson’s disease in southwestern Finland.
Neurology. 1999;52:302–308.
www.slm-neurology.com
23
ENJ 2009; 1: (1). September 2009
European Neurological Journal
review article
Imaging in Familial Frontotemporal Lobar
Degeneration With Mutations in MAPT or PGRN
Jennifer L. Whitwell1 and Keith A. Josephs2
Affiliations: Departments of 1Radiology and 2Neurology, Mayo Clinic, Rochester, MN, USA
Submission date: 29th May 2009, Revision date: 11th July 2009, Acceptance date: 8th August 2009
A B STRA C T
Background:
Familial frontotemporal lobar degeneration is most commonly related to mutations in the microtubule associated protein tau (MAPT) gene or the progranulin (PGRN) gene.
Methods:
Review of imaging findings in subjects with MAPT and PGRN mutations.
Results:
Patterns of atrophy vary across subjects yet patterns of anterior temporal dominant atrophy appear to be associated with MAPT mutations, while parietal lobe atrophy and significant asymmetry appear to be associated with PGRN
mutations.
Conclusions:
Imaging may be helpful in differentiating familial frontotemporal dementia patients with mutations in MAPT from those
with mutations in PGRN.
Keywords: frontotemporal lobar degeneration, tau, progranulin, structural imaging, functional imaging, voxel-based morphometry, genetic
Correspondence: Keith A. Josephs, MST, MD, MS, Department of Neurology, Mayo Clinic, 200 First Street SW, Rochester, MN
55905, USA. Tel: +1-507-538-1038; fax: +1-507-538-6012; e-mail: josephs.keith@mayo.edu
INTRODUCTION
and ubiquitin-positive neuronal inclusions in the frontotemporal cortices and hippocampus (FTLD-U) [11,
12]. Future treatments are likely to target the proteins
underlying these disorders and hence predicting the underlying pathology will become increasingly important.
Many studies have characterized the clinical [8, 13–15]
and imaging features of these mutation-carriers. This
review will discuss the imaging patterns that have been
identified in both MAPT and PGRN mutation carriers,
particularly with a view to determine whether imaging
could be useful in the differential diagnosis of these patients.
Frontotemporal lobar degeneration (FTLD) is a heterogenous progressive disorder that consists of a number
of different clinical and pathological variants and is associated with atrophy of the frontal and temporal lobes.
Approximately 40% of subjects have a family history and
an autosomal dominant pattern of inheritance [1, 2].
Some patients, particularly those with familial FTLD,
have mutations in the microtubule-associated protein
tau (MAPT) gene [3, 4] while others have been shown
to have mutations in the progranulin (PGRN) gene [5,
6]. Both of these genes, which are the two most commonly affected, are located on chromosome 17q21 and
mutations in these genes account in combination for
approximately 10% to 20% of all FTLD subjects [7, 8].
Mutations in MAPT are associated with deposits of the
hyperphosphorylated protein tau in neurons and glial
cells in the frontal and temporal cortices of the brain [9,
10]. In contrast, mutations in PGRN are associated with
deposition of TAR DNA binding protein 43 (TDP-43)ENJ 2009; 1: (1). September 2009
MAPT MUTATIONS
To date, 44 different pathogenic mutations in MAPT
have been identified in over 100 tauopathy families
[4, 16]. These mutations include missense mutations,
silent mutations, single codon deletions and intronic
mutations. Imaging has been reported in many clinical
studies that describe families with different MAPT mu25
www.slm-neurology.com
European Neurological Journal
Grey matter loss in MAPT
group compared to controls
Grey matter loss in PGRN
group compared to controls
p<0.001 corrected for multiple comparisons using false discovery rate
Figure 1. P
atterns of atrophy described by Whitwell et al [43] shown on glass brain renders. Red arrows illustrate regional differences
between groups
tations. Studies investigating patients with the P301L
and IVS10+3 mutations have reported consistent patterns of bilateral frontotemporal atrophy or hypometabolism on positron emission tomography (PET) [14,
17–22], with some showing asymmetric patterns [17, 19]
and others showing symmetric [14, 20] patterns. Patterns of atrophy and hypometabolism have been noted
in the frontotemporal cortex in patients with the N279K
mutation [23–26], showing relatively symmetric patterns, although cases from one carefully studied N279K
kindred showed atrophy on magnetic resonance imaging (MRI), predominantly in the temporal lobes [24, 25],
particularly the medial temporal lobes [25]. Atrophy of
the basal ganglia and brainstem has also been reported
[23, 26–28]. Patterns of atrophy in subjects with the
S305N and IVS10+16 mutations have also been found
to load on the temporal lobes [29–33], with early changes observed particularly in the amygdala in two S305N
cases [29].
ties also often noted [17, 19, 20, 23, 24, 29, 32, 33]. These
specific behavioral and cognitive features have all been
associated with temporal lobe abnormalities [36–40].
Parkinsonism is found particularly in the N279K subjects [23, 24] and may be associated with basal ganglia
and brainstem abnormalities.
Group studies using voxel-level automated techniques,
such as voxel-based morphometry (VBM), have since
been performed investigating a couple of the intronic
MAPT mutations which provide more detailed information concerning patterns of atrophy throughout the
brain. These techniques allow the assessment of brain
volume loss at every voxel throughout the image. Whitwell et al [41] assessed patterns of gray matter loss on
MRI using VBM in five symptomatic patients with the
IVS10+16 mutation compared to 20 healthy controls.
We found patterns of gray matter loss in the anterior
and medial temporal lobes and orbitofrontal cortex in
the IVS10+16 patients, with more severe involvement
of the right hemisphere. We also demonstrated that the
IVS10+16 patients showed greater anterior medial temporal lobe involvement than subjects with Pick’s disease,
another tau disorder, and subjects with FTLD-U pathology [41], suggesting that it could be useful to distinguish
these different pathologies. In another group study,
Spina et al [42] applied the technique of VBM to assess
gray matter atrophy in seven symptomatic patients with
A few studies have also examined imaging in asymptomatic mutation carriers and have reported both structural and functional changes before the onset of symptoms [24, 25, 34, 35]. Consistent with these patterns
of atrophy, subjects with mutations in MAPT typically
present with executive deficits and behavioral abnormalities, often showing a disinhibited phenotype, with
hyperphagia, memory problems and language difficulENJ 2009; 1: (1). September 2009
26
www.slm-neurology.com
Imaging in Familial Frontotemporal Lobar Degeneration With Mutations in MAPT or PGRN
PGRN MUTATIONS
the IVS10+3 mutation compared to 19 healthy controls.
They identified gray matter loss in the bilateral medial
temporal lobe, opercular cortex, insula and orbitofrontal
cortex, with a slightly greater involvement of the right
hemisphere. They also noted that, while the majority of
subjects had a clinical diagnosis of behavioral variant
frontotemporal dementia (FTD), memory impairment
and word finding difficulties were commonly identified
in the IVS10+3 patients. Therefore, although both studies assessed different mutations, they identified very
similar patterns of regional atrophy affecting particularly the medial temporal lobes.
The first reports of imaging in PGRN carriers came
from early clinical studies that described single cases.
These case studies typically reported variable patterns
of atrophy involving the frontal and temporal lobes, as
well as the parietal lobes [15, 46–52], consistent with
patterns typically observed in sporadic FTLD. Patterns
were noted to be variable across subjects both within
and between families [15, 49, 53]. A common feature
that was noted in these case reports however was asymmetry in the patterns of atrophy, with either the left
or right side showing the greatest atrophy [14, 15, 46,
47, 49, 51, 52]. Studies with larger numbers of subjects
which typically have a variety of different PGRN mutations have confirmed the findings of these reports and
demonstrated that a high proportion (64% to 76%) of
subjects with PGRN mutations show an asymmetrical pattern of atrophy [53, 54]. Subcortical white matter signal changes have also been reported in cases of
PGRN mutation carriers [49, 50, 53], typically observed
in the regions of maximal cortical atrophy. Subjects with
mutations in PGRN show varied clinical presentations
[8, 11, 13, 49, 51, 53, 54]: some patients present with
behavioral changes, often showing an apathetic phenotype, while others have presented with asymmetrical
disorders such as progressive nonfluent aphasia, or corticobasal syndrome, consistent with reporting of asymmetrical patterns of atrophy. Episodic memory loss and
Parkinsonism have also been reported [11, 53, 54].
A couple of more recent VBM studies by Whitwell and
colleagues have examined patterns of atrophy across
larger groups of patients with a variety of different mutations in MAPT [43, 44]. In the first study, we investigated 12 patients that represented seven families with
six different MAPT mutations [43]. We found that the
group as a whole showed gray matter loss in the frontal,
temporal and parietal lobes, but with the most severe
loss identified in the temporal lobes, particularly the anterior and medial temporal lobes (Figure 1). We then
took this a step further in the second study by investigating patterns of atrophy in six different MAPT mutations (P301L, V337M, N279K, S305N, IVS10+16 and
IVS10+3) [44]. We demonstrated that the most severe
loss in all these mutations was found in the temporal
lobes, although the mutations differed in which regions
of the temporal lobe were most affected. The patients
with IVS10+16, IVS10+3, N279K and S305N mutations all showed gray matter loss focused on the medial
temporal lobes, while those with P301L or V337M mutations showed gray matter loss focused on the lateral
temporal lobes, with a relative sparing of the medial
temporal lobe. Subjects from all MAPT mutation groups
performed poorly on neuropsychological tests of episodic
memory and confrontation naming, although the P301L
and V337M subjects performed on average better on episodic memory than the other mutation groups, perhaps
reflecting relative sparing of the medial temporal lobe.
There therefore appears to be differences in patterns of
temporal lobe atrophy across these MAPT mutations,
which may aid in the differentiation of the different mutation carriers. Interestingly, the P301L and V337M mutations act in a different way to the other mutations suggesting a possible relationship between the effect of the
mutation on tau and the resultant patterns of atrophy
in MAPT mutation carriers. The IVS10+16, IVS10+3,
N279K and S305N mutations all influence the alternative splicing of exon 10 [4, 45] thus changing the ratio
between 3R and 4R tau isoforms resulting in an increase
in 4R tau, whereas the P301L and V337M mutations do
not affect splicing of exon 10 but instead they affect the
structure and functional properties of the tau protein [4,
45]. How these different disease mechanisms may influence these anatomical changes is however unclear.
www.slm-neurology.com
A couple of group studies have been performed using
symptomatic PGRN mutation carriers. Whitwell et al
[55] used the technique of VBM to assess gray matter
atrophy in eight patients with mutations in PGRN that
had a pathological diagnosis of FTLD-U and to compare
them to eight patients with a pathological diagnosis of
FTLD-U that had screened negative for mutations in
PGRN. Clinical diagnosis was matched across the two
groups. We found that the PGRN carriers as a group
showed a widespread and severe pattern of loss involving the frontal, temporal and parietal lobes compared
to controls. In contrast, the PGRN noncarriers showed
a less severe pattern of gray matter loss restricted
mainly to the frontal and temporal lobes, with very little involvement of the parietal lobe. In fact, the PGRN
carriers showed significantly more frontal and parietal
gray matter loss than the PGRN noncarriers. This study
therefore suggested that PGRN is associated with more
severe disease than sporadic FTLD, particularly with
greater involvement of the frontal and parietal lobes
than PGRN noncarriers [55]. Le Ber et al [54] found
similar results in a voxel-level analysis of single photon emission computed tomography (SPECT) images
in which they compared 10 PGRN mutation carriers to
31 subjects that had no PGRN mutations. They found
more severe patterns of hypoperfusion in the PGRN
mutation carriers with greater involvement of the right
frontal, posterior temporal and parietal lobes. It is pos27
ENJ 2009; 1: (1). September 2009
European Neurological Journal
sible therefore that PGRN mutations result in a more
malignant form of FTLD-U. Smaller brain weights and
more cortical atrophy have also been observed at pathology in PGRN carriers compared to noncarriers [11].
They also performed group analyses on FDG–PET images and found hypometabolism in the frontal lobe and
left middle temporal gyrus in the asymptomatic carriers.
Similarly, a recent study that followed an individual patient with a PGRN mutation over multiple years found
atrophy of the frontal, temporal and parietal lobes, particularly the left angular gyrus, 18 months before the
onset of symptoms [59]. These imaging changes match
very well with those previously identified in symptomatic PGRN mutation carriers [54, 55], and suggest that
it is possible to identify the characteristic parietal lobe
MRI signature in asymptomatic subjects.
Both studies implicated the parietal lobe as being more
severely affected in PGRN mutation carriers. Parietal
lobe atrophy has been previously reported in the single
cases of PGRN mutation carriers [15, 46, 47, 50–52], and
was noted to be a common feature in larger cohorts [13,
48, 54]. Pathological studies have shown that parietal
degeneration is observed in PGRN mutation carriers
[11, 15, 52, 56], and is greater than the parietal degeneration observed in PGRN noncarriers with FTLD-U
[11]. Clinical studies have also found that deficits that
result from the parietal lobe, such as limb apraxia, dyscalculia, visuoperceptual/visuospatial dysfunction and
constructional disorders, are commonly found in PGRN
mutation carriers [13, 48, 54]. All of these results together suggest that the parietal lobe is differentially
affected in PGRN carriers and therefore may be useful
in diagnosis. It is important to note, however, that patterns of atrophy are variable and, while parietal atrophy
is a good signature for PGRN at the group level, it may
not be present in every case. Similarly, there is currently no strong evidence for whether patterns of atrophy
vary systematically across different PGRN mutations,
although one study did note the absence of any obvious
correlations between clinical phenotype and genotype
[54].
PGRN VERSUS MAPT MUTATIONS
The studies discussed so far have found trends for signature patterns of atrophy in both subjects with mutations in MAPT and subjects with mutations in PGRN. A
couple of group studies have also compared imaging features directly between symptomatic MAPT and PGRN
mutation carriers in order to determine whether the
imaging features differ. Beck et al [13] studied eight patients with mutations in PGRN and compared them to
nine patients with mutations in MAPT using volumetric
MRI measurements. They found firstly that the hemispheric asymmetry was greater in the PGRN mutation
carriers than in the MAPT mutation carriers, with no
asymmetry observed in the MAPT carriers, and secondly, that the ratio of the anterior half of the brain to
the posterior half of the brain was significantly different
from controls in the MAPT carriers but not in the PGRN
carriers. The asymmetry findings support the previous
studies discussed above that demonstrated asymmetric
patterns of atrophy in PGRN, and further show that it
differs from subjects with MAPT which show a much
more symmetric pattern of atrophy. This difference was
also observed in another study that compared MAPT
and PGRN patients using visual inspection of patterns
of atrophy [14]. The lack of anterior-posterior gradient
in PGRN most likely reflects the involvement of both
the frontal and parietal lobes, however, their technique
was not sensitive enough to be able to detect regional
atrophy.
Since the patterns of atrophy in PGRN carriers have
now been relatively well characterized, the important
next question is whether these patterns can be identified in subjects before they develop any symptoms of
the disease. A study by Borroni et al [57] investigated
structural brain changes in gray matter, using VBM,
and white matter, using diffusion tensor imaging (DTI),
in seven asymptomatic PGRN mutation carriers from
one PGRN family in which the proband presented with
progressive nonfluent aphasia. These were compared to
10 PGRN mutation noncarriers and 15 controls. They
found no differences between the groups using VBM,
although they found white matter changes in the left
uncinate fasciculus and left inferior occipitofrontal fasciculus in the asymptomatic subjects using DTI, suggesting that white matter changes precede symptom onset. This pattern of left hemisphere white matter tract
damage is consistent with that expected in subjects with
progressive nonfluent aphasia. In contrast, a study by
Cruchaga et al [58] has found that gray matter changes
do occur before onset of symptoms in PGRN carriers.
They investigated gray matter loss using VBM in three
asymptomatic PGRN mutation carriers from a progressive nonfluent aphasia family compared to a group of 11
age-matched noncarriers (consisting of two noncarrier
family members and nine controls). Grey matter volume
loss was identified in the left frontal, temporal and parietal lobes and precuneus in the asymptomatic carriers.
ENJ 2009; 1: (1). September 2009
Whitwell et al [43] have taken a more rigorous approach by examining differences between PGRN and
MAPT carriers at the voxel-level using VBM. We analyzed 12 patients with mutations in PGRN that were
matched by time from disease onset to scan to 12 patients with mutations in MAPT. Both the PGRN and
MAPT groups showed gray matter loss in frontal, temporal and parietal lobes compared to controls, although
the focus of loss differed across the groups with loss predominantly identified in posterior temporal and parietal lobes in PGRN and anteromedial temporal lobes in
MAPT (Figure 1, red arrows). Since the MAPT carriers
were approximately 14 years younger than the PGRN
carriers, we also compared each group to a specific agematched control cohort. This comparison showed that
28
www.slm-neurology.com
Imaging in Familial Frontotemporal Lobar Degeneration With Mutations in MAPT or PGRN
the MAPT group had greater loss than the PGRN group
when compared to healthy subjects of the same age. We
also performed direct comparisons across the PGRN
and MAPT groups and found that the MAPT patients
had significantly greater gray matter loss in the medial
temporal lobes, insula and putamen, than the PGRN patients.
ing could have potential value for future clinical trials.
However, many questions have still to be answered and
more work is needed before imaging can be considered a
biomarker of disease or a tool for differential diagnosis
in familial FTLD. It will be important for future studies to assess larger groups of subjects and to compare
the many different types of PGRN and MAPT mutations directly to determine whether patterns generalize
across mutations. Similarly, more detailed comparisons
should be performed between mutation subjects and
typical sporadic FTLD subjects. It will also be important
for studies to utilize other analysis techniques, such as
detailed volumetric measurements to better characterize patterns of atrophy and variability between subjects.
Other imaging modalities could also provide important
information, for example, DTI to further assess white
matter changes in the brain and magnetic resonance
spectroscopy to analyze brain metabolites. The asymptomatic mutation carriers are also a very valuable and
as yet understudied group of subjects. They provide the
unique opportunity for future studies to not only develop
early markers of disease but also to track how atrophy
progresses over time which will be essential knowledge
for future clinical trials that want to target early cases
and modify disease progression.
The patterns of atrophy identified in the PGRN carriers therefore support those identified in our earlier
study, with involvement of the frontal, temporal and
parietal lobes, and support the other PGRN studies
discussed above that have suggested that the parietal
lobe is important. However, the MAPT carriers also had
some parietal lobe atrophy although it appeared to be
a consequence of progressed disease rather than being
a focus of loss as with PGRN carriers. Therefore, in familial frontotemporal dementia, parietal lobe atrophy
may not be as good as anterior temporal lobe atrophy
for differentiating PGRN carriers from MAPT carriers.
Another interesting finding was that the degree of loss
when compared to age-matched controls was greater in
the MAPT carriers. One could infer from this result that
MAPT may result in a more aggressive disease process.
It is unclear however whether this is an effect of mutation or simply age, since age has been shown to correlate to rate of atrophy in patients with dementia [60].
However, it is also possible that the degree of loss in the
PGRN carriers may have been reduced due to asymmetry in the patterns of atrophy in these subjects which
would be averaged out in the VBM analysis.
Disclosures: The authors have no financial interests to disclose related to the contents of this article.
REFERENCES
1. Chow TW, Miller BL, Hayashi VN, Geschwind DH. Inheritance of
frontotemporal dementia. Arch Neurol. 1999;56:817-822.
2. Rizzu P, Van Swieten JC, Joosse M, et al. High prevalence of mutations in the microtubule-associated protein tau in a population
study of frontotemporal dementia in the Netherlands. Am J Hum
Genet. 1999;64:414-421.
3. Hutton M, Lendon CL, Rizzu P, et al. Association of missense
and 5’-splice-site mutations in tau with the inherited dementia
FTDP-17. Nature. 1998;393:702-705.
4. Rademakers R, Cruts M, van Broeckhoven C. The role of tau
(MAPT) in frontotemporal dementia and related tauopathies.
Hum Mutat. 2004;24:277-295.
5. Baker M, Mackenzie IR, Pickering-Brown SM, et al. Mutations in
progranulin cause tau-negative frontotemporal dementia linked
to chromosome 17. Nature. 2006;442:916-919.
6. Cruts M, Gijselinck I, van der Zee J, et al. Null mutations in progranulin cause ubiquitin-positive frontotemporal dementia linked
to chromosome 17q21. Nature. 2006;442:920-924.
7. Gass J, Cannon A, Mackenzie IR, et al. Mutations in progranulin are a major cause of ubiquitin-positive frontotemporal lobar
degeneration. Hum Mol Genet. 2006;15:2988-3001.
8. Pickering-Brown SM, Rollinson S, Du Plessis D, et al. Frequency
and clinical characteristics of progranulin mutation carriers
in the Manchester frontotemporal lobar degeneration cohort:
comparison with patients with MAPT and no known mutations.
Brain. 2008;131:721-731.
9. Hutton M. Missense and splice site mutations in tau associated
with FTDP-17: multiple pathogenic mechanisms. Neurology.
2001;56:S21-S25.
10. Taniguchi S, McDonagh AM, Pickering-Brown SM, et al. The
neuropathology of frontotemporal lobar degeneration with
respect to the cytological and biochemical characteristics of tau
protein. Neuropathol Appl Neurobiol. 2004;30:1-18.
LONGITUDINAL STUDIES
The majority of studies have only dealt with imaging
at one time-point with little being known about how
atrophy progresses over time in these different patient
groups. Rates of whole brain atrophy have been reported
to be between approximately 2.5% and 3.5% per year in
MAPT mutation carriers [29, 42], with one study finding no difference between the rates in MAPT and PGRN
mutation carriers [61]. Rates of atrophy in PGRN carriers (3.7% per year) have been shown to be significantly
higher than the 1.7% per year that has been reported
in sporadic FTLD-U [62, 63]. There is also a suggestion
from a couple of studies that the rates of whole brain atrophy are approximately linear over time in MAPT mutation carriers, although this conclusion was only based
on a small number of cases [29, 42].
SUMMARY
Imaging markers show a lot of promise to be useful in
the differential diagnosis of MAPT and PGRN mutation
carriers, and may therefore be able to play a role in predicting pathology in familial FTLD patients. Patterns of
temporal dominant atrophy appear to be associated with
MAPT mutations, while parietal lobe atrophy and significant asymmetry appear to be associated with PGRN
mutations. These studies therefore suggest that imagwww.slm-neurology.com
29
ENJ 2009; 1: (1). September 2009
European Neurological Journal
11. Josephs KA, Ahmed Z, Katsuse O, et al. Neuropathologic features
of frontotemporal lobar degeneration with ubiquitin-positive
inclusions with progranulin gene (PGRN) mutations. J Neuropathol Exp Neurol. 2007;66:142-151.
12. Mackenzie IR, Baker M, Pickering-Brown S, et al. The neuropathology of frontotemporal lobar degeneration caused by mutations in the progranulin gene. Brain. 2006;129:3081-3090.
13. Beck J, Rohrer JD, Campbell T, et al. A distinct clinical, neuropsychological and radiological phenotype is associated with
progranulin gene mutations in a large UK series. Brain.
2008;131:706-720.
14. Ghetti B, Spina S, Murrell JR, et al. In vivo and postmortem
clinicoanatomical correlations in frontotemporal dementia and
parkinsonism linked to chromosome 17. Neurodegener Dis.
2008;5:215-217.
15. Snowden JS, Pickering-Brown SM, Mackenzie IR, et al. Progranulin gene mutations associated with frontotemporal dementia
and progressive non-fluent aphasia. Brain. 2006;129:3091-3102.
16. Alzheimer Disease & Frontotemporal Dementia Mutation Database. http://www.molgen.ua.ac.be/FTDmutations
17. Bird TD, Nochlin D, Poorkaj P, et al. A clinical pathological
comparison of three families with frontotemporal dementia and
identical mutations in the tau gene (P301L). Brain. 1999;122(Pt
4):741-756.
18. Heutink P, Stevens M, Rizzu P, et al. Hereditary frontotemporal
dementia is linked to chromosome 17q21-q22: a genetic and
clinicopathological study of three Dutch families. Ann Neurol.
1997;41:150-159.
19. Kodama K, Okada S, Iseki E, et al. Familial frontotemporal
dementia with a P301L tau mutation in Japan. J Neurol Sci.
2000;176:57-64.
20. van Swieten JC, Stevens M, Rosso SM, et al. Phenotypic variation
in hereditary frontotemporal dementia with tau mutations. Ann
Neurol. 1999;46:617-626.
21. Neumann M, Mittelbronn M, Simon P, et al. A new family with
frontotemporal dementia with intronic 10+3 splice site mutation
in the tau gene: neuropathology and molecular effects. Neuropathol Appl Neurobiol. 2005;31:362-373.
22. Tolnay M, Grazia Spillantini M, Rizzini C, Eccles D, Lowe J,
Ellison D. A new case of frontotemporal dementia and parkinsonism resulting from an intron 10 +3-splice site mutation in the
tau gene: clinical and pathological features. Neuropathol Appl
Neurobiol. 2000;26:368-378.
23. Arima K, Kowalska A, Hasegawa M, et al. Two brothers with
frontotemporal dementia and parkinsonism with an N279K
mutation of the tau gene. Neurology. 2000;54:1787-1795.
24. Arvanitakis Z, Witte RJ, Dickson DW, et al. Clinical-pathologic
study of biomarkers in FTDP-17 (PPND family with N279K tau
mutation). Parkinsonism Relat Disord. 2007;13:230-239.
25. Frank AR, Wszolek ZK, Jack CR Jr, Boeve BF. Distinctive MRI
findings in pallidopontonigral degeneration (PPND). Neurology.
2007;68:620-621.
26. Pal PK, Wszolek ZK, Kishore A, et al. Positron emission tomography in pallido-ponto-nigral degeneration (PPND) family (frontotemporal dementia with parkinsonism linked to chromosome
17 and point mutation in tau gene). Parkinsonism Relat Disord.
2001;7:81-88.
27. Slowinski JL, Schweitzer KJ, Imamura A, et al. Brainstem atrophy on routine MR study in pallidopontonigral degeneration. J
Neurol. 2009;256:827-829.
28. Soliveri P, Rossi G, Monza D, et al. A case of dementia parkinsonism resembling progressive supranuclear palsy due to mutation
in the tau protein gene. Arch Neurol. 2003;60:1454-1456.
29. Boeve BF, Tremont-Lukats IW, Waclawik AJ, et al. Longitudinal
characterization of two siblings with frontotemporal dementia
and parkinsonism linked to chromosome 17 associated with the
S305N tau mutation. Brain. 2005;128:752-772.
30. Kobayashi K, Kidani T, Ujike H, et al. Another phenotype of frontotemporal dementia and parkinsonism linked to chromosome-17
(FTDP-17) with a missense mutation of S305N closely resembling
ENJ 2009; 1: (1). September 2009
Pick’s disease. J Neurol. 2003;250:990-992.
31. Doran M, du Plessis DG, Ghadiali EJ, Mann DM, PickeringBrown S, Larner AJ. Familial early-onset dementia with tau
intron 10 + 16 mutation with clinical features similar to those of
Alzheimer disease. Arch Neurol. 2007;64:1535-1539.
32. Janssen JC, Warrington EK, Morris HR, et al. Clinical features of
frontotemporal dementia due to the intronic tau 10(+16) mutation. Neurology. 2002;58:1161-1168.
33. Pickering-Brown SM, Richardson AM, Snowden JS, et al. Inherited frontotemporal dementia in nine British families associated
with intronic mutations in the tau gene. Brain. 2002;125:732751.
34. Kishore A, Wszolek ZK, Snow BJ, et al. Presynaptic nigrostriatal function in genetically tested asymptomatic relatives
from the pallido-ponto-nigral degeneration family. Neurology.
1996;47:1588-1590.
35. Alberici A, Gobbo C, Panzacchi A, et al. Frontotemporal dementia: impact of P301L tau mutation on a healthy carrier. J Neurol
Neurosurg Psychiatr. 2004;75:1607-1610.
36. Le Ber I, Guedj E, Gabelle A, et al. Demographic, neurological and behavioural characteristics and brain perfusion
SPECT in frontal variant of frontotemporal dementia. Brain.
2006;129:3051-3065.
37. Zamboni G, Huey ED, Krueger F, Nichelli PF, Grafman J. Apathy
and disinhibition in frontotemporal dementia: Insights into their
neural correlates. Neurology. 2008;71:736-742.
38. Mummery CJ, Patterson K, Price CJ, Ashburner J, Frackowiak
RS, Hodges JR. A voxel-based morphometry study of semantic
dementia: relationship between temporal lobe atrophy and semantic memory. Ann Neurol. 2000;47:36-45.
39. Thompson SA, Patterson K, Hodges JR. Left/right asymmetry of
atrophy in semantic dementia: behavioral-cognitive implications.
Neurology. 2003;61:1196-1203.
40. Petersen RC, Jack CR Jr, Xu YC, et al. Memory and MRI-based
hippocampal volumes in aging and AD. Neurology. 2000;54:581587.
41. Whitwell JL, Josephs KA, Rossor MN, et al. Magnetic resonance
imaging signatures of tissue pathology in frontotemporal dementia. Arch Neurol. 2005;62:1402-1408.
42. Spina S, Farlow MR, Unverzagt FW, et al. The tauopathy associated with mutation +3 in intron 10 of Tau: characterization of
the MSTD family. Brain. 2008;131:72-89.
43. Whitwell JL, Jack CR Jr, Boeve BF, et al. Voxel-based morphometry patterns of atrophy in FTLD with mutations in MAPT or
PGRN. Neurology. 2009;72:813-20.
44. Whitwell JL, Jack CR Jr, Boeve BF, et al. Atrophy patterns in
IVS10+16, IVS10+3, N279K, S305N, P301L and V337M MAPT
mutations. Neurology, in press.
45. Goedert M, Jakes R. Mutations causing neurodegenerative
tauopathies. Biochim Biophys Acta. 2005;1739:240-250.
46. Boeve BF, Baker M, Dickson DW, et al. Frontotemporal dementia
and parkinsonism associated with the IVS1+1G->A mutation
in progranulin: a clinicopathologic study. Brain. 2006;129:31033114.
47. Huey ED, Grafman J, Wassermann EM, et al. Characteristics of
frontotemporal dementia patients with a Progranulin mutation.
Ann Neurol. 2006;60:374-380.
48. Rohrer JD, Warren JD, Omar R, et al. Parietal lobe deficits in
frontotemporal lobar degeneration caused by a mutation in the
progranulin gene. Arch Neurol. 2008;65:506-513.
49. Spina S, Murrell JR, Huey ED, et al. Clinicopathologic features
of frontotemporal dementia with progranulin sequence variation.
Neurology. 2007;68:820-827.
50. Spina S, Murrell JR, Huey ED, et al. Corticobasal syndrome associated with the A9D Progranulin mutation. J Neuropathol Exp
Neurol. 2007;66:892-900.
51. Masellis M, Momeni P, Meschino W, et al. Novel splicing mutation
in the progranulin gene causing familial corticobasal syndrome.
Brain. 2006;129:3115-3123.
52. Mackenzie IR, Baker M, West G, et al. A family with tau-negative
30
www.slm-neurology.com
Imaging in Familial Frontotemporal Lobar Degeneration With Mutations in MAPT or PGRN
58. Cruchaga C, Fernandez-Seara MA, Seijo-Martinez M, et al. Cortical atrophy and language network reorganization associated with
a novel progranulin mutation. Cereb Cortex. 2009;19:1751-1760.
59. Rohrer JD, Warren JD, Barnes J, et al. Mapping the progression
of progranulin-associated frontotemporal lobar degeneration. Nat
Clin Pract Neurol. 2008;4:455-460.
60. Josephs KA, Whitwell JL, Ahmed Z, et al. Beta-amyloid burden is not associated with rates of brain atrophy. Ann Neurol.
2008;63:204-212.
61. Whitwell JL, Jack CR Jr, Boeve BF, et al. Patterns and rates of
brain atrophy in frontotemporal dementia with mutations in
MAPT or PGRN. Neurology 72 (Suppl 3); A149)
62. Whitwell JL, Jack CR Jr, Parisi JE, et al. Rates of cerebral
atrophy differ in different degenerative pathologies. Brain.
2007;130:1148-1158.
63. Josephs KA, Whitwell JL, Shiung M, et al. Cross-sectional and
longitudinal volumetric MRI analysis in FTLD-U with and without progranulin gene mutations. Neurology. 2007;68(12)(suppl
1):A272-A272.
frontotemporal dementia and neuronal intranuclear inclusions
linked to chromosome 17. Brain. 2006;129:853-867.
53. Kelley BJ, Haidar W, Boeve BF, et al. Prominent phenotypic
variability associated with mutations in Progranulin. Neurobiol
Aging. 2009;30:739-751.
54. Le Ber I, Camuzat A, Hannequin D, et al. Phenotype variability
in progranulin mutation carriers: a clinical, neuropsychological,
imaging and genetic study. Brain. 2008;131:732-746.
55. Whitwell JL, Jack CR Jr, Baker M, et al. Voxel-based morphometry in frontotemporal lobar degeneration with ubiquitin-positive
inclusions with and without progranulin mutations. Arch Neurol.
2007;64:371-376.
56. Mukherjee O, Pastor P, Cairns NJ, et al. HDDD2 is a familial
frontotemporal lobar degeneration with ubiquitin-positive, taunegative inclusions caused by a missense mutation in the signal
peptide of progranulin. Ann Neurol. 2006;60:314-322.
57. Borroni B, Alberici A, Premi E, et al. Brain magnetic resonance
imaging structural changes in a pedigree of asymptomatic progranulin mutation carriers. Rejuvenation Res. 2008;11:585-595.
www.slm-neurology.com
31
ENJ 2009; 1: (1). September 2009
European Neurological Journal
review article
Frontal and Periventricular Brain White Matter Lesions
and Cortical Deafferentation of Cholinergic and Other
Neuromodulatory Axonal Projections
N.I. Bohnen1,2,3, C.W. Bogan1 and M.L.T.M. Müller1
Affiliations: 1Functional Neuroimaging, Cognitive and Mobility Laboratory, Department of Radiology, University of Michigan,
Ann Arbor, MI, USA; 2Department of Neurology, University of Michigan, Ann Arbor, MI, USA; 3VA Ann Arbor Healthcare System,
GRECC, Ann Arbor, MI, USA
Submission date: 21st June 2009, Revision date: 10th July 2009, Acceptance date: 2nd September 2009
A B STRA C T
White matter fiber bundles form a spatial pattern defined by anatomical and functional architecture. Structural lesions
in the white matter may cause clinical symptoms because of disruption of fiber tracts. The clinical significance will depend
on the anatomic location of such lesions and whether the functional integrity of specific fiber bundles is affected. Unlike
more acute lesions of stroke or multiple sclerosis that may cause sudden sensorimotor deficits, white matter lesions of aging manifest with more subtle and gradual symptoms that are often cognitive in nature. Such cognitive symptoms have
been explained by strategically located white matter lesions in the deep forebrain that may disrupt cholinergic projection
fibers at their proximal origin. Recent in vivo imaging studies provide supportive evidence that periventricular white matter
lesions are associated with cortical cholinergic deafferentation in elderly with leukoaraiosis. White matter lesions at the
frontal horns, so-called ‘capping,’ are in close proximity to cholinergic axons that originate in the basal forebrain. Therefore, these lesions may result in more significant cortical deafferentation because of the more proximal axonal disruption.
A unique anatomic feature common to all cortical projections from subcortical neuromodulator systems (that not only
include the cholinergic but also the monoaminergic systems, such as dopamine, serotonin, and norepinephrine) is that
the proximal axons largely pass through the deep forebrain before fanning out to the cortex. It is thus plausible that deep
frontal white matter lesions may result in not only cholinergic but also variable monoaminergic cortical deafferentation.
Keywords: acetylcholine, aging, monoamines, MRI, white matter, PET
Correspondence: Nicolaas I. Bohnen, MD, PhD, Functional Neuroimaging, Cognitive and Mobility Laboratory, Departments
of Radiology and Neurology, The University of Michigan, 24 Frank Lloyd Wright Drive, Box 362, Ann Arbor, MI 48105-9755, USA.
Tel:1-734-998-8400 ; fax: 1-734-998-8403 ; e-mail: nbohnen@umich.edu
INTRODUCTION
ANATOMY OF CHOLINERGIC
PATHWAYS AND WML
White matter lesions (WML) are commonly observed
on magnetic resonance imaging (MRI) scans in older
adults and are thought to occur in the context of cardiovascular disease [1]. These age-associated WML have
been affiliated with cognitive decline, including dementia, and, also, depression and impaired mobility [2–4].
Given the diverse nature of these neurological consequences of WML, we postulate the hypothesis that the
clinical sequelae of WML in part reflect the disruption
of axonal projection fibers of neuromodulator systems
that travel from subcortical nuclei to the cortex. In this
paper, we will mainly focus on the cholinergic pathways
and present indirect and more direct evidence for the
disruption of cholinergic fibers by WML. Anatomic evidence for similar white matter disruptive mechanisms
of mono-aminergic neuromodulator systems (dopamine,
serotonin, norepinephrine) is also discussed.
ENJ 2009; 1: (1). September 2009
Several sites within the basal forebrain supply cholinergic innervation to the brain [5]. The medial septal nucleus (Ch1 cell group) and the vertical limb nucleus of
the diagonal band (Ch2) provide the major cholinergic
input to the hippocampus. Cholinergic neurons of the
horizontal limb nucleus of the diagonal band (Ch3) provide the major cholinergic input of the olfactory bulb,
and cholinergic neurons of the nucleus basalis of Meynert (nbM; Ch4) provide the principal cholinergic input
of the remaining cerebral cortex and amygdala [6]. The
trajectories of white matter pathways linking the nbM
with the cerebral cortex have been traced immunohistochemically in the human brain [5]. These cholinergic
pathways arise from the deep forebrain looping closely
around the anterior corpus callosum and the frontal
horns of the ventricles. The lateral pathway passes lat33
www.slm-neurology.com
European Neurological Journal
vessel cerebrovascular disease may increase the likelihood of expressing dementia in those with co-occurring
AD pathology [11]. A possible mechanism to explain this
observation is that WML may affect cholinergic projections and may exacerbate co-existing cortical cholinergic
deficits that are typical for AD. For example, Bocti et al
found that ratings of such strategic locations of WML
correlated better with cognitive impairment in AD
than more global measures of WML [12]. More findings
in this line were provided by Swartz et al who showed
that strategically located WML at the intersection with
cholinergic pathways, contribute to cognitive, especially executive, impairment in patients with AD [13]. In
their study, Swartz and colleagues compared ratings of
regional WML on brain MRI in a large series of elderly
with cognitive impairment to published immunohistochemical tracings of cholinergic pathways. They found
that moderate to severe cholinergic pathway involvement by WML was identified in 30% of patients with
AD and in 60% of patients with vascular dementia [13].
Figure 1. C
holinergic axonal projections in the brain orig­inating from the nucleus basalis of Meynert (left).
White matter lesions in the deep forebrain identified
on FLAIR MRI may partially disrupt these cholinergic
pathways (right)
eral to the ventricles through the external capsule before fanning out to innervate the cerebral cortex. The
medial pathway passes through the white matter deep
to the cingulate gyrus [5].
WML are typically located in more superficial subcortical areas but are also prominent adjacent to the ventricles, in particular at the frontal and occipital horns
[7]. Structural lesions in the white matter may cause
symptoms because of disruption of fiber tracts. The
more superficial or subcortical WML may disrupt the
functional connectivity of association fibers that convey
cortico-cortical connections. However, the more deeply
located lesions may disrupt long axonal projection fibers
of neuromodulator systems that travel from subcortical nuclei to the cortex, such as the cholinergic system.
As fibers entering the deep forebrain from lower brain
centers radiate fan-like through the cerebral white matter to the cortex, their density per unit of brain tissue
volume decreases along the way from their source to
destination [8]. Hence, it is plausible that WML that
are in close proximity to the cholinergic pathways, especially at their more proximal origin, are most likely to
disrupt these cholinergic projection axons (Figure 1).
This is consistent with evidence suggesting that WML
within the frontal white matter tracts are especially detrimental relative to WML in other lobar locations [9].
Finally, with respect to cholinergic treatment, patients
with AD who had strategically located subcortical WML
affecting cholinergic projections had a better cognitive
response, especially executive and working memory
functions, to cholinesterase inhibitors compared to patients without such WML [14]. Thus, cerebrovascular
compromise of the cholinergic pathways may be a factor
that contributes more selectively than does total nonselective white matter lesion burden in response to cholinergic therapy in AD.
IN VIVO IMAGING SUPPORT
FOR THE CHOLINERGIC FIBER
DISRUPTION HYPOTHESIS BY WML
Cholinergic In Vivo Imaging
Using PET or SPECT Techniques
Positron emission tomography (PET) or single-photon
emission computed tomography (SPECT) imaging studies allow the assessment of the regional distribution
and quantitative measurement of neurotransmitters,
enzymes or receptors in the living brain, and can be applied to examine cholinergic expression in vivo. Choline
acetyltransferase (ChAT) and acetylcholinesterase
(AChE) are the two ubiquitous constituents of cholinergic pathways of the human brain [15]. A traditional
presynaptic marker of cholinergic neurons, ChAT has
not been imaged successfully in vivo. Although there are
no radioligands for ChAT, there are radiotracers for the
vesicular acetylcholine transporter (VAChT) and AChE
that have been shown to map acetylcholine cells in the
brain and to have a good correspondence with ChAT [15,
16].
INDIRECT CLINICAL AND IMAGING
EVIDENCE OF THE CHOLINERGIC FIBER
DISRUPTION HYPOTHESIS BY WML
There are several observations that may provide support for the notion that strategically located WML may
disrupt cholinergic output fibers from the nbM. For example, a single case postmortem study of Cerebral Autosomal Dominant Arteriopathy with Subcortical Infarcts
and Leukoencephalopathy (CADASIL) demonstrated
that cortical cholinergic projections from the nbM could
be affected by purely subcortical ischemic lesions in the
absence of Alzheimer pathology [10].
AChE has been recognized since 1966 as a reliable
marker for brain cholinergic pathways, including in the
human brain [5, 17]. It is localized predominantly in
WML have also been recognized as a significant pathology in Alzheimer disease (AD), that is to say, small
ENJ 2009; 1: (1). September 2009
34
www.slm-neurology.com
Frontal and Periventricular Brain White Matter Lesions and Cortical Deafferentation
of Cholinergic and Other Neuromodulatory Axonal Projections
ent aspects of WML. A comprehensive scale was developed by Scheltens et al [27], in which WML burden is
assessed based on size and quantity of WML in a given
neuroanatomical location, including periventricular and
nonperiventricular WML. Periventricular hyperintensities are further separated into frontal, occipital, and lateral aspects. Other rating scales of WML are the BrantZawadzki et al scale [24] and the Cardiovascular Health
Study Scale [26], both of which place relatively more
emphasis on periventricular WML. However, unlike the
Scheltens scale, the latter two rating scales do not specifically assess regional periventricular areas.
Periventricular WML are often proportional to overall burden of WML [28, 29]. Given the neuroanatomical propensity to disrupt several projection axons, the
burden of WML around the frontal ventricular horns
may be of particular interest. A recent study by our
group found evidence that visual detection of frontal
horn ‘capping’ of WML, as defined by the Scheltens et
al scale, may serve as a simple screening biomarker for
functionally more significant WML in the context of cortical cholinergic deafferentation [30].
Figure 2. T ransaxial (left), coronal (middle), and sagittal (right)
slices of [11C]PMP AChE PET images (summed radioactivity images 0 to 25 minutes postinjection) and corresponding MRI slices showing normal AChE biodistribution
with most intense uptake in the basal ganglia, followed
by the cerebellum, with lower levels in the cortex
cholinergic cell bodies and axons. In the cortex, AChE
is present in axons innervating it from the basal forebrain [5]. There is also AChE in intrinsic cortical neurons and low levels of AChE are probably present in the
noncholinergic structures postsynaptic to the nucleus
basalis innervation [18]. AChE activity in the human
brain has been mapped using PET with the [11C]PMP
[19, 20] and [11C]MP4A [21] radioligands. These radioligands are acetylcholine analogues that serve as a selective substrate for AChE hydrolysis [22]. The hydrolyzed
radioligand becomes trapped as a hydrophilic product
locally in the brain following the AChE biodistribution.
Using [11C]PMP, Kuhl and colleagues [19] found a distribution of AChE activity that closely correlated with the
postmortem histochemical distribution in normal volunteers (Figure 2).
Volumetric assessment of the volume of WML around
the frontal horns may provide a more precise alternative
to the Scheltens et al scale for measuring frontal periventricular burden of WML. For example, the volume of
frontal horn caps on FLAIR or T2-weighted MR images
can be determined by tracing the volume of interest
(VOI) around the outline of the hyperintense caps on
multiple slices and summing the individual tracings into
a single volume (Figure 3).
The above-described methods of semiquantitative
visual assessment of WML burden are limited by the
fact that they are grader dependent. Intergrader variability may affect the reliability of these measures and
limit comparison across studies. Furthermore, with increasing magnet strength, white matter abnormalities
become more detailed on MRI. Punctuate and confluent
white matter areas can often be observed that appear
to be below a subthreshold hyperintensity, which sometimes are referred to as ‘dirty’ white matter. Inclusion
The VAChT is localized in the acetylcholine nerve terminals and carries acetylcholine from the cytoplasm
into the vesicles. Radiolabeling of these vesicular transporters would therefore provide a presynaptic marker of
cholinergic innervation. Several radioligands that target
the VAChT have been labeled [23]. Of these, only (–)-5[123I]iodobenzovesamicol (IBVM), a SPECT radiotracer,
has been used to image the living human brain [23].
VAChT SPECT could also be used to study the integrity
of cholinergic nerve terminals.
Visual Assessment of Periventricular
and Lobar WML on MRI
Although WML can be recognized on computed tomography (CT) scans, MRI scans are most commonly
used to identify WML, in particular, T2-weighted or
Fluid-Attenuated Inverse Recovery (FLAIR) sequences.
WML appear as punctuate or more confluent hyperintense areas on these sequences and are therefore also
often referred to as white matter hyperintensities. Several visual rating scales have been developed to estimate
WML burden [24–27], each with an emphasis on differwww.slm-neurology.com
Figure 3. E
xample of volumetric assessment of WML burden of
the periventricular frontal horn caps. VOI values are
drawn on the FLAIR MRI slices and summed to obtain an
estimate of WML volume of the frontal horns
35
ENJ 2009; 1: (1). September 2009
European Neurological Journal
intensity of cerebellar white matter voxels as a reference to define hyperintense supratentorial voxels. The
cerebellum was chosen as a reference because of the
clinical observation that age-associated WML, unlike
diseases such as multiple sclerosis, relatively spares the
cerebellum. For example, we reviewed brain MRI FLAIR
sequences of 104 community-dwelling subjects between
the ages of 20 and 85. Ratings of the Scheltens et al scale
confirmed that the cerebellum is overall spared for ageassociated white matter lesions. Ninety-five subjects
had a cerebellar score of 0; 6 subjects had a score of 1;
and 3 subjects had a score of 3. The mean score of cerebellar lesions was 0.1±0.5 which was <1% (0.66%) of
the total supratentorial white matter ratings in these
subjects (15.1±12.7).
We will now follow with a short technical description
of our automated reference-based WML identification
method. Volumetric SPGR (Spoiled Gradient Recall) sequences (TE=5, TR=25, flip angle=40 degrees, NEX=1,
slice thickness=1.5 mm) and fast fluid-attenuated inversion recovery (FLAIR) (TR/TE = 9002/56 ms Ef; TI
= 2200 ms, NEX = 1; slice thickness=5 mm) brain MRI
sequences (without contrast) were obtained on a Signa
1.5 T GE scanner (GE Medical Systems, Milwaukee, WI,
USA). All axial sequences were obtained with a 24 cm
field of view and a 192 × 256 pixel matrix.
Spatial preprocessing and WML identification were
done using standard routines and functions in the software package SPM5 [32, 33]. All MR images were normalized to the Montreal Neurological Institute (MNI)
standard ICBM-152 template brain. The following
steps were performed in SPM5 using default settings
(Figure 4):
Figure 4. M
ain steps involved in the automated method of identifying FLAIR hyperintensities showing normalized SPGR
(A) and normalized, coregistered FLAIR (B) slices. SPGR
segmented white matter mask (C) is used to exclude
cortical gray matter and extracerebral tissue on the
coregistered FLAIR volume (D). The template brain cerebellar mask (E) is used to identify the cerebellum volume
that is used for quantitative assessment of white matter
signal intensities on the FLAIR volume (F). Suprathreshold
voxels in the thresholded FLAIR white matter volume are
depicted in red color (G)
(1)Coregistration of FLAIR image to SPGR image for
each subject.
(2)Normalization of SPGR images to the ICBM-152
template brain and application of SPGR normalization parameters to the FLAIR image.
(3)Segmentation of SPGR images into white matter,
gray matter, and cerebrospinal fluid. The white
matter segmentation image is an underestimate
of the white matter because some WML appear
as gray matter on the SPGR. Therefore, step 8 is
performed below to accurately identify WML not
included in the original identification.
of these ‘dirty’ white matter areas may result in relative
overestimation of WML burden. More objective methods
are therefore needed for a more reliable WML burden
assessment.
Reference-Based Automated
Assessment of Supratentorial
Hyperintense White Matter Voxels
(4)Masking of FLAIR image by SPGR white matter
segmentation to produce a FLAIR white matter image (FLAIR WM image).
Automated routines have been developed to define
WML burden. These methods are typically based on
segmentation of white matter MRI series, thresholding
hyperintense white matter voxels, or ‘fuzzy’ neighboring cluster-based voxel analysis. A different approach is
based on the use of a reference region or tissue, such as
intensity of normal gray matter [31] or the cerebellum.
We have developed a routine where we use the mean
ENJ 2009; 1: (1). September 2009
(5)Masking of FLAIR WM image using template cerebellum VOI based on an averaged SPGR image
from 16 older healthy controls.
(6)Calculation of mean and standard deviation (SD)
of voxel intensities within the masked image of
FLAIR cerebellum white matter.
(7)Thresholding of FLAIR WM image to create FLAIR
36
www.slm-neurology.com
Frontal and Periventricular Brain White Matter Lesions and Cortical Deafferentation
of Cholinergic and Other Neuromodulatory Axonal Projections
Table 1. Mean (SD) of measures of WML burden and their age-corrected partial correlation coefficients and significance levels with
overall cortical AChE activity
Frontal periventricular WML
Scheltens et al ratings [27]
Volume of frontal horn
capping, mm3
Automated assessment of frontal
lobe WML (% hyperintense
voxels relative to cerebellum)
Cortical
Mean (SD)
R= –0.517
Mean (SD)
R= –0.511
Mean (SD)
R= –0.592
AChE activity
2.33 (1.28)
P < 0.05
0.87 (0.65)
P < 0.05
0.67 (0.88)
P < 0.05
Burden of frontal lobe WML (natural log transformation of the number of hyperintense voxels) from our reference-based automated
method is expressed as percentage of the total white matter volume.
WML mask based on cerebellar white matter
threshold (mean + 3SD). The FLAIR WML mask
identifies WML on the FLAIR; however, it is an underestimate of burden of WML due to the issue of
SPGR white matter segmentation described in step
3. Hence, step 8 is performed to identify WML adjacent to the FLAIR WML mask.
(8)Overlaying of FLAIR WML mask on normalized
FLAIR image and thresholding (mean + 3SD) of
voxels immediately surrounding identified WML to
create a more accurate FLAIR WML image.
(9)Identification of region-specific WML by masking of FLAIR WML image using regional masks
included in the Wake Forest University PickAtlas
software toolbox for SPM5 which includes the Talairach Daemon database.
(10)Summing suprathreshold voxels in each volume
of interest from the masked FLAIR WML image.
Brain regions were summed for left and right hemispheres.
Combined PET and MRI Studies
of Age-Associated WML and Cortical
Cholinergic Denervation
We recently reported on the in vivo findings of cortical AChE activity in subjects with variable degrees of
age-associated WML [30]. Nondemented community
dwelling middle-aged and elderly subjects (mean age
71.0±9.2; 55–84 years; n=18) underwent brain MRI and
AChE PET imaging. The severity of periventricular and
nonperiventricular WMH on FLAIR MRI images was
scored using the semiquantitative rating scale of Scheltens et al [27]. [11C]PMP AChE PET imaging was used
to assess cortical AChE activity [34]. The results of this
study showed that the severity of periventricular (Rs=
–0.52, P=0.04), but not nonperiventricular (Rs= –0.20,
ns), WML was inversely related to global cortical AChE
activity. Regional cortical cholinergic effects of periventricular WML were most significant for the occipital lobe
[30]. There was no significant effect of cerebral atrophy
to explain the study findings. These findings support a
regionally specific disruption of cholinergic projection
fibers by WML. Furthermore, the study found evidence
that visual detection of frontal horn ‘capping’ of WML
may serve as a simple screening biomarker for funcwww.slm-neurology.com
Figure 5. S catter plots showing cortical AChE activity with
frontal cap volume (upper figure) and cortical AChE
activity with reference-based automated frontal lobe
WML assessment (lower figure)
tionally more significant WML in the context of cortical
cholinergic deafferentation [30].
To further explore the utility of alternative WML burden assessment methods, we performed additional MRI
analyses of this data set using volumetric assessment
of the frontal horn caps and the automated referencebased method described above. We found that a large
volume of frontal horn capping was associated with
lower cortical AChE activity (Table 1; Figure 5). We
also found that quantitative assessment of the increased
number of frontal lobe WML hyperintense voxels was
associated with lower cortical AChE levels (Table 1).
However, the Scheltens et al rating scale for burden of
WML in the frontal lobe was not significantly associated
37
ENJ 2009; 1: (1). September 2009
European Neurological Journal
of the limbic system and the neocortex [38]. The nucleus
accumbens is the only noncortical innervation area of
the mesocortical dopamine system [39]. The mesocortical projections include isocortical areas, including the
mesial frontal, anterior cingulate, entorhinal, and perirhinal cortices, as well as allocortical areas including
the olfactory tubercle and bulb, piriform cortex, nucleus
accumbens and amygdaloid complex [39].
Figure 6. C
ortical pathways of monoaminergic neurotransmitter
systems. The dopamine cortical pathways (left image,
green color) originate from the ventral tegmental area.
Norepinephrine cortical pathways (middle image,
red color) originate from the locus ceruleus. Serotonin
cortical pathways originate from the raphe nuclei
(right image, orange color). Dopaminergic cortical
projections are more limited to the mesiofrontal cortex
whereas the other monoaminergic systems have more
widespread cortical projections
Cortical Projections of Norepinephrine (NE)
Norepinephrine in the central nervous system is produced by the locus ceruleus, which projects to virtually
all brain regions with the exception of the basal ganglia,
nucleus accumbens and olfactory tubercle [40, 41]. It
was the work of Ungerstedt in particular that demonstrated the extensive innervation of telencephalic structures by locus ceruleus neurons [42]. The ascending fibers of the locus ceruleus projection that enter the medial
forebrain bundle give rise to several distinct groups of
fibers. The largest of these is made up of fascicles of fibers that leave the medial forebrain bundle laterally and
part of these enter basal telencephalic areas, whereas
others continue into the external capsule. Finally, there
is a group of fibers that turn around the genu of the
corpus callosum and then run caudally in the cingulum
[40].
with cortical AChE activity, which may indicate a higher
sensitivity of the automated method to identify WML.
Although detailed ratings scales have been developed
to provide semiquantitative ratings of strategic locations of WML that may disrupt cholinergic projections
[12], our data indicate that limited assessment of periventricular WML provides a simplified rating tool to estimate the impact on cortical cholinergic hypofunction.
Furthermore, volumetric assessment of frontal horn
‘capping’ of WMH may serve as a simple screening biomarker for functionally more significant leukoaraiosis
[30].
Cortical Projections of Serotonin (5HT)
Serotonergic projections, which mainly originate in
the raphe nuclei of the brainstem have broad cortical
projections. Serotonergic projections to the cortex arise
primarily from the dorsal raphe nucleus and medial
raphe nucleus [43]. The dorsal raphe nucleus consists
primarily of ipsilateral projections to the frontal cortex
while the medial raphe nucleus projects bilaterally to
frontal, parietal and occipital cortices [43].
FRONTAL PERIVENTRICULAR
WHITE MATTER LESIONS MAY
ALSO DISRUPT NON-CHOLINERGIC
NEUROMODULATORY AXONAL
PROJECTIONS
Vertebrates have subcortical structures, known as neuromodulatory systems, which regulate fundamental behavior and provide the foundation for cognitive function
in higher organisms. Attention, emotion, goal-directed
behavior, and decision making all derive from the interaction between the neuromodulatory systems and areas
such as the amygdala, cortex, and hippocampus [35].
Ascending neuromodulatory systems include noradrenergic, dopaminergic, serotonergic, and cholinergic projections from the brainstem and basal forebrain regions
to broad areas of the central nervous system [36]. Each
of these neuromodulator systems has a subcortical origin and projects a specific neurotransmitter to variable,
often broad, areas of the cortex (Figure 6).
Monoaminergic Pathways and WML
The NE and 5HT pathways have more widespread
cortical projections whereas DA cortical projections are
more limited to the mesiofrontal cortex. WML may also
interrupt these monoaminergic pathways, which could
explain some of the other behavioral consequences of
age-associated WML such as depression and increased
risk of falling. Although specific in vivo cortical assessment of presynaptic DA and NE functions is limited
because of low binding levels at the cortex [44], [18F]6Fluorodopa (FDOPA) PET could be used to assess for
combined monoaminergic cortical deafferentation in the
presence of white matter lesions [45]. Serotonin transporter ligands, such as [11C]DASB [46], can be used for
specific assessment of serotonergic cortical cholinergic
deafferentation. Future studies are needed to further
explore the effects of WML on the deafferentation of
cortical monoaminergic pathways and the behavioral or
clinical consequences thereof.
The Mesocortical Dopamine (DA) System
There are two major subtypes of dopaminergic neurons in the brain: the neurons of the substantia nigra
pars compacta (A9 neurons) [37], which give rise to the
nigrostriatal pathway; and the A10 neurons of the ventral tegmental area (VTA), which give rise to the mesolimbic and mesocortical pathways that innervate parts
ENJ 2009; 1: (1). September 2009
38
www.slm-neurology.com
Frontal and Periventricular Brain White Matter Lesions and Cortical Deafferentation
of Cholinergic and Other Neuromodulatory Axonal Projections
CONCLUSIONS
human cerebral cortex. Brain Res. 1992;577:112–120.
16. Weihe E, Tao-Cheng JH, Schafer MK, Erickson JD, Eiden LE.
Visualization of the vesicular acetylcholine transporter in cholinergic nerve terminals and its targeting to a specific population of
small synaptic vesicles. Proc Natl Acad Sci USA. 1996;93:3547–
3552.
17. Shute CC, Lewis PR. Electron microscopy of cholinergic terminals and acetylcholinesterase-containing neurones in the
hippocampal formation of the rat. Z Zellforsch Mikrosk Anat.
1966;69:334–343.
18. Heckers S, Geula C, Mesulam MM. Acetylcholinesterase-rich
pyramidal neurons in Alzheimer’s disease. Neurobiol Aging.
1992;13:455–460.
19. Kuhl D, Koeppe R, Snyder S, Minoshima S, Frey K, Kilbourn M.
Mapping acetylcholinesterase activity in human brain using PET
and N-[11C]Methylpiperidinyl propionate (PMP). J Nucl Med.
1996;37(suppl):21P.
20. Kuhl DE, Koeppe RA, Minoshima S, et al. In vivo mapping of
cerebral acetylcholinesterase activity in aging and Alzheimer’s
disease. Neurology. 1999;52:691–699.
21. Iyo M, Namba H, Fukushi K, et al. Measurement of acetylcholinesterase by positron emission tomography in the brain of
healthy controls and patients with Alzheimer’s disease. Lancet.
1997;349:1805–1809.
22. Irie T, Fukushi K, Akimoto Y, Tamagami H, Nozaki T. Design and
evaluation of radioactive acetylcholine analogs for mapping brain
acetylcholinesterase (AChE) in vivo. Nucl Med Biol. 1994;21:801–
808.
23. Kuhl DE, Koeppe RA, Fessler JA, et al. In vivo mapping of cholinergic neurons in the human brain using SPECT and IBVM. J
Nucl Med. 1994;35:405–410.
24. Brant-Zawadzki M, Fein G, Van Dyke C, Kiernan R, Davenport L,
de Groot J. MR imaging of the aging brain: patchy white-matter
lesions and dementia. Am J Neuroradiol. 1985;6:675–682.
25. Fazekas F, Chawluk JB, Alavi A, Hurtig HI, Zimmerman RA.
MR signal abnormalities at 1.5 T in Alzheimer’s dementia and
normal aging. AJR Am J Roentgenol. 1987;149:351–356.
26. Longstreth WT Jr, Manolio TA, Arnold A, et al. Clinical correlates
of white matter findings on cranial magnetic resonance imaging
of 3,301 elderly people: the cardiovascular health study. Stroke.
1996;27:1274–1282.
27. Scheltens P, Barkhof F, Leys D, et al. A semiquantative rating
scale for the assessment of signal hyperintensities on magnetic
resonance imaging. J Neurol Sci. 1993;114:7–12.
28. DeCarli C, Fletcher E, Ramey V, Harvey D, Jagust WJ. Anatomical mapping of white matter hyperintensities (WMH): exploring
the relationships between periventricular WMH, deep WMH, and
total WMH burden. Stroke. 2005;36:50–55.
29. Barkhof F, Scheltens P. Is the whole brain periventricular? J
Neurol Neurosurg Psychiatry. 2006;77:143–144.
30. Bohnen NI, Muller ML, Kuwabara H, Constantine GM, Studenski SA. Age-associated leukoaraiosis and cortical cholinergic
deafferentation. Neurology. 2009;72:1411–1416.
31. de Boer R, Vrooman HA, van der Lijn F, et al. White matter lesion extension to automatic brain tissue segmentation on MRI.
Neuroimage. 2009;45:1151–1161.
32. Ashburner J, Friston K. Multimodal image coregistration and
partitioning – a unified framework. Neuroimage. 1997;6:209–217.
33. Ashburner J, Friston KJ. Unified segmentation. Neuroimage.
2005;26:839–851.
34. Nagatsuka S, Fukushi K, Shinotoh H, et al. Kinetic analysis of
[(11)C]MP4A using a high-radioactivity brain region that represents an integrated input function for measurement of cerebral
acetylcholinesterase activity without arterial blood sampling. J
Cereb Blood Flow Metab. 2001;21:1354–1366.
35. Krichmar JL. The neuromodulatory system: A framework for
survival and adaptive behavior in a challenging world. Adapt
Behav. 2008;16:385.
36. Briand LA, Gritton H, Howe WM, Young DA, Sarter M. Modulators in concert for cognition: modulator interactions in the
Recent in vivo imaging findings indicate that WML
previously described as ‘non-specific’, when in strategic
locations such as the frontal or periventricular regions,
can be associated with cortical cholinergic deafferentation. These findings support a regionally specific disruption of cholinergic projection fibers by WML and may
augur novel therapeutic approaches to treating cognitive symptoms of WML in the elderly. Neurochemical
imaging techniques are available to test the hypothesis
that frontal or periventricular forebrain WML may also
disrupt monoaminergic projections that pass through
the deep forebrain before innervating the neocortex.
Disclosures: The author has no financial interests to disclose
related to the contents of this article.
REFERENCES
1. Pantoni L, Garcia JH. Pathogenesis of leukoaraiosis. Stroke.
1997;28:652–659.
2. Gunning-Dixon FM, Raz N. The cognitive correlates of white
matter abnormalities in normal aging: A quantitative review.
Neuropsychology. 2000;14:224–232.
3. Au R, Massaro JM, Wolf PA, et al. Association of white matter
hyperintensity volume with decreased cognitive functioning: the
Framingham Heart Study. Arch Neurol. 2006;63:246–250.
4. Baezner H, Blahak C, Poggesi A, et al. Association of gait and balance disorders with age-related white matter changes: the LADIS
study. Neurology. 2008;70:935–942.
5. Selden NR, Gitelman DR, Salamon-Murayama N, Parrish TB,
Mesulam MM. Trajectories of cholinergic pathways within the
cerebral hemispheres of the human brain. Brain. 1998;121:2249–
2257.
6. Mesulam MM, Geula C. Nucleus basalis (Ch4) and cortical cholinergic innervation in the human brain: Observations based on the
distribution of acetylcholinesterase and choline acetyltransferase.
J Comp Neurol. 1988;275:216–240.
7. Wen W, Sachdev P. The topography of white matter hyperintensities on brain MRI in healthy 60- to 64-year-old individuals.
Neuroimage. 2004;22:144–154.
8. Pawlak M, Krejza J. A new visual scale to assess white matter hyperintensities within cholinergic pathways. Stroke. 2005;36:2064–
2065.
9. Schuff N, Capizzano AA, Du AT, et al. Different patterns of Nacetylaspartate loss in subcortical ischemic vascular dementia
and AD. Neurology. 2003;61:358–364.
10. Mesulam M, Siddique T, Cohen B. Cholinergic denervation in a
pure multi-infarct state: observations on CADASIL. Neurology.
2003;60:1183–1185.
11. Snowdon DA, Greiner LH, Mortimer JA, Riley KP, Greiner PA,
Markesbery WR. Brain infarction and the clinical expression of
Alzheimer disease. The Nun Study. JAMA. 1997;277:813–817.
12. Bocti C, Swartz RH, Gao FQ, Sahlas DJ, Behl P, Black SE. A
new visual rating scale to assess strategic white matter hyperintensities within cholinergic pathways in dementia. Stroke.
2005;36:2126–2131.
13. Swartz RH, Sahlas DJ, Black SE. Strategic involvement of cholinergic pathways and executive dysfunction: Does location of white
matter signal hyperintensities matter? J Stroke Cerebrovasc Dis.
2003;12:29–36.
14. Behl P, Bocti C, Swartz RH, et al. Strategic subcortical hyperintensities in cholinergic pathways and executive function decline
in treated Alzheimer patients. Arch Neurol. 2007;64:266–272.
15. Mesulam MM, Geula C. Overlap between acetylcholinesteraserich and choline acetyltransferase-positive (cholinergic) axons in
www.slm-neurology.com
39
ENJ 2009; 1: (1). September 2009
European Neurological Journal
prefrontal cortex. Prog Neurobiol. 2007;83:69–91.
37. Dahlstrom A, Fuxe K. Evidence for the existence of monoaminecontaining neurons in the central nervous system. I. Demonstration of monoamines in the cell bodies of brain stem neurons. Acta
Physiol Scand. 1964;62(suppl 232):231–255.
38. Bjorklund A, Lindvall O. Dopamine-containing systems in the
CNS. In: Bjorklund A, Hokfelt T, eds. Handbook of Chemical
Neuroanatomy. Amsterdam: Elsevier; 1984:55–122.
39. Moore RY, Bloom FE. Central catecholamine neuron systems:
anatomy and physiology of the dopamine systems. Annu Rev
Neurosci. 1978;1:129–169.
40. Moore RY, Bloom FE. Central catecholamine neuron systems:
anatomy and physiology of the norepinephrine and epinephrine
systems. Annu Rev Neurosci. 1979;2:113–168.
41. Berridge CW, Waterhouse BD. The locus coeruleus-noradrenergic
ENJ 2009; 1: (1). September 2009
system: modulation of behavioral state and state-dependent
cognitive processes. Brain Res Brain Res Rev. 2003;42:33–84.
42. Ungerstedt U. Stereotaxic mapping of the monoamine pathways
in the rat brain. Acta Physiol Scand. 1971;367:1–48.
43. O’Hearn E, Molliver ME. Organization of raphe-cortical projections in rat: a quantitative retrograde study. Brain Res Bull.
1984;13:709–726.
44. Bohnen NI, Frey KA. Imaging of cholinergic and monoaminergic neurochemical changes in neurodegenerative disorders. Mol
Imaging Biol. 2007;9:243–257.
45. Garnett ES, Firnau G, Nahmias C. Dopamine visualized in the
basal ganglia of living man. Nature. 1983;305:137–138.
46. Frankle WG, Huang Y, Hwang DR, et al. Comparative evaluation
of serotonin transporter radioligands 11C-DASB and 11C-McN
5652 in healthy humans. J Nucl Med. 2004;45:682–694.
40
www.slm-neurology.com
European Neurological Journal
review article
Restless Legs Syndrome and Peripheral
Neuropathy—A Critical Review
ET Hattan1, C Chalk2 and RB Postuma2
Affiliations: 1Section of Pulmonary and Critical Care Medicine, The University of Chicago Medical Center, Chicago, IL, USA;
2
Departments of Neurology and Neurosurgery and Medicine, McGill University, Montreal General Hospital, Montreal, Quebec,
Canada
Submission date: 21st June 2009, Revision date: 27th July 2009, Acceptance date: 7th August 2009
A B STRA C T
Much of our pathophysiologic understanding of the etiology of restless leg syndrome (RLS) incriminates abnormalities
within the central nervous system (CNS). However, peripheral neuropathy is classically listed as a risk factor for RLS. This discrepancy is difficult to reconcile. If there truly is a connection between neuropathy and RLS, it has important implications
for the screening and treatment of RLS, and it challenges our current concepts of RLS as a predominantly CNS disease.
The proposed association between RLS and peripheral neuropathy is based upon case reports, conflicting case–control
studies, and findings from pathological studies. Prevalence estimates of RLS among peripheral neuropathy patients range
from 5.2% to 37%. Initial reports found increased prevalence of RLS in patients with acquired neuropathy, but recently a
large blinded case–control study did not confirm these results. Furthermore, in this recent study, neuropathy subjects often
endorsed RLS-like symptoms, which could not be confirmed on diagnostic evaluation, suggesting that symptom overlap
between RLS and neuropathic pain may be a common confound. This study also showed an increased prevalence of
RLS selectively among hereditary neuropathy patients, which raises questions about the genetic relationship between RLS
and neuropathy. Small pathologic studies have detected features of subclinical sensory neuropathy in some RLS patients.
If confirmed, these findings may suggest the existence of a separate subclinical neuropathy/RLS syndrome, the nature of
which must be further delineated.
Keywords: restless legs syndrome, peripheral neuropathy, prevalence, review, genetics
Correspondence: Dr Ronald B. Postuma, Division of Neurology, L7-305 Montreal General Hospital, 1650 Cedar Ave., Montreal, Quebec, Canada H3G 1A4. Tel: +1-514-934-8058; fax: +1-514-934-8265; e-mail: ron.postuma@mcgill.ca
INTRODUCTION
tern of lower extremity paresthesia coupled with the
compulsion to move, worsening with rest and alleviated
by movement [3, 4]. These core features remain the basis of the current diagnostic criteria [5]. Ekbom coined
the name “asthenia crurum paraesthetica” or “restless
legs” and recognized the condition’s familial tendencies
and frequent clinical association with pregnancy and
anemia. He also alluded to potential clinical mimics by
distinguishing the presence of “creeping sensations” vs
“pain”. The latter were excluded from the diagnosis.
Ekbom specified that, despite its considerable clinical
morbidity, patients with RLS lacked objective evidence
of neurological abnormality; hence, RLS was reintroduced as a condition for which no associated neurological disease should be found [3, 4]. Ekbom’s distinction
between RLS and symptoms associated with identifiable
neurologic findings is a foreshadowing of a debate about
whether neuropathy mimics RLS or whether it causes
RLS [6].
Wherefore to some, when being abed they
betake themselves to sleep, presently in the
arms and legs, leapings and contractions
of the tendons, and so great a restlessness
and tossing of their members ensue, that the
diseased are no more able to sleep, than if
they were in the place of the greatest torture.
Penned in 1672, this description by Sir Thomas Willis,
a royal physician to King Charles I of England, may be
the earliest formal documentation of RLS [1]. Despite
this poetic entry into the medical literature, RLS then
proceeded to pass largely out of recognition until the
late 1800s when it resurfaced as “anxietas tibiarum”,
a sign of hysteria and/or neurosis [2]. It was not until 1944 that RLS regained medical creditability, when
Ekbom published an observational review of 34 cases.
In this publication, he characterized most of the syndrome’s salient clinical features; namely, a diurnal patENJ 2009; 1: (1). September 2009
41
www.slm-neurology.com
European Neurological Journal
METHODS
tion, and most population prevalence estimates suggest
a RLS prevalence of approximately 10%; therefore, this
study could also be interpreted as evidence of a lower
prevalence of RLS in patients with polyneuropathy.
The literature was examined using the PubMed search
engine. Keywords utilized were: RLS, peripheral neuropathy, review, prevalence, and genetics. As noted, the
goal was a critical review of the proposed relationship
between RLS and peripheral neuropathy (PN). Papers
were included if they were thought to have contributed
significantly to the RLS–PN literature. Furthermore,
references listed in reviewed papers were assessed and
reviewed and, accordingly, additional papers were then
reviewed based on relevant contributions to the desired
body of literature. Publication dates ranged from 1685
to 2009.
In 2006, Gemignani et al [15] examined 97 subjects
with suspected polyneuropathy, assessing for the prevalence of RLS. Neuropathic subjects were compared with
185 control subjects (full details of control selection procedures were not given). Clinical assessments were performed in all. All neuropathic patients underwent electrophysiologic evaluations; control subjects did not. The
prevalence of RLS was three times greater in patients
with acquired polyneuropathies compared with control
subjects [27/73 (37%) vs 17/185 (9%)], whereas the prevalence of RLS in possible hereditary neuropathies was
equivalent to that in control subjects [2/22 (9%) vs 17/185
(9%)]. The study concluded that RLS was frequently
found in patients with acquired polyneuropathies. One
limitation is that treatment of RLS with dopaminergic
medications was not attempted—RLS is generally exquisitely responsive to dopaminergic agents, such that
the absence of a treatment response argues against the
diagnosis [16]. Therefore, there is the consideration
that RLS could have mimicked symptoms of neuropathy [17]. Furthermore, iron studies were not performed,
suggesting that some of the “RLS” could have been
due to coexistent iron deficiency, a common comorbidity in patients with neuropathy. In 1997, Gemignani et
al [18] supported an association between RLS and acquired neuropathies when they found RLS in 33% (4/12)
of patients with neuropathy caused by essential mixed
cyroglobulinemia (EMC). Furthermore, compared with
EMC/neuropathy/non-RLS subjects, EMC/neuropathy/
RLS subjects were significantly more likely to have a
symmetrical sensory polyneuropathy, but overall neurophysiological features were not significantly different
(electrophysiologic data not given). Sural nerve biopsies
were preformed on three of the RLS and four of the nonRLS subjects, but no differences were noted. In 2007,
Gemignani et al [19] retrospectively found RLS in 33%
(33/99) of patients with documented diabetic neuropathy. Those with neuropathy and RLS were significantly more likely to have small fiber sensory dysfunction
than those without RLS [15/33 (45%) vs 15/66 (23%),
P=0.037]. Intervention with dopaminergic agents was
not attempted. However, neuropathic pain medications
were used and provided RLS relief in 11/20 of patients
(which raises the possibility that symptoms were indeed
neuropathic pain rather than true RLS). Moreover, neither of these latter studies was controlled or included
iron studies [18,19].
Early Links between RLS
and Peripheral Neuropathy
Shortly after his 1944 publication, Ekbom himself
nearly challenged his own proposed inclusion criterion
of a normal neurological examination. He described additional clinical associations with carcinoma and nutritional deficiency due to dietary restriction (dietary RLS
resulting from malabsorption was also described in 1970
by Banerji and Hurwitx [7, 8]). In 1947, Luft and Muller [9] reported RLS in a case of acute poliomyelitis. In
1966, Callaghan [10] documented RLS in five patients
with uremic neuropathy, all of whom had either clinical or electrophysiologic evidence of peripheral nerve
abnormalities. In 1967, Heinze et al [11] reported a case
of primary amyloidosis that presented as RLS and PN.
All these reports commonly suggested a possible association between RLS and PN, but all were uncontrolled,
observational reports of an anecdotal nature. In 1965,
Groman et al [12] retrospectively reviewed 27 RLS files
assessing for medical and psychosocial comorbidities.
Of the 27 subjects, 20 had coexisting symptoms of tension, anxiety, and depressive states, whereas three had
diabetes, two (8%) of whom had evidence of PN. They
concluded that RLS was most commonly associated with
mood disorders, but that similar symptoms could also be
seen in instances of neuropathy.
Case–Control Studies
in the Era of IRLS Criteria
In 1995, the official RLS diagnostic criteria were created [13], allowing direct comparison of results from
various case–control studies. The first study in the International Restless Legs Syndrome (IRLS) era was performed in 1996 by Rutkove et al [14] who examined 144
patients in a specialty neuropathy clinic for the presence
of RLS. All subjects underwent clinical history and examination, nerve conduction velocities (NCV), and electromyography (EMG). Eight (5%) screened positive for
RLS, and all of them showed signs of axonal neuropathy (severity unquantified). All eight neuropathy/RLS
patients reported a favorable response to levodopa. The
authors concluded that RLS might be associated with
polyneuropathy. However, there was no control populaENJ 2009; 1: (1). September 2009
Other studies have challenged links between RLS and
neuropathy. In 2008, Devigili et al [20] retrospectively
screened 150 patients referred for suspected sensory
neuropathy. Based on clinical examination, quantitative
sensory testing (QST), and skin biopsy examination, 83%
(124/150) had sensory neuropathy and 45% (67/124) had
42
www.slm-neurology.com
Restless Legs Syndrome and Peripheral Neuropathy—A Critical Review
pure small fiber neuropathy. All were screened for RLS
based on IRLS criteria. None of the 67 small fiber neuropathy patients were reported as fulfilling RLS criteria,
and only 4% (5/21) of those with mixed fiber neuropathy
fulfilled RLS criteria. In 2008, we reported a large case–
control study of RLS prevalence in 245 patients with PN
and 245 age- and sex-matched control subjects. Patients
were screened by telephone for symptoms of RLS by a
trained nurse, and all those who answered “yes” to three
of the four diagnostic questions were considered screen
positive. All RLS screen-positive subjects underwent a
confirmatory evaluation by a movement disorders specialist, blinded to neuropathy status. Some 26.5% of
neuropathy patients screened positive, compared with
10.2% of control subjects (P<0.0001), but the diagnosis was confirmed in only 46% of neuropathy patients
vs 80% of control subjects (P=0.005). This difference in
screen positivity and diagnostic confirmation rates was
also present in those who responded “yes” to all four criteria. After elimination of false positives, the prevalence
of RLS in neuropathy patients and control subjects was
not statistically different (12.2% vs 8.2%). However,
once neuropathy types were etiologically subclassified,
RLS was found to be significantly more likely in those
with hereditary neuropathy (22.3%) compared with control subjects (10.2 %) and those with acquired neuropathies (9.2%). Prevalence did not appear to be higher in
demyelinating HMSN I (in which positive symptoms are
generally absent). This suggests a differential effect of
neuropathy on RLS prevalence according to neuropathy
subtype [21].
ropathy actually have a family history of RLS? We could
not find evidence of this, as there was no difference in
RLS prevalence in “possible” vs “probable” hereditary
neuropathy (including genetically confirmed cases).
RLS is a highly hereditable disorder, which raises the
question of whether similar genes could underlie the
two disorders. Eight positive linkage regions for RLS
have been reported from studies of French Canadian,
Italian, American, Canadian, and German RLS families.
The BTBD9 gene on chromosome 6 and the Meis1 gene
on chromosome 2p14 confer a 50–70% increased risk of
RLS and, although much remains unknown, both are
implicated in embryonic limb formation and the development of spinal motor neuron connectivity in Drosophila.
The MAP2K5 gene may act in muscle differentiation
and/or neuroprotection of dopaminergic neurons, and
the LBXCOR1 gene may function in the development
of CNS sensory pathways. Although functional studies
are not yet complete, that these genes are implicated in
CNS sensory, spinal interneuron, and dopaminergic evolution and limb development could potentially explain
their associated presence in RLS [22–24]. Lastly, it is
conceivable that some of the genes responsible for various hereditary neuropathies could have CNS expression
and alter CNS dopamine or iron concentrations, which
could then contribute to the development of RLS. Note
that, as of the writing of this manuscript, none of the
reported chromosomal localizations for hereditary neuropathy appear to link to the chromosomal regions associated with RLS.
Reversing the Question—Peripheral
Neuropathy Prevalence in RLS
The discrepancy between RLS screen positivity and
confirmed RLS in people with neuropathy highlights
the inherent difficulties in diagnosing RLS in the presence of comorbid nerve damage. There is considerable
overlap in the symptoms of classic neuropathy and RLS.
For example, a patient with neuropathic paresthesia will
report uncomfortable sensations in the legs, and often
on cursory questioning may note exacerbation with rest
and relief with exercise (due to distraction), as well as
an escalation in symptoms at night, usually as an artifact of reduced distraction while in bed. Additionally,
cramps are painful and frequently worsen at night—on
initial questioning, those with cramps can report both
the “urge” to move (which on detailed questioning is
delineated as a need to reposition or “stretch out” the
cramp) and augmentation with rest (again as an artifact
of night-time inactivity). Careful questioning by practitioners familiar with RLS is often required to differentiate these conditions. The concern of RLS misdiagnosis
has been echoed in a recent systematic review of RLS
mimics, which includes a standardized approach to assess for RLS, to improve diagnostic reliability [16].
Another approach to study possible links between RLS
and neuropathy is to look at the question in reverse;
that is, to look for neuropathy in people with RLS. The
results of these studies may suggest intriguing links. In
1970, Harriman et al [25] looked at peripheral muscle
and nerve biopsies from 10 people with RLS. These biopsies were compared with five subjects with “RLS-plus”
(people with RLS-like symptoms who additionally suffered from “burning paresthesia”) and 10 normal control
subjects. No differences were found between the groups,
and they concluded a lack of “convincing evidence of an
organic basis for Ekbom’s syndrome”. On review, only
3/10 of the “RLS subjects” had symptomatic relief with
movement, suggesting possible diagnostic inaccuracy
(again, diagnostic criteria for RLS were not standardized until 1995, confounding all interpretation of literature published before 1995) [13]. Additional limitations
included unclear matching of groups and non-standardized biopsy techniques (samples were taken from various muscles and individual biopsy preparations varied).
The finding of increased RLS prevalence in hereditary but not acquired neuropathy is difficult to explain.
Misdiagnosis of hereditary neuropathy is one possibility—could patients who report a family history of neu-
In 1995, Iannaccone et al [26] looked for signs of axonal neuropathy in eight subjects with RLS. The study
targeted a “primary RLS” population, defined as RLS
without risk factors for neuropathy. NCV, EMG, and
www.slm-neurology.com
43
ENJ 2009; 1: (1). September 2009
European Neurological Journal
22 cases of “secondary” RLS (RLS associated with a peripheral etiological cause) with 20 patients with primary
idiopathic RLS. Using QST and quantitative nociceptor
axon reflex testing (QNART), they found significant evidence of peripheral sensory impairment in secondary
RLS but not in “primary” RLS. In contrast, in cases of
primary RLS, there were signs of CNS somatosensory
impairment not found in secondary RLS. Again, with
the caveat of possible “misdiagnosis” of RLS, this may
provide some evidence for a dichotomous cause.
quantitative thermal testing (QTT) electrophysiologic
data and sural nerve biopsies from RLS subjects were
compared with age-matched normative data. Six of eight
RLS subjects had unquantified and non-specific signs
of chronic reinnervation on EMG, and 7/8 had abnormal QTT values. Analysis of sural nerve biopsies found
RLS subjects to have had reduced nerve fiber densities
compared with matched norms. They concluded that
subjects with primary/idiopathic RLS had evidence of
peripheral nerve involvement and suggested that neurophysiologic testing be considered in routine RLS care
and management. Limitations include a small sample
size and possible lack of generalizability, considering
that all RLS patients had been treated medically for at
least a year and, hence, may have represented a severe
form of the syndrome.
It should be noted that these RLS cohort-based studies do not necessarily directly contradict the aforementioned case–control studies using neuropathy cohorts. In
the RLS-based studies, RLS patients showed evidence of
subclinical neuropathy and, hence, they would not have
been included in neuropathy clinic cohorts.
In 1996, Ondo and Jankovic [27] published a study of
54 patients with RLS (32/54 had isolated RLS and the
remaining 22/54 had other neurological conditions including Parkinson’s disease, tremor, and myoclonus).
All underwent interviews, 41/54 subjects had NVC/EMG
studies, and dopaminergic medication trials were attempted in all. Some 37% (15/54) had unquantified, unspecified abnormal electrophysiologic results, of which
47% (7/15) had clinical signs of neuropathy. A total of
92% of idiopathic RLS subjects had positive family histories vs 13% of neuropathic patients. Sporadic/neuropathic RLS cases had later onset and more rapid progression compared with familial/idiopathic RLS types,
and dopamine agents were the preferred treatment in
all. The authors suggested that there may be two major
etiological subgroups of RLS (idiopathic vs neuropathic)
that share a common pathophysiologic mechanism, possibly involving the dopaminergic system.
Future Considerations—Avoiding
Diagnostic Pitfalls
As illustrated, the proposed association between RLS
and PN has had a notably discordant history. Some of
the discrepancy can be attributed to chance, and some
explained by selection bias and differences in populations, especially variance in neuropathy subtypes. Some
inconsistencies are undoubtedly due to methodology—
in future prevalence studies, systemic sources of bias
must be addressed. Although there are standard RLS
criteria, a final diagnosis of RLS should be made exclusively by clinicians experienced in the diagnosis of RLS
and its pitfalls. Owing to the high prevalence of RLS
mimics in neuropathy, simple questionnaire screens will
almost certainly overestimate RLS prevalence. Blinding
of investigators who assign diagnosis can help to reduce
bias (although diligent clarification of symptom characteristics in patients with neuropathy will invariably lead
to inadvertent unblinding). The addition of a requirement for a positive response to dopaminergic therapy
may increase diagnostic accuracy, but as many patients
elect not to be treated, this approach may be problematic. Development of a biomarker for RLS could eliminate
potential sources of bias due to overlapping symptoms—
for example, genetic association studies linked the
BTBD9 gene to patient’s with periodic leg movements
of sleep (PLMS) without RLS, but not to patients with
RLS without PLMS [29].
In 2000, Polydefkis et al [5] further explored the association of neuropathy in RLS. They clinically, electrophysiologically (NCV, EMG, QTT), and pathologically
examined 22 subjects with RLS to look for signs of underlying polyneuropathy. Eight (36%) subjects had abnormal nerve conduction studies, intraepidermal nerve
fiber loss on skin punch biopsies, or both. In three of
these subjects, the only abnormality was intraepidermal
nerve fiber loss. RLS subjects with evidence of small
fiber disease had significantly older age of onset, higher
likelihood of pain, and decreased likelihood of a positive
family history of RLS. In conclusion, the authors again
proposed a formal schism in the diagnosis of RLS, such
that RLS should be divided into two phenotypically distinct populations: one painful, late-onset, non-familial
variant associated with small fiber polyneuropathy and
a second painless, early-onset, familial variant that does
not have concurrent neuropathy. Potential limitations
of this study included lack of matched control subjects,
lack of iron studies, and incomplete documentation of
response to dopaminergic therapy (again raising the
possibility of misdiagnosis of RLS).
CONCLUSIONS
The relationship between neuropathy and RLS is complex. There is no consistent evidence that overall RLS
prevalence is increased in individuals with clinically
diagnosed polyneuropathy. There is frequent overlap
between symptoms of RLS and neuropathy, and studies may easily be confounded by misdiagnosis. On the
other hand, there may be links between some types of
inherited neuropathy and RLS, raising the possibility of
genetic commonalities between the two conditions. In
addition, there is some preliminary evidence that a form
Finally, in 2004, Shattschneider et al [28] compared
ENJ 2009; 1: (1). September 2009
44
www.slm-neurology.com
Restless Legs Syndrome and Peripheral Neuropathy—A Critical Review
of subclinical neuropathy may exist in a proportion of
RLS patients. Further studies will be needed to clarify
these relationships, which will have important implications for the investigation and treatment of patients
with RLS.
common variants in three genomic regions. Nature Genet.
2007;39(8):1000–1006.
23. Trotti LM, Bhadriraju S, Rye DB. An update on the pathophysiology and genetics of restless legs syndrome. Curr Neurol Neurosci
Rep. 2008;8:281–287.
24. Trenkwalder C, Hogl B, Winklemann J. Recent advances in the
diagnosis, genetics and treatment of restless legs syndrome. J
Neurol. 2009;256:539–553.
25. Harriman DG, Taverner D, Woolf AL. Ekbom’s syndrome and
burning paraesthesia. Brain. 1970;93:393-406.
26. Iannaccone S, Zucconi M, Marchettini P, et al. Evidence of peripheral axonal neuropathy in primary restless legs syndrome.
Movement Disorders. 1995;10:2–9.
27. Ondo W, Jankovic J. Restless legs syndrome: clinicoetiologic correlates. Neurology. 1996;47:1435–1440.
28. Schattschneider J, Bode A, Wasner G, Binder A, Deuschl G, Baron
R, Idiopathic restless legs syndrome: abnormalities in central
somatosensory processing. J Neurol. 2004;251:977–982.
29. Stefansson H, Rye DB, Hicks A, et al. Genetic risk factors for
periodic limb movements in sleep. N Engl J Med. 2007;357:639–
647.
Disclosures: Dr Postuma has received compensation for consultation and speaking activities from Teva Neuroscience. Dr
Hattan and Dr Chalk, have no financial interests to disclose
related to the contents of this article.
REFERENCES
1. Willis T. The London Practice of Physick. 1 ed. London: Thomas
Bassett and William Crooke, 1685:1955.
2. Coccagna G, Vetrugno R, Lombardi C, Provini F. Restless legs syndrome: an historical note. Sleep Med. 2004;4:279–283.
3. Ekbom KA. Asthenia crurum paraesthetic (irritable legs). Acta
Med Scand. 1944;118:197–209.
4. Ekbom KA. Restless legs. Acta Med Scand Suppl. 1945;1:158.
5. Hening WA. Subjective and objective criteria in the diagnosis of
the restless legs syndrome. Sleep Med. 2004;5:285–292.
6. Polydefkis M, Allen RP, Hauer PM, Earley CJ, Griffin JW,
McArthur JC. Subclinical sensory neuropathy in late-onset restless legs syndrome. Neurology. 2000;55:1115–1121.
7. Ekbom KA. Restless legs syndrome. Neurology. 1960;10:868–873.
8. Banerji NK, Hurwitz LJ. Restless legs syndrome, with particular reference to its occurrence after gastric surgery. BMJ.
1970;4:774–775.
9. Luft R, Muller R. “Crampi” och “restless legs” vid akut poliomyelit. Nord Med. 1947;33:748–750.
10. Callaghan N. Restless legs syndrome in uremic neuropathy. Neurology. 1966;16:359–361.
11. Heinze EG, Frame B, Fine G. Restless legs and orthostatic hypotension in primary amyloidosis. Arch Neurol. 1967;16:497–500.
12. Gorman CA, Dyck PJ, Pearson JS. The restless leg syndrome.
Arch Intern Med. 1967,96:1571–1574.
13. Walters AS. The International Restless Legs Syndrome Study
Group, toward a better definition of the restless legs syndrome.
Movement Disorders. 1995;10(5):634–642.
14. Rutkove SB, Matheson JK, Logigian EL. Restless legs syndrome
in patients with polyneuropathy. Muscle Nerve. 1996;19(5):670–
672.
15. Gemignani F, Brindani F, Negrotti A, Vitetta F, Alfieri S, Marbini
A. Restless legs syndrome and polyneuropathy. Movement Disorders. 2006;21:1254–1257.
16. Allen RP, Picchietti D, Hening WA, Trenkwalder C, Walters AS,
Montplasir J. Restless legs syndrome: diagnostic criteria, special
considerations, and epidemiology. A report from the restless legs
syndrome diagnosis and epidemiology workshop at the National
Institutes of Health. Sleep Med. 2003;4:101–119.
17. Benes H, Walters A, Allen RP, Hening WA, Kohnen R. Definition
of restless legs syndrome, how to diagnose it, and how to differentiate it from RLS mimics. Movement Disorders. 2007;22:s401–
s408.
18. Gemignani F, Marbini A, Giovanni G, et al. Cryoglobulinaemic
neuropathy manifesting with restless legs syndrome. J Neurol
Sci. 1997;152:218–223 .
19. Gemignani F, Brindani F, Vitetta F, Marbini A, Calzetti S. Restless
legs syndrome in diabetic neuropathy: a frequent manifestation
of small fiber neuropathy. J Periph Nerv Syst. 2007;12:50–53.
20. Devigili G, Tugnoli V, Penza P, et al. The diagnostic criteria for
small fibre neuropathy: from symptoms to neuropathology. Brain.
2008;1–14.
21. Hattan E, Chalk C, Postuma RP. Is there an association between
RLS and peripheral neuropathy? Neurology. 2009;72:955–960.
22. Winkelmann J, Schormair B, Lichtner P, et al. Genomewide association study of restless legs syndrome identifies
www.slm-neurology.com
45
ENJ 2009; 1: (1). September 2009
European Neurological Journal
review article
Orthostatic Headache with and without
Cerebrospinal Fluid Leak: A Review
Andrea N. Leep Hunderfund and Bahram Mokri
Affiliation: Department of Neurology, Mayo Clinic College of Medicine, Rochester, MN, USA
Submission date: 20th June 2009, Revision date: 8th August 2009, Acceptance date: 17th August 2009
A B STRA C T
Orthostatic headache is a well-known complication of traumatic or iatrogenic dural puncture as well as overdraining
ventricular shunts for hydrocephalus. Orthostatic headache that starts spontaneously is most often due to spontaneous
cerebrospinal fluid (CSF) leak, but can also occur in the absence of CSF leak. The purpose of this article is to provide a
review of the causes, clinical characteristics, pathophysiologic mechanisms, evaluation, and management of orthostatic
headache with and without CSF leak. MEDLINE and PubMed searches were used to identify pertinent articles.
Keywords: headache, orthostatic headache, intracranial hypotension, cerebrospinal fluid leak, postural tachycardia syndrome
(POTS)
Correspondence: Bahram Mokri, Mayo Clinic, Neurology, 200 First Street SW, Rochester, MN 55905, USA. Tel: +1-507-2844036; fax: +1-507-284-4074; e-mail: bmokri@mayo.edu
INTRODUCTION
Table 1. Potential Etiologies of Orthostatic Headache
Orthostatic headache is defined as a headache that
is precipitated or significantly worsened in the upright
position and relieved or significantly improved with recumbency. The vast majority of orthostatic headaches
are due to cerebrospinal fluid (CSF) leak—traumatic,
iatrogenic, or spontaneous. While loss of CSF volume
plays a critical role in the pathogenesis of orthostatic
headache in these circumstances, the precise pathogenic
mechanism of postural headache remains controversial.
Orthostatic headache also occurs without CSF leak in a
minority of patients. Here, we review the causes, clinical
characteristics, pathophysiologic mechanisms, methods
of evaluation, and management of orthostatic headache
with and without CSF leak.
With CSF leak
Traumatic (e.g., motor vehicle accidents, sports injuries)
Iatrogenic
Lumbar puncture
Epidural catheterization
Spinal or cranial surgery
Spontaneous
Unknown cause
Weakness of the dural sac
Meningeal diverticula
Connective tissue matrix disorders
Mechanical factors
Trivial trauma
Osseous abnormalities (e.g., disk herniation,
spondylotic spur)
Without CSF leak
Occult spinal CSF leak
Increased compliance of the lower spinal CSF space
Postural tachycardia syndrome (POTS)
Mechanical or cervicogenic headache
Articles cited in this review were identified via a search
of MEDLINE and PubMed using the following search
terms: orthostatic headache, post-dural puncture headache, CSF leak, intracranial hypotension, CSF hypovolemia, CSF volume depletion, epidural blood patch, and
cervicogenic headache. The reference lists of relevant
papers were examined for additional articles of interest.
nipulations, etc.[1–5]. Iatrogenic CSF leaks may occur
following lumbar punctures, epidural catheterizations,
cranial or spinal surgeries, and other similar procedures
[6]. Although not a CSF leak in the classic sense, overdraining ventricular shunts for hydrocephalus can also
result in orthostatic headache [7–9]. These causes of orthostatic headache are often readily apparent from the
clinical history.
ETIOLOGY
Potential causes of orthostatic headache are listed in
Table 1.
Traumatic and Iatrogenic CSF Leaks
Spontaneous CSF Leaks
Traumatic CSF leaks include those resulting from motor vehicle accidents, sports injuries, manual spinal maENJ 2009; 1: (1). September 2009
CSF leaks can also occur spontaneously. Such cases are
47
www.slm-neurology.com
European Neurological Journal
Table 2. Types of Headache associated with CSF Leak
aging (MRI)). The latter also allows for delayed imaging and can reveal CSF leaks that may not be detected
by CT myelography or routine spine MRI—even when
highly T2-weighted images are obtained [24]. The sensitivity of this technique compared with CT myelography
has yet to be determined, but it can be a valuable next
step when CT myelography fails to reveal the site of
leakage. When CSF pressure is low, increasing the pressure to upper normal levels via intrathecal injection of
saline at the time of CT or MR myelography could further increase the likelihood of detecting the site of CSF
leak. This technique (referred to as positive pressure
MR myelography) has not been widely studied, however.
All these approaches, while theoretically appropriate
and sometimes successful, may still fail to reveal the
CSF leak.
Orthostatic headache (OH)
Initial non-OH (may precede typical OH by days or weeks)
Initial thunderclap headache (followed by typical OH) [32–34]
Non-orthostatic chronic daily headache [35, 36]
Chronic lingering non-OH (following months of typical OH)
Cervical or intrascapular pain (may be orthostatic) preceding OH,
accompanying OH, or without OH [37]
Second-half-of-the-day headache
Exertional headaches [38]
Intermittent headache of intermittent leak
Paradoxical postural headache (rare) [39, 40]
No headache (acephalgic form) [41]
almost always due to spontaneous leakage of CSF, usually at the level of the spine [10]. The precise cause of spontaneous spinal CSF leak is unknown in the majority of
patients [11]. A commonly suspected predisposing condition, however, is an underlying weakness of the dural
sac. A significant minority of patients with spontaneous
spinal CSF leaks display abnormalities on physical examination to suggest a disorder of connective tissue matrix (e.g., tall stature, arachnodactyly, highly arched palate, hyperextensible skin, hyperflexible joints) [12–14].
Furthermore, single or multiple meningeal diverticula
are frequently noted in patients with spontaneous CSF
leaks as well as in Marfan syndrome—a well-recognized
heritable disorder of connective tissue [15]. Mechanical factors may also play a role. Some patients describe
a traumatic event, often trivial, preceding the onset of
symptoms [16]. Finally, osseous spinal abnormalities
such as protruded disks or spondylotic spurs can rarely
puncture the dura, creating a CSF leak [17–22].
Radioisotope cisternography utilizing indium-111 allows for sequential imaging up to 48 h after initial introduction of the lumbar intrathecal tracer. The most common and most reliable abnormality is decreased activity
over the cerebral convexities at 24–48 h, which provides
indirect evidence of CSF leak. Early appearance of radioisotope in the kidneys and urinary bladder can also
provide indirect evidence of CSF leak, but one has to
be certain that this is not a consequence of inadvertent
partial epidural injection of radioisotope or “backwash”
of intrathecally injected radioisotope. The usefulness
of radioisotope cisternography for directly detecting
the site of CSF leak is relatively limited due to its poor
resolution. Furthermore, parathecal activity related to
meningeal diverticula may falsely mimic a site of CSF
leak [25].
Finally, some CSF leaks are not only slow-flowing, but
also intermittent. Although diagnostic studies may not
detect the presence or location of a leak, repeat studies
at a later date are not necessarily destined to fail if intermittent CSF leak is suspected.
Orthostatic Headache without CSF Leak
In a minority of patients with typical orthostatic headaches, extensive studies fail to reveal any direct or indirect evidence of intracranial hypotension or CSF leak
[23]. In such circumstances, etiologic possibilities include the following.
Increased Compliance of the Lower
Spinal CSF Space without Actual CSF Leak.
Another potential etiology of orthostatic headache
is increased compliance of the lower spinal CSF space
without an actual CSF leak [26]. In the upright position, this increase in compliance leads to an exaggerated
decrease in intracranial CSF pressure and therefore orthostatic headaches. These changes can occur as a result
of the increased compliance alone and do not require
any actual loss of CSF volume or decrease in the supine
lumbar CSF opening pressure. This phenomenon is addressed in more detail in the section on pathophysiologic
mechanisms of orthostatic headache.
Occult Spinal CSF Leak.
Very slow-flowing and intermittent CSF leaks may be
present in some such patients with orthostatic headache. Such leaks may evade detection at the time of evaluation—perhaps because of limitations in the resolution
of currently available imaging techniques [11].
Slow-flowing CSF leaks pose an additional diagnostic
challenge because the time interval between intrathecal
contrast administration and the acquisition of computed
tomography (CT) myelography images may be too brief
to allow for significant contrast extravasation to occur
through the slow-flowing leak [10]. This problem can be
addressed in some patients by (1) obtaining additional
delayed CT images a few hours later or (2) obtaining
magnetic resonance myelography (intrathecal injection
of gadolinium followed by spine magnetic resonance imENJ 2009; 1: (1). September 2009
Postural Tachycardia Syndrome (POTS).
POTS is a disorder typically seen in young women and
characterized by an increase in heart rate of over 30
beats per minute upon standing to an absolute rate of
at least 120 beats per minute [27]. The precise cause of
48
www.slm-neurology.com
Orthostatic Headache with and without Cerebrospinal Fluid Leak: A Review
Table 3. Clinical Manifestations of CSF Leak
POTS is unknown, but potential mechanisms include a
limited autonomic neuropathy (with peripheral denervation but intact cardiac autonomic innervation, poor
vasomotor tone, and venous pooling), beta receptor supersensitivity, reduced ability of the vagus nerve to slow
the heart, hypovolemia, disturbed sympathetic–parasympathetic balance, or disturbed brainstem mechanisms [28]. Typical manifestations of POTS include dizziness, decreased concentration, tremulousness, nausea,
and near-syncope or syncope in the upright position.
Rarely, orthostatic headache can be the most prominent
feature of the clinical presentation [28]. On the other
hand, patients with longstanding or intractable orthostatic headache may secondarily develop orthostatic intolerance due to a combination of prolonged bed rest,
hypovolemia, and deconditioning. Orthostatic headache
in this circumstance should not be attributed to POTS.
Common [11, 49]
Headache
Posterior neck pain or stiffness
Interscapular pain, less commonly low back pain [37]
Nausea with or without vomiting
Altered hearing (echoed, distant, muffled) or hearing loss [50, 51]
Disturbed sense of balance or dizziness
Photophobia
Visual blurring
Horizontal diplopia (unilateral or bilateral cranial nerve VI palsy)
[35, 52–56]
Uncommon
Non-horizontal diplopia due to cranial nerve III or IV palsies
[57–63]
Encephalopathy, obtundation, stupor, coma [64–71]
Visual field defects (superior binasal) [72]
Upper limb numbness, paresthesias, aches, radiculopathy [73–75]
Facial pain, numbness, weakness, or spasm [76, 77]
Ménière’s disease-like syndrome (labyrinthine hydrops) [78]
Frontotemporal dementia [79–81]
Parkinsonism, ataxia, bulbar manifestations [82]
Dorsal midbrain syndrome (episodic stupor and vertical gaze
palsy) [83]
Gait unsteadiness [84]
Difficulties with bowel and bladder control [85]
Quadriplegia [86]
Chorea [87]
Galactorrhea [88]
Decreased growth hormone secretion [89]
Amnestic syndrome [90]
Psychic akinesia (hypoactive, hypoalert behavior) [91]
Transtentorial herniation [92]
Acute respiratory failure [93]
Orthostatic Cervicogenic Headache.
Pain that originates from structures in the neck has
been increasingly recognized as a cause of headache.
Such headaches are thought to be referred head pains
due to the convergence of sensory input via the upper
and midcervical nerve roots on the descending tract of
V in the cervical spinal cord [29]. The neck muscles,
facet and uncovertebral joints, intervertebral disks, and
ligaments have each been implicated in the pathogenesis of cervicogenic headaches [29]. All these structures
also play a critical role in supporting the head in the
upright position. An obligatory diagnostic criterion for
cervicogenic headache is precipitation of symptoms by
neck movement or head positioning [30]. In some cases,
sitting or standing may be the specific position that triggers a cervicogenic headache. A cervicogenic headache
with prominent orthostatic features may therefore be
encountered [31].
experience has shown that the time to headache onset or
worsening after sitting or standing may be substantially
longer, however [43], and not all patients respond to epidural blood patching [44]. Improvement of the headache
after lying down often occurs within 30 min, although
this is also variable [11].
CLINICAL CHARACTERISTICS
Onset of orthostatic headache related to iatrogenic
dural puncture is usually temporally related to the precipitating event or procedure. For example, 90% of postdural puncture headaches occur within 3 days of the
procedure, and 66% start within the first 48 h [45, 46].
Onset of the initial headache in spontaneous spinal CSF
leak is variable. It can be abrupt (e.g., thunderclap [32]),
subacute reaching maximal intensity over minutes to
hours [11], or more insidious with the orthostatic features only becoming apparent over time.
Headache is the most common manifestation of CSF
leak. The headache is often (but not always) orthostatic.
For this reason, the clinical characteristics of orthostatic headache have been best described in the setting of
traumatic or spontaneous CSF leaks.
Other headache types reported in association with
CSF leak are summarized in Table 2. The orthostatic
headache and associated symptoms in patients with and
without CSF leak show substantial clinical similarity.
For this reason, clinical features alone are unlikely to
reliably differentiate between the two groups [23].
Orthostatic headaches related to traumatic or spontaneous CSF leaks may be frontal, frontotemporal, frontooccipital, occipital, or holocephalic in location [47]. They
are typically bilateral, dull in quality, and non-throbbing.
Occasionally, they may begin as a focal or unilateral headache and can evolve into a holocephalic headache if the
patient remains upright. The headaches are often aggravated by Valsalva maneuvers such as coughing, sneezing,
or straining. Similar headache characteristics have been
observed in patients without CSF leak [23, 28].
The defining feature of orthostatic headache is its
postural nature. This is reflected in the 2004 International Classification of Headache Disorders, 2nd edition
(ICHD-II) definition of headache related to spontaneous
intracranial hypotension. The orthostatic headache is
described as a headache that worsens within 15 min of
sitting or standing and resolves within 72 h of epidural
blood patching [42]. In spontaneous CSF leaks, clinical
www.slm-neurology.com
49
ENJ 2009; 1: (1). September 2009
European Neurological Journal
The severity of the headache varies considerably. Many
mild cases likely go undiagnosed. Approximately 85% of
post-dural puncture headaches are mild to moderate in
severity and can be adequately treated with common analgesics [48]. On the other hand, some patients are quite
incapacitated by orthostatic headache.
Orthostatic headaches with and without CSF leak also
have similar associated symptoms [23]. These include
posterior neck pain or stiffness, nausea and vomiting,
and cochleovestibular complaints such as altered hearing, tinnitus, dizziness, or a disturbed sense of balance
[11, 49]. Some of these symptoms may also be orthostatic in nature. Other symptoms associated with spontaneous CSF leaks are outlined in Table 3. Many of these
are also seen in traumatic CSF leaks, and a few can also
be seen in orthostatic headache without evidence of CSF
leak.
Figure 1. A
. Normal. The hydrostatic indifferent point (HIP) is
the location along the CSF axis where CSF pressure is
equal in the upright and horizontal positions. It is usually
located somewhere between the spinous processes of
C7 and T5. B. Increased compliance of the lower spinal
CSF space. The HIP shifts caudally, making the already
negative intracranial CSF pressure more negative and
the lumbar CSF pressure less positive than expected.
Compensatory dilatation of pain-sensitive intracranial
venous structures causes orthostatic headaches. Used
with permission from the Mayo Foundation
PATHOPHYSIOLOGIC MECHANISMS
The exact mechanism of orthostatic headache in CSF
leak is unknown. There have been at least four theories
to date.
Dilatation of Intracranial Venous Structures
The first theory is that CSF volume depletion causes
compensatory dilatation of intracranial venous structures. According to the Monroe–Kellie hypothesis, the
sum of the volumes of brain, CSF, and intracranial blood
remains constant inside the rigid skull. Therefore, a decrease in one should cause a reciprocal increase in either
or both of the remaining two [94]. The intracranial venous structures are pain-sensitive, and their dilatation
in turn may lead to headache.
pliance of the lower spinal CSF space is independent of
CSF hypotension or hypovolemia and, as such, may be
the unifying pathophysiologic mechanism of orthostatic
headache with and without CSF leak (with the exception of those due to mechanical or cervicogenic factors).
The hydrostatic indifferent point is a location along
the CSF axis where CSF pressure is equal in the upright
and supine positions. It is usually located somewhere between the spinous processes of C7 and T5 [100]. The CSF
above the hydrostatic indifferent point can be thought
of as effectively suspended from the cranial vault (creating a negative pressure cranially), while the CSF below
this point can be thought of as effectively resting on the
lower dural sac (creating a positive pressure caudally).
Sinking of the Brain
The second theory implicates sinking of the brain as
a consequence of CSF volume depletion. This causes
stretching of pain-sensitive suspending structures and
vessels in the upright position and hence orthostatic
headache [35, 95].
Sometimes, patients with spontaneous CSF leaks and
orthostatic headache have normal CSF opening pressures [96]. Thus, CSF volume loss (rather than CSF
hypotension) is thought to be the core pathogenic factor in both the above theories. In post-dural puncture
headaches, however, the degree of CSF leakage has not
been found to correlate with headache severity [97, 98].
Head MRI has also failed to show increased sinking of
the brain in the upright position [99].
When the compliance of the lower spinal CSF space
is increased, the hydrostatic indifferent point shifts
caudally in the direction of increased compliance [26].
Thus, in the upright position, more CSF is suspended
from the cranial vault, and the already negative intracranial CSF pressure becomes more negative (Figure 1).
When intracranial CSF pressure becomes more negative, compensatory dilatation of pain-sensitive intracranial venous structures occurs—again in accordance with
the Monroe–Kellie hypothesis—resulting in orthostatic
headache.
Increased Compliance of
the Lower Spinal CSF Space
A third theory is that caudal movement of the hydrostatic indifferent point in the upright position due
to increased compliance of the lower spinal CSF space
accounts for orthostatic headache [26]. Increased comENJ 2009; 1: (1). September 2009
Potential Mechanisms of Increased Compliance.
Levine and Rapalino [26] argue that compliance is increased in orthostatic headache with CSF leak when a
50
www.slm-neurology.com
Orthostatic Headache with and without Cerebrospinal Fluid Leak: A Review
dural hole or tear exposes CSF to the more compliant
surrounding tissues and structures. This is compounded
by a decrease in CSF volume, as a partially collapsed
lower thecal sac is more compliant than a normally distended one.
ed by cough but only in the upright position). Lumbar
CSF pressures in these two patients and in 25 normal
control subjects were measured in the supine, seated,
and standing positions. In the normal control subjects,
lumbar CSF pressures increased in the sitting relative
to the supine position and increased even further in the
standing position. In contrast, lumbar CSF pressures in
patients with orthostatic cough headache dropped in the
standing position compared with the sitting position—
reflecting increased compliance of the lower spinal CSF
space and inferior movement of the hydrostatic indifferent point.
In orthostatic headaches without CSF leak, increased
dural compliance alone occurs without any actual loss
of CSF volume [23]. Disorders of connective tissue matrix can increase the inherent distensibility of the dura.
Dilation of the dural sac has been demonstrated in patients with Marfan syndrome, a known disorder related
to abnormality of elastin and fibrillin. Over 90% of patients with Marfan syndrome have dural ectasia on lumbosacral MRI that increases in severity with increasing
patient age [101]. This observation is attributed to the
effect over time of lumbar CSF pressure in the upright
position, which progressively dilates what is an inherently more distensible thecal sac. Consistent with this,
clinical stigmata of an underlying disorder of connective
tissue matrix have been documented in two of six patients with orthostatic headache without CSF leak [23].
Epidural Hypotension
A final theory is that of epidural hypotension. Franzini et al [105] report that a markedly negative lumbar
epidural pressure was usually observed in their series
of patients with spontaneous intracranial hypotension.
They thus postulate that negative pressure within the
inferior vena cava (IVC) results in overdrainage of venous blood from the epidural spinal vein network via
large radicular veins through one-way valves. Decreased
lumbar epidural pressure and volume alter the gradient
between epidural pressure and intradural CSF pressure,
causing CSF to be aspirated into the epidural space and
veins. This process may be facilitated by the presence of
meningeal diverticula or intrinsic dural weakness. Focal areas of CSF aspiration may account for visible CSF
leaks, whereas diffuse aspiration along the dural surface
may account for orthostatic headaches without apparent CSF leak.
In orthostatic headache related to POTS, peripheral
venous pooling upon standing may decrease blood volume in the epidural venous plexus [28], resulting in an
orthostatic drop in lumbar CSF pressure due to caudal
movement of the hydrostatic indifferent point.
Demonstration of Caudal Movement of the
Hydrostatic Indifferent Point by Lumbar Puncture.
When the hydrostatic indifferent point shifts caudally,
not only does intracranial CSF pressure become more
negative, but the lumbar CSF pressure also becomes
less positive (Figure 1). The latter phenomenon can be
demonstrated by measuring lumbar CSF pressure in the
upright position. The lumbar CSF pressure in this position is found to be lower than expected and has been
referred to as an orthostatic drop in lumbar CSF pressure [102].
During standing and walking, the limb muscles actively pump venous blood from the peripheral towards the
central veins. This promotes drainage of smaller tributary veins (including the lumbar epidural veins) and
makes lumbar epidural pressure even more negative.
This would not only increase the amount of aspirated
CSF in the upright position, but also increase the overall
compliance of the lower spinal CSF space with resulting
caudal movement of the hydrostatic indifferent point
and headache in the upright position.
Orthostatic drops in lumbar CSF pressure have been
demonstrated in patients with orthostatic headache
both with and without CSF leak. Levine and Rapalino
[26] describe a patient with orthostatic headache due to
spontaneous spinal CSF leak in whom the lumbar CSF
pressure was normal at 90 mm of water in the supine
position but lower than expected at 280 mm of water
in the sitting position. (Lumbar CSF pressure in the
sitting position normally ranges from 320 to 630 mm
of water depending on the length of the torso with the
top of the CSF fluid column in the manometer normally
reaching somewhere between the occipital protuberance
and the spinous process of T2 [100].) Similar findings
are reported by Kunkle et al and by Nelson in patients
with post-lumbar puncture headaches [103, 104].
EVALUATION
On evaluation of patients with orthostatic headaches,
information should be obtained about any preceding
trauma, procedure with the potential for dural puncture, craniospinal surgery, or shunting for hydrocephalus. Personal or family history pointing to a disorder of
connective tissue matrix (e.g., tall stature, arachnodactyly, highly arched palate, hyperextensible skin, hyperflexible joints, mitral valve prolapse, retinal detachment,
aortic or intracranial aneurysms, carotid or vertebral
artery dissections) should be sought. The patient should
be queried regarding symptoms of orthostatic intolerance in the upright position (e.g., dizziness, decreased
concentration, tremulousness, nausea, near-syncope or
syncope). Physical examination should also assess for
Bono et al [102] also demonstrated orthostatic drops
in lumbar CSF pressure in two patients with orthostatic
cough headache without CSF leak (headache precipitatwww.slm-neurology.com
51
ENJ 2009; 1: (1). September 2009
European Neurological Journal
Table 4. Diagnostic Studies in Cerebrospinal Fluid (CSF) Leak
Diagnostic study
Characteristic findings in CSF leak
Notes
Brain magnetic resonance
imaging (MRI)
Pachymeningeal enhancement, pituitary hyperemia
and enlargement, subdural fluid collections, brain
sag, ventricular collapse, venous engorgement [35,
95, 107–118]
Can be normal in patients with documented spinal
CSF leak [116–118]; MRI abnormalities often improve with clinical improvement
Spine MRI
Spinal pachymeningeal enhancement (usually but
not always cervical), dilated epidural and occasionally intradural spinal veins, extra-arachnoid or
extradural CSF collections, meningeal diverticula,
nerve root sleeve ectasia [10, 25, 119–123]
Presence of extra-arachnoid or extradural fluid
collections can help identify the level of a CSF leak
(cervical, thoracic, lumbar) but very rarely identifies
the actual site of CSF leak [25]
Radioisotope cisternography
Delayed ascent of the tracer to the convexities, paucity of activity over the cerebral convexities on 24-h
images, early appearance of the tracer in the kidneys
and urinary bladder, parathecal activity at the level
or approximate site of leak (less common) [124–128]
Indium-111 is the radioisotope of choice; repeat
images can be obtained up to 48 h after radioisotope
injection; large meningeal diverticula may appear as
foci of parathecal activity that cannot be reliably distinguished from actual sites of CSF leak; extrathecal
injection or extravasation from intrathecal injection
can mimic parathecal activity resulting from CSF
leak and cause very early appearance of tracer in
kidneys and urinary bladder [25]
Myelography
Extradural extravasation of contrast, meningeal
diverticula, nerve root sleeve ectasia [85]
Computed tomography (CT) myelography is the
most reliable test to reveal the precise location of
the CSF leak or leaks [85]; high-speed multidetector
spiral CT (e.g., “dynamic CT myelography”) [129]
or digital subtraction myelography [130] may be
required to locate fast-flow CSF leaks; MR myelography using intrathecal gadolinium can also be
performed and may help to identify slow-flow CSF
leaks [24, 131–134]; both CT and MR myelography
can be done under positive pressure*
Lumbar CSF opening
pressure
Often low (less than 60 mm of water), occasionally
unmeasurable or even negative
Can be persistently normal in patients with documented symptomatic spinal CSF leak [96]
CSF analysis†
Protein may be normal or elevated (concentrations
of up to 100 mg/dL are not uncommon), glucose is
never low, leukocytes are normal or elevated with a
lymphocytic pleocytosis, erythrocyte count is normal
or elevated, CSF appearance can be clear or xanthrochromic, cytology is always normal, and microbiology always negative
Protein concentrations of up to 100 mg/dL are not
uncommon and concentrations as high as 1000 mg/
dL have been reported [135]; leukocyte elevations of
up to 50 cells/mm2 common and up to 222 cells/mm2
has been reported [35]; difficult and traumatic taps
common in CSF leak and dilation of epidural venous
plexus also increases likelihood of obtaining bloodtinged CSF [25]
*Elevating the CSF pressure to high normal levels through intrathecal injection of saline when CSF pressure is low.
†All CSF findings may vary in the same patient on different samples.
potential stigmata of a connective tissue matrix disorder and include measurements of heart rate and blood
pressure in the supine and upright positions. Placing patients in the Trendelenburg position (10–20 degrees of
head-down tilt) for 5 min to see if this alleviates or substantially improves their headache may also be a useful
screening test for intracranial hypotension, although it
is not diagnostic of this [106].
flex testing should be considered. Finally, an orthostatic
drop in lumbar CSF pressure in the upright position
might be demonstrated in some patients with orthostatic headache without CSF leak. This is not routinely
done in clinical practice. Furthermore, data on normal
values are scarce.
MANAGEMENT
In the majority of patients with spontaneous development of orthostatic headaches, diagnostic studies eventually reveal direct or indirect evidence of intracranial
hypotension or spinal CSF leak. Such studies include
brain MRI, spine MRI, radioisotope cisternography, CT
or MR myelography, and lumbar puncture. Characteristic findings in CSF leak for each study are outlined in
Table 4. In patients with prominent orthostatic intolerance—especially if young and female—autonomic reENJ 2009; 1: (1). September 2009
Most orthostatic headaches following iatrogenic dural
puncture resolve without treatment within 1 week [6,
49]. Spontaneous recovery can also occur in patients
with orthostatic headache due to spontaneous CSF leak.
The frequency with which this occurs is unknown. Patients encountered at tertiary care centers as referrals
now are frequently those with persistent symptoms,
past treatment failures, and atypical features.
52
www.slm-neurology.com
Orthostatic Headache with and without Cerebrospinal Fluid Leak: A Review
Traumatic or post-surgical CSF leaks may call for surgical correction. Well-selected cases of spontaneous CSF
leak can also be effectively treated with surgery when
conservative management and less invasive approaches
(such as EBP) have failed [155, 156] Thorough preoperative studies to identify the actual site of CSF leak are
critical, and the dural defects encountered may be complex [157]. Orthostatic headache due to overdraining
CSF shunts may likewise require surgical shunt valve
replacement or modification.
Conservative management of orthostatic headache involves bed rest, hydration, caffeine or theophylline administration, and abdominal binders—although with a
variable evidence base [48] and with only limited expectations of success in many cases. When such measures
fail, treatment of orthostatic headache involves selecting more targeted strategies.
In the majority of CSF leaks, epidural injection of autologous blood—commonly known as an epidural blood
patch (EBP)—is the treatment of choice [136]. EBPs can
provide relief through immediate and latent effects. The
immediate effect is through the creation of a dural tamponade, decreasing the volume and perhaps the compliance of the dural sac. The latent effect of EBP is related
to a tissue reaction provoked by the blood, which it is
hoped will seal the leak. The efficacy of EBP in iatrogenic post-dural puncture headaches is impressive, with
approximately 90% of patients achieving relief after the
first EBP and almost all after a second EBP [137]. In
contrast, many patients with spontaneous CSF leaks require more than one EBP, and the efficacy of each EBP
is about 30% [44].
When an occult CSF leak (e.g., slow flow or intermittent) is suspected, a trial of lumbar EBP is reasonable.
Any benefit from EBPs in orthostatic headache without
CSF leak is at best expected to be quite transient and
is likely related to the temporary decrease in lumbar
dural compliance created by the dural tamponade [23].
Durable methods of reducing compliance in the lower
dural sac merit are needed. Targeted EBPs or fibrin glue
injections are typically applied to already detected, confirmed, or highly suspected sites of CSF leak. As such,
they do not have an established role in the management
of orthostatic headache without CSF leak.
This discrepancy in efficacy is likely to the result of
several factors. Many spontaneous CSF leaks are due to
defects in the anterior aspect of the dura or in nerve root
sleeves with or without weeping meningeal diverticula,
whereas EBPs are placed posteriorly. Furthermore,
EBPs are often placed distant from the CSF leak—of
which there may be one or multiple. Finally, dural defects in spontaneous CSF leaks are often not simple
holes but rather congenitally attenuated zones of dura
with unsupported underlying arachnoid that is oozing
CSF from one or more sites [25].
One report describes partial improvement of symptoms
in a patient with spinal CSF leak following resection of a
strip of dura in the lower thecal sac, reducing its volume
[158]. The wisdom and value of such approaches have
not been established.
When orthostatic headache is related to POTS, treatment recommendations do not differ significantly from
those given to patients with more typical manifestations
of orthostatic intolerance. Such recommendations include oral hydration, increased salt intake, and a graded
exercise program. Wearing an abdominal binder may be
particularly helpful in patients with POTS, as doing so
increases intra-abdominal venous pressure. This pressure increase is presumably transmitted to the spinal
veins, which may in turn decrease the compliance of
the lower spinal CSF space and thus relieve orthostatic
headaches [28].
Targeted EBPs are somewhat (although not substantially) more effective but require diagnostic studies to
locate the site of CSF leakage [138–142] Epidural injections of fibrin glue [143–148] can also be considered.
These can be particularly helpful for small zones of CSF
leak that are well localized through diagnostic studies
and when the objective is to inject a smaller volume epidurally. Injection of fibrin glue mixed with homologous
blood has also been tried, although the mixture may become quite thick and create technical difficulties [105].
Additional proposed treatment options include epidural
infusion of saline [149, 150], epidural infusion of colloids (e.g., dextran) [151, 152], or intrathecal infusions
of fluid [65].
Finally, in orthostatic cervicogenic headaches, management efforts are often focused on identifying and addressing the underlying neck pathology generating the
headache. As such, possible treatment options include
cervical spine surgery, physical therapy, cervical spine
manipulation, or injections of related cervical structures with anesthetic or anti-inflammatory agents [29].
Symptomatic benefit is helpful in confirming the diagnosis of cervicogenic headache, and lack of benefit should
prompt reconsideration of the diagnosis [30, 159].
Lasting benefit after cessation of many of these approaches seems unlikely [6, 25] Prolonged infusions
also raise concerns regarding the risk of infection. Experience with both epidural and intrathecal infusions is
limited. The former should only be considered in refractory cases. The latter may have a role in the emergent
management of potentially life-threatening intracranial
hypotension (impending coma related to sinking of the
brain with brainstem compression or compromise) [153,
154].
www.slm-neurology.com
CONCLUSIONS
Orthostatic headache has long been recognized as a
potential complication of iatrogenic dural puncture and
can similarly complicate traumatic dural injuries or
overdraining CSF shunts. Spontaneous-onset orthostatic headaches have more recently been recognized as
53
ENJ 2009; 1: (1). September 2009
European Neurological Journal
the most common manifestation of spontaneous CSF
leaks—usually at the level of the spine. It is important
to emphasize, however, that spontaneous CSF leaks can
cause non-orthostatic headache or no headache at all.
Furthermore, not all orthostatic headaches are due to
CSF leak, although the clinical characteristics of orthostatic headache with and without CSF leak are often
similar.
10.
11.
12.
Loss of CSF volume plays a critical role in the pathogenesis of orthostatic headache with CSF leak. While a
unifying pathophysiologic change underlying orthostatic headache with and without CSF leak is sought, more
than one may exist. Caudal movement of the hydrostatic
indifferent point in the upright position due to increased
compliance of the lower spinal CSF space with a resulting orthostatic drop in lumbar CSF pressure may be the
underlying mechanism in most cases, however.
13.
Evaluation of patients with spontaneous-onset orthostatic headache should be aimed at identifying an underlying CSF leak, which is present in the majority of patients. When conservative measures fail, the mainstay
of orthostatic headache management due to spinal CSF
leak is the EBP. EBPs are more effective for iatrogenic
post-dural puncture headaches, less so for spontaneous
CSF leaks, and may provide only transient relief in orthostatic headache without leak. In proven CSF leaks,
surgical repair remains an option in carefully selected
cases.
16.
14.
15.
17.
18.
19.
20.
Disclosure: The authors have no financial interests to disclose related to the contents of this article.
21.
REFERENCES
1.
2.
3.
4.
5.
6.
7.
8.
9.
22.
Buchanan RJ, Brant A, Marshall LF. Traumatic cerebrospinal
fluid fistulas. In: Winn HR, ed. Youmans Neurological Surgery.
5th ed. Philadelphia: W.B. Saunders Co; 2004:5265–5272.
Morelli N, Gallerini S, Gori S, et al. Intracranial hypotension
syndrome following chiropractic manipulation of the cervical
spine. J Headache Pain. 2006;7(4):211–213.
Matthews MK, Frohman L, Lee HJ, et al. Spinal fluid leak after
chiropractic manipulation of the cervical spine. Arch Ophthalmol. 2006;124(2):283.
Donovan JS, Kerber CW, Donovan WH, Marshall LF. Development of spontaneous intracranial hypotension concurrent with
grade IV mobilization of the cervical and thoracic spine: a case
report. Arch Phys Med Rehabil. 2007;88(11):1472–1473.
Huntoon MA, Watson JC. Intracranial hypotension following
motor vehicle accident: an overlooked cause of post-traumatic
head and neck pain? Pain Pract. 2007;7(1):47–52.
Turnbull DK, Shepherd DB. Post-dural puncture headache:
pathogenesis, prevention, and treatment. Br J Anaesth.
2003;91(5):718–729.
Jackson IJ, Snodgrass SR. Peritoneal shunts in the treatment
of hydrocephalus and increased intracranial pressure: a 4-year
study of 62 patients. J Neurosurg. 1955;12(3):216–222.
Fox JL, McCullough DC, Green RC. Effect of cerebrospinal fluid
shunts on intracranial pressure and on cerebrospinal fluid dynamics. 2. A new technique of pressure measurements: results
and concepts. 3. A concept of hydrocephalus. J Neurol Neurosurg Psychiatry. 1973;36(2):302–312.
Rosenthal G, Pomeranz S, Spektor S, Yacoub M, Israel ZH.
Syndrome of overdrainage associated with disconnection of a
ENJ 2009; 1: (1). September 2009
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
54
ventriculoperitoneal shunt. Pediatr Neurosurg. 1999;31(3):124–
126.
Mokri B. Headaches caused by decreased intracranial pressure:
diagnosis and management. Curr Opin Neurol. 2003;16(3):319–
326.
Schievink WI. Spontaneous spinal cerebrospinal fluid leaks.
Cephalalgia. 2008;28(12):1347–1356.
Mokri B, Maher CO, Sencakova D. Spontaneous CSF
leaks: underlying disorder of connective tissue. Neurology.
2002;58(5):814–816.
Schievink WI, Gordon OK, Tourje J. Connective tissue disorders with spontaneous spinal cerebrospinal fluid leaks and
intracranial hypotension: a prospective study. Neurosurgery.
2004;54(1):65–70.
Schriver I, Schievink WI, Godfrey M, et al. Spontaneous
spinal cerebrospinal fluid leaks and minor skeletal features
of Marfan’s syndrome: a microfibrillopathy. J Neurosurg.
2002;96(3):483–489.
Davenport RJ, Chataway SJ, Warlow CP. Spontaneous intracranial hypotension from a CSF leak in a patient with Marfan’s
syndrome. J Neurol Neurosurg Psychiatry. 1995;59(5):56–59.
Schievink WI, Louy C. Precipitating factors of spontaneous
spinal CSF leaks and intracranial hypotension. Neurology.
2007;69(7):700–702.
Vishteh AG, Schievink WI, Baskin JJ, Sonntag VK. Cervical
bone spur presenting with spontaneous intracranial hypotension. Case report. J Neurosurg. 1998;89(3):483–484.
Winter SC, Maartens NF, Anslow P, Teddy PJ. Spontaneous
intracranial hypotension due to thoracic disc herniation. Case
report. J Neurosurg. 2002;96(3 Suppl):343–345.
Eross EJ, Dodick DW, Nelson KD, Bosch P, Lyons M. Orthostatic
headache syndrome with CSF leak secondary to bony pathology
of the cervical spine. Cephalalgia. 2002;22(6):439–443.
Rapport RL, Hillier D, Scearce T, Ferguson C. Spontaneous intracranial hypotension from intradural thoracic disc herniation.
Case report. J Neurosurg. 2003;98(3 Suppl):282–284.
Binder DK, Sarkissian V, Dillon WP, Weinstein PR. Spontaneous
intracranial hypotension associated with transdural thoracic
osteophyte reversed by primary dural repair. Case report. J
Neurosurg Spine. 2005;2(5):614–618.
Yokota H, Yokoyama K, Noguchi H, et al. Thoracic osteophyte
causing spontaneous intracranial hypotension. Cephalalgia.
2008;28(4):396–398.
Leep Hunderfund AN, Mokri B. Orthostatic headache without
CSF leak. Neurology. 2008;71(23):1902–1906.
Katramados A, Patel SC, Mitsias PD. Non-invasive magnetic
resonance myelography in spontaneous intracranial hypotension. Cephalalgia. 2006;26(9):1160–1164.
Mokri B. Low-cerebrospinal fluid volume headaches. In: Goadsby PJ, Silberstein SD, Dodick DW, eds. Chronic Daily Headache
for Clinicians. Hamilton: BC Decker; 2005:155–166.
Levine DN, Rapalino O. The pathophysiology of lumbar puncture headache. J Neurol Sci. 2001;192(1–2):1–8.
Low PA, Schondorf R, Novak V, et al. Clinical Autonomic
Disorders: Evaluation and Management, 2nd edn. Low PA, ed.
Philadelphia: Lippincott-Raven; 1997:681–697.
Mokri B, Low PA. Orthostatic headaches without CSF leak in
postural tachycardia syndrome. Neurology. 2003;61(7):980–982.
Haldeman S, Dagenais S. Cervicogenic headaches: a critical
review. Spine J. 2001;1(1):31–46.
Sjaastad O, Fredriksen TA, Pfaffenrath V. Cervicogenic headache: diagnostic criteria. The Cervicogenic Headache International Study Group. Headache. 1998;38(6):442–445.
Kim JH, Choi JY, Kim HJ, Oh K. Orthostatic headache as the
presenting symptom of cervical spine metastasis. Headache.
2008;48(1):161–163.
Schievink WI, Wijdicks EFM, Meyer FB, Sonntag VKH.
Spontaneous intracranial hypotension mimicking aneurysmal
subarachnoid hemorrhage. Neurosurgery. 2001:48(3):513–517.
Ferrante E, Savino A. Thunderclap headache caused by spon-
www.slm-neurology.com
Orthostatic Headache with and without Cerebrospinal Fluid Leak: A Review
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
taneous intracranial hypotension. Neurol Sci. 2005;26 Suppl
2:s155–s157.
Balgera R, Rigamonti A, Sozzi G, Agostoni E. An atypical
case of spontaneous intracranial hypotension. Neurol Sci.
2009;30(1):71–73.
Mokri B, Piepgras DG, Miller GM. Syndrome of orthostatic
headaches and diffuse pachymeningeal gadolinium enhancement. Mayo Clin Proc. 1997;72(5):400–413.
Kong DS, Park K, Nam DH, et al. Atypical spontaneous
intracranial hypotension (SIH) with nonorthostatic headache.
Headache. 2007;47(2):199–203.
Dunbar SA, Katz NP. Post-dural puncture thoracic pain without
headache: relief with epidural blood patch. Can J Anaesth.
1995;42(3):221–223.
Mokri B. Spontaneous CSF leaks mimicking benign exertional
headaches. Cephalalgia. 2002;22(10):780–783.
Mokri B, Aksamit AJ, Atkinson JL. Paradoxical postural headaches in cerebrospinal fluid leaks. Cephalalgia.
2004;24(10):883–887.
Liu H, Kaye A, Comarda N, Li M. Paradoxical postural cerebrospinal fluid leak-induced headache: report of two cases. J Clin
Anesth. 2008;20(5):383–385.
Mokri B, Atkinson JL, Piepgras DG. Absent headache despite
CSF volume depletion (intracranial hypotension). Neurology.
2000;55(11):1722–1724.
Headache Classification Subcommittee of the International
Headache Society. The International Classification of Headache
Disorders, 2nd edn. Cephalalgia. 2004;24 Suppl 1:9–160.
Mea E, Chiapparini L, Savoiardo M, et al. Application of HIS
criteria to headache attributed to spontaneous intracranial
hypotension in a large population. Cephalalgia. 2009;29(4):418–
422.
Sencakova D, Mokri B, McClelland RL. The efficacy of epidural blood patch in spontaneous CSF leaks. Neurology.
2001;57(10):1921–1923.
Leibold RA, Yealy DM, Coppola M, Cantees KK. Post-duralpuncture headache: characteristics, management, and prevention. Ann Emerg Med. 1993;22(12):1863–1870.
Reynolds F. Dural puncture and headache. Br Med J.
1993;306(6882):874–876.
Mokri B. Spontaneous intracranial hypotension spontaneous
CSF leaks. Headache Currents. 2005;2(1):11–22.
Harrington BE. Postdural puncture headache and the development of the epidural blood patch. Reg Anesth Pain Med.
2004;29(2):136–163.
Vandam LD, Dripps RD. Long-term follow up of patients who
received 10 098 spinal anesthetics; syndrome of decreased
intracranial pressure (headache and ocular and auditory difficulties). JAMA. 1956;161(7):586–591.
Lybecker H, Andersen T. Repetitive hearing loss following dural
puncture treated with autologous epidural blood patch. Acta
Anaesthesiol Scand. 1995;39(7):987–989.
Taki M, Nin F, Hasegawa T, et al. Case report: two cases of hearing impairment due to intracranial hypotension. Auris Nasus
Larynx. 2009;36(3):345–348.
Arjona A, Corral F, Mata M, et al. Paralysis of the 6th cranial
nerve secondary to spontaneous intracranial hypotension. Rev
Clin Esp. 1996;196(3):201–202.
Grueb M, Besch D, Mielke J, et al. Abducens nerve paresis in
spontaneous intracranial hypotension syndrome. Acta Ophthalmol Scand. 2005;83(2):269–270.
Guigon B, Tailla H, de la Marnière E, et al. Abducens nerve
palsy in spontaneous intracranial hypotension. J Fr Ophthalmol. 2004;27(4):392–396.
Kurbanyan K, Lessell S. Intracranial hypotension and abducens
palsy following upper spinal manipulation. Br J Ophthalmol.
2008;92(1):153–155.
Khemka S, Mearza AA. Isolated sixth nerve palsy secondary to spontaneous intracranial hypotension. Eur J Neurol.
2006;13(11):1264–1265.
www.slm-neurology.com
57. Brady-McCreery KM, Speidel S, Hussein MA, Coats DK. Spontaneous intracranial hypotension with unique strabismus due
to third and fourth cranial neuropathies. Binocul Vis Strabismus Q. 2002;17(1):43–48.
58. Follens I, Godts D, Evens PA, Tassignon MJ. Combined fourth
and sixth cranial nerve palsy after lumbar puncture: a rare
complication. Bull Soc Belge Ophthalmol. 2001;281:29–33.
59. Warner GT. Spontaneous intracranial hypotension causing a
partial third cranial nerve palsy: a novel observation. Cephalalgia. 2002;22(10):822–823.
60. Ferrante E, Savino A, Brioschi A, et al. Transient oculomotor
cranial nerves palsy in spontaneous intracranial hypotension. J
Neurosurg Sci. 1998;42(3):177–179.
61. Ohta H, Genmoto T, Yokota A. Spontaneous intracranial
hypotension associated with trochlear nerve palsy and bilateral
chronic subdural hematoma. No To Shinkei. 2004;56(2):169–
172.
62. Alonso Cánovas A, Martínez San Millán J, Novillo López ME,
Masjuán Vallejo J. Third cranial nerve palsy due to intracranial
hypotension syndrome. Neurologia. 2008;23(7):462–465.
63. Zada G, Pezeshkian P, Gianotta S. Spontaneous intracranial
hypotension and immediate improvement following epidural
blood patch placement demonstrated by intracranial pressure
monitoring. Case report. J Neurosurg. 2007;106(6):1089–1090.
64. Beck CE, Rizk NW, Kiger LT, et al. Intracranial hypotension
presenting with severe encephalopathy. Case report. J Neurosurg. 1998;89(3):470–473.
65. Binder DK, Dillon WP, Fishman RA, Schmidt MH. Intrathecal
saline infusion in the treatment of obtundation associated with
spontaneous intracranial hypotension: technical case report.
Neurosurgery. 2002;51(3):830–837.
66. Evan RW, Mokri B. Spontaneous intracranial hypotension
resulting in coma. Headache. 2002;42(2):159–160.
67. Ferrante E, Arpino I, Citterio A, Savino A. Coma resulting from
spontaneous intracranial hypotension treated with the epidural
blood patch in the Trendelenburg position pre-medicated with
acetazolamide. Clin Neurol Neurosurg. 2009 Jul 2 [Epub ahead
of print].
68. Kashmere JL, Jacka MJ, Emery D, Gross DW. Reversible coma:
a rare presentation of spontaneous intracranial hypotension.
Can J Neurol Sci. 2004;31(4):565–568.
69. Pleasure SJ, Abosch A, Friedman J, et al. Spontaneous intracranial hypotension resulting in stupor caused by diencephalic
compression. Neurology. 1998;50(6):1854–1857.
70. Sayer FT, Bodelsson M, Larsson EM, Romner B. Spontaneous
intracranial hypotension resulting in coma: case report. Neurosurgery. 2006;59(1):E204.
71. Schievink WI, Palestrant D, Maya MM, Rappard G. Spontaneous spinal cerebrospinal fluid leak as a cause of coma after
craniotomy for clipping of an unruptured intracranial aneurysm. J Neurosurg. 2009;110(3)521–524.
72. Horton JC, Fishman RA. Neurovisual findings in the syndrome
of spontaneous intracranial hypotension from dural cerebrospinal fluid leak. Ophthalmology. 1994;101(2):244–251.
73. Kleopa KA, Natsiopoulos K. Acute cervical radiculopathies in spontaneous intracranial hypotension. J Neurol.
2009;256(3):499–501.
74. Albayram S, Wasserman BA, Yousem DM, Wityk R. Intracranial
hypotension as a cause of radiculopathy from cervical epidural
venous engorgement: case report. AJNR Am J Neuroradiol.
2002;23(4):618–621.
75. Schabel JE, Wang ED, Glass PS. Arm pain as an unusual presentation of postdural puncture intracranial hypotension. Anesth
Analg. 2000; 91(4):910–912.
76. Carrero EJ, Agusti M, Fabregas N, Valldeoriola F, Fernandez C.
Unilateral trigeminal and facial nerve palsies associated with
epidural analgesia in labour. Can J Anaesth. 1998;45(9):893–
897.
77. Cheshire WP Jr, Wharen RE Jr. Trigeminal neuralgia in a
patient with spontaneous intracranial hypotension. Headache.
55
ENJ 2009; 1: (1). September 2009
European Neurological Journal
2009;49(5):770–773.
78. Portier F, de Minteguiaga C, Racy E, et al. Spontaneous intracranial hypotension: a rare cause of labyrinthine hydrops. Ann
Otol Rhinol Laryngol. 2002;111(9):817–820.
79. Hong M, Shah GV, Adams KM, et al. Spontaneous intracranial
hypotension causing reversible frontotemporal dementia. Neurology. 2002;58(8):1285–1287.
80. Sayao AL, Heran MK, Chapman K, et al. Intracranial hypotension causing reversible frontotemporal dementia and coma. Can
J Neurol Sci. 2009;36(2):252–256.
81. Walker L, DeMeulemeester C. Spontaneous intracranial
hypotension masquerading as frontotemporal dementia. Clin
Neuropsychol. 2008;22(6):1035–1053.
82. Pakiam AS, Lee C, Lang AE. Intracranial hypotension with
parkinsonism, ataxia, and bulbar weakness. Arch Neurol.
1999;56(7):869–872.
83. Fedi M, Cantello R, Shuey NH, et al. Spontaneous intracranial
hypotension presenting as a reversible dorsal midbrain syndrome. J Neuroophthalmol. 2008;28(4):289–292.
84. Nowak DA, Radiek SO, Zinner J, et al. Broadening the clinical
spectrum: unusual presentation of spontaneous cerebrospinal
fluid hypovolemia. Case report. J Neurosurg. 2003;98(4):903–
907.
85. Schievink WI, Meyer FB, Atkinson JL, Mokri B. Spontaneous
spinal cerebrospinal fluid leaks and intracranial hypotension. J
Neurosurg. 1996;84(4):598–605.
86. Schievink WI, Maya MM. Quadriplegia and cerebellar hemorrhage in spontaneous intracranial hypotension. Neurology.
2006;66(11):1777–1778.
87. Mokri B, Ahlskog JE, Leutmer PH. Chorea as a manifestation
of spontaneous CSF leak. Neurology. 2006;67(8):1490–1491.
88. Yamamoto M, Suehiro T, Nakata H, et al. Primary low cerebrospinal fluid pressure syndrome associated with galactorrhea.
Intern Med. 1993;32(3):228–231.
89. Inamo Y. A boy with growth disturbance caused by hypothalamic damage associated with intracranial hypotension syndrome
following a motor vehicle accident: case report. Endocr J.
2008;55(6):1113–1116.
90. Devine MJ, Mehta A, O’Riordan S. Spontaneous intracranial
hypotension manifesting as an amnestic syndrome. J Neurol.
2009;256(6):1002–1003.
91. Vetrugno R, Mascalchi M, Chierichetti F, et al. Hypoactive–hypoalert behavior (“psychic akinesia”) in intracranial hypotension syndrome. Neurology. 2008;71(18):1452–1454.
92. Chi NF, Wang SJ, Lirng JF, Fuh JL. Transtentorial herniation
with cerebral infarction and duret haemorrhage in a patient
with spontaneous intracranial hypotension. Cephalalgia. 2007
27(3):279–282.
93. Sung V. Acute respiratory failure complicating intracranial
hypotension. Headache. 2007;47(9):1340–1341.
94. Mokri B. The Monro–Kellie hypothesis: applications in CSF
volume depletion. Neurology. 2001;56(12):1746–1748.
95. Fishman RA, Dillon WP. Dural enhancement and cerebral
displacement secondary to intracranial hypotension. Neurology.
1993;43(3 Pt 1):609–611.
96. Mokri B, Hunter SF, Atkinson JL, Piepgras DG. Orthostatic
headaches caused by CSF leak but with normal CSF pressures.
Neurology. 1998;51(3)786–790.
97. Iqbal J, Davis LE, Orrison WW Jr. An MRI study of lumbar
puncture headaches. Headache. 1995 35(7):420–422.
98. Grant R, Condon B, Hart I, Teasdale GM. Changes in intracranial CSF volume after lumbar puncture and their relationship to post-LP headache. J Neurol Neurosurg Psychiatry.
1991;54(5):440–442.
99. Schievink WI, Tourje J. Upright MRI in spontaneous spinal cerebrospinal fluid leaks and intracranial hypotension. Headache.
2007;47(9):1345–1346.
100. Magnaes B. Body position and cerebrospinal fluid pressure. Part
2: clinical studies on orthostatic pressure and the hydrostatic
indifferent point. J Neurosurg. 1976;44(6):698–705.
ENJ 2009; 1: (1). September 2009
101. Fattori R, Nienaber CA, Descovich B, et al. Importance of dural
ectasia in phenotypic assessment of Marfan’s syndrome. Lancet.
1999;354(9182):910–913.
102. Bono F, Giliberto C, Lavano A, Quattrone A. Posture-related
cough headache and orthostatic drop in lumbar CSF pressure. J
Neurol. 2005;252(2):237–238.
103. Kunkle EC, Ray BS, Wolff HG. Experimental studies on
headache: analysis of the headache associated with changes in
intracranial pressure. Arch Neurol Psychiatry. 1943;49:323–358.
104. Nelson MO. Postpuncture headaches. A clinical and experimental study of the cause and prevention. Arch Derm Syphil.
1930;21:615–627.
105. Franzini A, Messina G, Nazzi V, et al. Spontaneous intracranial
hypotension syndrome: a novel speculative physiopathological
hypothesis and a novel patch method in a series of 28 consecutive patients. J Neurosurg. 2009 Jul [Epub ahead of print].
106. Rozen T, Swidan S, Hamel R, Saper J. Trendelenburg position:
a tool to screen for the presence of low CSF pressure syndrome
in daily headache patients. Headache. 2008;48(9):1366–1371.
107. Pannullo SC, Reich JB, Krol G, et al. MRI changes in intracranial hypotension. Neurology. 1993;43(5):919–926.
108. Atkinson JL, Weinshenker BF, Miller GM, et al. Acquired Chiari
I malformation secondary to spontaneous spinal cerebrospinal
fluid leakage and chronic intracranial hypotension syndrome in
seven cases. J Neurosurg. 1998;88(2):237–242.
109. Mokri B, Atkinson JL. False pituitary tumor in CSF leaks.
Neurology. 2000;55(4):573–575.
110. Shimazu N, Oba H, Aoki S, et al. Pituitary enlargement in
spontaneous intracranial hypotension on MRI. Nippon Igaku
Hoshasen Gakkai Zasshi. 1998;58(7):349–352.
111. Alvarez-Linera J, Escribano J, Benito-León J, et al. Pituitary enlargement in patients with intracranial hypotension syndrome.
Neurology. 2000;55(12):1895–1897.
112. Bakshi R, Mechtler LL, Kamran S, et al. MRI findings in
lumbar puncture headache syndrome: abnormal dural-meningeal and dural venous sinus enhancement. Clin Imaging.
1999;23(2):73–76.
113. Schievink WI, Maya MM, Moser FG, Tourje J. Spectrum of subdural fluid collections in spontaneous intracranial hypotension.
J Neurosurg. 2005;103(4):608–613.
114. Baryshnik DB, Farb RI. Changes in the appearance of venous
sinuses after treatment of disordered intracranial pressure.
Neurology. 2004;62(8):1445–1446.
115. Murros K, Fogelholm R. Spontaneous intracranial hypotension with slit ventricles. J Neurol Neurosurg Psychiatry.
1983;46(12):1149–1151.
116. Schievink WI, Tourje J. Intracranial hypotension without
meningeal enhancement on magnetic resonance imaging. Case
report. J Neurosurg. 2000;92(3):475–477.
117. Schoffer KL, Benstead TJ, Grant I. Spontaneous intracranial
hypotension in the absence of magnetic resonance imaging
abnormalities. Can J Neurol Sci. 2002;29(3):253–257.
118. Mokri B, Atkinson JL, Dodick DW, et al. Absent pachymeningeal gadolinium enhancement on cranial MRI despite symptomatic CSF leak. Neurology. 1999;53(2):402–404.
119. Rabin BM, Roychowdhury S, Meyer JR, et al. Spontaneous
intracranial hypotension: spinal MRI findings. AJNR Am J
Neuroradiol. 1998;19(6):1034–1039.
120. Chiapparini L, Farina L, D’Incerti L, et al. Spinal radiological
findings in nine patients with spontaneous intracranial hypotension. Neuroradiology. 2002;44(2):143–152.
121. Moayeri NN, Henson JW, Schaefer PW, et al. Spinal dural
enhancement on magnetic resonance imaging associated
with spontaneous intracranial hypotension. J Neurosurg.
1998;88(5):912–918.
122. Mokri B. Low cerebrospinal fluid pressure syndromes. Neurol
Clin. 2004;22(1):55–74.
123. Watanabe A, Horikoshi T, Uchida M, et al. Diagnostic value of
spinal MR imaging in spontaneous intracranial hypotension
syndrome. AJNR Am J Neuroradiol. 2009;30(1):147–151.
56
www.slm-neurology.com
Orthostatic Headache with and without Cerebrospinal Fluid Leak: A Review
124. Bai J, Yokoyama K, Kinuya S, et al. Radionuclide cisternography in intracranial hypotension syndrome. Ann Nucl Med.
2002;16(1):75–78.
125. Molins A, Alvárez J, Sumalla J, et al. Cisternographic pattern of
spontaneous liquoral hypotension. Cephalalgia. 1990;10(2):5–
65.
126. Moriyama E, Ogawa T, Nishida A, et al. Quantitative analysis
of radioisotope cisternography in the diagnosis of intracranial
hypotension. J Neurosurg. 2004;101(3):421–426.
127. Morioka T, Aoki T, Tomoda Y, et al. Cerebrospinal fluid leakage
in intracranial hypotension syndrome: usefulness of indirect
findings in radionuclide cisternography for detection and treatment monitoring. Clin Nucl Med. 2008;33(3):181–185.
128. Weber WE, Heidendal GA, de Krom MC. Primary intracranial
hypotension and abnormal radionuclide cisternography: report
of a case and review of the literature. Clin Neurol Neurosurg.
1991;93(1):55–60.
129. Leutmer PH, Mokri B. Dynamic CT myelography: a technique
for localizing high-flow spinal cerebrospinal fluid leaks. AJNR
Am J Neuroradiol. 2003;24(8):1711–1714.
130. Hoxworth JM, Patel AC, Bosch EP, Nelson KD. Localization of a
rapid CSF leak with digital subtraction myelography. AJNR Am
J Neuroradiol. 2009;30(3):516–519.
131. Tali ET, Ercan N, Krumina G, et al. Intrathecal gadolinium
(gadopentetate dimeglumine) enhanced magnetic resonance
myelography and cisternography: results of a multicenter study.
Invest Radiol. 2002;37(3):152–159.
132. Liong WC, Constantinescu CS, Jaspan T. Intrathecal gadolinium-enhanced magnetic resonance myelography in the detection of CSF leak. Neurology. 2006;67(8):1522.
133. Albayram S, Kilic F, Ozer H, et al. Gadolinium-enhanced MR
cisternography to evaluate dural leaks in intracranial hypotension syndrome. AJNR Am J Neuroradiol. 2008;29(1):116–121.
134. Yoo HM, Kim SJ, Choi CG, et al. Detection of CSF leak in
spinal CSF leak syndrome using MR myelography: correlation
with radioisotope cisternography. AJNR Am J Neuroradiol.
2008;29(4):649–654.
135. Mokri B, Parisi JE, Scheithauer BW, et al. Meningeal biopsy
in intracranial hypotension: meningeal enhancement on MRI.
Neurology. 1995;45(10):1801–1807.
136. Duffy PJ, Crosby ET. The epidural blood patch. Resolving the
controversies. Can J Anaesth. 1999;46(3):878–886.
137. Vilming ST, Titus F. Low cerebrospinal fluid pressure. In: Olefson J, Tpelt-Hanson P, Welch KMA, eds. The Headache. New
York: Raven Press; 1993:687–695.
138. Karst M, Hollenhorst J, Fink M, Conrad I. Computerized
tomography-guided epidural blood patch in the treatment of
spontaneous low cerebrospinal fluid pressure headache. Acta
Anaesthesiol Scand. 2001;45(5):649–651.
139. Dillo W, Hollenhorst J, Brassel F, et al. Successful treatment of a
spontaneous cervical cerebrospinal fluid leak with a CT guided
epidural blood patch. J Neurol. 2002;249(2):224–225.
140. Hayek SM, Fattouh M, Dews T, et al. Successful treatment of
spontaneous cerebrospinal fluid leak headache with fluoroscopically guided epidural blood patch: a report of four cases. Pain
Med. 2003;4(4):373–378.
141. Kantor D, Silberstein SD. Cervical epidural blood patch for low
CSF pressure headaches. Neurology. 2005;65(7):1138.
142. Rai A, Rosen C, Carpenter J, Miele V. Epidural blood patch at
C2: diagnosis and treatment of spontaneous intracranial hypotension. AJNR Am J Neuroradiol. 2005;26(10)2663–2666.
143. Cru JBP, Gerritse BM, van Dongen RTM, Schoonderwaldt HC.
Epidural fibrin glue injection stops persistent postdural puncture headache. Anesthesiology. 1999;91(2):576–577.
144. Patel MR, Caruso PA, Yousuf N, Rachlin J. CT-guided percutaneous fibrin glue therapy of cerebrospinal fluid leaks in the
spine after surgery. AJR Am J Roentgenol. 2000;175(2):443–
446.
145. Kamada M, Fujita Y, Ishii R, Endoh S. Spontaneous intracranial hypotension successfully treated by epidural patching with
www.slm-neurology.com
fibrin glue. Headache. 2000;40(10):844–847.
146. Schievink WI, Maya MM, Moser FM. Treatment of spontaneous intracranial hypotension with percutaneous placement of a fibrin sealant. Report of four cases. J Neurosurg.
2004;100(6):1098–1100.
147. Gladstone JP, Nelson K, Patel N, Dodick DW. Spontaneous CSF
leak treated with percutaneous CT-guided fibrin glue. Neurology. 2005;64(10):1818–1819.
148. Savoiardo M, De Simone T, Franzini A, et al. Spontaneous CSF
leak treated with percutaneous CT-guided fibrin glue. Neurology. 2006;66(5):782.
149. Gibson BE, Wedel DJ, Faust RJ, Petersen RC. Continuous
epidural saline infusion for the treatment of low CSF pressure
headache. Anesthesiology. 1988;68(5):789–791.
150. Vaquero Roncero LM, Sánchez Montero FJ, Muriel Villoria C.
Effectiveness of epidural administration of saline solutions to
prevent or treat postdural puncture headache. Rev Esp Anestesiol Reanim. 2004;51(10):589–594.
151. Aldrete JA. Persistent post-dural-puncture headache treated
with epidural infusion of dextran. Headache. 1994;34(5):265–
267.
152. Souron V, Hamza J. Treatment of postdural puncture headaches
with colloid solutions: an alternative to epidural blood patch.
Anesth Analg. 1999;89(5):1333–1334.
153. Akkawi N, Locatelli P, Borroni B, et al. A complicated case of
intracranial hypotension: diagnostic and management strategies. Neurol Sci. 2006;27(1):63–66.
154. Lasboo AA, Hurley MC, Walker MT, et al. Emergent imageguided treatment of a large CSF leak to reverse “in-extremis”
signs of intracranial hypotension. AJNR Am J Neuroradiol.
2008;29(9):1627–1629.
155. Maher CO, Meyer FB, Mokri B. Surgical treatment of spontaneous spinal cerebrospinal fluid leaks. Neurosurg Focus.
2000;9(1):e7.
156. Schievink WI, Morreale VM, Atkinson JL, et al. Surgical treatment of spontaneous spinal cerebrospinal fluid leaks. J Neurosurg. 1998;88(2):243–246.
157. Cohen-Gadol AA, Mokri B, Piepgras DG, et al. Surgical anatomy
of dural defects in spontaneous spinal cerebrospinal fluid leaks.
Neurosurgery. 2006;58(4 Suppl 2):ONS-238–245.
158. Schievink WI. A novel technique for treatment of intractable
spontaneous intracranial hypotension: lumbar dural reduction
surgery. Headache. 2009;49(7):1047–1051.
159. Merskey H, Bogduk N, eds. Classification of Chronic Pain:
Descriptions of Chronic Pain Syndromes and Definitions of Pain
Terms, 2nd edn. Seattle: IASP; 1994.
57
ENJ 2009; 1: (1). September 2009
European Neurological Journal
review article
Obesity, Diet, and Risk of Restless Legs Syndrome
Xiang Gao1,2 and Shivani Sahni3
Affiliations: 1Channing Laboratory, Department of Medicine, Brigham and Women’s Hospital, and Harvard Medical School,
Boston, MA, USA; 2Department of Nutrition, Harvard University School of Public Health, Boston, MA; 3Musculoskeletal Research,
Institute for Aging Research, Hebrew SeniorLife, and Harvard Medical School, Boston, MA, USA
Submission date: 26th June 2009, Revision date: 16th August 2009, Acceptance date: 3rd September 2009
A B STRA C T
The restless legs syndrome (RLS) is a common movement disorder, characterized by an almost irresistible urge to move
the legs in the evening or at rest. According to recent estimates, it affects ~5–15% of adults and often has a substantial
impact on sleep, daily activities, and quality of life. Although genetic susceptibility has been shown to play an important
role in the pathogenesis of RLS, there is evidence supporting possible environmental causes of RLS. In this review, we focus on obesity and dietary factors, including iron, B vitamins, vitamin E, vitamin C, and magnesium, as these factors are
modifiable. Both clinical and epidemiology studies suggest that obesity and dietary factors could be risk factors for RLS.
However, previous studies are limited by small sample sizes and retrospective or cross-sectional designs that preclude conclusions regarding causality. Therefore, further prospective studies examining the relation between obesity, diet, and the
risk of developing RLS should be a priority.
Keywords: Obesity, diet, risk factor, restless legs syndrome, iron deficiency, homocysteine
Correspondence: Xiang Gao, Channing Laboratory, Department of Medicine, Brigham and Women’s Hospital, and Harvard
Medical School, Boston, MA 02115, USA. Tel: +1-617-432-5080; fax: +1-617-432-2435; e-mail: xiang.gao@channing.harvard.edu
INTRODUCTION
vided strong evidence of the association of genetic variations with RLS [13, 14]. Additionally, non-genetic factors such as age, female sex, pregnancy, iron deficiency,
and other lifestyle factors have also been suggested to
play an important role in RLS [1, 2, 6]. In this review, we
focus on the potential roles of obesity and diet on RLS
risk because both are modified factors.
Restless legs syndrome (RLS) is a neurological disorder characterized by an almost irresistible urge to move
the legs [1, 2]. RLS is the most common movement disorder, affecting approximately 5–15% of the general population [1–3], with a substantial impact on sleep, daily
activities, and quality of life [4]. It has been shown that
approximately 50% of RLS patients reported an inability
to fall asleep and 61% reported disturbed or interrupted
sleep [5], which may result from an urge to move as well
as from related sensory symptoms in the legs, which are
worse at night and while at rest [1].
OBESITY AND RLS
The unfavorable role of obesity on dopamine status in
the central nervous system (CNS) has been demonstrated by several human and animal studies. A case–control study showed that obese individuals (n=10, body
mass index (BMI) >40 kg/m2) had a significantly lower
striatal dopamine D2 receptor availability than controls
(P<0.01) [15]. These findings were supported by observations from animal studies where obese rats had lower
D2 dopamine receptors [16, 17].
Cross-sectional studies have shown that subjects with
RLS have a significantly higher prevalence of depression, diabetes, cardiovascular disease, and a lower mental health score than subjects without RLS [6–11]. RLS
sufferers also have a reduced quality of life compared
with the general population, which is comparable with
that experienced by those with other serious chronic
medical conditions, such as type 2 diabetes mellitus,
chronic obstructive pulmonary disorder, or depression
[5]. Among patients with endstage renal disease, RLS
was associated with increased mortality [12].
Moreover, in obese individuals, D2 receptor levels have
been shown to be inversely associated with BMI (r=0.84)
[15]. Genetic studies have shown a link between obesity and variants of dopamine metabolism-related genes,
such as Taq 1, monoamine oxidase A, and monoamine
oxidase B [18, 19]. Vascular abnormality resulting from
obesity could be an alternative mechanism underlining
the possible association between obesity and RLS. Cardiovascular diseases have been shown to be positively
associated with RLS [20]. A recent study showed that
It has been suggested that RLS is associated with both
genetic and non-genetic factors. More than 50% of RLS
patients have a positive family history of this condition
[2]. Two recent genome-wide association studies proENJ 2009; 1: (1). September 2009
59
www.slm-neurology.com
European Neurological Journal
and weight gain were also positively associated with
the prevalence of RLS (P trend <0.01 for both). These
preliminary results suggest a possible role of obesity in
RLS. However, this observation needs to be replicated in
other populations with different cultural backgrounds
and lifestyles.
DIETARY FACTORS AND RLS
Iron Status
Since 1945, when Ekbom first proposed that RLS could
be secondary to iron deficiency [26], the role of iron in
the pathology of RLS has been investigated intensively.
Serological studies have observed a lower ferritin or a
higher transferrin concentration, indicating a decreased
iron sufficiency, in serum or cerebrospinal fluid (CSF)
among RLS patients, relative to control subjects [27–31].
These findings have been supported by imaging studies.
Using magnetic resonance imaging (MRI), it has been
shown that RLS patients have a significantly lower iron
concentration in brain, as assessed by an ”iron index”,
than control subjects [32, 33]. Further studies observed
that RLS severity was inversely associated with serum
ferritin levels [34, 35]. Clinical trials provide further evidence of a possible causal relationship between iron and
RLS: supplementation of iron, either orally [34, 36] or intravenously [30, 37], resulted in significant improvement
in RLS symptoms. However, it remains unclear whether
dietary iron content at normal intake level is associated
with RLS risk or progression. Iron could influence CNS
dopamine status via several mechanisms. Iron deficiency
may decrease dopamine synthesis, as iron is a cofactor
of tyrosine hydroxylase. Animal studies have shown that
iron deficiency reduced dopamine transporters and subsequently reduced dopamine uptake [38, 39].
Figure 1. A
djusted OR (95% CI) of RLS according to body mass
index in the Health Professionals Follow-up Study and the
Nurses’ Health Study II [25], adjusting for age, ethnicity,
smoking status, physical activity, use of antidepressants,
the Crown–Crisp phobia index, and presence of stroke,
hypertension, or myocardial infraction (each of them,
yes/no)
enhanced external counterpulsation treatment significantly improved the RLS symptoms [21].
Several epidemiologic studies have examined the crosssectional relationship between obesity and RLS. Most of
these studies [11, 22, 23], but not all [24], have reported
significant positive associations. Among 1803 men and
women aged 18 years or older, Phillips et al found that
each increase of 5 kg/m2 BMI was associated with a 31%
increased likelihood of having RLS [22]. In another
cross-sectional study conducted in five European countries (n=18 890), crude odds ratio (OR) for RLS was 1.22
(95% CI 1.0 to 1.5) for BMI of >27 vs 20–25 kg/m2 [11].
In a Korean population (n=9939), Kim et al found a significant association between BMI and RLS among women (OR=1.2 for BMI >25 vs ≤25 kg/m2) but not among
men (OR=1.1) [23]. In contrast, in a case–control study
including 103 RLS cases and 103 control subjects (mean
age 43 years for both groups) living in Mersin, Turkey,
Sevim et al reported a similar mean BMI between the
two groups (mean BMI 25.8 kg/m2 for both groups) [24].
One possible interpretation for failure to find significant
associations between BMI and RLS could be the small
sample size.
Because serum ferritin and transferrin concentrations
are affected by several factors, such as inflammation
and diet, in addition to body iron stores, the observed
associations between these biomarkers and RLS could
be confounded. This limitation could be overcome by the
use of the history of blood donation as a marker of body
iron levels. Because body iron stores can be reduced
greatly through regular blood donation, the contrast
between regular blood donors and non-donors with a
similar distribution of other RLS risk factors provides
a direct and powerful test of the hypothesis that depletion of body iron stores increases the risk of RLS. Few
studies have examined RLS status among blood donors.
In a small hospital-based cross-sectional study (n=109),
patients with repeated blood donation (≥5 times in their
lifetime) were five times more likely to have RLS and/
or periodic limb movement in sleep (PLMS) than those
without repeated blood donation (OR=5.1) [40]. A similar positive association between blood donation and
RLS was observed in a case–control study with 64 iron
deficiency anemic and 256 non-anemic control subjects
in an Indian population [41]. A cross-sectional study
We have been conducting analyses to examine the
cross-sectional association between obesity and RLS in
two ongoing US cohorts: the Health Professional Follow-up Study (23 575 men, mean age 67 years) and the
Nurses’ Health Study II (65 872 women, mean age 50
years) free from diabetes and arthritis (Figure 1) [25].
Information on RLS was assessed using a set of standardized questions recommended by the International
RLS Study Group. Multivariate OR for RLS was 1.42
(95% CI 1.3 to 1.6; P trend <0.0001) for subjects with
BMI >30 vs <23 kg/m2 and 1.60 (95% CI 1.5 to 1.8; P
trend<0.0001) for highest vs lowest waist circumference quintiles. BMI in early adulthood (age 18–21 years)
ENJ 2009; 1: (1). September 2009
60
www.slm-neurology.com
Obesity, Diet, and Risk of Restless Legs Syndrome
56]. Further, elevated homocysteine is associated with
cardiovascular and renal disease, which have been found
to co-occur with RLS. In a case–control study including
97 RLS patients and 92 healthy control subjects, Bachmann et al reported that RLS patients tended to have
higher serum homocysteine concentrations relative to
control subjects (11.7 vs 11.0 μmol/L), but the difference
was not significant [57].
reported a high prevalence of RLS among blood donors
(15% and 25% for men and women respectively; n=946)
[42]. This study also reported significant associations
between the presence of RLS and red cell distribution
width, a marker of iron deficiency, but not with dietary
iron intake [42]. In a small retrospective clinical study of
eight blood donors with RLS, six RLS cases were found
to have an onset at about the same time as or after blood
donations [43]. However, these results should be interpreted with caution because of small sample size and
the potential recall and selection bias associated with a
retrospective design.
Magnesium
A small case–control study showed that RLS patients
had a lower serum magnesium concentration relative to
control subjects [58]. Oral or intravenous administration of magnesium has been shown to relieve RLS symptoms [59, 60]. However, a recent case–control study
including 11 RLS cases failed to find a significant difference for serum and CSF magnesium concentrations
between cases and control subjects [61]. Magnesium inhibits N-methyl-D-aspartate receptors, which could be
involved in RLS through the activation and production
of inflammatory mediators [62]. Magnesium serves as a
calcium antagonist because of their chemical similarity
[63]. Reduction in magnesium–calcium competition due
to magnesium deficiency may lead to muscle cramping
[64]. This could confound the observed association between magnesium and RLS. However, to our knowledge,
no study has examined whether dietary intake of magnesium is associated with RLS risk or progression.
B Vitamins
The role of folate and vitamin B12 in RLS has been
suggested since the 1970s. In a series of clinical studies
of folate and RLS [44–47], Botez et al observed that (1)
RLS patients generally had low plasma, red blood cell,
and CSF folate concentrations; (2) RLS symptoms were
temporarily improved immediately after administration
of vitamin B12 as assessed by the Schilling test; and (3)
RLS was responsive to folic acid therapy. Further, in a
randomized clinical trial by Botez and Lambert [47], one
group of 11 pregnant women received a multivitamin
tablet daily containing 0.5 g of folic acid, B12, and iron,
whereas the other group (n=10) received the same multivitamin without folic acid. These women were followed
up to the 13th, 22nd, and 35th weeks of pregnancy and
6 weeks after delivery. At the end of the trial, only 1 out
of 11 women developed RLS in the folic acid group relative to 8 out of 10 in the control group (P=0.002). Similarly, a small cohort study including 45 pregnant women
found that those with RLS had lower plasma folate concentrations during preconception and at each trimester
than subjects without RLS [48].
A possible association between intake of antioxidants,
such as vitamin E, and vitamin C and RLS has been suggested by some studies [3, 65, 66]. However, there have
been no studies specifically examining whether longterm intake of such antioxidants influences RLS risk or
progression.
Folate is important for the generation of dopamine in
the CNS [49]. Folate and S-adenosyl-methionine, which
is modulated by folate and B12, influence the synthesis of CNS tetrahydrobiopterin, which is essential in
the conversion of tyrosine to L-dopa through tyrosine
hydroxylase [50, 51]. Interestingly, tetrahydrobiopterin
has a circadian change, which parallels the pattern seen
in RLS symptoms. A study including 30 RLS cases and
22 control subjects showed that tetrahydrobiopterin levels decreased significantly during the night among RLS
patients, but not among control subjects [52].
Summary
As reviewed above, both clinical and epidemiological
studies support the notions that obesity and nutritional
inadequacy could be modifiable risk factors for RLS.
However, these studies are limited by their cross-sectional design, small sample size, and failure to adjust for
several important cofounders. Therefore, further prospective studies examining the relation between obesity,
diet, and the risk of developing RLS should be a priority.
Further, potential interaction between these factors and
genetic susceptibility would also be of interest to explore
in future studies. Understanding the roles of obesity and
diet in RLS will not only improve our understanding of
the etiology of RLS but could potentially help to pursue
new treatment and prevention strategies.
Abnormalities in folate and B12 metabolic function
result in elevated homocysteine concentration, which
may also have a direct role in the pathogenesis of RLS
because of its toxic effect on dopaminergic neurons. In
an animal model of Parkinson’s disease (PD), folate deficiency and elevated homocysteine significantly sensitized dopaminergic neurons to a subtoxic dose of MPTP
[53]. Homocysteine also has a neurotoxic effect by activating the N-methyl-D-aspartate receptor, leading to
cell death [54, 55], or may be converted into homocysteic
acid, which also has an excitotoxic effect on neurons [54,
www.slm-neurology.com
Funding/support: The study was supported by NIH/NINDS
grant R01 NS048517. None of the sponsors participated in the
design of the study or in the collection, analysis, or interpretation of the data.
Disclosures: The authors have no financial interests to disclose related to the contents of this article.
61
ENJ 2009; 1: (1). September 2009
European Neurological Journal
REFERENCES
demiology of restless legs symptoms in adults. Arch Intern Med.
2000;160:2137–2141.
23. Kim J, Choi C, Shin K, et al. Prevalence of restless legs syndrome and associated factors in the Korean adult population:
the Korean Health and Genome Study. Psychiatry Clin Neurosci.
2005;59:350–353.
24. Sevim S, Dogu O, Camdeviren H, et al. Unexpectedly low prevalence and unusual characteristics of RLS in Mersin, Turkey.
Neurology. 2003;61:1562–1569.
25. Gao X, Schwarzschild MA, Wang H, Ascherio A. Obesity and
restless legs syndrome in men and women. Neurology. 2009;
72;1255–1261.
26. Ekbom K. Restless legs: a clinical study. Acta Med Scand.
1945;159:1–122.
27. Clardy SL, Earley CJ, Allen RP, Beard JL, Connor JR. Ferritin
subunits in CSF are decreased in restless legs syndrome. J Lab
Clin Med. 2006;147:67–73.
28. Mizuno S, Mihara T, Miyaoka T, Inagaki T, Horiguchi J. CSF iron,
ferritin and transferrin levels in restless legs syndrome. J Sleep
Res. 2005;14:43–47.
29. Earley CJ, Connor JR, Beard JL, Malecki EA, Epstein DK, Allen
RP. Abnormalities in CSF concentrations of ferritin and transferrin in restless legs syndrome. Neurology. 2000;54:1698–1700.
30. Earley CJ, Heckler D, Allen RP. The treatment of restless legs
syndrome with intravenous iron dextran. Sleep Med. 2004;5:231–
235.
31. Kryger MH, Otake K, Foerster J. Low body stores of iron and
restless legs syndrome: a correctable cause of insomnia in adolescents and teenagers. Sleep Med. 2002;3:127–132.
32. Allen RP, Barker PB, Wehrl F, Song HK, Earley CJ. MRI measurement of brain iron in patients with restless legs syndrome.
Neurology. 2001;56:263–265.
33. Earley CJ, P BB, Horska A, Allen RP. MRI-determined regional
brain iron concentrations in early- and late-onset restless legs
syndrome. Sleep Med. 2006;7:458–461.
34. O’Keeffe ST, Gavin K, Lavan JN. Iron status and restless legs
syndrome in the elderly. Age Ageing. 1994;24:200–203.
35. Sun ER, Chen CA, Ho G, Earley CJ, Allen RP. Iron and the restless legs syndrome. Sleep. 1998;21:371–377.
36. Davis BJ, Rajput A, Rajput ML, Aul EA, Eichhorn GR. A randomized, double-blind placebo-controlled trial of iron in restless
legs syndrome. Eur Neurol. 2000;43:70–75.
37. Sloand JA, Shelly MA, Feigin A, Bernstein P, Monk RD. A
double-blind, placebo-controlled trial of intravenous iron dextran
therapy in patients with ESRD and restless legs syndrome. Am J
Kidney Dis. 2004;43:663–670.
38. Burhans MS, Dailey C, Beard Z, et al. Iron deficiency: differential
effects on monoamine transporters. Nutr Neurosci. 2005;8:31–38.
39. Erikson KM, Jones BC, Beard JL. Iron deficiency alters dopamine
transporter functioning in rat striatum. J Nutr. 2000;130:2831–
2837.
40. Kryger MH, Shepertycky M, Foerster J, Manfreda J. Sleep disorders in repeat blood donors. Sleep. 2003;26:625–626.
41. Rangarajan S, D’Souza GA. Restless legs syndrome in Indian
patients having iron deficiency anemia in a tertiary care hospital.
Sleep Med. 2007;8:247–251.
42. Ulfberg J, Nystrom B. Restless legs syndrome in blood donors.
Sleep Med. 2004;5:115–118.
43. Silber MH, Richardson JW. Multiple blood donations associated
with iron deficiency in patients with restless legs syndrome. Mayo
Clin Proc. 2003;78:52–54.
44. Botez MI. Folate deficiency and neurological disorders in adults.
Med Hypotheses. 1976;2:135–140.
45. Botez MI, Cadotte M, Beaulieu R, Pichette LP, Pison C. Neurologic disorders responsive to folic acid therapy. Can Med Assoc J.
1976;115:217–223.
46. Botez MI, Fontaine F, Botez T, Bachevalier J. Folate-responsive
neurological and mental disorders: report of 16 cases. Neuropsychological correlates of computerized transaxial tomography and
radionuclide cisternography in folic acid deficiencies. Eur Neurol.
1. Kushida CA. Clinical presentation, diagnosis, and quality of life
issues in restless legs syndrome. Am J Med. 2007;120:S4–S12.
2. Allen RP, Picchietti D, Hening WA, Trenkwalder C, Walters AS,
Montplaisi J. Restless legs syndrome: diagnostic criteria, special
considerations, and epidemiology. A report from the restless legs
syndrome diagnosis and epidemiology workshop at the National
Institutes of Health. Sleep Med. 2003;4:101–119.
3. Garcia-Borreguero D, Odin P, Schwarz C. Restless legs syndrome:
an overview of the current understanding and management. Acta
Neurol Scand. 2004;109:303–317.
4. Kushida C, Martin M, Nikam P, et al. Burden of restless legs
syndrome on health-related quality of life. Qual Life Res.
2007;16:617–624.
5. Allen RP, Walters AS, Montplaisir J, et al. Restless legs syndrome
prevalence and impact: REST general population study. Arch
Intern Med. 2005;165:1286–1292.
6. Phillips B, Young T, Finn L, Asher K, Hening WA, Purvis C. Epidemiology of restless legs symptoms in adults. Arch Intern Med.
2000;160:2137–2141.
7. Rothdach AJ, Trenkwalder C, Haberstock J, Keil U, Berger K.
Prevalence and risk factors of RLS in an elderly population: the
MEMO study. Memory and Morbidity in Augsburg Elderly. Neurology. 2000;54:1064–1068.
8. Merlino G, Fratticci L, Valente M, et al. Association of restless
legs syndrome in type 2 diabetes: a case–control study. Sleep.
2007;30:866–871.
9. Phillips B, Hening W, Britz P, Mannino D. Prevalence and correlates of restless legs syndrome: results from the 2005 National
Sleep Foundation Poll. Chest. 2006;129:76–80.
10. Lopes LA, Lins C de M, Adeodato VG, et al. Restless legs syndrome and quality of sleep in type 2 diabetes. Diabetes Care.
2005;28:2633–2636.
11. Ohayon MM, Roth T. Prevalence of restless legs syndrome and
periodic limb movement disorder in the general population. J
Psychosom Res. 2002;53:547–554.
12. Winkelman JW, Chertow GM, Lazarus JM. Restless legs syndrome in end-stage renal disease. Am J Kidney Dis. 1996;28:372–
378.
13. Winkelmann J, Schormair B, Lichtner P, et al. Genome-wide association study of restless legs syndrome identifies common variants in three genomic regions. Nature Genet. 2007;39:1000–1006.
14. Stefansson H, Rye DB, Hicks A, et al. A genetic risk factor for
periodic limb movements in sleep. N Engl J Med. 2007;357:639–
647.
15. Wang GJ, Volkow ND, Logan J, et al. Brain dopamine and obesity.
Lancet. 2001;357:354–357.
16. Hamdi A, Porter J, Prasad C. Decreased striatal D2 dopamine
receptors in obese Zucker rats: changes during aging. Brain Res.
1992;589:338–340.
17. Huang XF, Zavitsanou K, Huang X, et al. Dopamine transporter
and D2 receptor binding densities in mice prone or resistant to chronic high fat diet-induced obesity. Behav Brain Res.
2006;175:415–419.
18. Noble EP, Blum K, Ritchie T, Montgomery A, Sheridan PJ. Allelic
association of the D2 dopamine receptor gene with receptorbinding characteristics in alcoholism. Arch Gen Psychiatry.
1991;48:648–654.
19. Need AC, Ahmadi KR, Spector TD, Goldstein DB. Obesity is
associated with genetic variants that alter dopamine availability.
Ann Hum Genet. 2006;70:293–303.
20. Winkelman JW, Shahar E, Sharief I, Gottlieb DJ. Association of
restless legs syndrome and cardiovascular disease in the Sleep
Heart Health Study. Neurology. 2008;70:35–42.
21. Rajaram SS, Shanahan J, Ash C, Walters AS, Weisfogel G. Enhanced external counter pulsation (EECP) as a novel treatment
for restless legs syndrome (RLS): a preliminary test of the vascular neurologic hypothesis for RLS. Sleep Med. 2005;6:101–106.
22. Phillips B, Young T, Finn L, Asher K, Hening WA, Purvis C. Epi-
ENJ 2009; 1: (1). September 2009
62
www.slm-neurology.com
Obesity, Diet, and Risk of Restless Legs Syndrome
1977;16:230–246.
47. Botez MI, Lambert B. Folate deficiency and restless-legs syndrome in pregnancy. N Engl J Med. 1977;297:670.
48. Lee KA, Zaffke ME, Baratte-Beebe K. Restless legs syndrome and
sleep disturbance during pregnancy: the role of folate and iron. J
Womens Health Gend Based Med. 2001;10:335–341.
49. Patrick LR. Restless legs syndrome: pathophysiology and the role
of iron and folate. Altern Med Rev. 2007;12:101–112.
50. Coppen A, Swade C, Jones SA, Armstrong RA, Blair JA, Leeming
RJ. Depression and tetrahydrobiopterin: the folate connection. J
Affect Disord. 1989;16:103–107.
51. Mann SP, Hill MW. Activation and inactivation of striatal tyrosine
hydroxylase: the effects of pH, ATP and cyclic AMP, S-adenosylmethionine and S-adenosylhomocysteine. Biochem Pharmacol.
1983;32:3369–3374.
52. Earley CJ, Hyland K, Allen RP. Circadian changes in CSF
dopaminergic measures in restless legs syndrome. Sleep Med.
2006;7:263–268.
53. Duan W, Ladenheim B, Cutler RG, Kruman, II, Cadet JL,
Mattson MP. Dietary folate deficiency and elevated homocysteine
levels endanger dopaminergic neurons in models of Parkinson’s
disease. J Neurochem. 2002;80:101–110.
54. Parnetti L, Bottiglieri T, Lowenthal D. Role of homocysteine in
age-related vascular and non-vascular diseases. Aging (Milano).
1997;9:241–257.
55. Lipton SA, Kim WK, Choi YB, et al. Neurotoxicity associated
with dual actions of homocysteine at the N-methyl-D-aspartate
receptor. Proc Natl Acad Sci USA. 1997;94:5923–5928.
56. Beal MF, Kowall NW, Swartz KJ, Ferrante RJ. Homocysteic
acid lesions in rat striatum spare somatostatin-neuropeptide Y
(NADPH-diaphorase) neurons. Neurosci Lett. 1990;108:36–42.
57. Bachmann CG, Guth N, Helmschmied K, Armstrong VW, Paulus
www.slm-neurology.com
W, Happe S. Homocysteine in restless legs syndrome. Sleep Med.
2008;9:388–392.
58. Popoviciu L, Asgian B, Delast-Popoviciu D, Alexandrescu A,
Petrutiu S, Bagathal I. Clinical, EEG, electromyographic and
polysomnographic studies in restless legs syndrome caused by
magnesium deficiency. Rom J Neurol Psychiatry. 1993;31:55–61.
59. Hornyak M, Voderholzer U, Hohagen F, Berger M, Riemann D.
Magnesium therapy for periodic leg movements-related insomnia and restless legs syndrome: an open pilot study. Sleep.
1998;21:501–505.
60. Bartell S, Zallek S. Intravenous magnesium sulfate may relieve restless legs syndrome in pregnancy. J Clin Sleep Med.
2006;2:187–188.
61. Walters AS, Elin RJ, Cohen B, Moller JC, Oertel W, Stiasny-Kolster K. Magnesium not likely to play a major role in the pathogenesis of restless legs syndrome: serum and cerebrospinal fluid
studies. Sleep Med. 2007;8:186–187.
62. Kapur N, Friedman R. Oral ketamine: a promising treatment for
restless legs syndrome. Anesth Analg. 2002;94:1558–1559, table
of contents.
63. Levine BS, Coburn JW. Magnesium, the mimic/antagonist to
calcium. N Engl J Med. 1984;310:1253–1255.
64. Altura BM, Altura BT. Magnesium and cardiovascular biology: an
important link between cardiovascular risk factors and atherogenesis. Cell Mol Biol Res. 1995;41:347–359.
65. Ayres S Jr, Mihan R. Leg cramps (systremma and “restless
legs” syndrome. Response to vitamin E (tocopherol). Calif Med.
1969;111:87–91.
66. Gao X, Chen H, Choi HK, Curhan G, Schwarzschild MA, Ascherio
A. Diet, urate, and Parkinson’s disease risk in men. Am J Epidemiol. 2008;167:831–838.
63
ENJ 2009; 1: (1). September 2009
European Neurological Journal
review article
Neuroimaging of Primary Progressive Aphasia
Jonathan D Rohrer and Nick C Fox
Affiliation: Dementia Research Centre, Department of Neurodegenerative Disease, UCL Institute of Neurology, University
College London, London, UK
Submission date: 17th July 2009, Revision date: 20th August 2009, Acceptance date: 1st September 2009
A B STRA C T
In this review, we discuss the neuroimaging features of primary progressive aphasia (PPA), a group of neurodegenerative
disorders characterized by an initial speech and language deficit. The PPA syndromes, semantic dementia (SD), progressive non-fluent aphasia (PNFA), and logopenic/phonological aphasia (LPA), are defined clinically, but there are emerging
patterns of clinico-imaging–pathological correlation. Each of the PPA subtypes has a distinctive initial pattern of atrophy
or hypometabolism affecting the left hemisphere language network that is consistent with the speech and language and
other cognitive/behavioral deficits present: SD is associated with disease affecting the anteroinferior temporal lobes, PNFA
with the left insula and inferior frontal lobes, LPA with the left posterior superior temporal and inferior parietal lobes, and
familial PPA caused by mutations in the progranulin gene with the left frontal, temporal, and parietal lobes. As we stand
on the verge of clinical trials in PPA, the combination of structural, functional, and molecular imaging holds the promise of
defining in vivo cohorts that are likely to have a common pathological target for disease-modifying treatments.
Keywords: primary progressive aphasia, frontotemporal dementia, frontotemporal lobar degeneration, logopenic aphasia, semantic dementia, progressive non-fluent aphasia
Correspondence: Nick Fox, Dementia Research Centre, Institute of Neurology, Queen Square, London WC1N 3BG, UK; Tel:
+44-207-829-8773; fax: +44-207-676-2066; e-mail: nfox@dementia.ion.ucl.ac.uk
INTRODUCTION
more heterogeneous. In the 1998 Neary criteria, the key
features of PNFA were agrammatism, phonemic paraphasias, and anomia [4], and PNFA is often known as
the agrammatic variant of PPA [11] with the underlying pathology either tau or TDP-43 pathology. However,
it has recently been recognized that one of the major
features of such patients is a motor speech impairment,
often characterized as an apraxia of speech [7, 18, 19],
and there are undoubtedly some patients in whom this
is the major cause of their speech production deficit either in combination with agrammatism/aphasia or independent of a true aphasia [18]. This appears to be
an important distinction as there is clear evidence that
the presence of an apraxia of speech is predictive of tau
pathology (corticobasal degeneration (CBD), progressive supranuclear palsy (PSP), or Pick’s disease) at post
mortem [18]. Although there had been limited reports
of a third variant of PPA earlier [1, 10], it was not until
the study by Gorno-Tempini et al in 2004 [7] that what
is now known as LPA or the logopenic/phonological variant of PPA was described in detail. The same group have
published a number of follow-up studies [8, 20–22], but
LPA remains a little studied disorder at present. One of
the main characteristics of this disorder is the presence
of word-finding pauses in speech giving the impression
of non-fluency. However, as there is no motor speech deficit or agrammatism, when patients do speak, the flow of
The term primary progressive aphasia (PPA) describes
a group of neurodegenerative disorders in which the
major cognitive domain involved is language [1–3]. It
overlaps with the frontotemporal lobar degeneration
(FTLD) spectrum of disorders [4] sharing the same
pathological and genetic causes [5]. The most well-defined clinical syndromes falling within the PPA group
are semantic dementia (SD), progressive non-fluent
aphasia (PNFA), and logopenic/phonological aphasia
(LPA) [6–9], although there remains some disagreement
over the exact number and types of syndromes that exist
within the PPA spectrum [10, 11]. Progressive impairment of semantic knowledge was initially described in
1975 [12], but was not given the name SD until much
later [13, 14] Early features in this disorder are empty,
circumlocutory speech with anomia and single word
comprehension difficulties secondary to verbal semantic
impairment [9]. A progressive fluent aphasia is the most
obvious initial feature of the disorder, leading many to
label SD the fluent variant of PPA. However, early nonverbal semantic impairment is also common, and this
becomes more prominent as the disease progresses [15,
16]. SD is a relatively homogeneous disorder with characteristic clinical and pathological features, usually being associated with TDP-43 pathology [9, 17]. Patients
with impairment of speech output (PNFA and LPA) are
ENJ 2009; 1: (1). September 2009
65
www.slm-neurology.com
European Neurological Journal
Figure 1. L ongitudinal series of coronal and axial T1 MR images from pathologically confirmed patients with SD (TDP-43-positive pathology type 1, Sampathu classification), PNFA (tau-positive Pick’s disease), LPA (Alzheimer’s disease pathology), and a patient
with PPA secondary to a progranulin mutation. Three scans, registered into the same space and separated by approximately 1
year, are shown in order to highlight the progression in atrophy, as described in the summary section. The images are shown in
radiological convention, i.e., left hemisphere on the right of the picture
This review concentrates on studies of neuroimaging
in PPA. The most common type has been cross-sectional
structural magnetic resonance imaging (MRI) or functional (positron emission tomography (PET)/single photon emission computed tomography (SPECT)) imaging
studies of patterns of atrophy or hypometabolism. These
studies, which provide insight into the topography of
neuronal loss or dysfunction in PPA, will be reviewed
initially in the next section. The majority of these studies have used voxel-wise whole-brain imaging methods
such as statistical parametric mapping (SPM), but some
have also looked at particular regions of interest such as
specific temporal lobe structures. More recently, there
have been a number of longitudinal imaging studies,
with some also providing data that allow estimates of
sample sizes that would be needed in clinical trials of
PPA.
speech is relatively fluent [8]. Impaired short-term phonological memory is also a key feature with associated
impaired sentence repetition and comprehension [7, 8].
The importance of separating LPA out from other PPA
disorders appears to be that a majority of the reported
cases coming to post mortem have Alzheimer’s pathology rather than the FTLD pathologies of abnormal tau or
TDP-43 inclusions [23]. It is important to note that, in a
parallel theme in the dementia literature, patients with
an atypical language variant of Alzheimer’s disease have
been described for a number of years [24, 25], mostly
in retrospective post-mortem series, and many of these
patients would undoubtedly fit the proposed criteria for
LPA. Some patients with PPA have a family history of
PPA or FTLD, and the majority of these patients seem
to have a mutation in the progranulin (GRN) gene [5].
Descriptions of such patients include the presence of a
non-fluent aphasia although with a prominent anomia
and often without motor speech impairment [26–28].
ENJ 2009; 1: (1). September 2009
The following section will look at brain–behavior cor66
www.slm-neurology.com
Neuroimaging of Primary Progressive Aphasia
a different neuroanatomical pattern, and these are described in the following sections. Examples of longitudinal series of structural images in patients with pathologically confirmed SD (TDP-43 positive type 1, Sampathu
classification), PNFA (tau-positive Pick’s disease), LPA
(Alzheimer’s pathology), and progranulin-associated
PPA are shown in Figure 1.
relations in PPA, mostly using structural imaging and
voxel-based morphometry (VBM) but also, in more recent work, using functional MRI (fMRI). These correlative brain–behavior studies provide insight into not only
the cognitive deficits that occur in PPA but also normal
language networks in the brain, providing information
complementary to other brain lesion (e.g., stroke) literature.
SD (or the temporal variant of FTLD) is the most comprehensively studied of the PPA subtypes in terms of
cross-sectional patterns of atrophy [7, 33–49]. Initial
VBM studies of clinically diagnosed SD identified an
asymmetrical pattern of atrophy affecting mainly the
anterior, inferior, and lateral temporal lobes, more so in
the left hemisphere [33, 34]. The findings of these studies were extended by detailed region of interest (ROI)
studies of temporal lobe structures, which showed that
the temporal pole, fusiform gyrus, entorhinal cortex,
inferior temporal gyrus, as well as the amygdala and
hippocampus were the most affected areas with relative
CROSS-SECTIONAL AND LONGITUDINAL
IMAGING: TOPOGRAPHY OF LOSS OR
DYSFUNCTION AND RELATIONSHIP TO
PATHOLOGY
Atrophy or hypometabolism in PPA is usually asymmetrical, being worse in the left hemisphere, with structural and functional imaging studies showing similar
findings [29, 30]. However, there are also left-handed
PPA patients described with greater right hemisphere
involvement [31, 32]. Each of the subtypes of PPA has
Figure 2. P
atterns of cortical thinning in SD (left temporal variant) compared with a control group of 29 subjects. The total group (right)
and three groups split according to disease severity (left, measured by extent of anomia on the Oldfield naming test) are
shown: group 1 (9 patients—least anomic, score >9), group 2 (11 patients, score 3–9), group 3 (8 patients—most anomic, score
<3). Effect size maps are shown with the colored bar representing percentage thickness difference. Reproduced with permission from [47]
www.slm-neurology.com
67
ENJ 2009; 1: (1). September 2009
European Neurological Journal
70). In those with the left temporal variant, there seems
to be increased right temporal lobe involvement as the
disease progresses as well as spread of atrophy within
the left hemisphere, particularly the more posterior
temporal areas and the orbitofrontal, anterior insular,
inferior frontal, and anterior cingulate lobes [21, 47,
65–67]. In the right temporal variant, limited evidence
suggests that a similar but mirror-image pattern of atrophy spread is seen [21]. Rates of whole-brain atrophy
in SD have been measured in some studies, and these
are similar to those seen in other neurodegenerative diseases (2.5% per year, [67]; 1.7% per year, [69]). Rates
of individual lobar change are greatest for the temporal
lobes [67, 70]. As SD is a relatively homogeneous clinicopathological syndrome, it is a prime candidate for clinical trials in which imaging biomarkers may well be used.
It appears that rates of temporal lobe volume change
require smaller sample sizes than whole-brain or ventricular measures [67, 69].
sparing of the superior temporal gyrus; there was also
the presence of an anteroposterior gradient with relative sparing of posterior cortical areas [35, 36]. Further
VBM studies showed that there may be involvement of
areas outside the temporal lobes in SD, particularly orbitofrontal, insular, and anterior cingulate cortices [37,
38]. This asymmetrical temporal, frontal, and anterior
cingulate pattern distinguishes SD from Alzheimer’s
disease (AD), which has more symmetrical hippocampal
atrophy involvement without an anteroposterior gradient [35] and greater posterior cingulate and parietal
lobe atrophy [39]. ROI studies (using either manual segmentation, e.g., [35, 50], or a visual rating scale, e.g.,
[51, 52]) have been shown to produce similar results to
VBM studies. Limited studies have used the more recently described technique of cortical thickness measurement [44, 47, 53], with findings similar to VBM and
ROI studies (see Figure 2).
However, one study also examined a pathologically confirmed cohort of patients with ubiquitin-positive (TDP43-positive) pathology showing the same characteristic
pattern of asymmetrical left greater than right anteroinferior temporal lobe involvement seen in the clinical cohorts [47]. A small VBM study of pathologically
confirmed patients found that patterns of atrophy were
similar in SD cases associated with both ubiquitin-positive and tau-positive FTLD pathology but, in the rare
cases with Alzheimer’s pathology, there was mainly left
hippocampal atrophy [49]. Similar patterns of asymmetrical temporal lobe hypometabolism have been found in
PET and SPECT imaging [54–57]. There have been a
couple of studies of white matter disease in SD [58, 59]
with one diffusion tensor imaging (DTI) study showing
particular involvement of the inferior longitudinal fasciculus with additional involvement of inferior frontooccipital fasciculus, callosal, and superior longitudinal
fasciculus tracts [59].
PNFA is less well studied than SD, and patterns of
neuroanatomical involvement are not quite so clear
[7, 18, 19, 40, 46–48, 71–73]. This is partly because of
the heterogeneity of PNFA and also the differences
in definition between research groups, e.g., it is likely
that patients with the LPA variant have been included
in previous studies of PNFA. Similar to SD, atrophy or
hypometabolism is usually asymmetrical and worse in
the left hemisphere. The most significantly affected areas are in the left inferior frontal lobe (particularly the
frontal opercular region) and anterior insula [7, 19, 47,
71]. However, left middle and superior frontal, superior
temporal, and caudate involvement are also frequently
reported in studies with less frequent involvement of
the anterior parietal lobes [7, 19, 47]. ROI studies are
limited in PNFA [51, 74–76], but have shown involvement of striatal structures, particularly the caudate.
Cortical thickness studies are also limited but show similar results to VBM and ROI studies [47] (see Figure 3).
There are few pathologically confirmed studies of PNFA,
and these have often studied mixed pathological groups
but, despite this, have shown fairly consistent findings
compared with the clinical studies, e.g., anterior insula
and inferior frontal involvement in mixed groups of taupositive patients [18, 47].
The majority of SD cases described in the literature
have asymmetrical left greater than right temporal
lobe atrophy, but there are a number of reports of the
opposite pattern with right greater than left temporal
lobe atrophy [7, 60–62]. This right temporal variant appears to be less common than the left temporal variant,
although this may simply represent an ascertainment
bias. (Of note, these are different from the rare lefthanded/right hemisphere-dominant individuals with semantic dementia described above.) Patients often have
initial behavioral symptoms rather than a progressive
aphasia [61], with the development of semantic impairment only later in the illness (leading some authors to
argue that this right temporal variant should be logically separated from the primary progressive aphasias,
e.g., [53]). The pattern of atrophy in these right temporal variants appears to be the mirror image of the left
temporal variant [21], although the underlying pathology remains unclear.
There are few longitudinal studies of PNFA [69, 77],
although it seems that, with disease progression, there
is spread from the left inferior frontal and insular cortex
to involve the superior temporal, middle and superior
frontal, and anterior parietal lobes [47, 77]. More posterior atrophy, particularly of the left anterior parietal
lobe, may herald the presence of an accompanying corticobasal syndrome (CBS) [77]. Rate of whole-brain atrophy is similar to SD (1.6% per year [69]), but there are
currently no studies of ROI biomarkers such as lobar or
sublobar volumes. It will be important to study PNFA
longitudinally in more detail, particularly targeting specific pathological subtypes.
Longitudinal studies in SD are less common [21, 63–
ENJ 2009; 1: (1). September 2009
68
www.slm-neurology.com
Neuroimaging of Primary Progressive Aphasia
Figure 3. P
atterns of cortical thinning in PNFA compared with a control group of 29 subjects. The total group (right) and three groups
split according to disease severity (left, measured by extent of anomia on the Oldfield naming test) are shown: group 1 (11 patients—least anomic, score >24), group 2 (11 patients, score 14–24), group 3 (6 patients—most anomic, score <14). Effect size
maps are shown with the colored bar representing percentage thickness difference. Reproduced with permission from [47]
LPA is the least studied of the three subtypes with
most imaging studies currently from the same research
group [8, 78, 79]. The most significantly atrophied areas in LPA are the left posterior superior temporal and
inferior parietal lobes and, to a lesser extent, posterior
cingulate and middle/inferior temporal lobe disease,
although the spread of atrophy outside these areas is
unclear: longitudinal studies of LPA will be required to
answer this. There are currently no large imaging studies of pathologically confirmed LPA.
aphasia and AD pathology, some of whom would have
fitted the proposed criteria for LPA, also showed left
temporo-parietal lobe atrophy [80].
GRN mutations have been associated with a nonfluent aphasia, although often with a prominent anomia [17, 26]. There are limited studies of the imaging
of GRN-PPA [28, 81, 82], although these have shown
asymmetrical left greater than right hemisphere atrophy (which may occur presymptomatically) affecting the
frontal, temporal, and (to a lesser extent) parietal lobes.
There appears to be more posterior atrophy than usually occurs in PNFA (and more anterior temporal lobe
atrophy than occurs in LPA).
However, one study used amyloid 11C PIB-PET imaging to assess the presence of amyloid pathology [20]: all
LPA patients (four out of four) had a positive PIB scan
compared with only one of six PNFA and one of five SD
patients. Interestingly, the 11C PIB binding was diffuse
in the LPA patients, similar to amnestic AD patients,
despite the presence of asymmetrical left temporo-parietal hypometabolism on FDG-PET imaging in the same
patients.
PPA, particularly PNFA, has been associated in small
studies with motor neurone disease (MND)/amyotrophic
lateral sclerosis (ALS), a disorder usually characterized
by TDP-43 pathology [83–87], with other studies highlighting an association of progressive motor speech impairment (without aphasia) with MND/ALS [88, 89].
Because these studies are small and often without de-
One retrospective study of patients with progressive
www.slm-neurology.com
69
ENJ 2009; 1: (1). September 2009
European Neurological Journal
tailed imaging, conclusions about patterns of atrophy
or hypometabolism in PPA with MND/ALS are difficult.
However, in single cases, bilateral (often worse on the
left) frontal or frontotemporal atrophy/hypometabolism
are reported [84, 86, 87].
linking such deficits with prefrontal atrophy [98–100].
Prosopagnosia develops in SD with greater right temporal lobe involvement [101], while emotional processing
deficits are also associated with right hemisphere atrophy, namely in the amygdala and orbitofrontal cortex
[38].
BRAIN–BEHAVIOR CORRELATIVE
IMAGING STUDIES
SUMMARY AND CONCLUSIONS
Each of the PPA subtypes has a distinctive pattern of
atrophy consistent with the speech and language and
other cognitive and behavioral deficits present. SD (left
temporal variant) is associated with left anterior temporal lobe disease consistent with a primary semantic
store deficit causing anomia, impaired single-word comprehension, and surface dyslexia. With disease progression, there is more posterior temporal involvement as
well as left inferior frontal, orbitofrontal, cingulate, and
right temporal lobe disease, consistent with the development of other linguistic deficits, behavioral impairment
such as disinhibition, emotional processing deficits, and
object and face associative agnosias. PNFA is associated
with anterior insula and inferior frontal lobe disease
consistent with the main deficits of agrammatism and
motor speech impairment (apraxia of speech). Spread
of disease to involve other areas in the frontal lobe,
superior temporal and anterior parietal areas, and the
caudate is consistent with the development of anomia,
impaired repetition, executive dysfunction and, later in
the disease, impaired single-word comprehension and
behavioral impairment. LPA is associated with posterior superior temporal and inferior parietal involvement
consistent with a phonological memory deficit, leading
to impaired sentence repetition and comprehension, as
well as impaired phonological access leading to anomia.
Patients with PPA have a wide variety of speech and
language deficits that differ between the subtypes: anomia and impaired single-word comprehension secondary
to a verbal semantic deficit in SD; agrammatism, motor speech impairment, anomia, and impaired repetition
in PNFA; and anomia and impaired sentence repetition
and comprehension secondary to a phonological memory deficit in LPA. Consistent with this, both structural
and functional MRI studies have shown involvement of
a distributed left hemisphere fronto-temporo-parietal
language network in PPA [90–92]. During the early
stages of the disease, all of the disorders have naming
deficits with anomia worse in SD than LPA and only a
mild impairment in PNFA. VBM studies suggest that
overlapping but distinct areas of the language network
correlate with anomia [6, 36, 93, 94]: in SD, anomia is
mostly associated with anterior temporal lobe atrophy,
whereas in PNFA, a more widespread network of areas is associated with anomia, particularly the inferior
frontal, lateral temporal, and anterior parietal lobes.
Semantic impairment in SD is associated with anterior
temporal lobe atrophy [43, 95]. Remarkably, there are
few studies of spontaneous speech in PPA, and those
performed have looked mainly at PNFA: apraxia of
speech has been associated with premotor and supplementary motor areas [18] as well as the insula and basal
ganglia [19], whereas early mutism in PNFA has been
associated with left pars opercularis and basal ganglia
atrophy [96]. Sentence comprehension is impaired in
both PNFA (for complex sentences) and LPA (for simple
and complex sentences) with one small fMRI study of
PNFA showing decreased activation in the left ventral
inferior frontal lobe areas known to be associated with
grammatical processing [97]. Reading deficits differ between the subtypes: surface dyslexia is seen in SD (i.e.,
inability to read irregular or exception words) and, in an
fMRI study, a group of SD patients (unlike cognitively
normal control subjects) did not activate anterior temporal lobe areas thought to be required for exception
word reading, but instead activated a left inferior parietal area not seen in normal individuals (which may
explain the regularization of exception words that SD
patients commonly exhibit) [79]. Phonological dyslexia
is seen in PNFA and LPA, i.e., particular difficulty reading nonsense or pseudowords, and is associated in PPA
with left temporo-parietal atrophy [21].
There still remain a number of unanswered questions
in the field of PPA neuroimaging. In particular, better
characterization of the LPA variant over longitudinal
studies is important, as well as its relationship to the
other variants. Similarly, it will be extremely useful to
study larger pathologically confirmed cohorts to identify ways of determining the underlying pathology of a
PPA syndrome in life, e.g., whether structural imaging
can distinguish TDP-43 SD from Pick’s disease SD, or
CBD PNFA from PSP PNFA. One way to study this may
be using classification methods such as support vector
machine learning algorithms. Such a study has been
performed recently in clinically described syndromes
of PPA (PNFA, SD, and LPA) [22] showing an accuracy
around 90–100% for most comparisons, although lower
at just over 80% for the comparison of PNFA and LPA.
The development of further molecular imaging methods
will also be important: initially, amyloid imaging will
help to define those with Alzheimer pathology with future tau or TDP-43 molecular imaging helping to distinguish other pathologies in life. Lastly, neuroimaging can
provide insights into the underlying vulnerable neural
networks affected in neurodegenerative disease: a re-
Non-linguistic deficits are also seen in PPA as the disease progresses. Executive dysfunction is seen in PNFA,
and patients have been included in correlative studies
ENJ 2009; 1: (1). September 2009
70
www.slm-neurology.com
Neuroimaging of Primary Progressive Aphasia
cent important study has shown that the VBM-derived
patterns of atrophy seen in PNFA and SD are similar
to structurally and functionally connected neural networks seen in cognitively normal individuals [102], suggesting that the PPA subtypes affect distinct selectively
vulnerable networks within the brain.
19.
20.
Disclosures: The authors have no commercial financial interests to disclose in relation to the contents of this article.
21.
Acknowledgments: This work was undertaken at UCLH/
UCL, which received a proportion of funding from the Department of Health’s NIHR Biomedical Research Centres funding
scheme. The Dementia Research Centre is an Alzheimer’s Research Trust Co-ordinating Centre. This work was also funded
by the Medical Research Council UK. JDR is supported by a
Brain Exit Scholarship.
22.
23.
24.
REFERENCES
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
25.
Mesulam MM. Slowly progressive aphasia without generalized
dementia. Ann Neurol. 1982;11(6):592–598.
Mesulam MM. Primary progressive aphasia. Ann Neurol.
2001;49(4):425–432.
Mesulam MM. Primary progressive aphasia–a language-based
dementia. N Engl J Med. 2003;349(16):1535–1542.
Neary D, Snowden JS, Gustafson L, et al. Frontotemporal lobar
degeneration: a consensus on clinical diagnostic criteria. Neurology. 1998;51:1546–1554.
Mackenzie IR, Rademakers R. The molecular genetics and
neuropathology of frontotemporal lobar degeneration: recent
developments. Neurogenetics. 2007;8(4):237–248.
Grossman M, McMillan C, Moore P, et al. What’s in a name: voxel-based morphometric analyses of MRI and naming difficulty in
Alzheimer’s disease, frontotemporal dementia and corticobasal
degeneration. Brain. 2004;127(Pt 3):628–649.
Gorno-Tempini ML, Dronkers NF, Rankin KP, et al. Cognition
and anatomy in three variants of primary progressive aphasia.
Ann Neurol. 2004;55(3):335–346.
Gorno-Tempini ML, Brambati SM, Ginex V, et al. The logopenic/
phonological variant of primary progressive aphasia. Neurology.
2008;71(16):1227–1234.
Hodges JR, Patterson K. Semantic dementia: a unique clinicopathological syndrome. Lancet Neurol. 2007;6(11):1004–1014.
Kertesz A, Davidson W, McCabe P, Takagi K, Munoz D. Primary
progressive aphasia: diagnosis, varieties, evolution. J Int Neuropsychol Soc. 2003;9(5):710–719.
Mesulam MM. Primary progressive aphasia: a 25-year retrospective. Alzheimer Dis Assoc Disord. 2007;21(4):S8–S11.
Warrington EK. The selective impairment of semantic memory.
Q J Exp Psychol. 1975;27(4):635–657.
Snowden JS, Goulding PJ, Neary D. Semantic dementia: a form
of circumscribed cerebral atrophy. Behav Neurol. 1989;2:167–
182.
Hodges JR, Patterson K, Oxbury S, Funnell E. Semantic dementia. Progressive fluent aphasia with temporal lobe atrophy.
Brain. 1992;115(Pt 6):1783–1806.
Bozeat S, Lambon Ralph MA, Patterson K, Garrard P, Hodges
JR. Non-verbal semantic impairment in semantic dementia.
Neuropsychologia. 2000;38(9):1207–1215.
Adlam AL, Patterson K, Rogers TT, et al. Semantic dementia
and fluent primary progressive aphasia: two sides of the same
coin? Brain. 2006;129:3066–3080.
Snowden J, Neary D, Mann D. Frontotemporal lobar degeneration: clinical and pathological relationships. Acta Neuropathol.
2007;114(1):31–38.
Josephs KA, Duffy JR, Strand EA, et al. Clinicopathological and
www.slm-neurology.com
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
71
imaging correlates of progressive aphasia and apraxia of speech.
Brain. 2006;129(Pt 6):1385–1398.
Ogar JM, Dronkers NF, Brambati SM, Miller BL, Gorno-Tempini
ML. Progressive nonfluent aphasia and its characteristic motor
speech deficits. Alzheimer Dis Assoc Disord. 2007;21(4):S23–30.
Rabinovici GD, Jagust WJ, Furst AJ, et al. Abeta amyloid and
glucose metabolism in three variants of primary progressive
aphasia. Ann Neurol. 2008;64(4):388–401.
Brambati SM, Rankin KP, Narvid J, et al. Atrophy progression in semantic dementia with asymmetric temporal involvement: a tensor-based morphometry study. Neurobiol Aging.
2009;30(1):103–111.
Wilson SM, Ogar JM, Laluz V, et al. Automated MRI-based classification of primary progressive aphasia variants. Neuroimage.
2009;47(4):1558-67.
Mesulam M, Wicklund A, Johnson N, et al. Alzheimer and frontotemporal pathology in subsets of primary progressive aphasia.
Ann Neurol. 2008;63(6):709–719.
Galton CJ, Patterson K, Xuereb JH, Hodges JR. Atypical and
typical presentations of Alzheimer’s disease: a clinical, neuropsychological, neuroimaging and pathological study of 13 cases.
Brain. 2000;123(3):484–498.
Alladi S, Xuereb J, Bak T, et al. Focal cortical presentations of
Alzheimer’s disease. Brain. 2007;130(Pt 10):2636–2645.
Snowden JS, Pickering-Brown SM, Mackenzie IR, et al.
Progranulin gene mutations associated with frontotemporal dementia and progressive non-fluent aphasia. Brain. 2006;129(Pt
11):3091–3102.
Snowden JS, Pickering-Brown SM, Du Plessis D, et al. Progressive anomia revisited: focal degeneration associated with
progranulin gene mutation. Neurocase. 2007;13(5):366–377.
Rohrer JD, Warren JD, Barnes J, et al. Mapping the progression
of progranulin-associated frontotemporal lobar degeneration.
Nature Clin Pract Neurol. 2008;4(8):455–460.
Desgranges B, Matuszewski V, Piolino P, et al. Anatomical and
functional alterations in semantic dementia: a voxel-based MRI
and PET study. Neurobiol Aging. 2007;28(12):1904–1913.
Panegyres PK, McCarthy M, Campbell A, Lenzo N, Fallon M,
Thompson J. Correlative studies of structural and functional
imaging in primary progressive aphasia. Am J Alzheimers Dis
Other Demen. 2008;23(2):184–191.
Drzezga A, Grimmer T, Siebner H, Minoshima S, Schwaiger M,
Kurz A. Prominent hypometabolism of the right temporoparietal and frontal cortex in two left-handed patients with primary
progressive aphasia. J Neurol. 2002;249(9):1263–1267.
Mesulam M, Weintraub S, Parrish T, Gitelman D. Primary progressive aphasia: reversed asymmetry of atrophy and right hemisphere language dominance. Neurology. 2005;64(3):556–557.
Mummery CJ, Patterson K, Wise RJ, Vandenberghe R, Price CJ,
Hodges JR. Disrupted temporal lobe connections in semantic
dementia. Brain. 1999;122(Pt 1):61–73.
Mummery CJ, Patterson K, Price CJ, Ashburner J, Frackowiak
RS, Hodges JR. A voxel-based morphometry study of semantic
dementia: relationship between temporal lobe atrophy and
semantic memory. Ann Neurol. 2000;47(1):36–45.
Chan D, Fox NC, Scahill RI, et al. Patterns of temporal lobe atrophy in semantic dementia and Alzheimer’s disease. Ann Neurol.
2001;49(4):433–442.
Galton CJ, Patterson K, Graham K, et al. Differing patterns of
temporal atrophy in Alzheimer’s disease and semantic dementia.
Neurology. 2001;57(2):216–225.
Rosen HJ, Gorno-Tempini ML, Goldman WP, et al. Patterns of
brain atrophy in frontotemporal dementia and semantic dementia. Neurology. 2002;58(2):198–208.
Rosen HJ, Perry RJ, Murphy J, et al. Emotion comprehension
in the temporal variant of frontotemporal dementia. Brain.
2002;125(Pt 10):2286–2295.
Boxer AL, Rankin KP, Miller BL, et al. Cinguloparietal atrophy
distinguishes Alzheimer disease from semantic dementia. Arch
Neurol. 2003;60(7):949–956.
ENJ 2009; 1: (1). September 2009
European Neurological Journal
40. Grossman M, Ash S. Primary progressive aphasia: a review.
Neurocase. 2004;10(1):3–18.
41. Halpern CH, Glosser G, Clark R, et al. Dissociation of numbers
and objects in corticobasal degeneration and semantic dementia.
Neurology. 2004;62(7):1163–1169.
42. Studholme C, Cardenas V, Blumenfeld R, et al. Deformation
tensor morphometry of semantic dementia with quantitative
validation. Neuroimage. 2004;21(4):1387–1398.
43. Davies RR, Graham KS, Xuereb JH, Williams GB, Hodges JR.
The human perirhinal cortex and semantic memory. Eur J Neurosci. 2004;20(9):2441–2446.
44. Gold BT, Balota DA, Cortese MJ, et al. Differing neuropsychological and neuroanatomical correlates of abnormal reading in
early-stage semantic dementia and dementia of the Alzheimer
type. Neuropsychologia. 2005;43(6):833–846.
45. Short RA, Broderick DF, Patton A, Arvanitakis Z, Graff-Radford
NR. Different patterns of magnetic resonance imaging atrophy
for frontotemporal lobar degeneration syndromes. Arch Neurol.
2005;62(7):1106–1110.
46. Schroeter ML, Raczka K, Neumann J, Yves von Cramon
D. Towards a nosology for frontotemporal lobar degenerations-a meta-analysis involving 267 subjects. Neuroimage.
2007;36(3):497–510.
47. Rohrer JD, Warren JD, Modat M, et al. Patterns of cortical thinning in the language variants of frontotemporal lobar degeneration. Neurology. 2009;72(18):1562–1569.
48. Lindberg O, Ostberg P, Zandbelt BB, et al. Cortical morphometric subclassification of frontotemporal lobar degeneration. AJNR
Am J Neuroradiol. 2009; 30(6):1233-9.
49. Pereira JM, Williams GB, Acosta-Cabronero J, et al. Atrophy patterns in histologic vs clinical groupings of frontotemporal lobar
degeneration. Neurology. 2009;72(19):1653–1660.
50. Good CD, Scahill RI, Fox NC, et al. Automatic differentiation of
anatomical patterns in the human brain: validation with studies
of degenerative dementias. Neuroimage. 2002;17(1):29–46.
51. van de Pol LA, Hensel A, van der Flier WM, et al. Hippocampal atrophy on MRI in frontotemporal lobar degeneration
and Alzheimer’s disease. J Neurol Neurosurg Psychiatry.
2006;77(4):439–442.
52. Davies RR, Scahill VL, Graham A, Williams GB, Graham KS,
Hodges JR. Development of an MRI rating scale for multiple
brain regions: comparison with volumetrics and with voxelbased morphometry. Neuroradiology. 2009; 51(8):491-503.
53. Mesulam M, Rogalski E, Wieneke C, et al. Neurology of anomia
in the semantic variant of primary progressive aphasia. Brain.
2009; 132(Pt 9):2553-65.
54. Soriani-Lefèvre MH, Hannequin D, Bakchine S, et al. Evidence
of bilateral temporal lobe involvement in primary progressive
aphasia: a SPECT study. J Nucl Med. 2003;44(7):1013–1022.
55. Diehl J, Grimmer T, Drzezga A, Riemenschneider M, Förstl H,
Kurz A. Cerebral metabolic patterns at early stages of frontotemporal dementia and semantic dementia. A PET study.
Neurobiol Aging. 2004;25(8):1051–1056.
56. Clark DG, Charuvastra A, Miller BL, Shapira JS, Mendez MF.
Fluent versus nonfluent primary progressive aphasia: a comparison of clinical and functional neuroimaging features. Brain
Lang. 2005;94(1):54–60.
57. Drzezga A, Grimmer T, Henriksen G, et al. Imaging of amyloid
plaques and cerebral glucose metabolism in semantic dementia
and Alzheimer’s disease. Neuroimage. 2008;39(2):619–633.
58. Chao LL, Schuff N, Clevenger EM, et al. Patterns of white matter atrophy in frontotemporal lobar degeneration. Arch Neurol.
2007;64(11):1619–1624.
59. Borroni B, Brambati SM, Agosti C, et al. Evidence of white
matter changes on diffusion tensor imaging in frontotemporal
dementia. Arch Neurol. 2007;64(2):246–251.
60. Thompson SA, Patterson K, Hodges JR. Left/right asymmetry of
atrophy in semantic dementia: behavioral-cognitive implications.
Neurology. 2003;61(9):1196–1203.
61. Seeley WW, Bauer AM, Miller BL, et al. The natural history
ENJ 2009; 1: (1). September 2009
62.
63.
64.
65.
66.
67.
68.
69.
70.
71.
72.
73.
74.
75.
76.
77.
78.
79.
80.
81.
72
of temporal variant frontotemporal dementia. Neurology.
2005;64(8):1384–1390.
Chan D, Anderson V, Pijnenburg Y, et al. The clinical profile of
right temporal lobe atrophy. Brain. 2009;132(Pt 5):1287–1298.
Chan D, Fox NC, Jenkins R, Scahill RI, Crum WR, Rossor MN.
Rates of global and regional cerebral atrophy in AD and frontotemporal dementia. Neurology. 2001;57(10):1756–1763.
Whitwell JL, Anderson VM, Scahill RI, Rossor MN, Fox NC.
Longitudinal patterns of regional change on volumetric MRI
in frontotemporal lobar degeneration. Dement Geriatr Cogn
Disord. 2004;17(4):307–310.
Diehl-Schmid J, Grimmer T, Drzezga A, et al. Longitudinal
changes of cerebral glucose metabolism in semantic dementia.
Dement Geriatr Cogn Disord. 2006;22(4):346–351.
Bright P, Moss HE, Stamatakis EA, Tyler LK. Longitudinal
studies of semantic dementia: the relationship between structural and functional changes over time. Neuropsychologia.
2008;46(8):2177–2188.
Rohrer JD, McNaught E, Foster J, et al. Tracking progression
in frontotemporal lobar degeneration: serial MRI in semantic
dementia. Neurology. 2008;71(18):1445–1451.
Czarnecki K, Duffy JR, Nehl CR, et al. Very early semantic
dementia with progressive temporal lobe atrophy: an 8-year
longitudinal study. Arch Neurol. 2008;65(12):1659–1663.
Knopman DS, Jack CR Jr, Kramer JH, et al. Brain and ventricular volumetric changes in frontotemporal lobar degeneration
over 1 year. Neurology. 2009;72(21):1843–1849.
Krueger CE, Dean DL, Rosen HJ, et al. Longitudinal rates of
lobar atrophy in frontotemporal dementia, semantic dementia,
and Alzheimer’s disease. Alzheimer Dis Assoc Disord. 2009;Jun
30. [Epub ahead of print]
Nestor PJ, Graham NL, Fryer TD, Williams GB, Patterson K,
Hodges JR. Progressive non-fluent aphasia is associated with
hypometabolism centred on the left anterior insula. Brain.
2003;126(Pt 11):2406–2418.
Zahn R, Buechert M, Overmans J, et al. Mapping of temporal and parietal cortex in progressive nonfluent aphasia and
Alzheimer’s disease using chemical shift imaging, voxel-based
morphometry and positron emission tomography. Psychiatry
Res. 2005;140(2):115–131.
Nestor PJ, Balan K, Cheow HK, et al. Nuclear imaging can
predict pathologic diagnosis in progressive nonfluent aphasia.
Neurology. 2007;68(3):238–239.
Looi JC, Lindberg O, Zandbelt BB, et al. Caudate nucleus volumes in frontotemporal lobar degeneration: differential atrophy
in subtypes. AJNR Am J Neuroradiol. 2008;29(8):1537–1543.
Chow TW, Izenberg A, Binns MA, et al. Magnetic resonance
imaging in frontotemporal dementia shows subcortical atrophy.
Dement Geriatr Cogn Disord. 2008;26(1):79–88.
Looi JC, Svensson L, Lindberg O, et al. Putaminal volume in
frontotemporal lobar degeneration and Alzheimer disease: differential volumes in dementia subtypes and controls. AJNR Am
J Neuroradiol. 2009;Jun 4. [Epub ahead of print]
Gorno-Tempini ML, Murray RC, Rankin KP, Weiner MW, Miller
BL. Clinical, cognitive and anatomical evolution from nonfluent progressive aphasia to corticobasal syndrome: a case report.
Neurocase. 2004;10(6):426–436.
Gorno-Tempini ML, Rankin KP, Woolley JD, Rosen HJ, Phengrasamy L, Miller BL. Cognitive and behavioral profile in a
case of right anterior temporal lobe neurodegeneration. Cortex.
2004;40(4-5):631–644.
Wilson SM, Brambati SM, Henry RG, et al. The neural basis
of surface dyslexia in semantic dementia. Brain. 2009;132(Pt
1):71–86.
Josephs KA, Whitwell JL, Duffy JR, et al. Progressive aphasia
secondary to Alzheimer disease vs FTLD pathology. Neurology.
2008;70(1):25–34.
Cruchaga C, Fernández-Seara MA, Seijo-Martínez M, et al.
Cortical Atrophy and language network reorganization associated with a novel progranulin mutation. Cereb Cortex.
www.slm-neurology.com
Neuroimaging of Primary Progressive Aphasia
2009;19(8):1751-60.
82. Borroni B, Alberici A, Premi E, et al. Brain magnetic resonance
imaging structural changes in a pedigree of asymptomatic progranulin mutation carriers. Rejuvenation Res. 2008;11(3):585–
595.
83. Caselli RJ, Windebank AJ, Petersen RC, et al. Rapidly progressive aphasic dementia and motor neuron disease. Ann Neurol.
1993;33(2):200–207.
84. Bak TH, O’Donovan DG, Xuereb JH, Boniface S, Hodges JR.
Selective impairment of verb processing associated with pathological changes in Brodmann areas 44 and 45 in the motor neurone disease-dementia-aphasia syndrome. Brain. 2001;124(Pt
1):103–120.
85. Tsuchiya K, Ozawa E, Fukushima J, et al. Rapidly progressive
aphasia and motor neuron disease: a clinical, radiological, and
pathological study of an autopsy case with circumscribed lobar
atrophy. Acta Neuropathol. 2000;99(1):81–87.
86. Catani M, Piccirilli M, Geloso MC, et al. Rapidly progressive
aphasic dementia with motor neuron disease: a distinctive clinical entity. Dement Geriatr Cogn Disord. 2004;17(1–2):21–28.
87. da Rocha AJ, Valério BC, Buainain RP, et al. Motor neuron
disease associated with non-fluent rapidly progressive aphasia: case report and review of the literature. Eur J Neurol.
2007;14(9):971–975.
88. Duffy J. Apraxia of speech in degenerative neurologic disease.
Aphasiology. 2006;20:511–527.
89. Duffy JR, Peach RK, Strand EA. Progressive apraxia of speech
as a sign of motor neuron disease. Am J Speech Lang Pathol.
2007;16:198–208.
90. Sonty SP, Mesulam MM, Thompson CK, et al. Primary progressive aphasia: PPA and the language network. Ann Neurol.
2003;53(1):35–49.
91. Vandenbulcke M, Peeters R, Van Hecke P, Vandenberghe R. Anterior temporal laterality in primary progressive aphasia shifts
to the right. Ann Neurol. 2005;58(3):362–370.
92. Sonty SP, Mesulam MM, Weintraub S, Johnson NA, Parrish
www.slm-neurology.com
TB, Gitelman DR. Altered effective connectivity within the
language network in primary progressive aphasia. J Neurosci.
2007;27(6):1334–1345.
93. McMillan C, Gee J, Moore P, Dennis K, DeVita C, Grossman M.
Confrontation naming and morphometric analyses of structural
MRI in frontotemporal dementia. Dement Geriatr Cogn Disord.
2004;17(4):320–323.
94. Amici S, Ogar J, Brambati SM, et al. Performance in specific
language tasks correlates with regional volume changes in progressive aphasia. Cogn Behav Neurol. 2007;20(4):203–211.
95. Nestor PJ, Fryer TD, Hodges JR. Declarative memory impairments in Alzheimer’s disease and semantic dementia. Neuroimage. 2006;30(3):1010–1020.
96. Gorno-Tempini ML, Ogar JM, Brambati SM, et al. Anatomical
correlates of early mutism in progressive nonfluent aphasia.
Neurology. 2006;67(10):1849–1851.
97. Cooke A, DeVita C, Gee J, et al. Neural basis for sentence comprehension deficits in frontotemporal dementia. Brain Lang.
2003;85(2):211–221.
98. Kramer JH, Quitania L, Dean D, et al. Magnetic resonance
imaging correlates of set shifting. J Int Neuropsychol Soc.
2007;13(3):386–392.
99. Fine EM, Delis DC, Dean D, et al. Left frontal lobe contributions
to concept formation: a quantitative MRI study of performance
on the Delis–Kaplan Executive Function System Sorting Test. J
Clin Exp Neuropsychol. 2008; 31(5):624-31.
100.Listerud J, Powers C, Moore P, Libon DJ, Grossman M. Neuropsychological patterns in magnetic resonance imaging-defined
subgroups of patients with degenerative dementia. J Int Neuropsychol Soc. 2009;15(3):459–470.
101.Josephs KA, Whitwell JL, Vemuri P, et al. The anatomic correlate of prosopagnosia in semantic dementia. Neurology.
2008;71(20):1628–1633.
102.Seeley WW, Crawford RK, Zhou J, Miller BL, Greicius MD.
Neurodegenerative diseases target large-scale human brain
networks. Neuron. 2009;62(1):42–52.
73
ENJ 2009; 1: (1). September 2009
European Neurological Journal
review article
Subcortical Gliosis and Leukodystrophy Overlap
Syndromes as a Cause of Late-onset Dementia
Russell H Swerdlow1, Bradley B Miller2, H Robert Brashear3 and Jeffrey M Burns1
Affiliations: : 1Departments of Neurology and Molecular and Integrative Physiology, University of Kansas School of Medicine,
Kansas City, KS, USA; 2Department of Pathology, University of Virginia Health System, Charlottesville, VA, USA; 3Janssen
Pharmaceutica, Titusville, NJ, USA
Submission date: 5th August 2009, Acceptance date: 24th August 2009
A B STRA C T
Neuroimaging is a routine part of the dementia syndrome evaluation. It frequently reveals perturbed white matter integrity. These perturbations, commonly referred to as leukoaraiosis or white matter disease, are often attributed to microvascular ischemia. When present, clinicians must decide whether white matter changes relate etiologically to clinical
cognitive decline. We recently described a large autosomal dominant kindred in which multiple affected members were
diagnosed initially with vascular dementia due to subcortical microischemic white matter disease. Subsequent brain autopsies revealed subcortical gliosis, features suggestive of late-onset leukodystrophy, and a lack of microvascular ischemic
disease. This review emphasizes that what is radiographically classified as microischemic white matter disease is not necessarily due to ischemia, and discusses the emerging realization that subcortical gliosis–leukodystrophy overlap syndromes
can cause late-life dementia.
Keywords: dementia, frontotemporal dementia, hereditary leukodystrophy with axonal spheroids, leukoaraiosis, leukodystrophy, subcortical gliosis, white matter disease
Correspondence: Russell H Swerdlow, Landon Center on Aging, MS 2012, 3901 Rainbow Blvd, Kansas City, KS 66160, USA.
Tel: +1-912-588-6970; fax: +1-913-588-0681; e-mail: rswerdlow@kumc.edu
INTRODUCTION: AN AUTOSOMAL
DOMINANT KINDRED WITH
FRONTOTEMPORAL DEMENTIA
AND LEUKOARAIOSIS
ly impaired retention and very impaired retrieval abilities. Freehand and copy drawing were good. He named
11 animals over 1 min and one F word over 1 min. He
performed poorly on bedside tests of set-shifting and
motor sequencing and showed ideomotor apraxia. Aside
from decreased ability to discriminate odors and mild
gegenhalten paratonia, his general neurologic examination was fairly unremarkable. The patient’s behavioral
and cognitive deficits progressively declined over the
next 4.5 years. He was admitted to a nursing home approximately 7 years after symptom onset, and died of
pulmonary complications at the age of 66.
A 61-year-old man was referred to a Memory Disorders
Clinic for progressive behavioral and personality changes. Symptoms had begun 3 years earlier and included
changes in hygiene. He had stopped bathing voluntarily and would only take “sponge baths”. He lost interest in his hobbies and developed stereotyped behaviors,
such as making daily visits to a discount store “whether
he needed something or not”. He started using expletives in his daily conversation, to the point it began to
hurt his business. He lost the ability to joke, gained 30
pounds, and could not “finish what he started”. A magnetic resonance imaging (MRI) scan obtained at the age
of 61 years showed leukoaraiosis of the subcortical white
matter capping the frontal horns (Figure 1). The MRI
white matter abnormalities were initially felt to represent a consequence of microvascular ischemia and to
suggest a diagnosis of vascular dementia.
The subject belonged to an extended kindred in which
affected members developed progressive dementia and
leukoaraiosis. Several kindred members had been diagnosed initially with vascular dementia based on radiographic findings, but autopsy studies instead revealed
profound subcortical gliosis and features typical of leukodystrophy. Based on the family history, the subject’s
diagnosis of microvascular dementia was changed to
subcortical gliosis. The subject’s brain autopsy ultimately also revealed extensive subcortical gliosis, findings
suggestive of leukodystrophy, and no vascular pathology.
His examination at the age of 61 showed anosognosia.
He scored 25/30 on the Mini Mental State Examination
(MMSE). Performance on memory testing showed mildENJ 2009; 1: (1). September 2009
The subject’s extended kindred was recently reported
[1]. Cognitive decline in autopsy-characterized members
75
www.slm-neurology.com
European Neurological Journal
dorsolateral prefrontal cortices proceed subcortically to
and through the basal ganglia and thalamus (Figure 2),
memory retrieval deficits with spared memory retention
are often characterized as frontal–subcortical dysfunction [3].
Figure 1. Frontal leuko­
araiosis is present on
the patient’s MRI
Frontal–subcortical circuitry projects anterior to the
lateral ventricle frontal horns, areas that frequently
show leukoaraiosis on neuroimaging studies of elderly
individuals. Given the profound leukoaraiosis in the patient described in the clinical vignette, it is not surprising that memory retrieval was impaired.
Other aspects of this subject’s clinical examination
suggest a disproportionate decline in executive skills
and also indicate frontal–subcortical dysfunction. Setshifting, motor sequencing, and ideomotor praxis are
skills that typically require intact DLPC function or
preserved projections leading to or from the DLPC. Poor
naming to letter (word association) with a more preserved ability to name to set (semantic fluency) suggests
frontal–subcortical dysfunction, as naming to letter is
more dependent on one’s frontal-mediated ability to organize word retrieval than is naming to set [4].
Figure 2. Frontal–subcortical
circuit that subserves memory retrieval. The dorsolateral prefrontal cortex projects
to the head of the caudate. Projections proceed
directly and indirectly (via
the globus pallidus external
portion and subthalamic
nucleus) to the globus
pallidus internal portion.
The internal globus pallidus
projects to the thalamus,
which accesses long-term
memory storage areas
Various symptoms in this patient also point to frontal–subcortical dysfunction. Diminished hygiene, diminished social judgment, weight gain, and stereotyped behaviors can reflect orbitofrontal and anterior cingulate
cortical damage, or else damage to projections leading to
and from these regions [5]. Therefore, from a signs and
symptoms perspective, it seems likely that the frontal
leukoaraiosis apparent on the patient’s MRI is relevant
to his clinical presentation.
of this family has begun as early as the fourth and as
late as the eighth decade. Insidious onset, gradual progression, and dysfunction of the frontal lobes or frontal–subcortical circuitry are consistent with syndromic
criteria for frontotemporal dementia [2]. In addition to
showing subcortical gliosis, autopsy histology reveals
findings typical of leukodystrophy syndromes.
NEUROIMAGING AND FRONTAL-BASED
COGNITIVE DYSFUNCTION
White matter lesions (also referred to as leukoaraiosis)
are frequently found on neuroimaging studies of elderly
subjects with and without cognitive impairment and appear as areas of hypertintense signal abnormalities on
T2 MRI or low density on computed tomography (CT).
Involved regions often include the periventricular white
matter, and periventricular involvement may excessively reside anterior to the frontal horns or posterior to the
occipital horns of the lateral ventricles. White matter
changes that extend outwards from or that are not contiguous to the lateral ventricles themselves can also appear as punctate or patchy lesions of the corona radiata
or centrum ovale. When present in the absence of an
obvious large vessel stroke, the etiology of white matter
lesions is often attributed to the spectrum of vascularrelated injury [6], largely because of their consistent association with age, hypertension, and other cardiovascular risk factors [7]. Although this etiologic attribution
may be accurate much of the time, there clearly are cases (such as the case discussed above) in which it is not.
CLINICAL FEATURES OF FRONTAL–
SUBCORTICAL DYSFUNCTION
In general, neuroanatomic localization of clinical deficits informs the neurologic differential diagnosis and
provides etiologic insight. The cognitive examination is
no exception. For adult patients with cognitive decline,
the clinician must decide whether signs and symptoms
are more referable to the medial temporal lobes, the
frontal lobes, or other brain regions. Medial temporal
dysfunction tends to appear as anterograde amnesia and
manifests as poor memory retention. Frontal dysfunction may also present with amnesia, but frontal-based
memory deficits typically show greater problems with
information retrieval than with retention.
Memory retrieval weakness that occurs in the presence
of more intact memory retention suggests that medial
temporal lobe-mediated information storage is still possible, and that the ability of the frontal lobe dorsolateral
prefrontal cortex (DLPC) to access stored information
is impaired. Because memory retrieval circuits from the
ENJ 2009; 1: (1). September 2009
Evidence indicates that the etiology of white matter lesions is heterogeneous and includes non-vascular
76
www.slm-neurology.com
Subcortical Gliosis and Leukodystrophy Overlap Syndromes as a Cause of Late-onset Dementia
a
b
c
d
Figure 3. N
europathological features. (a) Grossly, white matter degeneration in the frontal lobe is severe, with sparing of the superjacent cortex. (b, d) Microscopically, white matter pallor is evident at the gray matter–white matter junction, with sparing of the
U-fibers that are immediately subjacent to the cortex (hematoxylin & eosin (b) and Luxol fast-blue (d) stains; 20× total magnification; scale bar=1 mm). (c) High magnification shows myelin depletion with astrocytic gliosis and occasional axonal spheroids (hematoxylin & eosin; 400× total magnification; scale bar=20 μm)
causes [8]. Studies evaluating relationships between
the degree of leukoaraiosis and cognitive integrity have
been reported and suggest, in the absence of other clear
pathology, that cognitive integrity is not meaningfully
compromised unless leukoaraiosis is pervasive [9–11].
With the concomitant presence of Alzheimer’s diseaseassociated pathology, however, the threshold for leukoaraiosis-associated cognitive decline may be reduced [12].
Conversely, when leukoaraiosis is present, cognitive impairment [13] and dementia [14] resulting from Alzheimer’s disease may associate with reduced amounts of
Alzheimer’s disease histopathology.
of long-tract signs usually qualify for a histologic diagnosis of Alzheimer’s disease even when neuroimaging
reveals extensive white matter change that presumably
reflects pervasive microvascular ischemic disease [15].
Our subject’s signs and symptoms, however, correlated
with his leukoaraiosis. Neuroanatomic co-localization
of profound leukoaraiosis and clinical deficits suggests
the leukoaraiosis and clinical presentation were related.
However, the leukoaraiosis present on the MRI was not
due to microvascular ischemic disease, as was originally
suspected.
SUBCORTICAL GLIOSIS, LATE-ONSET
LEUKODYSTROPHIES, AND SUBCORTICAL
GLIOSIS–LEUKODYSTROPHY OVERLAP
When neuroimaging reveals leukoaraiosis in elderly
individuals with dementia, the cognitive examination
can provide insight into its clinical relevance. Elderly individuals with insidious and progressive memory retention failure, no clinical history of stroke, and a paucity
www.slm-neurology.com
Progressive subcortical gliosis was first described by
Neumann in 1949 as a variant form (Pick’s type 2) of
77
ENJ 2009; 1: (1). September 2009
European Neurological Journal
of the pathological changes leads one to identify the
white matter degeneration as the primary insult, with
the axonopathy, neuronopathy, gliosis, and macrophage
accumulation most likely occurring as secondary/reactive processes. The accumulation of ubiquitin, synuclein, tau, and APP in the axonopathic spheroids mirrors
changes that occur in as little as 6 h in acute head injury
[38–40]. As such, these proteins serve as markers for
neuronal reaction to axonopathic injury rather than as
links to the primary neurodegenerative disorders, such
as Alzheimer’s disease, with which they have been more
extensively associated.
Pick’s disease [16] and formally in 1967 by Neumann
and Conn [17] and others [18] as a distinct, albeit rare,
cause of dementia. It classically has an age of onset in the
fifth–seventh decade with a 4–7 year duration featuring
symptoms of disorientation, stereotypy, and dementia,
and a primarily frontotemporal pattern of cerebral atrophy. Its name refers to the presence of a pronounced
gliosis involving subcortical regions without severe involvement of the superjacent cortex or uniform myelin
loss. It has not generally been regarded as a well-defined
clinicopathologic entity, however, both because there is
no singular defining feature and because the associated
findings represent points on a spectrum associated with
other disorders featuring a leukodystrophy component.
The clinical and neuropathological features of our
adult-onset OLD syndrome are to mildly varying degrees shared with a number of other disorders that have
been described. A classification of OLD disorders into
pure, combined, and symptomatic forms was put forward in 1959 [41]. A similar disorder associated with
pigmented macrophages and other glia (pigmentary orthochromatic leukodystrophy, or POLD) was described
earlier, in 1936 [42]. Later, in 1984, a similar disorder
was described with attention drawn to the numerous axonal dilations, or spheroids (hereditary diffuse leucoencephalopathy with spheroids, or HDLS) [43]. Recently,
a similar disorder, proposed to be a distinct member of
this growing disease umbrella specified by its rapidity of
onset, was described [44].
Progressive subcortical gliosis, ill defined as it may
be, falls into the general category of adult-onset leukodystrophies. Leukodystrophies are, by definition, noninflammatory degenerative diseases primarily affecting
cerebral white matter. A number of the lysosomal storage disorders such as metachromatic leukodystrophy
[19], globoid cell leukodystrophy (Krabbe’s disease) [20],
Hurler syndrome and assorted mucopolysaccharidoses
[21–23], peroxisomal disorders (X-linked adrenoleukodystrophy), and other diseases (Alexander disease [24],
Canavan’s disease [25]) are leukodystrophies. Leukodystrophy can also be a prominent component of other
diseases, including Mendelian genetic [26–29], mitochondrial genetic [30–32], and infectious (cytomegalovirus) [33] diseases. Dementia can be a primary symptom
of many of these disorders, and frontotemporal-type
dementias have been associated with the axonopathy
that results from leukodystrophy [34–37]. Although
members of this group of disorders have variants that
can present in infancy, adolescence, and adulthood, it is
useful to divide them into (primarily) adult-onset and
childhood-onset disorders. They can be further divided
by the nature of their accumulated material into both
metachromatic and non-metachromatic (or orthochromatic) leukodystrophies (OLD).
Workers in this field are independently arriving at the
conclusion that each of these disorders may well represent points on a continuum of signs and symptoms
provoked either by a single genetic lesion with variable
penetrance and expressivity (depending on other genetic/environmental variables that may differ between
even closely related individuals) or, quite in opposition,
by multiple, different genetic lesions whose similarities
in expression are a result of their convergent effect on
a reductive, stereotypical response to disease in the central nervous system (CNS). Indeed, the CNS is particularly well known for the similarity in its responses to
diverse pathological insults, a fact that is neglected only
at great peril by the neuropathologist. Wider et al have
recently compared POLD and HDLS and, finding them
to overlap in almost every way, have suggested that they
be collectively referred to as ALSP (adult-onset leukoencephalopathy with axonal spheroids and pigmented
glia) [45]. Maillart et al have used the term LNS (adult
leukodystrophies with neuroaxonal spheroids) for these
conditions [44], although nomenclature was not the
primary focus of their report, and Moro-de-Casilla et al
have used the similar term LENAS (leucoencephalopathy with neuroaxonal spheroids) in their report [46].
We are in agreement that it would serve most interests
well to establish a unifying diagnostic term. Especially
as pigmented glia are not universally reported in these
disorders, whereas axonal spheroids are, we would favor
a term such as ALENAS (adult-onset leukoencephalopathies with neuroaxonal spheroids), essentially that of
The subcortical gliosis syndrome affecting the kindred
on which we recently reported [1] demonstrated neuropathological changes typical of OLD-type leukodystrophies. Grossly, there were multiple foci, primarily in the
frontal lobes, of white matter destruction, but there was
no frank destruction of the overlying cortex (Figure
3a). Microscopically, there were focal areas of secondary myelin loss (Figure 3b,d) with associated axonal
destruction and dilatation (“spheroids”). An inflammatory infiltrate per se was absent, but lipid-laden macrophages were prominent. The lesions were invested by a
robust astrocytic gliosis (Figure 3c). Axonal spheroids
were immunoreactive for alpha-synuclein, tau, ubiquitin, and amyloid precursor (APP) proteins. Juxtacortical U-fibers were spared. The cortex was not entirely
normal, as increased gliosis and scattered degenerating
neurons (“balloon neurons”) were in evidence.
As with other OLD syndromes, the kind and extent
ENJ 2009; 1: (1). September 2009
78
www.slm-neurology.com
Subcortical Gliosis and Leukodystrophy Overlap Syndromes as a Cause of Late-onset Dementia
REFERENCES
Moro-de-Casilla et al, as it offers the most encompassing
definition and highlights the adult-onset characteristic
of these disorders.
1. Swerdlow RH, Miller BB, Lopes MB, et al. Autosomal dominant
subcortical gliosis presenting as frontotemporal dementia. Neurology. 2009;72(3):260–267.
2. McKhann GM, Albert MS, Grossman M, Miller B, Dickson D,
Trojanowski JQ. Clinical and pathological diagnosis of frontotemporal dementia: report of the Work Group on Frontotemporal Dementia and Pick’s Disease. Arch Neurol. 2001;58(11):1803–1809.
3. Bonelli RM, Cummings JL. Frontal-subcortical dementias. Neurologist. 2008;14(2):100–107.
4. Martin A, Wiggs CL, Lalonde F, Mack C. Word retrieval to letter
and semantic cues: a double dissociation in normal subjects using
interference tasks. Neuropsychologia. 1994;32(12):1487–1494.
5. Mega MS, Cummings JL. Frontal-subcortical circuits and
neuropsychiatric disorders. J Neuropsychiatry Clin Neurosci.
1994;6(4):358–370.
6. Bernick C, Kuller L, Dulberg C, et al. Silent MRI infarcts and the
risk of future stroke: The cardiovascular health study. Neurology.
2001;57(7):1222–1229.
7. Longstreth WT, Jr., Arnold AM, Manolio TA, et al. Clinical
correlates of ventricular and sulcal size on cranial magnetic
resonance imaging of 3,301 elderly people. The Cardiovascular
Health Study. Collaborative Research Group. Neuroepidemiology.
2000;19(1):30–42.
8. Erkinjuntti T, Benavente O, Eliasziw M, et al. Diffuse vacuolization (spongiosis) and arteriolosclerosis in the frontal
white matter occurs in vascular dementia. Arch Neurol.
1996;53(4):325–332.
9. Miranda B, Madureira S, Verdelho A, et al. Self-perceived
memory impairment and cognitive performance in an elderly
independent population with age-related white matter changes. J
Neurol Neurosurg Psychiatry. 2008;79(8):869–873.
10. Schmidt R, Petrovic K, Ropele S, Enzinger C, Fazekas F. Progression of leukoaraiosis and cognition. Stroke. 2007;38(9):2619–
2625.
11. Longstreth WT Jr, Manolio TA, Arnold A, et al. Clinical correlates
of white matter findings on cranial magnetic resonance imaging
of 3301 elderly people. The Cardiovascular Health Study. Stroke.
1996;27(8):1274–1282.
12. Burns JM, Church JA, Johnson DK, et al. White matter lesions
are prevalent but differentially related with cognition in aging
and early Alzheimer disease. Arch Neurol. 2005;62(12):1870–
1876.
13. Esiri MM, Nagy Z, Smith MZ, Barnetson L, Smith AD, Joachim
C. Cerebrovascular disease and threshold for dementia in the
early stages of Alzheimer’s disease. Lancet. 1999;354(9182):919–
920.
14. Snowdon DA, Greiner LH, Mortimer JA, Riley KP, Greiner PA,
Markesbery WR. Brain infarction and the clinical expression of
Alzheimer disease. JAMA. 1997;277:813–817.
15. Nolan KA, Lino MM, Seligmann AW, Blass JP. Absence of vascular dementia in an autopsy series from a dementia clinic. J Am
Geriatr Soc. 1998;46(5):597–604.
16. Neumann MA. Pick’s disease. J Neuropathol Exp Neurol.
1949;8(3):255–282.
17. Neumann MA, Cohn R. Progressive subcortical gliosis, a rare
form of presenile dementia. Brain. 1967;90(2):405–418.
18. Verity MA, Wechsler AF. Progressive subcortical gliosis of Neumann: a clinicopathologic study of two cases with review. Arch
Gerontol Geriatr. 1987;6(3):245–261.
19. Gieselmann V. Metachromatic leukodystrophy: genetics,
pathogenesis and therapeutic options. Acta Paediatr Suppl.
2008;97(457):15–21.
20. Suzuki K. Globoid cell leukodystrophy (Krabbe’s disease): update. J Child Neurol. 2003;18(9):595–603.
21. Huntsman RJ, Seshia S, Lowry N, Lemire EG, Harder SL. Peripheral neuropathy in a child with Cree leukodystrophy. J Child
Neurol. 2007;22(6):766–768.
22. Bonnet-Dupeyron MN, Combes P, Boespflug-Tanguy O, Vaurs-
Whatever term might be agreeable to all, it is not a
trivial matter that one be identified. From our review
and those of others, it is almost certain that these disorders are more common than is currently estimated.
They are frequently misdiagnosed as vascular dementia, or as disorders in the vascular dementia spectrum
such as Binswanger’s disease. Additionally, ALENAS
spectrum disorders have been initially diagnosed as
early-onset rapidly progressive dementia [47], earlyonset frontal [48] and frontotemporal [49] dementias,
multiple-system atrophy [46], multiple sclerosis [50],
and Alzheimer’s disease [51]. As hospital autopsy rates
are in steep decline and without a compelling clinical
question to be answered, premortem diagnoses are rarely confirmed, and it is certain that the true incidence
and spectrum of expression of ALENAS disorders will
remain underappreciated. This deprives affected families and their caregivers of necessary (and otherwise
available) information and deprives investigators of the
substrates from which the genetic origin(s) of these disorders may be identified.
CONCLUSIONS
For middle-aged and elderly patients with symptoms
of progressive cognitive or behavioral change, frontal–
subcortical signs that exceed medial temporal-localizing
signs and extensive leukoaraiosis, it is important to
consider subcortical gliosis and leukodystrophy disorders in the differential diagnosis. Although traditionally
considered causes of childhood dementia, leukodystrophies are increasingly recognized to cause adult- and
even late-onset dementia. Relationships between subcortical gliosis and histologic white matter perturbations are apparent, and new classification schemes that
emphasize these relationships are emerging. A family
history of frontal–subcortical or frontotemporal dementia syndromes with accompanying leukoaraiosis should
raise suspicion of subcortical gliosis, leukodystrophy,
and subcortical gliosis–leukodystrophy overlap diseases.
Gene linkage studies of families with such histories will
ideally provide insight into mechanistic causes. In the
future, heightened clinical awareness will also hopefully promote clinicopathologic studies that define the
relevance of subcortical gliosis–leukodystrophy overlap
disorders to sporadic dementia cases. Finally, subcortical gliosis–leukodystrophy overlap disorders should be a
diagnostic consideration for patients traditionally felt to
have subcortical ischemic vascular dementia.
Disclosures: The author has no financial interests to disclose
related to the contents of this article.
www.slm-neurology.com
79
ENJ 2009; 1: (1). September 2009
European Neurological Journal
Barriere C. Absence of OLIG2 mutations in patients presenting with a severe Pelizaeus–Merzbacher-like leukodystrophy
associated with motor neuron dysfunction. Am J Med Genet B
Neuropsychiatr Genet. 2008;147B(4):538–539.
23. Shukla P, Balakrishnan P, Agarwal N, et al. Prenatal diagnosis of
megalencephalic leukodystrophy. Prenat Diagn. 2008;28(4):357–
359.
24. Johnson AB. Alexander disease: a leukodystrophy caused by a
mutation in GFAP. Neurochem Res. 2004;29(5):961–964.
25. Baslow MH. Canavan’s spongiform leukodystrophy: a clinical
anatomy of a genetic metabolic CNS disease. J Mol Neurosci.
2000;15(2):61–69.
26. Padiath QS, Saigoh K, Schiffmann R, et al. Lamin B1 duplications cause autosomal dominant leukodystrophy. Nature Genet.
2006;38(10):1114–1123.
27. de Kremer RD, Latini A, Suormala T, et al. Leukodystrophy and
CSF purine abnormalities associated with isolated 3-methylcrotonyl-CoA carboxylase deficiency. Metab Brain Dis. 2002;17(1):13–
18.
28. Barone R, Parano E, Trifiletti RR, Fiumara A, Pavone P. White
matter changes mimicking a leukodystrophy in a patient with
mucopolysaccharidosis: characterization by MRI. J Neurol Sci.
2002;195(2):171–175.
29. Gorospe JR, Singhal BS, Kainu T, et al. Indian Agarwal megalencephalic leukodystrophy with cysts is caused by a common MLC1
mutation. Neurology. 2004;62(6):878–882.
30. Hung PC, Wang HS. A previously undescribed leukodystrophy in
Leigh syndrome associated with T9176C mutation of the mitochondrial ATPase 6 gene. Dev Med Child Neurol. 2007;49(1):65–
67.
31. Magen D, Georgopoulos C, Bross P, et al. Mitochondrial hsp60
chaperonopathy causes an autosomal-recessive neurodegenerative disorder linked to brain hypomyelination and leukodystrophy. Am J Hum Genet. 2008;83(1):30–42.
32. Rahman S, Brown RM, Chong WK, Wilson CJ, Brown GK. A
SURF1 gene mutation presenting as isolated leukodystrophy.
Ann Neurol. 2001;49(6):797–800.
33. Tatli B, Ozmen M, Aydinli N, Caliskan M. Not a new leukodystrophy but congenital cytomegalovirus infection. J Child Neurol.
2005;20(6):525–527.
34. Tagawa A, Ono S, Inoue K, et al. A new familial adult-onset
leukodystrophy manifesting as cerebellar ataxia and dementia. J
Neurol Sci. 2001;183(1):47–55.
35. Johannsen P, Ehlers L, Hansen HJ. Dementia with impaired temporal glucose metabolism in late-onset metachromatic leukodystrophy. Dement Geriatr Cogn Disord. 2001;12(2):85–88.
36. Mascalchi M, Gavazzi C, Morbin M, et al. CT and MR imaging
of neuroaxonal leukodystrophy presenting as early-onset frontal
dementia. AJNR Am J Neuroradiol. 2006;27(5):1037–1039.
37. Kozian R, Sieber N, Thiergart S. Frontotemporal dementia
in metachromatic leukodystrophy. Fortschr Neurol Psychiatr.
2007;75(9):549–551.
38. Uryu K, Giasson BI, Longhi L, et al. Age-dependent synuclein
pathology following traumatic brain injury in mice. Exp Neurol.
2003;184(1):214–224.
39. Uryu K, Chen XH, Martinez D, et al. Multiple proteins implicated
in neurodegenerative diseases accumulate in axons after brain
trauma in humans. Exp Neurol. 2007;208(2):185–192.
40. McKee AC, Cantu RC, Nowinski CJ, et al. Chronic traumatic encephalopathy in athletes: progressive tauopathy after repetitive
head injury. J Neuropathol Exp Neurol. 2009;68(7):709–735.
41. Peiffer J. On non-metachromatic leukodystrophy. Arch Psychiatr
Nervenkr Z Gesamte Neurol Psychiatr. 1959;199:417–436.
42. Van Bogaert L, Yssen R. Le type tardif de la leukdystrophie progressive familiale. Rev Neurol (Paris). 1936;65:21–45.
43. Axelsson R, Roytta M, Sourander P, Akesson HO, Andersen O.
Hereditary diffuse leucoencephalopathy with spheroids. Acta
Psychiatr Scand Suppl. 1984;314:1–65.
44. Maillart E, Rousseau A, Galanaud D, et al. Rapid onset frontal
leukodystrophy with decreased diffusion coefficient and neuroax-
ENJ 2009; 1: (1). September 2009
onal spheroids. J Neurol. 2009.
45. Wider C, Van Gerpen JA, DeArmond S, Shuster EA, Dickson DW,
Wszolek ZK. Leukoencephalopathy with spheroids (HDLS) and
pigmentary leukodystrophy (POLD): a single entity? Neurology.
2009;72(22):1953–1959.
46. Moro-de-Casillas ML, Cohen ML, Riley DE. Leucoencephalopathy
with neuroaxonal spheroids (LENAS) presenting as the cerebellar subtype of multiple system atrophy. J Neurol Neurosurg
Psychiatry. 2004;75(7):1070–1072.
47. Goodman LA, Dickson DW. Nonhereditary diffuse leukoencephalopathy with spheroids presenting as early-onset, rapidly-progressive dementia. J Neuropathol Exp Neurol. 1995;54(3):471.
48. Mascalchi M, Gavazzi C, Morbin M, et al. CT and MR imaging
of neuroaxonal leukodystrophy presenting as early-onset frontal
dementia. AJNR Am J Neuroradiol. 2006;27(5):1037–1039.
49. Levin N, Soffer D, Biran I, et al. Leukoencephalopathy with neuroaxonal spheroids presenting as frontotemporal dementia. Isr
Med Assoc J. 2008;10(5):386–387.
50. Keegan BM, Giannini C, Parisi JE, Lucchinetti CF, Boeve BF,
Josephs KA. Sporadic adult-onset leukoencephalopathy with
neuroaxonal spheroids mimicking cerebral MS. Neurology.
2008;70(13 Pt 2):1128–1133.
51. Terada S, Ishizu H, Yokota O, et al. An autopsy case of hereditary
diffuse leukoencephalopathy with spheroids, clinically suspected
of Alzheimer’s disease. Acta Neuropathol (Berl). 2004;108(6):538–
545.
80
www.slm-neurology.com
www.slm-neurology.com