Perifere circulatie - Erasmus Critical Care

Perifere
circulatie
Jan Bakker
jan.bakker@erasmusmc.nl
Vasoconstriction during massive hemorrhage
Tools to assess the
peripheral circulation
The outside - Skin
Color - Temperature
Capillary Refill Time
Pulse Oximetry - Near Infrared Spectroscopy
Clinical assessment
capillary refill time
normal ≤ 2 sec in children
and young adults 4.5 sec in
older patients
Champion et al. Crit Care Med 1981;9:672-676 - Schriger et al. Ann Emerg Med 1988;17:932-935
Physiology of the skin
Circulatory bed
Renal
Cerebral
Skeletal muscle
Intestinal
Cutaneous
Sympathetic
control
Metabolic
control
Autoregulation
++
+
++
+
+++
+++
+++
+++
++
+++
+++
++
+
++
+
bases the diagnosis of shock.
a severe systemic illness, a critical curThe present study was unde
tailment of effective blood flow and a conlish whether quantitative c
sequent curtailment of the function of vital
temperature might provide a
organs is clinically recognized as circulatory
an Indication
Great blood
Toe aspressure,
Temperature
indication of the presence and
Reductionof inthearterial
shock.
Indication
Great
Toe as an
the lessened
Temperature
of theof
Severity
of Shock
shock. Such measure
mental
culatory
in urinary
output,
decline
inShock
blood
objective and yet available wi
and R.inadequacy
of HARRY
alertness,By HENRI
JOLY,
WEIL,flow
M.D.,
ofM.D.,the MAXSeverity
ofPH.D.
discomfort to the patient. The
the extremities with reduced or absent peSUMMARY
only a minor investment in
of
superficial
ripheral
pulsation,
HENRIchanges
R. constriction
JOLY,
M.D., AND
Bywhether
provideHARRY
objectiveWEIL,
indicationM.D., PH.D.
To establish
in skin temperature
wouldMAX
temperature measurementsand technical support, and the
of the and and
shock, hemodynamic
severity of
in Values
andafter admistheandobtained
hands
loss
ofof warmth
vessels,
3 hours
obtained with
the aid
digital computer.
sion and 3 hours prior to discharge death in 100 patients who presented with clinicaltential value for routine moni
SUMMARY
measured with standard
analyzed. Temperature
signs of circulatory shock
at
five
sites:
thepecially
finger,
large toe,
thermistor
the
digital
pad
of
the
third
probes
attractive.
provide
objective
indication
To establish whetherthechanges
in skin temperaturethewould
lateral portion of the thigh, and the rectum.
deltoid region of the
andtemperameasurementsthat an inde
temperature
of Athe
of shock,
hemodynamic
and
severity
presence
and Department
Shock
Unit
the
Research
From
and
between
the cardiac output of
correlation
demonstrated
significant
The
possibility
corrections
ture ofobtained
the toe (r with
increased
to 0.73 when
0.71).the
Correlations
of a digital
obtained 3 hours after admiscomputer.
Values
were
aidSouthern
of
California
School
Medicine,
University
for
in
analysis
provided
A
stepwise
regression
changes
ambient
temperature.
made
measurements
presented
with clinical of skin t
or death in 100 patients whotiple
3 hours prior
to predictability
discharge
sion
and improvement
in
for
index
when
the
values
of
cardiac
the
significant
Med- showed
and the sites
Los Angeles
Medicine
of signs
County-USC
measured
with standard would ser
of circulatory
analyzed.
Temperature
were
of other skin shock
included.
Discriminant
function analysiswas
temperatures
rectal
temperature
California.
Los
icalthermistor
Angeles,
Center,
of toe
measurement
outcome
correctlypad
predicts
patient
that
temperature
early probes
at five
sites:
finger, the large toe, the
the digital
of the
third
tenttheindication
of
67%
the
time.
of peripheral
andthigh, and
of the arm,bytheGrants
rectum.
Work
lateral HE-05570
portion of the
regionsupported
deltoid was
outputwould
and temperaA significant
between
correlation
was demonstrated
hence
quantitate a ci
Heart Institute,
U. the
S. cardiac
from
National
HE-07811
theWords:
Additional
Indexing
of
ture
the toeService,
were
to 0.73 when corrections were
(r 0.71).
Correlations
The
A. increased
Hartford
Health
and
Public
by
John
Skin temperatures
Rectal temperatures
Hemodynami
c data
characteristic of shock, was s
for changes
in ambient
temperature. A stepwise regression analysis provided no
made
Serum lactate
Burton's
index
Foundation,
Inc., New York.
in the Computing
forBurton.' Pre
when the work
values of
of cardiac indexprevious
predictability assistance
significant improvement
Sciences
obtained from
theskinHealth
wastemperatures
Computing
of other
sites were
showed
included.
Discriminant function
analysis
in
ha
our
unit
on
who
patients
which
the
AFacility,
on
clinician
the
course
of
feet
are
signs
FTER
TRAUMA,
or
during
physical
at
Los
California
of
Angeles,
University
of
toe temperature correctly predicts patient outcome
that an early measurement
bases
shock.
a
theHealth
diagnosis ofGrant
critical
cursevere
a
illness,
systemic
sequence of pancreatitis ind
of thebytime.
67% effective
Institutes
National
of
sponsored
to estabstudy
was
undertaken
The
present
a
conblood flow and
tailment of
creases
FR-3.
lish whether quantitative changes in skin in the skin temperatur
curtailment of the function of vital
sequent
AND
presence
were
a
or
was
were
arm,
was
were
were
no
were
an
temperature might provide a more competent
is clinically recognized as circulatory
organs Additional
Indexing
Words:
XXXIX,
January
1969
Circulation,
Volume
severity of cirindication of the presence and131
shock. Reduction in arterial blood pressure,
Skin temperatures
Hemodynami c data
Rectal temperatures
w w ._
bases the diagnosis of shock.
a severe systemic illness, a critical curThe present study was unde
tailment of effective blood flow and a conlish whether quantitative c
sequent
3h aftercurtailment
admission of the function of vital JOLY,
WEILmight provide a
temperature
organs is clinically recognized as circulatory
indication of the presence and
shock. Reduction in arterial blood pressure,
culatory shock. Such measure
decline in urinary output, lessened mental
objective and yet available wi
and inadequacy
of blood flow in
alertness, CARDIAC
INDEX
discomfort to the patient. The
reduced or absent pethe extremities with L/min/m2
only a minor investment in
ripheral pulsation, constriction
r= 0.71 of superficial
and technical support, and the
44and loss of warmth in the hands and
vessels,
tential value for routine moni
0
pecially attractive.
From the Shock Research Unit and Department of
The possibility that an inde
*
*
*0
California School
Medicine, University of 0 Southern
0
measurements of skin t
tiple
and the
Los Angeles
of Medicine
2*.
..County-USC Medrectal temperature would ser
ical Center, Los Angeles, California.
*
tent indication of peripheral
HE-05570 and
Work was supported by Grants
hence
would quantitate a ci
U. -5.24
S.
National Heart Institute,C.I=
HE-07811- from the.00
+ Ttoe
(0,286)
Public Health Service, and by The John A. Hartford
characteristic of shock, was s
Foundation, Inc., New York. Computing assistance
previous work of Burton.' Pre
I
from the Health Sciences Computing
was obtained
0 o
in our unit36on patients who ha
at Los Angeles,32
Facility, University24of California 28
of pancreatitis ind
Institutes of Health Grant
sponsored by National TEMPERATURE,
TOE OC sequence
creases "I-07in the skin temperatur
00 FR-3.
0
SRU1 6168 HIJ
EQUIENIt.A I'S
6541-
Figure 5
Circulation, Volume XXXIX, January 1969
131
,-
bases
shock.
a critical curthe
diagnosis
of
severe
a
illness,
systemic
the result
the flow
"early"
Thestudy was undeS
The present
conand a measurement.
tailment of effectiveofblood
lish whether quantitative c
sequent curtailment of the function of vital
temperature might provide a
organs is clinically recognized as circulatory
indication of the presence and
shock. Reduction in arterial blood pressure,
culatory shock. Such measure
lessened mental
decline in urinary output,
Differe
- p<O.00I
HOURS11.-1"
!!1.,13
wi
objective and yet available
alertness, and inadequacy of blood flow in
temper
discomfort to the patient. The
the extremities with reduced or absent peonly a minor investment in
of superficial
ripheral20pulsation,
lg constriction
p<O.TOE TEMP.
and technical support, predic
and the
28.2;.3
4OC
26.O;.2
in the- hands
and
of warmth
vessels, and loss
tential value for routine Whe
moni
pecially attractive. the
d
From the Shock Research Unit and Department of
The
an
that
inde
possibility
4 University of Southern California School
Medicine,
skin tt
measurements ofthe
tiple
of Medicine and the Los Angeles County-USC Medrectal temperature would
(fig.ser6
ical Center,
2 Los Angeles, California.
tent indication of peripheral
Work was supported by Grants HE-05570 and
dicted
hence
a ci
quantitate
would
Heart
U.
S.
National
HE-07811 1Lfrom the_I
Institute,
TOE
TEMP.
0
The John A.25.1
Hartford
and by° -4'C
Public Health
Service,
characteristic of shock,toe
waste
s
29.8+-.3
.2
00
Foundation, Inc., New York. Computing assistance
Pre
previous work of Burton.'
tures
was obtained from the Health Sciences Computing
in our unit on patients who ha
Facility, University of California at Los Angeles,
out ind
of
sequence of pancreatitis
sponsored by National Institutes of Health Grant
creases in the skin temperatur
this is
FR-3.
Figure 6
Circulation, Volume XXXIX, January 1969
ment
131
TOE-AMBIENT TEMPERATURE ° C
IADMISSION
SURVIVED
(56)
DIED
(44)
Peripheral perfusion in
cardiogenic shock
Odd’s ratio for day-30 mortality
‣Oliguria
‣Altered sensorium
‣Cold, clammy skin
Gusto I trial: 41.021 patients Am Heart J 1999;138:21-31
3.4 (2.8 - 4.2)
2.1 (1.7 - 2.5)
1.8 (1.5 - 2.3)
Skin temperature
• Skin temperature, Skin blood flow and Core/Ambient Temperature are related
• Better to use gradients
§ Central to peripheral
• Normal hemodynamics 3-7oC
• Limited by hypothermia, cold ambient temperature (<20oC), vasodilatory shock
ü Increased more than 2oC associated with hemodynamic improvement
ü Cardiogenic shock associated with a decrease in skin temperature, increase in skin temperature associated
with improvement
§ Peripheralprox to Peripheraldist
• Gradient related to fingertip bloodflow
• Gradient related to thresholds for vasodilation and vasoconstriction
• Few studies in critically ill patients
Skin temperature and
systemic circulation
• Cool vs. warm skin
§ Similar: Heart rate, blood pressure, PAOP, Hemoglobin, FiO2, PaO2, PaCO2
Cool
Cardiac Index 2.9 ± 1.2
Arterial pH 7.32 ± 0.2
SvO2 60 ± 4
Lactate 4.7 ± 1.5
Kaplan et al. J Trauma 2001;50:620-628
Warm
4.3 ± 1.2 *
7.39 ± 0.07 *
68 ± 8 *
2.2 ± 1.6 *
Skin color
H6, the arterial lactate levels, urinary output, and mottling
score were predictive of 14-day mortality. Lactate level
was associated with an increased risk of death
[\1.5 mmol/l OR 1, between 1.5 and 3 OR 3.8 (0.7–29.5),
[3 OR 9.6 (2.1–70.6), p = 0.01] as well as urinary output
\0.5 ml/kg/h [OR 10.8 (2.9–52.8), p = 0.001]. Mottling
score was the strongest predictor of mortality [score 0–1
OR 1; score 2–3 OR 16 (4–81); score 4–5 OR 74
(11–1,568), p \ 0.0001] (see Table 2). Fourteen-day
mortality according to the H6 mottling score increased
from 13% for a score of 0–1 to 70% for a score of 2–3 and
92% for a score of 4–5 (v2 test for trend p \ 0.001). We
analyzed time of death according to the mottling score.
We found that death occurred earlier in patients with a
higher score (p \ 0.0001). Patients with a mottling score
of 4–5 died mostly on day 1, whereas patients with a score
of 2–3 died on day 2 and day 3 after admission (see
Fig. 2).
Mottling score predicts survival in
septic shock
Ait-Oufella et al. Intensive Care Med 2011;37:801-807
moderate or ex
patients whose
Conversely, am
increased, only
survivors incre
ment [OR = 2
Table 3]. More
sive mottling a
mottling score b
increase in the
tion coefficient
An increase
increasing lact
urinary output
the cardiac in
SOFA score als
level of mottlin
Studied population
Mottling score
5
Table 2 Analysis of hemodynamic parameters at H6 reflecting Discussion
macro- and microcirculation and identification of three significant
risk factors of 14-day death in univariate analysis, urinary output,
Alterations in m
arterial lactate level and mottling score
4
3
2
1
SCORE 2
SCORE 4
Fig. 1 Left: the mottling score is based on a mottling area
extension on the legs. Score 0 indicates no mottling; score 1, a
Between February and August 2009, 60
development o
patients
wereUnivariate
included.
Baseline
characteris
Factor
analysis
Analysis mortality in pat
OR (95% CI)
p value
marized in Table
1. All of the
patients
[15]. To had
explo
focused
on
clin
MAP (mmHg)
and
h of and
adm
non
[65 required 1vasopressors within 6 assess
\65
1.9 (0.4, 10.5)
0.43
have
direct
ef
related
to pneumonia (45%) andillustrated
abdomi
Heart rate (beats/min)
by
[120
1.6 (0.4, 6.1)
shock patients.
(32%).
The mean
SOFA score computed
w
(90–120)
1
toe as a param
\90
1.1 (0.3, 4.0)
0.80
values
during
the first 6 h after ICU
admiss
Central venous
pressure (mmHg)
reported
a corr
[12
1 (0.3, 3.3)
toe temperatur
(8.5–14.5)
and
the SAPS II was
(8–12)
1
0.21 59 (45–71)
mixed ICU pop
\8
0.3 (0.1, 1.3)
capillary
refill
Cardiac
index
(l/min/m2)
patients
were
treated with norepinephrine
prognostic fact
[3
1
\3
(0.5, 4.1)(0.2–1)] during
0.53
study,
focus
(83%),
dose 1.40.60
the wefirst
Urinary output (ml/kg/h)
sign
of
shock
[0.5
1 and others with epinephrine
management
developed a cli
\0.5
10.8 (2.9, 52.8)
0.001
of mottling fro
Arterial
lactate
(mmol/l)
(17%),
dose 10.4 (0.4–2)]. The day-14
mort
\1.5
score is very ea
(1.5–3)
(0.7, 29.5)
it within
few m
45%
[95% CI3.8
(33–58)].
Distribution
of the
m
[3
26 (2.1,
70.6)
0.01
independent int
Mottling score
is0–1detailed in1 Table 1. To assess the
on theprogn
mottling
2–3
16 (4, 81)
(0.72, 0.97)].
mottling,
the74scores
4–5
(11, 1,568) were pooled
\0.0001as follows:
We perform
patients
withm
s
The cardiac index was
measured
transthoracic mild or
abnormalities
(0–1;
31 using
patients),
very limited to
echocardiography
MAP patients)
mean arterial pressure
severe as illust
17
and severe (4–5; 12 patients).
The Perfusion Index is calculated with a patented method using both, red and
infrared pulsations:
PI =k1*([ACred / DCred]-1) + k2*([ACinfrared / DCinfrared]-1)
I
LED
I
k1 and k2 wavelength-dependent factors
AC: the alternating component and DC: the constant component of the signal
PFI,
eraeraamic
artewas
tery
emce of
Philemwith
r by
emdif(9).
ded.
as a
-toe
a receiver
characteristic curve. A
healthy volunteers were included,
and aoperating
hypertension).
volunteers
h
PFI of 1.4 discriminated
best All
between
a valu
total of 216 measurements were
made. capillarycore-to-toe
add
refill timetemperand arterial
normal and abnormal
ade
The distribution of age in theature
healthy
difference
in
these
critically
ill
pauration
(96%
to
100%).
Use of a peripheral perfusion index derived from the pulse
sign
volunteers was normal: skewness,
tients0.06;
(area underGroup
the curve,
0.91;
95%
2. A total of 74 can
me
oximetry
signal
as
a
noninvasive
indicator
of
perfusion
confidence
– 0.98). Table 3 remedian, 36 yrs (inner quartile
range, interval,
were0.84
carried
out in the 37 pa
inte
ports
the
corresponding
sensitivity,
speciAlexandre Pinto Lima, MD; Peter Beelen, RN; Jan Bakker, MD, PhD
ied. Descriptive
statistics reveD
ficity, and likelihood
ratios.
PFI of 2.2 # 0.22 with a with
me
Objective: Peripheral perfusion in critically ill patients fre- ence >7°C. Peripheral perfusion index and arterial oxygen
put,
(inner
quently is assessed by use of clinical signs. Recently, the pulse saturation were measured by using
the Philipsquartile
Medical Systems range, 0.5–3.
0.3 - 10to reflect changes in periph- Viridia/56S monitor. In group 1, measurements were made before
oximetry signal has been suggested
by
eral perfusion. A peripheralmedian:
perfusion 1.4
index(IQR:
based on
analysis
of
and
after
a
meal.
In
group
2,
two
measurements
were
made,
with
summarizes hemodynamic
d
0.7 - 3.0)
the pulse oximetry signal has been implemented in monitoring the second measurement taken when the peripheral perfusion
dec
systems as an index of peripheral perfusion. No data on the profile had changed. A total of abnormal
216 measurements were
carried
peripheral
perfusio
The
variation of this index in the normal population are available, and out in group 1. The distribution of the peripheral perfusion index
clinical application of this variable in critically ill patients has not was skewed and values ranged from
0.3
to
10.0,
median
1.4
(inner
mal peripheral perfusion, perf
as
been reported. We therefore studied the variation of the peripheral quartile range, 0.7–3.0). Seventy-four measurements were carried
perfusion index in healthy adults and related it to the central-to- out in group 2. A significant correlation between the peripheral
mean doses of vasoactive equ
dru
toe temperature difference and capillary refill time in critically ill perfusion index and the core-to-toe temperature difference was
patients after changes in clinical signs of peripheral perfusion.
found (R ! .52; p < .001). A cutoff peripheral perfusion index
of p
nificant
relationship
between
value of 1.4 (calculated by constructing
a receiver
operating
Design: Prospective study.
characteristic curve) best reflected the presence of poor periphSetting: University-affiliated teaching hospital.
pale
PFI or core-to-t
patients. Changes and
in peripheral
Patients: One hundred eight healthy adult volunteers and 37 eral perfusion in critically ill perature
tech
perfusion index and changes in core-to-toe temperature differadult critically ill patients.
difference
was found.mea
A
! .52, p <
.001).
ence correlated significantly (Rture
Interventions: None.
Conclusions: The peripheral perfusion index distribution in the
Measurements and Main Results: Capillary refill time, periphChanges in the peripheral
eral perfusion index, and arterial oxygen saturation were mea- normal population is highly skewed.
exponential
relationship betw
Dop
sured in healthy adults (group 1). Capillary refill time, peripheral perfusion index reflect changes in the core-to-toe temperature
perfusion
index measurements
perfusion index, arterial oxygen saturation, central-to-toe tem- difference. Therefore, peripheral
the
core-to-toe
temperature
mic
used
to
monitor
peripheral
perfusion
in
critically
ill
paperature difference, and hemodynamic variables were measured can be
Figure 2. Relationship between peripheral
perfu2
–1213)
in critically ill patients (group 2) during different peripheral per- tients. (Crit Care Med 2002; 30:1210
was
found
(R
$ .52,difp " cha
.00
(PFI) and
core-to-toe
temperature
WORDSindex
: skin temperature;
peripheral
perfusion; shock;
heKEYsion
fusion profiles. Poor peripheral perfusion was defined as a capillary refill time >2 secs and central-to-toe temperature differ- modynamics; monitoring; central temperature
We found
a significant
lin
ference in all 74 measurements
in the
37 patients not
depa
studied. Displayed
is the
best fit curve
(logarithtion
between
changes
in
PFI
arly recognition of impaired clinical signs (1), from the central-to-toe
therefore, was to assess the
W
Crit Care Med 2002;30:1210-1213
mic) R2 " current
.52, pstudy,
#
.001.
Reference
lines
are the
organ perfusion is important temperature difference (2, 3, 5), or with variationin
of the
this perfusion
index
in
core-to-toe
temperatur
Figure
1. tissue
Frequency
ofDoppler
all
216
pe-healthy
to avoid
hypoxia that distribution
adults and volunteers
study the relationship
techniques such as laser
and capmedian
PFI
of healthy
and the refer- crea
2
2
Near InfraRed Spectroscopy
Detection of NIR-light
ρ
NIR-light
4 wavelengths
Emission probe
Detection probe
(multi-channel)
∂A
∂ρ
PFI - StO2
lary block
tic block
Axillary blocks
StO2 %
6 patients
Galvin et al. Anest Analg 2006;103:239-243
Peripheral vasoconstriction influences thenar
oxygen saturation as measured by NIRS
Lima et al. Intensive Care Med 2012 (in press)
•
•
•
•
•
8 healthy volunteers (26±6 yr)
use of cooling blanket
•
cooling of the periphery without change in core temperature
(tympanic temperature)
•
•
water temperature 32oC
measurements: baseline, 30’ after start cooling, 30’ after
suspension of cooling (rewarmed)
StO2 measured with 15 mm probe on thenar muscle
CO measured by NICOM (Cheetah)
Skin vasoconstriction: 50% decrease in Tskin-diff or PI
Peripheral vasoconstriction influences thenar
oxygen saturation as measured by NIRS
Lima et al. Intensive Care Med 2012;38(4):606-611
abnormal peripheral circualtion. This was associated
orse outcome in critically ill patients
with an abnormal StO2. Interestingly, although in all
Jansen,Research
Can Ince and Jan Bakker
used,
StO2
did not
Low tissuepatients
oxygen vasodilators
saturation atwere
the end
ofthe
early
goal-directed
normal
during the initial 8h of
ty Medical Centre Rotterdam, PO
Box 2040,to
3000
CA Rotterdam,
therapy is return
associated
with values
worse outcome
in critically ill patients
g of
rapy
cted
ce of
d to
with
copy
rate
NIRS
ssion
ated
rgan
ealth
bles:
ygen
sical
StO2
were
evels
2) at
edian
HE II,
cant
ables
here
lities
resuscitation.
Alexandre Lima, Jasper van Bommel, Tim C Jansen, Can Ince and Jan Bakker
The mortality of these patients was very high (50%)
whereas the mortality in the other patients was
absent!
Department of Intensive Care, Room HS3.20, Erasmus MC University Medical Centre Rotterdam, PO Box 2040, 3000 CA Rotterdam,
The
Netherlands
Critical
Care 2009, 13(Suppl 5):S13 (doi:10.1186/cc8011)
Corresponding author: Prof. Jan Bakker, jan.bakker@erasmusmc.nl
Introduction
Published:
30 November 2009
Critical Care 2009, 13(Suppl 5):S13 (doi:10.1186/cc8011)
A more
evaluation of tissue perfusion can be
This
article iscomplete
online at http://ccforum.com/content/13/S5/S13
achieved
by Central
adding
©
2009 BioMed
Ltd non-invasive assessment of peripheral
perfusion to global parameters [1]. Non-invasive monitoring of
peripheral perfusion is an alternative approach that allows
very early application throughout the hospital, including the
Abstract
Introduction
emergency department, operating room, and hospital wards.
Aonmore complete evaluation of tissue perfusion can be
The rationale
monitoring
peripheral
perfusion
is based
Introduction:
Theofprognostic
value
of continuous
monitoring
of
achieved by adding non-invasive assessment of peripheral
the concept
that peripheral
tissues
are the first
to reflect
early goal-directed
therapy
tissue
oxygen saturation
(StO2) during
hypoperfusion
in has
shock
and investigated.
the last to
perfusion to global parameters [1]. Non-invasive monitoring of
of
critically ill patients
not been
Wereperfuse
conducted during
this
prospective study
theperipheral
hypothesis that
the persistence
of
resuscitation
[1,2].to test
Poor
perfusion
may therefore
peripheral perfusion is an alternative approach that allows
careofadmission
is related to
low
StO2 levels following
be considered
an earlyintensive
predictor
tissue hypoperfusion
and
very early application throughout the hospital, including the
adverse
outcome.
a warning
signal of ongoing shock.
emergency department, operating room, and hospital wards.
Methods: We followed 22 critically ill patients admitted with
The rationale of monitoring peripheral perfusion is based on
increased
lactate practice,
levels (>3 non-invasive
mmol/l). Near-infrared
spectroscopy
In the clinical
monitoring
of peripheral
the concept that peripheral tissues are the first to reflect
rate new
(NIRS)
was used
the thenar eminence
perfusion
canto measure
be performed
easily StO
using
current
2 and the
hypoperfusion in shock and the last to reperfuse during
(RincStO
after a vascularspectroscopy
occlusion test. NIRS
of
StO2 increase such
technologies,
as2)near-infrared
(NIRS) [3].
resuscitation [1,2]. Poor peripheral perfusion may therefore
dynamic
measurementshas
werebeen
recorded
at intensive
caretoadmission
NIRS technology
used
as a tool
monitor tissue
and each 2-hour interval during 8 hours of resuscitation. All repeated
be considered an early predictor of tissue hypoperfusion and
oxygen saturation (StO2) in acutely ill patients [4]. In addition,
compared with Sequential Organ
StO2 measurements were further
a warning signal of ongoing shock.
the analysis of changes in StO2 during a vascular occlusion
Failure Assessment (SOFA), Acute Physiology
and Chronic Health
test, such
as a brief
episode
of forearm
ischemia,
has been
Evaluation
(APACHE)
II and
hemodynamic
physiological
variables:
used
as
a
marker
of
integrity
of
the
microvasculature
the clinical practice, non-invasive monitoring of peripheral
heart rate (HR), mean arterial pressure (MAP), central venous oxygen –In in
recovery
after
the
vascular
occlusion
test
particular,
the
StO
perfusion can be performed easily using current new
2
of peripheral circulation (physical
saturation (ScvO2) and parameters
[5-7]. These
have studied the correlationtechnologies,
of
examination
and reports,
peripheral however,
flow index (PFI)).
such as near-infrared spectroscopy (NIRS) [3].
measurements
and
outcome
where
there
intermittent
StO
2
Results: Twelve patients
were admitted with low StO2 levels (StO2
NIRS technology has been used as a tool to monitor tissue
are
only
limited
data
describing
whether
continuous
monitor<70%). The mean scores for SOFA and APACHE II scores were
oxygen saturation (StO ) in acutely ill patients [4]. In addition,
Although this was an abservational stu
limited number of patients the data sho
StO2 values can be present on admiss
intensive care and that this does not re
resuscitation even when using vasodil
addition the phenomenon is associate
mortality. Therefore more in depth exp
this is warranted.
Intensive Care Med (2005) 31:1316–1326
DOI 10.1007/s00134-005-2790-2
Alexandre Lima
Jan Bakker
REVIEW
Noninvasive monitoring of peripheral perfusion
‣Attractive way to assess the patient
‣Related to basic pathophysiological principles
‣Dynamic approach reveals new possibilities to monitor
Abstract Background: Early hemo- sociated with an increase in capillary
‣Responds to treatment
refill time. The temperature gradients
dynamic assessment of global paunclear
rameters in critically
ill patients fails peripheral-to-ambient, central-to-pe‣Correlation to outcome
ripheral and forearm-to-fingertip skin
to provide adequate information on
Received: 21 February 2005
Accepted: 4 August 2005
Published online: 17 September 2005
! Springer-Verlag 2005
This study was in part supported by materials provided by Hutchinson Technology
and a grant from Philips USA. Both authors
received a grant US $12,000 from Philips
USA and $10,000 from Hutchinson Tech-
tissue perfusion. It requires invasive
monitoring and may represent a late
intervention initiated mainly in the
intensive care unit. Noninvasive
monitoring of peripheral perfusion
can be a complementary approach
that allows very early application
throughout the hospital. In addition,
as peripheral tissues are sensitive to
alterations in perfusion, monitoring
of the periphery could be an early
marker of tissue hypoperfusion. This
are validated methods to estimate
dynamic variations in skin blood
flow. Commonly used optical methods for peripheral monitoring are
perfusion index, near-infrared spectroscopy, laser Doppler flowmetry
and orthogonal polarization spectroscopy. Continuous noninvasive
transcutaneous measurement of oxygen and carbon dioxide tensions can
be used to estimate cutaneous blood
flow. Sublingual capnometry is a
nfection, and Critical Care
The Journal of TRAUMA! Injury, Infection, and Critical Care
verity of
Can Near-Infrared Spectroscopy Identify the Severity of
Shock in Trauma Patients?
i, MD,
Blackbourne,
MD,
Bruce A. Crookes,
MD, Stephen M. Cohn, MD, FACS, Scott Bloch, BS, Jose Amortegui, MD,
obertRonald
Duncan,
PhD, and
Manning,
RN, Pam Li, RN, Matthew S. Proctor, BS, Ali Hallal, MD, Lorne H. Blackbourne, MD,
100
Mean ± SD
Robert Benjamin, MD, Dror Soffer, MD, Fahim Habib, MD, Carl I. Schulman, MD, Robert Duncan,
PhD, and
90
Kenneth G. Proctor, PhD
Thenar StO2 (%)
" 6% (n ! 707);80no shock, 83
! 85); mildBackground:
shock, 83 " 10%
Our (n
recent experimenderatetal
shock,
80
"
12%
(n
!
70
study showed
that peripheral muscle
ere shock,
45
"
26%
(n
! 14).
Normals
tissue oxygen saturation
(StO2), deter60 discrimiStO2mined
values noninvasively
clearly
by =near-infrared
(n
707)
normals
or
no
shock
patients
spectroscopy (NIRS), was more reliable
50shock (p <
tientsthan
withsystemic
severe
hemodynamics or invasive
oxygenation40
variables as an index of trau-
sion:matic
Decreased
mus- of this study
shock.thenar
The purpose
xygenwas
saturation
the range of theto establish
the normal
30reflects
f severe
hypoperfusion
and
nar muscle
StO2 in humans
and the relaed spectroscopy
may
be
a
20
tionship between shock state and StO2 in
od for
rapidly
and noninvatrauma
patients.
10 dysoxia.
sing changes
in tissue
Methods:
This was a prospective,
ords:nonrandomized,
Resuscitation endpoints,
observational, descrip0
ed spectroscopy,
Oxygen
de- volunteers
86.57
± 6.39 (n
tive study in normal human
ue oxygen
saturation.
! 707) and patients admitted to the resus-
citation area of our Level I trauma center
J Trauma.
2005;58:806
–816. a normal StO
(n
! 150).
To establish
2
range, an NIRS probe was applied to the normals, 87 " 6% (n ! 707); no shock, 83
thenar eminence of volunteers (normals). " 10% (n ! 85); mild shock, 83 " 10% (n
Subsequently, in a group of trauma pa- ! 19); moderate shock, 80 " 12% (n !
tients, an NIRS probe was applied to the 14); and severe shock, 45 " 26% (n ! 14).
thenar
No eminence and data were collected The thenar StO2 values clearly discrimiand
stored for offline analysis. StO2 mon- nated the normals or no shock patients
Shock
itoring was performed Mild
continuously and and the patients with severe shock (p <
(n = 98)
noninvasively, and values
were recorded 0.05).
Shock
Moderate
at 2-minute intervals. Five moribund
Conclusion: Decreased thenar mus(n = 19)
Shock
trauma patients were excluded. Members cle tissue oxygen saturation reflects the
= 14) of severeSevere
of our trauma faculty, blinded to StO2 (n presence
hypoperfusion and
Shock
values, classified each patient into one of near-infrared spectroscopy may be a
= 14)and noninvarapidly
four groups (no shock, mild shock, mod- novel method for (n
erate shock, and severe shock) using con- sively assessing changes in tissue dysoxia.
83.23
± 9.79
82.95
± 9.91
80.43 ± 11.90
± 26.07 endpoints,
ventional
physiologic
parameters.
Key Words:44.50
Resuscitation
Results: Mean " SD thenar StO2 Near-infrared spectroscopy, Oxygen devalues for each group were as follows: livery, Tissue oxygen saturation.
J Trauma. 2005;58:806 –816.
There was also no correlation with mean arterial pres(P ! 0.7, R2 ! 0.02), cardiac index (P ! 0.2, R2 !
[94 –99%],
P ! 0.04) (table 2).
Clinically,
22American
(77%)Society sure
Anesthesiology 97%
2009; 111:366
–71
Copyright
© 2009, the
of Anesthesiologists, Inc. Lippincott Williams & Wilkins, Inc.
Anesthesiology 2009; 111:366 –71
© 2009, the American
Society
of Anesthesiologists,
Williams &
Wilkins, Inc.
0.03),
hemoglobin
(P ! Inc.
0.8,Lippincott
R2 ! 0.02),
Ramsay
score
of the survivors had a capillary refill time Copyright
of less than
2 s,
2
(P ! 0.2, R ! 0.02) and norepinephrine dosage (P !
as compared with 13 (77%) of the nonsurvivors (P !
0.9, R2 ! 0.02). In contrast, the lactate plasma level was
1.00). The big toe temperature was hot in 18 (63%)
correlated with StO2 (fig. 3D).
survivors and 8 (61%) nonsurvivors (P ! 0.7). InterestFinally, in M.D.,†
a new cohort of nine patients, we tested
Marc Leone,
M.D.,the
Ph.D.,*
Blidi,were
M.D.,†significantly
François Antonini,
M.D.,*
Bertrand Meyssignac,
ingly,
StO2Sami
values
lower
in the
Sébastien
Bordon,
M.D.,†
Frédéric
Garcin,
M.D.,‡
Aude
Charvet,
M.D.,‡
Valéry
Blasco,
M.D.,*
Marc Leone,
M.D., Ph.D.,*
Blidi,
M.D.,† François
Bertrand
Meyssignac,
M.D.,†
distinct
sites: thenar,
masseter, and deltoid. Briefly,
nonsurvivors
thanSami
in the
survivors
(73% [68Antonini,
– 82%] vs.M.D.,*three
Jacques
Albanèse,
M.D.,
Ph.D.,§
Claude Garcin,
Martin, M.D.!!
Sébastien
Bordon,
M.D.,†
Frédéric
M.D.,‡
Aude
Charvet,
M.D.,‡
Valéry
Blasco,
M.D.,*
the two cohorts were similar in age (54 [40 –59] vs. 59
84% [81–90%], P ! 0.02) (fig. 1).
Jacques Albanèse, M.D., Ph.D.,§ Claude Martin, M.D.!!
[51– 69] yrs, P ! 0.3), sex ratio (33 vs. 33%, P ! 1), SAPS
A receiver operating characteristic analysis confirmed
IIspectroscopy
(50 [43–53]is vs.
47 [38 –58],monitorP ! 0.8), mean arterial
that Growing
StO2 was
significantly
associated
mortality Near-infrared
with an
Background:
evidence
suggests that
the microvasa noninvasive
4
cular dysfunction
is
the
key
element
of
the
pathogenesis
of
pressure
(84
[71–
85]
vs.
80
[72–
85] mmHg, P ! 0.9),
area under the curve at 71% (52–91%, P ! 0.03)
2).
ing,(fig.
providing
real-time feedback. Near-infrared specBackground:
Growing
evidence
that thethe
microvasseptic
shock. This study’s
purpose
was tosuggests
explore whether
Near-infrared
spectroscopy
is
a
noninvasive
urine
output
(50with
[45–120]
vs. 60
[40 –100] mlmonitor· kg–1 · h–1,
troscopy
monitors
only
vessels
a diameter
This association
was
not
found
for pathogenesis
Scvusing
O2, LEONE
lactate
ET
AL.
4 of less
cular dysfunction
is patients
the key
element
of the
ofplasma
outcome
of septic shock
after
early resuscitation
Near-infrared
ing,because
providing
real-time
than
the high
concentration
of 82]
blood
P ! 0.5),
and
ScvO2 feedback.
(76 [73–
vs.in78 [71– 84]spec%, P !
level,
and
norepinephrine
dosage
(fig.tissue
2).
A threshold
of
early
therapy
ispurpose
related to
their
muscle
septicgoal-directed
shock.
This
study’s
was
to explore
whether
the 1 mm
troscopy
monitors
only
vessels
with
a
diameter
of
less
AUCsto
:
0.71
and veins makes photon2 emergence unlikely.
oxygenation.
StOseptic
was
associated
with resuscitation
a sensitivity using
ofarteries
61% and
outcome of
shock
patients
after early
2 at 78%
Methods:
oxygen
(StO2) was
in tissue
thanlight
1 mm
high
concentration
of blood in
Near-infrared
(600because
– 800 nm)the
easily
crosses
biologic
early
goal-directed
therapy
is Nine
related
to monitored
their
muscle
aTissue
specificity
ofsaturation
87%.
(64%)
patients
whose
StO2 was
septic shock patients using a tissue spectrometer (InSpectra
tissues and
is absorbed
by hemoglobin,
myoglogin,
and
arteries
and veins
makes photon
emergence
unlikely.
oxygenation.
Model 325; Hutchinson Technology, Hutchinson, MN). For the
5
Table
2.
Hemodynamics
of
the
Patients
According
to
oxidized
cytochrome, as light
described
tool
Methods: Tissue oxygen saturation (StO2) was
monitored
in
Near-infrared
(600elsewhere.
– 800 nm) This
easily
crosses biologic
purpose of this retrospective study, the StO2 values
were colTheirpatients
Survivalusing a tissue spectrometer (InSpectra
septic
shock
can
quantify
microvascular
dysfunction
in
patients
with
lected at the first measurement done after the macrohemodytissues and is absorbed by hemoglobin, myoglogin, and
Modelvariables
325; Hutchinson
Technology,
Hutchinson,
MN). Forseptic
the shock.6
namic
(mean arterial
pressure, urine
output, central
oxidized cytochrome, as described elsewhere.5 This tool
Survivors
Nonsurvivors
purpose
of thisinretrospective
study,
the
St
O2 values were colvenous
saturation
oxygen)
were
optimized.
can hypothesize that near-infrared spectroscopy
Variables
(n ! 29)
(n ! 13) One
P Value
can quantify microvascular
dysfunction in patients with
Results:
the hemodynamic
variables
were corrected,
no
lected
at After
the first
measurement
done after
the macrohemodycan detect a potential microvascular
dysfunction in the
6
difference
was
observed
between
the
nonsurvivors
and
survishock.
namic variables
(mean
arterial pressure,
output,
central
Heart rate,
beats/min
100 urine
(85–114)
94 (88–115)
0.4 septic
patients
adequately
resuscitated from a macrohemodyvors,
withsaturation
the exception
of pulsewere
oximetry
saturation
Mean
arterial
pressure,
mmHg
79
(72–87) (94%
80 (71–84)
0.5 One can hypothesize that near-infrared spectroscopy
venous
in oxygen)
optimized.
namic
standpoint.
The
purpose of this study was to
[92–97%] vs.Urine
97% [94 –99%],
The 80
StO(40–100)
2 values were
ml/hP ! 0.04). variables
50 (32–80)no 0.3
Results: Afteroutput,
the hemodynamic
were corrected,
can detect
potential
in the
explore whether
the aoutcome
of microvascular
the septic shockdysfunction
pasignificantlyPulse
lower in thesaturation,
nonsurvivors than97
in(94–99)
the survivors
94
(92–97)
difference wasvs.oxygen
observed
between
the
nonsurvivors
and
survi- 0.04
(73% [68 – 82%]
84%
[81–90%],
P
!
0.02).
No
correlations
tients after
early resuscitation
early goal-directed
patients
adequately using
resuscitated
from a macrohemodypercentage
vors,found
withbetween
the exception
of pulse oximetry saturation (94%
were
the StOvariations,*
2 and SpO2 (P ! 0.7).
related
to
their
muscle
tissue
oxygenation.
Pulse pressure
9 (7–12)
8 (4–9) therapy
0.3was
namic standpoint. The purpose of this study was to
[92–97%]
vs. 97%
[94 –99%],
P ! 0.04).
StO2saturavalues were
Conclusions:
In septic
shock patients,
tissueThe
oxygen
percentage
explore whether the outcome of the septic shock pasignificantly
in the
nonsurvivors
than in
the28.
survivors
tion
below 78% lower
is associated
with
increased mortality
at day
Cardiac index,* (l/m²)
3.8 (3.1–4.2) 4.1 (3.4–4.3) 0.9
Further
investigations
are
required
to
determine
whether
the
(73% [68 –Oxygen
82%] vs.
84% [81–90%], P !780.02).
No correlations
central venous
(72–84)
79 (71–85)
0.9 tients after early resuscitation using early goal-directed
Materials
and Methods
correction
of an
impairedthe
level
of2 tissue
oxygen
may
were found
between
St
O
and
Sp
O2 (P saturation
!
0.7).
therapy was related to their muscle tissue oxygenation.
saturation, percentage
improve the outcome of these patients.
Oxygen
Saturation
Is Lower
in Nonsurvivors
than than
OxygenTissue
Tissue
Saturation
Is Lower
in Nonsurvivors
in Survivors after Early Resuscitation of Septic Shock
in Survivors after Early Resuscitation of Septic Shock
Conclusions:
septiclevel,
shock
patients,
oxygen
satura- 0.1
LactateIn
plasma
mmol/l
2.3 tissue
(1.4–2.9)
2.5 (1.5–4.7)
This study was retrospectively conducted in a 16-bed
tion below
78% is associated
with increased
mortality
day 28.
Hemoglobin,
g/dl
9.1 (8.6–11)
9.1at(8.6–11)
0.9
intensive
–1
Further
are
required
determine
whether
the 0.2 care unit of an 800-bed university hospital
Norepinephrine,
!grecommend
· kg · to 0.4
1.0 (0.5–2.0)
IN
septicinvestigations
shock,
guidelines
an (0.1–1.1)
early correc–1
(Hôpital Materials
Nord, Marseille,
France).
Informed consent and
and
Methods
correction
an
impaired
levelurine
of tissue
oxygen
may
min
tion
of meanof
arterial
pressure,
output,
and saturation
central
g. 1. or
Individual
values
of
oxygen
tissue
saturation
(St
O
)
are
approval
by
the
Ethics
Committee
were waived due to
1
Norepinephrine
h
36 (18–56) grow24 (14–61)
0.5
improve
the
outcome
ofduration,
these
patients.
2 Fig. 2. Receiver operating
mixed venous
saturation
in oxygen.
However,
characteristic curves. Tissue oxygen
Dobutamine, percentage
of
3 (10)
(10) the
1
naturewas
of
study.
Thissaturation
study
conducted
a 16-bed
ven for
the survivors
andthenonsurvivors
at 1day
28.observational
Horizontal
(Stthe
Oretrospectively
),
central
venous
saturation inin
oxygen
(ScvO2),
ing evidence
suggests that
microvascular
dysfunc2
patients
2
lactate
plasma
level,
and
norepinephrine
dosage
according
intensive care unit of an 800-bed university hospital to
is the key
element
in sepsis.recommend
This dysfunction
may correces !tion
median.
IN
septic
shock,
guidelines
an early
survival. The areas under the curves are 71% (52–91%, P !
3
Patients
be
associated
with
impaired
outcome.
Atoutput,
first
glance,
(Hôpital
Nord, Marseille, France). Informed consent and
0.03), 56% (31–70%, P ! 0.9), 38% (19 –58%, P ! 0.1), and 39%
* Data arterial
were available
in 30 (70%)
patients.
Data are
presented
as median
tion
of mean
pressure,
urine
and
central
Septic
shock
was
defined
according
to the criteria
ofwaivedSociety
the
correction
of
macrohemodynamics
does
not
preAnesthesiology
2009;
111:366
–71
Copyright
© 2009,
the
American
approval
by
the
Committee
were
due toof Anesthesiolo
(22–56%,
P !Ethics
0.2), respectively.
range) and number of patients1 (percentage).
7
or mixed(interquartile
venous saturation
in oxygen. However, grow-
elowclude
78%whether
did not
survive,
as compared
five
the International
Sepsis Definitions
All paor not
the microvascular
dysfunction with
the(36%)
observational
natureConference.
of the study.
ing evidence suggests that the microvascular dysfunctients received fluid expansion (crystalloids or 6% hycontinues.
ariables
ressure,
output
and all
cted for
red the
mission
mic staressure
pressor
n SOFA
A) were
e differand the
amples
tion of
tatemia
vel !2
luating
ipheral
with unA score
nted as
Differsted by
at were
itney U
ompare
ion beon and
mia, lomed. p
stically
study,
sion to
and 18
marizes
vorable evolution was significantly higher
in patients with abnormal peripheral perfusion (Table 4). Logistic regression analysis showed that the odds of unfavorable
evolution are 7.4 (95% confidence interval 2–19; p " 0.05) times higher for a
patient with abnormal peripheral perfusion than for a patient with normal peripheral perfusion. Differences in global
hemodynamic variables, such as heart
rate, mean arterial
central
veAlexandre
Lima,pressure,
MD; Tim
C. Jansen,
nous pressure, and urine output were not
The prognostic value of the subjective assessment of peripheral
perfusion in critically ill patients
Jan Bakker, MD, PhD
MD; Jasper van Bommel, MD, PhD; Can Ince, PhD;
eral per
sure, or
Objective: The physical examination of peripheral perfusion during the first 48 hours were analyzed (!-SOFA). Individual SOFA
addition
in patients with abnormal periphbased
touching the skin or measuring
capillary refill time has score was significantly
Numberon
of patients
50
Subjectivehigher
Evaluation
Age (yrs)
" mi
been
related to the prognosis 51
of (17–80)
patients with circulatory shock. eral perfusion than in those with normal peripheral perfusion (9that
culatory
Objective Parameters
Abnormal
# 23)peripheral flow
p index
7 " 2,(np #<27)
0.05). Tskin-diff
ItMale/female
is unclear, however, whether 39/11
monitoring peripheral
perfusion 3 vs. Normal
, Tc-toe(n, and
Sequential Organ Failure
8 (2–15)
temic va
afterAssessment
initial resuscitation
still provides information on morbidity in were congruent with the subjective assessment of peripheral
score admission
variable
$0.2 % 2.8
4.6 % 2.8
!0.01
Tskin-diff (°C)
perfusion.6.5The
proportion of patients
>0 was
critically
ill patients.
Therefore,23we
investigated whether
Acute Physiology
and Chronic
(13–35)
Tc-toe (°C)subjec% 3.4
10 %with
4.1 !-SOFA score
!0.01
reflect c
Evaluation
2.3 %higher
1.6
0.7 %abnormal
0.8
!0.01perfuin patients with
peripheral
tive Health
assessment
of IIperipheral perfusion could helpPFI
identify criti- significantly
in critic
Admission category
sion
(77%
vs.
23%,
p
<
0.05).
The
logistic
regression
analysis
cally
ill
patients
with
a
more
severe
organ
or
metabolic
dysfuncDelta
Pneumonia
9
Tskin-diff, forearm-to-fingertip skin-temperature gradient; Tc-toe, central-to-toe temperature differshowed
that
the
odds
of
unfavorable
evolution
are
7.4
(95%
tion
using
the
Sequential
Organ
Failure
Assessment
(SOFA)
score
to moni
Trauma
8
ence; PFI, peripheral flow index.
Abdominal
7
(16). In
confidence interval 2–19; p < 0.05) times higher for a patient with
and
lactate sepsis
levels.
Postoperative
5
whether
abnormal
peripheral perfusion. The proportion of hyperlactatemia
Design: Prospective observational
study. Table 3. Global hemodynamics
variables in abnormal and normal peripheral perfusion
Chronic obstructive
3
fusion in
was
significantly different
between patients
with abnormal and
Setting:
intensive care unit
in a university
Patients
with
abnormal
peripheral
perfusion
pulmonaryMultidisciplinary
disease
Cardiogenic shock
2
normal peripheral Peripheral
perfusionPerfusion
(67% vs. 33%, p < 0.05). The oddsdict
of an
hospital.
function
Hepatic
encephalopathy
Patients:
Fifty consecutive adult2 patients admitted to the in- hyperlactatemia by logistic regression analysis are 4.6 (95%
Hypovolemic/hemorrhagic
2
Normal (n # 27)
Abnormal (n # 23)
p
who pe
were more likely
to have
increased
confidence
interval
1.4 –15; plactate
< 0.05) times higher for a patient
tensive
shockcare unit.
perfusio
withless
abnormal
peripheral
Interventions: None.
Mediastinitis
2
HR (bpm)
90 % 22
% 20
0.53
had significantly
decreases
in perfusion.
lactate94levels
significa
Meningitis
2 Patients were
MAP (mm
Hg)
80 % 14 Subjective assessment
81 % 18 of peripheral perfusion
0.87
Conclusions:
Measurements and Main Results:
considered
to
vorable
OR 7.4
ofwith
increasing
Pancreatitis
2
CVPhad
(mm Hg)
14 % 6organ failure 13 % 7
0.84
physical
examination
following
initial
hemodynamic
resushave
abnormal
peripheral
perfusion
if
the
examined
extremity
had
crease in
Postcardiac arrest
2
Urine output (mL/hr)
111 % 83
72 % 30
0.14
citation in the first 24 hours of admission could identify hemoanCerebrovascular
increase in capillary
accident refill time 1(>4.5 seconds) or it was cool
toring p
Lung
cancer
1
HR, heart inspecrate; MAP, mean
arterial blood
pressure;
CVP, central
dynamically
stable
patients
with avenous
more pressure.
severe organ dysfunction
to the examiner hands. To address reliability of subjective
these pa
Systemic lupus erythematosus
1
tion
and palpation of peripheral 1perfusion, we also measured and higher lactate levels. Patients with abnormal peripheral perspite in
Urosepsis
Table
5.
Proportion
of
hyperlactatemia
and
norTable
4.
Proportion
of
patients
with
unfavorable
tion. Si
fusion
had
significantly
higher
odds
of
worsening
organ
failure
forearm-to-fingertip
skin-temperature
gradient
(T
),
centralSurvivor/nonsurvivor
35/15
skin-diff
mal blood
lactate
levels between
patients with
evolutionflow
("-SOFA
&0) and
favorable
analysis
to-toe temperature difference (Tc-toe), and peripheral
than
did evolution
patients with
normal
peripheral
perfusion
following
index.
("-SOFA score !0) stratified by normal and ab- abnormal and normal peripheral perfusiona
Values
are
given
as
mean
(range)
whereapan abno
TheCare
measurements were taken within 24 hours of admission to initial
resuscitation. (Crit Care Med 2009; 37:934 –938)
Crit
normal peripheral perfusiona
propriate.Med 2009; 37:934–938
ing resu
the intensive care after hemodynamic stability was obtained
refill;Perfusion
skin temperKEY WORDS: physical examination; capillary
Peripheral
main hy
Table 1. Demographic data of the patients
Table 2. Objective parameters of peripheral circulation according to the subjective evaluation of
peripheral perfusion
•
•
•
Medical Centre Rotterdam, PO Box 2040, 3000 CA Rotterdam,
of
py
ed
of
to
th
py
te
RS
on
ed
an
th
s:
Research
Low tissue oxygen saturation at the end of early goal-directed
therapy is associated with worse outcome in critically ill patients
Alexandre Lima, Jasper van Bommel, Tim C Jansen, Can Ince and Jan Bakker
Department of Intensive Care, Room HS3.20, Erasmus MC University Medical Centre Rotterdam, PO Box 2040, 3000 CA Rotterdam,
Critical Care 2009, 13(Suppl 5):S13 (doi:10.1186/cc8011) Available online http://ccforum.com/content/13/S5/S13
The Netherlands
Corresponding author: Prof. Jan Bakker, jan.bakker@erasmusmc.nl
Patients with circulatory failure (increased lactate)
Table 1
Patient
demographic
Published:
30 November
2009 data
This article is online at http://ccforum.com/content/13/S5/S13
Number
of Central
patients
© 2009
BioMed
Ltd
Introduction
Age (years)
enrolled in EGDT protocol (8 hours) immediately following admission
Critical Care 2009, 13(Suppl 5):S13 (doi:10.1186/cc8011)
22
62 (57 to 71)
A Male/female
more complete evaluation of tissue perfusion can be 16/6
achieved
by adding non-invasive assessment
Abstract
Introduction of peripheral 7 (5 to 9)
Sequential Organ Failure Assessment score
A more complete evaluation of tissue perfusion can be
perfusion
global
Introduction:
The to
prognostic
valueparameters
of
continuous
monitoring
Acute Physiology
and
Chronic
Health
Evaluation
II [1].
scoreofNon-invasive monitoring of
23 (16 to 30)
achieved by adding non-invasive assessment of peripheral
tissue oxygen saturation (StO2) during early goal-directed therapy
Admission category
peripheral
is an Wealternative
approach
that allows
perfusion
to global parameters
[1]. Non-invasive monitoring of
of
critically ill patientsperfusion
has not been investigated.
conducted
this
prospective
study
to test the hypothesisthroughout
that the persistence oftheperipheral
Septic
shock
pneumonia,
3 approach
abdominal that
sepsis,
1 meningitis
perfusion
is 3an
alternative
allows
very
early
application
hospital,
including
the
low StO2 levels following intensive care admission is related to
very early application
throughout the hospital,
including4 postoperative,
the
Circulatory failure not associated with sepsis
3 hypovolemic/hemorrhagic,
3 cardiogenic,
2 trauma
adverse
outcome.
emergency
department, operating room,
and
hospital
wards.
emergency
department,
operating
room,
and
hospital
wards.
circulatory
1 cerebrovascular accident, 2 postoperative
Methods:Without
We followed
22 failure
criticallyor illsepsis
patients admitted with
Theperfusion
rationale of monitoring
peripheralon
perfusion is based on
The
rationale
of
monitoring
peripheral
is
based
increased
lactate levels
Noradrenaline
use (>3 mmol/l). Near-infrared spectroscopy
the concept that peripheral tissues are 16
the(72%)
first to reflect
and
the
rate
(NIRS)
was
used
to
measure
the
thenar
eminence
StO
theNoradrenaline
concept
that peripheral2 tissues hypoperfusion
are the infirst
to
reflect
dose
(µg/kg/minute)
0.16
(0.07
to 0.24)
shock and the last to reperfuse
during
of StO2 increase (RincStO2) after a vascular occlusion test. NIRS
hypoperfusion
shock
the last
to reperfuse
during
resuscitation
[1,2]. Poor peripheral
perfusion
may therefore
dynamic
measurements
recorded
at intensiveand
care admission
Dobutamine
use were in
8 (36%)
and each 2-hour interval during 8 hours of resuscitation. All repeated
be considered an early predictor of tissue
Dobutamine dose (µg/kg/minute)
4.3 hypoperfusion
(3.6 to 6.3) and
resuscitation
[1,2]. Poor peripheral perfusion
may therefore
StO
2 measurements were further compared with Sequential Organ
a warning signal of ongoing shock.
Mechanical
ventilation
Failure
Assessment
(SOFA), Acute
Physiologypredictor
and Chronic Health
be
considered
an
early
of
tissue
hypoperfusion and 15 (68%)
Evaluation (APACHE) II and hemodynamic physiological variables:
Survivor/nonsurvivor
17/5of peripheral
In the clinical practice, non-invasive monitoring
heart
rate (HR), meansignal
arterial pressure
(MAP), central venous
oxygen
a warning
of ongoing
shock.
perfusion
can be performed easily using current new
parameters
circulation
(physical
saturation
Data (ScvO
expressed
number, of
asperipheral
median (25th
to 75th
percentile),
or as n (%).
2) andas
examination and peripheral flow index (PFI)).
technologies, such as near-infrared spectroscopy (NIRS) [3].
Medical Centre Rotterdam, PO Box 2040, 3000 CA Rotterdam,
of
py
ed
of
to
th
py
te
RS
on
ed
an
th
s:
Research
Low tissue oxygen saturation at the end of early goal-directed
therapy is associated with worse outcome in critically ill patients
Alexandre Lima, Jasper van Bommel, Tim C Jansen, Can Ince and Jan Bakker
Department of Intensive Care, Room HS3.20, Erasmus MC University Medical Centre Rotterdam, PO Box 2040, 3000 CA Rotterdam,
Critical Care 2009, 13(Suppl 5):S13 (doi:10.1186/cc8011)
The Netherlands
•Patienten met lage StO2 hadden:
•Hogere APACHE II score
Introduction
A more
complete
evaluation of tissue perfusion can be
Meer
orgaanfalen
•
achieved by adding non-invasive assessment
Abstract
Introduction of peripheral
perfusion
to global parameters
[1]. Non-invasive
Slechtere
klaring van
lactaat monitoring of
•
peripheral perfusion is an alternative approach that allows
Corresponding author: Prof. Jan Bakker, jan.bakker@erasmusmc.nl
Published: 30 November 2009
This article is online at http://ccforum.com/content/13/S5/S13
© 2009 BioMed Central Ltd
Critical Care 2009, 13(Suppl 5):S13 (doi:10.1186/cc8011)
A more complete evaluation of tissue perfusion can be
Introduction: The prognostic value of continuous monitoring of
achieved by adding non-invasive assessment of peripheral
tissue oxygen saturation (StO2) during early goal-directed therapy
perfusion to global parameters [1]. Non-invasive monitoring of
of critically ill patients has not been investigated. We conducted
this
prospective
studyapplication
to test the hypothesisthroughout
that the persistence oftheperipheral
perfusion
is an alternative
approach that allows
very
early
hospital,
including
the
low StO2 levels following intensive care admission is related to
very early application throughout the hospital, including the
adverse
outcome.
emergency
department, operating room,
anddepartment,
hospital
wards.
emergency
operating
room, and hospital wards.
Methods: We followed 22 critically ill patients admitted with
rationale of monitoring
peripheralon
perfusion is based on
The rationale of monitoring peripheralTheperfusion
is based
increased lactate levels (>3 mmol/l). Near-infrared spectroscopy
the concept that peripheral tissues are the first to reflect
the rate
(NIRS)
used to measure
the thenar
eminence StO2 and
the was
concept
that
peripheral
tissues
are the first to reflect
hypoperfusion in shock and the last to reperfuse during
of StO2 increase (RincStO2) after a vascular occlusion test. NIRS
hypoperfusion
shock
the last
to reperfuse
during
resuscitation
[1,2]. Poor peripheral
perfusion may therefore
dynamic
measurements were in
recorded
at intensiveand
care admission
and each 2-hour interval during 8 hours of resuscitation. All repeated
be considered an early predictor of tissue hypoperfusion and
resuscitation
[1,2]. Poor peripheral perfusion
may therefore
StO
2 measurements were further compared with Sequential Organ
a warning signal of ongoing shock.
Failure
Assessment (SOFA), Acute
Physiologypredictor
and Chronic Health
be considered
an early
of tissue hypoperfusion and
Evaluation (APACHE) II and hemodynamic physiological variables:
In the clinical practice, non-invasive monitoring of peripheral
heart
rate (HR), meansignal
arterial pressure
(MAP), central venous
oxygen
a warning
of ongoing
shock.
perfusion can be performed easily using current new
saturation (ScvO2) and parameters of peripheral circulation (physical
examination and peripheral flow index (PFI)).
technologies, such as near-infrared spectroscopy (NIRS) [3].
n=22
abnormal peripheral perfusion
n=14
normal peripheral perfusion
n=8
2 hours
29%
13%
n=8
n=2
8 hours
50%
50%
Mortality
StO2 ±60%
n=6
n=6
0%
0%
StO2 ±80%
StO2 and Outcome
•
•
•
•
march-september 2011
221 consecutive patients enrolled expected LOS: 2 days
complete initial resuscitation and stabilization
follow up to Day 28
Dynamics of StO2
mortality
Admission
Normal
StO2
n=221
(>75%) n=181
Evolution during
first 24h
ICU mortality
Normal
15%
n=160
Abnormal
n=21
Abnormal
(<75%) n=40
57%
Normal
n=15
13%
Abnormal
56%
n=25
Odds for mortality: persistent low StO2 during first 24h 7.9 (CI: 3-­‐21, P<0.001)
Odds for mortality: when StO2 decreased to <75% during first 24h 7.1 (CI: 2-­‐21, P<0.001)
Odds for mortality: persistent low StO2 and low peripheral perfusion 9.9 (CI: 3-­‐41ß, P<0.001)
National Incidence Study
1.
Erasmus MC Rotterdam
2.
SFG Rotterdam
3.
Maasstad Rotterdam
4.
RdG Delft
5.
OLVG Amsterdam
6.
Gelre Apeldoorn
7.
Jeroen Bosch Den Bosch
8.
Catherina Eindhoven
9.
MMC Veldhoven
10. Isala Zwolle
11. Gelderse Vallei Ede
12.
Canisius Nijmegen
Characteristics
‣
‣
‣
‣
‣
‣
‣
‣
‣
151 patients included
100 (66%) male
Age: 70 (60-77)
Temp: 36.7±1.6 0C
HR: 90±18 b/min
MAP: 83±17 mm Hg
SpO2: 96±9 %
Lactate: 1.5±1.3 mmol/L
Hb: 6.3±1.2 mmol/L
‣
‣
‣
‣
‣
SOFA: 6±3
APACHE II: 21±9
Days already in ICU: 9±11
Length of stay: 21±25
Mortality: 21 (14%)
Characteristics
StO2 ≤ 75% in 40 patients 26.5%
StO2 ≤ 70% in 26 patients
StO2 > 91% in 15 patients 10%
StO2≥ 95% in 5 patients
Abnormal StO2 is present in 36% of the ICU patients
70%
91%
Normal vs Abnormal StO2
Days in ICU
Age
APACHE
SOFA
LOS (days)
Mortality
Temp 0C
HR b/min
MAP mm Hg
SpO2 %
StO2 %
Lactate mmol/L
DOB ug/kg.min
Normal (n=110)
Abnormal (n=41)
8±8
67±14
20±9
6±3
19±23
14 (13%)
36.8±1.4
91±18
85±17
95±11
83±4
1.4±1.0
0.17±0.85
13±16
67±13
24±9 *
9±3 *
26±30
7 (17%)
36.5±2.0
90±17
79±13
98±2
73±18 *
1.7±1.8
0.82±2.18 *
Conclusion
‣
The periphery is the “canary of the body”
•
Responds rapidly to changes in blood flow
‣
Both temperature, color, perfusion parameters and oxygen saturation
parameters change in a similar direction
‣
Peripheral parameters can be collected within seconds to a few minutes
in every patient
‣
Peripheral parameters (stat and changes) are closely linked to morbidity
and mortality
‣
‣
Peripheral parameters respond to treatment
Unclear whether the patient will benefit from this