Document 6475774

526
An Unusual Case of Refractory Campylobacter jejuni Infection in a Patient with
X-Linked Agammaglobulinemia: Successful Combined Therapy with Maternal
Plasma and Ciproftoxacin
Ingo B. Autenrieth, Volker Schuster, Jann Ewald,
Dag Harmsen, and Hans W. Kreth
From Institut fur Hygiene und Mikrobiologie der Universitdt Wiirzburg,
and Kinderklink der Universitiit Wurzburg, Wilrzburg, Germany
An increasing body of evidence suggests that Campylobacter
jejuni is one of the most frequent enteropathogenic microorganisms causing enteritis and enterocolitis in humans [1]. Furthermore, extraintestina1 manifestations-including skin lesions
and chronic persistent infections-may occur, particularly in
immunocompromised individuals (such as patients with severe
hypogammaglobulinemia) [1-6]. Moreover, C. jejuni infections have been linked to immunopathologic sequelae, including reactive arthritis and Guillain-Barre syndrome [1].
Treatment with antibiotics such as erythromycin, imipenem,
and ciprofloxacin is recommended in severe cases (e.g., cases
in immunocompromised patients) [1,2]. However, a significant
number (5%-50%) of C. jejuni isolates have been demonstrated to be resistant to several antibiotics, including quinolones [I, 7, 8]. Hence, additional immunomodulating therapy
including intravenous and/or oral Ig preparations has been used
in the treatment of C.jejuni infections in immunocompromised
patients and has proved to be efficient [2, 9, 10].
In agreement with clinical data, recent work from our laboratory demonstrated that IgM antibodies may enhance the bactericidal effects of polymorphonuclear leukocytes (PMNLs)
against C. jejuni [10]. However, because of the lack of appropriate animal models for C. jejuni infections, there are practically no systematic data available on the immunobiology of
C. jejuni infections.
Herein, we report a case of refractory systemic infection with
an unusual hippurate-negative C. jejuni strain that occurred in
Received 5 January 1996; revised 2 April 1996.
Reprints or correspondence: Dr. Ingo B. Autenrieth, Institut fur Hygiene
und Mikrobiologie der Universitat Wurzburg, Josef-Schneider-Strazie 2, Bau
17, D-97080 Wiirzburg, Germany,
Clinical Infectious Diseases 1996;23:526-31
© 1996 by The University of Chicago. All rights reserved.
1058-4838/96/2303-0017$02.00
a patient with X-linked agammaglobulinemia (XLA); conventional therapy with antibiotics and Igs failed to cure the infection.
Case Report and Laboratory Investigations
A baby boy was born as the third child of healthy parents.
At 6 months of age, severe hypogammaglobulinemia was diagnosed; the count of circulating B cells was very low, but the
counts and functions of T cells were normal. Since his older
brother also had extremely low serum levels of Ig as well as
no circulating B cells, XLA was diagnosed and confirmed by
indirect DNA methods (including restriction fragment length
polymorphism analysis). The boy received regular intravenous
infusions of Ig (Endobulin containing pooled human Igs
[99.9%, IgG; 0.1%, IgA], Immuno, Heidelberg, Germany) at
a dose of 400 mg/kg of body weight every 4 weeks and was
doing well except for repeated episodes of otitis media.
At the age of 7 years, the boy was admitted to the Children's
Hospital of Wiirzburg Medical School (Wiirzburg, Germany)
because of a 4-month history of a generalized pruritic rash.
Physical examination revealed numerous grouped and disseminated maculopapular skin lesions with small vesicles or crusts
on the summit (figure 1). There was extensive excoriation. The
lesions were most pronounced on the trunk and on the extensor
surfaces of the extremities. There were no other abnormal physical findings. The patient was afebrile and had no history of
fever or chills during the preceding months.
Laboratory investigations revealed an erythrocyte sedimentation rate of 4 mm/h and a WBC count of 12.6 X 109/L
(57% neutrophils, 25% lymphocytes, 11% monocytes, 4% band
forms, 2% eosinophils, and 1% basophils). Three weeks after
the last dose of Endobulin, the serum concentration of IgG was
475 mg/dL. IgA, IgM, and IgE levels were undetectable. Total
complement activity (CH50, AP50) was normal.
Downloaded from http://cid.oxfordjournals.org/ by guest on September 9, 2014
An unusual hippurate-negative strain of Campylobacter jejuni caused a chronic refractory infection in a patient with X-linked agammaglobulinemia; this infection persisted for > 2 years despite
therapy with various antibiotics and immunoglobulins (Igs). To characterize the defense status of
this patient, several in vitro studies, including those with T cells and polymorphonuclear leukocytes
(PMNLs), were performed. T cell responses specific for C. jejuni were only weak in this patient.
Chemiluminescence and bacterial killing studies with PMNLs revealed that the bactericidal activity
of PMNLs against Campylobacter was enhanced more vigorously by maternal serum than by commercial Ig preparations. On the basis of these results, combined treatment with ciprofloxacin and
maternal plasma was initiated, and the C. jejuni infection was rapidly cured. This case report shows
that in vitro immunologic assays may be useful for characterizing immune functions of patients
with chronic or refractory C. jejuni infections, thus leading to individual treatment strategies.
em 1996;23 (September)
Refractory C. jejuni Infection and XLA
527
After admission, routine procedures were used to isolate
C. jejuni several times from blood and stool specimens from
the patient; cultures of the samples on plates of meat-yeast
extract agar were incubated for 48 hours at 42°C under microaerobic conditions (Anaerocult C, Merck, Darmstadt, Germany). The strain did not hydrolyze hippurate but was identified as C. jejuni by determination of ]6S rRN A gene sequences
according to the method of Giesendorf et al. [II] and as
C. jejuni serotype Lior II by slide agglutination according to
the method of Lior et al. [12]. A region of the 16S RNA
gene was amplified by PCR analysis with the Campylobacterspecific primer pair C422 and C490, which correspond to Escherichia coli positions 399-420 and 825-803, respectively [12].
The resulting PCR fragment (426 bp) was sequenced by the
Taq-cyc1e Dye Deoxy terminator method [13] in combination
with a 373A automatic sequencer (Perkin Elmer, Weiterstadt,
Germany) . Database searches revealed 100% sequence homology with C. jejuni. The presence of polymerase inhibitory substances and the specificity were controlled with the use of
human ,B-actin-specific primers [13], which amplified a fragment of 348 bp.
The C. jejuni strains repeatedly isolated from the patient
before and after antimicrobial therapy appeared to be identical
(determined by the above-mentioned methods) and exhibited an
identical pattern of antimicrobial susceptibility, thus suggesting
chronic persistent infection rather than reinfection with other
strains of C. jejuni. The strains were susceptible to gentamicin
(10 f..lg), tetracycline (30 Jig), imipenem (10 fig) , and ciprofloxacin (5 Ji.g) but were resistant to erythromycin (15 f..lg),
ampicillin (10 Jig). mezlocillin (30 fig), cefotiam (30 Jig). cefotaxime (30 Ji.g), and co-trimoxazole (25 Jig).
The boy was initially treated intravenously with amoxicillin
(500 mg three times a day for ]4 days) and doxycycline (200
mg/d for 10 days) and then with imipenem (500 mg twice a
day for 10 days). He also received bovine colostrum (10 g
orally twice a day for ]0 days; Lactobin, Biotest Pharma,
Dreieich, Germany). Moreover. Endobulin was substituted
by an intravenous Iglvl-containing preparation (Pentaglobin,
Downloaded from http://cid.oxfordjournals.org/ by guest on September 9, 2014
Figure 1. Campy/obaeterjejuni-induced skin lesions in a patient with X-linked agammaglobulinemia before (left) and 2 months after (right)
combined treatment with ciprofloxacin and maternal plasma.
Autenrieth et al.
528
CID 1996;23 (September)
Figure 2. Amplification
2
3
4
5
6
7
8
9
kb
3054
2036
1635
1018
516
~~~ -
298 =
220
154
Experimental Procedures
Proliferative responses of peripheral blood mononuclear
cells (PBMCs). PBMCs were isolated by Ficoll (Biochrom,
Berlin) density gradient centrifugation. PBMCs (105) were inoculated in triplicate into wells on 96-well microtiter plates
that contained RPMI 1640 tissue culture medium (Biochrom,
Berlin) supplemented with 10% AB serum (Sigma, Deisenho-
11
12
13
14
weight marker(Bethesda Research
Laboratories, Bethesda, MD; l-kilobase [kb1 DNA ladder); lane 2,
control C jej uni DNA (426 bp);
lane 3, control Campylobacter coli
DNA; lane 4, Cjejuni isolatefrom
the patient with X-linked agammaglobulinemia; lane 5, first skin
biopsy specimen from the patient;
lane 6, second skin biopsy specimen from the patient; lane 7. first
skin biopsy specimen plus control
C jejuni DNA; lane 8, Haem ophiIus infiuenzae: lanes 9 and 10, Vibrio species; lane II, human DNA
(PCR control with fJ-actin-specific
primers [348 bpj) ; lane 12, negative control (DNAextraction); lane
13. negative control (Pf' R reagents); and lane 14. molecular
weight marker (see lane I).
fen, Germany) and horse serum (Gibco BRL, Eggen stein, Germany). The cells were stimulated with 10 JLg of a heat-killed
preparation of C. jejuni, Listeria monocyt ogenes, Yersinia enterocolitica, or Salmonella typhimurium per milliliter or with
100 JLg of tetanu s toxoid per milliliter. Six days later, 0.5 JLCi
of eH]thymidine was added to each well. After a further 6
hours of incubation, cells were harvested, and eH]thymidine
uptake was determined.
PMNL functions. Blood was taken from the patient, his
mother , and unrelated healthy controls (laboratory personnel).
PMNLs were prepared as recently described [14]. Luminol
(Sigma)-amplified chemiluminescence (CL) of PMNLs was
performed, and the PMNL responses were measured by a microplate chemiluminometer (Hamamatsu Photonics, HCITsching, Germany) at 37°C as described previously [14]. Briefly,
105 PMNL s per well were incubated with RPMI 1640 cell
culture medium containing 2.5 X 10- 5 M luminol. For stimulation, viable C. jejuni (10- 100 bacteria per PMNL) or 10 JLg
of heat-killed C.jejuni or zymosan (Sigma, Munich) was added
to various Ig preparations or sera. Light emission was recorded
continuously for 90 minut es. The total light emission was calculated as previously described [14]. Three separate experiments with triplicate wells revealed comparable results.
Modulation of the bactericidal act ivity of PMNLs against
C. jej uni was shown by the above-described microplate assay
with various Ig preparations and sera . PMNLs were incubated with C. jejuni at a ratio of I: I00 for 30 minutes at
37°e. Subsequently, 30 JLg of gentamicin/mL was added to
kill extracellular bacteria. Two hours later, cells were
washed and lysed with PBS containing 0.5% bovine serum
albumin and 0.5% Tergitol (Fluka, Buchs, Switzerland).
Downloaded from http://cid.oxfordjournals.org/ by guest on September 9, 2014
Biotest Pharma). Although blood cultures became sterile , serial
stool cultures , spanning nearly 18 months , demonstrated persistent excretion of C. jejuni. Furthermore, there was no beneficial
effect of treatment on skin manife station s, which appeared to
be intractable.
Several months later, skin biopsy specimens were taken to
elucidate the etiology of the skin alterations. Although skin
aspirates and biopsy specimens were sterile, PCR techniques
revealed the presence of C. jejuni DNA in the skin lesions ,
thereby suggesting that the skin lesions were due to C. jejuni
infection (figure 2). Cultures of parallel blood specimens taken
at this time were negative .
Therefore, on the basis of the results obtained by the in vitro
studies (which are discussed below) and with the informed
consent of both parents, the decision was made to treat the boy
with maternal plasma (10 mL/[kg of body weight· w] for 5
weeks) in combination with ciprofloxacin (250 mg orally twice
a day for 3 weeks). Stool culture s became negative 5 days after
the start of treatment, and the skin lesions completely resolved
within 6 weeks (figure I) . During a 12-month follow-up period,
stool cultures remained negative , and the patient continues to
be asymptomatic.
10
of
Campylobacter jejuni DNA by
PCR analysis. Lane 1. molecular
em 1996;23 (September)
529
Refractory C. jej uni Infection and XLA
Duplicates of serial dilutions of these suspensions were
plated on meat-yeast extract agar, and cfu were counted after
incubation for 48 hours at 42°C.
Statistical analysis. Differences between mean values
were analyzed by the Student's t test. P values of < .05 were
considered statistically significant.
300
IV
u
c
'E 'iii
~~
E .=
IV
U
150
100
*
*
50
0
PMNL mother mother patient patient patient patient patient
Serum mother patient mother patient patient patient patient
Pentagl Endobul Polyglo
Ig prep.
Figure 3. Chemiluminescence responses of polymorphonuclear
leukocytes (PMNLs). The PMNLswere from a patient with X-linked
agammaglobulinemia and the patient's mother. The PMNL responses
were recorded when the cells were exposed to viable C. jejuni (light
bars) or zymosan (dark bars) in the presence or absence of various
Ig preparations (Ig prep.) and sera. Values of light emission are given
as the integrals of the area under the curve and represent the means
for triplicate wells. Values obtained with the patient's PMNLs and
serum represented 100%. Pentagl = Pentaglobin (Biotest Pharma,
Dreieich, Germany); Polyglo = Polyglobin N (Tropon Werke
GmbH & Co., KG, Cologne, Germany); Endobul = Endobulin (Immuno, Heidelberg, Germany). The asterisks indicate values that significantly (P < .05) differ from reference values (those for PMNLs
from the patient plus serum from the patient).
were exposed to C. jej uni in serum derived from the patien t' s
health y mother (P < .05; figure 3). Likewise, comparable results were obtained when the mother' s PMNL s (P < .05) or
PMNLs from unrelated health y control s (data not shown) were
used (figure 3), thus suggesting that certain components of the
patient ' s serum that are required for triggering C. jeju nispecific CL respo nses by PMNLs were lacking. In addition,
stimulation with zymosan, which is opsonized by complement
C3b, caused compara ble CL responses by PMNLs from the
patient and his mother (figure 3), thereby suggesting that C.
j ejlllli -specific compounds (such as antibodies ) were lacking
in the patient's serum .
Since it has not yet been establi shed that reactive oxygen
metabolites mediate the killing of C. jeju ni, the bactericidal
activity of PMNL s was directly determined. For this purpo se,
PMNLs were mixed with viable C. jej uni in various sera and
19 compounds, and a gentamicin assay was perform ed. The
results (figure 4) showed that the CL respons es of PMNL s that
are depicted in figure 3 correlated directly with bacterial killing
by PMNL s. Thus , apart from Pentaglobin, which significantl y
enhanced the bactericidal effects of PMNLs against C. jejuni
(P < .05), the commercial Ig preparat ions had no effec t and
even reduced killing. However, as observed for CL respons es.
addition of mate mal serum to the patient' s PMNLs most vigorously enhan ced the bactericidal activity against Campylobacter
(P < .05).
On the basis of these resu lts, the decision was made to
transfuse matemal plasma to enhance the patien t's defe nse
again st C. j ejuni infection. Since the patient's clini cal condition
Downloaded from http://cid.oxfordjournals.org/ by guest on September 9, 2014
A hippurate-negative C. jejuni strain caused a chronic refrac tory infection in a patient with XLA ; this infection persi sted
despite treatment with different antibiotics (amoxicillin, doxycycline, and imipenem) and intravenously and orally administered Ig preparations (Endobulin, Pentaglobin, and Lactobin).
This trea tmen t caused clearance of C. j ejuni from the blood
but not from the intestine and skin lesions .
The rationale for new treatment with matemal plasma and
ciprofloxacin for this patient was based on the findings of PCR
analysis, which revealed that C. j ejuni may pers ist not only in
the intestine (or the blood) but also in tissue (e.g., skin) lesions.
Moreover, the possibility that C. jej uni may persist intracellularly cannot be excluded, and thus , this persi stence may evade
substitution therapy with Ig or certain antibiotics (such as gentamicin) that only act extracellularly. Therefore, we believed
that combination therapy with antibiotics that act both intrace llularly and extracellularly and immunomodulating compounds
that act rather extracellularly should mediate a rapid clearance
of the bacteria from both the blood and the gut; on the other
hand , we thought that this combination therapy should stimu late and support the defense statu s of this patien t by eliminating
the pathogen at sites where antibiotics were not effective (e.g.,
skin abscesses).
Moreover, antibiotic therapy alone, in the absence of host
defense mechanisms, often appears to be unable to cure infec tions, espec ially those with intracellular pathogens (authors'
unpublished observations) . Further, a previous study [2] reported the beneficial effect of combination therapy with imipenem and plasma for chronic campylobacter infections.
Before the initiation of the new treatment, in vitro studies
were performed to investigate whether the patient's PMNLs
were principally able to kill the pathogen at least under experimental conditions and which immunologic compound might
be effective as immunomodulating treatment. To evaluate the
bacteri cidal activity of the patient' s PMNLs against Campy lobacter, CL responses of PMNLs were assayed in the presence
of various Ig preparations and sera . The patient's PMNLs
showed only weak CL responses when they were expo sed to
viable Cijejuni in the patient' s serum or commercial Ig preparation s (figure 3). Experiments with heat-killed C. jej uni revealed
comparable results (data not shown).
In agreement with recently publi shed results [10], only IgMcontaining preparations (Pentaglobin) enhanced the bactericidal
effects ofPMNLs against Campylobacter (P < .05). However,
the strongest responses to CL were observed when PMNLs
*
*
~ C 200
~.a-
.s:
Results and Discussion
250
IV~
530
.c:~
-
Autenrieth et al.
175
.~
.....
125
...
O ~
100
:::l0
-l'Cl
(,).Q
0
~
l'Cl
25
E
o
*
75
50
0
* *
100%
150
*
*
PMNL mother mother patient patient
Serum mother patient mother patient
Ig prep.
patient patient patient
patient patient patient
Pentagl Endobul PolygJo
was poor because of the chronic persistent C. jejuni infection,
we decided to start combination therapy with maternal plasma
and ciprofloxacin to rapidly clear the pathogen.
To study the effect of transferring maternal plasma to the
patient, blood was sampled at various intervals after transfusion, and Luminol-treated PMNLs were assayed for CL. The
results showed that immediately after transfusion the patient's
serum enhanced CL responses >30% when PMNLs were exposed to C. jejuni (P < .05; data not shown). These results
suggest that transfer of maternal plasma to the patient might
have enhanced the bactericidal activity of the patient's PMNLs
against Campy/obacter in vivo.
Therapy with ciprofloxacin (250 mg orally twice a day) was
continued for another 20 days. Five days after the therapy was
started, C. jejuni could no longer be detected in blood, skin,
and stool specimens, thus suggesting that the combination therapy cured the C. jejuni infection in this patient. FolIow-up for
a further 12 months confirmed that the chronic refractory
C. jejuni infection in this patient was cured with combination
treatment with maternal plasma and ciprofloxacin.
However, we do not yet know which component of maternal
plasma mediated the observed immunobiological effects. Surprisingly, we detected only insignificant low levels of C. jejuni-specific antibody in maternal serum by means of immunoblotting (data not shown). Moreover, it is not yet known
whether inhibitory components were present in the patient's
serum as suggested by the data presented in figures 3 and 4.
Thus, systemic studies are required to elucidate the therapeutic
effect of maternal plasma. On the other hand, we cannot exclude the possibility that ciprofloxacin therapy alone might
have caused eradication of the pathogen.
It is interesting that T cell responses specific for C. jejuni
in our patient were very weak (P < .05) only when they were
compared with those in a patient with acute C.jejuni infection,
while T cell responses upon exposure to a recalI antigen (tetanus toxoid) in our patient were pronounced compared with
those in controls (table 1). Whether these findings are because
of a so far unknown T celI deficiency in the patient, a T celI
tolerance due to persistent infection, or a deficiency in antigen
presentation as a result of the lack of phagocytosis of C. jejuni
in vivo is not yet clear. However, although the role ofT cells in
the host response to C.jejuni infection has yet to be established,
recent data from models of infection with other enteroinvasive,
extracellular bacteria (such as Y. enterocolitica) argue for such
a role [15, 16].
Finally, the strain isolated from our patient did not hydrolyze
hippurate but was identified as C. jejuni by 16S rRNA gene
sequencing. Although >99% of C. jejuni strains do hydrolyze
Table 1. Proliferative responses of PBMCs to various bacterial antigens.
Proliferative response* to indicated antigen (proliferation index')
PBMCs
Nil
Patient with XLA
188 ::':: 3
Mother ofpatient with XLA 280 ::':: 67
Patient with acute
Campylobacter jejuni
enteritis!
317 ::':: 157
HKC
443 ::':: 116 (204)
638 ::':: 110 (2.3)
10,164 ::':: 1,740 (32.1):
HKL
HKS
TT
1,354 ::':: 200 (7.2)
695 ::':: 195 (3 .7)
2,365 ::':: 800 (804) 2,062 ::':: 262 (704)1
2,710::':: 310 (1404)
3,464 ::':: 440 (1204)
10,620 ::':: 1,010 (56.5)
1,823 ::':: 695 (6.5)1
1,477 ::':: 220 (4.7)
632 ::':: 235 (2.0jI
337 ::':: 25
HKY
(1.1)
1,420 ± 52 (4.5)1
NOTE. Freshly isolated peripheral blood mononuclear cells (PBMCs; 10') were inoculated intriplicate into wells containing 200,'lL oftissue culture medium;
the cells were stimulated with 10 ,'lg of heat-killed C. jejuni (HKC), heat-killed Listeria monoeytogenes (HKL), heat-killed Yersinia enterocolitica (HKY), or
heat-killed Salmonella typhimurium (HKS) per milliliter or with 100 ,'lg oftetanus toxoid (TT) per milliliter and were incubated at 37°C for 6 days. XLA =
X-linked agammaglobulinemia.
• Means of [3H]thymidine uptake (cpm) ::':: 1 SO for triplicate wells.
t Mean ofantigenic proliferation divided by mean ofnonantigenic spontaneous proliferation.
: Values differing significantly (P < .05) from values for patient with XLA.
§ PBMCs from a 2-year-old patient with anacute C. jejuni infection were isolated and used asa positive control.
Downloaded from http://cid.oxfordjournals.org/ by guest on September 9, 2014
Figure 4. Number (cfu) ofviable Campy/obaeterjejuni in thepresence or absence of various Ig preparations (Ig prep.) and sera2 hours
after incubation with polymorphonuclear leukocytes (PMNLs) from
a patient with X-linked agammaglobulinemia and the patient's
mother; a gentamicin assay was used to determine the number (cfu)
of bacteria. Values represent the means for triplicate wells. Values
obtained with the patient's PMNLs and serum represented 100%.
Pentagl = Pentaglobin (Biotest Pharma, Dreieich, Germany); Polyglo
= Polyglobin N (Tropon Werke GmbH & Co., KG, Cologne, Germany); Endobul = Endobulin (Immuno, Heidelberg, Germany). The
asterisks indicate values thatdiffer significantly (P < .05) from reference values (those for PMNLs from the patient plus serum from the
patient).
CID 1996;23 (September)
em
1996;23 (September)
Refractory C. jejuni Infection and XLA
hippurate while Campylobacter coli strains do not, we and
other investigators [17] recommend that hippurate hydrolysis
[18] should not be used as a sole criterion for differentiation
between these Campylobacter species.
References
9. Hammarstrorn V, Smith CI, Hammarstrorn 1. Oral immunoglobulin treatment in Campylobacter jejuni enteritis [letter]. Lancet 1993:341:1036.
10. Autenrieth lB, SchwarzkopfA, Ewald JH, Karch H, Lissner R. Bactericidal
properties of Campylobacter jejuni-specific immunoglobulin M antibodies in commercial immunoglobulin preparations. Antimicrob Agents
Chemother 1995;39:1965~9.
11. Giesendorf BA, Quint WG, Henkens MH, Stegeman H, Huf FA, Nicsters
HG. Rapid and sensitive detection of Campylobacter spp. in chicken
products by using the polymerase chain reaction. Appl Environ MicrobioI1992;58:3804~8.
12. Lior H, Woodward DL, Edgar JA, Laroche LJ, Gill P. Serotyping of
Campylobacterjejuni by slide agglutination based on heat-labile antigenic factors. J Clin Microbiol 1982; 15:761 8.
13. Harmsen D, Heesemann J, Brabletz T, Kirchner T, Muller-Hermelink HK.
Heterogeneity among Whipple's-disease-associated bacteria [letter].
Lancet 1994; 343: 1288.
14. Ewald JH, Heesemann J, Rudiger H. Autenrieth lB. Interaction ofpolymorphonuclear leukocytes with Yersinia enterocolitica: role of the Yersinia
virulence plasmid and modulation by the iron-chelator desferrioxarnine
B. J Infect Dis 1994;170:140-50.
15. Autenrieth IB, Tingle A, Reske-Kunz A, Heesemann 1. T lymphocytes
mediate protection against Yersinia enterocolitica in mice: characterization of murine T-cell clones specific for Y enterocolitica. Infect Immun
1992;60:1140~9.
16. Autenrieth IB, Vogel U, Preger S, Heymer B, Heesemann.l. Experimental
Yersinia enterocolitica infection in euthymic and T-cell-deficient
athymic nude C57BL/6 mice: comparison of time course. histomorphology, and immune response. Infect Immun 1993:61 :2585-95.
17. Totten PA, Patton CM, Tenover FC, et a1. Prevalence and characterization
of hippurate-negative Campylobacter jejuni in King County, Washington. J Clin MicrobioI1987:25:l747-52.
18. Hwang MN, Ederer GM. Rapid hippurate hydrolysis method for presumptive identification of group B streptococci. J Clin Microbiol 1975: I:
114-5.
Downloaded from http://cid.oxfordjournals.org/ by guest on September 9, 2014
1. Nachamkin I. Campylobacter infections. Current Opinion in Infectious
Diseases 1993;6:72-6.
2. Kerstens PJ, Endtz HP, Meis JF, et a1. Erysipelas-like skin lesions associated with Campylobacter jejuni septicemia in patients with hypogammaglobulinemia. Eur J Clin Microbiol Infect Dis 1992; 11:842~ 7.
3. Perlman DM, Ampel NM, Schifman RB, et a1. Persistent Campylobacter
jejuni infections in patients infected with the human immunodeficiency
virus (HIV). Ann Intern Med 1988; 108:540-6.
4. Lever AM, Dolby JM, Webster AD, Price AB. Chronic campylobacter
colitis and uveitis in a patient with hypogammaglobulinaemia. Br Med
J Clin Res Ed 1984;288:531.
5. Chusid MJ, Coleman CM, Dunne WM. Chronic asymptomatic Campylobacter bacteremia in a boy with X-linked hypogammaglobulinemia.
Pediatr J Infect Dis 1987; 6:943-4.
6. Borleffs JC, Schellekens JF, Brouwer E, Rozenberg-Arska M. Use of an
immunoglobulin M containing preparation for treatment of two hypogammaglobulinemic patients with persistent Campylobacter jejuni infection. Eur J Clin Microbiol Infect Dis 1993; 12:772-5.
7. Kuschner RA, Trofa AF, Thomas RJ, et a1. Use of azithromycin for the
treatment of campylobacter enteritis in travelers to Thailand, an area
where ciprofloxacin resistance is prevalent. Clin Infect Dis 1995;21:
536-41.
8. Tee W, Mijch A, Wright E, Yung A. Emergence of multidrug resistance
in Campylobacterjejuni isolates from three patients infected with human
immunodeficiency virus. Clin Infect Dis 1995;21:634-8.
531