Liver Function in Early Lyme Disease

HAROLD W. HOROWITZ,1 BRAD DWORKIN,2 GILDA FORSETER,1 ROBERT B. NADELMAN,1 CHRIS CONNOLLY,1
BARBARA B. LUCIANO,3 JOHN NOWAKOWSKI,1 THOMAS A. OBRIEN,4 MARK CALMANN,4 AND GARY P. WORMSER1

To evaluate the frequency, pattern, and severity of

liver function test abnormalities in patients with Lyme
disease associated with erythema migrans (EM), 115 individuals with no other identifiable cause for liver function test abnormalities who presented with EM between
July 1990 and September 1993 were prospectively evaluated. For individuals with abnormal liver function tests,
common causes of hepatitis, including hepatitis A, B,
and C, were excluded. A local control group was used
for comparison. Forty-six (40%) patients had at least one
liver test abnormality, and 31 (27%) had more than 1
abnormality compared with 19 (19%) and 4 (4%) of controls, respectively (P .01 for each comparison). g-Glutamyl transpeptidase (28%) and alanine transaminase
(ALT) (27%) were the most frequently elevated liver
function tests among Lyme disease patients. Anorexia,
nausea, or vomiting was reported by 30% of patients, but
did not occur more frequently in patients with elevated
liver function tests compared with those with normal
values. Patients with early disseminated Lyme disease
were more likely to have elevated liver function studies
(66%) compared with patients with localized disease
(34%) (P .002). After antibiotic treatment, elevated liver
function tests improved or resolved in most patients.
Liver function test abnormalities are common in patients with EM but were mild, most often not associated
with symptoms, and improved or resolved by 3 weeks
after the onset of antibiotic therapy in most patients.
(HEPATOLOGY 1996;23:1412-1417.)

Abbreviations: EM, erythema migrans; ALT, alanine transaminase; AST,

aspartate transaminase; ELISA, enzyme-linked immunosorbent assay.
From The 1Department of Medicine, Divisions of 1Infectious Diseases and
2
Gastroenterology, Lyme Disease Diagnostic Center, Westchester County Medical Center and New York Medical College, Valhalla, NY; and the 3Employee
Health Service, Westchester County Medical Center, Valhalla, NY; and 4Ortho
Diagnostic Systems Inc., Raritan, NJ.
Received May 30, 1995; accepted January 23, 1996.
Supported in part by Cooperative Agreement No. U50/CCU 210280-01 from
the Centers for Disease Control and Prevention (G.P.W.) and by grants RO1AR41508 (R.B.N., J.N., G.P.W.) and RO1-AR43135 (G.P.W.) from the National
Institute of Arthritis and Musculoskeletal and Skin Diseases.
The contents of this manuscript are solely the responsibility of the authors
and do not represent the official views of the Centers for Disease Control and
Prevention or the National Institute of Arthritis and Musculoskeletal and Skin
Diseases.
Address reprint requests to: Harold W. Horowitz, M.D., Westchester County
Medical Center, Division of Infectious Diseases, Room 209 Macy Pavilion,
Valhalla, NY 10595.
Copyright q 1996 by the American Association for the Study of Liver
Diseases.
0270-9139/96/2306-0017$3.00/0

Lyme disease is a multisystemic infection caused by

the tick-borne spirochete, Borrelia burgdorferi. A great
deal of attention has been paid to the dermatologic,
musculoskeletal, cardiovascular, and neurological
manifestations of Lyme disease.1,2 Acute hepatitis has
been reported in the early stages of Lyme disease,3-6
and hepatic enzyme elevations have been reported in
20% to 50% of patients with the characteristic rash of
early Lyme disease, erythema migrans (EM).2,7-10 However, there have been no prospective evaluations of hepatic dysfunction in treated patients for whom other
major causes for liver disease were excluded. Moreover,
there have been few prospective evaluations of liver
function tests in a large group of patients in whom the
diagnosis of Lyme disease was confirmed by culture of
B. burgdorferi. To determine the frequency, severity,
and response to therapy of hepatitis associated with
Lyme disease, we prospectively evaluated patients presenting with EM for symptoms of hepatitis and liver
function abnormalities.
PATIENTS AND METHODS
The study population consisted of 124 consecutive untreated patients 16 years of age with a clinical diagnosis of
EM who were enrolled in one of two ongoing treatment trials
at the Lyme Disease Diagnostic Center of the Westchester
County Medical Center between July 1990 and September
1993. All patients fulfilled the Centers for Disease Control
and Prevention surveillance criteria for EM.11 At the initial
visit, the total number, size, and duration of EM lesions were
noted. All patients were questioned about the occurrence of
anorexia, nausea, and vomiting. Patients were treated with
either an oral doxycycline-based regimen of 100 mg twice
daily for 10 to 20 days or an intravenous ceftriaxone-based
regimen of 1 g every 12 hours. Informed consent was obtained
on study participants, and the studies were approved by the
Institutional Review Board at New York Medical College.
Patients were examined on day 10 of antibiotic therapy,
on days 20 to 30 (3-week visit), and at 3, 6, and 12 months
after the initial visit. Patients had serum biochemical analysis performed including alanine transaminase (ALT), aspartate transaminase (AST), alkaline phosphatase, g-glutamyl
transpeptidase, bilirubin and creatine phosphokinase before
treatment and 3 weeks after the initiation of treatment. Serum was stored from each visit at 0707C. Patients with any
abnormal liver function study were screened for evidence of
viral hepatitis. Lyme serologies were performed at each visit.
Skin or blood cultures for B. burgdorferi were performed on
99 patients before treatment. Liver biopsies were not performed in this study.

1412

5p0e$$0013

05-21-96 22:10:10

hepas

WBS: Hepatology

HEPATOLOGY Vol. 23, No. 6, 1996

HOROWITZ ET AL.

A retrospective convenience sample population was obtained to determine the frequency of abnormal liver function
studies in a population without Lyme disease. The charts
of 100 randomly selected employee health applicants at our
hospital from 1993 who had serum biochemistry profiles performed at preemployment evaluation were reviewed. Individuals who had exposure to blood products were excluded, as
were people with a history of active hepatitis, alcohol abuse,
or illicit intravenous drug use. None of these individuals had
a known history of Lyme disease. All controls had negative
serum studies for hepatitis B surface antigen and antibodies
to hepatitis C. Information regarding gastrointestinal symptoms were not generally available for controls.
Early localized Lyme disease was defined as a single EM
lesion without objective neurological involvement, carditis,
or arthritis. Early disseminated Lyme disease was defined
by the presence of multiple EM lesions or at least one EM
lesion with objective neurological or cardiac abnormalities or
arthritis.
Serum from visits at 3 months or 6 months after presentation was tested for the presence of total antibody to hepatitis
A by enzyme immunoassay by HAVAB EIA (Abbott Laboratories, Abbott Park, IL). All patients with evidence of HAV
antibodies had serum from the initial visit tested for hepatitis
A virus immunoglobulin M by HAVAB-M (Abbott Laboratories). Serum from 3 or 6 months after presentation was
tested for hepatitis B virus core antibody by enzyme-linked
immunosorbent assay (ELISA) (Ortho Diagnostic Systems,
Inc., Raritan, NJ). For patients who tested positive for hepatitis B core antibody, serum from the initial visit was tested for
hepatitis B virus surface antigen by ELISA (Ortho Diagnostic
Systems, Inc.). Serum from 3 or 6 months after presentation
was tested for hepatitis C virus antibody by ORTHO HCV
3.0 ELISA (Ortho, Diagnostic Systems, Inc.).
Antibodies to B. burgdorferi in serum were detected by a
polyvalent ELISA using Whittaker Stat ELISA kits (Whittaker Bioproducts, Inc., Walkersville, MD) according to the
manufacturers instructions. Culture of B. burgdorferi was
performed in modified Barbour-Stoenner-Kelly media using
techniques previously described.12,13
Statistical analysis was performed using x2, Fishers exact
test, or Students t test (all two-tailed) as indicated. For those
with abnormal liver function tests, a paired t test was used
to compare the mean value at baseline with the mean value
20 days later. All values are expressed as mean { SD.
RESULTS

One hundred fifteen patients were included in the

analysis. Of the initial 124 patients, 4 patients with
antibodies to hepatitis C virus and 5 others with a
history of ongoing alcohol abuse were excluded from
analysis. Eight of these 9 patients had at least one
abnormal liver function test. No patient had acute hepatitis A virus at presentation or acute or chronic hepatitis B virus infection. The only patient concurrently
taking a potentially hepatotoxic medication (trimethoprim-sulfamethoxazole) was included in the study
group. Seven (6%) patients with at least one abnormal
liver function test had an elevation in creatine phosphokinase levels (range, 203-1,090 U/L, normal 195
U/L). These patients are included in the study group.
Patients were significantly older, more likely to be
male, and more likely to have liver function abnormali-

5p0e$$0013

05-21-96 22:10:10

hepas

1413

TABLE 1. Comparison of Patients With EM

and a Control Group
Parameter

Patients
(n 115)

Controls
(n 100)

Age (yr)
Sex (male)
Any LFT elevation
ALT
AST
ALP
GGT
Total bilirubin
1 Liver test elevation
Liver test 2 times normal

44.0 { 14.3
75 (65%)
46 (40%)
31 (27%)
18 (16%)
22 (19%)
32 (28%)
3 (3%)
31 (27%)
16 (14%)

31.8 { 12.3
45 (45%)
19 (19%)
7 (7%)
4 (4%)
8 (8%)
4 (4%)
1 (1%)
4 (4%)
2 (2%)

.001
.005
.001
.001
.006
.028
.001
.625
.001
.002

Abbreviations: LFT, liver function test; ALP, alkaline phosphatase; GGT, g-glutamyl transpeptidase.

ties than controls (Table 1). A total of 75 (65%) of the

115 patients had antibodies to B. burgdorferi detected
in serum by days 20 to 30, and 49 (50%) of 99 patients
who were cultured grew B. burgdorferi from skin or
blood. Confirmation of Lyme disease by serology or culture was found in 82% of patients.
Among Lyme disease patients, 35 (30%) had gastrointestinal symptoms manifested by anorexia, nausea,
or vomiting. Abdominal tenderness, hepatomegaly, or
jaundice was not found in any patient on presenting
physical examination. Elevations in ALT and g-glutamyl transpeptidase values were the most commonly
seen abnormalities found among patients (Table 1).
Thirty-one (67%) of 46 Lyme disease patients with abnormal liver function studies had more than one abnormal test at presentation. Only 16 of 115 patients (14%)
had elevations of one or more values that were more
than twice normal compared with 2 (2%) of controls (P
.002). Patients with elevated liver function tests
were not more likely to have anorexia, nausea, or vomiting than patients with normal liver function studies
(Table 2).
The presence of fever, longer duration of EM before
treatment, increasing number of EM lesions, and diameter of the largest EM did not correlate with elevated
liver function tests (Table 2). Patients with elevated
liver function tests were not more likely to have Lyme
disease confirmed by culture of B. burgdorferi or by
serum ELISA reactivity (Table 2). Of 49 culture-positive patients, 17 (35%) had an abnormal liver function
test; and 44% of patients with culture or serological
confirmation of Lyme disease had at least one abnormal liver enzyme study. Patients with elevated liver
function tests, however, were more likely to have disseminated compared with localized Lyme disease (Table 2). Using ALT as a single marker for hepatic injury
yielded similar results to those above, except for a positive correlation between the number of EM lesions at
presentation and the likelihood of an abnormal ALT (P
.023) (data not shown).
Of the 46 patients presenting with an abnormal liver

WBS: Hepatology

1414 HOROWITZ ET AL.

HEPATOLOGY June 1996

TABLE 2. Comparison of Patients With EM and Normal or

Abnormal Liver Function Studies

Characteristic

Age (yr)
Sex (male)
Duration EM (days)
Size largest EM (cm)
Number EM lesions
Disseminated disease
Temperature
37.77C
GI symptoms
ELISA positive
Culture positive
ELISA or culture
positive

Normal LFT
(n 69)

Abnormal
LFT
(n 46)

43.6 { 13.8*
40 (58%)
7.5 { 6.6
15.6 { 6.9
3.3 { 12
13 (19%)

44.6 { 15
35 (76%)
8.5 { 7.6
18.6 { 12.8
7.3 { 15.7
22 (48%)

.632
.072
.504
.092
.124
.002

7 (10%)
18 (26%)
47 (68%)
32/56 (57%)

10 (22%)
17 (37%)
38 (83%)
17/43 (40%)

.148
.301
.129
.103

53 (77%)

41 (89%)

.153

Abbreviations: LFT, liver function test; GI, gastrointestinal.

* { SD.
ELISA positivity cumulative through week 3.
Culture positivity of skin or blood among 99 patients cultured.

test, 44 were retested after 3 weeks. Mean AST, ALT,

g-glutamyl transpeptidase, and alkaline phosphatase
values decreased significantly after antibiotic treatment (Fig. 1). Only 5 (32%) of the initial 16 patients
with a liver function test more than twice the upper
limit of normal had elevations that remained twice normal at follow-up. Among the 31 patients with more
than two liver function abnormalities at presentation,
7 (23%) of the 30 retested continued to have more than
two abnormalities after treatment.
Five patients who had normal liver function tests
at presentation developed abnormal values, including
four with ALT elevations, two with g-glutamyl transpeptidase elevations, and one with elevation of AST, at
the 20- to 30-day visit. There was no correlation between these new abnormalities and the antibiotic used
for treatment.
Posttreatment at the day 20 to 30 visit, a total of 26
patients (24%) had an abnormal liver function study.
The proportion of patients who had a single liver function test abnormality, more than two abnormal tests,
or at least one liver function test more than twice normal was not significantly different than the control
group (Table 3).
Further follow-up in 20 of the 21 patients who had
persistent elevations of liver tests at the 20- to 30-day
visit showed that 16 still had elevations 3 months after
presentation. Six of these 16, however, had lower values than on prior tests. At 9 months, 10 of the 21 still
had an abnormality. None had an abnormality more
than twice normal, and only 3 had more than one abnormal liver function study.
DISCUSSION

Elevations of liver function tests occurred in 40% of

our study population of patients with early-stage Lyme

5p0e$$0013

05-21-96 22:10:10

hepas

disease. These elevations were mild, were usually not

associated with anorexia, nausea, or vomiting, generally resolved or improved within 3 weeks of therapy,
and were not related to duration of disease, borrelial
culture result, or ELISA antibody status. Patients with
early disseminated Lyme disease, however, were significantly more likely to have one or more liver function
abnormalities than patients with localized disease.
Liver function abnormalities were most consistent with
mild hepatocellular injury.
This is the first prospective report of sequential liver
function studies of patients with early Lyme disease
in whom other common causes for liver disease were
excluded. In addition, a large number (49 of 99 tested)
of patients were proved to have Lyme disease by use
of culture. This is also the first study of liver function
abnormalities in Lyme disease in which a local control
group was included for comparison.
Although the control group was significantly younger
than patients, the small age differences between these
two groups is unlikely to have resulted in the observed
differences in liver function tests.14 The sex differences
between controls and patients would also not explain
the differences in liver function study abnormalities
between them, because sex (among nonpregnant
adults) is not a determinant of liver function test results among adults.14
Our data may, if anything, overestimate the frequency of Lyme-associated liver disease because of the
inclusion of seven (6%) patients who had elevations in
creatine phosphokinase. Some of the abnormalities of
AST and ALT in these patients may represent myositis,15 which also occurs in Lyme disease.16 If these patients are not included, then 37% of patients had liver
function abnormalities. Moreover, among controls, 19%
had abnormal liver studies. Depending on the linkage
of analytes, using two standards of deviation above the
mean as the upper limits of normal, 5% to 23% of controls (and patients) would be expected randomly to
have an abnormal study at any point in testing. At
least some of the abnormal liver function studies in the
patients are probably due to this statistical probability.
This study expands on previous observations of hepatic dysfunction in Lyme disease.2,7-10 Steere et al. reported that 10% of 314 patients with EM had symptoms
suggestive of hepatitis.2 AST and ALT elevations occurred in 19% and 15% of patients, respectively, in that
study.2 The authors noted that liver function elevations
generally returned to normal within several weeks, but
did not include the supportive data. We did not find
evidence of hepatomegaly, as was found in 5% of
Steeres cohort.
In two reports of antibiotic treatment trials of patients with early Lyme disease, liver function abnormalities were found in 28% and 50% of patients on
presentation.7,8 Liver function elevations were considered mild in both studies and returned to normal in
approximately 50% of patients between 2 and 4 weeks
after treatment.7,8 Melski et al. reported significant

WBS: Hepatology

HEPATOLOGY Vol. 23, No. 6, 1996

HOROWITZ ET AL.

1415

FIG. 1. Prospective evaluation of liver function tests in Lyme disease patients with EM at presentation and at 20 days after initiation
of treatment. j, mean values with 95% confidence bars. Dark horizontal bar represents the upper limit of normal for the liver function
test.

liver function abnormalities in 3 (20%) of 15 patients

who had EM, but noted that other patients had mild
abnormalities.9 Kazakoff et al. reported abnormal liver
function tests in 20 (27%) of 83 patients with EM who
were seen between 1989 and 1992.10 Few of these patients had confirmation of Lyme disease by serology

TABLE 3. Comparison of Liver Function Studies of Patients

With EM Posttreatment (3 weeks) Versus Controls
Parameter

Patients
(n 108)

Controls
(n 100)

Any LFT elevation

1 LFT elevation
Any LFT 2 times normal

26 (24%)
9 (8%)
2 (2%)

19 (19%)
4 (4%)
2 (2%)

.472
.256
1.0

Abbreviation: LFT, liver function test.

5p0e$$0013

05-21-96 22:10:10

hepas

(only 9.6%), and follow-up evaluations were not performed.

We are aware of only four cases in the English language literature of acute hepatitis presumed to be due
to Lyme disease in which other causes of acute hepatitis were excluded.3-6 In three of the four cases, improvement of liver test elevations occurred concomitantly
with antibiotic treatment.3-5 In the fourth case, the patient developed adult respiratory distress syndrome
and died.6 In none of these cases was there a positive
culture for B. burgdorferi from any site.
Hepatic involvement by B. burgdorferi is not suprising given the frequency of liver function abnormalities
in other spirochetal diseases, including infection
caused by Treponema pallidum (10%-50% with secondary syphilis),17,18 Leptospira sp. (over 40%),19,20 and
Borrelia recurrentis (over 30%).21 The relative lack of

WBS: Hepatology

1416 HOROWITZ ET AL.

HEPATOLOGY June 1996

severity of the hepatic involvement in Lyme disease

most closely resembles the hepatic involvement by T.
pallidum in secondary syphilis. However, the disproportionate increase of alkaline phosphatase without evidence for cholestasis found in patients with secondary
syphilis was not evident in our patients with early
Lyme disease.
The cause of the abnormal liver function tests in patients with EM is not known. Direct invasion of the
liver by B. burgdorferi and systemic or local host-mediated immune responses may be responsible individually or in combination for hepatocyte injury.
Early hepatic invasion by B. burgdorferi occurs in
several animal models.22,23 In Syrian hamsters and
Lewis rats, B. burgdorferi can be isolated from the liver
by 7 days and 2 weeks after inoculation, respectively.22,23 We are aware of only one case of Lyme diseaseassociated hepatitis in a human in which the organism was demonstrated histologically by Dieterle
stain.24 At the time of liver biopsy, the patient had not
received antibiotics. The patient clinically improved
with doxycycline therapy but did not have a repeat
liver biopsy.24 In two other cases of human Lyme disease, needle biopsy specimens of liver did not demonstrate Borrelia on histology.25
Although borrelial invasion may cause direct hepatocyte damage, the acute viral-like illness associated
with early B. burgdorferi dissemination raises the possibility that systemic cytokine release may be a factor
in hepatic enzyme elevations.26 Cytokines induced by
bacterial products such as lipopolysaccharide have
been demonstrated to cause hepatic dysfunction.27
Nonprotein fatty acid and carbohydrate antigenic moities, which may have lipopolysaccharide-like activity,
have been isolated from the B. burgdorferi cell wall;28,29
and B. burgdorferi has been reported to induce interleukin-1 production by human macrophages in in
vitro cell cultures.30
Immunologically mediated hepatic disease could also
contribute to the liver function abnormalities found in
early Lyme disease. Aberer et al. have demonstrated
that an antigenic component of the B. burgdorferi flagella protein is shared by several human tissues, including hepatocytes.31 Such molecular mimicry could lead to
local immune responses. Autoreactive T-cell lines have
been found against myelin basic protein in the cerebrospinal fluid of patients with neuroborreliosis.32
The natural history of B. burgdorferi liver involvement is not known in untreated humans. In one instance the possibility of late liver function abnormalities due to Lyme disease was raised.24 However,
reinfection with B. burgdorferi could not be distinguished from recurrence of disease in that case. Experimental animal models of Lyme disease using immunocompetent animals have demonstrated resolution of B.
burgdorferi culture positivity of liver by day 21 in Syrian hamsters22 and day 60 in Lewis rats,23 even without
treatment. However, in a severe combined immunodeficiency mouse model of Lyme disease, mononuclear cell

5p0e$$0013

05-21-96 22:10:10

hepas

infiltrates, granulomatous changes, and fibrosis of the

liver were demonstrated histologically in untreated
mice until at least 195 days after infection.33 If human
infection with B. burgdorferi parallels infection of nonimmunocompromised animals, direct Borrelia infection
of the liver occurs commonly, and is usually self-limited
without permanent sequelae.
The outcome of Lyme-associated liver abnormalities
appears to be excellent, at least when antibiotics are
used. However, 16 (15%) patients in the current study
still had mild liver function test elevations up to 3
months after treatment. These abnormalities may not
be related to Lyme disease, because 19% of controls
also had hepatic abnormalities, and most Lyme patients had no other sequelae of late Lyme disease.
Without confirmation of histopathological changes or
demonstration of B. burgdorferi in liver tissue, the interpretation of these results is open to question. Moreover, what effect antibiotic treatment had on the favorable outcome cannot be assessed.
In conclusion, although mild liver function test elevations occur commonly in patients with EM, they usually
do not result in clinically symptomatic liver disease and
typically improve or resolve by 3 weeks after antibiotic
therapy.
Acknowledgment: We thank Eleanor Bramesco for
assistance in manuscript preparation, Daniel Byrne for
statistical assistance, Susan Bittker and Denise Cooper
for their excellent laboratory assistance, and Drs. Arthur Karmen and Indrage Argani for helpful advice.
REFERENCES
1. Steere AC. Lyme disease. N Engl J Med 1989;321:586-96.
2. Steere AC, Bartenhagen NH, Craft JE, Gordon JH, Newman
JH, Rahn DW, Sigal LH, et al. The early clinical manifestations
of Lyme disease. Ann Intern Med 1983;99:76-82.
3. Chavanet P, Pillon D, Lancon JP, Waldner-Combernoux A, Maringe E, Portier H. Granulomatous hepatitis associated with
Lyme disease. Lancet 1987;ii:623-624.
4. Edwards KS, Kanengiser S, Li KI. Lyme disease presenting as
hepatitis and jaundice in a child. Pediatr Infect Dis J 1990;9:
592-593.
5. Sinusas K, Kazakoff MA, Macchia C. Hepatitis associated with
Lyme disease. JABFP 1992;5:635-637.
6. Kirsch M, Ruben FL, Steere AC, Duray PH, Norden CW, Winkelstein A. Fatal adult respiratory distress syndrome in a patient
with Lyme disease. JAMA 1988;259:2737-2739.
7. Massarotti EM, Luger SW, Rahn DW, Messner RP, Wong JB,
Johnson RC, Steere AC. Treatment of early Lyme disease. Am
J Med 1992;92:396-403.
8. Strle F, Preac-Mursic V, Cimperman J, Ruzic E, Maraspin V,
Jereb M. Azithromycin versus doxycycline for treatment of erythema migrans: clinical and microbiological findings. Infection
1993;21:83-88.
9. Melski JW, Reed KD, Mitchell PD, Barth GD. Primary and secondary erythema migrans in Central Wisconsin. Arch Dermatol
1993;129:709-716.
10. Kazakoff MA, Sinusas K, Macchia C. Liver function test abnormalities in early Lyme disease. Arch Fam Med 1993;2:409-413.
11. CDC case definitions for public health surveillance. Lyme disease. MMWR Morb Mortal Wkly Rep 1990;39(suppl RR-13):1921.
12. Wormser GP, Forseter G, Cooper D, Nowakowski J, Nadelman
RB, Horowitz H, Schwartz I, et al. Use of a novel technique of

WBS: Hepatology

HEPATOLOGY Vol. 23, No. 6, 1996

13.

14.
15.
16.

17.
18.
19.
20.
21.
22.

23.

HOROWITZ ET AL.

cutaneous lavage for diagnosis of Lyme disease associated with

erythema migrans. JAMA 1992;268:1311-1313.
Nadelman RB, Nowakowski J, Forseter G, Bittker S, Cooper D,
Goldberg N, McKenna D, Wormser GP. Failure to isolate Borrelia burgdorferi after antimicrobial therapy in culture-documented borreliosis associated with erythema migrans: report of
a prospective study. Am J Med 1993;94:583-588.
Kaplan MM. Laboratory tests. In: Schiff L and Schiff ER, eds.
Diseases of the liver. Philadelphia: Lippincott, 1993:108-144.
Helfgott SM, Karlson E, Beckman E. Misinterpretation of serum
transaminase elevation in occult myositis. Am J Med 1993;95:
447-449.
Horowitz HW, Sanghera K, Goldberg N, Pechman D, Kamer R,
Duray P, Weinstein A. Dermatomyositis associated with Lyme
disease; Case report and review of Lyme myositis. Clin Infect
Dis 1994;18:166-171.
Feher J, Somogyi T, Timmer M, Jozsa L. Early syphilitic hepatitis. Lancet 1975;ii:896-899.
Pareek SS, Ven D. Liver involvement in secondary syphilis. Dig
Dis Sci 1979;24:41-43.
Berman SJ, Tsai CC, Holmes K, Fresh JW, Watten RH. Sporadic
anicteric leptospirosis in South Vietnam. Ann Intern Med 1973;
79:167-173.
Lecour H, Miranda M, Magro C, Rocha A, Goncalves V. Human
leptospirosis: a review of 50 cases. Infection 1989;17:8-12.
Southern PM Jr, Sanford JP. Relapsing fever: a clinical and
microbiological review. Medicine 1969;48:129-149.
Goodman JL, Jurlovich P, Kodner C, Johnson RC. Persistent
cardiac and urinary tract infections with Borrelia burgdorferi in
experimentally infected Syrian hamsters. J Clin Microbiol 1991;
29:894-896.
Moody KD, Barthold SW, Terwilliger GA, Beck DS, Hansen GM,
Jacoby RO. Experimental chronic Lyme borreliosis in Lewis rats.
Am J Trop Med Hyg 1990;42:165-174.

5p0e$$0013

05-21-96 22:10:10

hepas

1417

24. Goellner MH, Agger WA, Burgess JH, Duray PH. Hepatitis
due to recurrent Lyme disease. Ann Intern Med 1988;108:707708.
25. Duray PH. The surgical pathology of human Lyme disease. Am
J Surg Pathol 1987;11(suppl 1):47-60.
26. Sikuler E, Guetta V, Keynan A, Neumann L, Schlaeffer F. Abnormalities in bilirubin and liver enzyme levels in adult patients
with bacteremia. Arch Intern Med 1989;149:2246-2248.
27. Shedlofsky SI, McClain CJ. Hepatic dysfunction due to cytokines. In: Kimball ES, ed. Cytokines and inflammation. Boca
Raton: CRC Press, 1991:235-261.
28. Cinco M, Banfi E, Balanzin D, Godeas C, Panfili E. Evidence for
(lipo) oligosaccharides in Borrelia burgdorferi and their serological specificity. FEMS Microbiol Immunol 1991;76:33-38.
29. Eiffert H, Lotter H, Jarecki-Khan K, Thomssen R. Identification
of an immunoreactive non-proteinaleous component in Borrelia
burgdorferi. Med Microbiol Immunol 1991;180:229-237.
30. Habicht GS, Beck G, Benach JL, Coleman JL, Leichtling KD.
Lyme disease spirochetes induce human and murine interleukin
1 production. J Immunol 1985;134:3147-3154.
31. Aberer E, Brunner C, Suchanek G, Klade H, Barbour A, Stanek
G, Lassmann H. Molecular mimicry and Lyme borreliosis: a
shared antigenic determinant between Borrelia burgdorferi and
human tissue. Ann Neurol 1989;26:732-737.
32. Martin R, Ortlauf J, Sticht-Groh V, Bogdahn U, Goldmann SF,
Mertens HG. Borrelia burgdorferi-specific and autoreactive Tcell lines from cerebrospinal fluid in Lyme radiculomyelitis. Ann
Neurol 1988;24:509-516.
33. Schaible UE, Gay S, Museteanu C, Kramer MD, Zimmer G, Eichmann K, Museteanu U, et al. Lyme borreliosis in the severe
combined immunodeficiency (scid) mouse manifests predominantly in the joints, heart, and liver. Am J Pathol 1990;137:811820.

WBS: Hepatology