951
Granulomatous Hepatitis and Necrotizing Splenitis Due to Bartonella henselae in
a Patient with Cancer: Case Report and Review of Hepatosplenic Manifestations
of Bartonella Infection
Thomas E. Liston and Jane E. Koehler
From the Division of Infectious Disease, Department of Pediatrics,
Kaiser Permanente Medical Center, Martinez, California; and the
Division of Infectious Disease, Department of Medicine,
University of California, San Francisco, California
Bartonella (formerly Rochalimaea [1]) infection often appears as skin lesions known as bacillary angiomatosis (BA)
but may affect the liver (bacillary peliosis hepatis; BP) and
spleen (bacillary splenitis) in both immunocompromised and
immunocompetent hosts [2]. Two bacterial agents have been
cultured directly from specimens of BA lesions: Bartonella
henselae and Bartonella quintana [3]. The occurrence of bartonella infection during chemotherapy for cancer was first reported in 1992; in that pediatric case and in two subsequent
cases involving adults, the vascular proliferative manifestation
of bartonella infection was described [4-6]. This report details
an adult patient receiving chemotherapy for cancer in whom
granulomatous visceral disease due to B. henselae developed;
we also review hepatosplenic manifestations of bartonella infection.
Case Report
A 54-year-old woman was admitted to the surgical service
of Kaiser Permanente Medical Center (Martinez, CA) in September 1992 after an annual mammogram showed a mass
(10 X 4 mm) in the left breast. Excisional biopsy revealed a
moderately well-differentiated infiltrating ductal carcinoma.
The patient received three courses of chemotherapy with cy-
Received 6 September 1995; revised 15 December 1995.
Financial support: Dr. Koehler received grant support from the NIH
(R29AI36075) and the University wide AIDS Research Program, University
of California. Dr. Koehler is a Pew Scholar in the Biomedical Sciences.
Reprints or correspondence: Dr. Thomas E. Liston, Department of Pediatrics,
Kaiser Permanente Medical Center, 200 Muir Road, Martinez, California
94553-4696.
Clinical Infectious Diseases
1996;22:951-7
© 1996 by The University of Chicago. All rights reserved.
1058-4838/96/2206-0007$02.00
toxan, adriamycin, and 5-fluorouracil sequentially with radiation therapy for 6 weeks and then a second cycle of three
planned courses of chemotherapy; each course of chemotherapy was 28 days.
The patient was again admitted to the hospital in March
1993 because of fever that developed 4 weeks after the fifth
course of chemotherapy. Her temperature rose to 105.1 OF
(40.6°C) and remained at this level for several days. She had
never received a blood transfusion and did not consent to HIV
antibody testing.
The patient was treated empirically with ceftriaxone and
gentamicin. Skin tests for tuberculosis and coccidioidomycosis
were negative, as were five blood cultures. The average complete leukocyte count was 5,000/mm3 (5 X 109/1), with 70%80% polymorphonuclear cells. A gallium scan 10 days after
admission showed increased uptake in the spleen. A CT. scan
done on day 14 showed multiple hypodense lesions in the
spleen and a few similar lesions in the liver (figure 1). A bone
marrow biopsy showed only hypocellularity.
The patient had been raised in Louisiana and recalled drinking fresh milk from cows there. She had traveled extensively
throughout the world but had not traveled recently and had
never visited South America or the Andes Mountains. She
disliked cats and denied any recent contact with a cat or dog
but visited people with pets and could have been bitten by
fleas.
Disseminated Mycobacterium tuberculosis infection was
considered the most likely diagnosis, and brucellosis was considered less likely. The antibiotics prescribed earlier were withdrawn; the patient was treated with isoniazid for 13 days and
pyrazinamide for 5 days, and administration of rifampin was
continued. Splenectomy was recommended for definitive diagnosis. At laparotomy no lesions were noted on the surface of
the liver, but a biopsy was performed, along with biopsy of a
lymph node. The spleen was removed.
Downloaded from http://cid.oxfordjournals.org/ by guest on September 9, 2014
Bacillary angiomatosis and the related disorders of bacillary peliosis hepatis and bacillary splenitis
are manifestations of infection with Bartonella henselae and Bartonella quintana in immunocompromised persons. B. henselae infection, but not B. quintana infection, is linked to contact with cats
and is presumed to cause visceral cat-scratch disease. We report a case of visceral infection by B.
henselae in an adult patient with cancer who was receiving chemotherapy and had had no contact
with a cat or dog. The patient-whose illness was eventually diagnosed on the basis of findings of
histologic, polymerase chain reaction, and serological studies-was treated with doxycycline and
rifampin, and the infection resolved. In addition, 41 cases of documented or suspected bartonella
infection of the liver, spleen, or both in immunocompetent or immunocompromised hosts are reviewed.
952
Liston and Koehler
em 1996;22 (June)
Figure 1. An abdominal CT
scan with oral and intravenous
contrast showing multiple small,
hypodense, round lesions in the
spleen (arrows) and liver of a patient with cancer.
Figure 2. A photograph of the gross anatomy of the spleen shows
heterogeneously sized white nodules.
infection [7], the diagnosis of bartonella infection was considered. However, Warthin-Starry silver stains performed at the
pathology department at Kaiser Permanente Medical Center
and the department of dermatology at the University of California (San Francisco) did not initially reveal the presence of
bacteria.
Microbiology. A special blood culture specimen was
drawn with use of the Isostat Isolator tube (Wampole Laboratories, Cranbury, NJ). Cultures were negative, although antibiotic treatment had been discontinued several days before the
culture specimens were drawn. Cultures of the spleen were
negative for fungi, Brucella species, and acid-fast bacilli.
Serology and peR. A CF test for Brucella was negative
(< 1:4). Although the initial test for B. henselae was negative,
a second specimen drawn 99 days later revealed IgO antibodies
to B. henselae (> 120 EIA units; normal reference value, < 12)
as well as IgM antibodies (79 EIA units), as determined at
Specialty Laboratories (Santa Monica, CA). Serology samples
were collected and sent to the Centers for Disease Control
and Prevention (CDC). Immunofluorescent antibody assays [8]
(performed at the CDC's Rickettsial Disease Laboratory) of
samples collected on day 21 showed antibodies to B. henselae
(1 :512) and B. quintana (> 1: 1,024); assays of sera collected
during the patient's convalescence (day 35) showed antibody
titers of > I: I,024 for B. henselae and > I: I,024 for
B. quintana.
Ultimately, Warthin-Starry staining of splenic tissue at the
Armed Forces Institute of Pathology (Washington, D.C.) revealed
a "few clumps ofsilvered bacilli in one granuloma. . . characteristic of cat scratch disease" (figure 3). Splenic tissue obtained at
Downloaded from http://cid.oxfordjournals.org/ by guest on September 9, 2014
Pathology. The spleen surface was covered with grayishwhite nodules that were 2-5 mm in size (figure 2). Histopathologic examination showed numerous necrotizing and caseating
granulomas, which stained negative for acid-fast bacilli and
did not produce a periodic acid-Schiff reaction; examination
of a nearby abdominal lymph node and one of several small
nodules obtained by liver biopsy revealed the same histopathologic characteristics.
Because similar nodules had been found in the spleen of a
patient with AIDS-related complex and presumed bartonella
em
1996;22 (June)
Hepatosplenic Bartonella Infections
953
Figure 3. A Warthin-Starry silver stain of the spleen showing
darkly staining bacilli (arrows)
consistent with Bartonella (original magnification, X 400).
Discussion
Human bartonella infections have only recently been reported in the United States and were initially described as
vascular proliferative lesions (now known as BA) in HIVinfected patients [10). Other syndromes reported in the past
several years that have been caused by Bartonella species include relapsing fever and bacteremia (in both immunocompromised and immunocompetent hosts [11, 12]) and endocarditis [13, 14). Recently, an infection first described many years
ago, cat-scratch disease (CSD), has been attributed to
B. henselae. This disease is usually self-limiting, develops as
a complication of a cat scratch, and manifests as a local, necrotizing lymphadenitis in the immunocompetent host [15, 16].
B. henselae has been cultured directly from specimens of
vascular proliferative BA lesions [3] and necrotizing granulomatous CSD lymph nodes [16]. Thus, the histopathologic manifestations of infection with B. henselae can be strikingly different and are apparently related to the degree of
immunocompromise in the host. Some patients whose immune
systems are not severely compromised appear to have disseminated granulomatous lesions, constitutional symptoms, and a
longer and more severe disease course than that of CSD. Our
patient had disseminated B. henselae disease, but whether the
severity of her illness was related to the chemotherapy for
breast cancer is unknown.
Bartonella species-associated lesions ofthe liver and spleen
can be categorized as one of two distinct types on the basis
of histopathologic findings: vascular proliferative lesions or
necrotizing granulomatous lesions. Vascular proliferative disease, or BP, differs histopathologically from BA and is characterized by multiple, cystic, blood-filled spaces in addition to
foci of necrosis [17]. A fibromyxoid stroma is often present and
contains clumps of Bartonella bacilli and a mixed inflammatory
infiltrate [17]. As with BA lesions, BP lesions are more common in HIV-infected patients.
Downloaded from http://cid.oxfordjournals.org/ by guest on September 9, 2014
splenectomy was sent to the CDC, where it was evaluated further.
Bacterial DNA was extracted from the splenic tissue; the template
DNA was amplified and then hybridized by dot blotting with use
of primers and probes, as previously described by Anderson et
al. (9). The probes are specific for differentiation of B. quintana
from B. henselae, and the species infecting this patient was determined to be B. henselae.
Treatment and outcome. Treatment with erythromycin for
suspected bartonella infection was begun on day 22. The patient
was unable to tolerate the drug because of gastrointestinal disturbance, and it was replaced with doxycycline (100 mg twice
daily) and rifampin (600 mg daily) on day 27. The patient
became afebrile the next day and was discharged 30 days after
admission. Cultures for M. tuberculosis were negative.
The patient continued to do well after completing 4 months
of treatment with oral doxycycline (100 mg b.i.d.) and rifampin
(600 mg daily). A follow-up CT scan done on day 54, ~ 1
month after splenectomy, showed resolution ofthe liver lesions.
Eighteen months following completion of chemotherapy for
breast cancer, no sign of recurrence of cancer was evident. The
patient declined a sixth course of chemotherapy, and there was
no evidence of bartonella infection during nearly 3 years of
follow-up.
954
Liston and Koehler
1996;22 (June)
hybridization probes, permitted rapid differentiation of the two
species.
The source of bartonella infections has been intensively investigated recently. A large case-control study [24] of 48 patients (5 immunocompetent patients, 1 transplant recipient, and
42 HIV-infected patients) revealed that BA and BP were significantly linked to owning a cat or being bitten or scratched
by a cat. Another study [25] showed a statistically significant
association between CSD and owning a kitten or being
scratched by a kitten. Further study [26] identified the domestic
cat as a major reservoir of B. henselae infection: all seven cats
of four patients with BA due to B. henselae had B. henselae
bacteremia.
In addition, 41 % of pet cats and cats at animal shelters in
the San Francisco Bay area whose blood specimens were cultured had B. henselae bacteremia. Viable B. henselae was isolated from fleas combed from one of the cats with bacteremia
[26], which indicates that fleas might serve as a vector of
B. henselae. Our patient had B. henselae infection yet disliked
cats and denied contact with them. Thus, the source of her
infection remains unknown.
Bartonella infections in patients with BA or BP respond
dramatically to treatment with erythromycin or doxycycline
[2]. Treatment of bartonella infection in immunocompetent
people with CSD and local lymphadenitis has not been studied
systematically, and because this infection is self-limiting, treatment may not be warranted for such patients. However, treatment is appropriate for patients with severe, disseminated,
granulomatous disease, and for adults it should include administration of erythromycin or doxycycline. Gentamicin or rifampin has also been used successfully to treat both BA and CSD
[2, 15], but neither is recommended as single-agent therapy for
seriously ill patients [2].
Our patient responded rapidly to treatment with doxycycline
and rifampin, and the hepatic lesions and systemic symptoms
resolved completely. For immunocompromised patients with
disseminated disease, including BP, treatment should last at
least 3 months; an adequate duration of treatment for immunocompetent patients is unknown, but treatment probably should
last at least I month.
Review of Previous Cases
Twenty-two articles that described patients with presumed hepatic or splenic disease due to bartonella infection were reviewed
[7, 17, 23, 27-45]. These reports contained sufficient clinical data
to confirm the diagnosis for 41 patients on the basis of at least
three ofthe following 10 criteria: consistent radiographic evidence
on CT scan films, ultrasonographic scan films, or both; compatible
gross pathology of multiple nodules; compatible histologic findings of granulomas or BP, with or without necrosis; positive
Warthin-Starry silver stains; positive PCR for Bartonella species;
successful culture of B. henselae; serological increase in titers
of antibody to Bartonella species; electron microscopic findings
compatible with Bartonella species; positive CSD skin test results;
Downloaded from http://cid.oxfordjournals.org/ by guest on September 9, 2014
The second type of hepatic pathological abnormality associated with bartonella infection resembles that found in the lymph
nodes of patients with CSD: necrotizing granulomas with leukocytic infiltration, palisading histiocytes, giant cells, and
sometimes stellate abscesses [18]. These histopathologic features are seen most commonly in immunocompetent hosts.
The hepatic and splenic lesions of both histopathologic types
are often multiple and can appear as hypoechoic or low-attenuation lesions scattered throughout the hepatic or splenic parenchyma on ultrasonographic or CT scans [19]. However, numerous other infectious and noninfectious conditions in either
immunocompetent or immunocompromised individuals may
appear identical, including abscesses due to fungi, protozoa,
and bacteria other than Bartonella species. In addition, the CT
appearance ofcancers such as lymphoma and Kaposi's sarcoma
can be indistinguishable from BP in HIV-infected patients. The
diagnosis of bartonella infection must be established by one or
more means: histopathologic examination of biopsy specimens,
serological evaluation, microbiological culture, and/or use of
PCR to detect Bartonella DNA in clinical specimens.
The most accessible diagnostic technique for a patient with
multiple hypodense lesions of the liver is biopsy. The risk of
hemorrhage from percutaneous liver biopsy for a patient with
BP remains theoretical, but viewing the liver directly during
biopsy may reduce this risk. The biopsied tissue should be
examined with hematoxylin-eosin staining, which demonstrates
either granulomatous hepatitis or peliosis hepatis. WarthinStarry staining reveals silver-staining bacilli in most cases of
BP because the number of bacteria present in lesions is much
greater in immunocompromised patients. However, in granulomatous hepatic lesions the bacilli are much more difficult to
find, and in our case the difficulty in visualizing organisms
may have been related to administration of antibiotic agents in
the 2 weeks before splenectomy.
Serological tests based on immunofluorescence [8] or EIA
[20] are now available. Testing of paired sera confirmed our
patient's bartonella infection, but the Bartonella species that
caused the infection could not be determined serologically;
in fact, the titer of antibodies to B. quintana was higher than
that to B. henselae, even though PCR showed B. henselae
infection.
Culture of Bartonella organisms from tissue or blood is
difficult but can be done with use of chocolate agar or heart
infusion agar with 5% rabbit blood [3, 11]. Isolates can be
identified by means of the Microscan Rapid Anaerobe panel
(Baxter Diagnostics, Deerfield, IL), which will distinguish between B. henselae and B. quintana [21].
Definitive identification of Bartonella species requires molecular characterization with citrate synthase PCR- restriction
fragment length polymorphism analysis of isolates [22] or 16S
rRNA gene amplification and sequencing [3, 23]. In a study
described by Anderson et al. [9], a PCR assay (utilized on our
patient's tissue) using degenerate primers to allow amplification of a 414-bp fragment of DNA from B. henselae and
B. quintana was developed. Internal oligonucleotides, used as
em
Downloaded from http://cid.oxfordjournals.org/ by guest on September 9, 2014
956
Liston and Koehler
Summary
This report summarizes the case of a cancer patient with a
severe febrile illness and visceral B. henselae disease and reviews the literature on hepatosplenic disease related to presumed bartonella infection. For our patient, the histopathologic
appearance was that of necrotizing granulomatous disease of
the spleen without angiomatous changes. Bartonella infection
must be considered as a cause of fever in patients with cancer
who are receiving chemotherapy as well as in patients who
are immunocompromised because of HIV infection or organ
transplantation.
Acknowledgments
At the Viral and Rickettsial Zoonoses Branch of the CDC, Mary
Jane Schmidt, M.D., arranged for serological and PCR testing of
spleen tissue obtained through biopsy; Judy Cooper, P.A., performed the serological testing; and Burt Anderson, Ph.D., performed the PCR studies. Investigators at the Armed Forces Institute
of Pathology (Washington, DC) performed the Warthin-Starry silver staining, and the Medical Editing Department of Kaiser Foundation Research Institute (Oakland, CA) provided editorial assistance.
References
1. Brenner DJ, O'Connor SP, Winkler HR, Steigervvalt AG. Proposals to
unify the genera Bartonella and Rochalimaea, with descriptions of Bartonella quintana comb. nov., Bartonella vinsonii comb. nov., Bartonella
henselae comb. nov., and Bartonella elizabethaecomb. nov.; and to
remove the family Bartonellaceae from the order Rickettsiales. Int J
Syst BacterioI1993;43:777-86.
1996;22 (June)
2. Koehler JE, Tappero JW. Bacillary angiomatosis and bacillary peliosis in
patients infected with human immunodeficiency virus. Clin Infect Dis
1993; 17:612-24.
3. Koehler JE, Quinn FD, Berger TG, LeBoit PE, Tappero JW. Isolation of
Rochalimaea species from cutaneous and osseous lesions of bacillary
angiomatosis. N Engl J Med 1992;326:1625-31.
4. Myers SA, Prose NS, Garcia JA, Wilson KH, Dunsmore KP, Kamino H.
Bacillary angiomatosis in a child undergoing chemotherapy. J Pediatr
1992; 121:574-8.
5. Torok L, Viragh SZ, Borka I, Tapai M. Bacillary angiomatosis in a patient
with lymphocytic leukaemia. Br J DermatoI1994; 130:665-8.
6. Mulvany NJ, Billson VR. Bacillary angiomatosis of the spleen. Pathology
1993;25:398-401.
7. Delahoussaye PM, Osborne BM. Cat-scratch disease presenting as abdominal visceral granulomas. J Infect Dis 1990; 161 :71-8.
8. Regnery RL, Olson JG, Perkins BA, Bibb W. Serological response to
"Rochalimaea henselae" antigen in suspected cat-scratch disease. Lancet 1992;339:1443-5.
9. Anderson B, Sims K, Regnery R, et a!. Detection of Rochalimaea henselae
DNA in specimens from cat scratch disease patients by PCR. J Clin
Microbiol 1994; 32:942-8.
10. Stoler MH, Bonfiglio TA, Steigbigel RT, Pereira M. An atypical subcutaneous infection associated with acquired immune deficiency syndrome.
Am J Clin PathoI1983;80:714-8.
11. Slater LN, Welch DF, Hensel D, Coody DW. A newly recognized fastidious gram-negative pathogen as a cause of fever and bacteremia. N Engl
J Med 1990;323:1587-93.
12. Lucey D, Dolan MJ, Moss CW, et a!. Relapsing illness due to Rochalimaea
henselae in immunocompetent hosts: implication for therapy and new
epidemiological associations. Clin Infect Dis 1992; 14:683-8.
13. Spach DH, Callis KP, Paauw DS, et al. Endocarditis caused by Rochalimaea quintana in a patient infected with human immunodeficiency
virus. J Clin Microbiol 1993; 31 :692-4.
.14. Daly JS, Worthington MG, Brenner DJ, et al. Rochalimaea elizabethae
sp. nov. isolated from a patient with endocarditis. J Clin Microbio1
1993; 31 :872-81.
15. Margileth AM. Antibiotic therapy for cat-scratch disease: clinical study
of therapeutic outcome in 268 patients and a review of the literature.
Pediatr Infect Dis J 1992; 11 :474-8.
16. Dolan MJ, Wong MT, Regnery RL, et a1. Syndrome of Rochalimaea
henselae adenitis suggesting cat-scratch disease. Ann Intern Med 1993;
118:331-6.
17. Perkocha LA, Geaghan SM, Yen TSB, et a1. Clinical and pathological
features ofbacillary peliosis hepatis in association with human immunodeficiency virus infection. N Eng1 J Med 1990;323:1581-6.
18. Schwartzman WA. Infections due to Rochalimaea: the expanding clinical
spectrum. Clin Infect Dis 1992; 15:893-902.
19. Wyatt SH, Fishman EK. Hepatic bacillary angiomatosis in a patient with
AIDS. Abdom Imaging 1993; 18:336-8.
20. Barka NE, Hadfield T, Patnaik M, Schwartzman WA, Peter JB. EIA for
detection of Rochalimaea henselae-reactive IgG, IgM, and IgA antibodies in patients with suspected cat-scratch disease [letter]. J Infect
Dis 1993; 167:1503-4.
21. Welch DF, Hensel DM, Pickett DA, San Joaquin VH, Robinson A, Slater
LN. Bacteremia due to Rochalimaea henselae in a child: practical identification of isolates in the clinical laboratory. J Clin MicrobioI1993;31:
2381-6.
22. Regnery RL, Anderson BE, Clarridge JE III, Rodriguez-Barradas MC,
Jones DC, Carr JR. Characterization of a novel Rochalimaea species,
R. henselae sp. nov., isolated from blood of a febrile, human immunodeficiency virus-positive patient. J Clin Microbiol 1992:30:265-74.
23. Reiman DA, Loutit JS, Schmidt TM, Falkow S, Tompkins LS. The agent
ofbacillary angiomatosis: an approach to the identification of uncultured
pathogens. N Engl J Med 1990;323:1573-80.
24. Tappero JW, Mohle-Boetani J, Koehler JE, et a1. The epidemiology of
bacillary angiomatosis and bacillary peliosis. JAMA.·1993;269:770-5.
Downloaded from http://cid.oxfordjournals.org/ by guest on September 9, 2014
(25 cases). In addition, results were abnormal for liver-spleen
scanning and CT scanning in 1 case, for a gallium study and
ultrasonographic studies in another case, and for MRI and CT
in a third case.
Including our case, gross pathological findings in 21 cases of
granulomatous bartonella infection included the typical small,
gray-white nodules or larger nodules. In cases of the peliotic
form of infection, numerous cystic blood-filled spaces were
evident. Histologic findings included granulomas (occasionally
caseating) in 21 cases, 15 of which involved immunocompetent
hosts; BP (vascular proliferative abnormalities) in 12 cases,
all of which involved immunocompromised hosts (two had
granulomas and BP); and inflammation, necrotizing lesions, or
abscess in 6 cases. The affected organs were the liver and
spleen in 22 cases, the liver only in 16, and the spleen only
in 4.
Warthin-Starry silver stains were positive in 27 cases, and
electron microscopy was positive in 13. PCR was positive in
six cases, and cultures were positive for Bartonella species in
one case. Serological studies were positive for Bartonella species in six cases. CSD skin tests were positive in 13 cases, but
exposure to kittens or cats was reported in only 25 cases.
em
cm
1996;22 (June)
Hepatosplenic Bartonella Infections
36. Golden SE. Hepatosplenic cat-scratch disease associated with elevated
anti-Rochalimaea antibody titers. Pediatr Infect Dis J 1993; 12:
868-71.
37. Slater LN, PithaJV, Herrera L, HughsonMD, MinK-W, ReedJA. Rochalimaea henselae infection in acquired immunodeficiency syndrome causing inflammatory disease without angiomatosis or peliosis: demonstration by immunocytochemistry and corroboration by DNA amplification.
Arch Pathol Lab Med 1994; 118:33-8.
38. Malatack JJ, Jaffe R. Granulomatous hepatitis in three children due to catscratch disease without peripheral adenopathy: an unrecognized cause
of fever of unknown origin. Am J Dis Child 1993; 147:949-53.
39. Doyle D, Eppes SC, Klein JD. Atypical cat-scratch disease: diagnosis by
a serologic test for Rochalimaea species. South Med J 1994; 87:
485-7.
40. Rappaport DC, Cumming WA, Ros PRo Disseminated hepatic and splenic
lesions in cat-scratch disease: imaging features. AJR Am J Roentgenol
1991; 156:1227-8.
41. Port J, Leonidas JC. Granulomatous hepatitis in cat-scratch disease: ultrasound and CT observations. Pediatr RadioI1991;21:598-9.
42. Cohen-Abbo A, Cheatham S, Edwards K. Disseminated cat-scratch disease
simulating neuroblastoma. Pediatr Infect Dis J 1992; 11:1058-60.
43. Weintrub P. Your diagnosis, please: fever of unknown origin in a fiveyear-old boy. Pediatr Infect Dis J 1991; 10:80-1, 83.
44. Reed JA, Brigati DJ, Flynn SD, et al. Immunocytochemical identification
of Rochalimaea henselae in bacillary (epithelioid) angiomatosis, parenchymal bacillary peliosis, and persistent fever with bacteremia. Am J
Surg PathoI1992; 16:650-7.
45. Dangman BC, Albanese BA, Kacica MA, Lepow ML, Wallach MT. Cat
scratch disease in two children presenting with fever of unknown origin:
imaging features and association with a new causative agent, Rochalimaea henselae. Pediatrics 1995;95:767-71.
Downloaded from http://cid.oxfordjournals.org/ by guest on September 9, 2014
25. Zangwill KM, Hamilton DH, Perkins BA, et al. Cat scratch disease in
Connecticut: epidemiology, risk factors, and evaluation of a new diagnostic test. N Engl J Med 1993; 329:8-13.
26. Koehler JE, Glaser CA, Tappero JW. Rochalimaea henselae infection: a
new zoonosis with the domestic cat as reservoir. JAMA 1994;271:
531-5.
27. Rocco VK, Roman RJ, Eigenbrodt EH. Cat scratch disease: report ofa case
with hepatic lesions and a briefreview ofthe literature. Gastroenterology
1985;89:1400-6.
28. Rizkallah MF, Meyer L, Ayoub EM. Hepatic and splenic abscesses in catscratch disease. Pediatr Infect Dis J 1988; 7: 191-5.
29. Lenoir AA, Storch GA, DeSchryver-Kecskemeti K, et al. Granulomatosis
hepatitis associated with cat scratch disease. Lancet 1988; 1: 1132-6.
30. Ma1atack JJ, Altman HA, Nard JA, Wiener ES, Urbach AH, McGregor
RS. Cat-scratch disease without adenopathy. J Pediatr 1988; 114:
101-4.
31. Kemper CA, Lombard CM, Deresinski SC, Tompkins LS. Visceral bacillary epithelioid angiomatosis: possible manifestations of disseminated
cat scratch disease in the immunocompromised host: a report of two
cases. Am J Med 1990;89:216-22.
32. Slater LN, Welch DF, Min K-W. Rochalimaea henselae causes bacillary
angiomatosis and pe1iosis hepatis. Arch Intern Med 1992; 152:602-6.
33. Steeper TA, Rosenstein H, Weiser J, Inampudi S, Snover DC. Bacillary
epithelioid angiomatosis involving the liver, spleen, and skin in an AIDS
patient with concurrent Kaposi's sarcoma. Am J Clin Pathol 1992;97:
713-8.
34. Tappero JW, Koehler JE, Berger TG, et al. Bacillary angiomatosis and
bacillary splenitis in immunocompetent adults. Ann Intern Med 1993;
118:363-5.
35. Apalsch AM, Nour B, Jaffe R. Systemic cat-scratch disease in a pediatric
liver transplant recipient and review of the literature. Pediatr Infect Dis
J 1993; 12:769-74.
957