Granulomatous hepatitis due to Bartonella henselae infection in an immunocompetent patient
© VanderHeyden et al; licensee BioMed Central Ltd. 2012
- Received: 27 July 2011
- Accepted: 23 January 2012
- Published: 23 January 2012
Bartonella henselae (B. henselae) is considered a rare cause of granulomatous hepatitis. Due to the fastidious growth characteristics of the bacteria, the limited sensitivity of histopathological stains, and the non-specific histological findings on liver biopsy, the diagnosis of hepatic bartonellosis can be difficult to establish. Furthermore, the optimal treatment of established hepatic bartonellosis remains controversial.
We present a case of hepatic bartonellosis in an immunocompetent woman who presented with right upper quadrant pain and a five cm right hepatic lobe mass on CT scan. The patient underwent a right hepatic lobectomy. Surgical pathology revealed florid necrotizing granulomatous hepatitis, favoring an infectious etiology. Despite extensive histological and serological evaluation a definitive diagnosis was not established initially. Thirteen months after initial presentation, hepatic bartonellosis was diagnosed by PCR studies from surgically excised liver tissue. Interestingly, the hepatic granulomas persisted and Bartonella henselae was isolated from the patient's enriched blood culture after several courses of antibiotic therapy.
The diagnosis of hepatic bartonellosis is exceedingly difficult to establish and requires a high degree of clinical suspicion. Recently developed, PCR-based approaches may be required in select patients to make the diagnosis. The optimal antimicrobial therapy for hepatic bartonellosis has not been established, and close follow-up is needed to ensure successful eradication of the infection.
- Granulomatous hepatitis
- Bartonella henselae
Cat scratch disease (CSD) is caused by B. henselae, a gram-negative, aerobic alpha Proteobacterium that is transmitted by the bite or scratch of a cat. The traditional diagnostic criteria for CSD include (1) contact with a cat and history of a scratch or other inoculation event, (2) positive cat scratch skin test reaction, (3) regional lymphadenopathy with no other apparent etiology, and (4) characteristic histopathologic features on biopsy . The development of PCR-based diagnostic assays has revealed a growing number of cases in which the traditional diagnostic criteria for CSD were absent .
Hepatic bartonellosis has been reported in 1-2% of CSD cases, and represents the third most common clinical manifestation after fever and lymphadenopathy. The typical symptoms include right upper quadrant pain, fevers, malaise, weight loss, chills and headaches .
Establishing the diagnosis of CSD-induced liver disease can be challenging:
Radiologic evaluation by contrast-enhanced CT may show a range of abnormalities, with lesions that are hypoattenuating, iso-attenuating, or show rim enhancement relative to the uninvolved liver . MRI imaging may show suggestive imaging patterns of the granulomas .
Histologically, hepatic bartonellosis is characterized by the presence of necrotizing granulomas. They are caused by the focal accumulation of activated macrophages, with a surrounding rim of lymphocytes and fibroblasts. However, hepatic granulomas are nonspecific and may be associated with a variety of liver diseases . Steiner silver staining, Bartonella-specific immunohistochemical stains, and serologic tests for Bartonella-specific antibodies  have been shown to have diagnostic utility but may be less sensitive and specific compared to PCR testing [8–12]. PCR assays do not rely on the presence of a humoral immune response, and may therefore be diagnostic at an earlier stage of infection, or during chronic infection in anergic patients.
Conventional bacterial cultures of liver tissue are rarely diagnostic of B. henselae hepatitis. However, Breitschwerdt and colleagues have recently developed an optimized enrichment culture method that enhances diagnostic detection and molecular typing (by DNA sequencing of the PCR amplicon) of B. henselae and other Bartonella spp. from venous blood. This highly sensitive method is able to detect active infection in patients who lack B. henselae antibodies and in whom conventional bacterial cultures have been negative [13, 14].
In this study, we describe how the use of optimized serum and tissue PCR assays resulted in the establishment of B. henselae hepatitis in a young, immunocompetent patient with severe granulomatous hepatitis.
In August, 2008, a 36 year-old woman with no prior history of liver disease presented to her primary care physician complaining of abdominal pain of five days duration. The pain was constant, sharp, localized to the right upper quadrant, non-radiating, and associated with nausea but no vomiting. Review of systems was positive for fatigue. Initial laboratory testing revealed only a mildly elevated alanine aminotransferase (ALT) of 42 IU/L. A right upper quadrant ultrasound demonstrated hepatomegaly and steatosis. Two weeks later, the patient presented to an outside hospital with worsening right upper quadrant pain, low-grade fevers, nausea and vomiting. An MRI of the abdomen revealed a two cm enhancing lesions of the right hepatic lobe. Fine needle biopsy of the lesion demonstrated a nonspecific, mixed inflammatory cellular infiltrate and steatohepatitis. There was no evidence of malignancy. Fungal and mycobacterial cultures were negative. The patient was discharged without a specific diagnosis, and no treatment was instituted.
Serological tests performed to identify the etiology of granulomatous hepatitis in the patient
Result Admission 1
Result Admission 2
Antibody to human immunodeficiency virus
Epstein Barr virus
B. quintana-IgM, IgG
Negative at 1:20, 1:60 dilution
Rapid plasma reagin
Coxiella bruniti (ELISA)
Negative at 1:20 dilution
Negative at 1:60 dilution
Cyrptococcal Ag (serum)
Histoplasma Ag (urinary)
Blastomycosis Ag (urinary)
On the third day of her hospitalization, the patient developed a temperature of 101.1°F. A comprehensive investigation for fever of unknown origin was initiated (see Table 1). Antibody titers for Bartonella quintana, and Brucella sp. were negative. An RPR was negative. Repeat HIV antibody testing was negative. A peripheral smear for malaria was negative. A transesophageal echocardiogram did not reveal any valvular vegetations. The patient refused a lumbar puncture. A percutaneous liver biopsy was performed and revealed extensive granulomatous hepatitis with occasional fibrin rings, with a background of mixed, micro- and macro-vesicular steatosis. Based on the presence of fibrin rings, the possibility of Q fever was entertained. However, ELISA testing for Coxiella burnetti antibodies was negative. Repeat special stains and cultures for AFB, fungal organisms, and cytomegalovirus were negative. Steiner silver stains for spirochetes and bacteria - including Bartonella sp.- were negative. The patient was empirically treated with a seven-day course of piperacillin and tazobactam. She remained afebrile for the remainder of her hospitalization. Based on the diagnosis of granulomatous hepatitis and the unrevealing workup for infectious organisms, the patient was started on empiric prednisone. Her liver enzymes, which had already been down-trending at the start of prednisone treatment, normalized over the next several weeks. Her symptoms resolved, and she was discharged home on a tapering dose of prednisone.
Based on the striking features of her hepatic granulomas, the possibility of hepatic Bartonella infection was raised, despite the negative Bartonella antibody titers. A sample of formalin-fixed liver tissue from the initial liver resection was sent to the University of Arkansas for PCR testing. Using previously validated assay conditions, a 153 bp fragment of the B. henselae 16S rRNA gene was amplified, and confirmed by Southern blot hybridization [15, 16]. Upon further questioning, the patient reported that she had intermittently come in contact with a cat while visiting her mother's house, although she did not recall any cat scratches or bites.
Based on the presumptive diagnosis of hepatic bartonellosis, the patient was started on azithromycin 250 mg daily, and her prednisone was discontinued. Two weeks later, the patient developed diarrhea and abdominal cramping that were attributed to her antibiotic. Azithromycin was discontinued, and the patient was started on a nine-week course of clarithromycin at a dose of 500 mg twice daily. Her symptoms completely resolved.
In May of 2010, the patient presented to our emergency room with recurrent right upper quadrant pain. On examination, she was afebrile and her vital signs were stable. There was no jaundice. Right upper quadrant tenderness was present in the area of the excisional scar. There was no palpable hepato-splenomegaly. No skin rashes were present. AST and ALT levels were elevated to 109 IU/L and 75 IU/L, respectively. The total bilirubin, alkaline phosphatase, and INR were normal. A liver-protocol CT revealed a 2.4 cm, low-density lesion at the previous surgical site, which was thought to be nonspecific. However, given her recurrent symptoms, she was empirically treated with a six-week course of ciprofloxacin 500 mg twice daily.
Our case illustrates the unique challenges of establishing and treating B. henselae-induced granulomatous hepatitis.
With regard to the establishment of the diagnosis, many of the typical clinical and laboratory features of CSD were absent. Similar to several case reports in the literature, our patient presented with atypical and nonspecific constitutional symptoms . Steiner silver stains of infected liver tissue and immunohistochemical assays did not reveal the culprit organism, and serologic tests failed to detect B. henselae-specific antibodies. A high degree of clinical suspicion, persistent attempts to establish a specific diagnosis, and the use of optimized enrichment cultures and PCR amplification were ultimately required to establish the diagnosis in our patient. Strikingly, PCR assays targeting three distinct sequences and performed at three different laboratories confirmed the presence of B. henselae at multiple time points in the patient's liver and serum, suggesting that this bacterium was the likely causative agent in this case.
With respect to treatment options, our patient's course challenges the traditional view of CSD as an inevitably self-limited disease. Recent guidelines have stressed the importance of prolonged treatment and antibiotic combination therapy in subgroups of patients with CSD, including individuals with HIV infection, bacillary angiomatosis, peliosis hepatis, retinitis, chronic bacteremia, endocarditis, chronic lymphadenopathy and neurological disorders . However, prospective studies validating these recommendations are lacking.
In vitro testing has demonstrated sensitivity of B. henselae strains to a wide spectrum of antibiotics including beta-lactams, macrolides, cephalosporins, aminoglycosides, flouroquinolones, doxycycline and rifampin . However, most of these agents are bacteriostatic and fail to eliminate the bacterium when used as monotherapy. In the only prospective, double-blind, placebo-controlled study on immunocompetent patients with CSD , azithromycin monotherapy resulted in improved lymphadenopathy but did not prevent dissemination or infectious complications. This disappointing result may have been due to the development of azithromycin resistance by B. henselae . Recurrent bacteremia has been described in several reports , even after a prolonged antibiotic course.
Our review of the published literature identified a small number of retrospective studies using antibiotic combination therapy: Arisoy and colleagues demonstrated the efficacy of rifampin and gentamycin in a small cohort of pediatric patients . In a single case report, the triple combination of doxycycline, erythromycin and azithromycin cleared B. henselae-induced, post liver-transplant granulomatous hepatitis [26, 27]. A combination of doxycycline (100 mg twice daily) and rifampin (300 mg twice daily) was used successfully to treat B. henselae retinitis . Prednisone has been suggested as adjunctive therapy in patients with antibiotic-refractory hepatic bartonellosis. However, immunosuppression should be used with caution, as it might contribute to the development of Bartonella endocarditis [29, 30].
Our patient was initially treated with surgical resection, reminiscent of a previous case report by Murano in which a giant, B. henselae-induced granuloma in a 10-year old child required a partial hepatectomy . This was followed by multiple courses of empiric antibiotic therapy, including piperacillin/tazobactam, azithromycin, clarithromycin, and ciprofloxacin. None of these treatments eradicated the infection. Following the establishment of B. henselae bacteremia and the demonstration that B. henselae genetic material was present in the patient's liver throughout the course of her disease, we opted to treat the patient with an 8-week course of rifampin (300 mg bid) and doxycycline (100 mg bid). Three months following this most recent treatment, the patient's liver enzymes have completely normalized for the first time, and her constitutional symptoms have not recurred. We are planning on a long-term follow-up and repeat enrichment blood cultures in the future.
Written informed consent was obtained from the patient for publication of this case report and any accompanying images. A copy of the written consent is available for review by the Editor-in-Chief of this journal.
The authors wish to thank Laura Lamps at the University of Arkansas for performing the 16S rDNA PCR, the Microbiology Department at the University of Washington for performing the RibC PCR, Nandhakumar Balakrishnan at North Carolina State University for performing the 16S-23S ITS PCR, and Dr. Keith Linder at North Carolina State University for the B. henselae immunohistochemistry. We also acknowledge Dr. Jonathan Lubens for his critical review of the manuscript.
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
- Bergmans AM, Groothedde JW, Schellekens JF, et al: Etiology of cat scratch disease: comparison of polymerase chain reaction detection of Bartonella (formely Rochalimaea) and Afipia felis DNA with serology and skin tests. J Infect Dis. 1995, 171: 916-923. 10.1093/infdis/171.4.916.View ArticlePubMedGoogle Scholar
- Hansmann Y, DeMartino S, Piemont Y, et al: Diagnosis of cat scratch disease with detection of Bartonella henseale by PCR: a study of patients with lymph node enlargements. J Clin Microbiol. 2005, 43: 3800-3806. 10.1128/JCM.43.8.3800-3806.2005.View ArticlePubMedPubMed CentralGoogle Scholar
- Lamps LW: Hepatic Granulomas with Emphasis on Infectious Causes. Adv Anat Pathol. 2008, 15: 309-318. 10.1097/PAP.0b013e31818a6477.View ArticlePubMedGoogle Scholar
- Mortele KJ, Segatto E, Ros PR: The infected liver: radiologic-pathologic correlation. Radiographics. 2004, 24: 937-955. 10.1148/rg.244035719.View ArticlePubMedGoogle Scholar
- Dangman BC, Albanese BA, Kacica MA, et al: 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: 767-771.PubMedGoogle Scholar
- Drebber U, Kasper HU, Ratering J, et al: Hepatic granulomas: histological and molecular pathological approach to different diagnosis: a study of 442 cases. Liver Int ISSN. 2008, 1478: 828-834.View ArticleGoogle Scholar
- Regnery RL, Olson JG, Perkins BA, Bibb W: Serological response to "Rochalimaea henselae" antigen in suspected cases of cat-scratch disease. Lancet. 1992, 339: 1443-1445. 10.1016/0140-6736(92)92032-B.View ArticlePubMedGoogle Scholar
- Dupon M, Savin De Larclause AM, Brouqui P, et al: Evaluation of serological response to Bartonella henselae, Bartonella quintana and Afipia felis antigens in 64 patients with suspected cat-scratch disease. Scand J Infect Dis. 1996, 28: 361-366. 10.3109/00365549609037920.View ArticlePubMedGoogle Scholar
- Caponetti GC, Pantanowitz L, Marconi S, et al: Evaluation of immunohistochemistry in identifying Bartonella henselae in cat-scratch disease. Am J Clin Pathol. 2009, 131: 250-256. 10.1309/AJCPMNULMO9GPLYU.View ArticlePubMedGoogle Scholar
- Adivor B, Kletter Y, Abulafia S, et al: Molecular diagnosis of cat scratch disease: a two-step approach. J Clin Microbiol. 1997, 35: 1924-1930.Google Scholar
- Anderson B, Sims K, Regnery R, et al: Detection of Rochalimaea henselae DNA in specimens from cat scratch disease patients by PCR. J Clin Microbiol. 1994, 32: 942-948.PubMedPubMed CentralGoogle Scholar
- La Scola B, Raoult D: Serological cross-reactions between Bartonella quintana, Bartonella henselae, and Coxiella burnetti. J Clin Microbiol. 1996, 34: 2270-2274.PubMedPubMed CentralGoogle Scholar
- Breitschwerdt EB, Maggi RG, Lantos PM, et al: Bartonella vinsonii subsp. berkhoffii and Bartonella henselae in a father and daughter with neurological disease. Parasites and Vectors. 2010, 3: 29-10.1186/1756-3305-3-29.View ArticlePubMedPubMed CentralGoogle Scholar
- Breitschwerdt EB, Maggi RG, Sigmon B, et al: Isolation of Bartonella quintana from a Woman and a Cat following Putative Bite Transmission. J Clin Microbiol. 2007, 45: 270-272. 10.1128/JCM.01451-06.View ArticlePubMedGoogle Scholar
- Anderson B, Sims K, Regnery R, et al: Detection of Rochalimaea henselae DNA in specimens from cat scratch disease patients by PCR. J Clin Microbiol. 1994, 32: 942-948.PubMedPubMed CentralGoogle Scholar
- Scott MA, McCurley TL, Vnencak-Jones CL, et al: Cat scratch disease: detection of Bartonella henselae DNA in archival biopsies from patients with clinically, serologically, and histologically defined disease. Am J Pathol. 1996, 149: 2161-2167.PubMedPubMed CentralGoogle Scholar
- Bereswill S, Hinkelmann S, Kist M, et al: Molecular analysis of riboflavin synthesis genes in Bartonella henselae and use of the ribC gene for differentiation of Bartonella species by PCR. J Clin Microb. 1999, 37: 3159-3166.Google Scholar
- Varanat M, Broadhurst J, Linder K, Maggi M, Breitschwerdt E: Identification of Bartonella henselaein 2 Cats With Pyogranulomatous Myocarditis and Diaphragmatic Myositis. Vet Pathol. 2011, Google Scholar
- Maman E, Bickels J, Ephros M, et al: Musculoskeletal manifestations of cat scratch disease. Clin Infect Dis. 2007, 45: 1921-1933.View ArticleGoogle Scholar
- Rolain JM, Brouqui P, Koehler JE, et al: Recommendations for treatment of human infections caused by Bartonella species; Minireview. Antimicrobiol Agents Chemother. 2004, 48: 1921-1933. 10.1128/AAC.48.6.1921-1933.2004.View ArticleGoogle Scholar
- Dorbecker C, Sander A, Oberle K, et al: In vitro susceptibility of Bartonella species to 17 antimicrobial compounds: comparison of Etest and agar dilution. J Antimicrob Chemother. 2006, 58: 784-788. 10.1093/jac/dkl341.View ArticlePubMedGoogle Scholar
- Bass JW, Freitas BC, Freitas AC, et al: Prospective randomized double blind placebo-controlled evaluation of azithromycin for treatment of cat-scratch disease. Pediatr Infect Dis J. 1998, 17: 447-452. 10.1097/00006454-199806000-00002.View ArticlePubMedGoogle Scholar
- Biswas S, Maggi RG, Papich MG, et al: Comparative activity of pradofloxacin, enrofloxacin and azithromycin against Bartonella henselae isolates derived from cats and a human. J Clin Microbiol. 2010, 48: 617-618. 10.1128/JCM.01287-09.View ArticlePubMedGoogle Scholar
- Sykes JE, Lindsay LL, Maggi RG, et al: Human coinfection with Bartonella henselae and two hemotropic mycoplasma variants resembling Mycoplasma ovis. J Clin Microbiol. 2010, 48: 3782-3785. 10.1128/JCM.01029-10.View ArticlePubMedPubMed CentralGoogle Scholar
- Arisoy ES, Correa AG, Wagner ML, et al: Hepatosplenic cat-scratch disease in children: selected clinical features and treatment. Clin Infect Dis. 1999, 28: 778-784. 10.1086/515197.View ArticlePubMedGoogle Scholar
- Bonatti H, Mendez J, Guerrero I, et al: Disseminated Bartonella infection following liver transplantation. Transpl Int. 2006, 19: 683-687. 10.1111/j.1432-2277.2006.00336.x.View ArticlePubMedGoogle Scholar
- Thudi KR, Kreikemeier JT, Phillips NJ, et al: Cat scratch disease causing hepatic masses after liver transplant. Liver Int. 2007, 27: 145-148.View ArticlePubMedGoogle Scholar
- Reed JB, Scale DK, Wong MT, et al: Bartonella henselae neuroretinitis in cat scratch disease. Diagnosis, management, and sequelae. Ophthalmol. 1998, 105 (3): 459-466. 10.1016/S0161-6420(98)93028-7.View ArticleGoogle Scholar
- Turner JW, Pien BC, Ardoin SA, et al: A man with chest pain and glomerulonephritis. Lancet. 2005, 365: 2062-10.1016/S0140-6736(05)66701-3.View ArticlePubMedGoogle Scholar
- Vikram HR, Bacani AK, DeValeria PA, et al: Bivalvular Bartonella henselae prosthetic valve endocarditis. J Clin Microbiol. 2007, 45: 4081-4084. 10.1128/JCM.01095-07.View ArticlePubMedPubMed CentralGoogle Scholar
- Murano I, Yoshii H, Kurashige H, et al: Giant hepatic granuloma caused by Bartonella henselae. Pediatr Infect Dis J. 2001, 20: 319-320. 10.1097/00006454-200103000-00023.View ArticlePubMedGoogle Scholar
- The pre-publication history for this paper can be accessed here:http://www.biomedcentral.com/1471-2334/12/17/prepub