Hepatic brucelloma (HB), a rare manifestation of brucellosis, refers to liver involvement in the form of abscess. A 35-year-old woman stockbreeder was admitted due to 1-month history of evening fever, sweating and weight loss, while she was on 3-week course of rifampicin/doxycycline for suspected brucellosis. On admission, she had hepatosplenomegaly and a systolic murmur, while cholestasis, increased inflammation markers and a strong-positive Wright-Coombs test were the main laboratory findings. As blood and bone marrow cultures were unrevealing, further investigation with CT imaging showed a central liver calcification surrounded by heterogeneous hypodense area being compatible with HB. Material from CT-guided drainage tested negative for Brucella spp. After failure to improve on a 10-week triple regiment, surgical excision was decided and Brucella spp were identified by PCR. Our case highlights challenges in establishing HB diagnosis, which should be considered on the right epidemiological context and when serological and radiological evidence favour its diagnosis.
- general practice / family medicine
- infectious diseases
- medical management
- occupational and environmental medicine
Statistics from Altmetric.com
- general practice / family medicine
- infectious diseases
- medical management
- occupational and environmental medicine
Human brucellosis is caused by Brucella spp, which are Gram-negative, non-spore-forming coccobacilli belonging to the Brucellaceae family. Brucellosis is the most common zoonotic infection worldwide with more than 500 000 reported cases annually.
Infection most commonly results after ingestion of contaminated food or via direct contact with infected animals (through ruptured mucosal surface or skin) in livestock workers, veterinarians or those working with dairy products. Inhalation of Brucella-containing aerosolised particles is a less frequent mode of infection.1 It is highly likely that the actual prevalence of the disease is underestimated in endemic areas (Mediterranean Basin, Latin America, Middle East) due to suboptimal recording and disease registering.
So far, approximately 60 cases of hepatic involvement in brucellosis presenting as liver abscess—known also as hepatic brucelloma (HB)—have been reported.2 Indeed, in a series of 905 patients with brucellosis diagnosed and followed in a single teaching hospital in Spain, HB was reported in only 1.5% of patients. Of note, in most cases, HB has been associated with the chronic forms of the disease, often in the absence of preceding Brucella-associated symptoms.3
Here, we present a case of HB opting to highlight the difficulties which physicians might face during work-up to establish a firm diagnosis of this clinical entity. Furthermore, we wanted to stress that these patients might often require a multimodal therapeutic approach.
A 35-year-old female dairy and livestock farmer from rural Greece was admitted to our department because of 1-month history of evening fever, lately up to 39°C, accompanied by rigours. The patient also reported foul smelling night sweats, generalised weakness and weight loss up to 5 kg although she had been under treatment by her family physician with oral rifampicin and doxycycline for suspected brucellosis in the last 3 weeks. Symptoms initially began with low-grade fever and anorexia, with the patient also recalling abdominal tenderness of the right upper quadrant during the first couple of days. Throughout this period, she visited multiple healthcare facilities and a positive Rose Bengal test had been reported. Her medical history included hypothyroidism, dyslipidaemia and anxiety disorder. The patient denied any travel history during the last 6 months or any current or past unsafe use of alcohol, herbal or supplements and nasal or intravenous drug abuse. On admission, clinical examination revealed a mild systolic murmur (1 out of 6, Levine scale), hepatosplenomegaly and poor dental hygiene.
Laboratory investigation revealed high gamma-glutamyl transpeptidase and alkaline phosphatase, increased serum markers of inflammation and anaemia (table 1). The remaining haematological, biochemical and microbiological markers including repeatedly blood, stools and urine cultures as well as the rheumatoid factors, the C3 and C4 complement components and the ferritin levels were within normal limits. A chest X-ray and ultrasound of the abdomen were also unrevealing.
At the same time, a Wright-Coombs test for Brucella melitensis was strongly positive at a titre of 1:2560, making brucellosis the most likely diagnosis. However, extended incubation of blood and bone marrow cultures tested negative for the isolation of Brucella spp. In addition, a bone marrow biopsy was negative for the detection of granulomas that could be possibly associated with brucellosis.
Based on the above clinical and laboratory findings, the differential diagnosis, apart from brucellosis, included also several other causes of persistent fever such as infective endocarditis and other zoonotic infections. More specifically, considering the patient’s profession and her residency in a rural region, we also anticipated leishmaniasis and Q-fever, which are endemic in our region.4–11 In this regard, transthoracic and trans-oesophageal echocardiograhic investigations proved negative for vegetation-like lesions while the serological work-up for zoonosis other than brucellosis (Coxiella, Rickettsia, Borrelia, Leptospira, Leishmania) was also negative.
The poor response to double antibiotic treatment with rifampicin and doxycycline for 3 weeks prior to admission led to further investigation. A chest and abdominal CT scan was performed for the detection of possible localised brucellosis and the exclusion of other causes of prolonged fever, including malignancies. The CT scan of the abdomen revealed a central liver calcification (2.3 cm) surrounded by a heterogeneous hypodense area with ill-defined borders (8.5 cm in diameter; figure 1A,B), which according to the medical literature was primarily compatible with HB. Moreover, other causes of hepatic abscesses and especially those caused by parasitic infections (Entamoeba and Echinococcus) tested negative.
The patient was set under triple antibiotic therapy—thus treated as a case of localised brucellosis—with oral rifampicin (900 mg/day), doxycycline (100 mg two times per day) and intramuscular streptomycin (1 g/day) for 3 weeks. Thereafter, the treatment was modified to oral rifampicin, doxycycline and oral ciprofloxacin (500 mg two times per day). As part of monitoring response to treatment, a new CT scan of the upper abdomen was performed after 2 weeks with no significant change of radiological findings. Therefore, a CT-guided drainage was attempted but with poor results. The minimal product of drainage also tested negative for Brucella spp by both extended incubation cultures and a broad-range PCR assay as previously described.12–15 After 10 weeks of triple treatment (6 weeks as inpatient and 4 weeks as outpatient) there was only partial clinical and laboratory improvement (cessation of fever and normalisation of inflammatory markers, but persistence of cholestasis; table 1). However, no radiological improvement was shown on a third CT. For these reasons, surgical excision of the lesion was deemed the best course of action opting for prevention of relapse and effective cure (figure 2). Biopsy of the tissue revealed the presence of non-specific necrosis and granulose lesions, while PCR from the same material identified Brucella spp that confirmed the originally suspected diagnosis as stockbreeders at least in our country are at high risk for contracting brucellosis either through direct contact with animals or by consumption of unpasteurised dairy products. After the establishment of HB diagnosis, the patient was discharged in good health and reassessed every 3 months for a total of 16 months being under oral triple antibiotic treatment (rifampicin, doxycycline and ciprofloxacin).
Outcome and follow-up
At last follow-up, 2 months after completion of a 16-month triple antibiotic treatment, the patient is still in good health and afebrile without any episode of relapse or drugs-related side effects. In addition, there was a significant improvement of the laboratory (table 1) and radiological findings (figure 1C,D).
Liver involvement presenting with hepatomegaly or mild derangement of liver function tests is reported in a significant proportion of patients with brucellosis. Granuloma formation is the typical feature on liver histology, which is considered as an effort of the immune system to restrict the infectious process.4 16 HB is a rare manifestation of brucellosis and most often appears as a prolonged recurrent disease that frequently remains asymptomatic for long periods. This is due to the ability of Brucella to evade the host immune responses and survive as intracellular pathogen in various organs in the body.17 In fact, the identification of circulating DNA even years after effective cure in asymptomatic patients underscores the potential of Brucella to survive leading subsequently to relapses and long-term sequelae, such as the development of hepatic abscess.18 19
As HB is a rare manifestation of brucellosis that often presents non-specific clinical features, its diagnosis can be proved highly challenging. Accordingly, in two series of patients with chronic hepatosplenic brucellosis from Spain, half of them, residing in rural areas, had a history of brucellosis dating back 2–35 years prior to their present illness.3 20 Diagnostic failure could be potentially catastrophic in terms of choice and duration of treatment as well as follow-up, with patients often being treated with standard antibiotic therapy for pyogenic abscesses instead of receiving Brucella-specific treatment.
Hereby, we wanted to highlight the difficulties, which clinicians might face in establishing the diagnosis of HB. Moreover, we wanted to emphasise on the challenges of the management of these patients that could pose for physicians. In particular, the lack of guidelines for the optimal choice and treatment duration further complicates the management of patients with HB, suggesting an individualised approach.
Indeed, the diagnosis of HB may pose difficulties as conventional microbiological techniques have reportedly severe limitations.3 21 In fact, as highlighted also in our case, blood and pus cultures often remain unrevealing. Moreover, the inability of B. melitensis, which is responsible for most of the Brucella cases in the Mediterranean Basin, to be recovered from pus relies mainly on its low inoculum in the pus or the preceding antibiotic treatment.
In cases of chronic and/or localised brucellosis serum agglutination tests revealing high IgG titres offer higher sensitivity compared with blood, bone marrow or pus cultures that are often negative.21 This supports the theory that the hepatic abscess emerges rather as a result of reactivation of brucellosis in the setting of chronic disease than during acute infection. Still, the use of serum agglutination tests should be interpreted cautiously in areas endemic for brucellosis. In this context, it is worth mentioning that in several cases low titre of antibodies is reported, which can hinder the diagnosis in patients with localised infection. Additionally, antibody titres have been used for the follow-up of patients during treatment, as the decline of the antibody titres has been associated with successful treatment outcome. Furthermore, substantially high titres can be maintained in a small proportion of patients, including ours, after successful antibiotic course and should not be considered as treatment failure in particular, when there is clinical improvement and cure.22
We were able to establish the diagnosis of brucellosis after identification of the organism by PCR in the liver tissue obtained during surgery. In fact, PCR assays have emerged as indispensable tools in the diagnosis of localised brucellosis, since it has been proven to have higher sensitivity compared with cultures and serology, even in patients who had received previous antibiotic treatment.2 3 23 Indeed, a study of 34 non-blood samples from patients with different forms of localised brucellosis showed that PCR had higher sensitivity (97.1%) compared with the conventional cultures (29.4%).23 As in our case, radiology and especially the CT imaging can provide significant help in the diagnosis of HB. The presence of central calcification surrounded by hypoechoic and hypodense area, often forming a single pseudo-tumour lesion, is considered characteristic but not specific for this clinical entity.2 24 25
Therapy of patients with HB is another issue that needs attention considering that no treatment guidelines exist regarding the choice and duration of antibiotics. The success of the treatment of brucellosis relies on the use of combination of antibiotics, while in the case of HB, a triple antibiotic regimen is endorsed for most cases.1 Still, no clear definition of treatment success or treatment failure exists and long-term use of antibiotics is apparently an effective option in a proportion of patients. Additional intervention with drainage or surgical excision of the lesion might be required, as this is the only therapeutic option offering complete cure.2 3
In conclusion, physicians should keep in mind brucellosis in the differential diagnosis as a rare cause of hepatic abscesses in areas of high prevalence, especially when serological and radiological evidence further favour the diagnosis. In these cases, the social history and epidemiological data seem of outmost importance for the achievement of effective patient care. Moreover, PCR assays can be proven helpful in cases of HB when other diagnostic techniques, including cultures and serology, fail to establish a firm diagnosis.
Hepatic brucelloma (HB) is a rare manifestation of brucellosis and its diagnosis can be highly challenging, as conventional microbiologic techniques, including serology and cultures, have reportedly severe limitations in such cases.
PCR assays have emerged as indispensable tools in the diagnosis of localised brucellosis, as they bear higher sensitivity compared with cultures and serology, even though patients have received previous antibiotic treatment.
Taking into account the lack of treatment guidelines regarding to the choice and duration of antibiotics, individualised treatment for HB is endorsed.
A multimodal approach consisting of antibiotics alongside with drainage and/or surgical excision of the lesion may be required.
Contributors GND and EIR had the original idea, designed the study and along with GV wrote the first draft of the manuscript. GV and EIR collected and summarised the published literature and the data of the patient. GND, EIR and GV were the principal treating physicians and made the clinical diagnosis while KT made the surgery and followed the patient. GND, EIR and KT made the final critical revision of the manuscript for important intellectual content. All authors have seen and approved the final version of the manuscript.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests None declared.
Patient consent for publication Obtained.
Provenance and peer review Not commissioned; externally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.