S2579 TB or Not TB: Tuberculous Peritonitis in a Patient With Decompensated Cirrhosis

Introduction: Diagnosis of tuberculous peritonitis poses a significant challenge for clinicians given its rarity in developed countries and non-specific findings. Management of tuberculous peritonitis adds an additional level of complexity especially in cirrhotic patients. There are currently no existing guidelines for treatment of tuberculosis in cirrhotics. Case Description/Methods: A thirty-nine-year-old Hispanic male with history of alcohol abuse presented to the emergency room with 2 months of malaise and weight loss. Physical exam was notable for fever, jaundice, and abdominal distention. Laboratory findings showed platelet count 137 x 103/μl, INR 2.0, albumin 2.5 g/dL, total bilirubin 3.7 umol/L, ALT 28 IU/L, AST 108 IU/L, alkaline phosphatase 96 IU/L; MELD-Na 25, Child-Pugh Class C. Ascitic fluid studies showed 1130 polymorphonuclear leukocytes, lymphocytic predominance, protein >2.5, serum-ascites albumin gradient >1. Interferon gamma release assays held conflicting results. Ascitic fluid adenosine deaminase levels were low. CT abdomen and pelvis revealed cirrhotic liver and non-specific omental infiltration. Ceftriaxone was started for presumed spontaneous bacterial peritonitis despite inconsistent fluid studies. Antibiotics were broadened due to persistent fevers and recurrent ascites without improvement. Empiric treatment with isoniazid, ethambutol, linezolid and levofloxacin was initiated with resolution of fevers. Omental biopsy revealed acid-fast stain positive for Mycobacterium Tuberculosis, confirming diagnosis of tuberculous peritonitis. Course was complicated by thrombocytopenia requiring multiple transfusions, thought related to linezolid which was replaced with cyclosporine. He then developed diffuse maculopapular rash and cyclosporine was switched to amikacin. Interestingly, our patient did not suffer complications due to hepatotoxicity. He clinically improved and was discharged with antitubercular treatment. Discussion: Tuberculous peritonitis has a high mortality rate which can in part be explained by difficulty in making an early diagnosis. Clinical findings are often insufficient to differentiate tuberculous peritonitis from alternative diagnoses. First line antitubercular treatment is not recommended with existing liver dysfunction. This case demonstrates the high index of suspicion required for diagnosis of tuberculous peritonitis as well as the imperativeness of initiating a tailored regimen in patients with underlying liver disease to avoid undue complications and mortality.Figure 1.: Ascitic fluid cytopathology: Scattered reactive appearing mesothelial cells and mild inflammation.