Spontaneous cryptococcal peritonitis in cirrhotic patients.S Sungkanuparph, A Vibhagool, R Pracharktam
Faculty of Medicine, Ramathibodi Hospital, Mahidol University, Bangkok, Thailand. , Thailand
Correspondence Address: Source of Support: None, Conflict of Interest: None PMID: 12432196
Source of Support: None, Conflict of Interest: None
Spontaneous bacterial peritonitis is a common complication in patients with cirrhosis and ascites. However, spontaneous peritonitis caused by Cryptococcus neoformans is uncommon. Delayed diagnosis of cryptococcal peritonitis often results in death. We describe three cases of spontaneous cryptococcal peritonitis in patients with decompensated cirrhosis. One case had associated symptomatic human immunodeficiency virus infection. Clinical awareness of this entity may lead to the early diagnosis and proper treatment.
Keywords: Adult, Ascitic Fluid, microbiology,Case Report, Cryptococcosis, complications,diagnosis,Cryptococcus neoformans, isolation &purification,Fatal Outcome, Female, Follow-Up Studies, Human, Liver Cirrhosis, Alcoholic, complications,diagnosis,Male, Middle Age, Peritonitis, complications,diagnosis,Risk Assessment, Severity of Illness Index,
Spontaneous bacterial peritonitis (SBP) is a common complication in patients with cirrhosis of liver with ascites. Clinically, it presents with fever, abdominal pain and abdominal tenderness. The diagnosis of SBP is mainly based on presence of over 250 neutrophil/mm in the ascitic fluid and by demonstration of bacteria on Gramís stained smear or culture. Spontaneous peritonitis caused by fungus, particularly Cryptococcus neoformans is rarely reported. Due to rarity and presentation indistinguishable from SBP, diagnosis of spontaneous cryptococcal peritonitis is usually delayed. We report three cases of spontaneous cryptococcal peritonitis in patients with decompensated cirrhosis.
A 34-year-old man with alcoholic cirrhosis, ascites and esophageal varices developed malignant lymphoma that responded to chemotherapy. The patient presented with hepatic encephalopathy and low-grade fever. Infection was considered as a precipitating cause of encephalopathy. Ascitic fluid, which was slightly turbid, revealed white blood cell†count 450 /mm3 with 80 % lymphocytes. Serum-ascites albumin gradient (SAAG) was 1.6 g/L. Gramís and acid fast stain of the ascetic fluid showed no organism. The patient was treated with cefotaxime and lactulose. Blood and ascitic fluid cultures did not reveal growth of any organism after 3 days. On the fifth day of admission, he developed high fever and massive bleeding from ruptured esophageal varices that resulted in death. Ascitic fluid culture in in-house haemoculture bottle (brain heart infusion with 0.2% agar) and urine culture taken from the first day of admission grew Cryptococcus neoformans after 7-day incubation.
A 42-year-old female with alcoholic cirrhosis presented with abdominal pain and ascites. Examination revealed oral thrush, hairy leukoplakia, and ascites. Abdomen was diffusely tender without rebound tenderness. No neck stiffness was detected. Serum anti-HIV antibody by ELISA was positive. CD4 cell count in blood was 75 /mm3. Ascitic fluid, which was clear, revealed white blood cell†count 200 /mm3 with 50% neutrophils. SAAG was 1.7 g/L, Gramís stain did not show any organism. Prothrombin time was 22.5 second with 32% (normal 12.14.5 second with 65-106%). The patient was treated with ceftriaxone and fresh frozen plasma. On fourth day of admission, the patient developed sepsis, hypoglycemia, and massive upper gastrointestinal bleeding, which resulted in death. Culture of ascitic fluid sent on the first day and blood on the fourth day grew C. neoformans.
A 64-year-old woman with history of breast cancer treated with modified mastectomy and chemotherapy and cirrhosis of liver due to chronic HBV infection presented with acute alteration of consciousness. On admission, the patient was drowsy, had jaundice, hypothermia and hypotension. Tense ascites and localised guarding were detected on abdominal examination. There was no stiffness of neck. Ascitic fluid examination revealed slight turbidity, white blood cell†count 340 /mm3 with 60% lymphocytes. Encapsulated, budding yeasts were incidentally detected at microscopic examination. India ink preparation confirmed presence of Cryptococcus. SAAG was 1.39 g/L. Though Ceftazidime and amphotericin B were given, the patient developed DIC and died in the next day. Ascites fluid culture recovered Cryptococcus neoformans. Blood culture was sterile.
Spontaneous cryptococcal peritonitis is a rare entity. From 1963 through 2000, only 26 cases have been reported in the English literature.,,,,,,, In our study, there were only 3 cases from 1992 to 2001 in our 800-bedded medical-school hospital. As cryptococcal peritonitis was not carefully looked for in the past, the incidence may be an underestimation. The incidence of this rare entity did not remarkably increase after AIDS epidemic though cryptococcosis is a common opportunistic infection in patients with AIDS. A previous study including case report and review of the literature in 1963-1999 revealed that only one of 11 cases had HIV infection and most of the cases (72.7%) had chronic liver disease. Review of earlier reports shows that spontaneous cryptococcal peritonitis developed in 2 groups: patients with chronic liver disease and patients with continuous ambulatory peritoneal dialysis.
All the patients in our study had underlying decompensated cirrhosis. The symptoms of these patients, i.e. abdominal pain, altered sensorium and fever were indistinguishable from those in patients with SBP. White blood cell†count of the ascitic fluid revealed pleocytosis as is found in SBP. A predominance of lymphocytes may be a clue for non-bacterial causes, particularly tuberculosis and less commonly, fungal peritonitis. However, a high SAAG (>1.1 g/dL) could suggest that tuberculosis was less likely. In such a situation, the possibility of spontaneous fungal peritonitis should be considered. Ascitic fluid from one case in which India ink preparation was performed showed encapsulated budding yeasts. This may emphasize the need for this test for initial screening in suspected cases. Serum cryptococcal antigen may be another good initial test., Unfortunately, we did not do this test in any of our patients. Thus, we recommend performing India ink preparation of the ascites fluid and serum cryptococcal antigen in cirrhotic patients with spontaneous peritonitis and when there is a predominance of lymphocytes in ascites fluid.
Concurrent positive urine and blood cultures in patient 1 and 2, in our series may suggest that spontaneous cryptococcal peritonitis is often associated with disseminated cryptococcosis. There is a previous study showing that spontaneous cryptococcal peritonitis may be an indicator of disseminated cryptococcosis.
Mortality rate in cirrhotic patients developing spontaneous cryptococcal peritonitis is high., The advanced cirrhosis, disseminated fungal sepsis and its delayed diagnosis may explain such high mortality. All our patients died during the same hospital admission. Clinical suspicion†of this entity may lead to the early diagnosis and better treatment and outcome.