Listeria monocytogenes peritonitis: presentation, clinical features, treatment, and outcome.

BACKGROUND: Spontaneous bacterial peritonitis (SBP) is a serious complication in cirrhotic patients. Gram (-) (E. coli, Klebsiella pneumoniae), and Gram (+) (Streptococci, Staphylococci) bacteria are most frequently cultured from patients'ascites. Listeria monocytogenes (Lm) is scarcely reported as a causative agent.

OBJECTIVE: Our objective is to describe Lm peritonitis as a clinical entity, including its presentation, clinical features, treatment, and the potential factors that might affect survival outcome.

DATA SOURCES: MEDLINE, Scholar.Google, Scopus databases, including English, Spanish, French, and German language papers published between 1966 and June 2011, and reference lists.

DATA EXTRACTION: investigators abstracted details about medical history, disease presentation, laboratory data, treatment and outcome.

DATA SYNTHESIS: One-hundred and twenty-eight cases with known survival outcome--eighty-six cirrhotics, seventeen individuals undergoing continuous ambulatory peritoneal dialysis and another twenty-five with other or no underline condition were reviewed. An additional number of twenty-five cases with unknown outcome were searched in Listeria studies published from 1990 to 2009 and were only used for calculating worldwide distribution.

CONCLUSION: Cirrhotics, mostly alcoholics, presented with fever and abdominal pain. Those who succumbed had significantly higher peripheral WBC count (15622 vs. 8155 cells/mm(3), p = 0.01) and (%) polymorphonuclear cells in differential count (83.3 vs. 71%, p = 0.001). Higher mortality was experienced in those with comorbidities, and those who presented with encephalopathy. Lower mortality was experienced in patients on continuous ambulatory peritoneal dialysis. Ascites was neutrocytic in 86% of the samples. In the sum of the cases mortality was 27.3%, with significantly highest rates in the elderly, in patients with bacteremia, immunosuppression, hematological malignancies, and lowest rates in those who presented with abdominal pain and in diabetics (type I or II). The latter observation was surprising and could be considered a single fortuitous fact. Initial appropriate treatment was associated with significantly better outcome (p = 0.002) than inappropriate; combination therapy with an aminoglycoside was superior to monotherapy (p = 0.038).