SPLENIC ABSCESS - ANOTHER LOOK AT AN OLD DISEASE

Objective: To study the changes in the incidence, causes, bacteriologi c profile, and management of a splenic abscess. Design: Retrospective case study. Setting: Tertiary, university referral center. Patients: T hirty-nine patients with a splenic abscess. Interventions: None. Main Outcome Measures: Demographics, signs and symptoms, causes, risk facto rs, diagnostic methods, bacteriologic profile, treatment, and outcome. Results: Patients presented at a mean age of 43 years (range, 2-83 ye ars), after a mean symptomatic period of 16 days, with fever (69%), ab dominal pain (56%), nausea and vomiting (38%), and splenomegaly (31%). The majority of abscesses represented metastatic infection (n = 19), and 11 were secondary to immunosuppression. Twelve patients had human immunodeficiency virus disease and 9 used intravenous drugs. In patien ts who underwent computed tomography, all had abnormal scans (n=33), w ith a well-defined abscess(es) in 28. Nine abscesses were polymicrobia l; monomicrobial isolates included grampositive organisms (23%), gram- negative organisms (31%), fungi (23%), and mycobacteria (23%). Patient s presenting before 1989 (1981-1988) (n=15) and those presenting after 1989 (1989-1996) (n=24) differed in risk factors (intravenous drug ab use, 0% vs 47% [P=.02]; hematologic malignancy, 43% vs 9% [P=.04]) and grampositive isolates (18% vs 64%; P=.06). Patients underwent splenec tomy (n=18), open drainage (n=4), medical therapy (n=10), or percutane ous drainage (n=5) with respective survival rates of 94%, 50%, 70%, an d 100%. Conclusions: In 1996, splenic abscesses are increasingly commo n. intravenous drug abuse and human immunodeficiency virus disease are significant risk factors, and the diagnosis should be considered in a patient with fever and abdominal pain who uses intravenous drugs. Ant imicrobial agents should be broad since 36% of abscesses were polymicr obial, and should include coverage of gram-positive organisms.