During the past eight years, 119 patients with abdominal abscesses underwent percutaneous catheter drainage (PCD), including 76 who had successful treatment by the initial PCD, 19 who had recurrent abscesses after removal of drainage catheters and 24 who were outright failures and either died of sepsis or required surgical drainage. This study was designed to identify outcome variables that might be used prospectively to assess the therapeutic efficacy of PCD. Outcome variables included abscess size, daily drainage volume and location, presence of a gastrointestinal fistula, age, bacteriologic factors and response of the pulse rate, body temperature and leukocyte count of the patient to PCD. Ninety of 119 patients (76 per cent) ultimately had successful drainage of abscesses by PCD alone. The over-all mortality rate was 16 per cent (19 of 119), with a 75 per cent mortality rate in the failure group. Neither abscess size, bacteriologic findings nor pulse rate correlated with outcome. PCD failure was significantly greater in patients greater than or equal to 60 years (p less than or equal to 0.01) and in patients with pancreatic abscesses versus other locations (p less than or equal to 0.04). Drainage volume was significantly greater in PCD failures than among PCD successes at greater than or equal to 3 days after PCD (p less than or equal to 0.03).(ABSTRACT TRUNCATED AT 250 WORDS)