Diagnostic peritoneal lavage (DPL) remains an accurate diagnostic test for intra-abdominal injury. This study examined the safety and efficacy of DPL performed by supervised residents in an urban trauma centre. A retrospective chart review was carried out of a one year experience (July 1994-June 1995). Junior surgical and senior emergency medicine residents rotating on the Trauma Service performed an open DPL by protocol under the direct supervision of senior surgical residents. Standard criteria were used for diagnosis. During this study, 1349 injured patients were admitted to the Trauma Service. Of these 525 patients underwent DPL. Complete records were available on 516 patients. The average age of the patients was 33 years and injury was primarily blunt (95%). There were 72 true positives, 428 true negatives, two false positive and seven false negative DPLs; for a sensitivity of 91.1%, specificity of 99.5% and accuracy of 96.9%. Complications occurred in 12 patients (2.3%): non-diagnostic DPL—7 (1.3%); intra-abdominal injury—4 (0.8%); wound complication—1 (0.2%). Seven patients underwent nontherapeutic laparotomy for a positive DPL. Thirty-four patients (6.6%) died, none from the DPL. DPL obviated the need for computed tomography scan of the abdomen and /or pelvis in 464 patients resulting in a cost saving of approximately $250 000. DPL performed by supervised junior surgical and senior emergency medicine residents is a safe and cost-effective method of evaluating patients with potential intra-abdominal injury.