Viewpoints regarding the use of computed tomography (CT) and diagnostic peritoneal lavage (DPL) in the evaluation of stable blunt abdominal trauma patients remain polarized and their respective roles are ill-defined. To further clarify their independent and combined value, the authors report the results of a prospective study of their use in adult patients satisfying the entry criteria of major blunt torso trauma, hemodynamic stability, equivocal and unreliable abdominal examination, and absence of both before abdominal surgery and unstable pelvic fractures. The 116 patients admitted to the study over an 11-month period were first submitted to a CT scan using a 9800 series GE scanner (General Electric Medical Systems; Milwaukee, WI). After its review with the faculty CT radiologist, the surgeon recorded his decision as to the need for the laparotomy. All patients were then subjected to DPL and the results recorded as recommending laparotomy based on finding aspiration of greater than 10 cc of blood or greater than or equal to 100,000 RBC/mm3 greater than W 500 WBC/mm3, or positive gram stain. A final decision was then formulated based on information gained from both studies. Actual need for operation or observation was recorded based on ultimate outcome in observed patients or upon findings at laparotomy. Initial laparotomy was performed in 22 patients; yet only 17 had injuries requiring repair. Unnecessary laparotomy was recommended by DPL in 15.5 percent of patients vs 0.8 percent by CT (P greater than .01), while inappropriate observation was recommended by DPL in 1.7% vs 6.9% by CT (P less than .01).(ABSTRACT TRUNCATED AT 250 WORDS)