Diagnostic peritoneal lavage versus abdominal computed tomography in blunt abdominal trauma: a review of prospective studies.

OBJECTIVE: To determine whether abdominal computed tomography (CT) or diagnostic peritoneal lavage (DPL) should be used in the evaluation of hemodynamically stable patients with blunt abdominal trauma and equivocal findings on physical examination. DATA SOURCE: MEDLINE. STUDY SELECTION: Prospective studies of hemodynamically stable trauma patients with blunt abdominal trauma and equivocal findings on physical examination that compared abdominal CT and DPL. DATA EXTRACTION: Data were extracted by a single observer. DATA SYNTHESIS: Most studies had excellent DPL results. The mean sensitivity was 98% (range from 90% to 100%), the mean specificity was 92% (range from 73% to 100%), the mean positive predictive value (PPV) was 82% (range from 57% to 92%), the mean negative predictive value (NPV) was 100% (range from 99% to 100%) and the mean accuracy was 93% (range from 80% to 98%). One study reported a low specificity (73%), PPV (57%) and accuracy (80%) for DPL, which may have been due to the loose criteria for red blood cells used in that study. The mean CT values were as follows: sensitivity 60% (range from 20% to 97%), specificity 98% (range from 91% to 100%), PPV 88% (range from 50% to 100%), NPV 84% (range from 76% to 93%) and accuracy 87% (range from 73% to 97%). In studies done in the mid-1980s the CT results were inferior, but they were improved in studies reported in the 1990s (sensitivity 88%, NPV 97%, accuracy 92%). These latest studies also suggest that CT and DPL are complementary rather than equivalent studies.
CONCLUSION: DPL should be performed if there are no contraindications and no associated injuries that would be better delineated by CT, in which case abdominal CT is indicated.