Diagnostic Peritoneal Lavage for Diagnosing Blunt Hollow Visceral Injury: The Accuracy of Two Different Criteria and Their Combination

Abstract

To test the usefulness of diagnostic peritoneal lavage (DPL) for identifying blunt hollow visceral injury with two different sets of criteria or a combination of the two. Fifty victims with physical examinations and/or computed tomography findings equivocal for blunt hollow visceral injury underwent DPL. Whether or not to perform surgery was determined based on Otomo's DPL criteria [lavage white blood cell counts (L-WBC) over lavage red blood cell counts (L-RBC) divided by 150 (L-WBC > or = L-RBC/150) in the presence of hemoperitoneum, or L-WBC over 500/mm(3) (L-WBC > or = 500) in the absence of hemoperitoneum]. The cell count ratio, a comparison of L-WBC, L-RBC, peripheral WBC (P-WBC), and peripheral RBC (P-RBC) [(L-WBC/L-RBC)/(P-WBC/P-RBC) > or = 1] were all calculated retrospectively. There were one and two false-positive cases based on Otomo's criteria and the cell count ratio, respectively, with corresponding accuracies of 97.8% and 95.7%, respectively. There were no false-positive or -negative cases according to the combined use of Otomo's criteria and cell count ratio, yielding an accuracy of 100%. Although each criterion alone is very accurate in predicting the presence of blunt hollow visceral injury, the combined use of the two would further improve the accuracy of the diagnosis and thereby reduce the number of unnecessary celiotomies.