Diagnostic accuracy of peritoneal lavage in patients with pelvic fractures

C. Mendez, K. D. Gubler and R. V. Maier
Department of Surgery, University of Washington School of Medicine, Seattle.

OBJECTIVE: To determine the accuracy of diagnostic peritoneal lavage (DPL) for the evaluation of intraabdominal injury in patients with a pelvic fracture as a result of blunt trauma. DESIGN: Retrospective cohort analysis. SETTING: Level I trauma center in metropolitan Seattle, Wash. PATIENTS: Four hundred ninety-seven consecutive patients admitted with pelvic fractures following blunt trauma during a 60-month period. OUTCOME MEASURES: Positive results of DPL, defined by one of the following: an immediate aspiration of more than 10 mL of gross blood; a red blood cell count of more than 0.0001 x 10(12)/L; a white blood cell count greater than 0.0005 x 10(9)/L; an elevated amylase, bilirubin, or creatinine level; or organic particles or bacteria in the effluent returned after installation of 1 L of crystalloid fluid lavaged in the peritoneal cavity. RESULTS: Two hundred eighty-six patients underwent DPL. For 80 patients (28.0%), results of DPL were positive and for 194 patients (67.8%) the results of DPL were negative. For two patients (0.7%), results of DPL were false positive for a sensitivity of 94%. For another two patients (0.7%), the results of DPL were false negative for a specificity of 99%. As a diagnostic modality, DPL has a positive predictive value of 98% and a negative predictive value of 97%. CONCLUSIONS: Diagnostic peritoneal lavage is a reliable method for the evaluation of intra-abdominal injury and should remain a standard component in the evaluation of patients following blunt injury with or without pelvic fractures.