Objective: Identify outcome predictors in blunt diaphragm rupture (BDR).
Design: Retrospective chart and trauma registry review.
Materials and methods: We reviewed records of patients with BDR from January 1987 through May 1994 for outcomes of mortality, intensive care unit stay, hospital stay, and ventilator days. Predictors tested were age, sex, Injury Severity Score (ISS), diagnostic delay, rupture side, head injury, and associated injuries. Stepwise regression models were developed and tested on an additional data base of 115 BDR records from four trauma centers.
Results: Thirty-two patients were identified. Age was the only significant predictor for all outcomes (p < 0.05). Age, ISS, and severe head injury were mortality predictors. In the larger data base, age and ISS remained predictive of mortality, but age was not predictive of morbidity.
Conclusions: Age and ISS are predictive of BDR mortality. No morbidity predictor was validated in the larger data base. These data emphasize that predictive models from a single institution should be applied cautiously.