Whole body computed tomography during trauma resuscitation—Effect on survival

Abstract

Introduction: An increasing number of trauma centres uses whole body computed tomography (WBCT) as a diagnostic tool for the early, primary trauma survey. There is no evidence to date suggesting that use of WBCT has a significant effect on the outcome of major trauma patients. Methods: In a retrospective, multicentre study we compared the probability of survival (Ps) in those blunt trauma patients who received WBCT during resuscitation to those who did not. Using data derived from the Trauma Registry of the German Trauma Society, we determined the Ps according to the trauma and injury severity score (TRISS), the revised injury severity classification score (RISC) and the standardized mortality ratio (SMR, observed/expected mortality). Results: 1494 (32.3%) of the identified 4621 patients received WBCT. The mean age was 42.6 years, 72.8% were male and mean ISS was 29.7. SMR calculated by TRISS was 0.745 (CI95% 0.633-0.859) for the WBCT vs. 1.023 (CI95% 0.909-1.137) for the non-WBCT group (p < 0.001). RISC-score calculation revealed a SMR of 0.865 (CI95% 0.774-0.956) for the WBCT vs. 1.034 (CI95% 0.959-1.109) for the non-WBCT group (p = 0.017), respectively. This means that the observed mortality rates in the WBCT-group are significantly lower than predicted by the TRISS/RISC. Multivariate adjustment for hospital level, year of trauma and potential centre effects confirmed that WBCT is an independent predictor for survival (p ≤ 0.002). The "number needed to scan" is 17 based on TRISS and 32 based on RISC calculation. Discussion: Integration of WBCT into early trauma care significantly increases the P s of polytrauma patients. The relative reduction of the mortality rate based on the TRISS is 25% (CI95% 14-37%) and 13% (CI95% 4-23%) based on the RISC. Despite potentially harmful effects of increased radiation exposure WBCT is recommended as a standard diagnostic tool during early resuscitation phase of polytrauma patients (Lancet, in press).