Serial lactate and admission SOFA scores in trauma: an analysis of predictive value in 724 patients with and without traumatic brain injury.

Abstract

Arterial lactate, base excess (BE), lactate clearance, and Sequential Organ Failure Assessment (SOFA) score have been shown to correlate with outcome in severely injured patients. The goal of the present study was to separately assess their predictive value in patients suffering from traumatic brain injury (TBI) as opposed to patients suffering from injuries not related to the brain. A total of 724 adult trauma patients with an Injury Severity Score (ISS)≥16 were grouped into patients without TBI (non-TBI), patients with isolated TBI (isolated TBI), and patients with a combination of TBI and non-TBI injuries (combined injuries). The predictive value of the above parameters was then analyzed using both uni- and multivariate analyses. The mean age of the patients was 39years (77% males), with a mean ISS of 32 (range 16-75). Mortality ranged from 14% (non-TBI) to 24% (combined injuries). Admission and serial lactate/BE values were higher in non-survivors of all groups (all p<0.01), but not in patients with isolated TBI. Admission SOFA scores were highest in non-survivors of all groups (p=0.023); subsequently septic patients also showed elevated SOFA scores (p<0.01), except those with isolated TBI. In this group, SOFA score was the only parameter which showed significant differences between survivors and non-survivors. Receiver operating characteristic (ROC) analysis revealed lactate to be the best overall predictor for increased mortality and further septic complications, irrespective of the leading injury. Lactate showed the best performance in predicting sepsis or death in all trauma patients except those with isolated TBI, and the differences were greatest in patients with substantial bleeding. Following isolated TBI, SOFA score was the only parameter which could differentiate survivors from non-survivors on admission, although the SOFA score, too, was not an independent predictor of death following multivariate analysis.