Frailty: an independent predictor of burns mortality following in-patient admission

Abstract

IntroductionBurn injury in the elderly is associated with increased morbidity and mortality. It is not uncommon for biological age, or frailty, to differ from chronological age in this patient group and thus predicting individual clinical outcomes remains challenging. It has been previously shown that Rockwood’s Clinical Frailty Scale, a global clinical measure of fitness and frailty in older people, can be a useful adjunct for predicting outcomes for elderly patients with burns >10% TBSA. We refine our previous work to investigate the impact of frailty on mortality of elderly patients with thermal burns of any size admitted to a burns unit and explore its role as a meaningful adjunct to the modified Baux score. MethodsA retrospective analysis of case notes for all patients ≥65years admitted to our burns centre as an in-patient during an 8-year period was performed with standard demographics, burn injury parameters, length of stay and mortality outcomes collected. Measures of frailty were reviewed and statistically analysed to assess the impact of biological aging on clinical outcome in order to assess how the modified Baux score may be developed for the elderly using Frailty Score. Results239 patients met the inclusion criteria. Mean age was 77years (range: 65–99years) and mean burn size was 14.46% TBSA (Range: 0.1–98% TBSA). The modified Baux and Frailty Score were both independent predictors of mortality (p<0.0001). Increased premorbid Frailty Score was associated with increased in-hospital (OR: 2.33, 95% CI: 1.63–3.34) and one-year mortality (OR: 3.13, 95% CI: 2.22–4.41) independent of burn size compared to the modified Baux Score (IHM OR: 1.09; 95% CI: 1.07–1.13, 1yr M: OR 1.08; 95% CI: 1.05–1.11). The Frailty Score (>3) was a much more sensitive predictor of one-year mortality (Sensitivity: 83.9%; Specificity: 66.4%) than the modified Baux (>97) (Sensitivity: 59.8%; Specificity: 82.9%). A Frailty Score >3 when combined with the modified Baux score demonstrated increased area under ROC curve for both in-hospital (0.89 (95% CI: 0.85–0.94); p=0.02) and one-year (0.88 (95% CI: 0.84–0.92); p=0.02) mortality when compared to the modified Baux alone. ConclusionWe demonstrate that Frailty Score can be used to independently predict in-hospital and one-year mortality for thermal burns of any size in the elderly admitted as an in-patient to a burns unit. We also find that the Frailty Score can be employed in combination with the modified Baux score to improve mortality prediction. We recommend that Frailty Score is integrated into the modified Baux score and used to focus burn care resources appropriately.