Combination of red cell distribution width and American Society of Anesthesiologists score for hip fracture mortality prediction.

UNLABELLED: The prognostic value of red cell distribution width (RDW) and a combination of RDW and the American Society of Anesthesiologists (ASA) score for long-term hip fracture mortality remains unknown. Our data showed that both RDW and ASA were independent risk predictors. A combination of these two parameters may provide a more powerful strategy for the prediction of hip fracture mortality.

INTRODUCTION: Red cell distribution width (RDW) has recently been suggested as an independent predictor of prognosis in a variety of disorders. The American Society of Anesthesiologists (ASA) system has been widely used to stratify patients for outcome evaluations. However, the prognostic value of RDW and a combination of RDW and the ASA score for long-term hip fracture mortality has yet to be studied.

METHODS: This prospective cohort study included 1402 subjects from 2000 to 2011 with a follow-up study over a 2 year period. Cox proportional hazards models with a bootstrap validation were used to evaluate associations of RDW, ASA, and a combination of both with long-term mortality. The global fit and the area under the receiver operating characteristic (ROC) curve (AUC) for model discrimination were further analyzed.

RESULTS: Both RDW and ASA exhibited as independent risk predictors of 2-year mortality. The population with elevation of either RDW or ASA increased the risk of mortality (bootstrap validated hazard ratio (HR) 1.971 95 % confidence interval (CI) [1.336-3.005] p < 0.01) while those with an increase in both assessments (bootstrap validated HR 2.667 95 % CI [1.526-4.515] p < 0.01) were at the highest risk for mortality. The addition of the combination of ASA and RDW improved the discrimination power of risk prediction models (AUC increased from 0.700 to 0.723, p < 0.05).

CONCLUSION: Both RDW and ASA exhibited as independent risk predictors of 2-year hip fracture mortality. The combination of these two readily available parameters may provide a more powerful and effective strategy for the assessment of all-cause mortality in hip fracture patients.