Which frailty measure is a good predictor of early post-operative complications in elderly hip fracture patients?

Introduction Current pre-operative assessment using, e.g., American Society of Anaesthesiologists score does not accurately predict post-operative outcomes following hip fracture. The multidimensional aspect of frailty syndrome makes it a better predictor of post-operative outcomes in hip fracture patients. We aim to discover which frailty measure is more suitable for prediction of early post-operative outcomes in hip fracture patients. Methods Hundred consecutive hip fracture patients seen by the orthogeriatric service were included. We collected baseline demographic, functional and comorbidity data. In addition to ASA, a single blinded rater measured frailty using two scales (i) modified fried criteria (MFC) and (ii) reported edmonton frail scale (REFS). The MFC adopted a surrogate gait speed measure with two questions: (i) Climbing one flight of stairs and (ii) Ability to walk 1 km in the last 2 weeks. Immediate post-operative complications during the inpatient stay were taken as the primary outcome measure. Results Subjects had mean age of 79.1 +/- 9.6 years. Sixty six percent were female and 87 % of Chinese ethnicity. Eighty two percent had surgery, of which 37.8 % (n = 31) had post-operative complications. Frailty, measured by MFC (OR 4.46, p = 0.04) and REFS (OR 6.76, p = 0.01) were the only significant predictors of post-operative complications on univariate analyses. In the hierarchical logistic regression model, only REFS (OR 3.42, p = 0.04) predicted early post-operative complications. At 6 months follow-up, REFS significantly predicted [basic activities of daily living (BADL)] function on the multivariable logistic regression models. (BADL, OR 6.19, p = 0.01). Conclusions Frailty, measured by the REFS is a good predictor of early post-operative outcomes in our pilot study of older adults undergoing hip surgery. It is also able to predict 6 months BADL function. We intend to review its role in longer-term post-operative outcomes and validate its potential role in pre-operative assessment of older adults undergoing hip surgery.