Cut-off values of preoperative knee extensor strength and hip abductor strength for predicting good walking ability after total knee arthroplasty.

Abstract

Total knee arthroplasty (TKA) reduces pain and improves physical function; however, not all patients have successful outcomes after surgery. To identify these patients would be critical information for improving rehabilitation programs. The purpose of this study was to clarify the cut-off values of lower extremity muscle strength for predicting postoperative good walking ability. Timed Up and Go test of 105 patients was measured at 6months postoperatively, and participants were divided into good (< 9.1s) and poor (≥ 9.1s) walking ability. Both sides of knee extensor strength (KES) and hip abductor strength (HAS) were measured using hand-held dynamometer preoperatively. Receiver operating characteristic (ROC) curve analysis was used to identify cut-off values for classifying the participants into the two groups. Of the 105 patients, 54 were allocated in the poor walking ability group, whereas 51 were allocated in the good walking ability group. KES and HAS were significantly greater in the good walking ability group than in the poor walking ability group. ROC curve analysis revealed that the cut-off value for KES was 0.79 Nm/kg (area under the curve (AUC) 0.68; sensitivity 64.7%; specificity 68.5%) on the involved side and 0.86 Nm/kg (AUC 0.73; sensitivity 84.6%; specificity 55.6%) on the uninvolved side, and for HAS was 0.57 Nm/kg (AUC 0.71; sensitivity 60.8%; specificity 71.7%) on the involved side and 0.61 Nm/kg (AUC 0.76; sensitivity 66.7%; specificity 77.4%) on the uninvolved side. The cut-off values of preoperative KES and HAS for predicting good walking ability after TKA are 0.79 Nm/kg on the involved side and 0.86 Nm/kg on the uninvolved side, and 0.57 Nm/kg on the involved side and 0.61 Nm/kg on the uninvolved side, respectively. We should provide enhanced pre- and post-operative rehabilitation programs for patients with muscle strength lower than these values.