Comparison of the ALR-RSI, AOFAS and Karlsson scores for the identification of the readiness to return to sport at the pre-injury level following surgery for chronic lateral ankle instability

Abstract

Compare the value of three foot and ankle scores (AOFAS, Karlsson and ALR-RSI), to discriminate patients who returned to sport at the preinjury level from those who did not following surgery to repair Chronic Lateral Ankle Instability. The secondary outcome was to define a cutoff score for the most favorable score. A total of 159 patients (86 men and 73 women) were identified from a prospective database of patients who underwent primary CLAI surgery. Seventy-two of the patients (37 men, 35 women; mean ± SD age, 33.8 ± 11years) underwent an arthroscopic Boström procedure and 87 patients (49 men, 38 women; mean ± SD age, 37.1 ± 11 years) underwent anatomical reconstruction of the ATFL and CFL with a gracilis tendon graft using an all-arthroscopic approach. Inclusion criteria were patients at least 18 years of age with a history of foot and/or ankle surgery and a full understanding of the written language. Patients were excluded if they did not practice a sport. After a mean 29.7 (±8.5) months of follow-up, patients and physicians filled out the three scores (ALR-RSI, AOFAS, and Karlsson) at the same time as well as another questionnaire about their return to sport at the preinjury level. The mean scores for each score for the entire population were 74.9 (±27.4) for the ALR RSI, 37 (±26.5) for the Karlsson, and 80.9 (±15.7) for the AOFAS. Among the 159 included patients, only 64 returned to sport at the preinjury level while 95 patients did not return to sport. The ALR RSI score was the only score to have acceptable validity, while the validity and the differentiation of patients who returned to sport at the preinjury level were poor with the Karlsson and AOFAS scores (ALR RSI: area under ROC curve = 0.70; 95% CI, 0.6–0.77; Karlsson: area under ROC curve = 0.53; 95% CI, 0.43–0.63; AOFAS: area under ROC curve = 0.61; 95% CI, 0.52–0.70). The ALR-RSI found to be most favorable COM with a cutoff score of 83 points for a RTS at the preinjury level corresponding to a sensitivity of 63% and specificity of 71%. We validated to ALR-RSI to differentiate the RTS at the preinjury level with a cut-off score of 83 points, while the AOFAS and Karlsson scores were not validated.