Preoperative Short Form Health Survey Score Is Predictive of Return to Play and Minimal Clinically Important Difference at a Minimum 2-Year Follow-up After Anterior Cruciate Ligament Reconstruction

Abstract

Background: There is increased interest in understanding the preoperative determinants of postoperative outcomes. Return to play (RTP) and the patient-reported minimal clinically important difference (MCID) are useful measures of postoperative outcomes after anterior cruciate ligament reconstruction (ACLR). Purpose: To define the MCID after ACLR and to investigate the role of preoperative outcome scores for predicting the MCID and RTP after ACLR. Study Design: Case-control study; Level of evidence, 3. Methods: There were 294 active athletes enrolled as part of an institutional ACL registry with a minimum 2-year follow-up who were eligible for inclusion. A questionnaire was administered to elicit factors associated with RTP. Patient demographic and clinical data as well as patient-reported outcome measures were captured as part of the registry. Outcome measures included the International Knee Documentation Committee (IKDC) subjective knee evaluation form, Lysholm scale, and 12-Item Short Form Health Survey (SF-12) physical component summary (PCS) and mental component summary (MCS). Preoperative outcome score thresholds predictive of RTP were determined using a receiver operating characteristic (ROC) with area under the curve (AUC) analysis. The MCID was calculated using a distribution-based method. Multivariable logistic models were fitted to identify predictors for achieving the MCID and RTP. Results: At a mean (±SD) follow-up of 3.7 ± 0.7 years, 231 patients were included from a total 294 eligible patients. The mean age and body mass index were 26.7 ± 12.5 years and 23.7 ± 3.2 kg/m2, respectively. Of the 231 patients, 201 (87.0%) returned to play at a mean time of 10.1 months. Two-year postoperative scores on all measures were significantly increased from preoperative scores (IKDC: 50.1 ± 15.6 to 87.4 ± 10.7; Lysholm: 61.2 ± 18.1 to 89.5 ± 10.4; SF-12 PCS: 41.5 ± 9.0 to 54.7 ± 4.6; SF-12 MCS: 53.6 ± 8.1 to 55.7 ± 5.7; P < .001 for all). The corresponding MCID values were 9.0 (IKDC), 10.0 (Lysholm), 5.1 (SF-12 PCS), and 4.3 (SF-12 MCS). Preoperative score thresholds predictive of RTP were the following: IKDC, 60.9; Lysholm, 57.0; SF-12 PCS, 42.3; and SF-12 MCS, 48.3. These thresholds were not independently predictive but achieved significance as part of the multivariable analysis. In the multivariable analysis for RTP, preoperative SF-12 PCS scores above 42.3 (odds ratio [OR], 2.73; 95% CI, 1.09-7.62) and SF-12 MCS scores above 48.3 (OR, 4.41; 95% CI, 1.80-10.98) were predictive for achieving RTP; an ACL allograft (OR, 0.26; 95% CI, 0.06-1.00) was negatively predictive of RTP. In the multivariable analysis for the MCID, patients with higher preoperative scores were less likely to achieve the MCID ( P < .0001); however, a higher preoperative SF-12 MCS score was predictive of achieving the MCID on the IKDC form (OR, 1.27; 95% CI, 1.11-1.52) and Lysholm scale (OR, 1.08; 95% CI, 1.00-1.16). Medial meniscal injuries, older age, and white race were also associated with a decreased likelihood for achieving the MCID. Conclusion: Preoperative SF-12 MCS and PCS scores were predictive of RTP after ACLR; patients scoring above 42.3 on the SF-12 PCS and 48.3 on the SF-12 MCS were more likely to achieve RTP. Additionally, we defined the MCID after ACLR and found that higher SF-12 MCS scores were predictive of achieving the MCID on knee-specific questionnaires.