Assessment of the minimum clinically important difference in pain, disability, and quality of life after anterior cervical discectomy and fusion

Object Treatment effectiveness following spine surgery is usually gauged with the help of patient-reported outcome (PRO) questionnaires. Although these questionnaires assess pain, disability, and general health state, their numerical scores lack direct, clinically significant meaning. Thus, the concept of minimum clinically important difference (MCID) has been introduced, which indicates the smallest change in an outcome measure that reflects clinically meaningful improvement to patients. The authors set out to determine anterior cervical discectomy and fusion (ACDF)–specific MCID values for the visual analog scale (VAS), Neck Disability Index (NDI), 12-Item Short-Form Health Survey (SF-12), and EQ-5D (the EuroQol health survey) in patients undergoing ACDF for cervical radiculopathy. Methods Data on 69 patients who underwent ACDF for cervical radiculopathy were collected in the authors' web-based, prospective registry during the study enrollment period. Patient-reported outcome questionnaires (VAS–neck pain [NP]), VAS–arm pain [AP], NDI, SF-12, and EQ-5D) were administered preoperatively and 3 months postoperatively, allowing 3-month change scores to be calculated. Four established calculation methods were used to calculate anchor-based MCID values using the North American Spine Society (NASS) patient satisfaction scale as the anchor: 1) average change, 2) minimum detectable change (MDC), 3) change difference, and 4) receiver operating characteristic (ROC) curve analysis. Results Sixty-one patients (88%) were available at follow-up. At 3 months postoperatively, statistically significant improvement (p < 0.001) was observed for the following PROs assessed: VAS-NP (2.7 ± 3.3), VAS-AP (3.7 ± 3.6), NDI (23.2% ± 19.7%), SF-12 physical component score (PCS; 10.7 ± 9.7), and EQ-5D (0.20 ± 0.23 QALY). Improvement on the SF-12 mental component score (MCS) trended toward significance (3.4 ± 11.4, p = 0.07). The 4 MCID calculation methods generated a range of MCID values for each of the PROs: VAS-NP 2.6–4.0, VAS-AP 2.4–4.2, NDI 16.0%–27.6%, SF-12 PCS 7.0–12.2, SF-12 MCS 0.0–7.2, and EQ-5D 0.05–0.24 QALY. The maximum area under the curve (AUC) was observed for NDI (0.80), and the minimum AUC was observed for SF-12 MCS (0.66) and EQ-5D (0.67). Based on the MDC approach, the MCID threshold was 2.6 points for VAS-NP, 4.1 points for VAS-AP, 17.3% for NDI, 8.1 points for SF-12 PCS, 4.7 points for SF-12 MCS, and 0.24 QALY for EQ-5D. The mean improvement in patient scores at 3 months surpassed the MCID threshold for VAS-NP, NDI, and SF-12 PCS but not for VAS-AP, SF-12 MCS, and EQ-5D. Conclusions The ACDF-specific MCID is highly variable depending on the calculation technique used. The MDC approach seems to be most appropriate for MCID calculations in the ACDF population, as it provided a threshold value above the 95% confidence interval of nonresponders (greater than the measurement error) and was closest to the average change of most PROs reported by responders. When the MDC method was applied with the NASS patient satisfaction scale as the anchor, the MCID thresholds were 2.6 points for VAS-NP, 4.1 points for VAS-AP, 17.3% for NDI, 8.1 points for SF-12 PCS, 4.7 points for SF-12 MCS, and 0.24 QALY for EQ-5D.