Background context:

The minimal clinically important difference (MCID) represents the smallest change in an outcome measure recognized as clinically meaningful to a patient after receiving a clinical intervention. Most studies that discussed the MCIDs for lumbar spinal stenosis (LSS) included mixed pathologies or procedures despite that the MCID value should be different depending on the intervention. Moreover, despite the efficacy of adopting percentage-change improvement for the MCID threshold, there are limited reports and discussions in the field of lumbar surgery.

Purpose:

The aim of the present study was to elucidate the MCIDs for the Oswestry Disability Index (ODI), EuroQOL 5-dimension 3-level (EQ-5D-3L), physical component summary (PCS) of the Short Form of the Medical Outcomes Study, and Numeric Rating Scale (NRS) in patients with degenerative LSS treated with decompression surgery without fusion.

Study design/setting:

A multicenter retrospective cohort study was performed.

Patient sample:

A total of 422 patients who underwent decompression surgery for LSS and answered a complete set of questionnaires were included in the study. Patients who underwent endoscopic or revision surgery were excluded.

Outcome measures:

Preoperative and 1-year postoperative scores of each health-related quality of life questionnaires (HRQOLs) and patient satisfaction questionnaire response Methods: The patient satisfaction question was used as an anchor, and the cutoff values were estimated based on absolute point improvement from baseline using a receiver-operating characteristic (ROC) curve analysis and the “mean change” method for MCIDs. The MCID values for percentage-change in HRQOLs were also calculated using ROC curve analysis. The three cutoff values for each HRQOL were validated using the Youden index for determining the most robust MCIDs.

Results:

Of the patients, 356 (84.4%) were at least “somewhat satisfied” with the treatment results. The two cutoff values of absolute point-change in each HRQOL, which were estimated by two different anchor-based methods, were similar. The area under the curve of the ROC curve for percentage-change tended to be higher than that for absolute point-change. Moreover, the Youden index of the percentage-change in each HRQOL was higher than that of the absolute point-change calculated by either the “mean change” method or the ROC curve analysis. Based on these results, it was proposed that MCID was 42.4% for percentage-change in ODI, 22.0% for EQ-5D-3L, 13.7% for PCS, 25.0% for NRS (low back pain), 55.6% for NRS (leg pain), 22.2% for NRS (leg numbness).

Conclusions:

The MCIDs of HRQOLs were calculated in patients with LSS treated with decompression surgery without concomitant fusion procedure. The MCID cutoffs based on percentage-change from baseline were more effective than those of absolute point-change.