Is pain relief enough? Early response discordance after lumbar decompressive surgery: a prospective cohort study

Background Different patient-reported outcome measures (PROMs) may classify the same patient differently as a treatment responder after lumbar decompressive surgery. We aimed to quantify inter-instrument agreement on minimal clinically important difference (MCID)-based responder status at one month postoperatively, characterize multi-dimensional response discordance, and identify independent one-month PROM predictors of one-month patient dissatisfaction. Methods In this prospective observational cohort study, 158 consecutive adults undergoing lumbar decompressive surgery for disc herniation ( n = 77) or spinal stenosis ( n = 81) at a single neurosurgical centre were enrolled; all underwent single-level decompression of the symptomatic level. Of these, 134 (84.8%) completed paired baseline and one-month assessments using the Visual Analogue Scale for leg pain (VAS), Oswestry Disability Index (ODI), and EQ-5D-5 L utility index. MCID thresholds were derived from external sources (VAS ≥ 1.5 points, ODI ≥ 10 points, EQ-5D-5 L ≥ 0.10). Inter-instrument agreement was quantified using Cohen’s kappa (κ). Patient dissatisfaction was defined by the satisfaction subscale of the Swiss Spinal Stenosis Questionnaire (SSSQ > 12). Independent predictors of dissatisfaction were identified by multivariable linear regression with bootstrapped confidence intervals. Results All instruments showed significant improvement at one month (all p < 0.001). MCID-based responder rates differed markedly: VAS 82.1%, EQ-5D-5 L 83.6%, and ODI 54.5%. Inter-instrument agreement was slight to fair (κ = 0.19–0.29); Only 47.0% of patients achieved MCID across all three instruments simultaneously, while 94.0% achieved it on at least one — producing a 47-percentage-point disagreement zone. A total of 39.6% (53/134) of patients experienced VAS and/or EQ-5D-5 L improvement without corresponding ODI improvement. 23% (31/134) of patients were dissatisfied at one month. In multivariable regression adjusted for age, sex, and diagnosis (R² = 0.40), one-month EQ-5D-5 L items of mobility limitation and anxiety/depression, together with residual functional disability (ODI total), independently predicted dissatisfaction; residual pain did not. Conclusions Instrument selection alone changes responder classification rates by nearly 30% points at one month after lumbar decompressive surgery. Dissatisfaction is driven by residual functional limitation and psychological distress, not residual pain. Multi-dimensional PROM assessment with diagnosis-informed instrument selection may help identify, at the early postoperative visit, patients requiring targeted rehabilitation and psychological support. Trial registration Not applicable.