Predictors of Discordance Between Neurological and Quality-of-Life Recovery After Surgery for Degenerative Cervical Myelopathy: A Retrospective Cohort Study
In brief
Neurological and quality-of-life recovery diverged in 43% of neck surgery patients
In a multicentre cohort, 42.6% of patients improved meaningfully in neurological function or quality of life, but not both, one year after surgery for degenerative cervical myelopathy. Higher starting scores in one domain predicted a mismatch in the other, and patients with neurological-only recovery had less improvement in pain-related disability. The findings show why both outcomes matter when assessing recovery.
- Journal
- Spine (Q1)
- Published
- 6 November 2026
- Study design
- Prospective / inception cohort
- Evidence level
- Level 2, Moderate (CEBM 2b)
- Authors
- Joshua Bougadis, Mohammed Ali Alvi, Karlo M Pedro, Ava Bidari, Michael G Fehlings
- PMID
- 43008133
- DOI
- 10.1097/BRS.0000000000005924
Why clinicians should know about it
- Picked for Spine Surgery (top studies of the week, 15 November 2026): Recovery profiles and predictors after degenerative cervical myelopathy surgery
Abstract
STUDY DESIGN: Retrospective analysis of prospectively collected multicentre cohort data. OBJECTIVE: Quantify discordance between neurological recovery (modified Japanese Orthopaedic Association [mJOA]) and quality-of-life (QoL) recovery (SF-36 PCS) after surgery for degenerative cervical myelopathy (DCM) and identify predictors of each profile. SUMMARY OF BACKGROUND DATA: Neurological and patient-reported outcomes capture different dimensions of recovery after DCM surgery, but the frequency, structure, and baseline correlates of discordance between these outcome domains remain incompletely characterized. METHODS: Retrospective analysis of prospective multicentre AO Spine CSM-NA and CSM-I cohorts. Recovery used published minimum clinically important differences (MCID) for mJOA and SF-36 PCS at 12 months, with 6-month values carried forward when unavailable. Patients were classified as concordant recovery, neurological-only recovery, QoL-only recovery, or no recovery. Discordant recovery was defined as achieving one MCID without the other. Baseline predictors were estimated by multinomial logistic regression using multiple imputation (m=20) for missing covariates. RESULTS: Of 353 patients, 310 were classifiable: 88 (28.4%) concordant, 50 (16.1%) neurological-only, 82 (26.5%) QoL-only, 90 (29.0%) no recovery; 132 (42.6%) discordant. Higher baseline PCS predicted neurological-only versus concordant recovery (OR 4.42 per SD, 95% CI 2.34-8.35), whereas higher baseline mJOA predicted QoL-only (OR 5.67, 2.95-10.90) and no recovery (OR 4.23, 2.25-7.95; all P <0.001). Older age, higher Nurick grade, lower mental component score and greater frailty also predicted non-concordant profiles. In exploratory models, ossification of the posterior longitudinal ligament predicted no recovery (OR 5.71, 2.17-15.0) and number of compressed levels predicted neurological-only recovery (OR 1.85, 1.17-2.92). Neurological-only patients improved less in pain-related disability (NDI Q1) than concordant or QoL-only patients (P <0.02). CONCLUSIONS: Neurological and QoL recovery were discordant in over 40% of patients and governed by different baseline variables. Neurological-only recovery reflects limited room for measurable QoL gain, compounded by persistent pain, rather than surgical failure. LEVEL OF EVIDENCE: III.
Abstract as published, via PubMed.
For healthcare professionals. The summary is generated by AI from the published abstract, and the evidence level is assigned automatically from the study design on the Oxford CEBM hierarchy. Neither is medical advice. Read the full paper before changing practice.