Clinical outcomes associated with adding pelvic correction manipulation to adjacent needling for lumbar disc herniation: a prospective comparative cohort study

Vollständiger Abstract

Worum geht es in dieser Arbeit?

Background Lumbar disc herniation (LDH) commonly causes persistent pain, disability, and recurrence despite conservative care. Whether pelvic correction manipulation adds durable benefit to adjacent needling remains uncertain. Objective This study aimed to evaluate the association of adding pelvic correction manipulation to adjacent needling with disability, pain, and recurrence in LDH. Methods This single-center prospective comparative cohort study with retrospective long-term outcome ascertainment enrolled consecutive adults with symptomatic, imaging-confirmed LDH in Wuxi, China, from December 2019 to December 2020. Patients received adjacent needling plus pelvic correction manipulation or adjacent needling alone according to clinician–patient choice. The primary outcome was change in Oswestry Disability Index (ODI) at treatment completion. Secondary outcomes included pain, Japanese Orthopedic Association score, ODI response, recurrence, and adverse events through 24 months. Baseline-adjusted mixed-effects models, propensity-score overlap weighting, multiple imputation, and sensitivity analyses were used. Results Among 208 patients, 106 received combined therapy and 102 adjacent needling alone. Combined therapy was associated with greater ODI reduction at treatment completion (adjusted difference, −3.40 points; 95% CI, -5.40 to -1.40; P = 0.001) and at 24 months (−4.40; 95% CI, −7.30 to −1.50; P = 0.003). At 24 months, ≥50% ODI improvement occurred in 58.9% versus 40.5% (adjusted RR, 1.41; 95% CI, 1.06–1.87), and recurrence in 23.6% versus 39.2% (adjusted HR, 0.55; 95% CI, 0.34–0.89). Exploratory alignment, mobility, and inflammatory findings also favored combined therapy. No serious treatment-related events occurred. Conclusion Adding pelvic correction manipulation was associated with modestly better long-term disability outcomes and lower recurrence, but randomized trials are needed to confirm causality. Because treatment was not randomized, these findings should be read as associations rather than proof of a causal effect, and unmeasured confounding cannot be excluded.