Effectiveness of selective dorsal rhizotomy for spastic cerebral palsy: a systematic review and single-arm meta-analysis.

BACKGROUND

Selective dorsal rhizotomy (SDR) is used to treat spastic cerebral palsy (CP), but its effectiveness remains debated due to heterogeneous, predominantly non-randomized evidence.

METHODS

This systematic review and single-arm meta-analysis was prospectively registered in PROSPERO (CRD420251132796) and conducted in accordance with PRISMA 2020 and the Cochrane Handbook. PubMed, Embase, and Web of Science were searched for studies evaluating SDR in pediatric CP. Eligible designs included case series, retrospective or prospective cohorts, and randomized trials reporting motor or spasticity outcomes. The primary endpoints were Gross Motor Function Measure-66 (GMFM-66) and Modified Ashworth Scale (MAS). Risk of bias was assessed using the ROBINS-I tool. Mean change values with 95% confidence intervals (CI) were pooled using random-effects models (REML). Statistical heterogeneity was evaluated with the I2 statistic, and sensitivity analyses (including leave-one-out and varying correlation coefficients for missing SDs) were performed.

RESULTS

Sixteen studies including 756 children were analyzed. SDR was associated with significant improvements in GMFM-66 at the shortest (mean change 3.29; 95% CI 1.54-5.03; I2=71%) and longest (mean change 3.71; 95% CI 2.00-5.42; I2=72%) available follow-up. Spasticity was significantly reduced, with a pooled mean change in MAS of -1.98 (95% CI -3.26 to -0.70; I2=99%). Sensitivity analyses confirmed the stability of results across assumptions, and leave-one-out analyses did not materially alter effect estimates. All included studies were judged to have a moderate risk of bias.

CONCLUSIONS

SDR is associated with clinically meaningful and durable improvements in gross motor function and reductions in spasticity in children with spastic CP.