Association of multidisciplinary perioperative enhanced recovery pathways with perioperative outcomes in pediatric scoliosis correction surgery: a systematic review and meta-analysis

Background Multidisciplinary perioperative enhanced recovery pathways are increasingly utilized in pediatric scoliosis correction surgery, yet the robustness of their pooled outcome estimates compared to usual care remains inadequately characterized. This systematic review and meta-analysis evaluates the pathway-level associations of these protocols with both clinical and operational outcomes. Methods We systematically searched PubMed, EMBASE, CINAHL, Scopus, Web of Science, and the Cochrane Library from inception through March 31, 2026. Eligible studies compared structured multidisciplinary pathways with usual care in pediatric scoliosis populations. Outcomes included postoperative complications, 30-day unplanned readmission, operative duration, and length of hospital stay. Outcome definitions and ascertainment were extracted as reported by the original studies, including the reported follow-up period and measurement reference point where available. Because postoperative complications could encompass clinically heterogeneous events and length of hospital stay could be influenced by institution-specific discharge criteria and measurement conventions, no post hoc uniform definition was imposed across studies; relevant differences in outcome ascertainment were considered when interpreting the pooled evidence. Pooled odds ratios (ORs) and weighted mean differences (WMDs) with 95% confidence intervals (CIs) were derived using random-effects models. Given the inherent methodological heterogeneity of real-world observational data, restricted maximum likelihood (REML) estimation with Knapp-Hartung adjustment was applied as a prespecified sensitivity analysis for outcomes demonstrating extreme statistical heterogeneity (I 2 > 90%). DerSimonian-Laird was the primary analysis for all outcomes. Results Twenty studies were included in the review, with different subsets contributing to each outcome-specific meta-analysis. Twelve studies ( n = 1,874) revealed no statistically significant association with postoperative complications (OR: 0.75; 95% CI: 0.43 to 1.31; I 2 = 34.5%). Seven studies ( n = 1,386) demonstrated no statistically significant association with 30-day readmissions (OR: 1.21; 95% CI: 0.73 to 2.02; I 2 = 0.0%). While the primary DerSimonian–Laird analysis showed a significant reduction in operative duration (WMD: −26.05 min; 95% CI: −48.54 to −3.55), this finding became statistically non-significant under the more conservative REML with Knapp–Hartung adjustment (WMD: −22.46 min; 95% CI: −59.23 to 14.32; p = 0.197). In contrast, fifteen studies demonstrated a statistically significant but uncertain association with length of hospital stay (WMD: −1.75 days; 95% CI: −2.18 to −1.32; p < 0.001), which remained robust after conservative adjustment ( p < 0.001). Extreme heterogeneity (I 2 > 95%) was observed for all continuous outcomes, and significant publication bias was detected for length of hospital stay (Egger's test, p = 0.003). Conclusion This meta-analysis provides a quantitative signal that multidisciplinary enhanced recovery pathways are associated with a reduced length of hospital stay, suggesting an improvement in perioperative discharge efficiency. However, the evidence for reduced operative duration is statistically fragile and inconclusive, and no statistically significant associations were found for complications or readmissions. The overall certainty of the evidence is very low due to the predominance of retrospective designs, extreme heterogeneity, and publication bias.