Enhanced recovery after surgery (ERAS) pathways in gynecologic oncology: A systematic review and meta-regression of perioperative domains.

Abstract

e17537 Background: Enhanced Recovery After Surgery (ERAS) pathways are multimodal perioperative care programs designed to reduce surgical stress and accelerate postoperative recovery. While ERAS implementation in gynecologic oncology (GynOnc) has been associated with improved outcomes, domain-specific effects are unclear, and limited data exists on its impact in marginalized populations of social risk. Methods: We conducted a systematic review and meta-analysis of ERAS pathways versus usual care in adults undergoing GynOnc surgery. Eligible studies included randomized controlled trials (RCTs) and nonrandomized studies (NRS) reporting perioperative outcomes. ERAS exposure was defined as multi-stage, evidence-based domains per ACOG and ERAS Society guidelines. Primary outcomes were length of stay (LOS), postoperative complications (overall and Clavien-Dindo ≥III), 30-day readmission, reoperation, and mortality. Random-effects meta-analyses were conducted, with meta-regression to display the impact of different ERAS domains on primary outcomes. Risk of bias and all stages were completed by individual reviewers. The protocol is registered on OSF. Results: Inclusion criteria were met in 46 studies, with 31 studies providing data for meta-analysis. There were 5823 ERAS patients (EG) vs. 5119 controls (CG). Weighted mean age was 58.58 in EG and 58.19 in CG. Mean difference in hospital LOS was 2.21 [95%CI 1.28-3.14; i2=97%] days shorter in RCTs and 1.2 [0.88-1.43; i2=98%] in NRS compared to usual care. Complications were reduced by 40% [RR 0.60, (0.41-0.87); i2=49%] in RCTs and 21% [RR 0.79, (0.69-0.90); i2=73%] in NRS, compared to usual care. Multivariable meta-regression in RCTs showed early diet prolonged LOS by 3.63 days (34% of between-study variability), while 8 other domains independently shortened LOS by up to 1.62 days. There was no dose-response relationship between number of ERAS domains and LOS. Conclusions: ERAS displayed significant improvements in patient outcomes and is associated with reduced LOS regardless of implementation intensity. This supports framing ERAS as a threshold intervention rather than a linear “more is better” exposure. Protocol harmonization via structured ERAS protocols may optimize the perioperative care of GynOnc patients. Further understanding of the interaction of domains in such analysis is warranted. Consistent reporting methodology in ERAS studies is integral to deciphering which domains hold priority, while limited equity data is available to assess if ERAS implementation holds similar weight in different patient groups.