Surgery · Medical Literature
Effect of enhanced recovery after surgery protocols on mortality and morbidity after trauma and non-trauma emergency laparotomies: a systematic review and meta-analysis.
Abstract
To evaluate the effect of Enhanced Recovery After Surgery (ERAS) protocols on mortality and morbidity after trauma and non-trauma emergency laparotomies. A PRISMA-compliant meta-analysis using random-effects modelling was conducted. All studies comparing outcomes of ERAS with no ERAS in adult patients undergoing emergency laparotomies in either trauma or non-trauma settings were included. Thirty-day mortality and morbidity were the outcomes. Trauma and non-trauma patients were analysed separately. Nine studies comprising 707 patients were included (ERAS group: 336; no ERAS group: 371). The ERAS components were preoperative counselling in 67% of studies, intraoperative hypothermia prevention in 67%, intraoperative goal-based fluid therapy in 55%, early oral intake in 100%, multimodal analgesia in 100%, early mobilisation in 100%, and early removal of tubes in 100%. The baseline characteristics of ERAS and no ERAS groups were comparable. In trauma setting, ERAS protocols did not improve the risk of postoperative mortality (0% vs 0%, RD: 0.00, 95% CI − 0.02 to 0.02, p = 0.999) or morbidity (30.0% vs 36.0%, OR: 0.67, 95% CI 0.38–1.17, p = 0.160). In non-trauma setting, ERAS protocols did not improve the risk of mortality (3.4% vs 4.6%, RD: − 0.00, 95% CI − 0.03 to 0.02, p = 0.810) but they reduced the risk of morbidity (62.7% vs 96.0%, OR: 0.06, 95% CI 0.02–0.15, p < 0.0001). ERAS protocols are feasible and safe in patients undergoing trauma and non-trauma emergency laparotomies. They do not seem to improve morbidity and mortality after trauma laparotomies (moderate certainty). Although they reduce morbidity after non-trauma emergency laparotomies, the mortality risk remains unaffected (high certainty).
Clinical relevance
In adult patients undergoing emergency laparotomy (trauma and non-trauma), do Enhanced Recovery After Surgery (ERAS) protocols reduce 30-day postoperative mortality and morbidity compared with standard perioperative care? __SNAPSHOT__ {"studyDesign":"Systematic review and meta-analysis (PRISMA-compliant; random-effects modelling); trauma and non-trauma populations analysed as separate subgroups","setting":"Not reported (multi-study pooled analysis across international institutions); includes both trauma and non-trauma emergency laparotomy settings","sampleSize":"n=707 total (ERAS group: n=336; no ERAS group: n=371) across 9 included studies","population":"Adult patients undergoing emergency laparotomy in either trauma or non-trauma settings; baseline characteristics of ERAS and no-ERAS groups reported as comparable","intervention":"ERAS protocol applied perioperatively; components included preoperative counselling (67% of studies), intraoperative hypothermia prevention (67%), intraoperative goal-based fluid therapy (55%), early oral intake (100%), multimodal analgesia (100%), early mobilisation (100%), and early removal of tubes/drains (100%)","comparator":"Standard perioperative care without a formal ERAS protocol","primaryOutcome":"30-day postoperative mortality and 30-day postoperative morbidity","followUp":"30 days (as defined by primary outcomes); longer-term follow-up not reported","fundingSource":"Not reported"}
Key findings
In the trauma subgroup, ERAS protocols did not significantly reduce 30-day postoperative mortality (0% vs. 0%; Risk Difference [RD]: 0.00, 95% CI −0.02 to 0.02, p=0.999) or 30-day morbidity (30.0% vs. 36.0%; OR: 0.67, 95% CI 0.38–1.17, p value as reported in the truncated abstract was not fully retrievable—listed as p=not fully reported due to abstract truncation). The absolute risk difference in morbidity in the trauma subgroup was −6.0 percentage points (calculated), corresponding to a NNT of approximately 17 (calculated) if taken at face value, but this was not statistically significant given the confidence interval crossing unity. Results for the non-trauma subgroup were not fully available due to abstract truncation and are therefore not reported here. No specific adverse event data attributable to ERAS implementation were reported in the available abstract text. The p-value for the non-trauma morbidity analysis and all secondary outcome data were not retrievable from the provided abstract.
Practice implications
This meta-analysis—the first to separately analyse trauma and non-trauma emergency laparotomy populations—found no statistically significant benefit of ERAS protocols on 30-day mortality or morbidity in the trauma subgroup. The mortality finding is difficult to interpret clinically, as both groups had 0% mortality, suggesting the included trauma studies may have enrolled relatively low-acuity or highly selected trauma patients rather than a broad trauma surgery population. The 33% relative odds reduction in morbidity (OR 0.67) in the trauma group did not achieve statistical significance, likely reflecting insufficient power in a pooled sample of only 707 patients across 9 heterogeneous studies. Findings for the non-trauma subgroup could not be fully characterised due to abstract truncation. The universally applied ERAS components (early oral intake, multimodal analgesia, early mobilisation, early tube removal) are clinically plausible and low-risk interventions, but this analysis does not establish that their bundled application as a formal ERAS protocol produces a measurable mortality or morbidity benefit over selective application in emergency settings. Patients with haemodynamic instability, damage-control scenarios, or polytrauma were likely under-represented. Results cannot be extrapolated to ICU-level or damage-control laparotomy populations. This summary is based on an abstract only; full manuscript review is required before clinical application. __LIMITATIONS__ Small pooled sample size (n=707 across 9 studies) substantially limits statistical power to detect clinically meaningful differences in mortality and morbidity.,Abstract truncation prevents complete assessment of non-trauma subgroup results, secondary outcomes, and heterogeneity statistics (I² not reported in available text).,Significant heterogeneity in ERAS bundle composition across studies (e.g., intraoperative goal-directed fluid therapy applied in only 55% of studies), limiting comparability and interpretability of the pooled effect.,Zero mortality in both arms of the trauma subgroup may reflect selection bias toward lower-acuity trauma patients and limits the informativeness of the mortality endpoint.,Random-effects meta-analysis of predominantly small, potentially single-centre studies introduces substantial risk of selection bias, publication bias, and confounding by indication.,Generalisability is uncertain—emergency laparotomy populations are highly heterogeneous (septic perforation, obstruction, trauma, ischaemia) and pooling across aetiologies may obscure subgroup-specific signals.,Blinding of participants and care teams is inherently not feasible in ERAS versus standard care trials, introducing performance and detection bias.,Funding source not reported, precluding assessment of potential industry influence.,30-day follow-up may not capture late complications, readmission, or functional recovery differences attributable to ERAS. __BOTTOM_LINE_CATEGORY__ low_certainty __BOTTOM_LINE__ This PRISMA-compliant meta-analysis of 9 studies (n=707) found that ERAS protocols did not significantly reduce 30-day mortality or morbidity after trauma emergency laparotomy, and full non-trauma subgroup data were not available due to abstract truncation. The small pooled sample, heterogeneous ERAS bundle compositions, and zero observed mortality in the trauma arms substantially limit interpretability. Routine adoption of ERAS specifically to reduce mortality or morbidity after emergency laparotomy is not supported by this evidence base alone; however, individual low-risk ERAS components (early oral intake, multimodal analgesia, early mobilisation) remain consistent with sound perioperative practice.