Enhanced Postoperative Care Units: What the latest evidence shows
Published manuscripts on the topic of EPOCs are well summarised in two recent systematic reviews:
Review one
Schockaert BR, van Bruchem RM, Engel MF, Stolker RJ, van Lier F, Hoeks SE. Outcomes following extended postoperative recovery unit admission in noncardiac surgery: A systematic review and meta-analysis. Eur J Anaesthesiol. 2025 May 1;42(5):407-418. doi: 10.1097/EJA.0000000000002145. Epub 2025 Mar 6. PMID: 40062429; PMCID: PMC11970609.
This analysis included both traditional ICUs and intermediate care units (EPUs).
The principal findings were:
Mortality
The overall pooled 30-day or in-hospital mortality across 15 studies was 3% (95% CI 2–6%), with a wide prediction interval of 0–31% reflecting substantial heterogeneity (I²=99%).
The key finding was the subgroup difference between care settings: EPC unit patients had a pooled mortality of 2% (95% CI 1–4%) compared to 8% (95% CI 4–14%) in ICU patients (χ²=7.99; P<0.01).
The post-hoc comparative analysis of three studies directly comparing EPC units with ward care showed a mortality OR of 0.65 (95% CI 0.39–1.07) favouring EPC, which did not reach statistical significance.
Hospital length of stay
Pooled hLOS across six studies was 8.6 days (95% CI 5.9–11.3), though this was a descriptive finding without a comparator group and must be interpreted cautiously given the heterogeneous data.
Surgery cancellation
Heller et al. reported a marked reduction in surgery cancellation after establishing a postoperative surgical unit (34/843 vs 186/503), though this was a single uncontrolled observation.
Optimal care receipt
Turner et al. found mortality of 1.2% in patients who received their requested level of postoperative care versus 3.1% in those who did not (P<0.038).
Their conclusions were: "This review found an overall pooled mortality of 3 (95% CI, 2 to 6)% after extended postoperative recovery in noncardiac surgery, with lower mortality observed among patients managed in EPC units. However, considerable variability in the definitions, operational capacities and admission criteria warrants careful interpretation. This emphasises the need for standardisation and future research, while also acknowledging the inherent diversity of clinical practices across different healthcare settings."
Review two
Tran L, Stern C, Harford P, Ludbrook G, Whitehorn A. Effectiveness and Safety of Enhanced Postoperative Care Units for Noncardiac, Non-Neurological Surgery: A Systematic Review. A A Pract. 2025 Aug 5;19(8):e02008. doi: 10.1213/XAA.0000000000002008. PMID: 40762387.
30-day mortality — two quasi-experimental studies (Stahlschmidt et al., P=0.015; Wang et al., P=0.046) reported significant reductions favouring EPCU over standard ward care, though neither reported effect sizes with confidence intervals.
Days at home at 30 days — Ludbrook et al. (2023) found a mean difference of 1.74 additional days at home (95% CI 0.11–3.36; P=0.03) in favour of EPCU. No significant difference was found at 90 days.
Hospital length of stay — one study (Koning et al.) reported a significantly shorter stay in favour of EPCU (geometric mean ratio 0.77, 95% CI 0.66–0.91; P=0.002), though two other studies found the opposite (significantly longer stays in EPCU patients).
Unplanned ICU admissions — Swart et al. (2017) reported a striking and significant reduction (0/68 EPCU vs 22/139 usual care; P=0.00015).
Return to theatre — two quasi-experimental studies (Stahlschmidt et al. and Swart et al. 2017) reported significant reductions in unplanned surgical reintervention.
MET calls — two of three studies reporting this outcome (Costa-Pinto et al. and Stahlschmidt et al.) found significant reductions in post-discharge MET calls favouring EPCU.
Health economics — the Leaman & Ludbrook cost-effectiveness analysis demonstrated dominance, with 4.3 additional DAH at 90 days and a reduction in hospital costs of A$1,081 per patient (ICER −251).
All outcomes carried a GRADE certainty rating of very low, and the overall findings were mixed, with no benefit demonstrated across most studies for most outcomes.
Their conclusions were: "While the majority of included studies were assessed to be of satisfactory quality, there was significant clinical and methodological heterogeneity. However, while the current body of research is not yet comprehensive, it does suggest potential benefits of this model that warrant serious consideration and further exploration. Further research should prioritize the standardization of EPCUs, identify which specific factors are responsible for the observed positive effects and determine the patient cohorts most likely to benefit. Further expansion of EPCUs should be approached with caution, guided by careful evaluation."