Enhanced recovery after surgery (ERAS) protocols are patient-centred, interprofessional perioperative care models aimed at reducing hospital stays, complications, healthcare costs, and improving patient satisfaction. Evidence for ERAS in cardiac surgery is scarce; therefore, we aimed to assess its efficacy and safety in minimally invasive heart valve surgery (MIHS).

Interdisciplinary Perioperative Care in Minimally Invasive Heart Valve Surgery (INCREASE) is a randomized-controlled trial that compares ERAS with standard care in MIHS. The intervention group (IG) received ERAS-based treatment; the control group (CG) standard care. Co-primary end-points were in-hospital stay for cardiovascular reasons within 1 year postoperatively (superiority hypothesis) and physical performance (6-minute walk test [6MWT]) at discharge (non-inferiority hypothesis). Superiority hypothesis was tested primarily in surgically treated patients (modified intention to treat population), while non-inferiority hypothesis was tested in per protocol population.

In total, 201 patients scheduled for MIHS were randomized (IG: 101; CG: 100). Total weighted in-hospital stay within 1 year was non-significantly shorter in the IG (adjusted mean difference −2.7 days, 95% confidence interval (CI) (−6.3, 0.9), 1-sided P = .07). Sensitivity analyses favoured the IG (adjusted ratio 0.71, 95% CI, 0.6, 0.85). For 6MWT, adjusted mean difference was −14%, 95% CI (−43%, 15%) exceeding the non-inferiority margin of −15% (1-sided P = .48). Without baseline adjustment, 6MWT values revealed a mean difference between groups of −1%, 95% CI (−10%, 8%), reaching non-inferiority of the IG. Safety outcomes were comparable with the exception of an unexpected higher rate of post-operative atrial fibrillation in patients randomized to ERAS.

ERAS is feasible and safe in MIHS. The trial did not demonstrate a reduction in length of hospital stay during the first postoperative year in the population studied. Exploratory analyses suggest that further evaluation of ERAS concepts is justified.