Enhanced Recovery Protocols Reduce Mortality Across Eight Surgical Specialties at Academic and University-affiliated Community Hospitals

OBJECTIVE: To determine if implementation of a simplified ERP across multiple surgical specialties in different hospitals is associated with improved short and long-term mortality. Secondary aims were to examine ERP effect on length of stay, 30-day readmission, discharge disposition, and complications. SUMMARY BACKGROUND DATA: Enhanced recovery after surgery and various derivative ERPs have been successfully implemented. These protocols typically include elaborate sets of multimodal and multidisciplinary approaches, which can make implementation challenging or are variable across different specialties. Few studies have shown if a simplified version of ERP implemented across multiple surgical specialties can improve clinical outcomes. METHODS: A simplified ERP with 7 key domains (minimally invasive surgical approach when feasible, pre-/intra-operative multimodal analgesia, postoperative multimodal analgesia, postoperative nausea and vomiting prophylaxis, early diet advancement, early ambulation, and early removal of urinary catheter) was implemented in 5 academic and community hospitals within a single health system. Patients who underwent nonemergent, major orthopedic or abdominal surgery including hip/knee replacement, hepatobiliary, colorectal, gynecology oncology, bariatric, general, and urological surgery were included. Propensity-matched, retrospective case-control analysis was performed on all eligible surgical patients between 2014 and 2017 after ERP implementation or in the 12 months preceding ERP implementation (control population). RESULTS: A total of 9492 patients (5185 ERP and 4307 controls) underwent ERP eligible surgery during the study period. Three thousand three hundred sixty-seven ERP patients were matched by surgical specialty and hospital site to control non-ERP patients. Short and long-term mortality was improved in ERP patients: 30 day: ERP 0.2% versus control 0.6% ( P = 0.002); 1-year: ERP 3.9% versus control 5.1% ( P < 0.0001); 2-year: ERP 6.2% versus control 9.0% ( P < 0.0001). Length of stay was significantly lower in ERP patients (ERP: 3.9 ± 3.8 days; control: 4.8 ± 5.0 days, P < 0.0001). ERP patients were also less likely to be discharged to a facility (ERP: 11.3%; control: 14.8%, P < 0.0001). There was no significant difference for 30-day readmission. All complications except venous thromboembolism were significantly reduced in the ERP population (P < 0.02). CONCLUSIONS: A simplified ERP can uniformly be implemented across multiple surgical specialties and hospital types. ERPs improve short and long-term mortality, clinical outcomes, length of stay, and discharge disposition to home.