Whipple ERAS: Identifying patients that fall "Off pathway"

Abstract

Presenter: Carlos Ayala MD, PhD | Stanford University Background: Pancreaticoduodenectomy (PD) is a complex surgical procedure that requires attentive perioperative care. Enhanced Recovery After Surgery (ERAS) protocols have emerged to enhance functional recovery, decrease complications, and reduce hospital length of stay (LOS). We successfully implemented an ERAS pathway for PD to optimize patient care; however, our current mean LOS is still over the proposed ERAS target of 7 days. This study focuses on detailing those patients that fell “off pathway” to determine causes of failure and areas of ERAS improvement. Methods: Four surgeons worked closely with the hospital quality team to develop our own institutional ERAS pathway for pancreatic resections. The ERAS pathway for PD was implemented in 2017. This study period includes all PD procedures that utilized the ERAS pathway over 2 years (May 2017 - May 2019). Robotic PD procedures were excluded. We specifically reviewed all medical records pertaining to those patients that failed to meet target LOS. Results: A total of 142 PD procedures utilized the ERAS pathway over two years. We observed a LOS reduction from pre-ERAS vs post-implementation (mean LOS 10.5 vs 8.4 days, p<0.05); but still failed to achieve our target LOS for the whole cohort. There were 65 patients (46%) that failed to meet the 7 day LOS. The most common reason to fail pathway LOS was for either ileus or delayed gastric emptying (36/65, 55%), leading to a longer LOS of 12 days. Within this subgroup, patients who required a nasogastric tube during any time of admission had longer LOS (13 vs 9 days, p<0.05). Additional reasons for the remaining “off pathway” patients (29/65, 45%) not meeting target LOS are in Table 1. These primarily included non-gastrointestinal reasons for stay with the top three including work-up based on leukocytosis and/or concern for pancreatic leak (9/29, LOS 12), desire for additional “night” of observation (7/29, LOS 8), and orthostatic hypotension (3/29, LOS 9). Of these additional 29 patients, 9 patients underwent computed tomography (on or after POD 7) and only 2 patients received an inpatient intervention during the extra LOS (1 drain study, 1 percutaneous drain manipulation). ERAS implementation did not increase perioperative complications, 30-day readmission, or mortality. Renal, respiratory, cardiac, and superficial site infections all remained below NSQIP benchmarks. Conclusion: The most common reasons for PD pathway failure included slow return of gastric or bowel function which are perhaps inevitable in some patients undergoing PD. The remaining patients not meeting ERAS target were often kept for observation without additional intervention. This group represents an actionable cohort to target for improving LOS through surgeon awareness rather than pathway modification.