The association of length of hospital stay with readmission after elective pancreatic resection

Objective: The aim of this study was to identify if prolonged length of hospital stay was protective for certain post-discharge complications requiring readmission after pancreatectomy. Methods: Patients undergoing elective pancreatectomy from 2012 to 2013 were identified in the American College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP). Multivariate regression analyses were conducted to elucidate the association of length of hospital stay (LOS), in-hospital complications, and post-discharge complications with 30-day hospital readmission. Results: Mean LOS for the readmitted versus not readmitted cohort was 9.01 and 10.36, respectively (P<0.001). Approximately half of hospital readmissions occur within 7 days of discharge from the hospital. The readmission rate (after hospital discharge) for patients with any in-hospital complication was 13.2%. On the other hand, the readmission rate for patients with any post-discharge complication was 70.8%. Readmission rates were above 85% for patients with the following post-discharge complications: organ/space surgical site infection (SSI), pneumonia, ventilator dependence greater than 48 h, progressive renal insufficiency, sepsis, and septic shock. In a multivariate logistic regression model, prolonged LOS reduced the odds of any post-discharge complication requiring readmission (OR=0.68, P=0.01). Specifically, prolonged LOS reduced the odds of a post-discharge organ/space SSI requiring readmission (OR=0.72, P=0.02). Conclusion: Readmission after pancreatectomy primarily occurs due to a new post-discharge event. Furthermore, increased LOS is protective for readmission for post-pancreatectomy complications, particularly those due to post-discharge organ space SSIs. Our findings suggest that solely focusing on reducing LOS in pancreatectomy may lead to the unintended consequence of increasing readmission rates.