Epidural analgesia for hepatopancreatobiliary operations and postoperative urinary tract infections: an unrecognized association of "best-practices" and adverse outcomes

Background: Thoracic epidural analgesia (TEA) is considered "best-practices" for pain-control following HPB operations. It is unknown if TEA increases the risk of UTI. We sought to examine the association of TEA and UTI following HPB operations. Methods: A retrospective cohort study of patients undergoing elective HPB operations was performed (ACS-NSQIP [2014-2016]). Patients were categorized by TEA utilization. The primary outcome was UTI. Multivariable logistic regression models were created to examine the association of TEA with UTI; including sensitivity and interaction analyses for age and gender. Results: Among 28,571 patients included, 5764 (20.2%) had TEA. UTI occurred more frequently with TEA (3.5% vs. 2.2%, p < 0.01). After multivariable analysis, TEA was associated with increased risk of UTI (1.59 [1.34-1.89]); when stratified by age and gender, the association persisted with an incremental increased risk observed in males over 70 years (1.91 [1.41-2.59]). UTI was associated with increased risk of sepsis (16.8% vs. 5.6%, P < 0.001), LOS (9 versus 6 days, P < 0.001) and readmission rates (21.4% vs. 12.3%, P < 0.001). Conclusion: Despite TEA recommended as a best-practice standard for HPB operations, the increased risk of UTI calls for evaluation of current practices and consideration of alternative strategies for high-risk vulnerable populations - elderly males.