Association of preoperative biliary drainage technique with postoperative outcomes among patients with resectable hepatobiliary malignancy

• Authors: Q. Lina Hu, American College of Surgeons; Jason B. Liu, American College of Surgeons; Ryan J. Ellis, American College of Surgeons; Jessica Y. Liu, Emory University; Anthony D. Yang, Northwestern University Feinberg School of Medicine; Michael I. D'Angelica, Memorial Sloan-Kettering Cancer Center; Clifford Y. Ko, American College of Surgeons; Ryan P. Merkow, American College of Surgeons
• Journal Title: HPB
• Volume: Volume 22, Number 2
• Publisher: Oxford | 2020-02-01, Pages 249-257
• Type of Work: Article | Post-print: After Peer Review
• Publisher DOI: http://doi.org/10.1016/j.hpb.2019.06.011
• Subjects: Health Sciences, Medicine and Surgery; Health Sciences, Health Care Management; Health Sciences, Oncology
• Keywords: Aged; Biliary Tract Neoplasms; Cholestasis; Databases, Factual; Disease-Free Survival; Drainage; Endoscopy; Female; Humans; Liver Neoplasms; Logistic Models; Male; Middle Aged; Patient Selection; Postoperative Complications; Propensity Score; Quality Improvement; Retrospective Studies; Stents; Survival Rate

Abstract

Background: Endoscopic biliary stenting (EBS) and percutaneous transhepatic biliary drainage (PTBD) are two techniques used for preoperative biliary drainage prior to hepatobiliary resection. The objectives of this study were to determine predictors of the drainage technique selection and to evaluate the association between drainage technique and postoperative outcomes. Methods: Using ACS NSQIP data (2014–2017), patients who underwent preoperative biliary drainage prior to hepatobiliary resection for malignancy were identified. Separate multivariable-adjusted, propensity score (PS) adjusted, and PS matched logistic regression models were constructed to evaluate the association between drainage technique and postoperative outcomes. Results: Of 527 patients identified, 431 (81.8%) received EBS and 96 (18.2%) received PTBD. Patients who underwent PTBD had more preoperative co-morbidities, including higher ASA class, recent weight loss, and hypoalbuminemia (all p < 0.05). After multivariable adjustment, PTBD was significantly associated with 30-day DSM (OR 1.92, 95% CI 1.24–2.97, p = 0.004), overall SSI (OR 1.74, 95% CI 1.10–2.76, p = 0.019), and superficial SSI (OR 2.08, 95% CI 1.20–3.60, p = 0.010). These findings remained significant for both PS-adjusted and PS-matched models. Conclusion: Patients undergoing hepatobiliary resection selected for PTBD had significantly more preoperative co-morbidities and nutritional deficits. Compared to EBS, PTBD was associated with significantly higher odds of postoperative morbidity and mortality.