• Koya Yasukawa, Akira Shimizu, Takahide Yokoyama, Koji Kubota, Tsuyoshi Notake, Hitoshi Seki, Akira Kobayashi, and Yuji Soejima.
• Division of Gastroenterological, Hepato-Biliary-Pancreatic, Transplantation, and Pediatric Surgery, Department of Surgery, Shinshu University School of Medicine, Matsumoto, Japan. Electronic address: kouyayasu@shinshu-u.ac.jp.
• J. Am. Coll. Surg. 2020 Dec 1; 231 (6): 658-669.

BackgroundNonalcoholic fatty liver disease (NAFLD) is a complication of pancreaticoduodenectomy (PD). We conducted a randomized clinical trial to determine if high-dose digestive enzymes prevented the development of NAFLD after PD.Study DesignThis parallel-group, nonblinded, multicenter study enrolled patients undergoing elective PD at Shinshu University School of Medicine, from June 2011 to April 2017. Patients were randomly assigned to receive normal-dose (Excelase: 3.0 g/day [Meiji Seika Pharma Holdings Co, Ltd]) or high-dose digestive enzyme treatment (Excelase: 3.0 g/day; Pancreatin [Tokyo Chemical Industry Co Ltd]: 3.0 g/day; Berizym [Kyowa Pharmaceutical Industry Co Ltd]: 3.0 g/day; and Toughmac-E [Ono Pharmaceutical Co, Ltd]: 3.0 g/day) within 1 week after surgery. Because patients in the control group switched interventions upon receiving a diagnosis of NAFLD, intention-to-treat analysis was used. The primary endpoint was incidence of NAFLD within 1 year, and the secondary endpoints were the incidences of NAFLD at 1, 3, 6, and 12 months and the rate of improvement in NAFLD with high-dose transfer in the control group. The secondary analysis comprised assessment of risk factors for the development of NAFLD.ResultsEighty-four patients were randomly assigned (42 per group), 80 of whom were finally analyzed (39 normal-dose, 41 high-dose). The incidence of NAFLD was significantly lower in the high-dose (8 of 41) compared with the normal-dose (25 of 39) patients (p < 0.001). Multivariate analysis identified normal-dose (odds ratio [OR] 14.65, p < 0.001), total protein ≤ 6.5g/dL (OR 9.01, p = 0.018), pre-albumin ≤ 22.0 mg/dL (OR 7.71, p = 0.018), and pancreatic function diagnostic test ≤ 70% (OR 6.66, p = 0.009) as independent risk factors. There were no adverse effects. The model was accurate (c-index = 0.92) and reliable (Hosmer-Lemeshow test p = 0.32).ConclusionsHigh-dose administration of digestive enzymes significantly reduced the onset of NAFLD after PD compared with normal-dose administration. Registration number: UMIN000005595 (http://www.umin.ac.jp/ctr/).Copyright © 2020 American College of Surgeons. Published by Elsevier Inc. All rights reserved.

19 keywords...

hide…