• Koichi Taniguchi, Ryusei Matsuyama, Yasuhiro Yabushita, Yuki Homma, Yohei Ota, Ryutaro Mori, Daisuke Morioka, and Itaru Endo.
• Department of Gastroenterological Surgery, Yokohama City University Graduate School of Medicine, Yokohama, Japan.
• J Hepatobiliary Pancreat Sci. 2020 Dec 1; 27 (12): 950-961.

Background/PurposeEarly drain removal (EDR) based on drain fluid amylase level (DFA) after pancreaticoduodenectomy excluded 15%-40% patients from EDR because of inappropriate DFA.MethodsOf 198 pancreatoduodenectomy cases, we used the first 105 cases as an exploration cohort to construct the optimal criteria for EDR on postoperative day (POD)4 that were applied to the subsequent 93 cases used as the validation cohort. After that, we examined another 142 patients to further assess the efficacy of the new EDR criteria.ResultsOf the four independent predictors of clinically relevant postoperative pancreatic fistula (CR-POPF) ([1] soft pancreas, [2] positive drain fluid culture on POD1, and [3] serum C-reactive protein [CRP] ≥13 mg/dL on POD4) in the exploration cohort, EDR was applied to cases in the validation cohort meeting the [2] and/or [3], enabling 96% (89/93) applicability of EDR. Outcomes were improved in the validation cohort compared to the exploration cohort; CR-POPF: 8.6% vs 25.7%, P = .005; Dindo-Clavien grade ≥ 3 complications: 23.7% vs 41.9%, P = .007; and median hospital stay (day): 21 vs 27, P = .005. The subsequent 142 patients showed 92% (131/142) applicability of EDR and 5.6% (8/142) incidence of CR-POPF.ConclusionsOur new criteria for EDR, without DFA, enabled ≥ 90% applicability of EDR and reduced CR-POPF.© 2020 Japanese Society of Hepato-Biliary-Pancreatic Surgery.

16 keywords...

hide…