A Simple Algorithm for Drain Management after Pancreaticoduodenectomy

Abstract

Pancreatic fistula (PF) continues to be the Achilles’ heel of pancreaticoduodenectomy (PD) with both morbidity and mortality linked to its occurrence. The optimal drain management strategy after PD remains unclear. We evaluated drain amylase (DA) levels on postoperative Day (POD) 0 to 5 in 76 consecutive patients undergoing PD to determine the patterns associated with PF. Of these 76 patients, eight patients (11%) developed Grade A, B, or C PF by International Study Group of Pancreatic Fistula criteria. POD 1 DA levels correlated closely with PF rates when high (greater than 5000 U/L, 100% PF rate) and low (less than 100 U/L, 2% PF rate). In patients with intermediate POD 1 DA (100 to 5000 U/L), 42 and 74 per cent had low DA levels on POD 3 and 5, respectively, and the PF rate was four of 31 (13%). Overall, the temporal pattern of decreasing DA levels after PD correlates closely with the risk of PF, and only two patients (5%) developed PF after early DA levels had normalized. Based on these data, we propose an algorithm of monitoring DA daily with drain removal when the level is less than 100 U/L. In our patient group drain removal would have occurred on a mean of 1.8 days and median 1 day after surgery.

Predictive value of drain pancreatic amylase concentration for postoperative pancreatic fistula on postoperative day 1 after pancreatic resection

“…In Fig. 3 , the Fagan plot in DPA1 for overall POPF [ 14 , 16 – 18 , 20 , 22 – 28 ] implied that when the pretest probabilities were respectively 25%, 50%, 75% the positive post-test probabilities (post-test (+)) were 0.59, 0.81, 0.93 and the negative post-test probabilities (post-test (−)) were 0.06, 0.16, 0.36. For CR-POPF, [ 5 , 15 , 18 , 19 , 21 , 29 ] when the pretest probabilities were respectively 25%, 50%, 75% the post-tests (+) were 0.66, 0.86, 0.95, and the post-tests (−) were 0.10, 0.26, 0.51.…”

Meta-analysis of drain amylase content on postoperative day 1 as a predictor of pancreatic fistula following pancreatic resection

“…The full text of each of the remaining 49 articles was examined in more detail. This led to the exclusion of a further 36 studies, leaving 13 that were subsequently included in the meta‐analysis ( Fig . ).…”

“…All studies graded POPF using the ISGPF criteria. Drain amylase levels were evaluated as a predictor of POPF of any grade in seven studies, as a predictor of clinically relevant POPF in five studies and as a predictor of both in one study. The prevalence of POPF of any grade ranged from 0·11 to 0·43, and the prevalence of clinically relevant POPF ranged from 0·03 to 0·39.…”

“…Four studies were retrospective. The QUADAS‐2 instrument highlighted concerns of bias in the index test category in seven studies as their threshold was not prespecified ( Fig. S1 , supporting information).…”

“…Several preoperative, intraoperative and postoperative risk factors have been analysed previously and found to be associated with POPF. A number of recent studies have emphasized the importance of early postoperative drain amylase values, by evaluating the accuracy of this simple test in predicting POPF.…”