Prophylactic abdominal drainage for pancreatic surgery.

Rationale: This is the fourth update of a Cochrane review first published in 2015 and last updated in 2021. The use of surgical drains is a very common practice after pancreatic surgery. The role of prophylactic abdominal drainage to reduce postoperative complications after pancreatic surgery is controversial.

Objectives: To assess the benefits and harms of routine abdominal drainage after pancreatic surgery; to compare the effects of different types of surgical drains; and to evaluate the optimal time for drain removal.

Search methods: We searched CENTRAL, MEDLINE, three other databases, and five trials registers, together with reference checking and contact with study authors, to identify studies for inclusion in the review. The search dates were 20 April 2024 and 20 July 2024.

Eligibility criteria: We included randomised controlled trials (RCTs) in participants undergoing pancreatic surgery comparing (1) drain use versus no drain use, (2) different types of drains, or (3) different schedules for drain removal. We excluded quasi-randomised and non-randomised studies.

Outcomes: Our critical outcomes were 30-day mortality, 90-day mortality, intra-abdominal infection, wound infection, and drain-related complications.

Risk of bias: We used the Cochrane RoB 1 tool to assess the risk of bias in RCTs.

Synthesis methods: We synthesised the results for each outcome using meta-analysis with the random-effects model where possible. We used GRADE to assess the certainty of evidence for each outcome.

Included studies: We included 12 RCTs with a total of 2550 participants. The studies were conducted in North America, Europe, and Asia and were published between 2001 and 2024. All studies were at overall high risk of bias.

Synthesis of results: We considered the certainty of the evidence for intra-abdominal infection for the comparison of early versus late drain removal following pancreaticoduodenectomy to be moderate, downgraded due to indirectness. We considered the certainty of the evidence for the other outcomes to be low or very low, mainly downgraded due to high risk of bias, inconsistency, indirectness, and imprecision. Drain use versus no drain use following pancreaticoduodenectomy We included two RCTs with 532 participants randomised to the drainage group (N = 270) and the no drainage group (N = 262) after pancreaticoduodenectomy. The evidence is very uncertain about the effect of drain use on 30-day mortality (risk ratio (RR) 0.49, 95% confidence interval (CI) 0.07 to 3.66; 2 studies, 532 participants), 90-day mortality (RR 0.25, 95% CI 0.06 to 1.15; 1 study, 137 participants), intra-abdominal infection rate (RR 0.85, 95% CI 0.21 to 3.51; 2 studies, 532 participants), and wound infection rate (RR 0.85, 95% CI 0.55 to 1.31; 2 studies, 532 participants) compared with no drain use. Neither study reported on drain-related complications. Drain use versus no drain use following distal pancreatectomy We included two RCTs with 626 participants randomised to the drainage group (N = 318) and the no drainage group (N = 308) after distal pancreatectomy. There were no deaths at 30 days in either group. The evidence is very uncertain about the effect of drain use on 90-day mortality (RR 0.16, 95% CI 0.02 to 1.35; 2 studies, 626 participants), intra-abdominal infection rate (RR 1.20, 95% CI 0.60 to 2.42; 1 study, 344 participants), and wound infection rate (RR 2.12, 95% CI 0.93 to 4.87; 2 studies, 626 participants) compared with no drain use. Neither study reported on drain-related complications. Active versus passive drain following pancreaticoduodenectomy We included three RCTs with 441 participants randomised to the active drain group (N = 222) and the passive drain group (N = 219) after pancreaticoduodenectomy. The evidence is very uncertain about the effect of an active drain on 30-day mortality (RR 1.24, 95% CI 0.30 to 5.07; 2 studies, 321 participants), intra-abdominal infection rate (RR 0.58, 95% CI 0.06 to 5.43; 3 studies, 441 participants), and wound infection rate (RR 0.92, 95% CI 0.44 to 1.90; 2 studies, 321 participants) compared with a passive drain. None of the studies reported on 90-day mortality. There were no drain-related complications in either group (1 study, 161 participants; very low-certainty evidence). Early versus late drain removal following pancreaticoduodenectomy We included three RCTs with 557 participants with a low risk of postoperative pancreatic fistula, randomised to the early drain removal group (N = 279) and the late drain removal group (N = 278) after pancreaticoduodenectomy. Low-certainty evidence suggests that early drain removal may result in little to no difference in 30-day mortality (RR 0.99, 95% CI 0.06 to 15.45; 3 studies, 557 participants) and wound infection rate (RR 1.07, 95% CI 0.47 to 2.46; 3 studies, 557 participants) compared with late drain removal. Moderate-certainty evidence shows that early drain removal probably results in a slight reduction in intra-abdominal infection rate compared with late drain removal (RR 0.45, 95% CI 0.26 to 0.79; 3 studies, 557 participants). Approximately 58 (34 to 102 participants) out of 1000 participants in the early removal group developed intra-abdominal infections compared with 129 out of 1000 participants in the late removal group. There were no deaths at 90 days in either study group (2 studies, 416 participants). None of the studies reported on drain-related complications.

Authors' conclusions: The evidence is very uncertain about the effect of drain use compared with no drain use on 90-day mortality, intra-abdominal infection rate, and wound infection rate in people undergoing either pancreaticoduodenectomy or distal pancreatectomy. The evidence is also very uncertain whether an active drain is superior, equivalent, or inferior to a passive drain following pancreaticoduodenectomy. Moderate-certainty evidence suggests that early drain removal is probably superior to late drain removal in terms of intra-abdominal infection rate following pancreaticoduodenectomy for people with low risk of postoperative pancreatic fistula.

Funding: None.

Registration: Registration: not available. Protocol and previous versions available via doi.org/10.1002/14651858.CD010583, doi.org/10.1002/14651858.CD010583.pub2, doi.org/10.1002/14651858.CD010583.pub3, doi.org/10.1002/14651858.CD010583.pub4, and doi.org/10.1002/14651858.CD010583.pub5.