Randomized clinical trials

Ghaneh P, Palmer D, Cicconi S, Jackson R, Halloran CM, Rawcliffe C et al, for the European Study Group for Pancreatic Cancer.

Lancet Gastro Hepatol 2023; 8: 157-168.

This study included 83 evaluable patients from 478 screened. Neoadjuvant treatment did not reduce the resection rate. Neoadjuvant treatment improved 1-year disease-free survival in both groups combined: 59 versus 39% for immediate surgery (hazard ratio 0.53, 95 per cent confidence interval 0.28 to 0.98, P=0.016.

Comment: The evidence for neoadjuvant chemotherapy is accumulating, but the number of suitable patients may be small.

Seifert L, von Renesse J, Seifert AM, Sturm D, Meisterfeld R, Rahbari NN et al.

BJS Open 2023; 7: zrac163.

Bile leak rates were similar (14.6 per cent interrupted versus 10.3 per cent continuous sutures) in this study that included 82 procedures.

Comment: No advantage in using interrupted sutures, which took longer.

Nakachi K, Ikeda M, Konishi M, Nomura Skatayama H, Kataoka T, et al  on behalf of the Hepatobiliary and Pancreatic Oncology Group of the Japan Clinical Oncology Group (JCOG-HBPOG).

Lancet 2023; 401: 195-203.

In this study that included 440 procedures, after a median of 45.4 months, adjuvant S-1 therapy given for four cycles improved overall survival from 67.6 to 77.1 per cent, P=0.008.

Comment: A benchmark study for this rare condition, that provides a real hope of improved survival.

Ishii K, Yokoyama Y, Yonekawa Y, Hayashi D, Kinoshita F, Kuwatsuka Y et al.

Br J Surg 2023; 110: 159–165.

This trial that included 218 procedures did not show any difference in intra-operative blood loss with (mean 695ml), or without tranexamic acid (667 ml). Other secondary outcomes were also similar.

Comment: Surprising, since tranexamic acid does seem to be beneficial in most operations with significant blood loss.

Merdrignac A, Garnier J, Dokmak S, Regenet N, Lesurtel M, Mabrut JY, et al.

Ann Surg 2022; 276: 769-775.

Use of the reinforced stapler actually increased the overall rate of pancreatic fistula compared to the standard device: 83 versus 67.7 per cent (P=0.021), in this study that included 199 procedures. The rates of clinically relevant fistula and other complications was similar.

Comment: Disappointing results.

Smits FJ, Henry AC, Besselink MG, Busch OR, van Eijck CH, Arntz M et al, for the Dutch Pancreatic Cancer Group.

Lancet 2022; 399: 1867-1875.

This multicentre study included 1748 pancreatic resections. Use of the algorithm reduced the composite risk of complications from 14 to 8 per cent, relative risk 0.48, 95 per cent confidence interval 0.38 to 0.61, P<0.0001.

Comment: There is no substitute for specialist care.

Van Braak WG, Ponten JEH, Loozen CS, Schots JPM, van Geloven AAW, Donkervoort SC et al.
Br J Surg 2022 109: 267-273.

Some 226 patients with mild to moderate cholecystitis were included. Use of antibiotics reduced the rate of all infections (7.1 versus 12.6 per cent), and in particular surgical site infections (5.3 versus 12.1 percent; P=0.01).

Comment: Antibiotics recommended.

Bressan AK, Isherwood S, Bathe OF, Dixon E, Sutherland FR, Ball C.
Ann Surg 2022; 275: 281-287.

A single preoperative dose of 500mg methylprednisolone reduced the rate of postoperative complications (31.2 versus 47.3 per cent, P=0.042), the rate of organ space site infections (6.5 versus 17.6 per cent, P=0.036) and hospital stay (8.9 versus 12.5 days, P=0.015) in this study that included 151 procedures.

Comment: This is an extraordinary result, and hard to explain on scientific grounds. Fertile ground for more research.

Hewitt, DB, Nissen N, Hatoum H, Musher B, Seng J, Coveler AL et al.
Ann Surg 2022; 275: 45-53.

In this study that included 303 patients, the addition of neoadjuvant immunotherapy did not improve progression-free survival (12. versus 13.4 months in controls) or overall survival (14.3 versus 14.9 months, respectively).

Comment: disappointing negative result.

Toyama H, Matsumoto I, Mizumoto T, Fujita H, Tsuchida S, Kanbara Y, et al.
Ann Surg 2021; 274: 935-944.

This multicentre trial included 212 procedures. The rate of delayed gastric emptying was similar: 15.6 per cent after retrocolic, and 12.6 per cent after antecolic reconstruction. The rates of other complications were also similar.

Comment: Hasn’t this been studied enough now?

Welsch T, Müssle B, Korn S, Sturm D, Bork U.
Br J Surgery 2022; 109: 37-45.

A total of 445 procedures were included. In the intention to treat analysis, there was no significant reduction in haemorrhage (2.9 versus 7.1 per cent, odds ratio 0.39, 95 per cent confidence interval 0.15 to 1.02, P=0.071). In the per protocol analysis the effect was more positive: odds ratio 0.26, 0.09 to 0.8, P=0.017.

Comment: The authors reasonably conclude falciform wrap should be considered during pancreaticoduodenectomy.