In a Dutch nationwide trial reported by Smits et al in The Lancet, researchers in the Dutch Pancreatic Cancer Group found that algorithm-based early postsurgical care was associated with a reduced risk of the composite outcome of bleeding requiring invasive intervention, organ failure, and 90-day mortality vs usual care in patients who had undergone pancreatic resection.

The open-label trial included all eligible patients undergoing pancreatic resection in the Netherlands (17 centers) during a 22-month period between January 2018 and November 2019. A total of 1,748 patients were randomly assigned to algorithm-based care (n = 863) or usual care (n = 885). A total of 75% of patients in each group underwent resection for pancreatic cancers; 4% to 5% underwent resection for chronic pancreatitis, and 20% to 21% underwent resection for other conditions. The algorithm determined when to perform abdominal computed tomography and radiologic drainage, start antibiotic treatment, and remove abdominal drains. The algorithm was incorporated into a smartphone app that included daily evaluation of clinical and biochemical markers. Daily evaluation using the algorithm was conducted from postoperative day 3 to postoperative day 14. The primary outcome measure was a composite of bleeding that required invasive intervention, organ failure, and 90-day mortality.

Key Findings

The primary outcome occurred in 73 (8%) of 863 patients in the algorithm-based care group vs 124 (14%) of 885 patients in the usual care group (adjusted risk ratio [RR] = 0.48, 95% confidence interval [CI] = 0.38–0.61, P < .0001).

Significant improvements with algorithm-based care were observed in each of the components of bleeding that required intervention (5% vs 6%, RR = 0.65, 95% CI = 0.42–0.99, P = .046), organ failure (5% vs 10%, RR = 0.35, 95% CI = 0.20–0.60, P = .0001), and 90-day mortality (3% vs 5%, RR = 0.42, 95% CI = 0.19–0.92, P = .029).

Significant reductions in the primary outcome in the algorithm-based care group were observed in both low/medium-volume centers (25 [9%] of 291 patients vs 42 [14%] of 294 patients; RR = 0.49, 95% CI = 0.25–0.68) and in high-volume centers (48 [8%] of 572 patients vs 82 [14%] of 591 patients, RR = 0.46, 95% CI = 0.32–0.66). The outcomes included a nonsignificant reduction in 90-day mortality in both low/medium-volume centers (3% vs 7%, RR = 0.35, 95% CI = 0.11–1.16) and high-volume centers (3% vs 4%, RR = 0.48, 95% CI = 0.18–1.32).

The investigators concluded, “The algorithm for the early recognition and minimally invasive management of complications after pancreatic resection considerably improved clinical outcomes compared with usual care. This difference included an approximate 50% reduction in mortality at 90 days.”

Hjalmar C. van Santvoort, PhD, of the Regional Academic Cancer Centre Utrecht, St Antonius Hospital Nieuwegein, and University Medical Centre Utrecht, is the corresponding author for The Lancet article.

Disclosure: The study was funded by the Dutch Cancer Society and University Medical Centre Utrecht. For full disclosures of the study authors, visit thelancet.com.