Early versus late necrosectomy in severe necrotizing pancreatitis.

Debate as to whether surgery in severe necrotizing pancreatitis (SNP) should be done early or late has been present ever since the disease was described. There are no prospective, randomized studies addressing this specific issue. Patients with SNP, documented clinically, with Ranson's criteria, and dynamic pancreatography (DP) findings were randomly allocated in two groups for treatment. Group A included early necrosectomy (within 48 to 72 hours of onset) and group B, late necrosectomy (at least 12 days after onset). Both groups continued with open packing and staged necrosectomies. Cultures were obtained at each laparotomy and necrosis was verified histologically in all instances. During a 36-month study period, 150 patients with unequivocal acute pancreatitis were admitted for treatment. Forty-one with SNP initially entered the study; there were 5 drop outs. Patients in group A (25) and group B (11) had no difference in distribution by gender or mean age, etiology, mean Ranson's signs (4 versus 3.8), DP findings, rate of infected necrosis, or necrosectomies required per patient. Although the mortality rate (58% versus 27%) did not reach statistical significance, the odds ratio for mortality was 3.4 times higher in group A, which made us finish the study. This prospective, randomized study from a single institution clearly demonstrates that early intensive conservative treatment with late necrosectomy for selected cases is the current rationale approach for SNP.