Immediate drainage of infected necrotizing pancreatitis within 24 hours of diagnosis did not lead to significantly fewer complications compared to delayed drainage, according to a small randomized controlled trial.
In an intention-to-treat analysis of 104 patients, the mean Complete index of complications (CCI) over 6 months was similar for those who received either immediate drainage within 24 hours of diagnosis or delayed drainage to walled necrosis (57 vs. 58 out of 100, respectively), reported Marc G. Besselink, MD, PhD, University of Amsterdam in the Netherlands, and colleagues.
However, patients in the immediate drainage group underwent more invasive procedures, including necrosectomy and catheter drainage, than those in the delayed drainage group (mean 4.4 versus 2.6 procedures), but had similar mortality rates at 6 months (13% vs. 10%, respectively; RR 1.25, 95% CI 0.42-3.68), the authors wrote in the New England Journal of Medicine.
Interestingly, 39% of the patients in the delayed drainage group were treated with antibiotics instead of drainage, and 17 out of 19 survived, they noted.
“These results suggest that an initial conservative approach with antibiotics is warranted when infected necrosis is diagnosed,” the authors wrote.
The current standard for the treatment of infected necrotizing pancreatitis involves a gradual approach, starting with catheter drainage, the authors said, an approach supported by recent American Association of Gastroenterology guidelines “even in the early stages of the disease”.
However, Besselink’s group noted that it was not known whether immediate catheter drainage could improve patient outcomes.
“As the present study shows, it is best to delay non-operative drainage in clinically stable patients until the development of walled necrosis, which typically occurs 30 days or more after the onset of pancreatitis,” Todd wrote H. Baron, MD, of the University of North Carolina at Chapel Hill, in a accompanying editorial.
“Differentiating infected necrosis from sterile necrosis with a concomitant continuous systemic inflammatory response during the first few weeks after onset of pancreatitis can be difficult, but there are established criteria to guide clinicians,” noted Baron. “The distinction between infected and sterile necrosis is essential because infected necrosis is associated with significantly higher mortality, requires the initiation of antibiotics that enter pancreatic tissue (appropriate to the culture data available) and often results in intervention. percutaneous, endoscopic or surgical (alone or in combination). “
The NEEDLE (Delayed or Immediate Drainage of Infected Necrotizing Pancreatitis) focused on 22 centers and included patients with acute pancreatitis who developed infected necrotizing pancreatitis and who could benefit from endoscopic transluminal drainage or percutaneous drainage guided by image within 35 days of symptom onset.
Besselink and colleagues evaluated 104 patients from August 2015 to October 2019 and randomized them 1: 1 to receive immediate drainage (n = 55) or delayed drainage (n = 49) after walled necrosis.
The infected necrosis was confirmed by the presence of gas in the pancreatic and peripancreatic necrosis on a CT scan with contrast media injection, or a positive culture by fine needle aspiration or a positive Gram stain within 14 days of l onset of acute pancreatitis.
The primary endpoint of the study was the CCI score of randomization at 6 months, with follow-up at 3 and 6 months. Clavien-Dindo standings was used to classify complications.
The patients had a mean age of 59 years and 58% were men. About two-thirds had gallstones as the cause of their pancreatitis.
On average, immediate catheter drainage occurred 24 days after symptom onset, while delayed drainage occurred 34 days after symptoms. Fifty-one of 55 patients in the immediate drainage group underwent drainage within 24 hours of randomization. Pancreatic and peripancreatic necrosis was “largely or completely encapsulated” in 60% of patients in the immediate group and 70% in the delayed group, the authors reported.
No significant difference occurred in the incidence of major complications between the immediate and delayed groups, respectively, including new organ failure (25% vs. 22%), visceral perforation, or enterocutaneous fistula ( 9% vs. 8%), bleeding (15% vs. 20%), pancreatic-cutaneous fistula (11% vs. 8%) or wound infection (0% vs. 1%).
The average length of hospital stay was 59 days in the immediate group versus 51 days in the delayed group, and the length of ICU stay did not differ between groups (12 days for each group), noted the authors.
The analysis had several limitations, the researchers recognized, including the small sample size and the fact that CCI is only designed to assess postoperative complications. In addition, the trial allowed both intensifying and endoscopic surgical approaches, although the endoscopic route “has gradually become the preferred treatment strategy,” the authors noted. In addition, not all necrotic collections could be reached endoscopically.
Funding was provided by the Dutch government, Amsterdam UMC-University of Amsterdam and the NutsOhra Fund.
Besselink did not report any conflicts of interest. Some co-authors reported relationships with industry.
Baron reported affiliations with Cook Endoscopy, Olympus, Boston Scientific, Ambu, WL Gore, and Medtronic.