Time for a Changing of Guard: From Minimally Invasive Surgery to Endoscopic Drainage for Management of Pancreatic...

Endoscopic drainage (ED) with or without necrosectomy, and minimally invasive surgical necrosectomy (MISN) have been increasingly utilized for treatment of symptomatic sterile and infected pancreatic walled-off necrosis (WON). We conducted this systematic review to compare the safety of ED with MISN for management of WON.

We searched several databases from inception through November 9, 2017 to identify comparative studies evaluating the safety of ED versus MISN for management of WON. MISN could be performed using video-assisted retroperitoneal debridement or laparoscopy. We evaluated difference in mortality, major organ failure, adverse events, and length of hospital stay.

Six studies (2 randomized controlled trials and 4 observational studies) with 641 patients (326 ED and 315 MISN) were included in this meta-analysis. Rates of mortality for ED and MISN were 8.5% and 14.2%, respectively. Pooled odds ratio (OR) with 95% confidence interval was 0.59 (0.35-0.98), I=0% in favor of ED. On subgroup analysis: no difference in mortality was seen based on randomized controlled trials [OR, 0.65 (0.08-5.11)], while ED had improved survival in observational studies [OR, 0.49 (0.27-0.89)]. Development of new major organ failure rates after interventions were 12% and 54% for ED and MISN, respectively. Pooled OR was 0.12 (0.06-0.31), I=25% in favor of ED. For adverse events, pooled OR was 0.25 (0.10-0.67), I=70% in favor of ED. There was no difference in risk of bleeding [OR, 0.68 (0.44-1.05)], while ED was associated with a significantly lower rate of pancreatic fistula formation [OR, 0.20 (0.11-0.37)], I=0%. Length of stay was also lower with ED, pooled mean difference was -21.07 (-36.97 to -5.18) days.

When expertise is available, ED is the preferred invasive management strategy over MISN for management of WON as it is associated with lower mortality, risk of major organ failure, adverse events, and length of hospital stay.