Endoscopic transluminal necrosectomy in necrotising pancreatitis: a systematic review

We performed a systematic review to assess the outcome of endoscopic transluminal necrosectomy in necrotising pancreatitis with additional focus on indication, disease severity, and methodological quality of studies. We searched the literature published between January 2005 and June 2013. Cohorts, including patients with (infected) necrotising pancreatitis, undergoing endoscopic necrosectomy were included. Indication, disease severity, and methodological quality were described. The main outcomes were mortality, major complications, number of endoscopic sessions, and definitive successful treatment with endoscopic necrosectomy alone. After screening 581 papers, 14 studies, including 455 patients, fulfilled the eligibility criteria. All included studies were retrospective analyses except for one randomized, controlled trial. Overall methodological quality was moderate to low (mean 5, range 2-9). Less than 50 % of studies reported on pre-procedural severity of disease: mean APACHE-II score before intervention was 8; organ failure was present in 23 % of patients; and infected necrosis in 57 % of patients. On average, four (range 1-23) endoscopic interventions were performed per patient. With endoscopic necrosectomy alone, definitive successful treatment was achieved in 81 % of patients. Mortality was 6 % (28/460 patients) and complications occurred in 36 % of patients. Bleeding was the most common complication. Endoscopic transluminal necrosectomy is an effective treatment for the majority of patients with necrotising pancreatitis with acceptable mortality and complication rates. It should be noted that methodological quality of the available studies is limited and that the combined patient population of endoscopically treated patients is only moderately ill.