Early and long-term results after necrosectomy for necrotizing pancreatitis

Abstract

Abstract Background Necrotizing pancreatitis is a difficult clinical problem associated with significant mortality and morbidity rates. Indications for surgical intervention, and the timing and type of intervention, remain controversial issues. A 10-year experience in the management of this condition in a tertiary referral centre is reviewed; treatment policies and early and long-term results are reported. Methods Between 1987 and 1997, 39 consecutive patients (32 men, seven women), with a mean age of 47 (range 13–74) years, underwent necrosectomy for severe necrotizing pancreatitis. Diagnosis was confirmed by contrast-enhanced computed tomography. Twenty-nine patients were referred from other hospitals in Northern Ireland. The aetiology included gallstones (15), alcohol (ten), idiopathic (seven), surgery (two), endoscopic retrograde cholangiopancreatography (two), blunt trauma (one), hyperlipidaemia (one) and pancreas divisum (one). Mean Acute Physiology And Chronic Health Evaluation (APACHE) II score on admission or arrival was 2 (range 5–21). Indications for surgical intervention were clinical deterioration with development of multiple organ failure, despite maximal medical support in the intensive care unit (ICU), or positive fine-needle aspiration for micro-organisms or radiographic evidence of infected necrosis (i.e. presence of gas within peripancreatic collection). The mean interval from onset to intervention was 23 (range 9–47) days. In 13 patients a conventional approach (necrosectomy with primary closure over drains) was used, 14 patients underwent planned staged necrosectomy with delayed closure over drains and in 12 patients necrosectomy with open laparostomy was undertaken. Results There were five deaths and 14 patients experienced significant hospital morbidity (fistula, ten; pseudocyst, two; renal failure, two). Variables which correlated with mortality rate were higher APACHE II score, acute renal failure requiring dialysis and early intervention. There was no difference in either mortality or morbidity rate between infected and sterile necrosis, or between the three surgical techniques. Long-term morbidity occurred in 20 surviving patients (incisional hernia, ten; endocrine pancreatic insufficiency, nine; exocrine pancreatic insufficiency, two; pseudocyst, two; chronic renal failure, two; recurrent pancreatitis, one; and chronic pain, one). Conclusion A low mortality rate can be achieved in patients with severe necrotizing pancreatitis with aggressive surgical intervention. Early maximal medical treatment and ICU support is essential before intervention, which should preferably be avoided early in the course of the disease and should be based on well defined criteria. The long-term morbidity rate remains high, although not always severe. This emphasizes the need for close follow-up.