Long-term results following different extent of resection in chronic pancreatitis

To evaluate, whether the indication related varying extent of resection in chronic pancreatitis has a predictable impact on long-term outcome.

One hundred and twenty-six patients consecutively underwent surgery for chronic pancreatitis from March 1987 to September 1997. Three treatment categories were defined: Pancreatoduodenectomy, duodenum-preserving resection and drainage procedures, and left-sided pancreatectomy. Main outcome measures were late mortality, pain scores preoperatively and at follow-up, body-weight change, percentage of insulin dependent diabetes, patient's and physician's satisfaction with surgery.

Forty-one patients underwent pancreatoduodenectomy, 59 drainage procedures, and 26 left-sided pancreatectomy, respectively. Hospital mortality was 1 (2.4%), 4 (6.8%), and 1 (3.8%) (P = NS), totaling 4.8%. After an average follow-up of 5.2 years, late mortality was 10 (24.4%), 9 (15.3%), and 4 (15.4%) (P = NS) for a total of 23 (18.3%). Two patients (1.6%) died of unsuspected pancreatic cancer. Three patients (2.4%) had to be reoperated upon for pain relapse. The mean pain score was 8.8 preoperatively and 2.1 at late follow-up and not different among groups. Body-weight gain averaged 3.0, 4.0, and 3.4 kg, with no significant differences. Percentage of insulin dependency in all patients rose from 14% prior to surgery to 30% at reevaluation, and was very similar in all treatment categories.

The different kind and level of invasiveness of the surgical procedures did not significantly influence the late outcome. High rates of late mortality and deterioration of endocrine function are to a greater extent sequelae of comorbidity and the progression of the underlying pancreatic disease.