Reappraisal of a method of reconstruction after pancreatoduodenectomy

After pancreatoduodenectomy (PD), pancreatic leak and the functional pancreatic and gastrointestinal disorders are the most important complications. Still there is no single method which takes care of all of them. After identifying the various reasons behind these complications, the senior author started performing the present method in the 1980s. Since then we have been able to bring these complications to a very low incidence. The present study is designed to substantiate the claims of various advantages of this method of PD and reconstruction and to explain the rationale behind this method.

This is a retrospective study of 225 consecutive PD procedures done for periampullary cancers, during the period of 1993-2004. The shortterm and long-term results assessed were mortality rate, morbidity rate, early reoperation, survival, steatorrhea, pancreatic enzyme supplementation requirement, occurrence of bile gastritis, dumping, new onset of diabetes, marginal ulcers, cholangitis, postoperative weight trends and frequency of hospital readmission for symptom management. Follow-up was done 6 monthly in all the patients with the aim of diagnosis of recurrence and assessment of long-term gastrointestinal and pancreatic function and nutritional status. In addition, in 15 patients, upper gastrointestinal endoscopy (UGIE) with gastric and jejunal biopsy, 99Tc-HIDA scan, determination of fecal fat loss after a standard 100-g fat diet for three days, fecal elastase-1 measurement (ELISA) and MRCP were done to objectively document the changes in gastrointestinal and pancreatic function. The data were compared with the results available in the literature.

The mean age was 56 years with a range of 27-85 years. There were 130 males and 95 females. Preoperatively 18 patients had diabetes and preoperative weight loss varied from 5-30 kilograms with a mean of 12 kilograms. Of all the PD cases 57 were for ampullary, 70 were for lower end cholangiocarcinoma and 98 were for pancreatic head cancer. The postoperative complications occurred in the form of intra-abdominal bleed (5), pancreaticojejunostomy leak (12), intra-abdominal abscess (4) and pneumonia (5). Delayed gastric emptying was not seen in any of the patients. In 8/12 patients with PJ leak the closure was achieved with the conservative treatment. The 30-day mortality was 6/225 (2.66%). The causes were sepsis in 3, intra-abdominal bleed in 2 and pulmonary embolism in 1 patient. There was no mortality related to PJ leak. The median follow-up was of 36 months. The overall 5-year survival for ampullary, lower end cholangio- and pancreatic head carcinoma were 65%, 25% and 20% respectively. After surgery none of the patients had clinical evidence of steatorrhea, gastritis, peptic ulcer disease, cholangitis, dumping and there was no new case of diabetes. After 6-12 months 80% of the patients gained weight similar to their preoperative levels. UGIE with gastric and jejunal biopsies, 99Tc-HIDA scan, fecal fat loss estimation, fecal elastase estimation and MRCP were done in 15 patients and were found to be normal.

Our method of PD and reconstruction produces encouraging results with respect to PJ leak, mortality, DGE, malabsorption, bile gastritis, dumping, marginal ulcers and diabetes. We recommend this technique as a safe and effective method even to the low volume centers.