A systematic review of methodology in nutritional trials in patients undergoing pancreatoduodenectomy.

Background: Malnutrition is highly prevalent amongst pancreatoduodenectomy (PD) patients and remains a major modifiable risk factor for postoperative morbidity and mortality. However, studying nutritional interventions is difficult due to multiple methodological issues. The aim of this review is to report and compare the administration of different nutritional therapies as well as reporting of nutritional outcomes in randomised controlled trials (RCTs) of PD patients. Methods: A systematic review of RCTs reporting the effect of different nutritional interventions after PD was performed according to the PRISMA guidelines. Results: In the 20 suitable studies identified, there was widespread variation in when oral diet was started, ranging between postoperative day (POD)1 and POD7. Regarding enteral tube feeding, multiple different feed types were used and feed administration was highly variable, for both rate and duration. In terms of outcome types, 17 (85%) studies reported biochemical outcomes with serum albumin being the most common (12 studies, 57%) and 12 (57%) reported anthropometric outcomes with body mass index (BMI) being the most common (7 studies, 33%). Conclusions: Amongst the published RCTs investigating the effect of different nutritional routes after PD, widespread variation in the methodology was evident. This has implications for both interpretation of results and for design of future nutritional trials. Multiple biochemical and anthropometric outcomes were reported but quality of life measures, such as the SF-36 recommended by a recently published core outcome set (COS), were not used.