Perioperative optimization with nutritional supplements in patients undergoing gastrointestinal surgery for cancer: A randomized, placebo-controlled feasibility clinical trial.

BACKGROUND: Perioperative nutritional supplementation may improve outcomes. Trials have not investigated the role of combination strategy using different types of nutritional supplements. METHODS: We conducted a single-site randomized pilot trial, among gastrointestinal cancer patients undergoing surgery, comparing perioperative nutritional supplements versus placebo (1 placebo to each supplement), to determine feasibility of a larger trial. Intervention, administered in sequence, included: protein supplementation (preoperative day 30-6), protein supplementation rich in arginine and omega-6 (preoperative day 5-1, and postoperative day 1-5), and carbohydrate loading (surgery day). Primary outcome was enrollment. Secondary outcomes included participant compliance with study supplements (target ≥70% of total packets). We planned protocol modifications to improve enrollment and compliance. Postoperative complications were described. RESULTS: Over 18 months, 495 patients were screened, 144 were deemed eligible, and 71 consented to participate, resulting in an enrollment fraction of 71/144 (49%, 95% confidence interval: 41%-57%). 'Too much burden' was the most common reason for refusal to participate (34%). Participants' median overall compliance with study packets was 80%. Protocol modifications (decreasing the interval from enrollment to surgery from 4 to 2 weeks and decreasing length of baseline assessment) did not impact enrollment or compliance. Postoperative complications were similar between control (18/31 [58%, 95% confidence interval: 4-74]), and intervention (22/34 [65%, 95% confidence interval: 48-79]) arms, with a higher proportion of infectious complications in the control arm (16/31, 52% vs 12/34, 35%). CONCLUSION: Results from this pilot suggest a larger phase III trial is feasible. Postoperative infectious complications were common, making this a suitable outcome of interest.