Nutritional Route In Esophageal Resection Trial II

Esophageal Neoplasms

Esophagectomy, Nutrition, Oral feeding, Enteral feeding, Pulmonary complications, Anastomotic leakage, Quality of life

The aim of this study is to investigate the effects of early start versus delayed start of oral intake on postoperative functional recovery after an esophagectomy.

Early oral feeding is safe and beneficial in most types of gastro-intestinal surgery and is a crucial part of fast track or enhanced recovery protocols. Following esophagectomy, fast track programs are increasingly being applied, resulting in a reduced length of stay, perioperative morbidity and hospital charges. In a previous feasibility study (Nutrient I trial) has been shown that direct oral intake following esophagectomy was feasible and did not result in an increase of major complications. It remains unclear what the best strategy is for nutrition in the early postoperative phase following esophagectomy. The aim of this study is to investigate the effects of early start versus delayed start of oral intake on postoperative recovery and pulmonary complications following esophagectomy. Patients will be recruited at the outpatient clinic of the Catharina Hospital Eindhoven and the Hospital Group Twente.Patients will enroll in the early oral feeding group or the delayed oral feeding group. The early oral feeding group will receive liquid oral nutrition directly after surgery. Patients in the delayed oral feeding group will receive enteral tube feeding (via a jejunostomy) for 5-7 days (standard protocol in The Netherlands). After these 5-7 days a liquid oral diet is initiated. Patient characteristics and clinical parameters are registered in an electronic database. All surgical complications are classified using the Clavien-Dindo classification15 (Appendix II). An independent investigator will check functional recovery daily. Data is provided by local investigator via an online checklist (electronic case report form / website). Quality of life and Symptoms are scored using the European organization for research and treatment of cancer questionnaire (QLQ-C30/EORTC-OG25)16 at baseline (5 weeks after neoadjuvant treatment), 6 weeks, 3 months and 6 months postoperatively. Cost-effectiveness is scored via EQ-5D-5L, iMTA Medical Consumption Questionnaire (iMCQ) and Productivity Cost Questionnaire (iPCQ). Nutritional status will be measured using the actual caloric and protein intake versus the needed caloric and protein intake. Long-term outcome parameters such as local recurrence, overall and cancer-specific survival will be registered in a database until the end of the study. The sample size calculation is based on functional recovery as primary outcome. Functional recovery is the time to surgical recovery according to the definition described previously. Patients in our historic cohort (1 week artificial nutrition) were at least considered functional recovered by a mean of 12 days post-surgery. Patients in the NUTRIENT I were at least functional recovered at day 10 postoperative (mean) following an esophagectomy. In this trial we consider a reduction of 2 days in functional recovery as clinically significant. Using a Power of 80%, an alpha of 5% and a standard deviation of 4 days, a total of 128 patients (64 patients in each group) is needed to show this difference. It is expected that the primary outcome will not be normally divided and therefore an extra 15% inclusion is necessary, requiring a total of 148 patients (74 patients in each group). Interim-analyses will be performed following 50 and 100 patients and will be checked by an independent physician. All analyses will be done according to the intention-to-treat approach in which all randomized patients are included, regardless of adherence to study protocol. Occurrences of the primary and secondary endpoints are compared between the treatment groups. Results are presented as risk ratios with corresponding 95% confidence intervals. A two-tailed P \< 0.05 is considered statistically significant. To compare the groups the data will be tested for normal distribution and an unpaired T-test will be performed when appropriate, otherwise the Mann-Whitney U or Chi-square tests.

Netherlands