Relationship Between Early Oral Intake Post Pancreaticoduodenectomy and Chyle Leakage: A Retrospective Cohort Study.

BACKGROUND

Early oral intake is strongly recommended according to the enhanced recovery after surgery (ERAS) guidelines because it can reduce complications and improve recovery. However, early oral intake has been indicated to be associated with chyle leakage (CL) after pancreatic surgery, which may lead to worsening of existing malnutrition and impeded recovery. This study investigated the relationship between early oral intake and CL and identified risk factors for CL to reduce its occurrence and promote recovery after pancreaticoduodenectomy.

MATERIALS AND METHODS

All patients who underwent pancreaticoduodenectomy between June 2014 and June 2018 were identified retrospectively. Patients were divided into the early-oral-intake and control groups according to whether they had early oral intake according to ERAS protocols. CL and other clinicopathological characteristics were recorded. Univariable and multivariable analyses assessed CL risk factors.

RESULTS

Early oral intake improved recovery, leading to a shorter postoperative hospital stay for the early-oral-intake group in comparison to that of the control group [13.6 (range, 12-68) vs. 17.8 (range, 14-83) days;  = 0.047] without increasing the incidence of CL and other complications. CL was diagnosed significantly earlier in the early-oral-intake group than in the control group [4.6 (range 3-5) vs. 6.7 (range 3-9) days;  = 0.001]. Early oral intake did not increase the grade severity ( = 0.845) or the costs ( = 0.241) or prolong postoperative hospital stays ( = 0.611). A primary diagnosis of malignancy, para-aortic lymph node dissection, lymphatic invasion, lymph node metastases, the number of harvested nodes, and the number of positive nodes were significantly associated with CL ( < 0.05), whereas early oral intake was not ( = 0.525). Multivariate analyses demonstrated that para-aortic lymph node dissection ( = 0.039) and the number of harvested nodes ( = 0.001) were independent risk variables.

CONCLUSION

This study provides significant evidence that early oral intake after pancreaticoduodenectomy is not associated with CL. The identification of the independent risk factors for CL can help prevent it.