Use of prophylactic and reactive feeding tubes in patients with esophageal cancer: A single-centre cohort study

Esophageal cancer (EC) patients suffer from significant cachexia prior to and during treatment with chemotherapy and radiation (CRT). Feeding tubes can become the primary form of nutritional support for these patients during and after completion of treatment. Our primary objective was to identify factors that can predict EC patients at high risk of requiring feeding tube insertion.

A retrospective cohort review was completed. All patients with an EC diagnosis made from January 1, 2013 to December 31, 2018 were included. Baseline characteristics of all patients and those receiving feeding tubes were collected. A multivariate logistic regression was performed comparing the group that required a reactive feeding tube insertion to those who did not require any feeding tube insertion to identify risk factors.

A total of 350 patients were included in the study, and 132/350 (38%) patients received a feeding tube, with gastrostomy tube being the most common type inserted (72%). 82 out of 132 (62%) patients had feeding tube inserted prophylactically, while 50/132 (38%) patients had feeding tube inserted reactively. Amongst patients who had a feeding tube inserted, median age was 67 years old, with 74% being male. Fifty-one out of 132 (39%) were adenocarcinoma, 75/132 (57%) were squamous cell carcinoma, 5/132 (4%) were neuroendocrine, and 1/132 (1%) was lymphoma. Severe dysphagia (OR 19.9, 95% CI 2.6-151.2, p < 0.001) at diagnosis and decision to undergo chemotherapy (OR 2.8, 95% CI 1.3-5.8, p=0.008) appeared to be predictors for reactive feeding tube insertion. Complications relating to feeding tubes were seen in 24/50 (48%) of the reactive insertion group and 34/82 (41%) of the prophylactic insertion group (p=0.48).

In this single-centre cohort study, severe dysphagia and undergoing chemotherapy were identified as risk factors for requiring a feeding tube later on. Future prospective studies can further explore other risk factors that may predict those at high risk of requiring a feeding tube, especially as it appears that those who have feeding tubes inserted reactively may have more complications than those who have feeding tube inserted prophylactically.

The authors.

Has not received any funding.

G. Pond: Honoraria (self): Takeda; Advisory / Consultancy: Merck, Astra-Zeneca, Profound Medical; Shareholder / Stockholder / Stock options: Roche; Full / Part-time employment: Roche Canada. I. Menjak: Disclosures not submitted. All other authors have declared no conflicts of interest.