Impact of early oral hydration on pharyngocutaneous fistula following total laryngectomy

To the Editor, I read with interest the article entitled “Early oral hydration protects against pharyngocutaneous fistula after total laryngectomy or total pharyngolaryngectomy.” Le Flem et al compared a historical control (did not receive early oral hydration) to the intervention group (received early oral hydration). I commend the authors on their reduced rate of pharyngocutaneous fistula (PCF) from 50% (2015-2017) to 20% (2017-2019) following their intervention. The authors acknowledged the high rate of PCF in the historical control group between 2015 and 2017. It is unclear whether this high rate represented an anomaly, or if the rate at their department has been consistently high prior to the 2015. This is important, as the use of a historical control could overestimate the effect of their intervention: if the historical control is systematically different from concurrent control. Of note, the authors compared baseline characteristics between the groups, and patients in the early oral hydration group had significantly higher rate of comorbidities supporting their findings. In addition, the reported median rate (interquartile range) of PCF following total laryngectomy in 52 individual studies over the past 20 years was 21.5% (16.6%27.2%); pooled rate of 21% (1808/8605). The rate of 20% reported by Le Flem et al following introduction of early oral hydration appeared comparable to most reports without routine early oral feeding/hydration. The effect of early oral feeding on PCF has been previously examined. A meta-analysis of four randomized controlled trials (180 patients) and four observational studies (490 patients) compared early vs late oral feeding/ hydration. They found no significant difference in rate of PCF between the groups. These meta-analyses were limited by heterogeneous regimen for early feeding following laryngectomy. Another randomized controlled trial evaluated tolerance of early oral feeding alone following laryngectomy. They found that patients who had early oral feeding alone did not meet their calorific and protein requirement during postoperative day 1 to 4 and required complementary nutrition via other route. Other factors previously identified to increase risk of PCF in meta-analyses including previous radiotherapy and advance tumor stage were not statistically significant in this current study by Le Flem et al. This may be due to type II error, given the small sample size. A PubMed database search (from inception to July 2020) found no other study reporting significant reduction in the rate of PCF with early oral hydration/feeding. However, previous studies have not shown increased risk of PCF with early oral feeding. Thus, some authors have considered this option due to the potential benefit of facilitating early discharge, reduce cost, and complications associated with nasoenteral feeding. Well-designed controlled studies with a sample size calculation/justification are needed to validate the findings by Le Flem et al.