37 Background: Patients with HNSCC who undergo definitive chemoradiation (dCRT) frequently require nutritional support via G-tubes; despite this, the factors associated with the timing of G-tube placement remain poorly understood. Methods: We performed a retrospective analysis of patients with HNSCC who received dCRT and required G-tube placement between 01/2018 – 01/2024. We segregated G-tube timing into two groups as determined by the intent of the treating clinician: “prophylactic” placement (ordered in anticipation of dCRT start) or “reactive” placement (ordered in response to weight loss or other adverse clinical indicators). We then evaluated associated clinical variables (patient, disease, treatment) and outcomes (weight loss, G-tube duration, healthcare utilization). Differences across G-tube groups were assessed using t-tests or Wilcoxon Rank Sum tests for continuous measures and Fisher’s exact tests or logistic regression for binary variables. Results: We identified 86 patients; 59 (69%) prophylactic and 27 (31%) reactive G-tubes were placed. At baseline, patients had a mean age of 63, BMI of 28 kg/m 2 , and Charlson Comorbidity Index of 4. Patients were 12% non-White, 2% non-English speaking, and 10% unhoused; 27% had a non-metropolitan address, 67% had public insurance, 31% were smokers and 26% had a history of alcohol dependence. The most common site was the oropharynx (66%; 75% HPV+). Across HNSCC types, 38% had stage IV disease and 60% had preexisting dysphagia/odynophagia. Most received weekly (83%) cisplatin (84%) and bilateral (95%) radiation (RT). We collected RT dose data (parotids, esophagus, pharyngeal constrictors) and found no association between RT dose and G-tube group. Patients with prophylactic G-tubes were significantly more likely to have preexisting odynophagia/dysphagia (69% vs 41%, p = 0.017). Compared to the prophylactic group, reactive G-tube placement occurred more often among patients living in non-metropolitan settings (44% vs 19%, p = 0.018) and was associated with a trend towards a higher mean percent weight loss (12% vs 10%; p = 0.128) and cachexia (per int’l criteria; 89% vs 74%, p = 0.159). We observed no difference in the median G-tube duration or in the rate of persistent G-tube use at last follow-up or death. Notably, we found that patients with reactive G-tubes were hospitalized more frequently (74% vs 36%, p = 0.001) and had a longer median length of stay (LOS; 6 vs 0 days, p = 0.001); these findings remained statistically significant when controlling for other factors (OR 3.76 [95% CI: 1.20-12.65]; p = 0.026). Conclusions: We found that reactive G-tubes were associated with increased hospitalization and LOS, suggesting that select patients may benefit from earlier G-tube placement.
Read full abstract