Association of percutaneous endoscopic gastrostomy (PEG) tube with unplanned hospitalization and clinical outcomes among patients with head and neck cancer.

343 Background: Despite intensity modulated radiation therapy for patients with head and neck cancer, oral mucositis pain remains common leading to hospitalization. However, there is a heterogeneous practice pattern in prophylactic percutaneous endoscopic gastrostomy (PEG) tube use, and it is unclear whether increased hospitalization is driven by those with versus without prophylactic PEG tube. To address this knowledge gap, we performed an observational cohort study to evaluate the association of PEG tube with hospitalization and other clinical outcomes. Methods: Our single institution database was queried for patients with non-metastatic head and neck cancer diagnosed between October 2011 and December 2023 who received definitive radiation or chemoradiation. Reasons for hospitalization during and within 3 months after treatments were summarized. Fine-Gray multivariable analysis (MVA) was performed to identify variables associated with time to first hospitalization. Logistic MVA was used to identify variables associated with first hospitalization and readmission. Cox MVA and Fine-Gray MVA were performed to analyze survival and tumor recurrence outcomes. Bonferroni correction was used for multiple comparisons with P < 0.025 for statistical significance. Results: A total of 560 patients were identified (n = 141 [25.2%] with prophylactic PEG tube, n = 203 [36.3%] with no PEG tube, n = 216 [38.6%] with therapeutic PEG tube). Median follow up was 44.9 months (95% confidence interval [CI] 43.0-46.6). Overall, 249 (44.5%) and 94 (16.8%) were hospitalized once and multiple times, respectively. Three most common reasons for hospitalization were: failure to thrive (n = 192 [33.9%]), fever or infection (n = 83 [14.7%]), and low blood count (n = 50 [8.8%]). On Fine-Gray MVA for time to first hospitalization, no PEG tube was associated with less hospitalization (adjusted hazards ratio [aHR] 0.64, 95% CI 0.45-0.91, p = 0.01), but not therapeutic PEG tube (aHR 1.16, 95% CI 0.84-1.59, p = 0.38) when compared to prophylactic PEG tube. Similar findings were noted on logistic MVA (no PEG tube: adjusted odds ratio [aOR] 0.58, 95% CI 0.36-0.92, p = 0.023; therapeutic PEG tube: aOR 1.24, 95% CI 0.79-1.95, p = 0.36). When no PEG tube and therapeutic PEG tube were each compared to prophylactic PEG tube, there was no statistical significance for readmission, overall survival, progression-free survival, locoregional failure, and distant failure. Similar findings were noted when stratified by p16 status. Conclusions: Our study suggested that nearly half of patients were hospitalized at least once during and within 3 months after treatments, and the most common reason was failure to thrive. Those who never had PEG tube had the lowest hospitalization rate, while prophylactic PEG tube was not associated with less hospitalization compared to those with therapeutic PEG tube.