Gastrostomy Tubes in Patients with Advanced Head and Neck Cancer

Objectives: To measure the percentage of patients requiring gastrostomy tubes (G‐tubes) and the timing of their placement, in addition to studying whether pretreatment variables (T stage, tumor site, N stage) and intratreatment variables (weight loss during treatment) are valid predictors for the need for G‐tube placement. Study Design: Retrospective case control study. Methods: Retrospective review of a prospectively collected database with chart reviews was carried out on 477 patients treated on a regimen of intraarterial cisplatin and concurrent radiation. The likelihood of requiring a G‐tube was calculated by studying several independent variables using chi‐square analysis and the unpaired t test. Results: Two hundred twenty (46%) patients had G‐tubes placed, with mean time of placement being on day 20 after the initiation of treatment. Patients with T4 disease were statistically more likely to require a G‐tube (50% vs. 37%, P < .05). Patients were less likely to require a G‐tube when the primary site of tumor involved the larynx or paranasal sinuses ( P < .05). There were no statistically significant differences between the two groups with regards to patient age, sex, N stage, and weight changes over the course of treatment. The odds of requiring a G‐tube were highest for tumors involving the hypopharynx (80%), oral cavity (52%), and oropharynx (49%). Conclusions: Although clinical judgment appears to be sufficient for deciding when G‐tube placement is warranted in our population, the threshold for placing a tube should be lower in those patients who had a higher likelihood of requiring a G‐tube during the course of treatment, including those with advanced disease of the oral cavity, oropharynx, and hypopharynx.