Dosimetric predictors of laryngeal edema

Purpose: To investigate dosimetric predictors of laryngeal edema after radiotherapy (RT). Methods and Materials: A total of 66 patients were selected who had squamous cell carcinoma of the head and neck with grossly uninvolved larynx at the time of RT, no prior major surgical operation except for neck dissection and tonsillectomy, treatment planning data available for analysis, and at least one fiberoptic examination of the larynx within 2 years from RT performed by a single observer. Both the biologically equivalent mean dose at 2 Gy per fraction and the cumulative biologic dose-volume histogram of the larynx were extracted for each patient. Laryngeal edema was prospectively scored after treatment. Time to endpoint, moderate or worse laryngeal edema (Radiation Therapy Oncology Group Grade 2+), was calculated with log rank test from the date of treatment end. Results: At a median follow-up of 17.1 months (range, 0.4-50.0 months), the risk of Grade 2+ edema was 58.9% +/- 7%. Mean dose to the larynx, V30, V40, V50, V60, and V70 were significantly correlated with Grade 2+ edema at univariate analysis. At multivariate analysis, mean laryngeal dose (continuum, hazard ratio, 1.11; 95 % confidence interval, 1.06-1.15; p < 0.001), and positive neck stage at RT (NO-x vs. N +, hazard ratio, 3.66; 95% confidence interval, 1.40-9.58; p = 0.008) were the only independent predictors. Further stratification showed that, to minimize the risk of Grade 2+ edema, the mean dose to the larynx has to be kept <= 43.5 Gy at 2 Gy per fraction. Conclusion: Laryngeal edema is strictly correlated with various dosimetric parameters; mean dose to the larynx should be kept <= 43.5 Gy. (c) 2007 Elsevier Inc.