The results of Kaplan-Meier and multivariate analyses of etiological factors related to the outcome of combined pharmacological therapy against laryngeal granuloma.

CONCLUSIONS: The present results indicate that the diagnosis of gastroesophageal reflux disease (GERD) is an independent etiological factor predicting retardation of the resolution of laryngeal granuloma.

OBJECTIVES: To assess the effects of combined usage of an inhaled corticosteroid plus tranilast and/or a proton pump inhibitor on the size of granulomatous lesions, and to reveal etiological factors related to the outcome using the Kaplan-Meier method and a subsequent multivariate analysis.

METHODS: Sixty-two patients with laryngeal granuloma were enrolled. An inhaled corticosteroid plus tranilast (300 mg/day) and rabeprazole (20 mg/day) were administered to all of the patients, and only to those diagnosed to have GERD, respectively. The size of granulomatous lesion was measured for each patient at the initial visit and every 4 weeks. At 48 weeks, the Kaplan-Meier plots for lesion disappearance rate were compared between groups with and without each of the etiological factors, followed by Cox proportional-hazards regression.

RESULTS: The 48-week lesion disappearance rates for the whole population were 82.3%. Although the Kaplan-Meier analysis exhibited significant differences between patients separated by GERD diagnosis, phonotrauma, and habitual smoking, only GERD were identified as a real independent etiological factor affecting the resolution of the lesion by a multivariate analysis using Cox's proportional-hazards regression.