Objective: This study compared the effectiveness of five different techniques of intratympanic gentamicin administration for Meniere's disease. Data Sources: A MEDLINE search of the English language literature from 1978 to 2002 was performed using the key words "intratympanic," "gentamicin," "therapy," "Meniere's," and "disease." Study Selection: Inclusion criteria to select articles for meta-analysis were clear description of gentamicin delivery technique, clearly reported vertigo control results, and report of hearing loss posttreatment. Seven studies (n = 218) describing the multiple daily dosing technique (delivery three times per day for greater than or equal to4 d), two studies (n = 84) describing, the weekly dosing technique (weekly injections for four total doses), eight studies (n = 253) of the low-dose technique (one to two injections with retreatment for recurrent vertigo), four studies (n = 156) of continuous microcatheter delivery; and six studies (n = 269) of the titration technique (daily or weekly doses until onset of vestibular symptoms, change in vertigo, or hearing loss) were entered into the model. Data Extraction: Vertigo control results were stratified into complete, substantial, or poor control. Hearing results were separated by profound, partial, or no hearing loss. Individuals undergoing caloric testing were separated by degree of vestibular ablation (complete versus partial) and analyzed for vertigo control (n = 301) and hearing loss (n = 333) after treatment. Data Synthesis: Comparisons between the rates of complete vertigo control, effective vertigo control (complete plus substantial control), overall hearing loss (partial plus profound), and profound hearing loss by delivery method were based on a parametric empirical Bayes analysis using binomial generalized linear models and backward variable selection (joining). Relative risk for vertigo control and hearing loss by partial or complete ablation was examined study by study using residual maximum likelihood to carry out a parametric empirical Bayes analysis. Conclusion: The titration method of gentamicin delivery demonstrated significantly better complete (81.7%, p = 0.001) and effective (96.3%, p < 0.05) vertigo control compared with other methods. The low-dose method of delivery demonstrated significantly worse complete vertigo control (66.7%, p < 0.001) and trends toward worse effective vertigo control (86.8%, p = 0.05) compared with other methods. The weekly method of delivery trends toward less overall hearing loss (13.1%, p = 0.08), and the multiple daily method demonstrated significantly more overall hearing loss (34.7%, p < 0.01) compared with other groups. No significant difference in profound hearing loss was found between groups. Degree of vestibular ablation after gentamicin therapy is not significantly correlated with the resulting vertigo control or hearing loss status.
