Retrospective study of risk factors for tympanic membrane perforation in the early period after intratympanic steroid injection

Abstract

ObjectiveIntratympanic steroid injection (ITSI) can be an effective treatment for sudden sensorineural hearing loss or Meniere's disease. Tympanic membrane (TM) perforation after ITSI is a major complication which needs additional treatment. The purpose of this study is to assess the factors associated with TM perforation after ITSI. MethodsWe obtained the clinical data of patients who underwent ITSI treatment at the Department of Otolaryngology JR Tokyo General Hospital from April 2013 to March 2021. The data included age, sex, treated side, number of injections, average interval between injections, TM anesthesia with Zentöl solution, which contains phenol, any history of diabetes and any concurrent use of oral or intravenous steroids. We evaluated the association between these variables and TM perforation after ITSI using the Student's t-test, the chi-squared test, univariate logistic regression analysis and multivariate logistic regression analysis. TM perforation was defined as a case in which perforation was observed at least once during outpatient visits. ResultsRecords of 205 ears in 190 patients were analyzed. The overall proportion of TM perforation in the early period after ITSI was 12.7% (26 out of 205 ears), which decreased to 9.3% (19 out of 205 ears) and 5.9% (12 out of 205 ears) at the 1- and 3-month follow-ups, respectively. The proportion of TM perforation in the early period after ITSI without TM anesthesia was 3.5% (5 ears out of 145 ears), which decreased to 1.4% (2 ears) or 0% at the 1- or 3- month follow-ups, respectively. The use of tympanic anesthetics which contain phenol was significantly associated with TM perforation in univariate logistic regression analysis (odds ratio: 15.08, 95% confidence interval: 5.34–42.56, p < 0.001) and in multivariate analysis (odds ratio: 20.76, 95% confidence interval: 6.31–68.3, p < 0.001). All TM perforation cases without TM anesthesia healed spontaneously or with paper tympanic closure treatment. TM perforation in 6 ears out of 21 ears with TM anesthesia did not heal during the follow-up. ConclusionThe overall proportion of TM perforations from the early period after ITSI was 12.7%, 9.3% at the 1-month post-ITSI outpatient follow-up, 5.9% at the 3-month post-ITSI outpatient follow-up. Tympanic anesthesia was significantly associated with TM perforation after ITSI, which indicated that TM anesthesia with solutions containing phenol is not recommended for ITSI.