Water protection after tympanostomy (Shepard) tubes does not decrease otorrhea incidence – retrospective cohort study()

INTRODUCTION: Myringotomy for tube insertion is the most common otologic surgery. Otorrhea is a frequent complication of this procedure and, to prevent it, most surgeons strongly recommend avoiding contact with water as this is thought to adversely impact on post-operative quality of life. OBJECTIVE: To understand the benefit of this recommendation. METHODS: Observational study – retrospective cohort study comparing the incidence of post-operative otorrhea and its impact on patients’ quality of life, in two groups of patients comprising children under 10 years of age who underwent bilateral myringotomy and tube placement for chronic otitis media with effusion between May 2011 and May 2012. One group received water protection care after surgery, the other did not. Data was collected through telephonic interview, after one year of follow up (one year after the procedure). Water exposure without protection was considered the exposure event. Incidence of otorrhea and perceived impact on quality of life were the outcome measures. Results were compared after logistic regression. RESULTS: We included 143 children: 116 were not exposed to water without protection and 27 were exposed. In the not exposed group 36.2% had at least one episode of otorrhea, compared to 40.0% of the exposed group. Odds ratio for otorrhea on exposed was 1.21 (95% CI 0.51–2.85, p = 0.6). Negative impact on quality of life was reported by parents of 48.2% on the not exposed children, compared to 40.7% on the exposed group. This difference was not significant (p = 0.5). CONCLUSION: We found that recommending water protection did not have beneficial effect on the incidence of otorrhea after myringotomy with tubes on chronic otitis media with effusion. However, such measures did not appear to have a negative impact on quality of life. This is a populational observational study with few cases (143 cases); these final statements would be better stated by a very large populational study with another large control group.