Traumatic perforation of the tympanic membrane: Etiologies and risk factors for healing and intervention

ABSTRACT

@#<p style="text-align justify;"><strong>OBJECTIVE:</strong> To study various etiologies of traumatic tympanic membrane perforation; evaluate the factors involved in healing of traumatic tympanic membrane perforation; and identify patients with perforations unlikely to benefit from conservative management.<br /><strong>METHODS:</strong><br /><strong>Design:</strong> Prospective observational study<br /><strong>Setting:</strong> Tertiary Government Medical College and Hospital<br /><strong>Participants:</strong> 64 consecutive cases of traumatic tympanic membrane perforation seen over one year were followed for 3 months. Perforations were assessed in terms of size, etiology, condition of edge and other associated factors or combinations of factors with regards to spontaneous healing using descriptive statistics and chi-square tests.<br /><strong>RESULTS:</strong> Of the 64 cases, 51 perforations healed while 13 did not. There were significant associations between tympanic membrane condition after 3 months and explosive mode of injury (?2 = 23.30; p=.00001) as well as with size of perforation ((?2 = 25.75; p=.00001). The risk of persistence of a tympanic membrane perforation was 34.57 times more among patients with a perforation size >50% compared to those with perforation size ?50% [OR-34.57 (6.28, 190.14); p= .00001]. Combined, explosive etiology and perforation size >50% were significantly associated with non-healing ((?2 = 37.60; p = .00001). There were no significant associations with the condition of the edge of the perforation and upper respiratory tract infection.<br /><strong>CONCLUSIONS:</strong> An explosive etiology and tympanic membrane perforation size >50% may be significant risk factors predicting non-healing of the perforation. Risk stratification of patients having one or both of these risk factors with early intervention for those with both, and close monitoring for those with any one of these may lessen unnecessary morbidity. Bigger multicenter future studies are necessary to confirm these initial findings.</p>