Transtympanic Gentamicin vs. Steroids in Refractory Meniere's Disease

Meniere's disease is characterised by episodic spontaneous vertigo attacks with hearing loss, ringing sounds and fullness in the ear. In one out of five patients, standard first line medical treatment is not effective in controlling vertigo attacks. For these incapacitated patients, gentamicin injections through the ear drum is a well established minimally invasive treatment. Major surgery of the balance organs or nerve, risking complete hearing loss, CSF leak, meningeal infections, are rarely performed nowadays. Gentamicn is very effective in controlling vertigo and acts by chemical ablation of end organs. As hearing and balance organs are entwined around each other, gentamicin treatment does not come without the risk of hearing loss. In fact, meta-analysis shows hearing deterioration in 13% to 35% percent of gentamicin treated patients. On the other hand, steroids are drug of choice for autoimmune inner ear disease and commonly used for sudden hearing loss. They are non toxic drugs without any known side effects during local treatment in ear. We will compare the two in this randomised, double blind trial.

Primary Outcome Measures

1. Vertigo Attacks
   The number of vertigo attacks between 18-24months follow-up were taken retrospectively during a face-to-face appointment at 24 months follow-up and compared to 6 month pre-enrollment baseline (as per Committee on Hearing and Equilibrium guidelines).

Secondary Outcome Measures

1. Change in Hearing
   Hearing was measured as ipsilesional pure-tone threshold at Baseline, 1month, 2months, 6months, 12month, 18months and 24 months follow-up. Hearing level was taken as the average threshold across 0.5, 1, 2 and 3KHz.
2. Change in Speech Discrimination
   Speech discrimination was measured at Baseline, 1month, 2months, 6months, 12month and 24 months follow-up. Speech discrimination was assessed by means of ipsilesional suprathreshold word recognition (%). Arthur Boothroyd's isophonemic word lists (AB wordlists, Guymark, Southampton) comprising sets of 10 words were played to the ipsilesional ear at the low-frequency pure-tone threshold of 0·5, 1 and 2 kHz +30dB with masking sound in the contralesional ear if necessary. The formula for masking level was: low-frequency pure-tone threshold in ipsilesional ear - bone conduction mean threshold (0·5, 1 and 2KHz) in contralesional ear - 40dB. Speech loudness and masking were rounded to the nearest 5dB. Step increments and decrements of 10dB for speech loudness and masking were used to attain the maximum speech discrimination score.