Two-year outcome of ventilation tubes in a randomized controlled trial of persistent childhood otitis media with effusion

Abstract

Introduction. Resolving the conflict between commissioners and providers of treatment in otitis media with effusion requires a cost-effectiveness trial in which the outcomes reflect health gain for the whole child. TARGET (Trial of Alternative Regimens of Glue Ear Treatment) is an MRC-funded randomized controlled trial of surgical management (bilateral ventilation tubes with and without adjuvant adenoidectomy) versus non-surgical management in children aged between 3.25 and 6.75 years. Methods. The entry criterion was bilateral otitis media (OME) associated with a pure-tone average (PTA) >20 dB HL in the better hearing ear, documented as persistent over a 12-week watchful waiting period. The watchful waiting period of 12 weeks is supported as appropriate for routine practice, because 48% of children meeting the criterion at their first visit showed sufficient resolution to no longer satisfy the entry criterion. Certain groups of children were not randomized because of overriding parental or doctor concern, or because the parents refused participation or randomization. These groups did not differ significantly on any of the variables likely to affect the outcome. Some children with a PTA >40 dB HL were excluded from randomization but were fully followed-up following surgery. Results were found to be largely similar to those for the randomized children with this degree of hearing impairment. Of the 370 randomized children, 124 had bilateral Shephard ventilation tubes inserted under a general anaesthetic within 6 weeks, 127 had similar surgery but with the addition of adenoidectomy, and 119 had non-surgical management. Following randomization, all children were followed up 3, 6, 12, 18 and 24 months later. This paper presents an initial analysis, on an intention-to-treat basis, of the outcomes in the children having ventilation tubes alone against the non-surgically managed children. The adjuvant effect of adenoidectomy will be reported subsequently. Results. The average effect of management, summarized over the first and both years, for the hearing, the reported hearing difficulty, the parental report of the child's behaviour, the child's general health and the parental report of quality of family life are shown in Tables 2, 3, 4, 5 and 6 respectively. The differences in the effect between ventilation tubes and non-surgical management were all significant (P < 0.05) except when indicated by NS (not significant). The effect of surgery is reported both as a difference and as a proportion of the standard deviation in the sample prior to randomization. Using the latter scale, a difference of 0.33 SD is a modest but worthwhile effect, 0.66 SD a substantial effect, 1.00 SD a large effect, and 1.33 SD a very large effect. Table 2. Average effect of management on hearing loss (HL) Mean dB HL over 1 year 2 years The mean HL at randomization was 32.7 dB HL (SD 6.1). Ventilation tubes 17.6 18.5 No surgery 23.4 21.8 Difference 5.8 3.3 Size of effect 0.9 SD 0.6 SD Table 3. Average effect of management on reported hearing difficulty Mean score over 1 year 2 years The mean score at randomization was 10.4 (SD 3.2). Ventilation tubes 4.6 4.9 No surgery 8.1 7.8 Difference 3.5 2.9 Size of effect 1.3 SD 1.3 SD Table 4. Average effect of management on behaviour Mean score over 1 year 2 years The disease effect of otitis media (OME) on behaviour was 0.67 SD. The mean score at randomization was 13.2 (SD 5.7). Ventilation tubes 10.5 9.7 No surgery 12.2 12.0 Difference 1.7 2.3 Size of effect 0.3 SD 0.6 SD Table 5. Average effect of management on general health Mean score over 1 year 2 years The disease effect of otitis media (OME) on general health was 0.68 SD. The mean score at randomization was 1.2 (SD 0.8). Ventilation tubes 0.8 0.9 No surgery 1.0 1.0 Difference 0.2 0.1 Size of effect 0.4 NS Table 6. Average effect of management on quality of life (QOL) Mean QOL over 1 year 2 years The disease effect of otitis media (OME) on QOL was 1.01 SD. The mean QOL at randomization was 16.1 (SD 4.4). Ventilation tubes 18.5 18.9 No surgery 17.6 18.3 Difference 0.9 0.6 Size of effect 0.4 SD NS Conclusions. The substantial beneficial effect of ventilation tubes on the hearing in children with bilateral, persistent OME has been confirmed up to 2 years. This is mirrored by a substantial and significant improvement in behaviour. There is a modest but significant improvement in general health and parental quality of life up to 1 year. There is a potential additional effect on these outcomes of adenoidectomy which will be subsequently reported.