Skip navigation

Autoinflation for glue ear


Clinical bottom line

Autoinflation for glue ear may be of benefit. In three trials using manufactured autoinflation balloons in 386 children, 32% improved using the balloon compared with 14% without. The NNT was 5.5 (3.8 to 10).


DD Reidpath et al. Systematic review of autoinflation for treatment of glue ear in children. British Medical Journal 1999 318: 1177-1178.


The review looked at MEDLINE and the Cochrane Library for randomised trials of autoinflation. A further unpublished randomised trial was found.


There were six studies in all, three using balloon autoinflation and three other studies. None of the studies used blinding when measuring outcomes. No two studies were comparable in terms of length of treatment, method of inflation and level of data analysis (child versus ear). Three used manufactured nasal balloons, two toy balloons, and one an anaesthetic mask.

For the three using manufactured nasal balloons, 63/195 (32%) of children were cured with the balloon, compared with 27/191 (14%) without it (Figure 1). The relative benefit was 2.3 (1.6 to 3.4) and the NNT was 5.5 (3.8 to 10).

Figure 1: Cure rates with manufactured nasal balloons compared with no treatment controls.


There is some limited evidence that manufactured nasal balloons are helpful, but the trials were of relatively poor quality.