Comparison of various treatments for sleep bruxism using determinants of number needed to treat and effect size.

OBJECTIVE

Sleep bruxism (SB) is associated with temporomandibular pain, headaches, tooth wear, and disruption of the bed partner's sleep. The aim of this report was to compare SB treatments from various experimental studies to guide the selection of a treatment for a large sample size study.

METHODS

After a literature search, randomized controlled studies of 7 pharmacologic treatments and 3 oral devices were included. The number needed to treat (NNT) was calculated from raw data from the sleep laboratory at the Hôpital du Sacré-Coeur, Montréal or from published articles when sufficient data were available. The effect size (ES) was calculated for all included studies. In the most effective treatments, the NNT ranged from 1 to 4, while a high ES was above 0.8.

RESULTS

The treatments with the best NNT and ES results were the mandibular advancement device (MAD) and clonidine. The NNT (+/-95% CI) and ES were 2.2 (1.4 to 5.3) and 1.5 for the MAD, and 3.2 (1.7 to 37.3) and 0.9 for clonidine, respectively. An NNT of 3.8 (1.9 to -69.4) and an ES of 0.6 were observed with the occlusal splint, with a reduction of 42% in the SB index. NNT could not be calculated for clonazepam, although the ES was 0.9.

CONCLUSIONS

Although the NNT and ES results seem to indicate that the MAD and clonidine are the most promising experimental treatments, both treatments were associated with side effects (ie, discomfort for the MAD; REM suppression and morning hypotension for clonidine). The occlusal splint and clonazepam seem to be acceptable short-term alternatives, although further longitudinal, large sample size randomized controlled trials in SB management are needed.