Snoring and arousals: a retrospective analysis

The purpose of this retrospective analysis was to search for possible associations between snoring and arousals. We searched our data base containing more than 2,000 records and selected only patients who 1) had objective measurements of snoring, 2) were not taking sedating medication, 3) did not have sleep apnea (apnea/ hypopnea index < 10) and 4) did not have periodic leg movements (myoclonus index < 5). This procedure left 367 patients available for analysis. We hypothesized that arousals observed in these patients were associated with snoring, and we performed univariate and multivariate regression analysis with arousals as the dependent variable, and age, body mass index, snoring, maximum nocturnal sound intensity and nocturnal oxygen saturation as the independent variables. The results showed that only snoring and mean nocturnal oxygen saturation were significant (p < 0.05) but weak determinants of arousals, accounting for only 7% of their variance. To examine whether snorers have more arousals than non-snorers, we compared a control group of non-snorers (< 50 snores/hour of sleep), with a group of heavy snorers (> 400 snores/hour of sleep). We found that the arousal index was significantly but weakly higher in snorers than non-snorers [mean +/- standard deviation (SD) = 14 +/- 8 vs. 10 +/- 6, p < 0.002]. Conversely, patients within the highest arousal quartile snored significantly more than those within the lowest quartile (snoring index 293 +/- 292 vs. 179 +/- 282, p < 0.008, respectively). We conclude that despite the limitations of this retrospective analysis, there appears to be an association between snoring and arousals, warranting further, properly designed prospective studies.