Effect of oxygen desaturation threshold on determination of OSA severity during weight loss

Purpose Weight loss leads to improvement of obstructive sleep apnea (OSA), based on frequency of respiratory events (apnea-hypopnea index, AHI). However, AHI does not incorporate the severity of individual obstruction events. The American Academy of Sleep Medicine suggests two alternative oxygen desaturation thresholds (ODT) for scoring of hypopneas. We hypothesize that lowering the ODT level increases the determined impact of weight loss on OSA severity. We investigate this during weight change with AHI and adjusted AHI. Adjusted AHI is a novel parameter incorporating both severity and number of the events. Methods Ambulatory polygraphic data of 54 OSA patients (F 15/M 39, 51.7 ± 8.4 years), divided into weight loss (>5 %, n = 20), control (weight change 0–5 %, n = 26), and weight gain (>5 %, n = 8) groups, were evaluated at baseline and after 5-year follow-up. Effect of ODT (ODT2%–ODT6%) on AHI and adjusted AHI was investigated. Results The greatest changes in AHI (decrease in weight loss group and increase in weight gain group) were observed with ODT2%. Changes in AHI diminished with increasing ODT. In weight loss group, adjusted AHI showed a similar but non-significant trend. In contrast, the higher ODT was used in weight gain group, the greater increase in adjusted AHI resulted. Using adjusted AHI instead of AHI, led to a smaller number of patients (20 vs. 55 %, ODT3%) whose OSA severity category improved along weight loss. Conclusions Weight loss significantly reduced AHI. This reduction was highly dependent on selected ODT. The change in adjusted AHI did not occur in the same extent. This was expected as the more severe events which tend to remain during the weight loss have greater importance in adjusted AHI, while the event severity is neglected in AHI.