Patients with obstructive sleep apnea (OSA) have been described to have increased levels of inflammatory cytokines (particularly TNF-α) and have severely disturbed sleep architecture. Serum inflammatory markers, even in normal individuals, have been associated with abnormal sleep architecture. Not much is known about the role the TNF receptor plays in the inflammation of OSA nor if it is associated with changes in sleep architecture or arousals during the night. We hypothesized that the TNF receptor might play an important role in the inflammation as well as sleep architecture changes in patients with OSA.
Thirty-six patients with diagnosed (AHI > 15) but untreated OSA were enrolled in this study. Baseline polysomnograms as well as TNF-α and soluble TNF receptor I (sTNF-RI) serum levels were obtained on all patients. We evaluated the association between serum levels of TNF-α and sTNF-RI with various polysomongraphic characteristics, including sleep stages and EEG arousals.
sTNF-RI levels were significantly correlated with snore arousals (r value 0.449, p value 0.009), spontaneous movement arousals (r value 0.378, p value 0.025), and periodic limb movement arousals (r value 0.460, p value 0.008). No statistically significant correlations were observed with TNF-α to any polysomnographic variables. To control for statistical significance with multiple comparisons, a MANOVA was performed with TNF-α and sTNF-RI as dependent variables and sleep architecture measures and arousals as independent variables. The model for sTNF-RI was statistically significant (F value 2.604, p value 0.03), whereas the model for TNF-α was not, suggesting sleep quality significantly affects sTNF-RI. Hierarchal linear regression analysis demonstrated that sTNF-RI was independently associated with spontaneous movement arousal index scores after controlling for age, body mass index, and sleep apnea severity.
These findings suggest that sTNF-RI is associated with arousals during sleep, but not with other measures in patients with OSA.