AUTHOR=Pham Luu V. , Schwartz Alan R. , Jun Jonathan C. 

TITLE=Oxyhemoglobin Saturation Overshoot Following Obstructive Breathing Events Mitigates Sleep Apnea-Induced Glucose Elevations

JOURNAL=Frontiers in Endocrinology

VOLUME=Volume 9 - 2018

YEAR=2018

URL=https://www.frontiersin.org/journals/endocrinology/articles/10.3389/fendo.2018.00477

DOI=10.3389/fendo.2018.00477

ISSN=1664-2392

ABSTRACT=Background: Obstructive sleep apnea (OSA) causes nocturnal intermittent hypoxia (IH) and is associated with disturbances in glucose regulation and diabetes. We recently showed that CPAP withdrawal in patients with moderate-severe OSA leads to dynamic within-night glucose elevations in proportion to the severity of IH. In the present study, we examined the relationships between nocturnal IH signal dynamics and nocturnal glucose. We hypothesized that specific patterns of IH lead to alterations in glucose throughout the night.
 Methods: 27 non-diabetic CPAP-adherent patients with moderate-severe OSA (18 men, age 48.9 ± 10.3 years and BMI 36.3 ± 7.6 kg/m2) underwent polysomnography with concurrent blood sampling every 20-30 minutes, on CPAP and after 3-night CPAP withdrawal in random order. To characterize IH dynamics, we calculated mean SPO2, and the frequency and magnitude of SPO2 changes relative to baseline during the 10-minute period preceding each glucose measurement during the CPAP withdrawal night. We further characterized SPO2 variations by the frequency and amplitude of SPO2 reductions and rises (desaturations and resaturations, respectively), which were calculated relative to a baseline defined by the mean spo2 during CPAP night. We used mixed-effects linear regressions models to examine the associations between nocturnal plasma glucose and AHI, mean SPO2, and the frequency of desaturations and resaturations at varying amplitudes. We constructed multi-variable models to examine the independent associations between these IH components, OSA severity and nocturnal glucose. 
Results: In single predictor models, glucose was not significantly associated with AHI, or with the frequency of desaturations of any amplitude (p>0.05). Glucose was inversely associated with mean SPO2 (p<0.01). SPO2 peaks that did not fully recover to baseline were associated with a 0.49 mg/dL glucose elevation (p=0.01). In contrast, SPO2 peaks that exceeded baseline by >1% predicted a 0.46 mg/dL glucose reduction.  In multivariable models, the protective effect of resaturation overshoot against glucose elevation was significant with adjustment for mean SPO2 and AHI (p<0.05). 
Conclusions:  Hypoxia severity predicted nocturnal glucose elevations. Resaturation overshoot independently predicted glucose change in a dose-dependent manner. These findings suggest that hypoxia from OSA contributes to nocturnal glucose elevations, which can be mitigated by maneuvers that improve oxygenation.