Blood Pressure is Positively Associated with Glycosylated Hemoglobin but not the Hypoxic Ventilatory Response in Obstructive Sleep Apnea Patients

Abstract

Study ObjectiveThe hypoxic ventilatory response (HVR) is considered an index of peripheral chemoreflex sensitivity. Previous work has suggested that an increase in the HVR is responsible for acute nocturnal blood pressure surges that ultimately lead to chronic elevations in blood pressure. Previous results have also indicated that glycosylation of hemoglobin (HbA1C) impairs nitric oxide (NO) relaxation of vascular smooth muscle leading to chronic elevations in blood pressure. Given these findings, we hypothesized that the HVR and HbA1C would be positively associated with blood pressure during sleep before and 3 weeks after treatment with mild intermittent hypoxia (MIH) in humans with obstructive sleep apnea (OSA).MethodsEight hypertensive participants with OSA were exposed to twelve 2‐minute episodes of hypoxia (PETO2 ≈ 50 mmHg) separated by 2‐minute normoxic intervals during wakefulness for 15 days over 3 weeks. PETCO2 was sustained at 2 mmHg above baseline during MIH administration. HbA1C was measured prior to completion of the protocol on Day 1 and 15. Moreover, blood pressure was measured over a period of 24 hours, before and after completion of the MIH protocol.ResultsOn Day 1, the HVR was negatively associated with HbA1C (−0.74, p < 0.04) and systolic (SBP) (R = −0.73, p < 0.05), diastolic (DBP) (R = −0.69, p < 0.05) and mean arterial blood pressure (MAP) (R = −0.71, p < 0.05) recorded during the sleep period of a 24 hour cycle. In contrast, HbA1C was positively associated with measures of SBP (R = 0.82, p < 0.01), DBP (R = 0.81, p < 0.01) and MAP (R = 0.86, p < 0.01) during sleep. Following treatment with MIH, blood pressure during sleep decreased (Before vs. After MIH: SBP 131.9 ± 4.1 vs. 122.9 ± 3.1, DBP 77.8 ± 3.9 vs. 72.8 ± 3.9, MAP 95.8 ± 3.8 vs. 89.5 ± 3.4, P < 0.02). The change (Δ) in SBP during sleep following MIH was correlated to the Δ HbA1C (R = 0.85, p ≤ 0.01). Moreover, the degree to which SBP dipped before vs. after MIH was also correlated to Δ HbA1C (R = 0.87, p ≤ 0.01). In contrast, Δ HVR was not correlated to the Δ in blood pressure parameters.ConclusionAlthough peripheral chemoreflex sensitivity may be responsible for acute surges in blood pressure, this factor does not appear to be a strong predictor of chronic changes in blood pressure. In contrast, modifications in local signaling molecules may ultimately have the greatest impact on chronic modifications in blood pressure. Indeed, our findings lend support to previous results which suggest that increases in blood pressure may be coupled to the impairment of nitric oxide in response to glycosylated hemoglobin. This relationship may exist in the presence of a blunted HVR.Support or Funding InformationDepartment of Veterans Affairs (I01CX000125 & 15SRCS003 JHM, 1IK1RX002945 GSP), and NIH (R56HL142757 & R01HL142757 JHM).