SFSS - Svensk förening för sömnforskning och sömnmedicin

Holger Kraiczi

Department of Clinical Pharmacology, Sahlgrenska Universitetssjukhus

Cardiovascular dysfunction in obstructive sleep apnea

Abstract: This thesis addresses relationships between the presence or severity of obstructive sleep apnea (OSA) and blood pressure (BP) as well as other signs of cardiovascular dysfunction.

In a cross-sectional investigation of a sleep laboratory cohort of 81 patients, a dose response relationship was observed between the number of apneas plus hypopneas per hour of sleep (apnea-hypopnea index, AHI; measured with polysomnography) and office systolic and diastolic BP (SBP and DBP), daytime ambulatory DBP as well as nighttime ambulatory SBP and DBP (all ambulatory measurements taken at 20-minute intervals). In essence, these relationships remained significant after adjustments for age, gender, use of antihypertensive agents, smoking, body mass index, history of coronary artery disease, hypercholesterolemia or serum C-peptide concentration. In contrast, relationships between AHI and left ventricular (LV) muscle thickness (M-mode echocardiography) were constituted by a complex interaction of AHI and LV muscle size with BP and other confounding variables.

In a more homogeneous sample of 20 male subjects with OSA but without clinically overt cardiovascular disease or risk, increased degree of apnea-related oxygen hemoglobin desaturation during sleep, but not AHI, was related to the thickness of the interventricular septum in diastole, to impaired LV diastolic function and to reduced endothelium-dependent dilation of the brachial artery.

Long-term intermittent hypoxia (LTIH) is one important characteristic of OSA. Exposure of Wistar-Kyoto (WKY) and spontaneously hypertensive rats (SHR) to LTIH (hypoxia/normoxia cycles of 90+90 sec, 80 hours daily for 70 days) resulted in an augmented BP increase but a dampened heart rate response to a brief hypoxic stimulus. In the SHR, but not WKY, LV weight-to-body weight ratio and LV levels of atrial natriuretic peptide were increased after LTIH, when compared with normoxic control animals. Unstimulated BP was not influenced by LTIH in the current setting.

Augmented BP response to short-term hypoxia has also been reported in OSA patients, and an increased vasoconstrictor sensitivity of resistance vessels was suggested as one explanation for this phenomenon. In the present thesis, this hypothesis was supported by the finding that infusion of angiotensin II into the brachial artery induced a stronger forearm vasoconstriction (calculated from plethysmographically measured blood flow velocity and intraarterial BP) in OSA patients than in control subjects.

A clinical study in 40 OSA patients treated during two periods of 6 weeks each (incomplete block design) demonstrated that atenolol (50 mg o.d.) reduced office DBP more effectively than amlodipine (5 mg), enalapril (20 mg), hydrochlorothiazide (25 mg) or losartan (50 mg). Among the secondary outcome variables, nighttime ambulatory SBP and DBP were more effectively lowered with atenolol than with amlodipine, enalapril or losartan.

This thesis supports the hypothesis that OSA is independently associated with several aspects of cardiovascular dysfunction. Moreover, the presence of OSA may have direct consequences for cardiovascular risk factor intervention.

Svensk förening för sömnforskning och sömnmedicin