Case control study of cerebrovascular damage defined by magnetic resonance imaging in patients with OSA and normal matched control subjects

Study Objectives: To assess whether MRI detectable evidence of silent cerebrovascular disease is more prevalent in patients with obstructive sleep apnea (OSA) when compared to carefully matched control subjects. Design and Setting: Case-control study of patients with OSA attending a specialist sleep clinic and matched control subjects drawn from the normal community. Participants: Forty-five sleep clinic patients with moderate to severe OSA and excessive daytime sleepiness, matched to 45 control subjects without excessive sleepiness or evidence of OSA on a sleep study. Matched variables included age, body mass index (BMI), alcohol and cigarette consumption, treated hypertension, and ischaemic heart disease. Interventions: N/A Measurements and Results: All subjects underwent 24-hour ambulatory blood pressure recordings (before treatment in OSA patients) and sagittal T1, axial T2, and coronal dual echo cerebral MRI imaging to detect clinically silent abnormalities related to hypertensive cerebrovascular disease; areas of high signal foci in deep white matter (DWM), lacunae, and periventricular hyperintensity. Lacunae/high signal foci in DWM and/or periventricular hyperintensity were present in 15 (33%) OSA subjects and 16 (35%) controls, despite significant increases in mean daytime diastolic blood pressure (4.6mmHg, p<0.05), and both nighttime diastolic (7.2mmHg, p<0.001) and systolic blood pressures (9.2mmHg, p<0.05) in OSA subjects. These data exclude more than a 17% excess prevalence of MRI detected minor cerebrovascular disease in the OSA patients, with 95% confidence. Conclusions: Sub-clinical cerebrovascular disease is prevalent in both clinic patients with OSA and their matched control subjects. Despite the increased arterial blood pressures, there is, however, no apparent excess of MRI-evident subclinical cerebrovascular disease in patients with OSA compared to appropriately matched control subjects.