Associations of Nocturnal Blood Pressure With Cognition by Self-Identified Race in Middle-Aged and Older Adults: The GENOA (Genetic Epidemiology Network of Arteriopathy) Study

Abstract

Background: Whether the association of blood pressure (BP) during sleep (nocturnal BP) with cognition differs by race is unknown.

Methods and results: Participants in the GENOA (Genetic Epidemiology Network of Arteriopathy) Study underwent ambulatory BP measurements, brain magnetic resonance imaging, and cognitive function testing (the Rey Auditory Verbal Learning Test, the Digit Symbol Substitution Task, and the Trail Making Test Part B) between 2000 and 2007. We examined multivariable linear regression models of the nocturnal BP-cognition association. Among 755 participants (mean age, 63 years; 64% women; 42% self-identified black race; 76% taking antihypertensive medication), mean nocturnal systolic BP (SBP)/diastolic BP was 126/69 mm Hg, daytime SBP/diastolic BP level was 139/82 mm Hg, and mean reduction in SBP from day to night (dipping) was 9%. Among the entire sample, a race interaction was observed in Digit Symbol Substitution Task and Trail Making Test Part B (both P<0.15). Race-stratified analyses showed that a 1-SD increase in nocturnal SBP levels was associated with poorer Digit Symbol Substitution Task and log-transformed Trail Making Test Part B scores (unstandardized regression coefficient [95% confidence interval]: -1.98 [-3.28 to -0.69] and 0.06 [0.004-0.12]; both P<0.05) in black but not white individuals. Additional adjustments for white matter hyperintensity volumes or brain atrophy, measured via brain magnetic resonance imaging, did not change the results. Results were similar when nocturnal SBP dipping was assessed as the exposure, yet daytime SBP levels yielded no association with cognition.

Conclusions: Nocturnal SBP measurements may be useful in assessing the potential risk for lower cognitive function in middle-aged and older adults, particularly in black individuals.

Keywords: blood pressure; cognition; nocturnal blood pressure; race and ethnicity.

Figure 1

Cognitive function by nocturnal systolic blood pressure (SBP) dipping quartiles. A, Adjusted odds ratio (OR; 95% confidence interval [CI]) of low Mini‐Mental State Examination (MMSE; defined as the lowest quartile of the distribution of the MMSE scores) by nocturnal SBP dipping quartiles. The Trail Making Test Part B (TMT‐B) scores were log transformed because of skewed distributions. The first quartile group was defined as a reference. B through E, Adjusted means (95% CIs) of each cognitive test score by nocturnal SBP dipping quartile. The first quartile group was defined as a reference. Adjustment factors included age, sex, race, education, clinical characteristics (body mass index, estimated glomerular filtration rate, prevalent diabetes mellitus, duration of hypertension, use of antihypertensive medications, prevalent stroke, and clinic SBP levels), and 24‐hour mean SBP levels. All P values shown were for trend tests. Statistical significance was defined as P<0.05. DSST indicates Digit Symbol Substitution Task; and RAVLT, Rey Auditory Verbal Learning Test.