Midlife blood pressure is associated with the severity of white matter hyperintensities: analysis of the UK Biobank cohort study

Abstract

Aims: White matter hyperintensities (WMH) progress with age and hypertension, but the key period of exposure to elevated blood pressure (BP), and the relative role of systolic BP (SBP) vs. diastolic BP (DBP), remains unclear. This study aims to determine the relationship between WMH and concurrent vs. past BP.

Methods and results: UK Biobank is a prospective community-based cohort of 40-69-year olds from 22 centres, with magnetic resonance imaging in a subgroup of over 40 000 people at 4-12 years after baseline assessment. Standardized associations between WMH load (WMH volume normalized by total white matter volume and logit-transformed) and concurrent vs. past BP were determined using linear models, adjusted for age, sex, cardiovascular risk factors, BP source, assessment centre, and time since baseline. Associations adjusted for regression dilution bias were determined between median WMH and usual SBP or DBP, stratified by age and baseline BP.In 37 041 eligible participants with WMH data and BP measures, WMH were more strongly associated with concurrent SBP [DBP: β = 0.064, 95% confidence interval (CI) 0.050-0.078; SBP: β = 0.076, 95% CI 0.062-0.090], but the strongest association was for past DBP (DBP: β = 0.087, 95% CI 0.064-0.109; SBP: β = 0.045, 95% CI 0.022-0.069), particularly under the age of 50 (DBP: β = 0.103, 95% CI 0.055-0.152; SBP: β = 0.012, 95% CI -0.044 to 0.069). Due to the higher prevalence of elevated SBP, median WMH increased 1.126 (95% CI 1.107-1.146) per 10 mmHg usual SBP and 1.106 (95% CI 1.090-1.122) per 5 mmHg usual DBP, whilst the population attributable fraction of WMH in the top decile was greater for elevated SBP (19.1% for concurrent SBP; 24.4% for past SBP). Any increase in BP, even below 140 for SBP and below 90 mmHg for DBP, and especially if requiring antihypertensive medication, was associated with increased WMH.

Conclusions: WMH were strongly associated with concurrent and past elevated BP with the population burden of severe WMH greatest for SBP. However, before the age of 50, DBP was more strongly associated with WMH. Long-term prevention of WMH may require control of even mildly elevated midlife DBP.

Keywords: Diastolic blood pressure; Hypertension; Magnetic resonance imaging; Small vessel disease; White matter hyperintensities.

Figure 1

Increase in median white matter hyperintensities across usual blood pressure groups, stratified by age. Median white matter hyperintensities are presented on a logarithmic scale as a ratio of median white matter hyperintensities per patient group compared to the reference group (age < 50 years and systolic blood pressure <120 mmHg or diastolic blood pressure <70 mmHg), stratified by age (<50, 50–60, ≥60 years) and blood pressure at baseline (diastolic blood pressure <70, 70–80, 80–90, ≥90 mmHg; systolic blood pressure <120, 120–130, 130–140, ≥140 mmHg). Diastolic blood pressure plotted in blue on the left and systolic blood pressure plotted in grey on the right. Lines represent a linear fit across all points within age group on a logarithmic scale.

Figure 2

Median white matter hyperintensities load, stratified by age, definition of hypertension, and treatment at baseline. HT, hypertensive; NT, normotensive.