Orthostatic Hypotension and Risk of Clinical and Subclinical Cardiovascular Disease in Middle-Aged Adults

Abstract

Background: Although orthostatic hypotension (OH) is a well-recognized manifestation of neuropathy and hypovolemia, its contribution to cardiovascular disease (CVD) risk is controversial.

Methods and results: Participants with OH, defined as a decrease in blood pressure (systolic ≥20 mm Hg or diastolic ≥10 mm Hg) from the supine to standing position, were identified during the first visit of the ARIC (Atherosclerosis Risk in Communities) Study (1987-1989) within 2 minutes of standing. All participants were followed up for the development of myocardial infarction, heart failure, stroke, fatal coronary heart disease (CHD), any CHD (combination of silent, nonfatal, and fatal CHD or cardiac procedures), and all-cause mortality. Participants were assessed for carotid intimal thickness and plaque during the first visit. Detectable high-sensitivity troponin T (≥5 ng/L) and elevated NT-proBNP (N-terminal pro-B-type natriuretic peptide; ≥100 pg/mL) were determined in blood collected during the second visit (1990-1992). All associations were adjusted for known CVD risk factors. In 9139 participants (57% women; 23% black; mean age, 54±5.7 years), 3% had OH. During follow-up (median, 26 years), OH was associated with myocardial infarction (hazard ratio [HR], 1.88; 95% confidence interval [CI], 1.44-2.46), congestive heart failure (HR, 1.65; 95% CI, 1.34-2.04), stroke (HR, 1.83; 95% CI, 1.35-2.48), fatal CHD (HR, 2.77; 95% CI, 1.93-3.98), any CHD (HR, 2.00; 95% CI, 1.64-2.44), and all-cause mortality (HR, 1.68; 95% CI, 1.45-1.95). OH was also associated with carotid intimal thickness (β, 0.05 mm; 95% CI, 0.04-0.07 mm), carotid plaque (odds ratio, 1.51; 95% CI, 1.18-1.93), detectable high-sensitivity troponin T (odds ratio, 1.49; 95% CI, 1.16-1.93), and elevated NT-proBNP (odds ratio, 1.92; 95% CI, 1.48-2.49).

Conclusions: OH identified in community-dwelling middle-aged adults was associated with future CVD events and subclinical CVD. Further research is necessary to establish a causal role for OH in the pathogenesis of CVD.

Keywords: NT‐proBNP; cardiovascular disease; heart failure; high‐sensitivity troponin; mortality; orthostatic hypotension; stroke.

Figure 1

Adjusted restricted cubic splines of the relationship (solid line) between postural change in systolic blood pressure (SBP) or diastolic blood pressure (DBP; per mm Hg) with incident myocardial infarction (A and B), incident congestive heart failure (C and D), incident stroke (E and F), fatal coronary heart disease (G and H), any coronary heart disease (I and J), and mortality from any cause (K and L). Participants were followed up through December 31, 2015. The models were expressed relative to the 50th percentile of postural change in SBP or DBP (0.85 and 3.5 mm Hg, respectively), with 4 knots specified using Harrell's method. All models used Cox proportional hazards models to determine hazard ratios shown on natural log scale and were adjusted for age, sex, race‐center, estimated glomerular filtration rate, body mass index, resting heart rate, high‐density lipoprotein cholesterol, low‐density lipoprotein cholesterol, triglycerides, total cholesterol, diabetes mellitus status, hypertension status, antihypertensive medication use in the past 2 weeks, self‐reported dizziness, alcohol consumption, education level, physical activity, and smoking status. The plot was truncated at the 0.5th and 99.5th percentiles of postural change in SBP or DBP. In addition, overlaid are kernel density plots depicting the distribution of postural change in systolic or diastolic blood pressure by participants who had the outcome of interest (red dash) vs those who did not have the outcome of interest (solid).

Figure 2

Adjusted restricted cubic splines of the relationship (solid line) between postural change in systolic blood pressure (SBP) or diastolic blood pressure (DBP; per mm Hg) with carotid intimal thickness (measured in mm; A and B) or the presence of carotid plaque (C and D). Gray shade represents 95% confidence intervals. The models were expressed relative to the 50th percentile of postural change in SBP or DBP (0.85 and 3.5 mm Hg, respectively), with 4 knots specified using Harrell's method. Linear regression was used for carotid intimal thickness. Logistic regression was used for the presence of carotid plaque (odds ratios shown on natural log scale). Linear or logistic regression models were adjusted for age, sex, race‐center, estimated glomerular filtration rate, body mass index, resting heart rate, high‐density lipoprotein cholesterol, low‐density lipoprotein cholesterol, triglycerides, total cholesterol, diabetes mellitus status, hypertension status, antihypertensive medication use in the past 2 weeks, self‐reported dizziness, alcohol consumption, education level, physical activity, and smoking status. The plot was truncated at the 0.5th and 99.5th percentiles of postural change in SBP or DBP. In addition, overlaid are kernel density plots depicting the distribution of postural change in SBP or DBP by participants with carotid plaque (red dash) vs those without carotid plaque (solid). Because carotid intimal thickness is a continuous outcome, the kernel density depicts the overall distribution alone (solid).

Figure 3

Adjusted restricted cubic splines of the relationship (solid line) between postural change in systolic blood pressure (SBP) or diastolic blood pressure (DBP; per mm Hg) with detectable high‐sensitivity troponin T (hs‐cTNT) ≥5 ng/L (A and B) or elevated NT‐proBNP (N‐terminal pro‐B‐type natriuretic peptide) ≥100 pg/mL (C and D). Gray shade represents 95% confidence intervals. The models were expressed relative to the 50th percentile of postural change in SBP or DBP (0.85 and 3.5 mm Hg, respectively), with 4 knots specified using Harrell's method. All models used logistic regression to determine odds ratios shown on natural log scale and were adjusted for age, sex, race‐center, estimated glomerular filtration rate, body mass index, resting heart rate, high‐density lipoprotein cholesterol, low‐density lipoprotein cholesterol, triglycerides, total cholesterol, diabetes mellitus status, hypertension status, antihypertensive medication use in the past 2 weeks, self‐reported dizziness, alcohol consumption, education level, physical activity, and smoking status. The plot was truncated at the 0.5th and 99.5th percentiles of postural change in SBP or DBP. In addition, overlaid are kernel density plots depicting the distribution of postural change in SBP or DBP by participants with either detectable hs‐cTnT or elevated NT‐proBNP present (red dash) vs absent (solid).