Cardiovascular Autonomic Function and Incident Chronic Obstructive Pulmonary Disease Hospitalizations in Atherosclerosis Risk in Communities

Abstract

Rationale: The autonomic nervous system extensively innervates the lungs, but its role in chronic obstructive pulmonary disease (COPD) outcomes has not been well studied. Objective: We assessed relationships between cardiovascular autonomic nervous system measures (heart rate variability [HRV] and orthostatic hypotension [OH]) and incident COPD hospitalization in the multicenter ARIC (Atherosclerosis Risk In Communities) study. Methods: We used Cox proportional hazards regression models to estimate hazard ratios and 95% confidence intervals between baseline (1987-1989) autonomic function measures (HRV measures from 2-minute electrocardiograms and OH variables) and incident COPD hospitalizations through 2019. Adjusted analyses included demographic data, smoking status, lung function, comorbidities, and physical activity. We also performed analyses stratified by baseline airflow obstruction. Results: Of the 11,625 participants, (mean age, 53.8 yr), 56.5% were female and 26.3% identified as Black. Baseline mean percentage predicted forced expiratory volume in 1 second was 94 ± 17% (standard deviation), and 2,599 participants (22.4%) had airflow obstruction. During a median follow-up time of 26.9 years, there were 2,406 incident COPD hospitalizations. Higher HRV (i.e., better autonomic function) was associated with a lower risk of incident COPD hospitalization. Markers of worse autonomic function (OH and greater orthostatic changes in systolic and diastolic blood pressure) were associated with a higher risk of incident COPD hospitalization (hazard ratio for the presence of OH, 1.5; 95% confidence interval, 1.25-1.92). In stratified analyses, results were more robust in participants without airflow obstruction at baseline. Conclusions: In this large multicenter prospective community cohort, better cardiovascular autonomic function at baseline was associated with a lower risk of subsequent hospitalization for COPD, particularly among participants without evidence of lung disease at baseline.

Keywords: autonomic nervous system; chronic obstructive; cohort studies; disease exacerbation; pulmonary disease.

Figure 1.

Associations between log-transformed heart rate variability measures and incident chronic obstructive pulmonary disease hospitalizations in all participants (black) and stratified by those with and without baseline airflow obstruction (gray). Hazard ratios represent the change in risk for a 1–standard deviation increase in the exposure of interest. Hazard ratios are from proportional hazards models and include age, sex, education, race-center, forced expiratory volume in 1 second (FEV₁) percentage predicted, current smoking status, smoking pack-years, coronary artery disease, heart failure, diabetes, body mass index, and physical activity as covariates. Participants with airflow obstruction were defined as those with an FEV₁/forced vital capacity ratio <0.7 at baseline; participants without airflow obstruction had an FEV₁/forced vital capacity ratio ⩾0.7 at baseline. There were 11,625 participants; 2,599 had airflow obstruction and 9,026 did not. Heart rate variability measures include standard deviation of normal R-R intervals, root mean square of sequential differences in normal R-R intervals, mean R-R interval, low-frequency power, high-frequency power, and low-frequency/high-frequency ratio (LF/HF). HF = high-frequency power; LF = low-frequency power; RMSSD = root mean square of successive differences in normal R-R intervals; SDNN = standard deviation of normal R-R intervals.

Figure 2.

Associations among orthostatic hypotension (OH), changes in systolic and diastolic blood pressure, and incident chronic obstructive pulmonary disease hospitalization for all participants (black) and stratified by those with and without baseline airflow obstruction (gray). Adjusted hazard ratios represent the change in risk for the presence versus absence of OH and a 10–mm Hg decrease in systolic and diastolic blood pressure. Hazard ratios are from proportional hazards models and include age, sex, education, race-center, forced expiratory volume in 1 second (FEV₁) percentage predicted, current smoking status, smoking pack-years, coronary artery disease, heart failure, diabetes, body mass index, and physical activity as covariates. Participants with airflow obstruction were defined as those with an FEV₁/forced vital capacity ratio <0.7 at baseline; participants without airflow obstruction had an FEV₁/forced vital capacity ratio ⩾0.7 at baseline. There were 10,574 participants with OH measurements; 2,403 had airflow obstruction and 8,171 did not. Orthostatic hypotension is defined as a 20–mm Hg decrease in systolic blood pressure and/or a 10–mm Hg decrease in diastolic blood pressure. DBP = diastolic blood pressure; SBP = systolic blood pressure.

Figure 3.

Interactions of heart rate variability and (A) sex, (B) smoking status, and (C) lung function on chronic obstructive pulmonary disease hospitalization risk. Adjusted hazard ratios represent the change in risk for a 1–standard deviation increase in the exposure of interest. Models are adjusted for age, sex, height, education, race-center, forced expiratory volume in 1 second (FEV₁) percentage predicted, current smoking status, pack-years of smoking, coronary artery disease, heart failure, diabetes, body mass index, and physical activity. Adjusted hazard ratios represent the change in risk for the presence of orthostatic hypotension or a 10–mm Hg decrease in systolic or diastolic blood pressure. Interaction P values were obtained by including cross-product terms in the models. For these analyses, the 10,574 participants were stratified by sex (n = 5,953 female, 4,621 male), smoking status (n = 2,782 current, 3,344 former, and 4,442 never-smokers), and lung function (n = 900 with FEV₁ <70% predicted, 1,108 with FEV₁ 71–80% predicted, 2,150 with FEV₁ 81–90% predicted, 2,959 FEV₁ with 91–100% predicted, and 4,421 with FEV₁ >100% predicted). HF = high-frequency power; LF = low-frequency power; RMSSD = root mean square of successive differences in normal R-R intervals; SDNN = standard deviation of normal R-R intervals.

Figure 4.

Interactions of orthostatic hypotension and (A) sex, (B) smoking status, and (C) lung function on chronic obstructive pulmonary disease hospitalization risk. Adjusted hazard ratios represent the change in risk for the presence versus absence of orthostatic hypotension and a 10–mm Hg decrease in systolic or diastolic blood pressure. DPB = diastolic blood pressure; SBP = systolic blood pressure. Models are adjusted for age, sex, height, education, race-center, FEV₁ percent predicted, current smoking status, pack years of smoking, coronary artery disease, heart failure, diabetes, BMI, and physical activity. Adjusted hazard ratios represent the change in risk for the presence of OH or a 10 mm Hg decrease in SBP or DBP. Interaction p-values were obtained by including cross-product terms in the models. For these analyses, the 10,574 participants were stratified by sex (n = 5,953 females and 4,621 males), smoking status (n = 2,782 current, 3,344 former, and 4,442 never smokers) and lung function (n = 900 FEV₁ <70% predicted, 1,108 FEV₁ 71–80% predicted, 2,150 FEV₁ 81–90% predicted, 2,959 FEV₁ 91–100% predicted, and 4,421 FEV₁ >100% predicted). COPD = chronic obstructive pulmonary disease; DPB = diastolic blood pressure; OH = orthostatic hypotension; SBP = systolic blood pressure.