Abstract
Background
Premature atrial contractions (PACs) are associated with increased risk of atrial fibrillation (AF) and ischemic stroke. Although lifestyle and risk factor modification reduces AF incidence, their relationship to PACs frequency is unclear. We assessed the association of Life’s Simple 7 (LS7) and individual LS7 factors in midlife with PACs frequency in late life in the Atherosclerosis Risk in Communities (ARIC) Study.
Methods
We followed 1924 participants from ARIC clinic Visit 3 (1993–95) to Visit 6 (2016–17) when a 2-week continuous heart rhythm monitor (Zio®XT Patch) was applied. LS7 factors were assessed at Visit 3 and a composite score was calculated. PACs frequency was categorized as minimal (<0.1%), occasional (≥0.1%-5%) and frequent (>5%). Logistic regression was used to evaluate the association of LS7 score and individual factors with PACs frequency.
Results
Each 1-point LS7 score increase was associated with lower odds of frequent PACs vs. no PACs (OR [95% CI]: 0.87 [0.78, 0.98]) and frequent PACs vs. occasional PACs (OR [95% CI]: 0.88 [0.79, 0.98]). Of the individual LS7 factors, compared with ideal physical activity, poor physical activity was associated with 81% higher odds of frequent PACs vs. no PACs. Compared with ideal BMI, poor BMI was associated with 41% higher odds of occasional PACs vs. no PACs.
Conclusion
Lifestyle risk factors, particularly physical activity and BMI, are associated with higher odds of PACs frequency. More research is needed to determine whether modifying these risk factors in midlife would prevent frequent PACs, and thereby prevent AF and stroke in older age.
Keywords: Life’s simple 7, premature atrial contractions, risk prediction, prevention
Introduction
Premature atrial contractions (PACs) are frequently encountered and have long been considered a benign entity. However, more recently this assumption has been questioned because a growing number of studies have shown an association between higher PACs frequency and greater risk of atrial fibrillation (AF).1,2 Further, PACs have also been found to be associated with an increased risk of ischemic stroke independent of clinical AF,3,4 though this has been postulated to be related to the occurrence of subclinical episodes of AF. The relationship between higher PACs frequency and increased AF risk has motivated efforts to suppress PACs, with elimination of PAC triggers of AF forming the basis of catheter ablation of paroxysmal AF. Despite the clinical significance of high frequency of PACs, the determinants are poorly understood. Prior studies examining risk factors for PACs have been limited by the cross-sectional nature of study designs and the imprecise quantification of PACs based on a single 24-hour Holter recording, which is confounded by day-to-day variability in PACs frequency.5,6
While lifestyle and modifiable risk factors in early to midlife are associated with higher risk of coronary heart disease (CHD) and AF in late life, the relationship between these risk factors in midlife and risk of PACs frequency in late life is unclear.7,8 The American Heart Association has defined a set of metrics that represents ideal cardiovascular health known as Life’s Simple 7 (LS7), which includes seven modifiable health behaviors and risk factors (physical activity, total cholesterol, diet, blood pressure, body mass index (BMI), fasting blood glucose, and smoking status).9 A more favorable LS7 risk factor profile is associated with lower AF incidence8 but its association with PACs frequency has not yet been studied. Therefore, in this study, we evaluated the prospective association of LS7 score and individual LS7 risk factors in midlife with PACs frequency in late life within the Atherosclerosis Risk in Communities (ARIC) Study, a prospective community-based cohort study.
METHODS
Study Population and Design
The ARIC study is a community based prospective cohort study that was initiated in 1987–1989. At inception, the ARIC study recruited 15,792 participants aged between 45–64 years from four USA communities (Forsyth County, NC; Jackson, MS; Washington County, MD; suburbs of Minneapolis-St. Paul, MN).10 Since the baseline exam, six additional visits have been completed: visit 2(1990–92), visit 3(1993–95), visit 4(1996–98), visit 5(2011–13), visit 6 (2016–17) and visit 7 (2018–19).
Visit 3 (1993–95) served as the baseline for this analysis. In 2016–17, 4,003 participants attended the Visit 6 exam when they were aged between 75–94 years. During this exam, all participants were invited to wear a Zio® XT Patch for a duration of 14 days; exclusion criteria included history of cardiac electronic device implantation or skin allergic reaction to adhesive tape. At the end of the recording period, participants removed the device and mailed the heart rhythm monitor to iRhythm Technologies Inc., where recorded ECG data were processed using a proprietary algorithm and a report was generated. Of 3,680 Visit 6 participants who were eligible to wear the Zio® XT Patch, 2,650 (72.0%) agreed to participate. Of the 2,650 devices, 34 (1.3%) were lost or returned without data resulting in a total of 2,616 Zio® XT Patch devices with analyzable data. We excluded participants with race other than black or white and race other than white from the Minneapolis and Washington County centers respectively due to small numbers (n=12), participants with AF (n=350) (prior diagnosis of AF or AF detected on Zio® XT Patch), who were missing 1 or more LS7 components at Visit 3 (n=156), did not have an echocardiogram performed at Visit 5 (n=165), or wore the Zio® XT Patch for less than 2 days (n=9), leaving 1,924 participants for this analysis (Figure 1). Participants with AF were excluded for two key reasons. First, the presence of AF will suppress manifestation and detection of PACs by the Zio XT Patch, thus rendering the quantification of PAC frequency inaccurate. Second, our primary goal is to examine the relationship of cardiovascular health to a well-established precursor to AF (frequent PACs) and not to AF itself.
Figure 1.
Flow of Study Participants: The Atherosclerosis Risk in Communities Study.
Life’s Simple 7
LS7 components and score were assessed at Visit 3. Cigarette smoking status was categorized as never, former or current. Former smokers were further classified according to quit date (>12 months versus =≤12 months). Weight and standing height were measured by trained field center staff, and body mass index (BMI) was calculated as measured weight in kilograms divided by standing height in meters squared. Systolic and diastolic blood pressure were measured 3 times, and the average of the last 2 measurements was used. Total blood cholesterol concentration was assessed by standard enzymatic procedures.11 Blood glucose levels were measured by the modified hexokinase/glucose-6-phosphate dehydrogenase method. Physical activity was assessed using the modified Baecke questionnaire, which defines 3 semicontinuous indices ranging from 1 (low) to 5 (high) for physical activity during sports, during leisure time, and at work.12 Total physical activity was then defined as minutes per week of moderate or vigorous exercise based on metabolic equivalent values and the number of months annually a participant partook in the activity. Usual dietary intake was determined using a modified Willett Food Frequency Questionnaire that was administered by trained interviewers to ensure accuracy and completeness.13 Nutritional content information of each food item, released by the US Department of Agriculture, was used to estimate intake of micro- and macronutrients. The healthy diet score was calculated as the sum of the scores for each of 5 individual components: fruits and vegetables (≥4.5 cups per day), fish (≥2 3.5-oz servings per week), fiber-rich whole grains (≥3 1-oz-equivalent serving per day), sodium (<1500 mg/day) and sugar-sweetened beverages (≤450 kcal/week). The diet score ranged from 0 to 5, with a lower score being unhealthy.
Each individual LS7 component was categorized as poor, intermediate, or ideal according to the American Heart Association’s LS7 criteria.9 Ideal health status was ≥150 min/wk of moderate activity or ≥75 min/wk of vigorous activity or ≥150 min/wk of moderate and vigorous activity, total cholesterol <200 mg/dL, blood pressure <120/80 mmHg, BMI <25 kg/m2, fasting blood glucose <100 mg/dL, nonsmokers or quit >12 months ago, and a healthy diet score (>4 components). Intermediate levels were: 1–149 min/wk of moderate activity or 1–74 min/wk of vigorous activity or 1–149 min/wk of moderate and vigorous physical activity, total cholesterol 200–239 mg/dL, systolic blood pressure 120–139 and diastolic blood pressure of 80–89 mmHg, BMI 25-<30kg/m2, fasting blood glucose 100–125 mg/dl or treated to goal, quit smoking <12 months ago, and a diet score of 2–3. Poor health status included no physical activity, total cholesterol ≥240 mg/dL, blood pressure ≥140/90 mmHg, BMI ≥30kg/m2, fasting blood glucose ≥126 mg/dL, current smokers, and a diet score of 0–1. Participants who were taking medications to achieve target levels for blood cholesterol, blood pressure, or blood glucose were classified as intermediate for each respective health factor. An overall LS7 composite score ranging from 0 to 14 was calculated, in which each risk factor was given points of 0, 1, or 2 for poor, intermediate, or ideal, respectively. A higher score indicates better health. The composite score was categorized as inadequate (0–4), average (5–9), or optimum (10–14) cardiovascular health.
Premature atrial contractions
PACs count was calculated based on the number of isolated, couplet, and triplet PACs per day [PACs count per day = # isolated PACs + 2 * (# couplet PACs) + 3 * (# triplet PACs)]. PACs frequency was calculated as the proportion of PACs over total number of heart beats per day averaged over the 14 day Zio® XT Patch monitor recording. No participant in this study had 0 PACs/day averaged over the 14 days of monitoring. We classified PACs frequency into three categories: minimal (<0.1%), occasional (≥0.1–5%) and frequent (>5%). We evaluated PAC burden as 3 categories for 2 reasons. First, the Zio XT Patch is a patch monitor that is widely used in clinical practice and its standard report uses these 3 categories and cutoffs (minimal *0.1%+, occasional [≥0.1–5%], and frequent [>5%]). Thus, by evaluating the established reporting categories of a widely used patch monitor, our findings are directly relevant to clinical practice. Second, in our analysis, we did not find evidence of a linear association between LS7 and PACs frequency, and LS7 was associated with increased odds of frequent PACs, but not occasional PACs. Hence, exploring PACs frequency as a continuous variable would be misleading given this lack of linear association. Also, the limited number of study participants in the frequent PACs category (<100) makes a dose-response analysis in that group unfeasible.
Covariates
Covariates included age (years), sex, ARIC field center, CHD, heart failure (HF), stroke and medication use [betablockers (yes/no), calcium channel blockers (yes/no), digoxin (yes/no), anti-arrhythmic drugs (yes/no)] which were obtained through Visit 6 while race and education level (less than high school education, high school graduate or equivalent, college or above) were self-reported by participants and obtained at Visit 1. Alcohol consumption status (never, former and current drinkers) was self-reported by participants and obtained at Visit 3. CHD was defined as a self-reported history of physician-diagnosed myocardial infarction or prior coronary revascularization at Visit 1 and adjudicated events thereafter. The Gothenburg criteria were used to define prevalent HF at Visit 1 and subsequent events were identified based on ICD codes from hospitalizations.11 Stroke was defined as a self-reported history of physician-diagnosed stroke at Visit 1 and adjudicated events thereafter. In addition, left ventricular ejection fraction (LVEF) and left atrial volume index (LAVI) were measured from 2D-echocardiograms performed during Visit 5.
Statistical Analysis
Baseline characteristics are reported as frequency and percentage for categorical variables, while continuous variables are reported as mean ± standard deviation or median (interquartile range).
Multinominal logistic regression was used to compute the odds ratios (OR) and 95% confidence intervals (CI) for the association between LS7 composite score at Visit 3 and PACs frequency (minimal, occasional and frequent) at Visit 6. We also computed ORs for the association between PACs frequency and each LS7 component factor individually, as well as per 1-unit increment in overall LS7 score.
Multivariable models were adjusted as follows: Model 1 was adjusted for age, sex, race/ARIC field center, education and drinking status. Model 2 was further adjusted for CHD, HF, stroke, and medication use. Finally, to determine whether cardiac structural and functional remodeling, as detected by echocardiography, might explain the association between LS7 and PACs frequency, Model 3 adjusted for Model 2 co-variates plus LVEF and LAVI. In all models, inverse probability weighting was used to account for attrition due to death, visit 6 non-attendance, or not wearing the Zio® XT Patch. Logistic models were used to estimate the probabilities of being alive at visit 6, attending visit 6, and being eligible and agreeing to wear the Zio® XT Patch. Weights for each individual were the inverse of the product of the estimated probabilities. For all analyses, the ideal health category or optimum risk factor level was the reference group. All analyses were conducted using SAS software (version 9.4; SAS Institute Inc., Cary, NC). A two-tailed p-value <0.05 indicated statistical significance.
Results
A total of 1924 individuals, 59.2% females and 23.1% African-Americans, were included in the study (Table 1). The participants were aged 56.6 ± 4.4 years at Visit 3 and 78.8 ± 4.7 years at Visit 6. The mean LS7 score at Visit 3 was 8.7±2.1; 52 (2.7%) participants had an inadequate score, 1141 (59.3%) participants had an average score, and 731 (38%) participants had an optimum score (Table 2).
Table 1.
Baseline Characteristics of Study Participants by Premature Atrial Contractions Frequency Categories: The Atherosclerosis Risk in Communities Study.
| Total Sample (n=1924) | Minimal PAC, <0.1% (n=407) | Occasional PAC, ≥0.1% – 5% (n=1419) | Frequent PAC, >5% (n=98) | p-values | |
|---|---|---|---|---|---|
| Demographics | |||||
| Age, years | 56.4 ± 4.4 | 55.2 ± 3.9 | 56.6 ± 4.5 | 57.0 ± 4.2 | <0.001* |
| Male sex | 785 (40.8) | 140 (34.4) | 591 (41.7) | 54 (55.1) | <0.001* |
| Black race | 444 (23.1) | 120 (29.5) | 301 (21.2) | 23 (23.5) | 0.5 |
| < High school education, | 211 (10.9) | 36 (8.9) | 161 (11.4) | 14 (14.3) | 0.73 |
| Current alcohol consumption, | 1115 (58.0) | 236 (58.0) | 815 (57.4) | 64 (65.3) | 0.14 |
| Clinical Characteristics | |||||
| Body mass index (kg/m2) | 27.7 ± 4.8 | 27.4 ± 5.0 | 27.7 ± 4.7 | 28.2 ± 5.0 | 0.26 |
| Coronary heart disease | 102 (5.3) | 24 (5.9) | 75 (5.3) | 3 (3.1) | 0.59 |
| Heart failure | 64 (3.3) | 10 (2.5) | 50 (3.5) | 4 (4.1) | 0.52 |
| Stroke | 258 (13.4) | 46 (11.3) | 194 (13.7) | 18 (18.4) | 0.31 |
| LV ejection fraction (%) | 65.8 ± 5.5 | 65.5 ± 5.8 | 65.9 ± 5.3 | 65.3 ± 7.2 | 0.49 |
| LA volume index, ml/m2 | 24.4 ± 6.6 | 22.6 ± 5.8 | 24.7 ± 6.6 | 27.2 ± 7.5 | <0.001* |
| Beta-blockers | 563 (29.3) | 124 (30.5) | 409 (28.8) | 30 (30.6) | 0.46 |
| Calcium channel blockers | 476 (24.7) | 100 (24.6) | 350 (24.7) | 26 (26.5) | 0.54 |
| Anti-arrhythmic drugs | 8 (0.4) | 2 (0.5) | 6 (0.4) | 0 (0) | 0.35 |
All categorical variables expressed as n (%)
All continuous variables expressed as mean ± standard deviation.
Baseline at ARIC Visit 3 (1993–95) and PAC burden measured at Visit 6 (2016–17)
Echocardiographic parameters of LA volume index and LV ejection fraction were measured at Visit 5.
Table 2.
Life’s Simple 7 Characteristics of Study Participants at Visit 3 by Premature Atrial Contractions Frequency Categories: The Atherosclerosis Risk in Communities Study.
| Total Sample (n=1924) | Minimal PAC, <0.1% (n=407) | Occasional PAC, ≥0.1% - 5% (n=1419) | Frequent PAC, >5% (n=98) | p-value | |
|---|---|---|---|---|---|
| LS7 composite score | |||||
| Inadequate (0–4), N (%) | 52 (2.7) | 15 (3.7) | 35 (2.5) | 2 (2) | 0.07 |
| Average (5–9), N (%) | 1141 (59.3) | 230 (56.5) | 841 (59.3) | 70 (71.4) | |
| Optimum (10–14), N (%) | 731 (38.0) | 162 (39.8) | 543 (38.3) | 26 (26.5) | |
| Individual LS7 risk factors | |||||
| Physical activity | |||||
| Poor, N (%) | 626 (32.6) | 132 (32.4) | 455 (32.1) | 39 (39.8) | 0.53 |
| Intermediate, N (%) | 468 (24.3) | 94 (23.1) | 351 (24.7) | 23 (23.5) | |
| Ideal, N (%) | 830 (43.1) | 181 (44.5) | 613 (43.2) | 36 (36.7) | |
| Total cholesterol | |||||
| Poor, N (%) | 270 (14.0) | 62 (15.2) | 193 (13.6) | 15 (15.3) | 0.48 |
| Intermediate, N (%) | 760 (39.5) | 165 (40.5) | 551 (38.8) | 44 (44.9) | |
| Ideal, N (%) | 894 (46.5) | 180 (44.2) | 675 (47.6) | 39 (39.8) | |
| Blood pressure | |||||
| Poor, N (%) | 203 (10.6) | 39 (9.6) | 149 (10.5) | 15 (15.3) | 0.37 |
| Intermediate, N (%) | 755 (39.2) | 161 (39.6) | 552 (38.9) | 42 (42.9) | |
| Ideal, N (%) | 966 (50.2) | 207 (50.9) | 718 (50.6) | 41 (41.8) | |
| Body mass index | |||||
| Poor, N (%) | 507 (26.3) | 95 (23.3) | 387 (27.3) | 25 (25.5) | 0.14 |
| Intermediate, N (%) | 829 (43.1) | 173 (42.5) | 606 (42.7) | 50 (51.0) | |
| Ideal, N (%) | 588 (30.6) | 139 (34.2) | 426 (30.0) | 23 (23.5) | |
| Fasting blood glucose | |||||
| Poor, N (%) | 122 (6.3) | 29 (7.1) | 81 (5.7) | 12 (12.2) | 0.10 |
| Intermediate, N (%) | 659 (34.3) | 132 (32.4) | 496 (35.0) | 31 (31.6) | |
| Ideal, N (%) | 1143 (59.4) | 246 (60.4) | 842 (59.3) | 55 (56.1) | |
| Smoking status | |||||
| Poor, N (%) | 234 (12.2) | 64 (15.7) | 156 (11.0) | 14 (14.3) | 0.06 |
| Intermediate, N (%) | 23 (1.2) | 2 (0.5) | 20 (1.4) | 1 (1.0) | |
| Ideal, N (%) | 1667 (86.6) | 341 (83.8) | 1243 (87.6) | 83 (84.7) | |
| Diet | |||||
| Poor, N (%) | 803 (41.7) | 157 (38.6) | 598 (42.1) | 48 (49.0) | 0.14 |
| Intermediate, N (%) | 1081 (56.2) | 245 (60.2) | 787 (55.5) | 49 (50.0) | |
| Ideal, N (%) | 40 (2.1) | 5 (1.2) | 34 (2.4) | 1 (1.0) |
PAC – Premature atrial contractions, LS7 – Life’s Simple 7.
Percentages are column percentages.
Baseline at ARIC Visit 3 (1993–95) and PAC burden measured at Visit 6 (2016–17)
The mean wear time of the Zio® XT Patch at Visit 6 was 12.6 ± 2.6 days. The median (IQR) PACs frequency was 0.2% (0.7%); 407 (21.2%) individuals had minimal PACs, 1419 (73.8%) participants had occasional PACs and 98 (5.1%) participants had frequent PACs. Compared to participants with minimal or occasional PACs, participants with frequent PACs were older, had a greater proportion of males and had a higher LAVI. There was a higher proportion of blacks in the minimal PACs category than the occasional and frequent PACs categories. The PACs frequency distribution for each LS7 score is shown in Figure 2.
Figure 2.
Distribution of Premature Atrial Contraction Frequency per Life’s Simple 7 (LS7) Category
The boundaries of the box represent quartiles 1 and 3 (i.e. interquartile range [IQR]), with the line representing the median value. The whiskers are calculated as ±1.5×IQR. The diamond represents the mean value.
LS7 composite score and PACs frequency
Table 3 shows the weighted ORs (95% CI) of association of LS7 score and categories with PAC frequency. Compared to participants in the optimal LS7 category, those in the average LS7 category had higher odds of frequent PACs than minimal PACs, (Model 2 ORs [95% CIs]: 1.71 [1.01, 2.89]). This relationship persisted even after adjusting LAVI and LVEF (Model 3 ORs [95% CIs]: 1.74 [1.03, 2.94]). Each 1-point increase in LS7 score was associated with a 12% decrease in odds of frequent PACs (Model 2 OR [95% CI]: 0.88 [0.78, 0.98)]) as compared with minimal PACs, which persisted even after adjusting for LVEF and LAVI (Model 3 OR [95% CI]: 0.87 [0.78, 0.98]). While no association was seen between LS7 categorization and frequent PACs as compared to occasional PACs, each 1-point increase in LS7 score was associated with a lower odds of frequent PACs as compared to occasional PACs (Model 2 OR [95% CI]: 0.88 (95% [0.79,0.97]). These results remained unchanged after further adjustment for LVEF and LAVI (Model 2 OR [95% CI]: 0.88 (95% [0.79,0.98]) . We did not observe any significant association between LS7 categories and occasional PACs vs minimal PACs.
Table 3.
Weighted* Odds Ratios (95% Confidence Intervals) of Premature Atrial Contractions Frequency with Overall Life’s Simple 7 (LS7) Score: The Atherosclerosis Risk in Communities Study, 1993–2017
| Premature atrial contractions frequency | |||||||||
|---|---|---|---|---|---|---|---|---|---|
| Occasional PAC vs Minimal PAC | Frequent PAC vs. Minimal PAC | Frequent PAC vs. Occasional PAC | |||||||
| Model 1 | Model 2 | Model 3 | Model 1 | Model 2 | Model 3 | Model 1 | Model 2 | Model 3 | |
| LS 7 health categories | |||||||||
| Inadequate | 0.81 (0.43, 1.53) | 0.79 (0.42, 1.49) | 0.78 (0.41, 1.49) | 0.83, (0.18, 3.83) | 0.78 (0.17, 3.67) | 0.74 (0.16, 3.51) | 1.03 (0.24, 4.45) | 0.99 (0.23, 4.38) | 0.95 (0.21, 4.19) |
| Average | 1.08 (0.85, 1.38) | 1.08 (0.84, 1.38) | 1.10 (0.85, 1.41) | 1.73 (1.03, 2.91) | 1.71 (1.01, 2.89) | 1.74 (1.03, 2.94) | 1.60 (0.99, 2.59) | 1.58 (0.97, 2.58) | 1.58 (0.97, 2.59) |
| Optimum | 1 (Ref) | 1 (Ref) | 1 (Ref) | 1 (Ref) | 1 (Ref) | 1 (Ref) | 1 (Ref) | 1 (Ref) | 1 (Ref) |
| Per 1-point increase | 0.99 (0.94, 1.05) | 0.99 (0.94,1.06) | 0.99 (0.94, 1.05) | 0.88 (0.79, 0.98) | 0.88 (0.78, 0.98) | 0.87 (0.78, 0.98) | 0.88 (0.80, 0.97) | 0.88 (0.79, 0.97) | 0.88 (0.79, 0.98) |
Individual LS7 components and PACs frequency
Amongst the various LS7 components, the majority of the study population had ideal scores for smoking status (86.6%), fasting blood glucose (59.4%) and blood pressure (50.2%) while only 2.1% of participants had an ideal score for diet (Table 2). Supplementary Table 1 shows the ORs (95% CI) of association of individual LS7 risk factors with PACs frequency.
Occasional PACs vs minimal PACs
Compared to those with ideal BMI, participants with poor BMI had higher odds of occasional PACs compared to no PACs (Model 2 ORs [95% CIs]: 1.48 [1.08, 2.04]). This association remained unchanged after adjusting for LAVI and LVEF (Model 3 ORs [95% CIs]: 1.41 [1.02, 1.94]). Those with intermediate BMI had nominally higher odds of occasional PACs as compared to minimal PACs (Model 2 ORs [95% CIs]: 1.14 [0.87, 1.49]); however, this association was not statistically significant. No association was seen between the other LS7 components and occasional PACs as compared to minimal PACs.
Frequent PACs vs Minimal PACs
Poor physical activity was associated with higher odds of frequent PACs than minimal PACs (Model 2 ORs [95% CIs]: 1.70 [0.99, 2.92]). This association was more robust after adjusting for LAVI and LVEF (Model 3 ORs [95% CIs]: 1.81 [1.05, 3.12]). Although participants with intermediate physical activity score also had higher odds of frequent PACs than minimal PACs, the association was not statistically significant. While poor scores for BMI, blood pressure, and blood cholesterol were associated with higher odds of frequent PACs than minimal PACs, these associations were not statistically significant likely owing to the small number of participants in these subgroups. No association was seen between the other LS7 components and frequent PACs vs minimal PACs.
Frequent PACs vs Occasional PACs
We did not find any association between the various LS7 components and frequent PACs as compared to occasional PACs.
Discussion
In this community-based cohort study involving two weeks of continuous rhythm monitoring, we observed that each 1-point increment in LS7 score (more favorable cardiovascular health) in midlife was associated with 12% lower odds of frequent PACs vs. minimal and occasional PACs in late life. This association was independent of intercurrent cardiovascular events and not explained by abnormalities in LA size and LV function. Of the individual LS7 risk factors, a poor score for physical activity was associated with higher odds of frequent PACs vs. minimal PACs, while a poor score for BMI was associated with higher odds of occasional PACs vs. minimal PACs. Collectively, our findings underscore the potential importance of lifestyle and risk factor modification in preventing frequent PACs, which is an established risk factor for AF and ischemic stroke.1–4,5
Physical activity has a J-shaped relationship with AF: higher risk of AF at extreme levels of physical activity while regular guideline-directed levels of exercise have consistently been shown to lower the risk of AF and AF burden.14,15 Additionally, poor levels of physical activity have also been found to be associated with a greater AF burden.18 Consistent with current literature, in this study, participants from a general population with lower levels of physical activity were seen to have a higher frequency of PACs independent of left atrial size. A similar association of higher frequency of PACs in individuals with lower levels of physical activity was described previously by Conen et al. and Kerola et al. Regular exercise positively modulates other cardiovascular risk factors such as obesity, diabetes mellitus and hypertension and also improves endothelial function, inflammation, and autonomic function thereby attenuating their deleterious effects on atrial structural and electrical remodelling and potentially lowering the risk for PACs.5,6,15–17
Obesity is an important risk factor for AF and AF burden.18,19 However, prior studies assessing risk factors for PACs reported contrasting findings: Conen and colleagues did not find any association between BMI and PACs while Kerola and colleagues reported a higher frequency of PACs in individuals with a lower BMI, concluding that obesity increases the risk of AF by creating a substrate for AF without necessarily increasing triggers for AF. By contrast, in our study, we found an association between a poor score for BMI in midlife and higher odds of occasional PACs vs. minimal PACs and higher odds of frequent PACs vs. minimal PACs, although the latter association was not statistically significant. These findings suggest that obesity, particularly after a long duration of more than 20 years, may promote triggers for AF. Obesity has been shown to alter the structural and electrophysiological properties of the left atrium in addition to contributing to autonomic dysfunction.20 These changes likely contribute to the higher PACs frequency seen in patients with a poor score for BMI. The conflicting results of earlier studies may be due to the time lag between the onset of obesity and genesis of PACs and the cross-sectional nature of prior studies.
Interestingly, the association of both poor physical activity and obesity with higher odds of frequent PACs did not attenuate after adjusting for LAVI even though these modifiable risk factors are believed to predispose to PACs and AF by inducing structural remodelling of the atria. The initial stages of atrial remodelling consist of subcellular changes with alterations in expression of membrane ion channels, gap junction proteins and other intracellular proteins followed by extracellular deposition of fibrous tissue with alterations in LA size and LA contractile function being seen only in the later stages of remodelling.20 Since current standards for noninvasive detection of LA structural remodelling are based on identification of alterations in LA size, either by echocardiography or by magnetic resonance imaging, the sensitivity of these techniques for detection of early stages of LA remodelling are poor. This low sensitivity of these imaging techniques likely explains the lack of attenuation of the observed associations after adjusting for LAVI.
The two studies that have sought to define risk factors for frequent PACs suffer from certain design limitations. For example, Conen et al. found increasing age, increasing height, cardiovascular disease, lower levels of physical activity, and lower HDL-cholesterol levels to be associated with higher PACs frequency.5 Similarly, Kerola and colleagues identified increasing age, increasing height, lower levels of physical activity, and a history of myocardial infarction to be associated with higher frequency of PACs.6 Both studies, however, were cross-sectional; hence, temporality of exposure vs. outcome cannot be firmly established. Furthermore, both studies employed a single 24-hour Holter to quantify PACs, which is imprecise given the day-to-day variability of PACs frequency. Moreover, while echocardiographic evaluation was not used in the study by Conen et al., Kerola et al. included echocardiographic information on left atrial size in only around 65% of patients. Additionally, left atrial size was assessed using left atrial diameter, which is a less optimal measure than left atrial volume index.
Our research addresses the limitations of prior studies and advances the field on several fronts. First, we focused our investigation on modifiable risk factors to maximize public health impact. Second, we assessed the longitudinal association of modifiable risk factors in midlife with PACs frequency in older age, a follow-up of more than 20 years. Third, we measured PACs frequency using 2-week continuous heart rhythm monitoring, which better accounted for the day-to-day variability in PACs frequency, thus providing a more precise measurement of our outcome. Finally, left atrial volume was measured echocardiographically and indexed to body surface area as the LAVI which is a more optimal measure of LA size and remodelling.
Several limitations of our study need to be considered. First, this study evaluates the presence of risk factors at only one time point in midlife. Since information on all LS7 risk factors were not collected at all visits, we were unable to evaluate the association between a change in LS7 score over time and frequency of PACs in late life. Second, selection bias may have occurred due to death, visit non-attendance, or not wearing the Zio XT Patch. However, we accounted for this by using inverse probability weighting in the models. Second, heart rhythm data from continuous heart rhythm monitors were not available at baseline. Although PACs can be ascertained from 10-second 12-lead ECGs at baseline, the latter has very low sensitivity in detecting PACs. Thus, relying on 10-second 12-lead ECGs to determine presence of PACs at baseline would introduce substantial misclassification bias. Finally, the distribution of participants with different scores in each LS7 component was non-uniform with small sample sizes in some of the components and a small number of participants with an inadequate LS7 score (52 of 1,924 [2.7%]) (Table 2), which might have limited statistical power for the analyses of some of the individual LS7 factors.
Conclusion
In this longitudinal community-based study, less optimal lifestyle and modifiable risk factors in midlife, particularly poor physical activity and obesity, are associated with greater PACs frequency. These associations are not mediated by alterations in left atrial size or left ventricular systolic function. More research is needed to determine whether modifying lifestyle risk factors in midlife would prevent frequent PACs, and thereby prevent AF and stroke in older age.
Supplementary Material
Highlights.
Lifestyle risk factors play an important role in the pathogenesis of premature atrial contractions (PACs)
Obesity and poor physical activity in midlife were associated with an increased frequency of PACs in late life
A lower Life’s Simple 7 score was associated with a higher odds of frequent PACs as compared to occasional and minimal PACs
Acknowledgements
The authors thank the staff and participants of the ARIC study for their important contributions.
Sources of Funding
The Atherosclerosis Risk in Communities study has been funded in whole or in part with Federal funds from the National Heart, Lung, and Blood Institute, National Institute of Health, Department of Health and Human Services, under contract numbers (HHSN268201700001I, HHSN268201700002I, HHSN268201700003I, HHSN268201700004I, and HHSN268201700005I). In addition, this work was funded by R01 HL126637. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Heart, Lung, and Blood Institute or the National Institutes of Health. L.Y.C. is supported by grants R01 HL126637 and R01 HL141288 from the National Heart, Lung, and Blood Institute of the National Institutes of Health. M.R.R. was supported by grant T32HL007024 from the National Heart, Lung, and Blood Institute of the National Institutes of Health.
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Conflict of interest: None
Disclosures There are no conflicts of interest.
Data Availability Data available on request.
References
- 1.Haïssaguerre M, Jaïs P, Shah DC, et al. Spontaneous Initiation of Atrial Fibrillation by Ectopic Beats Originating in the Pulmonary Veins. N Engl J Med 1998; 339: 659–666. [DOI] [PubMed] [Google Scholar]
- 2.Binici Z, Intzilakis T, Nielsen OW, et al. Excessive supraventricular ectopic activity and increased risk of atrial fibrillation and stroke. Circulation 2010; 121:1904–1911. [DOI] [PubMed] [Google Scholar]
- 3.Engstrom G, Hedblad B, Juul-Moller S, et al. Cardiac arrhythmias and stroke: increased risk in men with high frequency of atrial ectopic beats. Stroke 2000; 31: 2925–2929. [DOI] [PubMed] [Google Scholar]
- 4.Larsen BS, Kumarathurai P, Falkenberg J, et al. Excessive Atrial Ectopy and Short Atrial Runs Increase the Risk of Stroke Beyond Incident Atrial Fibrillation. J Am Coll Cardiol 2015; 66: 232–241. [DOI] [PubMed] [Google Scholar]
- 5.Conen D, Adam M, Roche F, et al. Premature Atrial Contractions in the General Population. Circulation 2012; 126: 2302–2308. [DOI] [PubMed] [Google Scholar]
- 6.Kerola T, Dewland TA, Vittinghoff E, et al. Predictors of atrial ectopy and their relationship to atrial fibrillation risk. EP Eur 2019; 21: 864–870. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Berenson GS, Srinivasan SR, Hunter SM, et al. Risk factors in early life as predictors of adult heart disease: the Bogalusa Heart Study. Am J Med Sci 1989; 298:141–151. [DOI] [PubMed] [Google Scholar]
- 8.Garg PK, O’Neal WT, Chen LY, et al. American Heart Association’s Life Simple 7 and Risk of Atrial Fibrillation in a Population Without Known Cardiovascular Disease: The ARIC (Atherosclerosis Risk in Communities) Study. J Am Heart Assoc. 2018. April 17; 7(8): e008424. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Lloyd-Jones DM, Hong Y, Labarthe D, et al. Defining and setting national goals for cardiovascular health promotion and disease reduction: the American Heart Association’s strategic Impact Goal through 2020 and beyond. Circulation 2010; 121: 586–613. [DOI] [PubMed] [Google Scholar]
- 10.The Atherosclerosis Risk in Communities (ARIC) Study: design and objectives. The ARIC investigators. Am J Epidemiol 1989; 129: 687–702. [PubMed] [Google Scholar]
- 11.Siedel J, Hagele EO, Ziegenhorn J, et al. Reagent for the enzymatic determination of serum total cholesterol with improved lipolytic efficiency. Clin Chem 1983; 29:1075–1080. [PubMed] [Google Scholar]
- 12.Baecke JA, Burema J, Frijters JE. A short questionnaire for the measurement of habitual physical activity in epidemiological studies. Am J Clin Nutr 1982; 36: 936–942. [DOI] [PubMed] [Google Scholar]
- 13.Willett WC, Sampson L, Stampfer MJ, et al. Reproducibility and validity of a semiquantitative food frequency questionnaire. Am J Epidemiol 1985; 122: 51–65. [DOI] [PubMed] [Google Scholar]
- 14.Elliott AD., Mahajan R, Pathak RK, et al. Exercise Training and Atrial Fibrillation. Circulation 2016; 133: 457–459. [DOI] [PubMed] [Google Scholar]
- 15.Andersen K, Farahmand B, Ahlbom A, et al. Risk of arrhythmias in 52 755 long-distance cross-country skiers: a cohort study. Eur Heart J 2013; 34: 3624–3631. [DOI] [PubMed] [Google Scholar]
- 16.Snowling NJ, Hopkins WG. Effects of different modes of exercise training on glucose control and risk factors for complications in type 2 diabetic patients: a meta-analysis. Diabetes Care 2006; 29: 2518–2527. [DOI] [PubMed] [Google Scholar]
- 17.Soares-Miranda L, Sattelmair J, Chaves P, et al. Physical Activity and Heart Rate Variability in Older Adults: The Cardiovascular Health Study. Circulation 2014; 129: 2100–2110. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Wane Wendy, Norby Faye L, Rooney Mary R, et al. Association of Life’s Simple 7 with Atrial Fibrillation Burden (From the Atherosclerosis Risk in Communities Study). Am J Cardiol 2020. December 15;137:31–38. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Guglin M, Maradia K, Chen R, et al. Relation of Obesity to Recurrence Rate and Burden of Atrial Fibrillation. Am J Cardiol 2011; 107: 579–582. [DOI] [PubMed] [Google Scholar]
- 20.Mahajan R, Lau DH, Brooks AG, et al. Electrophysiological, Electroanatomical, and Structural Remodeling of the Atria as Consequences of Sustained Obesity. J Am Coll Cardiol 2015; 66: 1–11. [DOI] [PubMed] [Google Scholar]
- 21.Thijssen Victor L. J. L., Ausma Jannie. Borgers Marcel. Structural remodelling during chronic atrial fibrillation: act of programmed cell survival. Cardiovascular Research, October 2001; 52 (1): 14–24. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.