Clinical Epidemiology

Two-Week Burden of Arrhythmias across CKD Severity in a Large Community-Based Cohort: The ARIC Study

Esther D. Kim, Elsayed Z. Soliman, Josef Coresh, Kunihiro Matsushita and Lin Yee Chen
JASN March 2021, 32 (3) 629-638; DOI: https://doi.org/10.1681/ASN.2020030301

Significance Statement

CKD is associated with sudden cardiac death and atrial fibrillation (AF). However, other arrhythmia types and different arrhythmic burden measures, such as presence and frequency of an arrhythmia, are not well characterized in CKD. To comprehensively quantify the burden of arrhythmias across CKD severity in 2257 community-dwelling older adults aged 71–94 years, the authors examined associations of major arrhythmias with CKD measures using 2-week noninvasive electrocardiogram monitoring. After adjusting for confounders, CKD—especially as measured by albuminuria—was associated with the presence of AF and with nonsustained ventricular tachycardia. Additionally, eGFR was associated with less frequent atrioventricular block, whereas albuminuria was associated with more frequent ventricular ectopy. Use of a novel 2-week monitoring approach revealed a broader range of arrhythmias associated with CKD than previously reported.

Abstract

Background CKD is associated with sudden cardiac death and atrial fibrillation (AF). However, other types of arrhythmia and different measures of the burden of arrhythmias, such as presence and frequency, have not been well characterized in CKD.

Methods To quantify the burden of arrhythmias across CKD severity in 2257 community-dwelling adults aged 71–94 years, we examined associations of major arrhythmias with CKD measures (eGFR and albuminuria) among individuals in the Atherosclerosis Risk in Communities study. Participants underwent 2 weeks of noninvasive, single-lead electrocardiogram monitoring. We examined types of arrhythmia burden: presence and frequency of arrhythmias and percent time in arrhythmias.

Results Of major arrhythmias, there was a higher prevalence of AF and nonsustained ventricular tachycardia among those with more severe CKD, followed by long pause (>30 seconds) and atrioventricular block. Nonsustained ventricular tachycardia was the most frequent major arrhythmia (with 4.2 episodes per person-month). Most participants had ventricular ectopy, supraventricular tachycardia, and supraventricular ectopy. Albuminuria consistently associated with higher AF prevalence and percent time in AF, and higher prevalence of nonsustained ventricular tachycardia. When other types of arrhythmic burden were examined, lower eGFR was associated with a lower frequency of atrioventricular block. Although CKD measures were not strongly associated with minor arrhythmias, higher albuminuria was associated with a higher frequency of ventricular ectopy.

Conclusions CKD, especially as measured by albuminuria, is associated with a higher burden of AF and nonsustained ventricular tachycardia. Additionally, eGFR is associated with less frequent atrioventricular block, whereas albuminuria is associated with more frequent ventricular ectopy. Use of a novel, 2-week monitoring approach demonstrated a broader range of arrhythmias associated with CKD than previously reported.

CKD is associated with an increased risk of sudden cardiac death (SCD),1 implying a high burden of cardiac arrhythmias in CKD. Previous studies, however, have focused mainly on atrial fibrillation (AF) or studied patients on dialysis,2⇓⇓⇓⇓⇓⇓–9 which mainly reflects the difficulty in detecting arrhythmias that can often be asymptomatic and transient. Recent findings demonstrating that bradyarrhythmia can lead to SCD,10,11 and that the vast majority of persons with CKD are not on dialysis,12 highlight our incomplete understanding of the range of tachy- and bradyarrhythmias that are associated with CKD.

Recent advancements in wearable continuous electrocardiogram (ECG) devices, which can monitor for a longer time, allow us to investigate other arrhythmias that are asymptomatic and transient in populations other than patients on dialysis. In addition, they can provide information on different measures of arrhythmic burden, such as the frequency of episodes and percent time in arrhythmias. We therefore used data from a 2-week continuous ECG monitoring device in the Atherosclerosis Risk in Communities (ARIC) study, to comprehensively examine various arrhythmias (e.g., AF, ventricular tachycardia, long pause, atrioventricular block) and different burden measures (i.e., prevalence, frequency, and percent time) across CKD severity. We hypothesized that more severe CKD is associated with a broader range of arrhythmias, especially when burden measures beyond mere presence or absence of arrhythmias are investigated.

Methods

Study Design

This study is a cross-sectional analysis, using data from visit 6 (2016–2017) of the ARIC study, which was originally designed to investigate the natural history of atherosclerotic disease from mid- to late-life.13 The ARIC study design has been published previously.13 Briefly, 15,792 participants were recruited during 1987–1989 from four communities in the United States (Forsyth County, NC; Jackson, MS; Minneapolis, MN; and Washington County, MD) and completed the first study visit (visit 1). The participants subsequently completed six study visits (visit 2 in 1990–1992, visit 3 in 1993–1995, visit 4 in 1996–1998, visit 5 in 2011–2013, visit 6 in 2016–2017, and visit 7 in 2018–19). Additionally, they were contacted annually (semiannually, beginning in 2012) to obtain updated information on medical history and lifestyle. In visit 6, 4003 participants attended the study examination, which included assessment for cardiovascular disease, cognitive function, and their risk factors; biospecimen collection; medication assessment; and 2-week continuous ECG monitoring. The ARIC study was approved by the institutional review board of each participating center, and written informed consent was obtained from participants at each study visit.

Study Population

Of 4003 visit-6 participants, those who did not have a pacemaker or an implantable cardioverter defibrillator were asked to undergo 2-week ECG monitoring, as detailed below. We excluded participants who did not undergo the continuous ECG monitoring (n=1387), were missing CKD measures (n=194), and were non-White and non-Black (n=6). We also excluded those who were missing any covariate of interest (n=147) (Supplemental Figure 1). The final analytical sample of this study was 2257.

Kidney Disease Measures

The primary exposures of interest were the two key measures of CKD: eGFR, which indicates kidney function; and albuminuria, which reflects kidney damage. eGFR was calculated on the basis of the CKD Epidemiology Collaboration equation, using cystatin C,14 which was measured in serum using Gentian Cystatin C reagent on the Roche Modular P Chemistry analyzer (Roche Diagnostics, Indianapolis, IN), with a laboratory interassay coefficient of variation of 6.1% and a mean concentration of 0.94 mg/L. We used cystatin C rather than serum creatinine because there is concern that age-related reduced muscle mass can overestimate eGFR in older individuals, when using a creatinine-based equation.15 Albuminuria was measured using albumin-creatinine ratio (ACR), which was calculated by dividing urine albumin by urine creatinine, as recommended by the Kidney Disease: Improving Global Outcomes clinical practice guideline for CKD.16 Urine albumin was measured in urine on the ProSpec nephelometric analyzer, using an immunoturbidometric method (Dade Behring GMBH, Marburg, Germany), with a laboratory interassay coefficient of variation of 4.0% and a mean concentration of 19.7 mg/L. Urine creatinine was measured on a Roche Modular P Chemistry Analyzer, using a creatinase enzymatic method (Roche Diagnostics, Indianapolis, IN) that yielded a coefficient of variation of 4.5% and a mean concentration of 17 mg/dl.

Arrhythmia Measurement

The Zio XT Patch (iRhythm Technologies Inc., San Francisco, CA) is a US Food and Drug Administration (FDA)-approved ECG skin patch that is s single-lead, noninvasive, lightweight, and water-resistant monitor that allows ECG recording for up to 14 days.17 Compared with the standard 24-hour Holter monitor, the Zio Patch is demonstrated to detect more arrhythmic events and shown to be more comfortable to wear.17 ARIC study participants were asked to wear the Zio Patch for up to 14 days, and mail the device back to the manufacturer, where the ECG data were analyzed using an FDA-approved proprietary algorithm and underwent technical review. The following arrhythmias were detected by the Zio Patch and were included in this study: major types, including AF, nonsustained ventricular tachycardia, long pause lasting >3 seconds, and Mobitz type 2 or third-degree atrioventricular block; and minor types, including ventricular ectopy, supraventricular tachycardia (not including AF), and supraventricular ectopy.

As a result of continuous monitoring, we were able to examine different measures of arrhythmic burden. We first examined the presence of arrhythmias, defined as the dichotomous information of the occurrence of any arrhythmic episode during the monitoring period, and we examined this for all types of arrhythmias. We also assessed the frequency of arrhythmia during the monitoring period, which was defined as the number of episodes per day, and this was measured for all types of arrhythmias except for AF. For AF, instead, we examined the percent time in arrhythmia because the frequency of episodes will not appropriately reflect the burden of longer-lasting AF (e.g., the frequency equals one in the case of continuous AF). More specific definitions are listed in Supplemental Table 1.

To avoid false-positive findings, in the ARIC study, E.Z.S. and his team of physician ECG readers at Wake Forest University downloaded the standard report from the iRhythm website on a daily basis, and verified the accuracy of reported major arrhythmias.

Covariate Measurements

All variables were assessed at ARIC study visit 6, except for education, which was measured at visit 1. Educational background was categorized as less than high school, graduated from high school or vocational school, or obtained college or graduate/professional degrees. Body mass index was calculated by dividing body mass (in kilograms) by height (in meters) squared. Cigarette smoking and alcohol drinking was self-reported and categorized as current or noncurrent. Diabetes was defined as fasting plasma glucose ≥126 mg/dl, nonfasting glucose ≥200 mg/dl, using medication for diabetes, or self-reported physician diagnosis of diabetes. Total cholesterol and HDL cholesterol were measured using an enzymatic method and the Olympus HDL-Cholesterol test, respectively.18 Prevalent cardiovascular disease included coronary heart disease, heart failure, and stroke, which included self-reported prevalence at visit 1 and relevant events between visit 1 and visit 6, adjudicated by physician reviewers on the basis of medical records from hospitalization data. Systolic and diastolic BP were measured in a sitting position after 5 minutes of rest, using a validated automatic sphygmomanometer (OMRON HEM-907 XL). The average of the last two of three measures was used. Use of antihypertensive medication in the past 4 weeks was assessed by self-report, and confirmed with drug containers, if available. Similarly, the use of antiarrhythmic medications was recorded by inspecting the participants’ drug containers for all classes of antiarrhythmic medications. For certain types of antiarrhythmic medications that may have overlapping indications (i.e., β-blockers may be used for some arrhythmias or hypertension), we used self-reported data to confirm if individuals have hypertension, and classified the medications as antihypertensive drugs accordingly. We also identified medications that are commonly reported to be QT-prolonging (Supplemental Table 2),19 as QT prolongation can affect the risk of arrhythmias.20

Statistical Analyses

We examined the distributions of all variables. and summarized them using mean and SD for normally distributed continuous variables, median and interquartile interval for non-normally distributed variables, and frequency and percentage for categorical variables. To summarize the rates of arrhythmias, we divided the total number of episodes observed during the monitoring period by the total person-month contributed. The 95% percentile confidence intervals (95% CIs) were estimated by bootstrapping the rate estimate using 1000 replicates.

For better interpretability, eGFR was examined per −15 ml/min per 1.73 m2, and ACR was log-transformed (using a log base of 4) to accommodate its non-normal distribution.

The associations between each CKD measure and the presence of each arrhythmia were estimated using Poisson regression, with robust variance to accommodate the high prevalence of some arrhythmias.21

The relationships between CKD measures and the frequency of arrhythmia episodes were estimated using a zero-inflated Poisson regression, with robust variance to account for an excess of zero counts, taking into consideration the total analyzable wear time of the device as the offset variable to explore frequency as the rate of arrhythmias across CKD status. The zero-inflation portion of the model included the same set of covariates as the Poisson model. The associations between CKD measures and percent time in AF were examined using linear regression.

For all associations, we adjusted for potential confounders in a series of models: model 1 was unadjusted; model 2 adjusted for age, sex, race, study center, and education; model 3 additionally included body mass index, current smoking, alcohol use, systolic BP, antihypertensive medication, diabetes, prevalent cardiovascular disease, total cholesterol, HDL cholesterol, and antiarrhythmic and QT-prolonging medications. Depending on the main exposure of interest (eGFR or ACR), we additionally adjusted for ACR or eGFR in model 3. When examining minor arrhythmias originating from the supraventricular region (i.e., supraventricular tachycardia and supraventricular ectopy), we excluded participants who had continuous AF because continuous AF would likely preclude participants from experiencing these supraventricular arrhythmias.

To test the robustness of our findings, we performed the following sensitivity analyses. We ascertained additional prevalent cases of the major types of arrhythmias (AF, nonsustained ventricular tachycardia, long pause, and atrioventricular block) using hospitalization records collected within 1 year of visit 6. We chose a time frame of 1 year to capture current cases and avoid additionally including older events that may have been treated and no longer relevant. Using the additional prevalent events of arrhythmias, we re-examined the associations between CKD measures and the presence of arrhythmias. Because antiarrhythmic medications can potentially affect the risk of various arrhythmias, we also performed the main set of analyses after excluding participants using antiarrhythmic medications.

All analyses were performed in R version 3.5.1 and Stata/SE version 15.1 (StataCorp, College Station, TX). A two-tailed P<0.05 was considered statistically significant.

Results

Baseline Characteristics

Among 2257 participants, the mean age was 79 years and 58% had CKD as defined by eGFR<60 ml/min per 1.73 m2 or ACR≥30 mg/g (Table 1). The majority were female (57%), White (76%), current drinkers (52%), and taking antihypertensive medication (76%). Many had diabetes (33%) and prevalent cardiovascular disease (24%). The majority of participants wore the Zio Patch for the full 2 weeks (median wear time of 13.7 days [interquartile interval, 12.7–13.9]).

View this table:
Table 1.

Baseline characteristics in 2257 study participants

Descriptive Statistics of Arrhythmias

Of the major arrhythmias, nonsustained ventricular tachycardia was the most prevalent (30.2%), followed by AF (7.4%), long pause (2.7%), and atrioventricular block (1.8%) (Table 2). The prevalence of minor arrhythmias were higher. Supraventricular ectopy was the most common (99.9%), followed by ventricular ectopy (98.8%) and supraventricular tachycardia (89.8%).

View this table:
Table 2.

Burden of major and minor arrhythmias

Prevalence Ratio for Major Arrhythmias across CKD Measures

In unadjusted models, eGFR demonstrated significant prevalence ratios (PRs) for AF (PR, 1.32; 95% CI, 1.18 to 1.47), nonsustained ventricular tachycardia (PR, 1.14; 95% CI, 1.08 to 1.20), and long pause (PR, 1.38; 95% CI 1.13 to 1.68) (model 1 in Figure 1, Supplemental Table 3). However, after adjusting for demographic and clinical factors and ACR (in model 3), these associations were no longer significant. On the other hand, the associations of ACR with AF and nonsustained ventricular tachycardia were robust across all models (PR, 1.27; 95% CI, 1.13 to 1.44 for AF and PR, 1.07; 95% CI, 1.01 to 1.13 for ventricular tachycardia in model 3). Its association with long pause was statistically significant only in model 1. Neither eGFR nor ACR was significantly associated with atrioventricular block, although both lower eGFR and higher ACR demonstrated inverse associations in models 3.

Figure 1.
Figure 1.

The associations of eGFR with AF, nonsustained ventricular tachycardia, and long pause are no longer significant after covariatte adjustment while the associations of ACR with AF and nonsustained ventricular tachycardia are significant in all models.

Percent Time in AF and Frequency of Other Major Arrhythmias

Both eGFR and ACR were generally associated with higher percent time in AF, but only the association involving ACR remained statistically significant in all models (Figure 2A, Supplemental Table 4). Lower eGFR and higher ACR demonstrated similar associations with a higher frequency of long pause episodes; however, these associations were no longer statistically significant after adjusting for other confounders in model 3 (Figure 2B, Supplemental Table 4). Neither eGFR nor ACR showed significant associations with frequency of ventricular tachycardia. As for the frequency of atrioventricular block, both lower eGFR and higher ACR demonstrated inverse relationships in model 1, but the relationship was robust only for eGFR across the three models.

Figure 2.
Figure 2.

eGFR and ACR were generally associated with higher percent ttime in AF but only the associations involving ACR remained statistically significant. Lower eGFR and higher ACR demonstrated similar associations with a higher frequency of long pause episodes.

Prevalence and Frequency of Minor Arrhythmias across CKD Measures

Lower eGFR and higher ACR were not associated with the prevalence of ventricular ectopy (Supplemental Figure 2, Supplemental Table 5), and only ACR was significantly associated with a higher frequency of ventricular ectopy in all models (Supplemental Figure 3, Supplemental Table 6). Neither eGFR nor ACR was significantly associated with the prevalence or frequency of supraventricular tachycardia. For supraventricular ectopy, eGFR and ACR similarly demonstrated significant associations with its prevalence in models 1 or 2, but not with its frequency.

Sensitivity Analysis

Using hospitalization records, we ascertained 26 additional prevalent cases of AF, one additional ventricular tachycardia, three additional long pause, and three additional atrioventricular block events. When we used the additional events to re-examine the relationship between CKD measures and the prevalence of the major arrhythmias, the findings remained similar: eGFR was associated with AF, nonsustained ventricular tachycardia, and long pause only in models 1 and 2, but not in model 3; whereas ACR was robustly and consistently associated with a higher prevalence of AF and nonsustained ventricular tachycardia (Supplemental Table 7). When we excluded participants taking antiarrhythmic medications, the results remained similar (Supplemental Table 8). Similarly, when we adjusted for the use of β-blockers and calcium-channel blockers, the main associations remained similar, and when we stratified the results by the use of β-blockers or calcium-channel blockers, the associations were also similar (Supplemental Tables 9 and 10).

Discussion

In this community-based study with excellent adherence to a 2-week continuous ECG monitor, we found a higher prevalence of clinically significant arrhythmias, such as AF and nonsustained ventricular tachycardia, among those with more severe CKD. When comparing eGFR and ACR, we found that ACR demonstrated stronger and more consistent associations, which indicates that albuminuria is more strongly related to arrhythmias. Of the major types of arrhythmias, AF demonstrated the most consistent association with CKD measures, regardless of the type of burden examined (i.e., the prevalence and percent time in arrhythmia). Although the presence of other major arrhythmias such as long pause and atrioventricular block were not associated with CKD measures, when we further examined other types of arrhythmic burden, we found that lower eGFR was associated with lower frequency of atrioventricular block. CKD measures were not strongly associated with minor arrhythmias; however, ACR was associated with a higher frequency of ventricular ectopy.

Despite several published studies suggesting that CKD is associated with a higher burden of AF,3,22⇓–24 there still remains an important knowledge gap on the prevalence and burden of AF across the severity of CKD.25 In past studies, the prevalence of AF ranged from <1% to >49% in CKD,3,22,24,26⇓⇓⇓–30 which is a fairly large range that can be attributed to different study populations (e.g., inpatients, population-based cohorts, patients receiving dialysis, etc.), different definitions of CKD (e.g., eGFR calculated using various formulae and filtration markers, proteinuria, dialysis status, etc.), and different methods used to record arrhythmias (e.g., Holter monitors, clinic or study visit ECGs, self-report, etc.). In our study of community-dwelling older adults using continuous ECG monitoring, we report a 2-week AF prevalence of around 7% (higher in more severe CKD) and a rate of 1.9 episodes per person-month, which is similar to the rate among those with moderate-to-severe CKD and type 2 diabetes who underwent a similar single-lead ECG monitoring for 11.3±3.9 days.30 The independent association between CKD and the prevalence of AF in our study was consistent with previous studies.3,22,23 Moreover, we uniquely observed that CKD severity was also strongly associated with higher percent time of AF, further supporting the close link between CKD and AF.

We also found that CKD, especially as measured by ACR, was associated with a higher prevalence of nonsustained ventricular tachycardia, which corroborates past studies reporting that individuals with CKD have a higher risk of incident life-threatening ventricular arrhythmias as captured by implantable cardioverter defibrillators.1,31 We also found that the frequency of ventricular ectopy (although not the binary presence of ventricular ectopy) was higher among those with higher albuminuria, which has not been widely reported in the past. Taken together, our findings suggest that CKD is associated with a higher risk of ventricular arrhythmias of varying severity (ectopic beats to nonsustained tachycardia), which can explain the high burden of SCD in CKD.1,32 Our finding that ventricular arrhythmias occur at a higher prevalence and frequency is, however, in contrast to the recent ECG monitoring study of moderate-to-severe patients with CKD and type 2 diabetes, which showed that ventricular arrhythmias occurred at the lowest rate compared with other types of cardiac arrhythmias.30 The differences found between the studies may be explained by multiple factors: we present a larger and older cohort of community-dwelling individuals (sample size of 2257 versus 38), who underwent a longer duration of continuous ECG monitoring, and represent a larger spectrum of CKD severity. These two studies also used different wearable ECG patches.

In addition to AF and ventricular arrhythmias, quantifying the burden of other arrhythmias is especially important because more recent studies have implicated the role of bradyarrhythmia in SCD among the CKD population.10,11 In this context, we observed a higher prevalence and frequency of long pause in CKD versus no CKD. Although these relationships were not statistically significant after adjusting for all potential confounders, the findings are in line with recent small studies (n<70) reporting bradycardia as an important terminal arrhythmia, leading to SCD in patients on dialysis.10,11 Although further research is warranted to better understand the relationship between long pause and SCD in CKD, there may be a need to pay attention to bradycardia in research and clinical practice and research in patients with CKD.

We unexpectedly observed that those with more severe CKD, as measured by lower eGFR, had a lower frequency of atrioventricular block. The pathophysiology underlying this association is unclear, and we are not aware of any previous reports showing this association. It is possible that less frequent episodes of heart block among those with more severe CKD might indicate longer durations of block; however, we cannot test this theory because the Zio Patch did not report durations of arteriovenous block. Thus, this observation needs to be confirmed in other studies, and if so, the clinical effect of arteriovenous block as detected by the Zio Patch should be also evaluated.

When comparing eGFR and ACR as measures of CKD, ACR was more strongly associated with the presence of major arrhythmias. Although only a few studies have examined both eGFR and albuminuria with prevalent AF,23,28,33 one study similarly demonstrated that urinary ACR was more strongly associated with the presence of AF compared with eGFR.23 Our study extended our knowledge to ventricular tachycardia and other arrhythmias. The precise mechanism underlying the associations of ACR with arrhythmias is unknown; however, ACR is also considered a marker of microvascular damage, a pathophysiological condition known to play an important role in cardiac dysfunction. Indeed, ACR has been shown to strongly predict incident heart failure.34 Thus, it is possible that ACR could broadly reflect subclinical cardiac abnormalities that increase vulnerability to major arrhythmias.

In light of our findings, our study overall demonstrates the importance of examining different types of arrhythmia burden, as this provided additional insights into the relationships between CKD and various arrhythmias that would have been missed by relying only on the presence versus absence of arrhythmias. Further, it highlights the value of continuous monitoring devices, and supports recent statements from the American Heart Association emphasizing the importance of moving beyond examining arrhythmias only as a binary entity.35 Although the prognostic value of our findings cannot be investigated at this moment, the ARIC study is an ongoing prospective study, and thus will be able to eventually address this important question.

There are several limitations of our study. Although it has better detection than 24-hour Holter monitoring,17 2-week monitoring with the Zio Patch may still have missed some prevalent cases. Indeed, additional yield of detecting AF by repeating two sequences of Zio Patch has been reported.36 Also, approximately 1000 participants did not agree to wear Zio Patch. Participants who underwent continuous monitoring were slightly younger, more likely to be White, and have less heart failure and better kidney function and albuminuria than those who did not; however, the two groups were very similar otherwise (Supplemental Table 11). The cross-sectional design of this study is certainly not a limitation for examining the 2-week prevalent burden of arrhythmias; nonetheless, because of the study design, we cannot establish any temporality in the associations between CKD and arrhythmias. We also could not examine the duration of episodes for most arrhythmias as we did not have access to this information; however, future studies consider examining this in addition to the frequency of episodes to better understand the relationships. Despite the limitations, the strength of our study is that we were able to address important gaps and limitations in the literature by examining various arrhythmias and the different types of arrhythmic burden in association with CKD severity using both eGFR and albuminuria.

In conclusion, the presence of major arrhythmias, specifically AF and nonsustained ventricular tachycardia, are higher among individuals with more severe CKD. Those with worse CKD also had a higher frequency of long pause and ventricular ectopy. Using a novel 2-week monitoring approach, our study found a broader range of arrhythmias associated with CKD than previously reported.

Disclosures

L.Y. Chen reports scientific advisor or membership as an Editorial Board member of Circulation: Arrhythmia and Electrophysiology. K. Matsushita reports receiving personal fees from Akebia, grants and personal fees from Fukuda Denshi, and grants and personal fees from Kyowa Kirin, outside the submitted work. All remaining authors have nothing to disclose.

Funding

L.Y. Chen reports receiving National Institutes of Health funding via grant R01HL126637. K. Matsushita reports receiving National Institutes of Health grants, during the conduct of the study. The ARIC study has been funded, in whole or in part, with Federal funds from the National Heart, Lung, and Blood Institute, National Institutes of Health, Department of Health and Human Services, under contracts HHSN268201700001I, HHSN268201700002I, HHSN268201700003I, HHSN268201700005I, and HHSN268201700004I.

Acknowledgments

The authors thank the staff and participants of the ARIC study for their important contributions.

Supplemental Material

This article contains the following supplemental material online at http://jasn.asnjournals.org/lookup/suppl/doi:10.1681/ASN.2020030301/-/DCSupplemental.

Supplemental Figure 1. Study flow diagram.

Supplemental Figure 2. Associations of eGFR and ACR with prevalence of minor arrhythmias.

Supplemental Figure 3. Associations of eGFR and ACR with frequency of minor arrhythmias.

Supplemental Table 1. Definitions of arrhythmia burden.

Supplemental Table 2. List of drugs commonly associated with QT-prolongation and torsades de pointes.

Supplemental Table 3. Associations of eGFR and ACR with prevalence of major arrhythmias.

Supplemental Table 4. Associations of eGFR and ACR with percent time in AF and frequency of nonsustained ventricular tachycardia, long pause, and atrioventricular block.

Supplemental Table 5. Associations of eGFR and ACR with prevalence of minor arrhythmias.

Supplemental Table 6. Associations of eGFR and ACR with frequency of minor arrhythmias.

Supplemental Table 7. Associations of eGFR and ACR with prevalence of major arrhythmias using additional arrhythmia events captured from hospitalization records.

Supplemental Table 8. Associations of eGFR and ACR with prevalence of major arrhythmias, after excluding participants taking antiarrhythmic medications.

Supplemental Table 9. Associations of eGFR and ACR with prevalence of major arrhythmias, after adjusting for β-blockers and calcium-channel blockers

Supplemental Table 10. Associations of eGFR and ACR with prevalence of major arrhythmias, stratified by β-blockers and calcium-channel blockers

Supplemental Table 11. Baseline characteristics comparing participants who did and did not undergo 2-week continuous ECG monitoring.

Footnotes

  • Published online ahead of print. Publication date available at www.jasn.org.

References

  1. ↵
    1. Suzuki T,
    2. Agarwal SK,
    3. Deo R,
    4. Sotoodehnia N,
    5. Grams ME,
    6. Selvin E, et al
    .: Kidney function and sudden cardiac death in the community: The Atherosclerosis Risk in Communities (ARIC) study. Am Heart J 180: 46–53, 2016pmid:27659882
    OpenUrlPubMed
  2. ↵
    1. Alonso A,
    2. Lopez FL,
    3. Matsushita K,
    4. Loehr LR,
    5. Agarwal SK,
    6. Chen LY, et al
    .: Chronic kidney disease is associated with the incidence of atrial fibrillation: The Atherosclerosis Risk in Communities (ARIC) study. Circulation 123: 2946–2953, 2011pmid:21646496
    OpenUrlAbstract/FREE Full Text
  3. ↵
    1. Deo R,
    2. Katz R,
    3. Kestenbaum B,
    4. Fried L,
    5. Sarnak MJ,
    6. Psaty BM, et al
    .: Impaired kidney function and atrial fibrillation in elderly subjects. J Card Fail 16: 55–60, 2010pmid:20123319
    OpenUrlCrossRefPubMed
  4. ↵
    1. Lee YS
    : Bradyarrhythmia can increase the plasma level of N-terminal pro-brain natriuretic peptide.Int J Arrhythm 17: 112–117, 2016
    OpenUrl
  5. ↵
    1. Liao JN,
    2. Chao TF,
    3. Liu CJ,
    4. Wang KL,
    5. Chen SJ,
    6. Lin YJ, et al
    .: Incidence and risk factors for new-onset atrial fibrillation among patients with end-stage renal disease undergoing renal replacement therapy. Kidney Int 87: 1209–1215, 2015pmid:25587708
    OpenUrlCrossRefPubMed
  6. ↵
    1. Marcos EG,
    2. Geelhoed B,
    3. Van Der Harst P,
    4. Bakker SJL,
    5. Gansevoort RT,
    6. Hillege HL, et al
    .: Relation of renal dysfunction with incident atrial fibrillation and cardiovascular morbidity and mortality: The PREVEND study. Europace 19: 1930–1936, 2017pmid:28108549
    OpenUrlPubMed
  7. ↵
    1. Molnar AO,
    2. Eddeen AB,
    3. Ducharme R,
    4. Garg AX,
    5. Harel Z,
    6. McCallum MK, et al
    .: Association of proteinuria and incident atrial fibrillation in patients with intact and reduced kidney function. J Am Heart Assoc 6: e005685, 2017pmid:28684642
    OpenUrlAbstract/FREE Full Text
  8. ↵
    1. Ohsawa M,
    2. Tanno K,
    3. Okamura T,
    4. Yonekura Y,
    5. Kato K,
    6. Fujishima Y, et al
    .: Standardized prevalence ratios for atrial fibrillation in adult dialysis patients in Japan. J Epidemiol 26: 272–276, 2016pmid:26804038
    OpenUrlPubMed
  9. ↵
    1. Shen CH,
    2. Zheng CM,
    3. Kiu KT,
    4. Chen HA,
    5. Wu CC,
    6. Lu KC, et al
    .: Increased risk of atrial fibrillation in end-stage renal disease patients on dialysis: A nationwide, population-based study in Taiwan. Medicine (Baltimore) 95: e3933, 2016pmid:27336884
    OpenUrlPubMed
  10. ↵
    1. Roy-Chaudhury P,
    2. Tumlin JA,
    3. Koplan BA,
    4. Costea AI,
    5. Kher V,
    6. Williamson D, et al
    .; MiD Investigators and Committees: Primary outcomes of the Monitoring in Dialysis Study indicate that clinically significant arrhythmias are common in hemodialysis patients and related to dialytic cycle. Kidney Int 93: 941–951, 2018pmid:29395340
    OpenUrlPubMed
  11. ↵
    1. Wong MCG,
    2. Kalman JM,
    3. Pedagogos E,
    4. Toussaint N,
    5. Vohra JK,
    6. Sparks PB, et al
    .: Bradycardia and asystole is the predominant mechanism of sudden cardiac death in patients with chronic kidney disease. J Am Coll Cardiol 65: 1263–1265, 2015pmid:25814235
    OpenUrlFREE Full Text
  12. ↵
    1. United States Renal Data System
    , 2017 USRDS Annual Data Report: Epidemiology of Kidney Disease in the United States National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases Bethesda, MD 2017. Available at https://www.usrds.org/media/2285/2017_volume_1_ckd_in_the_us.pdf. Accessed November 11, 2017
  13. ↵
    1. The ARIC Investigators
    : The Atherosclerosis Risk in Communities (ARIC) study: Design and objectives. Am J Epidemiol 129: 687–702, 1989
    OpenUrlCrossRefPubMed
  14. ↵
    1. Inker LA,
    2. Schmid CH,
    3. Tighiouart H,
    4. Eckfeldt JH,
    5. Feldman HI,
    6. Greene T, et al
    .; CKD-EPI Investigators: Estimating glomerular filtration rate from serum creatinine and cystatin C [published correction appears in N Engl J Med 367: 681, 2012]. N Engl J Med 367: 20–29, 2012pmid:22762315
    OpenUrlCrossRefPubMed
  15. ↵
    1. Fan L,
    2. Levey AS,
    3. Gudnason V,
    4. Eiriksdottir G,
    5. Andresdottir MB,
    6. Gudmundsdottir H, et al
    .: Comparing GFR estimating equations using cystatin C and creatinine in elderly individuals. J Am Soc Nephrol 26: 1982–1989, 2015pmid:25527647
    OpenUrlAbstract/FREE Full Text
  16. ↵
    Kidney Disease: Improving Global Outcomes (KDIGO) CKD Work Group, KDIGO 2012 clinical practice guideline for the evaluation and management of chronic kidney disease.Available at https://kdigo.org/wp-content/uploads/2017/02/KDIGO_2012_CKD_GL.pdf. Accessed November 11, 2017
  17. ↵
    1. Barrett PM,
    2. Komatireddy R,
    3. Haaser S,
    4. Topol S,
    5. Sheard J,
    6. Encinas J, et al
    . Comparison of 24-hour Holter monitoring with 14-day novel adhesive patch electrocardiographic monitoring. Am J Med 127: 95.e11–97, 2014
    OpenUrlCrossRefPubMed
  18. ↵
    1. ARIC
    : Visit 5/NCS Analysis Manual, 2015. Available at: https://www2.cscc.unc.edu/aric/sites/default/files/public/listings/V5%20NCS%20Analysis%20Manual_150901%20v1.pdf. Accessed November 11, 2017
  19. ↵
    1. Thompson JL,
    2. Crossman RR
    : Drug-Induced QT Prolongation. U.S. Pharmacist, 2007. Available at: https://www.uspharmacist.com/article/drug-induced-qt-prolongation. Accessed November, 2018
  20. ↵
    1. Yap YG,
    2. Camm AJ
    : Drug induced QT prolongation and torsades de pointes. Heart 89: 1363–1372, 2003pmid:14594906
    OpenUrlFREE Full Text
  21. ↵
    1. Zou G
    : A modified Poisson regression approach to prospective studies with binary data. Am J Epidemiol 159: 702–706, 2004pmid:15033648
    OpenUrlCrossRefPubMed
  22. ↵
    1. Baber U,
    2. Howard VJ,
    3. Halperin JL,
    4. Soliman EZ,
    5. Zhang X,
    6. McClellan W, et al
    .: Association of chronic kidney disease with atrial fibrillation among adults in the United States: Reasons for Geographic and Racial Differences in Stroke (REGARDS) study. Circ Arrhythm Electrophysiol 4: 26–32, 2011pmid:21076159
    OpenUrlAbstract/FREE Full Text
  23. ↵
    1. McManus DD,
    2. Corteville DC,
    3. Shlipak MG,
    4. Whooley MA,
    5. Ix JH
    : Relation of kidney function and albuminuria with atrial fibrillation (from the Heart and Soul study). Am J Cardiol 104: 1551–1555, 2009pmid:19932791
    OpenUrlCrossRefPubMed
  24. ↵
    1. Soliman EZ,
    2. Prineas RJ,
    3. Go AS,
    4. Xie D,
    5. Lash JP,
    6. Rahman M, et al
    .; Chronic Renal Insufficiency Cohort (CRIC) Study Group: Chronic kidney disease and prevalent atrial fibrillation: The Chronic Renal Insufficiency Cohort (CRIC) [published correction appears in Am Heart J 160: 1190, 2010]. Am Heart J 159: 1102–1107, 2010pmid:20569726
    OpenUrlCrossRefPubMed
  25. ↵
    1. Turakhia MP,
    2. Blankestijn PJ,
    3. Carrero JJ,
    4. Clase CM,
    5. Deo R,
    6. Herzog CA, et al
    .; Conference Participants: Chronic kidney disease and arrhythmias: Conclusions from a Kidney Disease: Improving Global Outcomes (KDIGO) controversies conference. Eur Heart J 39: 2314–2325, 2018pmid:29522134
    OpenUrlPubMed
  26. ↵
    1. Ananthapanyasut W,
    2. Napan S,
    3. Rudolph EH,
    4. Harindhanavudhi T,
    5. Ayash H,
    6. Guglielmi KE, et al
    .: Prevalence of atrial fibrillation and its predictors in nondialysis patients with chronic kidney disease. Clin J Am Soc Nephrol 5: 173–181, 2010pmid:20007681
    OpenUrlAbstract/FREE Full Text
  27. ↵
    1. Chua SK,
    2. Shyu KG,
    3. Lu MJ,
    4. Hung HF,
    5. Cheng JJ,
    6. Lee SH, et al
    .: Association between renal function, diastolic dysfunction, and postoperative atrial fibrillation following cardiac surgery. Circ J 77: 2303–2310, 2013pmid:23979566
    OpenUrlCrossRefPubMed
  28. ↵
    1. Suzuki S,
    2. Sagara K,
    3. Otsuka T,
    4. Kanou H,
    5. Matsuno S,
    6. Uejima T, et al
    .: Estimated glomerular filtration rate and proteinuria are associated with persistent form of atrial fibrillation: Analysis in Japanese patients. J Cardiol 61: 53–57, 2013pmid:23085035
    OpenUrlPubMed
  29. ↵
    1. Yonezawa Y,
    2. Horinaka S,
    3. Shirakawa C,
    4. Kogure Y
    : Estimated glomerular filtration ratio is a better index than creatinine clearance (Cockcroft-Gault) for predicting the prevalence of atrial fibrillation in the general Japanese population. Hypertens Res 41: 451–459, 2018pmid:29559690
    OpenUrlPubMed
  30. ↵
    1. Akoum N,
    2. Zelnick LR,
    3. de Boer IH,
    4. Hirsch IB,
    5. Trence D,
    6. Henry C, et al
    .: Rates of cardiac rhythm abnormalities in patients with CKD and diabetes. Clin J Am Soc Nephrol 14: 549–556, 2019pmid:30890578
    OpenUrlAbstract/FREE Full Text
  31. ↵
    1. Dalal D,
    2. de Jong JS,
    3. Tjong FV,
    4. Wang Y,
    5. Bruinsma N,
    6. Dekker LR, et al
    .: Mild-to-moderate kidney dysfunction and the risk of sudden cardiac death in the setting of acute myocardial infarction. Heart Rhythm 9: 540–545, 2012pmid:22079886
    OpenUrlPubMed
  32. ↵
    1. Deo R,
    2. Khodneva YA,
    3. Shlipak MG,
    4. Soliman EZ,
    5. Judd SE,
    6. McClellan WM, et al
    .: Albuminuria, kidney function, and sudden cardiac death: Findings from the Reasons for Geographic and Racial Differences in Stroke (REGARDS) study. Heart Rhythm 14: 65–71, 2017pmid:27523775
    OpenUrlPubMed
  33. ↵
    1. Ohyama Y,
    2. Imai M,
    3. Kurabayashi M
    : Estimated glomerular filtration rate and proteinuria are separately and independently associated with the prevalence of atrial fibrillation in general population. PLoS One 8: e79717, 2013pmid:24223186
    OpenUrlPubMed
  34. ↵
    1. Matsushita K,
    2. Coresh J,
    3. Sang Y,
    4. Chalmers J,
    5. Fox C,
    6. Guallar E, et al
    .; CKD Prognosis Consortium: Estimated glomerular filtration rate and albuminuria for prediction of cardiovascular outcomes: A collaborative meta-analysis of individual participant data. Lancet Diabetes Endocrinol 3: 514–525, 2015pmid:26028594
    OpenUrlCrossRefPubMed
  35. ↵
    1. Chen LY,
    2. Chung MK,
    3. Allen LA,
    4. Ezekowitz M,
    5. Furie KL,
    6. McCabe P, et al
    .; American Heart Association Council on Clinical Cardiology; Council on Cardiovascular and Stroke Nursing; Council on Quality of Care and Outcomes Research; and Stroke Council: Atrial fibrillation burden: Moving beyond atrial fibrillation as a binary entity: A scientific statement from the American Heart Association. Circulation 137: e623–e644, 2018pmid:29661944
    OpenUrlAbstract/FREE Full Text
  36. ↵
    1. Steinhubl SR,
    2. Waalen J,
    3. Edwards AM,
    4. Ariniello LM,
    5. Mehta RR,
    6. Ebner GS, et al
    .: Effect of a home-based wearable continuous ECG monitoring patch on detection of undiagnosed atrial fibrillation: The mSToPS randomized clinical trial. JAMA 320: 146–155, 2018pmid:29998336
    OpenUrlCrossRefPubMed
Back to top

In this issue

View Selected Citations (0)
Print
Download PDF
Email Article
Citation Tools
Request Permissions
Two-Week Burden of Arrhythmias across CKD Severity in a Large Community-Based Cohort: The ARIC Study
Esther D. Kim, Elsayed Z. Soliman, Josef Coresh, Kunihiro Matsushita, Lin Yee Chen
JASN Mar 2021, 32 (3) 629-638; DOI: 10.1681/ASN.2020030301
del.icio.us logo Digg logo Reddit logo Twitter logo Facebook logo Google logo Mendeley logo

Jump to section

Keywords