Author + information
- Received January 14, 2016
- Revision received March 14, 2016
- Accepted March 24, 2016
- Published online November 1, 2016.
- Wesley T. O’Neal, MD, MPHa,
- Waqas T. Qureshi, MD, MSb,
- Michael J. Blaha, MD, MPHc,
- Zeina A. Dardari, MSc,
- Jonathan K. Ehrman, PhDd,
- Clinton A. Brawner, PhDd,
- Elsayed Z. Soliman, MD, MSc, MSb,e and
- Mouaz H. Al-Mallah, MD, MScd,f,∗ ()
- aDepartment of Internal Medicine, Wake Forest School of Medicine, Winston-Salem, North Carolina
- bDepartment of Internal Medicine, Section on Cardiovascular Medicine, Wake Forest School of Medicine, Winston-Salem, North Carolina
- cJohns Hopkins Ciccarone Center for the Prevention of Heart Disease, Baltimore, Maryland
- dDivision of Cardiovascular Medicine, Henry Ford Hospital, Detroit, Michigan
- eEpidemiological Cardiology Research Center, Wake Forest School of Medicine, Winston-Salem, North Carolina
- fKing Saud bin Abdul Aziz University for Health Sciences, King Abdullah International Medical Research Center, King Abdul Aziz Cardiac Center, Ministry of National Guard, Health Affairs, Riyadh, Saudi Arabia
- ↵∗Reprint requests and correspondence:
Dr. Mouaz H. Al-Mallah, Cardiac Imaging, King Abdul-Aziz Cardiac Center, King Abdul-Aziz Medical City (Riyadh), National Guard Health Affairs Department Mail Code: 1413, P.O. Box 22490, Riyadh 11426, Kingdom of Saudi Arabia.
Objectives The goal of this study was examine the association between chronotropic incompetence and incident atrial fibrillation (AF).
Background Patients with an inadequate heart rate response during exercise may have abnormalities in sinus node function or autonomic tone that predispose to the development of AF.
Methods The association between heart rate response and incident AF was examined in 57,402 patients (mean age 54 ± 13 years, 47% female, 64% white) free of baseline AF who underwent exercise treadmill stress testing from the Henry Ford ExercIse Testing (FIT) Project. Age-predicted maximum heart rate (pMHR) values <85% and chronotropic index values <80% were used to define chronotropic incompetence. Cox regression, adjusting for demographic characteristics, cardiovascular risk factors, medications, coronary heart disease, heart failure, and metabolic equivalent of task achieved, was used to compute hazard ratios (HRs) and 95% confidence intervals (CIs) for the association between chronotropic incompetence and incident AF.
Results Over a median follow-up of 5.0 years (25th to 75th percentiles: 2.6 to 7.8 years), a total of 3,395 (5.9%) participants developed AF. pMHR values <85% were associated with an increased risk of developing AF (HR: 1.33; 95% CI: 1.22 to 1.44). Chronotropic index values <80% also were associated with an increased risk of AF (HR: 1.28; 95% CI: 1.19 to 1.38). Using varying cutoff points to define chronotropic incompetence, the associations of pMHR and chronotropic index with AF remained significant.
Conclusions Our analysis suggests that patients with inadequate heart rate response during exercise have an increased risk of developing AF.
Exercise stress testing is widely used to detect the presence of obstructive coronary artery disease (1). In addition, this testing is used to assess heart rate and atrioventricular conduction response (2). During exercise, heart rate increases linearly with workload and oxygen demand and can be expected to rise 10 beats/min per metabolic equivalent of task (MET) achieved (1). The inability of heart rate to augment cardiac output to match metabolic demands during exercise has been defined as chronotropic incompetence, and its presence is associated with an increased risk of coronary heart disease and mortality (3–5).
Chronotropic incompetence is commonly described in individuals with underlying sinus node dysfunction and conduction abnormalities. It has also been suggested that an increased prevalence of chronotropic incompetence exists in patients with chronic atrial fibrillation (AF) (6). Potentially, those individuals with chronotropic incompetence have sinus node dysfunction or abnormalities in autonomic tone that precede the development of arrhythmias such as AF (7). However, to our knowledge, this hypothesis has not been fully explored. The purpose of the present analysis, therefore, was to examine the association between chronotropic response during exercise and incident AF in the Henry Ford ExercIse Testing (FIT) Project, a racially diverse registry of men and women aimed at elucidating the association between cardiorespiratory fitness and cardiovascular outcomes.
Details of the design, procedures, and methods used in the FIT Project have been described previously (8). Briefly, the project population included 69,885 consecutive patients who underwent physician-referred exercise treadmill stress testing in the Henry Ford Health System–affiliated hospitals and ambulatory care centers throughout the metropolitan area of Detroit, Michigan, between 1991 and 2009. Data regarding treadmill testing, medical history, and medications were collected by laboratory staff at the time of testing. Follow-up data were collected from electronic medical records and an administrative claims database. Patients <18 years of age at the time of testing or those who underwent pharmacological stress testing, modified Bruce protocol tests, and other non–Bruce protocol tests were excluded from the database. The FIT Project was approved by the Henry Ford Health System institutional review board.
The present analysis examines the association between chronotropic incompetence and the risk of new-onset AF. Patients with a history of AF (n = 1,975) or valve surgery (n = 579) were excluded. Also excluded were patients with missing baseline characteristics, medication data, and/or follow-up data (n = 9,929).
Demographic and clinical characteristics were obtained at the time of treadmill testing. Age, race, sex, and smoking status were self-reported. Diabetes mellitus was defined as a previous diagnosis of diabetes, use of hypoglycemic medications including insulin, or a database-verified diagnosis of diabetes. Obesity was identified by the clinician at the time of the test. Hypertension was defined as a previous diagnosis of hypertension or a database-verified diagnosis. Blood pressure at the time of the test was not used to diagnose hypertension. Hyperlipidemia was defined according to a previous diagnosis of any major lipid abnormality or a database-verified diagnosis of hypercholesterolemia or dyslipidemia. Coronary heart disease was defined as a history of previous myocardial infarction, coronary angioplasty, or coronary artery bypass graft surgery. Heart failure was defined as a previous clinical diagnosis of systolic or diastolic heart failure.
Exercise stress testing
Exercise treadmill stress testing was conducted by using the Bruce protocol (9). Resting heart rate was measured from the baseline electrocardiogram, and blood pressure was manually measured before each stress test with each participant in the upright position. Heart rate was measured continuously during testing, and blood pressure values were measured every 3 min. Peak heart rate and blood pressure were the values recorded closest to the end of the exercise test for each participant. Initial treadmill speed was set at 2.7 km/h and increased to 4.0, 5.4, 6.7, 8.0, 8.8 km/h on minutes 3, 6, 9, 12, and 15, respectively. Peak METs were calculated by using the treadmill/electrocardiogram control on the basis of peak exercise workload.
Age-predicted maximum heart rate (pMHR) was computed as a percentage of the maximum predicted value according to the following formula: % predicted = (peak heart rate/[220 − age]) × 100%. Chronotropic incompetence for the main analysis was defined as <85% of the pMHR. The presence of chronotropic incompetence was also evaluated and defined by using different values varying between 60% and 95% of the pMHR with 5% incremental steps (<95%, <90%, <85%, <80%, <75%, <70%, <65%, and <60%). The chronotropic index was also examined by using the following formula: ([peak heart rate − resting heart rate]/[(220 − age) − resting heart rate]) × 100%. Chronotropic index values <80% are considered significant for chronotropic incompetence (4).
AF events were ascertained through linkage with administrative claim files from services delivered by the system-affiliated group practice and/or reimbursed by the system’s health plan. These files included the appropriate International Classification of Diseases-Ninth Revision, code for AF; a new diagnosis was considered present when the appropriate code (427.31) was identified in at least 3 separate follow-up encounters. Patients were censored when they lost contact with the health system.
Categorical variables are reported as frequency and percentage, and continuous variables are reported as mean ± SD. Follow-up time was defined as the date of exercise stress testing until the date of AF, death, censoring, or end of follow-up (March 2010). Kaplan-Meier estimates were used to compute cumulative incidence curves for AF by chronotropic incompetence, and the differences in estimates were compared by using the log-rank procedure. Cox regression was used to compute hazard ratios (HRs) and 95% confidence intervals (CIs). We also constructed a restricted cubic spline model to examine the graphical dose–response relationship between pMHR and AF at the 5th, 50th, and 95th percentiles (10). Multivariable models were constructed as follows: model 1 adjusted for age, sex, and race; model 2 adjusted for model 1 covariates plus resting heart rate, smoking, hypertension, diabetes, obesity, hyperlipidemia, coronary heart disease, heart failure, antihypertensive medication use, lipid-lowering medication use, aspirin, and METs achieved. We tested for interactions between the main effect variable and age (stratified according to median age), sex, race (white vs. nonwhite), hypertension, and coronary heart disease. A secondary analysis was performed to examine the association between pMHR and AF with varying cutoff points (lower values have been suggested among populations taking medications that influence heart rate and among older adults) (11,12). In addition, the association between chronotropic incompetence and AF was examined by using chronotropic index with similar cutoff points (<95%, <90%, <85%, <80%, <75%, <70%, <65%, and <60%). We also examined the association between chronotropic incompetence and AF after excluding participants who reported the use of heart rate–modifying therapies (beta-blockers, calcium channel blockers, digoxin, or amiodarone) to determine if the association between chronotropic incompetence and AF was materially altered.
The test statistic of Grambsch and Therneau (13) was used to check the proportional hazards assumption. Statistical significance was defined as p < 0.05 for the main effect model and tests for interaction. SAS version 9.3 (SAS Institute, Inc., Cary, North Carolina) was used for all analyses.
A total of 57,402 patients (mean age 54 ± 13 years, 47% female, 64% white) were included in this analysis. A total of 13,013 (23%) patients did not reach 85% of the pMHR and were classified as having chronotropic incompetence. Baseline characteristics stratified according to chronotropic incompetence are shown in Table 1.
Over a median follow-up of 5.0 years (25th to 75th percentiles: 2.6 to 7.8 years), a total of 3,395 (5.9%) participants developed AF. AF developed in 11% (n = 1,398) of patients with chronotropic incompetence and in 4.5% (n = 1,997) of patients without chronotropic incompetence. A higher incidence rate (per 1000 person-years) was observed for patients with chronotropic incompetence (incidence rate: 20.6; 95% CI: 19.6 to 21.7) than those without (incidence rate: 8.0; 95% CI: 7.7 to 8.4). The cumulative incidence curves of AF events by chronotropic incompetence are shown in Figure 1 (log-rank test: p < 0.0001). The cumulative incidence of AF increased with decreasing cutoff points to define chronotropic incompetence (Figure 2).
Chronotropic incompetence was associated with a 33% increase in the risk of AF development after adjustment for demographic characteristics and potential confounders (Table 2). Each 10% decrease in pMHR was associated with an 8% increase in the risk for AF. The results remained similar when stratified by sex, race, and hypertension. The association between chronotropic incompetence and AF was stronger for younger patients compared with older patients. The relationship was also stronger for those without coronary heart disease than for those with coronary heart disease. The proportional hazards assumption was not violated in our analysis.
A dose–response relationship was observed between pMHR and AF, with the risk increasing for lower values of pMHR (Figure 3). Chronotropic incompetence remained a significant predictor of incident AF with varying cutoff points of pMHR to define chronotropic incompetence (Table 3). Similar results were obtained with chronotropic index to define chronotropic incompetence.
When we examined the association between chronotropic incompetence and AF after excluding participants who reported heart rate–modifying therapies (n = 17,310), pMHR <85% (HR: 1.25; 95% CI: 1.09 to 1.43) and chronotropic index <80% (HR: 1.19; 95% CI: 1.07 to 1.33) remained significantly associated with AF. The association between chronotropic incompetence and AF after excluding patients who reported heart rate–modifying therapies is shown in Online Tables 1 and 2.
In this analysis from the FIT registry, we showed that the inability to achieve adequate heart rate response during exercise is independently associated with an increased risk for the development of AF. The risk of AF remained after excluding participants who reported taking heart rate–modifying therapies. In addition, we explored several cutoff points to define chronotropic incompetence, and all were associated with AF development. To our knowledge, our findings are the first to report that chronotropic incompetence is associated with the development of AF. Our data also alert practitioners to the increased risk for AF development in patients with chronotropic incompetence.
Several reports have examined the predictive ability of chronotropic incompetence. Data from the Framingham Heart Study have shown that the inability to achieve 85% of the pMHR during exercise is predictive of incident coronary heart disease and total mortality (3). In a cohort of adults referred for symptom-limited exercise treadmill stress testing at the Cleveland Clinic, there was an increased risk of death (HR: 1.84; 95% CI: 1.13 to 3.00) among those failing to achieve 85% of the pMHR (4). Similarly, an increased risk of death was observed in adults with chronotropic incompetence (<85% pMHR) independent of coronary artery disease in a cohort of adults who were not receiving beta-blockers (5).
Previous reports have largely focused on all-cause mortality and neglected the risk of arrhythmia development. An examination of patients undergoing exercise testing to determine the indication for rate-responsive pacing before primary pacemaker implantation or pacemaker replacement noted an increased prevalence of chronotropic incompetence in patients with chronic AF (6). However, no studies have explored the AF risk associated with chronotropic incompetence. Our findings also indicate that younger individuals with chronotropic incompetence have a higher risk of AF compared with older adults. Maximum heart rate decreases with age, and the higher risk in younger participants likely reflects that younger persons are more likely to be labeled as having chronotropic incompetence at higher cutoff points (12). In addition, a stronger association between chronotropic incompetence and AF was observed among those without coronary heart disease. This finding possibly reflects the increased likelihood of chronotropic incompetence to detect autonomic dysfunction in persons without coronary heart disease rather than chronotropic incompetence due to ventricular dysfunction in persons with coronary artery disease (14).
The underlying mechanisms that explain the association between inadequate heart rate response and AF are currently unknown. A potential mechanism includes an inappropriate autonomic response that favors parasympathetic dominance (15). This possibility is supported by observations of increased AF risk in those with low resting heart rates, suggesting that underlying sinus node dysfunction predisposes to AF (7,16–18). Therefore, it is plausible that patients who are unable to appropriately increase their heart rate during exercise represent a group with sinus node dysfunction that predisposes to AF development. In addition, persons with chronotropic incompetence possibly have increased myocardial fibrosis and abnormal left atrial remodeling, and both of these conditions have been associated with AF (15,19–22). Several AF risk factors (e.g., older age, hypertension) have been associated with sinus node dysfunction, and this association is another explanation to link both conditions (23–25). However, our findings remained statistically significant after adjusting for several of these common risk factors.
Current guidelines recommend the use of exercise stress testing to assess chronotropic competence, arrhythmias, and response to implanted device therapy (1). This report largely focuses on abnormalities regarding impulse initiation and conduction during exercise to identify individuals who will benefit from rate-responsive therapies. Our data suggest that exercise stress testing is also able to identify those who are more likely to develop AF. Also, the observed increased mortality risk associated with chronotropic incompetence potentially is partially related to undetected AF and its well-known thromboembolic complications (26). In addition to the mortality risk associated with chronotropic incompetence, our findings alert practitioners to a group of patients in whom targeted programs to identify AF events are possibly beneficial. Therefore, the identification of AF in patients with chronotropic incompetence potentially will reduce mortality by providing this high-risk group with therapies that are known to influence survival in AF (e.g., anticoagulation) (27).
AF events were identified by using administrative claim files that are specific to the Henry Ford Health System, and any cases that occurred in other health systems were possibly missed. However, given that many patients followed up in this health maintenance organization, it is likely that a small number of AF episodes were missed. In addition, nonpermanent cases (e.g., paroxysmal) potentially were missed. We were unable to ascertain AF events by using Holter monitors or event recorders because these data were not collected in our dataset. Our results suggest that an increased risk for AF exists with varying cutoff points to define chronotropic incompetence and the criteria used. Unfortunately, we were unable to determine a specific cutoff point to label patients as high risk, and further research is needed to determine the clinical value in which AF risk is greatest. Although other definitions exist to define pMHR during exercise, the definitions used were clinically relevant and allow for easy comparison with previous research (28). Furthermore, we included several covariates in our multivariable models that likely influenced the development of AF, but we acknowledge that residual confounding remains a possibility. For example, we were unable to account for left atrial size in our analysis.
We have shown that chronotropic incompetence is associated with an increased risk of AF. In addition to assessing sinus node function and conduction defects, exercise stress testing is able to identify patients who are at risk for developing this common arrhythmia. Further research is needed to determine the clinically relevant cutoff point to define chronotropic incompetence in which a closer evaluation for the detection of AF is warranted.
COMPETENCY IN MEDICAL KNOWLEDGE: Patients with chronotropic incompetence detected on exercise stress testing have an increased risk for the development of atrial fibrillation.
TRANSLATIONAL OUTLOOK: The well-known increased mortality associated with chronotropic incompetence is possibly related to undetected atrial fibrillation and its well-known thromboembolic complications. Our findings alert practitioners to a group of patients in whom targeted programs to identify atrial fibrillation events are beneficial.
The authors thank the patients and support staff who participated in The FIT Registry.
For supplemental tables, please see the online version of this article.
Dr. Qureshi is funded by Ruth L. Kirschstein NRSA Institutional Training Grant 5T32HL076132-10. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- atrial fibrillation
- confidence interval
- Henry Ford ExercIse Testing Project
- hazard ratio
- metabolic equivalent of task
- age-predicted maximum heart rate
- Received January 14, 2016.
- Revision received March 14, 2016.
- Accepted March 24, 2016.
- American College of Cardiology Foundation
- Fletcher G.F.,
- Ades P.A.,
- Kligfield P.,
- et al.
- Gibbons R.J.,
- Balady G.J.,
- Bricker J.T.,
- et al.
- Lauer M.S.,
- Okin P.M.,
- Larson M.G.,
- Evans J.C.,
- Levy D.
- Tanaka H.,
- Monahan K.D.,
- Seals D.R.
- Grambsch P.,
- Therneau T.
- Iwasaki Y.K.,
- Nishida K.,
- Kato T.,
- Nattel S.
- Grundvold I.,
- Skretteberg P.T.,
- Liestol K.,
- et al.
- Baldesberger S.,
- Bauersfeld U.,
- Candinas R.,
- et al.
- Patton K.K.,
- Ellinor P.T.,
- Heckbert S.R.,
- et al.
- Patton K.K.,
- Heckbert S.R.,
- Alonso A.,
- et al.
- Soliman E.Z.,
- Prineas R.J.,
- Case L.D.,
- Zhang Z.M.,
- Goff D.C. Jr..
- Jensen P.N.,
- Gronroos N.N.,
- Chen L.Y.,
- et al.
- Shroff G.R.,
- Solid C.A.,
- Herzog C.A.
- Corley S.D.,
- Epstein A.E.,
- DiMarco J.P.,
- et al.