Methods

Details of CHS have previously described. Briefly, CHS is a prospective population-based cohort study of risk factors for coronary heart disease and stroke in subjects 65 years and older. A total of 5,888 participants with Medicare eligibility in the United States were recruited from 4 field centers located in the following locations: Forsyth County, North Carolina; Sacramento County, California; Washington County, Maryland; and Pittsburgh, Pennsylvania. Subjects were followed with semiannual contacts, alternating between telephone calls and surveillance clinic visits. CHS clinic examinations ended in June of 1999 and since that time, 2 yearly phone calls to participants were used to identify events and collect data. The institutional review board at each site approved the study, and written informed consent was obtained from participants at enrollment.

For the purpose of this analysis, participants were excluded if they had AF at baseline, major intraventricular conduction delays (including complete bundle branch blocks and/or QRS duration ≥120 ms), or had missing baseline covariate data or missing follow-up data.

Identical electrocardiographs (Mac PC, Marquette Electronics Inc., Milwaukee, Wisconsin) were used at all clinic sites, and resting, 10-second standard simultaneous 12-lead ECGs were recorded in all participants. All ECGs were processed in a central laboratory (initially at Dalhousie University, Halifax, Nova Scotia, Canada and later at the Epidemiological Cardiology Research Center, Wake Forest School of Medicine, Winston-Salem, North Carolina). The method and prevalence of ECG abnormalities in CHS have been previously reported. Computerized automated analyses of the electrocardiographic data were performed, which included selective averaging to obtain the frontal P-wave axis. Axis values between 0° and 75° were considered normal.

Baseline AF cases were identified from the initial study ECG or by self-reported history of a physician diagnosis. AF cases also were identified during the annual study ECGs that were performed every year until 1999. Additionally, hospitalization discharge data were used to identify AF cases using International Classification of Diseases codes 427.31 and 427.32. Hospital diagnosis codes for AF ascertainment have been shown to have a positive predictive value of 98.6%.

Participant characteristics were collected during the initial CHS interview and questionnaire. Age, gender, race, income, and education were self-reported. Annual income was dichotomized at $25,000, and education was dichotomized at “high school or less.” Smoking was defined as ever (e.g., current or former) or never smoker. Participants’ blood samples were obtained after a 12-hour fast at the local field center. Measurements of total cholesterol, high-density lipoprotein cholesterol, and plasma glucose were used in this analysis. Diabetes was defined as a self-reported history of a physician diagnosis, a fasting glucose value ≥126 mg/dl, or by the current use of insulin or oral hypoglycemic medications. Blood pressure was measured for each participant in the seated position, and systolic measurements were used in this analysis. Heart rate was measured from the ECG at rest. The use of aspirin, statins, and antihypertensive medications was self-reported. Body mass index was computed as the weight in kilograms divided by the square of the height in meters. Hypertension was defined as blood pressure values >140/90 or by the use of antihypertensive medications. Baseline coronary heart disease was determined by self-reported history or by medical record adjudication of the following diagnoses: myocardial infarction, angina pectoris without myocardial infarction, coronary revascularization procedures (angioplasty and coronary artery bypass graft surgery). Baseline cases of stroke and heart failure were identified by self-reported history of a physician diagnosis followed by review of medical records.

Categorical variables are reported as frequency and percentage, whereas continuous variables are recorded as mean ± SD. Statistical significance for categorical variables was tested using the chi-square method and the Wilcoxon rank-sum procedure for continuous variables. Follow-up time was defined as the time between the initial study visit until one of the following: AF development, death, loss to follow-up, or end of follow-up (December 31, 2010). Kaplan–Meier estimates were used to compute cumulative incidence of AF by abnormal P-wave axis, and the difference in estimates was compared using the log-rank procedure. Cox regression was used to compute hazard ratios and 95% confidence interval for the association between abnormal P-wave axis and incident AF. Multivariate models were constructed as follows: model 1 adjusted for age, gender, race, education, and income; model 2 adjusted for model 1 covariates plus smoking, diabetes, coronary heart disease, stroke, heart failure, heart rate, systolic blood pressure, body mass index, total cholesterol, high-density lipoprotein cholesterol, antihypertensive medications, aspirin, and statins. We tested for interactions between our main effect variable and age (dichotomized at 75 years), gender, and race (white vs black). We also constructed a restricted cubic spline model to examine the graphical dose–response relation between P-wave axis and AF at the 5th, 50th, and 95th percentiles. Statistical significance for our main effect models and interaction terms was defined as p <0.05. SAS, version 9.3, (Cary, North Carolina) was used for all analyses.

Results

A total of 4,274 participants with complete data were used in this analysis. Baseline characteristics are reported in Table 1 .

Over a median follow-up of 12.1 years, a total of 1,274 participants (30%) developed AF. The incidence rate of AF was higher for those with abnormal P-wave axis (26 cases per 1,000 person-years) than normal P-wave axis (24 cases per 1,000 person-years). The unadjusted cumulative incidence for AF by abnormal P-wave axis is shown in Figure 1 . A U-shaped association was observed when P-wave axis was included in a restricted cubic spline model as a continuous variable ( Figure 2 ).

Cumulative incidence of atrial fibrillation (log-rank p = 0.082).

Risk of atrial fibrillation across P-wave axis. Each hazard ratio was computed with the median P-wave axis value of 63° as the reference and was adjusted for age, gender, race, education, income, smoking, diabetes, coronary heart disease, stroke, heart failure, heart rate, systolic blood pressure, body mass index, total cholesterol, HDL cholesterol, antihypertensive medications, aspirin, and statins.

In a multivariate Cox regression analysis, abnormal P-wave axis was associated with a 17% increase in the risk of AF ( Table 2 ). The association was consistent when we stratified the analysis by age, gender, and race ( Table 3 ).

Table 2

P-wave axis and risk of atrial fibrillation (n = 4,274)

P-wave Axis	Cases	Person-years	Model 1 ∗ HR (95%CI)	P-value	Model 2 † HR (95%CI)	P-value
Normal	932	39,174	1.0	–	1.0	–
Abnormal	342	13,159	1.10 (0.97, 1.25)	0.12	1.17 (1.03, 1.33)	0.02

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

1. Use of Mechanical Circulatory Support in Percutaneous Coronary Intervention in the United States
2. Comparison of Adherence to Guideline-Based Cholesterol Treatment Goals in Men Versus Women
3. Usefulness of C-Reactive Protein Plasma Levels to Predict Exercise Intolerance in Patients With Chronic Systolic Heart Failure
4. High-Output Heart Failure from a Hepatic Hemangioma With Exertion-Induced Hypoxia

Stay updated, free articles. Join our Telegram channel

Join