Study design and population

The Kailuan Study is a prospective population-based cohort study involving 101,510 men and women aged 18–98 years in the Kailuan community in Tangshan of China. The design and methods of the study have been described in detail previously [19], [20], [21]. The protocol was approved by the Ethics Committee of both Kailuan General Hospital and Beijing Tiantan Hospital in compliance with the Declaration of Helsinki.

In brief, all participants provided a written informed consent and underwent a clinical examination and a standardized interview including questions on smoking and alcohol consumption status, socioeconomic parameters, physical activity, histories of arterial hypertension, coronary heart disease, diabetes mellitus, hyperlipidemia, and stroke, with the current treatments of these diseases. Anthropomorphic parameters, such as body height, weight, and waist circumference were measured. The body mass index was calculated as the ratio of body weight (kilograms) divided by the square of body height (meters). Blood pressure was measured on the left arm to the nearest 2 mm Hg using a mercury sphygmomanometer with a cuff of appropriate size. Fasting blood samples were biochemically examined for the concentration of glucose, high-density lipoproteins (HDL), low-density lipoproteins (LDL), triglycerides, total cholesterol, and high-sensitive C-reactive protein (hs-CRP).

Hypertension was defined as having a history of hypertension, a systolic blood pressure (SBP) ≥140 mmHg, a diastolic blood pressure (DPB) ≥90 mmHg, or using antihypertensive medications. Diabetes mellitus was diagnosed if the subject had a history of diabetes mellitus, was currently using insulin or oral hypoglycemic agents, or had a fasting blood glucose level was ≥126 mg/dL. Hyperlipidemia was defined as having a history of hyperlipidemia, total blood cholesterol levels ≥220 mg/dL, triglyceride levels ≥150 mg/dL, or taking antihyperlipidemic medications.

The study participants were refrained from smoking and drinking of coffee, tea, or alcohol for at least three hours nor performing any exercise for at least 30 minutes prior to the RHR measurement. A 10-second 12-lead electrocardiography was used to measure the RHR after the participants rested in supine position for 5 minutes. The number of R-R intervals (number of QRS complexes - 1) was divided by the time difference between the first and last beat, and the result was converted to beats per minute (bpm) [20].

The interview and all examinations were carried out by specially trained medical doctors and nurses.

Follow-ups and outcomes

The participants were followed up by face-to-face interviews at every two-year medical examination routine until December 31, 2010 unless CVDs events or death occurred. The follow-ups were performed by hospital physicians, research physicians, and research nurses. For the participants without face-to-face follow-ups, their outcome information was obtained by referring to death certificates from provincial vital statistics offices, discharge summaries from the 11 hospitals, or medical records from medical insurance companies [22], [23].

The primary outcome was the first occurrence of any CVDs including either the first nonfatal CVDs event or a CVDs death. Any stroke event included ischemic stroke or hemorrhagic stroke event The diagnosis of all CVDs events was confirmed through a medical record review, using the World Health Organization criteria [24], [25].

Statistical analyses

We stratified the study population into quintiles based on the RHR (Q1 group: participants with RHR ≤66 bpm, Q2: participants with the RHR range of 67–70 bpm, Q3 group: participants with the RHR range of 71–74 bpm, Q4: participants with the RHR range of 75–80 bpm, and Q5 group: participants with RHR ≥81 bpm). Continuous variables were described by means ± standard derivation (SD) and categorical variables were described by percentages. We used the ANOVA test for non-paired samples of normally distributed parameters and the Kraskal-Wallis test for non-parametric variables. The Chi-squared test was applied for the comparison of categorical variables. We used a Cox proportional hazards model to calculate hazard ratios (HRs) and their 95% confidence intervals (CIs) for all CVD, stroke, MI, ischemic stroke, hemorrhagic stroke and all-cause death, respectively by treating the lowest RHR quintile (Q1 group) as the reference category. Person-years of follow-up were calculated as the time from baseline assessment to either the development of the first endpoint of interest, censoring, or the end of follow-ups (December 31, 2010). We created three multivariate-adjusted models. Model 1 was adjusted for age and sex. In addition to adjusting for the confounders in Model 1, Model 2 also adjusted for the following confounders: average monthly income of each family member, education level, marital status, body mass index, waist circumference, smoking status, drinking status, and physical activity., Model 3 adjusted for hs-CRP, hypertension, diabetes mellitus, and dyslipidemia apart from the confounders that were adjusted in Model 2.

Finally, we also conducted the above three Models by treating the RHR as a continuous variable in order to test the liner relationship between RHR and CVD events.

Two-sided P-values were reported for all analyses. A P-value <0.05 was considered to be statistically significant. All statistical analyses were performed by SAS software, version 9.3 (SAS Institute Inc., Cary, NC, USA).

All-cause cardiovascular disease

Compared with the lowest RHR group, The highest HR (95% CI) for the risk of all-cause CVDs was 1.40 (1.21–1.61) in Model 1 and 1.39 (1.20–1.61) in Model 2, respectively (Table 2). However, there was no significant relationship between RHR and all-cause CVD after adjusting all confounders in Model 3. After treating heart rate as a continuous variable, the results were similar.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Hazard ratios (95% CI) for cardiovascular disease and all-cause death according to quintiles of resting heart rate.

https://doi.org/10.1371/journal.pone.0110985.t002

Myocardial infarction

After controlling for all the potential confounders in Model 3, participants in the highest RHR group (≥81 bpm) had a statistically significant increased risk of 45% for MI when compared with those in the lowest RHR group (≤66 bpm) (Table 2). The risk of MI was significantly increased by 10% with an increase of 10 bpm (Table 2).

Any stroke, ischemic stroke, and hemorrhagic stroke

There was a significant association between increased RHR and any stroke and ischemic stroke in Model 1 and Model 2. But after adjusting for all confounders in Model 3, the relationship disappeared completely (Table 2). There was no significant association between RHR and hemorrhagic stroke in the three Models (Table 2).

All-cause death

All-cause mortality was increased in a graded trend from the lowest to the highest RHR group (1.45%, 1.38%, 1.64%, 1.85%, and 2.31%, respectively) (Table 2). After adjusting for all confounders, participants in the Q4 group (75–80 bpm) and those in the Q5 group (≥81 bpm) had more risk for all-cause death than those in the reference group (≤66 bpm) and HRs were 1.38 (95% CI, 1.19–1.61) and 1.51 (95% CI, 1.29–1.77), respectively. After treating heart rate as a continuous variable, the risk for all-cause death was increased by 18% per 10 bpm (HR: 1.18 [95% CI, 1.13–1.23]) after adjusting all confounders (Table 2).