DOI: https://doi.org/10.21203/rs.3.rs-1452038/v1
We aim to analyze the age and gender difference in blood pressure(BP) associations with cardiovascular outcomes and diabetes in middle aged and elderly Chinese population.
We included a total of 4,405 adults aged 40–80 years from the China Health and Nutrition Survey (CHNS) cohort. Cox proportional hazards regression models were used to estimate hazard ratios and 95% CIs. Restricted cubic spline analyses were used to explore linear and nonlinear relationships of BP with disease risk.
With a mean follow-up of 12.2 years, 627 individuals progressed to at least one condition. Multivariable adjusted Cox models showed a strong positive association between BP and cardiovascular outcomes. Each 10 mm Hg higher usual systolic BP was significantly associated with myocardial infarction(MI)(HR[95%CI]: 1.16[1.07–1.27]), stroke(HR[95%CI]: 1.31[1.23–1.40]), and diabetes(HR[95%CI]: 1.1[1.03–1.17]). In subgroup analysis, BP was strongly associated with cardiovascular outcomes in young age group and associated with diabetes in female individuals.
BP was more strongly associated with stroke than with MI and diabetes. The strength of these associations declined with increasing age. Besides, either systolic BP or diastolic BP was more strongly associated with diabetes in female individuals than male individuals.
Hypertension is one of the main causes of disease burden in China and worldwide[1][2]. Several prospective studies have demonstrated a strong, direct, linear, and continuous relationship between blood pressure(BP) and cardiovascular disease (CVD) risk[3][4], and randomized trials of antihypertensive treatments have proved the reversibility of most of the excess risks after initial treatment[5]. As part of the metabolic syndrome, hypertension and diabetes are closely associated with obesity and frequently occur together in an individual[6]. One study[7] found that individuals with hypertension had 2.34 times higher risk of developed type 2 diabetes compared with individuals without hypertension.
In China, the prevalence of hypertension rapidly increased in the past 30 years. However, awareness, treatment and control of hypertension had declined or remained unchanged in China from 2000 to 2010[8], while they had increased significantly in the developed countries[9]. Although hypertension is an established cause of cardiovascular outcomes, uncertainty remains about the age and gender difference in BP associations with the CVD and diabetes risk in China. In order to better reduce disease burden, more evidence on the association between hypertension and consequent outcome is needed. Thus, we investigated the association of BP levels with adverse cardiovascular outcomes and diabetes among 4,405 adults from nine provinces around China.
The China Health and Nutrition Survey (CHNS) was an ongoing open cohort survey of ten waves ( 1989, 1991, 1993, 1997, 2000, 2004, 2006, 2009, 2011 and 2015). The samples of nine provinces in China were obtained by a multistage, random cluster process, and the detailed CHNS samples and cohort profile information were described elsewhere [10]. The institutional review committees of the University of North Carolina at Chapel Hill and the National Institute of Nutritional and Food Safety, China Centre for Disease Control and Prevention approved the survey protocols and the process of obtaining informed consent[11].
For this study, data were drawn from the 2000–2015 CHNS cycles. We included Chinese residents with BP data collected at age 40 to 80 years and with subsequent follow-up time available(n = 6,245). After excluding persons who had a history of stroke, myocardial infarction(MI) or diabetes(n = 299), persons who were followed up only until 2006 and did not participate in the later follow-up survey in 2015(n = 985), and those without complete BP measure data at baseline(n = 556). As a result, 2,083 men and 2,322 women were available for analysis.
Participators received detailed physical examinations that included weight, height, arm and head circumference, mid-arm skinfold measurements, systolic blood pressure(SBP) and diastolic blood pressure(DBP). BP was measured 3 times after a rest in a seat, and the mean BP values were used in our study.
Detailed information on sociodemographic and lifestyle factors was collected at each wave using a structured questionnaire. Smoking status was classified into two categories as follows: current smoker or not. Drinking status was divided into two groups: current drinker or not.
Incident CVD and diabetes were defined by a self-report history during the follow-up period (2000–2015). The primary outcome in our study was a composite of the first episode of myocardial infarction(MI), stroke or diabetes during the observation period. Cases were censored at the date of diagnosis of MI, stroke, diabetes or the final visit, whichever came first.
The baseline characteristics of study participants were presented as mean ± standard deviation for continuous variables, and number (percentage) for categorical variables. Cox proportional hazards models were used to explore the relationship between baseline BP category and disease incidence. In categorical analyses, participants were divided into four groups according to baseline SBP(80–119 mm Hg, 120–139 mm Hg, 140–159 mm Hg,and 160–249 mm Hg). We chose the 80-119mmHg SBP category as the reference group. Fully adjusted models were adjusted for potential confounders including age, gender(male or female), body mass index(BMI, kg/m2), current drinker(no or yes), current smoker(no or yes), taking antihypertensive drugs(no or yes) and site(rural or urban). Tests for linear trend were computed by modeling the median values of each category as a continuous variable in regression models.
We used restricted cubic splines with four knots at the 5th, 35th, 65th, and 95th centiles to flexibly model the association of BP level with disease risk[10]. Moreover, we did stratified analyses to explore whether the association of BP level (10 mmHg increase in SBP; 5 mmHg increase in DBP) with cardiovascular events and diabetes varied across age, gender, BMI, taking antihypertensive drugs, smoking status, drinking status, and site. The main analyses are of disease associations with SBP, but the corresponding analyses for DBP are provided in the appendix.
The baseline characteristics of the 4405 participants were shown in Table 1. Within a mean follow-up of 12.2 years, 627 (14.2%) of the individuals progressed to at least one condition, of whom 304(48.5%) had diabetes, 162 (25.8%) had MI, and 242(38.6%) had stroke; 72(11.5%) had 2 or 3 of these conditions. At baseline, age, BMI and prevalence rate of taking antihypertensive drugs increased with SBP category.
Table 2 displays the association of SBP category with MI, stroke and diabetes. For the incidence of MI, compared with participants with SBP 80–119 mmHg, the multivariable adjusted HRs[95% CIs] were 1.20[0.79–1.81], 1.93[1.18–3.16], and 2.35[1.26–4.39] for MI in participants with SBP 120–139, 140–159, 160–249 mmHg, respectively; a significant positive trend in categorized SBP for incident MI(P for trend 0.001) was observed; each additional 10 mmHg of SBP was associated with a 16% higher probability of developing MI(HRs[95% CIs]: 1.16[1.07–1.27]). For the incidence of stroke, compared with participants with SBP 80–119 mmHg, the multivariable adjusted HRs[95% CIs] were 1.64[1.12–2.41], 3.12[2.03–4.81], and 5.65[3.47–9.20] for stroke in participants with SBP 120–139, 140–159, and 160–249 mmHg, respectively; a significant positive trend in categorized SBP for incident stroke(P for trend < 0.001) was observed; each additional 10 mmHg of SBP was associated with a 31% higher probability of developing stroke(HRs[95% CIs]: 1.31[1.23–1.40]). For the incidence of diabetes, compared with participants with SBP 80–119 mmHg, the multivariable adjusted HRs[95% CIs] were 1.48[1.10-2.00], 1.60[1.10–2.34], and 1.54[0.93–2.56] for diabetes in participants with SBP 120–139, 140–159, and 160–249 mmHg, respectively; a significant positive trend in categorized SBP for incident diabetes(P for trend 0.030) was observed; each additional 10 mmHg of SBP was associated with a 10% higher probability of developing diabetes(HRs[95% CIs]: 1.10[1.03–1.17]). Multivariable adjusted restricted cubic spline analyses showed a linear association of SBP with MI (P for linearity 0.001), stroke (P for linearity < 0.001), diabetes(P for linearity 0.005) and composite outcome(P for linearity < 0.001), respectively (Fig. 1).
In subgroup analysis, either SBP or DBP was more strongly associated with MI and stroke in young age group than elder age group(Fig. 2 and Fig. 3; Figure S2 and Figure s3 in appendix). For the incidence of diabetes, either SBP or DBP was more associated with diabetes in female participants, but not associated with diabetes in male participants(Fig. 4 and Figure s4 in appendix). The associations for DBP were broadly consistent with those for SBP: there were linear associations of usual DBP with stroke, diabetes and composite outcome, and J shape relation of DBP with MI throughout the BP range examined(50–120 mm Hg usual DBP)(Figure s1 in appendix). The strengths of these associations for 5 mmHg higher usual DBP were equivalent to about 10 mm Hg higher usual SBP(Figure s2, s3 and s4 in appendix).
| Characteristics | Baseline SBP level(mmHg) | |||
|---|---|---|---|---|
| 80–119(n = 1835) | 120–139(n = 1800) | 140–159(n = 553) | 160–249(n = 217) | |
| Age(year) | 51.14 ± 8.33 | 54.08 ± 9.43 | 59.48 ± 9.92 | 60.46 ± 9.89 |
| Gender | ||||
| female | 1075(58.58%) | 852(47.33%) | 275(49.73%) | 120(55.3%) |
| male | 760(41.42%) | 948(52.67%) | 278(50.27%) | 97(44.7%) |
| BMI(kg/m^2) | 22.35 ± 2.89 | 23.43 ± 3.19 | 24.20 ± 3.38 | 25.10 ± 4.14 |
| DBP(mmHg) | 71.75 ± 8.11 | 81.53 ± 7.44 | 89.67 ± 9.64 | 98.98 ± 12.21 |
| Antihypertensive drugs | ||||
| no | 1793(97.71%) | 1687(93.72%) | 458(82.82%) | 140(64.52%) |
| yes | 15(0.82%) | 72(4%) | 87(15.73%) | 73(33.64%) |
| Current smoker | ||||
| no | 1253(68.28%) | 1130(62.78%) | 373(67.45%) | 146(67.28%) |
| yes | 552(30.08%) | 632(35.11%) | 171(30.92%) | 67(30.88%) |
| Current drinker | ||||
| no | 1169(63.71%) | 1054(58.56%) | 351(63.47%) | 141(64.98%) |
| yes | 614(33.46%) | 695(38.61%) | 185(33.45%) | 69(31.8%) |
| Site | ||||
| urban | 558(30.41%) | 617(34.28%) | 200(36.17%) | 68(31.34%) |
| rural | 1277(69.59%) | 1183(65.72%) | 353(63.83%) | 149(68.66%) |
| Follw-up(year) | 12.57 ± 3.54 | 11.64 ± 3.95 | 10.53 ± 4.31 | 9.56 ± 4.64 |
| Composite outcome | ||||
| no | 1688(91.99%) | 1542(85.67%) | 412(74.5%) | 136(62.67%) |
| yes | 147(8.01%) | 258(14.33%) | 141(25.5%) | 81(37.33%) |
| Number of individuals across categories may not sum to the given number because of missing data. | ||||
| MI | stroke | diabetes | ||||
|---|---|---|---|---|---|---|
| HR[95%CI] | P | HR[95%CI] | P | HR[95%CI] | P | |
| SBP(mmHg) | ||||||
| 80–119 | 1.00 | 1.00 | 1.00 | |||
| 120–139 | 1.20[0.79–1.81] | 0.388 | 1.64[1.12–2.41] | 0.011 | 1.48[1.10-2.00] | 0.010 |
| 140–159 | 1.93[1.18–3.16] | 0.009 | 3.12[2.03–4.81] | < .001 | 1.60[1.10–2.34] | 0.014 |
| 160–249 | 2.35[1.26–4.39] | 0.007 | 5.65[3.47–9.20] | < .001 | 1.54[0.93–2.56] | 0.094 |
| P for trend | 0.001 | < .001 | 0.030 | |||
| 10mmHg increase | 1.16[1.07–1.27] | 0.001 | 1.31[1.23–1.40] | < .001 | 1.10[1.03–1.17] | 0.006 |
| All models were adjusted for age, gender, BMI, antihypertensive drugs, smoking status, drinking status, and site. | ||||||
In this study of 4405 patients accruing 627 disease events during 12.2 years mean follow-up, BP was positively associated with the adverse cardiovascular outcome and diabetes with adjustment for potential risk factors, and this association was steepest for stroke compared with MI and diabetes(Fig. 1–4). These linear associations were similar to those from most previous observational studies including the Prospective Studies Collaboration[4] and the Framingham study (in which J-shape associations with DBP occurred only when accompanied by SBP > 140 mm Hg)[13], and is supported by trial evidence in elderly people that shows significant and sustained risk reduction for cardiovascular endpoints with lowering of BP[14].
In our result, BP level was more associated with stroke than MI(Fig. 1), which has also been suggested by the Prospective Studies Collaboration[4] and another observational study[15]. Further, in a meta-analysis that included 147 RCTs, Law et al[16]calculated the CVD risk reduction for the average active treatment versus control trial difference in SBP (10 mm Hg). The reductions in stroke and coronary heart disease(CHD) incidence rates were similar to the benefits expected on the basis of a 10 mm Hg difference in SBP in the Prospective Studies Collaboration meta-analysis of observational studies. Benefit for the same difference in BP was greater for stroke, reflecting the higher risk of elevated BP for cerebral vessels compared with the coronary circulation. In addition, our result also showed that the risk of MI might be higher in low DBP level group(Figure s1). This result was similar to a previous observational study[17], which reported that low DBP was associated with subclinical myocardial damage and CHD events among adults with SBP ≥ 120mmHg, and thus elevated pulse-pressure.
For subgroup analysis, either SBP or DBP was more strongly associated with the risk of adverse cardiovascular outcomes in individuals aged 40–60 years than in those aged 60–80 years(Fig. 2 and Fig. 3; Figure s2 and Figure s3 in appendix). Other observational studies demonstrated that the proportional associations of BP with all major cardiovascular diseases become less steep with age, although they remain strong and direct even in the oldest age groups[15][18]. Besides, we also found that either SBP or DBP was associated with diabetes in female individuals, but not associated with diabetes in male individuals(Fig. 4 and Figure s4 in appendix). Findings from the Women’s Health Study[19] also suggested that BP and BP progression are independent predictors of incident type 2 diabetes among women individuals. In contrast, one study[20] in men did not find an significant association between BP and new-onset diabetes. The sex difference for the association of BP and diabetes needs to be demonstrated in further studies. Further, the association was consistently weaker in people who received antihypertensive drugs at baseline than in those who did not. Similar to our result, high-quality meta-analyses have demonstrated the effectiveness of BP lowering for prevention of CVD[5][16][21][22].
Our study has several limitations. First, diabetes, MI and stroke are based on self-report, which may have introduced bias in outcome ascertainment. Second, we have not accounted for changes in BP and other risk factors over time. A more detailed analysis of lifetime risks would take repeated measures into consideration to account for potential changes and would adjust for the time-dependent effect of medications taken long term. Finally, some other risk factors (such as cholesterol values, air pollutants and dietary patterns) are shown to contribute to CVD risk, we were unable to involve them in present analyses due to the lack of these information in this survey.
In conclusion, our study found the relationship between BP and cardiovascular outcomes and diabetes in the Chinese population. BP is strongly and positively related to the risk of cardiovascular outcomes and diabetes, and this positive association was more steep for stroke than for MI. Either SBP or DBP was more strongly associated with diabetes in female individuals than male individuals. These results provide more evidence for guidelines or health policies of primary prevention of CVD and diabetes with the management of BP level. Individual-level and population-level interventions to control BP level are needed to reduce the disease burden in China.
Ethics approval and consent to participate
This research uses data from the China Health and Nutrition Survey (CHNS). The institutional review committees from the University of North Carolina at Chapel Hill and the National Institute for Nutrition and Food Safety, China Centre for Disease Control and Prevention, approved the survey protocols and instruments and the process for obtaining informed consent for the survey. All participants and/or their parents/guardians provided written informed consents for their participation in the survey.
Consent for publication
Not applicable.
Availability of data and materials
Data from China Health and Nutrition Survey was used in this study, which can be downloaded at http://www.cpc.unc.edu/projects/china/data/datasets.
Competing interests
The authors declare that there is no conflict of interest.
Funding
Not applicable.
Author contributions statement
QY and QDJ conceptualized and designed the study, carried out the initial analyses, drafted the initial manuscript, and reviewed and revised the manuscript;
QY and QDJ critically reviewed and revised the manuscript; and all authors approved the final manuscript for submission.
Acknowledgements
We acknowledge CHNS database for providing their platforms and contributors for uploading their meaningful datasets.