Body weight, weight change and the risk of cardiovascular disease in patients with hypertension: a primary-care cohort study

Obesity and cardiovascular disease (CVD) often co-occur. However, the effects of excessive body weight and weight change on CVD in patients with hypertension are not clearly established. We examined the associations of BMI, weight change and the risk of CVD in patients with hypertension. Our Data were drawn from the medical records of primary-care institutions in China. A total of 24,750 patients with valid weight measurements attending primary healthcare centers were included. Body weight were grouped in BMI categories of underweight ( < 18.5 kg/m2), healthy weight (18.5–22.9 kg/m2), overweight (23.0–24.9 kg/m2) and obesity ( ≥ 25.0 kg/m2). Weight change over 12 months was divided into: gain >4%, gain 1–4%, stable (−1 to 1%), loss 1–4%, and loss ≥4%. Cox regression analyses were used to estimate hazard ratio (HR) and 95% confidence interval (95% CI) between BMI, weight change and the risk of CVD. After multivariable adjustment, patients with obesity were related to higher risks of CVD (HR = 1.48, 95% CI: 1.19–1.85). Higher risks were seen in participants with loss ≥4% and gain >4% of body weight compared to stable weight (loss ≥4%: HR = 1.33, 95% CI: 1.04–1.70; gain >4%: HR = 1.36, 95% CI: 1.04–1.77). Obesity and weight change of loss ≥4% and gain >4% were related to higher risks of CVD. Close monitoring and appropriate interventions aimed at achieving an optimal weight are needed to prevent adverse cardiovascular outcomes for patients with hypertension.


INTRODUCTION
Hypertension has become a major global health issue [1][2][3]. Substantial evidence has stated that hypertension is an established risk factor for cardiovascular disease (CVD) [4,5], which is the leading cause of death in modern society [6]. Excessive body weight as an important component of the metabolic syndrome predisposes to insulin resistance and promotes development of hypertension. Weight control is therefore widely promoted as a major preventive measure to reduce the burden of hypertension and subsequent CVD. In accord with this concept, current guidelines for prevention and treatment of hypertension and of CVD recommend weight reduction to improve disease status and outcome [7][8][9][10].
However, the effects of excessive body weight on CVD in patients with hypertension are less clear [11]. The LIFE (Losartan Intervention for Endpoint reduction in hypertension) study found that higher risks of cardiovascular death were observed in groups with obesity than in healthy weight groups [12]. However, a recent study suggested a survival benefit in overweight patients with hypertension compared to subjects with healthy weight [13]. Current evidence on body weight and cardiovascular outcomes mainly reported on single-time assessment of body weight, prospective studies on weight change affecting CVD are limited. Moreover, previous studies were mainly conducted in developed countries, particularly in North American and European regions. These studies have shown inconsistent results because of differences in study design, participants, and outcome definitions. Therefore, further research is urgently required.
A recent study showed nearly half of Chinese adults (35-75 years) had hypertension, highlighting a heavy burden of hypertension in China [14]. Several guidelines for patients with hypertension have been formulated in China, and weight management is highly recommended. However, there is a paucity of studies on body weight and CVD in individuals with hypertension. Therefore, we aimed to explore the relationship between body weight, weight change and the risk of CVD among patients with hypertension based on the Health Management Program of hypertension (HMPH) of primary-care institutions in China.

METHOD
This study was approved by the ethics committee of Huazhong University of Science and Technology (Number:2022S004).

Data resource
Our Data were drawn from the medical records of HMPH in community healthcare centers from Xixiang subdistrict, Shenzhen, China. A total of 38 community healthcare centers are located at Xixiang subdistrict, covering 33 communities, and providing medical service for over 1.5 million residents. According to guidelines of the HMPH, general practitioners in the primary care institutions carry out health management program for patients with hypertension, including regular follow-up evaluations and systematic health examinations. The main procedures of HMPH are conducted as follows. First, general doctors set up health records for patients with hypertension, which contains information including demographics (such as age and sex), physical checkup information (such as height and weight measurement), and disease-related information (such as duration of hypertension). Patients are then followed up every three months and undergone health checkups every year. The quarterly follow-up evaluations gather information on physical checkups (such as height, weight, and blood pressure measurement), lifestyle characteristics (such as smoking status, drinking habit and exercise frequency), and treatment modalities of the patients (such as use of antihypertensive drugs, oral antidiabetic drugs or insulin, and lipid lowering drugs). Annual health checkups contain diagnosis of CVD (such as coronary heart disease and stroke).

Body weight and weight change assessment
Body mass index (BMI) was assessed as ratio of body weight and height squared. BMI in the health record was viewed as baseline BMI and was categorized according to the WHO recommendations for Asian populations criteria: underweight ( < 18.5 kg/m 2 ), healthy weight (18.5-22.9 kg/m 2 ), overweight (23.0-24.9 kg/m 2 ) and obesity ( ≥ 25.0 kg/m 2 ) [15]. We defined weight change assessment period as 12 months and all the participants should undergo assessment period to obtain information on weight change. During this period, patients were followed up quarterly, and weight was measured during each follow-up. Weight value in the health record was viewed as baseline weight. Weight change was calculated as the percentage of the difference between baseline and 12-month follow-up. Previous study showed that 4% was an optimal threshold for defining clinically important weight change and 1% was a common cut-point for studies on weight change [16,17]. Therefore, based on relevant studies and our data, we divided weight change into five categories: gain >4%, gain 1-4%, stable -1 to 1%, loss 1-4%, and loss ≥4%.

Study population
This study was a dynamic cohort study. We enrolled 33,823 patients with hypertension from the HMPH between January 2010 and March 2020. The exclusion criteria were as follows: (1)

Definition of covariates
Data on covariates were obtained from the last follow-up record of 12month weight assessment period to obtain the updated information. Demographics, lifestyle characteristics, treatment modalities of the patients and disease information were analyzed. Demographics included age, sex, length of education and marital status. Lifestyle characteristics included smoking status, drinking habit, and exercise status. Treatment modalities included the use of antihypertensive drugs, oral antidiabetic drugs or insulin, and the use of lipid-lowering treatments. Disease information consisted of the value of systolic blood pressure (SBP), diastolic blood pressure (DBP), and durations of hypertension.

Statistical analysis
In the descriptive analysis, means ( ± standard deviation) for continuous variables and frequency (percentage) for categorical variables. For comparison of baseline characteristics between weight change groups, we used analysis of variance for continuous variables and χ 2 test for categorical variables. Events rates between baseline BMI groups and weight change categories were analyzed by adjusted Cox regression model. The survival and disease-free probability of primary outcomes according to groups was calculated using Kaplan-Meier curves, and a logrank test was performed to analyze differences. The proportional-hazards assumption was evaluated using the schoenfeld residuals test with the logarithm of the cumulative hazards function based on Kaplan-Meier estimates. Data analysis showed that all models fulfilled proportional hazards assumption (Appendix Fig. 1). Cox proportional hazards models were then used to calculate the hazard ratio (HR) and 95% confidence interval (95% CI) between categories of BMI, weight change with CVD, where healthy weight (BMI:18.5-22.9 kg/m 2 ) and stable weight (−1 to 1%) was defined as the reference. To control for potential confounding factors, we adjusted for age and sex in the first model. The second model was further adjusted for demographics, lifestyle characteristics, disease information and treatment modalities. Restricted cubic spline curves

Baseline characteristics of the study population grouped by weight change
The characteristics of the 24,750 participants grouped according to their 12-month weight change was listed in Table 1. Of all participants, 52.27% had a stable weight. The proportion of patients with gain >4%, gain 1-4%, loss 1-4%, and loss ≥4% of weight was 11.01%, 11.32%, 12.77%, and 12.64%, respectively. In contrast to subjects with stable weight, patients with gain >4% of body weight were more likely to be current smokers and drinkers. Individuals with loss ≥4% of weight were younger and related to higher baseline BMI.

Risk of cardiovascular disease and baseline BMI
During a mean follow-up period of 3.60 years, 525 individuals developed CVD. HR and 95% CI for outcomes in BMI subgroups after adjustment for age, sex and after full adjustment for all available covariables were shown Table 2. Adjusted risk models showed that patients with obesity had higher risks of CVD compared to those with healthy weight (HR = 1.48, 95% CI: 1.19-1.85). When baseline BMI was used as a continuous variable, we found patients with higher baseline BMI were at higher risks of CVD ( Fig. 2A). Table 2 reported HR and 95% CI of CVD based on different degrees of weight change. After adjusting for all confounding variables, including baseline BMI, the risk of CVD was significantly  (Fig. 2B). Patients with stable weight had the lowest risk of CVD, and the risk increased with weight fluctuation.

Subgroup analysis
We performed stratified analysis by age ( < 65 vs. ≥65 years), sex (male vs. female) and BMI groups (non-obese: BMI < 25 kg/m 2 vs. obese: BMI ≥ 25 kg/m 2 ). We observed that participants over 65 with fluctuating weight were at greater risks for CVD than patients under 65 (Fig. 3C). And participants with obesity who had weight gain were more likely to develop CVD (Fig. 3E).

DISCUSSION
Based on the HMPH of primary health care institutions in China, we explored the association between baseline BMI and weight change with CVD among patients with hypertension in a Fig. 3 Subgroup analyses of association between the baseline BMI and weight change categories with cardiovascular disease stratified by baseline BMI, age, and sex. Hazard ratio and 95% confidence intervals of outcomes by the BMI groups or weight change categories. A for baseline BMI and cardiovascular disease stratified by age. B for baseline BMI and cardiovascular disease stratified by sex. C for weight change and cardiovascular disease stratified by age. D for weight change and cardiovascular disease stratified by sex. E for weight change and cardiovascular disease stratified by baseline BMI.
real-world medical setting. Our data revealed obesity might be a risk factor for CVD among patients with hypertension. Patients with loss ≥4% and gain >4% of body weight were significantly associated with increased risks of cardiovascular outcomes compared to those with stable weight, adjusting for confounding variables, including baseline BMI. The relationship between BMI and CVD among patients with hypertension has been poorly studied. Our findings were in line with an 8-year follow-up study reported that obesity was positively related to higher risk of CVD [18], and serval studies based on general population also reported similar findings [19][20][21]. The underlying mechanism between excessive body weight and CVD may be as follows. First, obesity increases the workload of the heart, thereby causes the progress of left ventricular hypertrophy [22,23], which is associated with poor cardiovascular outcomes [24]. In addition, severe overweight and obesity are related to the activation of inflammatory response [25,26], and thus increases vascular thromboxane receptor gene expression [27], which may be associated with prognosis of adverse cardiovascular outcomes [28]. Glitches in the innate autophagy and gut microbiome homeostasis networks may also serve as equally important elements among link obesity to atherosclerosis, which may contribute to CVD [29].
The relationships of body weight and cardiovascular outcomes weres reported in most observational studies based on single-time assessment of body weight. Limited evidence addressed the association between weight change and CVD among patients with hypertension. In this study, we identified patients with 4% of weight gain was associated with higher risks of adverse cardiovascular events compared to those with stable body weight. The evidence for the associations between weight gain and high risks of cardiovascular events among general population are compelling, and a recent meta-analysis of 23 prospective cohort studies suggested that weight gain could increase the risk of multiple cardiovascular events [30]. Several biological mechanisms may underlie the associations between weight gain and CVD. Weight gain intensifies the activities of sympathetic neural [31], which increases insulin resistance [32] and reduces endothelial, kidney, and heart function to accelerate the process of CVD [33]. Moreover, weight gain is associated with increased levels of multiple inflammation-sensitive plasma proteins [34], which could contribute to inflammation, metabolic syndrome, and CVD.
We found weight loss ≥4% during one year was a warning sign of harmful cardiovascular outcomes. Although limited evidence studied on the impacts of weight loss on CVD among patients with hypertension, serval observational studies on weight change in patients with other chronic disease did not show a beneficial survival of weight reduction for improved prognosis. The PROactive study [35] and ORIGN study [36] reported that weight loss was associated with higher risks of negative cardiovascular outcomes. However, the mechanisms underlying the association weight loss and CVD remain unclear [17,37]. The possible mechanisms are as follows. Weight loss may be related to disorder of nutrients intake or absorption and is a common sign of nutritional deficiencies [38], which causes electrolyte imbalances, cardiac arrhythmias, and thus promotes CVD [17].
We observed that patients over 65 had higher risks of CVD due to weight change. However, the underlying mechanism seems to be complex and remains unclear. Previous studies stated that aging is accompanied with deterioration of cardiovascular homeostasis and metabolic disorders, which are interconnected adverse vascular and cardiac phenotypes responsible for multiple CVD [39,40]. Further research is needed to clarify if recommendations on weight management should differentiate more clearly between elder and younger patients with hypertension.
We believed that this study was one of the few studies conducted on weight change and CVD among Chinese patients with hypertension. However, there are several limitations. First, selection of study subjects who had received 12 months of followup service might have selection bias. A total of 5,205 patients without 12-month follow-up records were not included in the analysis. These subjects were comparable to analysis who were included with respect to gender distribution (Appendix Table 1). However, relatively more smokers, drinkers and infrequent exercisers were excluded. Second, with a mean follow-up of only 3.60 years, we may observe insufficient cardiovascular outcomes and have underestimated the incidence of CVD. Third, our study was an observational study and no intervention on body weight was performed. Therefore, we were unable to recognize the reasons of weight change in patients with hypertension. The specific reasons for weight change might have different effects on CVD [41,42]. Fourth, information on covariates was collected only once in this study. That meant we did not consider the covariate information of participants could change, which might have a potential effect on the results. Fifth, previous study suggested to account for confounders both at enrolment and the first follow-up to better recognize the effects of weight change [41]. However, we had information only at one time point. And the confounders were accounted only once in this study, which might also have impacts on the results. Finally, the study was conducted in Shenzhen. Nationwide studies are needed to further investigate the correlation between weight change and cardiovascular events in patients with hypertension.
In conclusion, this cohort study of community-dwelling adults with hypertension demonstrates that individuals with obesity generally exhibited more detrimental effects than healthy weight patients. Moreover, patients with loss ≥4% and gain >4% of body weight over 12 month had higher risks of CVD than patients with stable weight. Therefore, appropriate interventions aimed at achieving an optimal weight may be needed to prevent adverse health outcomes in patients with hypertension.

DATA AVAILABILITY
The datasets and computer code analyzed during the current study are available from the corresponding author on reasonable request.