Introduction

Left ventricular hypertrophy (LVH) is a serious cardiac disease characterized by ventricular wall thickening, increased myocardial weight, and myocardial remodeling, which further leads to heart failure [1, 2]. LVH often occurs in hypertensive patients and is the most common complication of hypertension [3, 4]. LVH significantly increases the risk of coronary atherosclerotic heart disease, heart failure, stroke, and death in hypertensive patients [2, 5, 6]. Studies have shown that obesity increases the accumulation of fat in the heart and increases the risk of hypertensive LVH [7, 8]. Therefore, early improvement of obesity assessment and early identification and reversal of LVH in patients with obesity-related hypertension can significantly reduce the risk of cardiovascular events and death. [9, 10].

WWI is a new obesity index. Previous studies have shown that WWI is positively associated with fat mass and negatively associated with muscle mass [11]. The study of approximately 1 million Korean adults found that WWI was positively associated with cardiovascular mortality and was best at predicting the risk of cardiometabolic disease and death, which has been proposed to assess obesity [11, 12]. Studies have shown that obesity increases fat accumulation in the heart, which is often associated with LVH [7]. Traditionally, body mass index (BMI) and waist circumference (WC) have been used to identify obesity and related cardiometabolic diseases. However, recently, a large epidemiological study showed that WWI was the best predictor of cardiovascular mortality compared with BMI/WC [12]. Given that LVH is an independent risk factor for cardiovascular disease, the relationship between WWI and LVH in hypertensive population remains unclear. These findings suggest that epidemiological studies should further explore the association between WWI and LVH. We hypothesized that WWI could be used as an effective indicator for the early identification of LVH risk in hypertensive population. Therefore, we conducted a cross-sectional study to clarify the association between WWI and LVH and to assess whether it was influenced by other relevant factors. In addition, subgroup analyses were performed to test the robustness of our results.

Methods

Study population

Our study protocol and exclusion criteria were described in detail in previous studies [6, 13]. This study used a multi-stage cluster sampling method to conduct a cross-sectional survey of rural residents aged 40–75 years in Xinyang County, central China during 2004–2005. A total of 13,444 villagers (5270 males and 8174 females) completed the survey. A total of 5421 hypertensive patients were enrolled. Among them, 4805 patients completed the measurement of left ventricular weight. Ninety participants were excluded due to a lack of anthropometric data. Figure 1 shows the recruitment process of the study population. This study was approved by the ethics committees of Fuwai Hospital and local hospitals. The research process follows ethical standards. All study subjects provided written consent forms and all investigators received uniform training at the Chinese Academy of Medical Sciences (Beijing, China) before the start of the study.

Fig. 1
figure 1

The recruitment process of the study population

Full size image

Covariate measurements and definitions

Demographic data, past medical history, smoking, and alcohol consumption were recorded accurately. Smoking history refers to past or current regular smoking behavior. Drinking history refers to regular drinking behavior in the past or at present. All subjects underwent routine physical examinations, including height, weight, WC, hip circumference, and resting blood pressure (blood pressure measured by a trained professional with a standard mercury sphygmomanometer). Participants should rest for at least five minutes and refrain from caffeinated beverages or exercise before taking blood pressure measurements. Measurements were taken three times, at least 30 s apart, and averaged over three times. Use the same height–weight measuring machine and ruler. During the measurement, the subjects wore light-colored clothes, took off their shoes and hats, and stood at attention. The height and weight of the subjects were accurate to 0.1 cm and 0.1 kg, respectively. We measured the WC of the standing subjects with a piece of soft tape located between the lowest rib and the iliac crest (to the nearest 0.1 cm).

WWI was a new anthropometric index to assess adiposity by standardizing WC for weight and it was calculated with the following formula: WWI = WC (cm)/\(\sqrt {{{{{{{{\mathrm{weight}}}}}}}}{{{{{\rm{(kg)}}}}}}}\) [12]. BMI was calculated with the following formula: BMI = weight (kg)/height2 (m2). All patients were fasted for at least 12 h in the morning to collect blood from the antecubital vein. Quantitative analysis of fasting blood glucose (FPG), total cholesterol (TC), low-density lipoprotein cholesterol (LDL-C), high-density lipoprotein cholesterol (HDL-C), triglycerides (TG), electrolytes and renal functional indicators, etc. Hypertension was defined as diastolic blood pressure (DBP) of ≥90 mmHg, systolic blood pressure (SBP) of ≥140 mmHg, physician diagnosis, or current hypertension medication (defined by World Health Organization in 1999) [14].

Echocardiography measurements

A HP 5500 (Phillips Healthcare System, USA) or HDI 3000 (ATL, USA) color Doppler ultrasound diagnostic instrument was used to make M-mode, two-dimensional (2D) and color Doppler recordings from parasternal long-axis and short-axis Windows. 2D and color Doppler assessments were also performed from the apical window to produce 2, 3, and 4 chamber images. We also measured end-systolic diameter, end-diastolic diameter, interventricular septal thickness, and posterior left ventricular wall thickness. LV interior dimensions, interventricular septum, and posterior wall thickness were measured during three cardiac cycles, end-diastolic and systolic, as recommended by the American Society of Echocardiography [15]. Left ventricular mass (LVM) was calculated by using the following equation: 0.8 × 1.04[(IVS + LVEDD + PW)3 − LVEDD3] × 0.6. Interventricular septal thickness (IVS) is the interventricular septum, posterior wall thickness (PW) is the posterior wall, and left ventricular end-diastolic diameter (LVEDD) is left ventricular end-diastolic diameter. LVM was divided by height2.7 to obtain LVMI (LVMIh2.7). LVH was diagnosed with the criteria of LVM ≥ 49.2 gm−2.7 in males and 46.7 gm−2.7 in females [16]. Transthoracic echocardiography was performed by 2 echocardiographers trained in echocardiography at Fuwai Hospital, Chinese Academy of Medical Sciences [1].

Statistical analyses

SPSS 26.0 statistical software was used for data management and statistical analysis. According to the baseline WWI, the subjects were divided into four groups Q1(<10.16), Q2(10.16–10.67), Q3(10.68–11.19), and Q4(≥11.20), and the parameters of the four groups were compared. The Kolmogorov–Smirnoff test is used to check whether the data has a normal distribution. Data are reported as mean ± standard deviation for continuous variables and as percentages for categorical variables. The clinical variables and echocardiographic data of all subjects were grouped by WWI quartile, the continuous variables were analyzed by ANOVA, and the categorical variables were compared by chi-square test. Logistic regression analysis was used to calculate ORs and 95% CIs. The sequence model is established. Firstly, we refer to the previous high-quality literatures in related fields, summarize the confounding corrected by most literatures, and make the next adjustment as an alternative. Secondly, after we have determined the confounders that need to be adjusted and alternative, we can make the final confirmation of alternative confounders through the results of univariate analysis. Variables with statistical significance (P < 0.05) were included in subsequent models for analysis. However, because in univariate analysis, the differences between the results could not fully reflect the relationship between the alternative confounders and the study factors or outcomes. Because univariate analysis only makes a comparison between two variables, and ignores the influence of other effects. To reduce this effect, we can appropriately relax the P value threshold to 0.1, 0.2, or even 0.4, etc. In addition, although there is no significant difference in factor analysis, the real effect between them is likely to be underestimated, masked, or reinforced. Broadening the P value can include as many alternative confounders as possible. We adjusted these variables into the model according to the steps above. Model 1 is the unadjusted variable. In model 2, adjustments were made for age and sex. In addition to confounding factors analyzed in Model 2, pulse, BMI, WC, SBP, DBP FPG, alanine aminotransferase (ALT) and aspartate aminotransferase (AST), TC, TG, LDI-C, HDL-C, serum creatinine (SCr), serum uric acid (SUA), smoking, drinking, and coronary heart disease, diabetes, stroke history were adjusted in Model 3 on the basis of Model 2. In addition, the ORs and 95% CIs of LVH and WWI in each subgroup were obtained by stratified analysis according to sex, age and BMI. The differences were considered significant if a two-tailed P value is less than 0.05.

Results

Clinical characteristics of the study population

Table 1 shows the clinical and demographic characteristics of the study population by WWI categories: as compared with the lowest WWI category (<10.68 cm/√kg), the highest WWI category (≥11.25 cm/√kg) had higher age, female proportion, BMI, WC, SBP, DBP, FPG, ALT, TC, TG, LDL-C, SCr, SUA, but lower HDL-C. In terms of echocardiographic parameters, from Q1–Q4, IVST, RWT, and LVMI parameters increased significantly with increasing WWI (all P < 0.001). The prevalence of LVH was higher in patients in the highest quartile of WWI compared with patients in the lowest quartile of WWI. The prevalence of LVH was 34.1%, 38.4%, 42.2%, and 51.5%, respectively. As this study was conducted in a remote area of China in 2004, the local rural medical level was relatively backward at that time, and the awareness rate of hypertension, drug utilization rate, and control rate were low. Participants take a wide variety of antihypertensive drugs, most of which are proprietary Chinese medicines, and the proportion is low, less than 20%. The use of statins and antidiabetic drugs was also low, at less than 5.0% and 4.8%, respectively.

Table 1 Baseline characteristics of study participants by weight-adjusted waist index
Full size table

Association between WWI and LVH

Table 2 shows the association between WWI and the risk of LVH. As shown in Table 2, WWI was associated with an increased risk of LVH in univariate analysis (Model 1). After adjustment for age and sex (Model 2) and further adjustment for pulse rate, BMI, WC, SBP, DBP, FPG, ALT, AST, TC, TG, LDL-C, HDL-C, SCr, SUA, and smoking, drinking, coronary heart disease, diabetes, stroke history (model 3), WWI and LVH remained independently associated. Compared with the lowest WWI category Q1(<10.16 cm/√kg), a WWI of Q2(10.16–10.67)cm/√kg, Q3(10.68–11.19) cm/√kg increased the probability of LVH (OR 1.33, 95% CI 1.08–1.63) and 1.50 (1.19–1.89), as did the highest WWI category (≥11.20 cm/√kg) (OR 2.28, 95% CI 1.74–2.99). We further check the multicollinearity for WC and WWI. The results showed that there were free from excessive collinearity. In addition, we found that the prevalence of LVH increased with WWI quartiles in all models (P for trend < 0.001).

Table 2 Logistic regression analysis for the association between LVH and weight-adjusted waist index
Full size table

Subgroup analyses for association between WWI and LVH

The association between the highest WWI category (≥11.20 cm/√kg) and the risk of LVH was evaluated by the subgroups sex, age, and BMI, with the lowest WWI category (<10.16 cm/√kg) as the reference. After controlling for all covariates except for the stratified variable, the relationship between WWI and LVH remained in nearly every subgroup analysis, but was not significant at male (Fig. 2). In addition, we conducted an interaction analysis of the WWI quartile with each stratified factor and found no interaction between them.

Fig. 2
figure 2

Subgroup analyses for the association between the risk of LVH and WWI categories

Full size image

Discussion

We conducted a cross-sectional study to analyze the association between WWI and LVH risk in hypertensive patients. The major finding is that the prevalence of LVH increased with WWI quartiles and that there was a significant correlation between WWI and LVH. This association was independent of sex, age, lifestyle, BMI, WC, and various cardiovascular and cerebrovascular risk factors. The stratified analysis also explored whether the relationship between WWI and LVH was influenced by gender. In fact, the relationship between WWI and LVH was more significant in female than in male. The underlying mechanism is currently unknown, but one possible explanation could be the effect of sex hormones, or it could be caused by differences in fat distribution between men and women. In addition, the proportion of males and females involved in the study is not balanced, with men far less than women, resulting in the limited statistical power of men in this study. The ASIAN-HF registry also showed gender differences. The study demonstrated that females with heart failure are more likely to have concentric left ventricular geometry than males [17]. A previous study showed found that lower estradiol levels and higher testosterone/estradiol ratios were associated with an increased risk of heart failure, which is more common in older women [18]. Decreased estradiol during menopause may affect vascular and cardiac remodeling processes, resulting in different heart failure phenotypes [19]. Our results provide new insights into the pathophysiology of LVH in men and women and provide new impetus for further research. Finally, we stratified study subjects by age and BMI, and the associations persisted.

LVH refers to cardiac enlargement characterized by cardiomyocyte hypertrophy and fibrous connective tissue hyperplasia. LVH is one of the most common heart damages associated with high blood pressure and can increase the risk of heart failure [3, 4]. However, so far, the pathogenesis of LVH is not fully understood, which may be related to age, sex, weight, race, genetic susceptibility, metabolic status (such as insulin resistance), hypertension, etc. It has been proposed that there is an inverse relationship between insulin sensitivity and LV wall thickness in subjects with essential hypertension [20, 21]. Leptin also increases the activity of extracellular matrix metalloproteinases, can induce interstitial fibrosis, and is involved in the inflammatory response during the development of LVH [22]. Disruption of the leptin signaling pathway within the heart causes LVH [23].

Obesity is a major public health problem worldwide [24]. The prevalence of obesity has increased dramatically worldwide over the last decades and nearly a third of the world’s population now classified as overweight or obese [25]. Obesity can increase the risk of diabetes, several cancers, and cardiovascular disease, and has also been found to be an independent risk factor for congestive heart failure [26,27,28]. Previous studies have shown that obesity is a risk factor for LVH and an early marker of myocardial damage in obese people [29,30,31,32]. In recent years, several studies have found that there is a positive correlation between obesity and left ventricular mass. According to the calculation method of left ventricular mass index, the incidence of LVH in obese patients ranged from 9.5% to 48.4% [33]. Obese patients are often accompanied by increased blood volume, decreased peripheral vascular resistance, and increased heart rate, which together lead to increased stroke volume, ultimately manifested as increased cardiac output and volume load. The increase in volume and pressure load will further increase the cardiomyocyte volume, change the cardiomyocyte volume, change the collagen matrix composition, and ultimately lead to LVH [34]. The mechanism of obesity-related LVH has not been fully elucidated, and it is believed that it is mainly caused by hemodynamic factors and neurohumoral factors [34, 35]. Different adipose tissues have different effects on the diseases. Visceral adipocytes have high lipolysis activity and anti-lipolysis effect of insulin, and can secrete inflammatory factors that act on blood vessels. With the increase of visceral fat, tumor necrosis factor-α, interleukin-6, C-reactive protein and other cytokines increased significantly, which can promote insulin secretion and induce pancreatic β-cell apoptosis. These factors may be involved in the process of LV remodeling hypertrophy, suggesting that visceral fat accumulation is associated with LVH.

WWI is a novel weight and waist-based obesity index with good predictive power for heart disease incidence and mortality in the Korean population [12]. Li et al. recently reported that WWI was significantly associated with an increased risk of hypertension in a rural Chinese cohort study [36]. A recent study showed that WWI is positively associated with fat mass and negatively associated with muscle mass in older adults [11]. A recent prospective study of our own showed that WWI was associated with all-cause mortality in elderly Chinese [37]. The current study verified the association between this index and LVH in Chinese adults with hypertension. The index is simple to operate, economical, and practical. It can be applied at all levels of medical and healthcare facilities, especially in areas that lack medical standards or require large-scale data research. Therefore, for people with high WWI, early assessment of target organ damage and initiation of treatment can reduce the risk of cardiovascular and cerebrovascular diseases and improve the prognosis.

Although our study recruited a large number of hypertensive patients and effectively reduced selection bias, it still had some limitations. Prospective follow-up and intervention studies were not performed in the cross-sectional study. High WWI is independently correlated with LVH, suggesting that obesity may lead to the occurrence and development of LVH. However, the causal relationship between WWI and LVH cannot be conclusively concluded. The proportion of male subjects in this study was relatively low (33.8%). Therefore, the effect of gender on the association between WWI and LVH needs to be further elucidated by expanding the proportion of male subjects.

Conclusion

In this study, WWI was independently associated with LVH, especially in females, suggesting that WWI can be used to identify patients at high risk for LVH in hypertensive populations.