Obesity and Risk of Incident Left Ventricular Hypertrophy in Community‐Dwelling Populations With Hypertension: An Observational Study

Abstract

Background

The aim of this study was to evaluate the association between obesity and risk of incident left ventricular hypertrophy (LVH) in community‐dwelling populations with hypertension and investigate whether this association would be attenuated by a lower achieved systolic blood pressure (SBP).

Methods and Results

We used the EMINCA (Echocardiographic Measurements in Normal Chinese Adults) criteria, which were derived from healthy Chinese populations to define LVH. A total of 2069 participants with hypertension and without LVH (obesity 20.4%) were included. The association between obesity and risk of incident LVH was evaluated using Cox proportional hazard models and stratified by achieved follow‐up SBP levels (≥140, 130–139, and <130 mm Hg). These analyses were also assessed using the American Society of Echocardiography/European Association of Cardiovascular Imaging criteria, which were derived from European populations to define LVH. After a median follow‐up of 2.90 years, the rates of incident LVH in the normal‐weight, overweight, and obese groups were 13.5%, 20.3%, and 27.8%, respectively (P<0.001). In reference to normal weight, obesity was associated with increased risk of incident LVH (adjusted hazard ratio [aHR], 2.51 [95% CI, 1.91–3.29]), which was attenuated when achieved SBP was <130 mm Hg (aHR, 1.78 [95% CI, 0.99–3.19]). This association remained significant when achieved SBP was ≥140 mm Hg (aHR, 3.45 [95% CI, 2.13–5.58]) or at 130 to 139 mm Hg (aHR, 2.32 [95% CI, 1.23–4.36]). Differences in these findings were noted when LVH was defined by the American Society of Echocardiography/European Association of Cardiovascular Imaging criteria.

Conclusions

Obesity was associated with incident LVH and an SBP target <130 mm Hg might be needed to attenuate this risk in patients with hypertension and obesity.

Nonstandard Abbreviations and Acronyms

ASE/EACVI

American Society of Echocardiography/European Association of Cardiovascular Imaging

EchoNoRMAL

Echocardiographic Normal Reference Ranges of the Left Heart

EMINCA

Echocardiographic Measurements in Normal Chinese Adults

LVMI

left ventricular mass index

SPRINT

Systolic Blood Pressure Intervention Trial

STEP

Semaglutide Treatment Effect in People With Obesity

Clinical Perspective.

What Is New?

• Obesity is associated with increased risk of incident left ventricular hypertrophy (LVH) as defined by the EMINCA (Echocardiographic Measurements in Normal Chinese Adults) criteria in community‐dwelling populations with hypertension.

• A lower achieved systolic blood pressure level (<130 mm Hg) at follow‐up attenuates the risk of incident LVH associated with obesity, and differences in these findings are noted when LVH is defined by the American Society of Echocardiography/European Association of Cardiovascular Imaging criteria.

What Are the Clinical Implications?

• In patients with obesity and hypertension, a lower systolic blood pressure target may be needed to attenuate the risk of obesity and hypertension on LVH development, and the EMINCA criteria should be routinely adopted to define LVH for the Chinese populations in daily clinical practice and future research.

As estimated, 1.3 billion people globally are living with hypertension. ¹ The awareness and control rates are low, particularly in low‐ and middle‐income countries such as China, with only 15.3% having their blood pressure (BP) controlled. ² The left ventricle is a primary target for hypertension, with long‐standing high BP resulting in left ventricular hypertrophy (LVH). ³ LVH is independently associated with cardiovascular events and death in patients with hypertension. ⁴ , ⁵ , ⁶ Therefore, prevention of LVH is crucial to reduce these risks.

Over the past 5 decades, the prevalence of obesity has increased worldwide to pandemic proportions, ⁷ where the prevalence of obesity increased from 3.2% to 10.8% in adult men and from 6.4% to 14.9% in adult women from 1975 to 2014. ⁸ In China, the prevalence of obesity in adults reached 16.4%. ⁹ Obesity is a common comorbidity of hypertension. ¹⁰ , ¹¹ Obesity not only leads to BP elevation but also makes BP difficult to control, attributing to insulin resistance and aldosterone overproduction. ¹² Obesity is also an important risk factor of LVH, ³ , ¹³ and BP elevation is one of the mechanisms by which obesity induces LVH. ¹⁴ , ¹⁵ , ¹⁶ SPRINT (Systolic Blood Pressure Intervention Trial) suggested that intensive BP control with a target systolic BP (SBP) <120 mm Hg significantly reduced the incidence of LVH in comparison with those with a target SBP <140 mm Hg. ¹⁴ The STEP (Semaglutide Treatment Effect in People With Obesity) trial reported similar findings. ¹⁷ Considering the synergistic effects of hypertension and obesity on LVH development, ¹⁸ , ¹⁹ we hypothesized that a lower SBP target of <130 mm Hg might be needed to mitigate the adverse effects of obesity and hypertension on LVH development.

Echocardiography is commonly used to diagnose LVH in daily clinical practice, ²⁰ but the criteria recommended by the American Society of Echocardiography/European Association of Cardiovascular Imaging (ASE/EACVI) were derived from European populations. Results of the international EchoNoRMAL (Echocardiographic Normal Reference Ranges of the Left Heart) study have shown ethnic differences in the reference values of echocardiographic parameters. ²¹ The EMINCA (Echocardiographic Measurements in Normal Chinese Adults) study also found that the ASE/EACVI criteria might lead to misdiagnosis of LVH in Chinese patients with hypertension, ²² arguing for the necessity of applying ethnic‐specific criteria to define LVH. Accordingly, in this study, using the EMINCA criteria to define LVH, ²² we aimed to (1) evaluate the association between obesity and incident LVH in community‐dwelling populations with hypertension and (2) to investigate whether this association would be attenuated by a lower achieved SBP level (eg, <130 mm Hg) at follow‐up.

Method

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Study Design and Participants

This is an observational study of community‐dwelling patients with hypertension, which was initiated in 2010 in the Liaobu County, Dongguan, Guangdong Province, China. The aims of this study were to evaluate the prevalence and monitor the management of hypertension in the community. Patients with hypertension were screened and enrolled during the government‐sponsored annual health examination between 2010 and 2016. In brief, hypertension was defined on the basis of self‐report, current use of antihypertensive drugs, or BP ≥140/90 mm Hg at the index enrollment.

The exclusion criteria were (1) lack of echocardiographic examination at follow‐up, (2) missing information on height or body weight at baseline or follow‐up, and (3) presence of LVH at baseline. Written informed consent was obtained from participants, and this study was approved by the Clinical Research Ethic Committee of the Guangdong Provincial People's Hospital and the Liaobu County Health Department.

Data Collection and Study Variables

Demographics (age and sex), smoking status, prior medical history, and current medication use were recorded using a standardized questionnaire by trained health care staff during baseline examination. Briefly, physical inactivity was defined as exercise <1 time per week.

A blood sample was taken after overnight fasting and was used to measure fasting plasma glucose, lipid profiles, serum uric acid, and serum creatinine, which was used to calculate estimated glomerular filtration rate on the basis of the Modification of Diet in Renal Disease formula. ²³ Chronic kidney disease was defined as an estimated glomerular filtration rate <60 mL/min per 1.73 m². Body weight and height were measured using the standardized method and were used to calculate body mass index (BMI), which was normally distributed (Figure 1). Based on the recommendations of the Working Group on Obesity in China, ²⁴ overweight was defined as a BMI ≥24 kg/m² and obesity as a BMI ≥28 kg/m². Accordingly, participants were divided into 3 body weight groups: normal weight (BMI <24 kg/m²), overweight (≤BMI 24 kg/m² to <28 kg/m²), and obese (BMI ≥28 kg/m²).

Figure 1. Distribution of BMI.

BMI was normally distributed. BMI indicates body mass index.

BP Measurements

According to the Chinese Hypertension Guidelines, ²⁵ 2 BP measurements with a 1‐minute interval were performed using the Omron HEM‐7051 device (Omron HealthCare, Kyoto, Japan), and the average value was recorded. If the first 2 BP readings differed by >5 mm Hg, an additional measurement was performed, and the mean value of 3 readings was calculated and recorded. Baseline SBP level was defined as the value obtained at the baseline echocardiographic examination, and achieved SBP level at follow‐up was defined as the value obtained at the follow‐up echocardiographic examination.

Echocardiographic Examination and LVH Definition

Echocardiographic examination was performed by experienced sonographers using a Vivid S6 M4S‐RS Probe (GE Ving‐Med) equipped with a 2.5‐ to 3.5‐MHz phased‐array probe. All the procedures were performed in accordance with the guideline recommendations. ²⁶ Left ventricular (LV) end‐diastolic and end‐systolic diameters, interventricular septum, and LV posterior wall thicknesses in diastole (interventricular septum and left ventricular posterior wall) were used to calculate LV mass. Based on a prior report, ²⁷ LV mass was indexed to height^1.7 and an LV mass index (LVMI) >81 g/m^1.7 for men and >78 g/ m^1.7 for women was defined as LVH. ²² Relative wall thickness >0.51 was defined as concentric remodeling for both sexes. ²² Based on the presence of LVH and concentric remodeling, LV geometry was categorized into normal, concentric remodeling, eccentric LVH, and concentric LVH. In addition, the ASE/EACVI criteria were used to define LVH, with an LVMI >81 g/m^1.7 for men and >60 g/m^1.7 for women, and concentric remodeling was defined as relative wall thickness >0.42 for both sexes. ²⁶

Statistical Analysis

Continuous variables were presented as mean±SD if normally distributed; otherwise, they were presented as median and interquartile range. Categorical variables were presented as frequency (proportion). ANOVA or the Kruskal–Wallis H test was used to compare continuous variables, and the χ² test was used to compare categorical variables among the 3 body weight groups. Cox proportional hazard models were used to evaluate the association between overweight and obesity and risk of incident LVH, and the covariates in the models included age, sex, comorbidities (smoking status, physical inactivity, diabetes, estimated glomerular filtration rate, coronary heart disease, and stroke), achieved SBP level, and antihypertensive drugs at follow‐up. The normal‐weight group served as the reference. The association between overweight and obesity and risk of incident LVH was analyzed, stratified by achieved SBP level at follow‐up (≥140, 130–139, and <130 mm Hg). Hazard ratios (HRs) and 95% CIs were reported. The proportional hazards assumption as assessed using Schoenfeld residuals was not violated. Finally, we compared the prevalence of diabetes by incident LVH status and achieved SBP levels in the 3 body weight groups. A 2‐sided P value <0.05 was considered as statistically significant. All the analyses were performed using R version 3.4 (R Foundation for Statistical Computing, Vienna, Austria).

Results

A total of 2879 community‐dwelling patients with hypertension had 2 echocardiographic examinations between 2010 and 2016 (Figure 2). Among these participants, based on the EMINCA criteria, 2069 participants without LVH at baseline included 888 (42.9%) men and 1181 (57.1%) women (Table 1). The mean BMI was 25.05 kg/m², and the proportion of participants with normal weight, overweight, and obesity was 40.2%, 39.4%, and 20.4%, respectively. Based on the ASE/EACVI criteria, 1291 participants without LVH at baseline included 888 (68.8%) men and 403 (31.2%) women (Table S1). The mean BMI was 24.69 kg/m², and the proportion of participants with normal weight, overweight, and obesity was 43.1%, 39.7%, and 17.3%.

Figure 2. Study flowchart.

A total of 2879 hypertensive individuals had 2 echocardiographic examinations between 2010 and 2016, and after excluding those missing information on body weight or height at baseline or follow‐up (n=21), and those with LVH at baseline on the basis of the EMINCA criteria (n=789), 2069 hypertensive individuals without LVH at baseline were included. When using the ASE/EACVI criteria, there were 1291 hypertensive individuals without LVH at baseline. ASE/EACVI indicates American Society of Echocardiography/European Association of Cardiovascular Imaging; EMINCA, Echocardiographic Measurements in Normal Chinese Adults; and LVH, left ventricular hypertrophy.

Table 1.

Baseline Characteristics

	Normal‐weight group (n=832)	Overweight group (n=816)	Obese group (n=421)	P value	P value normal weight vs overweight	P value normal weight vs obese	P value overweight vs obese
Demographics
Age, y	64.77±11.85	60.26±12.02	58.12±12.27	<0.001	<0.001	<0.001	0.003
Women, n (%)	494 (59.4)	442 (54.2)	245 (58.2)	0.09	0.04	0.73	0.19
Anthropometrics
Height, cm	156.13±8.78	158.42±8.55	157.52±8.35	<0.001	<0.001	0.007	0.07
Weight, kg	52.76±7.66	64.98±7.69	75.82±10.34	<0.001	<0.001	<0.001	<0.001
Body mass index, kg/m2	21.57±1.86	25.81±1.12	30.47±2.52	<0.001	<0.001	<0.001	<0.001
Heart rate, bpm	72.91±10.50	73.30±10.31	74.10±10.60	0.17	0.45	0.06	0.20
Systolic blood pressure, mm Hg	139.25±17.16	138.32±15.85	138.83±15.59	0.52	0.25	0.67	0.59
Diastolic blood pressure, mm Hg	82.70±11.36	84.34±10.29	85.54±10.81	<0.001	0.002	<0.001	0.05
Comorbidities
Current smoking, n (%)	232 (28.0)	207 (25.4)	92 (21.9)	0.06	0.25	0.02	0.19
Physical inactivity, n (%)	382 (46.0)	349 (42.8)	179 (42.5)	0.34	0.21	0.27	0.98
Dyslipidemia, n (%)	278 (33.7)	388 (47.7)	230 (55.4)	<0.001	<0.001	<0.001	0.01
Diabetes, n (%)	94 (11.4)	122 (15.1)	89 (21.2)	<0.001	0.03	<0.001	0.00
Coronary heart disease, n (%)	23 (2.8)	26 (3.2)	15 (3.6)	0.73	0.71	0.54	0.85
Stroke, n (%)	24 (2.9)	15 (1.8)	5 (1.2)	0.11	0.21	0.09	0.53
Chronic kidney disease, n (%)	92 (11.2)	76 (9.4)	36 (8.7)	0.29	0.25	0.2	0.76
Atrial fibrillation, n (%)	3 (0.4)	5 (0.6)	4 (1.0)	0.43	0.69	0.35	0.76
Laboratory
Creatinine, μmol/L	78.51±39.62	77.77±27.72	75.34±24.43	0.26	0.66	0.13	0.13
Estimated glomerular filtration rate, mL/min per 1.73 m2	85.49±24.63	87.66±26.52	90.19±26.63	0.009	0.08	0.002	0.11
Total cholesterol, mg/dL	199.72±40.21	203.44±45.70	205.36±43.38	0.06	0.08	0.02	0.48
Triglyceride, mg/dL*	113.25 [82.18, 159.25]	144.75 [101.15, 206.67]	162.60 [118.85, 236.00]	<0.001	<0.001	<0.001	<0.001
LDL‐C, mg/dL	100.96±27.07	103.52±29.97	105.71±29.29	0.02	0.06	0.005	0.22
HDL‐C, mg/dL	53.52±12.45	49.18±14.07	48.23±12.95	<0.001	<0.001	<0.001	0.25
Fasting plasma glucose, mmol/L	5.06±1.14	5.29±1.70	5.54±1.65	<0.001	0.001	<0.001	0.01
Uric acid, μmol/L	372.10±101.35	400.07±111.53	426.75±108.38	<0.001	<0.001	<0.001	<0.001
Medications
ACEI/ARB, n (%)	416 (50.0)	452 (55.4)	245 (58.2)	0.01	0.03	0.007	0.37
Calcium channel blocker, n (%)	341 (41.0)	344 (42.2)	191 (45.4)	0.33	0.66	0.15	0.30
β Blocker, n (%)	93 (11.2)	105 (12.9)	45 (10.7)	0.43	0.32	0.86	0.30
Diuretic, n (%)	41 (4.9)	41 (5.0)	37 (8.8)	0.01	1	0.01	0.01
Antihypertensive, n (%)	592 (71.2)	613 (75.1)	313 (74.3)	0.17	0.07	0.26	0.81
No. of antihypertensive drugs	1.07±0.87	1.17±0.88	1.24±0.95	0.005	0.02	0.002	0.2
Statins, n (%)	77 (9.3)	87 (10.7)	64 (15.2)	0.006	0.38	0.002	0.02
Antidiabetic, n (%)	56 (6.7)	74 (9.1)	54 (12.8)	0.002	0.09	<0.001	0.05
MRA, n (%)	1 (0.1)	1 (0.1)	3 (0.7)	0.08	1	0.22	0.22
Echocardiographic variables
Interventricular septum, mm	9.53±1.19	9.84±1.22	9.87±1.25	<0.001	<0.001	<0.001	0.72
LVPW, mm	9.30±1.12	9.48±1.19	9.47±1.17	0.004	0.002	0.02	0.83
LVESD, mm	26.04±3.56	27.03±3.88	27.39±3.61	<0.001	<0.001	<0.001	0.11
LVEDD, mm	42.67±4.34	44.04±4.38	44.66±4.13	<0.001	<0.001	<0.001	0.01
LVMI, g/m1.7	61.09±10.41	64.70±9.43	66.77±8.58	<0.001	<0.001	<0.001	<0.001
Relative wall thickness	0.44±0.08	0.44±0.08	0.43±0.08	0.03	0.18	0.006	0.12
LVEF, %	69.26±7.13	68.75±7.13	69.02±7.69	0.35	0.14	0.57	0.54

ACEI/ARB indicates angiotensin‐converting enzyme inhibitor/angiotensin receptor blocker; HDL‐C, high‐density lipoprotein cholesterol; LDL‐C, low‐density lipoprotein cholesterol; LVEDD, left ventricular end diastolic diameter; LVEF, left ventricular ejection fraction; LVESD, left ventricular end‐systolic diameter; LVMI, left ventricular mass index; LVPW, left ventricular posterior wall thickness; LVS, left ventricular septum thickness; and MRA, mineralocorticoid receptor antagonist.

^*Presented as median (interquartile range).

Baseline Characteristics

As shown in Table 1, baseline SBP was similar between the 3 body weight groups, and participants in the obese group were younger and had a higher diastolic BP and estimated glomerular filtration rate. In addition, they were more likely to have cardiometabolic disorders and use an angiotensin‐converting enzyme inhibitor/angiotensin receptor blocker and a diuretic, and the mean number of antihypertensive drugs was higher in the obese group. Patients with obesity had a higher LVMI but a lower relative wall thickness at baseline.

Crude Rate and Cox Proportional Hazard Analysis for Incident LVH

After a median follow‐up of 2.90 (interquartile range, 2.09–3.86) years, the crude rates of incident LVH in the normal‐weight, overweight, and obese groups were 13.5%, 20.3%, and 27.8%, respectively (P value <0.001; Table 2). In reference to normal weight, overweight (adjusted HR [aHR], 1.66 [95% CI, 1.29–2.12]) and obesity (aHR, 2.51 [95% CI, 1.91–3.29]) were associated with higher risk for incident LVH. When applying the ASE/EACVI criteria, the crude rates of incident LVH in the 3 body weight groups were 24.1%, 28.9%, and 33.6%, respectively (P=0.02). After adjusting for covariates, overweight and obesity remained associated with higher risk for incident LVH.

Table 2.

Crude Rate and Cox Proportional Hazard Analysis for Incident LVH

EMINCA criteria	Normal‐weight group (n=832)	Overweight group (n=816)	Obese group (n=421)	P value	P value normal weight vs overweight	P value normal weight vs obese	P value overweight vs obese
BMI at follow‐up, kg/m2	21.78±2.70	25.71±2.13	29.75±3.23	0.001	<0.001	<0.001	<0.001
SBP at follow‐up, mm Hg	136.81±16.71	136.85±15.28	137.68±16.43	0.62	0.95	0.37	0.37
LVMI at follow‐up, g/m1.7	65.52±26.35	68.74±14.38	72.24±14.66	<0.001	0.002	<0.001	<0.001
Incident LVH, n (%)	112 (13.5)	166 (20.3)	117 (27.8)	<0.001	<0.001	<0.001	0.004
Unadjusted HR (95% CI)	Reference	1.58 (1.24–2.00)*	2.21 (1.71–2.87)*	<0.001†	…	…	…
Adjusted HR (95% CI)	Reference	1.66 (1.29–2.12)*	2.51 (1.91–3.29)*	<0.001†	…	…	…

ASE/EACVI criteria	Normal‐weight group, (n=556)	Overweight group, (n=512)	Obese group, (n=223)	P value	P value, normal weight vs overweight	P value, normal weight vs obese	P value, overweight vs obese
BMI at follow‐up, kg/m2	21.58±2.86	25.74±2.02	29.58±3.53	<0.001	<0.001	<0.001	<0.001
SBP at follow‐up, mm Hg	136.06±17.13	136.67±15.47	137.32±16.32	0.60	0.54	0.348	0.60
LVMI at follow‐up, g/m1.7	62.93±13.81	67.47±14.36	71.82±14.22	<0.001	<0.001	<0.001	<0.001
Incident LVH, n (%)	134 (24.1)	148 (28.9)	75 (33.6)	0.02	0.08	0.009	0.232
Unadjusted HR (95% CI)	Reference	1.22 (0.97–1.54)	1.54 (1.16–2.05)*	<0.001†	/	/	/
Adjusted HR (95% CI)	Reference	1.42 (1.11–1.80)*	2.10 (1.53–2.87)*	<0.001†	/	/	/

Adjustment for age, sex, comorbidities (smoking status, physical inactivity, diabetes, estimated glomerular filtration rate, coronary heart disease, and stroke), achieved SBP at follow‐up, and antihypertensive therapy at follow‐up. ASE/EACVI indicates American Society of Echocardiography/European Association of Cardiovascular Imaging; BMI, body mass index; EMINCA, Echocardiographic Measurements in Normal Chinese Adults; HR, hazard ratio; LVH, left ventricular hypertrophy; LVMI, left ventricular mass index; and SBP, systolic blood pressure.

^* P<0.05.

^† P‐trend value.

Association Between Obesity and Risk of Incident LVH Stratified by Achieved SBP at Follow‐Up

When applying the EMINCA criteria, in the subgroup with achieved SBP ≥140 mm Hg at follow‐up, the crude rate of incident LVH was significantly higher in the overweight and obese groups in comparison with the normal‐weight group (both P<0.001; Table 3). When achieved SBP was at 130 to 139 mm Hg, the difference remained significant between the normal‐weight and obese groups (P=0.006), and such difference disappeared when achieved SBP was <130 mm Hg (P=0.06). After adjusting for covariates, overweight and obesity remained independently associated with higher risk for incident LVH in those with achieved SBP ≥140 mm Hg. When achieved SBP was at 130 to 139 mm Hg, obesity remained associated with higher risk for incident LVH, which was attenuated when achieved SBP was <130 mm Hg (Table 4). When applying the ASE/EACVI criteria, differences in the incidence of LVH between the normal‐weight and obese groups were noted, and the LVH risk associated with obesity was attenuated when achieved SBP was at 130 to 139 mm Hg (Tables 3 and 4).

Table 3.

Rate of Incident LVH in the 3 Body Weight Groups According to Achieved SBP at Follow‐Up

EMINCA criteria	Achieved SBP at follow‐up (mm Hg)			P value*	P value†	P value‡	P value§
	Normal‐weight group, n (%)	Overweight group, n (%)	Obese group, n (%)
≥140	44 (12.8)	83 (24.8)	59 (32.6)	<0.001	<0.001	<0.001	0.07
130–139	29 (13.4)	46 (19.4)	29 (26.1)	0.016	0.11	0.006	0.20
<130	39 (14.4)	37 (15.2)	29 (22.5)	0.10	0.90	0.06	0.10

ASE/EACVI criteria	Achieved SBP at follow‐up (mm Hg)			P value*	P value†	P value‡	P value§
	Normal‐weight group, n (%)	Overweight group, n (%)	Obese group, n (%)
≥140	40 (18.3)	70 (32.6)	36 (37.5)	<0.001	<0.001	<0.001	0.47
130–139	39 (27.9)	36 (25.9)	19 (32.8)	0.62	0.81	0.60	0.42
<130	55 (27.9)	42 (26.6)	20 (29.0)	0.92	0.87	0.98	0.83

ASE/EACVI indicates American Society of Echocardiography/European Association of Cardiovascular Imaging; EMINCA, Echocardiographic Measurements in Normal Chinese Adults; LVH, left ventricular hypertrophy; and SBP, systolic blood pressure. Bold values are statistically significant (P<0.05).

^*Overall.

^† Normal weight vs overweight.

^‡ Normal weight vs obese.

^§ Overweight vs obese.

Table 4.

Association Between Obesity and Risk of Incident LVH Stratified by Achieved SBP at Follow‐Up

EMINCA criteria	Normal‐weight group	Overweight group	Obese group
Achieved SBP at follow‐up (mm Hg)	Adjusted hazard ratio (95% CI)
≥140	Reference	2.15 (1.37–3.36)*	3.45 (2.13–5.58)*
130–139	Reference	1.67 (0.94–2.96)	2.32 (1.23–4.36)*
<130	Reference	1.18 (0.68–2.03)	1.78 (0.99–3.19)

ASE/EACVI criteria	Normal‐weight group	Overweight group	Obese group
Achieved SBP at follow‐up, mm Hg	Adjusted hazard ratio (95% CI)
≥140	Reference	2.27 (1.41–3.66)*	3.85 (2.17–6.82)*
130–139	Reference	1.12 (0.64–1.95)	1.77 (0.88–3.54)
<130	Reference	1.04 (0.64–1.69)	1.23 (0.63–2.37)

Adjusted for age, sex, and comorbidities (smoking, physical inactivity, diabetes, estimated glomerular filtration rate, coronary heart disease, and stroke).

ASE/EACVI indicates American Society of Echocardiography/European Association of Cardiovascular Imaging; EMINCA, Echocardiographic Measurements in Normal Chinese Adults; LVH, left ventricular hypertrophy; and SBP, systolic blood pressure.

^* P<0.05.

LV Geometry at Follow‐Up

Based on the EMINCA criteria, >60% of participants had normal geometry at follow‐up in the 3 body weight groups (Figure 3A), and participants were more likely to develop concentric LVH than eccentric LVH. While based on the ASE/EACVI criteria, only nearly 30% of participants had normal geometry at follow‐up in the 3 body weight groups, and concentric remodeling was the most prevalent geometry pattern (Figure 3B). Furthermore, participants were more likely to develop eccentric LVH than concentric LVH.

Figure 3. LV geometry at follow‐up.

A, Based on the EMINCA criteria, at follow‐up, the proportion of participants with normal geometry in the normal‐weight, overweight, and obese groups was 77.0%. 70.7% and 62.2%; concentric remodeling was 9.5%, 8.9% and 10.0%; concentric LVH was 9.6%, 13.5%, and 21.6%; and eccentric LVH was 3.8%, 6.9%, and 6.2%. B, Based on the ASE/EACVI criteria, at follow‐up, the proportion of participants with normal geometry in the normal‐weight, overweight, and obese groups was 32.9%, 30.5%, and 28.3%; concentric remodeling was 43.0%, 40.6%, and 38.1%; concentric LVH was 5.8%, 8.2%, and 9.4%; and eccentric LVH was 18.3%, 20.7%, and 24.2%. ASE/EACVI indicates American Society of Echocardiography/European Association of Cardiovascular Imaging; cLVH, concentric left ventricular hypertrophy; CR, concentric remodeling; eLVH, eccentric left ventricular hypertrophy; EMINCA, Echocardiographic Measurements in Normal Chinese Adults; LV, left ventricular; and LVH, left ventricular hypertrophy.

Prevalence of Diabetes by Incident LVH Status and Achieved SBP Levels in the 3 Body Weight Groups

Among the 3 body weight groups, there was no difference in the prevalence of diabetes in those with and without incident LVH (Table S2). Similarly, there was no difference in the prevalence of diabetes in the 3 achieved SBP level groups (Table S3).

Discussion

In this study, we use the EMINCA criteria to evaluate the association between obesity and risk of incident LVH in Chinese community‐dwelling populations with hypertension. In addition, we investigate whether this association would be attenuated with a lower achieved SBP level at follow‐up. We found that (1) obesity was associated with increased risk of incident LVH in people with hypertension; (2) this association was attenuated when achieved SBP level at follow‐up was <130 mm Hg; and (3) differences in these findings were noted when LVH was defined by the ASE/EACVI criteria.

Obesity and Risk of Incident LVH

The mechanisms underlying obesity‐related LVH development are multifactorial. ²⁸ , ²⁹ In specific, aldosterone overproduction in obesity excessively stimulates mineralocorticoid receptor in the heart, resulting in cardiac fibrosis and myocardial hypertrophy. ¹² Furthermore, insulin resistance and chronic systemic inflammation in obesity are associated with endothelial dysfunction and vasoconstriction, which leads to BP elevation and increase in cardiac afterload. ¹² Prior studies have demonstrated that high body weight is independently associated with LVMI progress and LVH development. ²⁸ , ²⁹ , ³⁰ , ³¹ For example, Lee et al found that higher baseline BMI was associated with a greater increase in LVMI at follow‐up after adjusting for covariates including hypertension. ³² Bello et al evaluated the association between BMI and LVMI in a cross‐sectional study, and the results supported a positive correlation between BMI and LVMI after adjusting for covariates, including SBP and antihypertensive drugs. ³³ In patients with severe obesity, Litwin et al reported that baseline BMI was an independent predictor for LVM increase at follow‐up. ³⁴ Results of our study are consistent with prior reports and support the independent relationship between obesity and LVH development in community populations in China with hypertension. Considering the high prevalence of obesity in people with hypertension and its adverse effects on LVH development, besides BP control, additional efforts are needed to reduce body weight so as to better prevent LVH development in these populations.

Influences of Achieved SBP Level at Follow‐Up on Risk of Incident LVH

We found that the crude rate of LVH was 27.8% in patients with obesity and hypertension after a median follow‐up of 2.90 years, which was 2‐fold higher than their normal‐weight counterparts (13.5%), supporting the synergistic effects of hypertension and obesity on LVH development. ¹⁸ , ¹⁹ The mean SBP at follow‐up was 137.6 mm Hg in the obese group; however, nearly one‐fourth of the patients with obesity and hypertension still developed LVH. These findings suggest that a more intensive BP control might be needed to further mitigate the residual risk. This notion is supported by a prior study showing that treated patients with obesity and hypertension with an average SBP <120 mm Hg at follow‐up had a lower incidence of LVH than their counterparts with an average SBP >140 mm Hg at follow‐up. ³⁵ Furthermore, our study also showed a clear trend in reducing LVH incidence with a lower achieved SBP level at follow‐up in the overweight and obese groups.

In patients with severe obesity (BMI range, 35–92 kg/m²) from the United States, Avelar et al found that BP elevation appeared to intensify the effects of obesity on LVH development, ¹⁸ again supporting the synergistic effects of elevated BP and increased body weight. We recruited participants with modest obesity (mean BMI, 30.47 kg/m²), which provided complementary evidence by demonstrating that even modest obesity was associated with high LVH risk. Extending previous study, ¹⁸ our results indicate that the adverse effect of obesity could be attenuated by a lower achieved SBP level (<130 mm Hg). In aggregate, these results suggest that more intensive SBP control may be needed to prevent LVH development in patients with obesity and hypertension, which deserves further study.

Comparisons of the EMINCA Criteria and the ASE/EACVI Criteria

Consistent with a prior report, ²² our results suggest that for Chinese patients with hypertension, the ASE/EACVI criteria might lead to misclassification of LV geometry, which were largely due to the different cutoffs of relative wall thickness. The ASE/EACVI criteria would also overestimate the incidence of LVH, which were due to a relatively low cutoff of LVMI for women. Furthermore, when applying the EMINCA criteria, there was a relatively higher proportion of women without LVH at baseline than men. In contrast, when using the ASE/EACVI criteria, the proportion of women without LVH at baseline was lower than that in men. These findings highlight the importance of implementing ethnic‐specific diagnostic criteria, which might influence the sex‐specific burden of LVH.

Clinical Implications

Hypertension and obesity are highly prevalent in China and globally, ² , ⁹ and both are important risk factors of cardiovascular events and death, which cause substantial health and economic loss. Focusing on primary prevention of cardiovascular disease at the population level would be cost effective, ³⁶ , ³⁷ especially in low‐ and middle‐income countries where health resources are limited and uptake of novel medication therapy is stagnant. ³⁸ Our study provides insights into the primary prevention of cardiovascular disease, that is, a lower SBP target (<130 mm Hg) might confer greater benefits on preventing LVH development in patients with obesity and hypertension, which will likely be associated with a lower cardiovascular and mortality risk.

Strengths and Limitations

The current study is strengthened by its relatively large sample size and complete echocardiographic data at baseline and follow‐up in community‐dwelling populations. In addition, there are some limitations that should be noted. First, this is a single‐center study, and selection bias cannot be avoided. Second, as an observational study, these findings can be used only for hypothesis generation. Unmeasured confounding factors such as myocardial infarction, heart failure, or renal failure at follow‐up could influence the relationship between body weight and risk of incident LVH. Third, compared with longitudinal studies that had multiple BP measurements at follow‐up, ³⁴ , ³⁹ SBP was measured only on 1 occasion at follow‐up in this study, which might not sufficiently reflect the effects of long‐term BP control on risk of incident LVH. Fourth, in recent years, novel diagnostic and therapeutic approaches have been introduced in the management of hypertension and related cardiovascular disease, while no additional echocardiographic measurements were performed after the end of 2016. Therefore, we were unable to examine whether these novel strategies could have influenced the relationship between body weight and risk of incident LVH.

Conclusions

Obesity is an independent risk factor of LVH in community populations with hypertension, and this association is attenuated when a more stringent target SBP is achieved at follow‐up. Future studies may be warranted to investigate whether intensive BP control could prevent LVH development and reduce subsequent cardiovascular events in patients with obesity and hypertension.

Sources of Funding

The current study was supported by the Climbing Plan of Guangdong Provincial People's Hospital (Deng Feng Ji Hua 2 020 022), Guangdong Provincial Clinical Research Center for Cardiovascular disease (2020B1111170011), Guangdong Provincial Key Laboratory of Coronary Heart Disease Prevention (2017B030314041), and the Key Area Research & Development Program of Guangdong Province (No. 2019B020227005). A.E.S. is supported by a National Health and Medical Research Council Leadership Investigator Grant (Application ID 2017504).

Disclosures

None.

Supporting information

Table S1

Table S2

Table S3