Evaluation of the associations of body height with blood pressure and early‐stage atherosclerosis in Chinese adults

Abstract

Body height has been recently related to the risk of coronary heart disease and metabolic risk factors. However, data are scarce regarding the relationship between body height and early‐stage atherosclerotic changes, especially in Chinese individuals. In this study, we aimed to comprehensively examine the associations of body height with early‐stage atherosclerosis and blood pressure in Chinese adults. Carotid‐femoral pulse wave velocity (cfPWV), carotid‐radial pulse wave velocity (crPWV), carotid artery‐dorsalis pedis pulse wave velocity (cdPWV), and body height were measured in 5098 men and women. All samples were obtained from a community‐based health examination survey in central China. After adjusting for sex, age, weight, fasting glucose level, lipid level, creatinine, and heart rate, low body heights were significantly associated with higher cfPWV, crPWV, and blood pressure (all P for trend <.01), whereas no significant association was found between body height and cdPWV. In addition, we found a significant interaction between prehypertension status and body height in relation to cfPWV, after adjusting for covariates (P for interaction = .0024). The associations were stronger in participants with prehypertension than in those with normal blood pressure. Compared to the group with the tallest stature and normal blood pressure, individuals in the group with the shortest stature and prehypertension had nearly a 2.5 m/s higher cfPWV. These results indicate that short body height was associated with an increased risk of early‐stage atherosclerosis in Chinese adults, independent of traditional cardiometabolic risk factors. Prehypertension might modify the association between body height and cfPWV.

1. INTRODUCTION

The relationship between body height and cardiovascular risk has long been noted, ¹ , ² , ³ and the causality of body height with coronary heart disease was recently established through genetic analyses. ⁴ , ⁵ , ⁶ In addition, it has been reported that body height is closely related to cardiovascular risk factors such as lipid levels and hypertension. ⁷ , ⁸ , ⁹ The present study reported that the blood pressure to body height ratio is a useful screening tool for prehypertension and hypertension in some populations. However, few studies have assessed whether body height affects early‐stage atherosclerotic changes, and the data from Chinese studies are even sparse. ¹⁰

Aortic stiffness is an established marker for early‐stage atherosclerosis. Pulse wave velocity (PWV) is the gold standard for assessing arterial stiffness and has been widely used as an indicator of early‐stage atherosclerosis. ¹¹ , ¹² PWV measured at different arterial sites may reflect atherosclerotic alterations at central arteries (eg, cfPWV) or peripheral arteries (eg, crPWV and cdPWV). The cfPWV is considered an important parameter of aortic arterial stiffness, ¹³ whereas cdPWV and crPWV are indicators of early alterations in peripheral vascular dynamics. ¹⁴

We hypothesize that short body height was associated with an increased risk of early‐stage atherosclerosis. In the present study of a large sample of Chinese adults, we aimed to comprehensively investigate the relationship between body height and PWV measures for central and peripheral arterial stiffness. We particularly assessed whether traditional cardiovascular risk factors such as blood pressure (BP) and levels of lipids, creatinine, and glucose modified the associations.

2. METHODS

2.1. Participants

This study was designed as part of the Cardiometabolic Risk in Chinese Study. We performed a community‐based health examination survey of 7431 individuals (age range, 18‐93 years) who were sequentially recruited from residents living in the urban area of central China. Details of this study have been presented elsewhere. ¹⁵ , ¹⁶ , ¹⁷ , ¹⁸ , ¹⁹ All participants completed a series of medical examinations and blood laboratory tests and simultaneously provided demographic details and medical history at the time of their clinical consultation. People who were taking any medication or had any clinically manifested illness, and participants with missing PWV values or omitted blood sampling were excluded. We enrolled 5098 participants with appropriate and sufficient data in this study.

The study was reviewed and approved by the ethics committee of the Central Hospital of Xuzhou, Affiliated Hospital of Medical School of Southeast University, China, and all participants provided written informed consent for participation.

2.2. Assessment of body height

Body height was recorded to the nearest 0.5 cm without shoes using a standardized wall‐mounted body height board. Measured body height was categorized in quintiles according to sex. The average measured body heights among the participants were 171.21 cm (standard deviation [SD]: 5.83 cm) and 158.87 cm (SD: 5.79 cm) for men and women, respectively.

2.3. Assessment of PWVs

All measurements were performed in a quiet room with controlled ambient temperature. The cfPWV, a marker of central aortic stiffness, was measured using an automatic waveform analyzer (Complior System; Artech‐Medical Corp.) after 5 minutes of bed rest in the supine position. Carotid and femoral artery pressure waveforms were recorded with multiarray tonometry sensors at the left femoral artery and left carotid arteries. The electrocardiography examination was performed by placing electrodes on both wrists. The cdPWV and crPWV, markers of peripheral arterial stiffness, were obtained in a similar way, with the pulse wave being measured simultaneously in the right radial dorsum of the foot and right carotid arteries. The PWV was based on the distance/time ratio (meters/second); it was calculated as the path length between arterial sites of interest divided by the time delay between the foot of the respective waveforms and expressed as m/s. ²⁰ Sixteen PWVs were measured for each participant. We averaged 10 PWVs after removing the three maximum and three minimum values.

2.4. Other anthropometric measures

Body weight was recorded to the nearest 0.1 kg using a standard beam balance scale with participants wearing light indoor clothing and no shoes. Body mass index (BMI) was calculated as weight (in kg) divided by body height (in meters squared). BP was measured three times consecutively in the seated position by trained clinic nurses, using a mercury sphygmomanometer after a resting period of at least 5 minutes in the supine position. The participant's arm was placed at the level of the heart, and the recorded BP value was the mean of three measurements. Participants who had a sitting BP ≥140/90 mm Hg or who were taking antihypertensive drugs regularly were defined as hypertensive, and prehypertension was defined as systolic BP (SBP) between 120 and 139 mm Hg or diastolic BP (DBP) between 80 and 89 mm Hg according to the seventh report of the Joint National Committee on Prevention, Detection, Evaluation and Treatment of High BP. ²¹

2.5. Assessment of biomarkers

Venous blood sampling from all participants was obtained after overnight fasting (8‐12 hours). The blood was transferred into glass tubes and allowed to clot at room temperature for 1‐3 hours. Subsequently, clotting serum was immediately separated by centrifugation for 15 minutes at 1760 g. Blood samples were collected to measure glucose, serum uric acid, serum creatinine, total triglyceride (TG), total cholesterol (TC), high‐density lipoprotein cholesterol (HDL‐C), and low‐density lipoprotein cholesterol (LDL‐C) levels using an autoanalyzer (Type 7600; Hitachi Ltd.). Participants underwent a 75‐g oral glucose tolerance test (OGTT). Blood samples were drawn at 120 minutes after the glucose or carbohydrate load. The glycated hemoglobin A1c (HbA1c) was measured using high‐performance liquid chromatography (HLC‐723G7 hemoglobin HPLC analyzer; Tosoh Corp.) according to the standard method. The fasting insulin level was measured by the radioimmunoassay method (Pharmacia).

2.6. Statistical analyses

Data management and statistical analysis were conducted using SAS statistical software (version 9.1; SAS Institute, Inc). The relationship between body height levels (in quintiles) with BP or PWVs was examined using general linear regression models after adjusting for covariates, including sex, age, weight, BP, lipid profiles, and heart rate (HR), when they were not the strata variables. Multivariate linear regression models were used to test the potential interactions of body height with prehypertension status in relation to the cfPWV. All the reported P‐values are two‐tailed. Variables with P‐values <.05 were considered statistically significant.

3. RESULTS

3.1. Characteristics of the study participants by body height levels

The study population was represented by 56.32% of men, with an average age of 46.69 years. The baseline information of all variables was shown in Table S1. Table 1 shows the study participants' characteristics according to body height levels (in quintiles). Age, BMI, levels of fasting glucose, fasting insulin, 2‐hour OGTT, HbA1c, TC, and serum creatinine showed significant differences across body height groups, with a decreasing trend as the body height increased. We have also shown the levels of association (correlations) for all variables in relation to the levels of cfPWV, crPWV, and cdPWV as Table S2.

TABLE 1.

Participants' characteristics by body height

	Body height (in quintiles)					P‐value
	Q1 (women <154.1, men <166.1)	Q2 (women 154.1‐158, men 166.1‐170)	Q3 (women <158.1‐160, men 170.1‐172)	Q4 (women 160.1‐164, men 172.1‐176)	Q5 (women >164, men >176)
N	1156	1300	743	1061	838
Sex (male)	646	741	400	604	480	.614
Age, y	54.92 ± 13.77	49.74 ± 12.52	47.90 ± 11.80	45.46 ± 11.75	43.54 ± 10.71	<.001
BMI, kg/m2	25.24 ± 3.32	24.77 ± 3.09	24.69 ± 3.13	24.37 ± 3.15	24.22 ± 3.22	<.001
HR, bpm	70.56 ± 11.02	70.78 ± 10.28	70.08 ± 10.01	70.28 ± 9.58	70.28 ± 9.91	.533
Fasting glucose level, mmol/L	5.06 ± 0.54	5.03 ± 0.52	4.99 ± 0.49	5.01 ± 0.50	4.98 ± 0.51	.001
Fasting insulin level, IU/mL	8.22 ± 5.05	8.83 ± 5.86	8.84 ± 5.01	8.89 ± 4.81	9.27 ± 5.28	<.001
2‐h OGTT, mmol/L	6.53 ± 1.54	6.27 ± 1.54	6.11 ± 1.39	6.07 ± 1.52	5.82 ± 1.34	<.001
HbA1c level, %	5.36 ± 0.43	5.31 ± 0.40	5.26 ± 0.39	5.24 ± 0.38	5.22 ± 0.38	<.001
Total cholesterol level, mmol/L	5.13 ± 0.96	5.05 ± 0.89	5.01 ± 0.91	4.97 ± 0.89	5.00 ± 0.89	.001
Triglyceride level, mmol/L	1.54 ± 1.23	1.58 ± 1.56	1.53 ± 1.44	1.54 ± 1.37	1.59 ± 1.67	.886
HDL‐C level, mmol/L	1.25 ± 0.32	1.25 ± 0.29	1.25 ± 0.28	1.26 ± 0.31	1.25 ± 0.30	.797
LDL‐C level, mmol/L	3.04 ± 0.80	3.00 ± 0.76	2.98 ± 0.77	2.97 ± 0.75	2.96 ± 0.75	.584
SUA level, umol/L	299.88 ± 78.78	297.44 ± 79.13	296.46 ± 83.12	296.98 ± 79.79	297.61 ± 80.90	.962
Serum creatinine level, mg/dL	66.76 ± 15.82	65.42 ± 13.06	64.61 ± 12.96	65.02 ± 12.58	66.53 ± 13.12	.023

Data are presented as a mean ± SD. A linear regression model was used to test trend for continuous variables.

Abbreviations: 2‐h OGTT, 2‐hour blood glucose level of the oral glucose tolerance test; BMI, body mass index; bmp, beats per min; HbA1c, glycated hemoglobin A1c; HDL‐C, high‐density lipoprotein cholesterol; HR, heart rate; LDL‐C, low‐density lipoprotein cholesterol; SUA, serum uric acid.

3.2. Association between body height and BP

Table 2 displays the associations between BP and body height. In the unadjusted model, short body heights were significantly associated with an increasing trend of SBP and DBP in a dose‐dependent pattern. Adjustment for age, sex, weight, levels of fasting glucose, TG, TC, HDL‐C, LDL‐C, creatinine and HR did not significantly change the associations (P for trend <.001).

TABLE 2.

Associations of body height with blood pressure

	Body height (in quintiles)					P‐value for trend
	Q1 (women <154.1, men <166.1)	Q2 (women 154.1‐158, men 166.1‐170)	Q3 (women <158.1‐160, men 170.1‐172)	Q4 (women 160.1‐164, men 172.1‐176)	Q5 (women >164, men >176)
SBP
Model 1	129.68 ± 18.41	127.21 ± 17.96	124.52 ± 16.03	124.07 ± 15.91	123.45 ± 15.14	<.001
Model 2	126.38 ± 17.04	123.71 ± 16.54	121.58 ± 14.81	122.14 ± 15.18	122.40 ± 14.68	<.001
Model 3	126.38 ± 17.04	123.71 ± 16.54	121.58 ± 14.81	122.14 ± 15.18	122.40 ± 14.68	<.001
DBP
Model 1	79.78 ± 11.15	79.35 ± 11.66	78.06 ± 10.98	79.54 ± 11.34	79.15 ± 11.32	.041
Model 2	79.61 ± 10.12	79.35 ± 11.66	78.06 ± 10.67	78.96 ± 9.27	78.88 ± 11.37	<.001
Model 3	79.61 ± 10.12	79.35 ± 11.66	78.06 ± 10.67	78.96 ± 11.20	78.88 ± 11.24	<.001

Model 1: unadjusted.

Model 2: adjusted for age, sex, and weight.

Model 3: adjusted for age, sex, weight, levels of fasting glucose, total cholesterol, triglyceride, high‐density lipoprotein cholesterol, low‐density lipoprotein cholesterol, creatinine, and heart rate.

Data are presented as mean ± SD.

Abbreviations: DBP, diastolic blood pressure; SBP, systolic blood pressure.

3.3. Association between body height and markers of central and peripheral arterial stiffness

We examined the associations between PWVs and body height in quintiles (Table 3). In an unadjusted model, short body heights were significantly associated with an increasing trend of cfPWV, crPWV, and cdPWV in a dose‐dependent pattern. After adjustment for age, sex, weight, levels of fasting glucose, TG, TC, HDL‐C, LDL‐C, creatinine, and HR, the association with cdPWV was not significant, whereas the associations with cfPWV and crPWV remained significant (P for trend = .007 and .003 respectively).

TABLE 3.

Associations of body height with central and peripheral arterial stiffness

	Body height (in quintiles)					P‐value for trend
	Q1 (women <154.1, men <166.1)	Q2 (women 154.1‐158, men 166.1‐170)	Q3 (women <158.1‐160, men 170.1‐172)	Q4 (women 160.1‐164, men 172.1‐176)	Q5 (women >164, men >176)
cfPWV
Model 1	11.01 ± 2.21	10.74 ± 1.96	10.53 ± 1.71	10.52 ± 1.62	10.47 ± 1.71	.003
Model 2	10.85 ± 2.03	10.54 ± 1.82	10.50 ± 1.58	10.52 ± 1.70	10.42 ± 1.53	.007
Model 3	10.85 ± 1.21	10.54 ± 1.96	10.51 ± 1.81	10.52 ± 1.57	10.42 ± 1.53	<.001
crPWV
Model 1	10.03 ± 1.45	10.23 ± 1.59	10.18 ± 1.58	10.27 ± 1.55	10.33 ± 1.67	<.001
Model 2	10.35 ± 1.43	10.40 ± 1.59	10.39 ± 1.60	10.34 ± 1.52	10.45 ± 1.68	.003
Model 3	10.35 ± 1.43	10.40 ± 1.56	10.39 ± 1.60	10.34 ± 1.52	10.45 ± 1.68	.006
cdPWV
Model 1	9.77 ± 1.56	9.68 ± 1.59	9.60 ± 1.49	9.56 ± 1.40	9.52 ± 1.39	<.001
Model 2	9.83 ± 1.59	9.61 ± 1.49	9.58 ± 1.51	9.56 ± 1.34	9.61 ± 1.38	.425
Model 3	9.83 ± 1.59	9.61 ± 1.49	9.58 ± 1.51	9.56 ± 1.34	9.61 ± 1.38	.234

Model 1: unadjusted.

Model 2: adjusted for age, sex, and weight.

Model 3: adjusted for age, sex, weight, levels of fasting glucose, total cholesterol, triglyceride, high‐density lipoprotein cholesterol, low‐density lipoprotein cholesterol, creatinine, and heart rate.

Abbreviations: cdPWV, carotid artery‐dorsalis pedis pulse wave velocity. PWVs are presented as a mean (standard deviation); cfPWV, carotid‐femoral pulse wave velocity; crPWV, carotid‐radial pulse wave velocity.

3.4. Interaction between body height and BP on central arterial stiffness

We further examined whether cardiovascular risk factors such as BP, levels of lipids, creatinine, glucose, and HR modified the relationship between body height and cfPWV. We grouped the strata factors of SBP into three categories (tertiles): low (≤121.63 ± 16.39 mm Hg), medium (>121.63 ± 16.39 mm Hg, ≤125.52 ± 16.37 mm Hg), and high (>125.52 ± 16.37 mm Hg). We found that SBP significantly interacted with body height (P for interaction = 0.0039), and body height and SBP showed an additive effect on cfPWV (Figure 1). Compared to the group with the tallest stature and lowest SBP, individuals in the group with the shortest stature and highest SBP had nearly a 3 m/s higher cfPWV. There was no significant interaction of body height with the levels of lipids, creatinine, glucose, and HR in relation to the cfPWV. We have also analyzed the interaction between body height and BP on crPWV. However, it is not statistically significant (P for interaction = 0.246; Figure S1).

FIGURE 1.

Combined effect of body height and systolic blood pressure (low, medium, and high levels) on the risk of arterial stiffness. The mean values of cfPWV are presented. The analysis was adjusted for age; sex; weight; levels of fasting glucose, lipids, and creatinine; and heart rate. Abbreviations: cfPWV, carotid‐femoral pulse wave velocity; SBP, systolic blood pressure; Q1, women 164, men >176

In addition, we found a significant interaction between prehypertension status and body height in relation to the cfPWV, after adjusting for sex, age, weight, levels of fasting glucose, lipids, and HR (P for interaction = 0.0024). The associations were stronger in participants with prehypertension than in those with normal BP (Figure 2).

FIGURE 2.

Interaction between body height and prehypertension in relation to cfPWV. The mean values of cfPWV are presented. The analysis was adjusted for age; sex; weight; levels of fasting glucose, lipids, and creatinine; and heart rate. Abbreviations: cfPWV, carotid‐femoral pulse wave velocity; Q1, women 164, men >176

4. DISCUSSION

In this study, we found that short body height was significantly associated with increasing BP and PWV measures for central and peripheral arterial stiffness in Chinese adults. The associations were independent of conventional cardiometabolic risk factors. Moreover, we found that BP modified the relationship between body height and cfPWV.

Our findings that body height was inversely related to BP are consistent with those of several previous studies. ⁷ , ²² In a British birth cohort study, it was also found that short body height was associated with a higher pulse pressure and SBP, independent of numerous potential confounders. ⁷ It has been proposed that adverse conditions that impair prenatal growth might affect adult body height and changes in the structure and function of the developing vasculature later, which would lead to an increased risk of high BP in adult life. ²³ , ²⁴ In addition, shorter stature is also associated with adverse socio‐economic circumstances, which has been related to a higher risk of high BP. ²⁵ , ²⁶

Aortic stiffness is widely recognized as an indicator of early atherosclerosis. ²⁷ , ²⁸ The cfPWV is considered the gold standard for assessing aortic stiffness, and it has been related to cardiovascular complications. ²⁹ , ³⁰ The noninvasive assessment of PWV, as the supreme index of arterial stiffness, is crucially dependent on the measurement of the travel distance of the arterial pulse wave. ³¹ In line with the result of a previous study of a Chinese population, ³² we found a significant inverse association between body height and PWV, which may be partly explained by the faster return of the reflected wave in shorter people. The travel time of the arterial wave and timing of wave reflection depend on transmission path length, which is determined by body height. ³³ Consequently, reflected waves in those with shorter body height are more likely to arrive early, resulting in increased central pressure augmentation.

Intriguingly, we found that the inverse effects of body height on PWV appeared more evident in people with prehypertension than in those with normal BP. Notably, on average, the cfPWV was much higher in individuals with prehypertension than in those with normal BP (Figure 2). Arterial stiffness is strongly associated with BP. ³⁴ , ³⁵ A recent study in European natives demonstrated that the PWV was positively related to central or peripheral systolic BP and confirmed the graded increase in the PWV from normotensive to high‐normal and hypertension. ³⁶ The precise mechanisms underlying such an additive effect between body height and BP remain unclear. Some studies have shown that the PWV reflects functional or even structural changes in the vascular wall structure as part of the evolving hypertension process and is therefore a harbinger of further BP elevation in the future. ³⁶ Elevated BP, particularly SBP, increases pulsatile aortic wall stress, which accelerates elastin degradation. ³⁷ , ³⁸ However, our data also indicate that the associations between body height and cfPWV were independent of BP, suggesting other mechanisms might be also involved to explain the stronger associations in patients with high BP. Short stature has been related to various metabolic changes. We assume that certain metabolic changes may interact with high BP in promoting atherosclerosis. Inflammation‐mediated proatherogenic activation and the regulation of vascular tone and endothelial function are all factors that may lead to a significant increase of arterial stiffness in essentially prehypertensive patients. ³⁹ , ⁴⁰ Our findings might have important clinical implications for the prevention and intervention of cardiovascular risk at an early stage. Further investigations are needed to verify the postulation.

The sample size of the present study was large, which ensures sufficient power to detect the inverse effects of body height on arterial stiffness and interactions with BP. However, several limitations of this study warrant consideration. First, this study was a cross‐sectional analysis; thus, we could not define a causal association between body height and PWVs. Second, several potential confounders such as childhood social class, educational attainment, adult social class, family history of early cardiovascular disease, central aortic pressure levels, and other lifestyle factors in adulthood (tobacco smoking, alcohol consumption, and lack of physical exercise) were unavailable in our cohort. However, these factors more likely attenuated the associations toward null. The ankle‐brachial index and intima‐media thickness are complementary parameters to evaluate the severity of early atherosclerosis. ⁴¹ , ⁴² However, these potential markers were not measured in our study. Finally, as the study was performed in a Chinese population, the results may not be generalizable to other ethnic groups. Further studies in other populations of different ethnicities are warranted to verify our findings.

In summary, we found that short body height was associated with markers of an increased risk of early‐stage atherosclerosis in Chinese adults, independent of other cardiometabolic risk factors. BP might modify these associations.

CONFLICT OF INTEREST

None declared.

AUTHOR CONTRIBUTIONS

Qinqin Qiu contributed to data analysis, interpretation of results, and manuscript writing. Xiangyu Meng and Yanjun Li contributed to data analysis and manuscript writing. Xuekui Liu contributed to statistical analysis of data and interpretation of results. Fei Teng, Yu Wang, and Jun Liang contributed to data analysis and interpretation of results. Xiu Zang contributed to experimentation and data collection. Yun Wang contributed to data collection. Jun Liang also participated in the concept and design of the study, and reviewed and edited the manuscript.

Supporting information

Supplementary Material

Qiu Q, Meng X, Li Y, et al. Evaluation of the associations of body height with blood pressure and early‐stage atherosclerosis in Chinese adults. J Clin Hypertens. 2020;22:1018–1024. 10.1111/jch.13870