Carotid Intima-media Thickness and its Association with Conventional Risk Factors in Low-income Adults: A Population-based Cross-Sectional Study in China

Carotid intima-media thickness (CIMT) is an established predictor of cardiovascular disease and stroke. However, risk factors associated with CIMT remain unclear. Therefore, we aimed to identify factors associated with CIMT in a low-income Chinese population. Stroke-free and cardiovascular disease-free residents aged ≥45 years were recruited. B-mode ultrasonography was performed to measure CIMT. The mean age of participants (n = 3789) was 59.92 years overall, 61.13 years in men, and 59.07 years in women (P < 0.001). Male sex, older age, low education level, smoking, hypertension, and high systolic blood pressure, fasting blood glucose, and low-density lipoprotein cholesterol levels were independent determinants of mean CIMT. Mean CIMT was higher by 18.07 × 10−3 mm in hypertensive compared to normotensive participants (P < 0.001), by 19.03 × 10−3 mm in men compared to women (P < 0.001), and by 9.82 × 10−3 mm in smokers compared to never smokers (P < 0.001). However, mean CIMT decreased by 1.07, 0.37, and 2.36 × 10−3 mm per 1-unit increase in education level, diastolic blood pressure, and triglycerides, respectively. It is important to manage conventional risk factors in low-income populations to decrease stroke incidence.


Discussion
This is the first report on the characteristics and factors associated with CIMT in a low-income population with a high incidence of stroke. Compared with previous reports from non-Asian populations, we found a lower mean CIMT in this population with a high incidence of stroke and high prevalence of conventional risk factors [4][5][6][7][8] .
Positive correlations were observed between mean CIMT and older age, male sex, current smoking, SBP, FBG, and LDL-C, an inverse relationship was found between mean CIMT and education level. However, inconsistent with that seen in previous studies, a negative association between mean CIMT and DBP and TG was observed in this study. The Gutenberg-Heart Study, which was a community-based study, reported median CIMT values of 0.62mm in women and 0.65 mm in men 17 . A significant trend of higher CIMT with increasing age was observed in a community-based study in Taiwan 10 . Additional studies found that mean CIMT was higher in black subjects compared with white subjects aged 35-85 years (0.80 mm vs. 0.75 mm; P < 0.001) 19 and South Asian subjects aged >45 years (0.64 mm vs. 0.61 mm; P = 0.022) 20 . In line with these studies, we found an elevated mean CIMT with increased age. The age-specific mean CIMT was 0.54 mm in those aged 45-54 years, 0.57 mm in those aged 55-64 years, 0.59 mm in those aged 65-74 years, and 0.61 mm in those aged ≥75 years.
A number of studies have confirmed the association between CIMT and the incidence of stroke in Western and Asian populations 9-12,21-25 . Hazard ratios increased from the lowest to the highest tertile of CIMT for every CVD outcome and for all-cause mortality after adjusting for conventional risk factors, carotid plaque groups, and C-reactive protein levels; the highest tertile was associated with an 84% increased risk of composite CVD events, a 54% increased risk of all-cause mortality, and a doubling of risk for CVD-related death 9 . The Rotterdam Study found that for a mean CIMT increase of 0.163 mm, the odds ratios (adjusted for age and sex) were 1.41 (1.25 to 1.82) for stroke and 1.43 (1.16 to 1.78) for myocardial infarction 21 . The Cardiovascular Health Study showed that a maximal CIMT increase of 0.2 mm resulted in age-and sex-adjusted HRs of 1.33 (1.21 to 1.48) for myocardial infarction and 1.37 (1.25 to 1.51) for stroke 22 .
Inconsistent with previous studies, the mean CIMT was lower in this population with a high incidence of stroke, with the quartile of 0.51 mm, 0.55 mm, and 0.61 mm, respectively. The lower level of mean CIMT was in contrast with the higher incidence of stroke and prevalence rates of risk factors, such as hypertension, diabetes, and obesity, in this population [4][5][6][7][8] . The prevalence of hypertension was higher in this population than in Taiwanese; with the prevalence of 57.2% for men and 56.2% for women in those aged 45-54 years in this study 7 , while 28.2% of men and 23.6% of women in previous study in Taiwan 18 . However, the mean CIMT in the same age group was 0.54 mm overall, 0.66 mm in men, and 0.61 mm in women in Taiwan. The disparity of low mean CIMT and high incidence of stroke indicates that it is necessary to further evaluate the mean CIMT values in different ethnic populations and regions.
Low socio-economic status, defined as low education, low income, or employed in a manual occupation, has been shown to be associated with increased CIMT [26][27][28][29][30][31] . Consistent with these findings, mean CIMT decreased by 1.07 × 10 −3 mm per 1-year increase in education level in this study.
The association between conventional risk factors and CIMT has been established in previous studies; hypertension, diabetes mellitus, smoking, and higher levels of SBP, FBG, TG, and LDL-C have been shown to be associated with increased CIMT 18-20,32,33 , though an the inverse relationship was found between CIMT and HDL-C 33,34 .  Similar relationships were found in this study between CIMT and hypertension, current smoking, and higher levels of SBP, FBG, and LDL-C. However, in contrast with previous studies that found a significant correlation between SBP and mean CIMT [35][36][37][38] , a negative correlation was observed between CIMT and DBP and TG in the current study. The exact causes are unknown, and further prospective large population-based studies are needed. There were several limitations in this study. First, the study population was small and from a local town in Tianjin, China, so the findings may not be generalized to the Chinese population. Second, the cross-sectional study design may have led to a selection bias. However, including only stroke-and CVD-free participants may have reduced this bias. Finally, information regarding medication use and additional blood measurements, such as HDL-C compositions, was absent and may have affected the results. However, lower rates of medication use for hypertension (1.5%) and dyslipidaemia (1.02%) in this population may reduce the impact of medication use on the mean CIMT.

Conclusions
This study was the first to report the association between CIMT and relevant risk factors in a low-income population with a high incidence of stroke in China. In this study, we evaluated mean CIMT and relevant risk factors amongparticipants aged 45 years and over. There was a disparity between lower mean CIMT and higher incidence of stroke in this population. Established risk factors, including age, sex, education level, hypertension, smoking, SBP, DBP, FBG, and LDL-C, were significantly associated with mean CIMT; however, we observed a negative correlation between mean CIMT and TG. These findings suggest that there are distinct differences in mean CIMT and relevant risk factors between Chinese and Western populations, especially among those with a low socio-economic status. Therefore, it is crucial to explore the primary determinants of stroke and CVD in low-income populations in China in order to decelerate the dramatic increase in stroke incidence and to reduce the burden of stroke.

Study population.
This was a population-based cross-sectional study conducted from April 2014 to January 2015. The study population was from the Tianjin Brain Study [5][6][7] , and the study design was described in a previous study 16 . In brief, the total population included 14,251 persons distributed within 18 administrative villages. Approximately 95% of participants were low-income farmers, with a per capita disposable income of < $1600 US in 2014 39 . All residents aged 45 years and over without CVD were recruited to this study, while those with a previous history of CVD were excluded.
All investigative protocols were approved by the ethics committee of Tianjin Medical University General Hospital and the ethics committee of Peking University First Hospital; the methods were carried out in accordance with the approved guidelines, and informed consent was obtained from all participants.
Information collection and risk factor definitions. All variables in this study were evaluated by trained epidemiological researchers through face-to-face interviews. A prespecified questionnaire was used to collect all information in this study.
Demographic information, including name, sex, date of birth, and educational level, were obtained from previous records. All participants were categorised into four age groups: 45-54 years, 55-64 years, 65-74 years, and ≥ 75 years. Educational level was categorised into three groups according to educational years: illiteracy (without education), 1-6 years group, and > 6 years group.
Previous individual and family medical histories, which included hypertension, diabetes mellitus, stroke, transient ischemic attack, and coronary heart disease, were obtained according to patient self-reporting or previous records.
Lifestyle characteristics included cigarette smoking, passive smoking, and alcohol consumption. Cigarette smoking was defined as smoking more than 1 cigarette per day for at least 1 year, and participants were categorised as never smokers, ever smokers (ceased smoking for at least 6 months), and current smokers. Alcohol consumption was defined as drinking more than 500 grams of alcohol per week for at least 1 year, and participants were categorised into the never alcohol consumption, ever alcohol consumption (temperance for at least 6 months), and current alcohol consumption groups.
Physical examination. Measurements of blood pressure (including systolic blood pressure [SBP] and diastolic blood pressure [DBP]), height, and weight were performed in the local village clinic during the baseline survey; levels of fasting blood glucose (FBG), total cholesterol (TC), triglycerides (TG), high-density lipoprotein cholesterol (HDL-C), and low-density lipoprotein cholesterol (LDL-C) in serum were measured at the Ji County People's Hospital. Carotid ultrasonography and 12-lead echocardiography were also performed by a professional. Body mass index was calculated as weight (kg) divided by the square of height (m 2 ).
Hypertension was defined as SBP ≥ 140 mmHg or DBP ≥ 90 mmHg or taking medications for hypertension. Diabetes was defined as FBG ≥ 7.0 mmol/L or taking insulin or oral hypoglycaemic medications.
Ultrasonography measurements. One trained technician blinded to individuals' previous disease histories performed all the ultrasound exams. The patients were examined in the supine position using B mode ultrasonography (Terason 3000; Burlington, MA, US) with a 5-12 MHz linear array transducer. CIMTs at the near and far walls of the common carotid artery were measured on the left and right, and 3 values were obtained: the maximum CIMT, minimum CIMT, and average CIMT. Images were obtained and digitally stored according to a standard protocol.
Scientific RepoRts | 7:41500 | DOI: 10.1038/srep41500 Statistical analyses. Continuous variables are presented as means with standard deviations, and were compared between groups using the Student's t-test for two groups and analysis of variance for more than two groups. Categorical variables are presented as numbers with frequencies, and were compared using chi-square tests. Multiple linear regression analyses were used to evaluate the associations between traditional risk factors and CIMT. We performed linear regression analyses to evaluate the determinants of CIMT after adjusting for age, educational levels, SBP, DBP, FBG, TC, TG, HDL-C, and LDL-C (continuous variables), and sex, smoking status, alcohol consumption, hypertension, and diabetes (categorical variables). The relationships are presented as unstandardised regression coefficients (β 1 ) with standard errors and standardized regression coefficients (β 2 ). P values < 0.05 in two-tailed tests were considered statistically significant. SPSS for Windows (version 19.0; SPSS Inc., Chicago, IL, USA) was used for analyses.