Prevalence of carotid atherosclerosis and carotid plaque in Chinese adults: A systematic review and meta-regression analysis

Prevalence of carotid atherosclerosis and carotid plaque in Chinese adults: A systematic review and meta-regression analysis

Atherosclerosis 276 (2018) 67e73 Contents lists available at ScienceDirect Atherosclerosis journal homepage: www.elsevier.com/locate/atherosclerosis...

937KB Sizes 0 Downloads 33 Views

Atherosclerosis 276 (2018) 67e73

Contents lists available at ScienceDirect

Atherosclerosis journal homepage: www.elsevier.com/locate/atherosclerosis

Prevalence of carotid atherosclerosis and carotid plaque in Chinese adults: A systematic review and meta-regression analysis Peige Song a, b, Wei Xia c, Yajie Zhu d, Manli Wang a, Xinlei Chang a, Shuai Jin e, Jingpin Wang e, Lin An a, * a

Department of Maternal and Child Health, School of Public Health, Peking University, Beijing, China Centre for Global Health Research, University of Edinburgh, Edinburgh, Scotland, United Kingdom School of Nursing, University of Hong Kong, Hong Kong, China d George Institute for Global Health, University of Oxford, Oxford, United Kingdom e School of Nursing, Peking University, Beijing, China b c

a r t i c l e i n f o

a b s t r a c t

Article history: Received 19 April 2018 Received in revised form 10 July 2018 Accepted 13 July 2018

Background and aims: The national representative prevalence of carotid atherosclerosis (CAS) or carotid plaque (CP) in the general Chinese population has never been estimated. We aim to generate the prevalence and number of people with CAS and CP in the general Chinese population. Methods: We searched China National Knowledge Infrastructure, Wanfang Data, VIP Database for Chinese Technical Periodical, PubMed, Embase and Medline. Articles reporting the prevalence of CAS or CP in the general Chinese population were included. A multilevel mixed-effects logistic regression was used to estimate the age- and gender-specific prevalence of CAS and CP. The effects of risk factors for CAS were assessed by a random-effects meta-analysis. Results: The prevalence of CAS and CP increased with advanced age. Males had a higher prevalence of CAS and CP than females consistently across all age groups. Overall, 27.22% and 20.15% of Chinese people aged 30e79 years were with CAS and CP, respectively, in 2010, equivalent to 207.73 million and 153.82 million affected individuals. With demographic ageing, the number of people affected by CAS and CP will increase to 267.25 million and 199.83 million, respectively, by 2020. In addition, current smoking, hypertension and diabetes were found to be risk factors for CAS. More than 70% of the national CAS cases were in rural China in 2010. Conclusions: CAS and CP are highly prevalent in China. The huge disease burden of CAS and CP calls for efforts on effective preventive health strategies and early-detection of CVDs in people with CAS or CP, especially in rural areas. © 2018 Elsevier B.V. All rights reserved.

Keywords: China Carotid atherosclerosis Carotid plaque Prevalence

1. Introduction Cardiovascular diseases (CVDs), including cerebrovascular disease (stroke), coronary heart disease (CHD) and ischemic brain attack, are the most prevalent cause of mortality and morbidity worldwide [1,2]. In 2015, CVDs affected 422.7 million people and accounted for 17.9 million deaths globally [2]. Disproportionally, more than 80% of CVD deaths occur in low and middle-income countries (LMICs) [3,4]. In China, the largest LMIC in the world, CVDs are also a leading cause of death and represent a severe public

* Corresponding author. Department of Maternal and Child Health, School of Public Health, Peking University, 38 Xueyuan Road, Beijing, 100191, China. E-mail address: [email protected] (L. An). https://doi.org/10.1016/j.atherosclerosis.2018.07.020 0021-9150/© 2018 Elsevier B.V. All rights reserved.

health problem. With rapid urbanization and population ageing, it is expected that CVD morbidity and mortality will keep rising in the next decade in China [5,6]. Atherosclerosis, a generalized disease characterized by the accumulation of lipids and fibrous elements in the large arteries, is worldwide an important cause of CVDs [1,7,8]. Atherosclerosis mainly occurs at the carotid bifurcation and the proximal internal carotid artery [9]. Carotid intima-media thickness (cIMT), the distance from the lumen-intima interface to the media-adventitia interface of the artery wall, has been suggested to be independently associated with CVD [8,10e13]. By using non-invasive imaging, cIMT could be simply measured and a value of 1.0 mm is generally suggested as abnormal [14e16]. Carotid plaque (CP) refers to a focal cIMT1.5 mm protruding into the lumen or the presence

68

P. Song et al. / Atherosclerosis 276 (2018) 67e73

of focal wall thickening that is  0.5 mm or 50% than the surrounding vessel wall [14,17,18]. In population-based studies, as well as in clinical practice, cIMT has been widely adopted as a surrogate marker for the presence of carotid atherosclerosis (CAS) or CP [10,11,14,15,19]. In 2013, the 66th World Health Assembly adopted the Global action plan for the prevention and control of noncommunicable diseases 2013e2020, in which a target of reduction of premature mortality from CVDs worldwide by 25% by 2025 was set [3,20,21]. Despite the fact that there have already been abundant research and data on CVDs, CAS and CP are still relatively neglected in terms of research priority [22,23]. In addition, due to the lack of a standardized definition and measurement of cIMT, as well as the large disparity of study design and study population between studies, the reported prevalence of CAS and CP varied considerably [10,23,24]. Furthermore, those individual epidemiological investigations are generally small in sample size or limited to a specific geographic area. Therefore, the generality of those results is largely restricted. In China, where an upward trend of CVD morbidity and mortality is very likely in the foreseen future, a detailed understanding of the epidemiology of CAS, the major cause of CVDs, has become imperative. Data from the China National Stroke Prevention Project (CSPP) showed that the standardized prevalence of CAS among asymptomatic Chinese aged 40 years was 36.2% in 2010. However, this prevalence estimate might be relatively lower than that in the general population where asymptomatic and symptomatic people should simultaneously exist [25]. Furthermore, the temporal trend of CAS prevalence has never been revealed in previous investigations. To fill this gap of knowledge, in this study, we set out to determine the prevalence of ultrasonography-based CAS and CP in general Chinese population by using a systematic review and meta-analysis, and to estimate and project the number of people with CAS or CP in China from 2000 to 2020. The major risk factors for CAS and subnational-level number of CAS cases were also investigated. 2. Materials and methods

reported the numerical estimates of CAS or CP prevalence in general Chinese population. Studies that were hospital- or health care facility-based were excluded because an overestimation was likely. Similarly, studies that were in a sample with potentially higher risks of CVDs were also excluded. In the included studies, the assessment of CAS or CP should be based on ultrasound imaging. Furthermore, to ensure our ability of synthesizing prevalence data from different studies, the definitions of CAS and CP should be consistent and accord with the 2016 European Guidelines on cardiovascular disease prevention in clinical practice and the Mannheim Carotid Intima-Media Thickness and Plaque Consensus respectively, where CAS was defined as a cIMT 1.0 mm and CP as 1) a focal structure that encroaches into the arterial lumen of 0.5 mm or 50% than the surrounding wall; or 2) a cIMT 1.5 mm [14,18]. For multiple publications that were based on the same single study, the most recent one or the one with the most comprehensive results or the largest sample size was included. 2.2. Data extraction From the included studies, two authors (M. Wang and X. Chang) independently extracted relevant information, which included study characteristics (authors, publication year, investigation year, study setting, sampling method, case assessment and definition), characteristics of the investigated sample (sample size, gender and age) and prevalence estimates (CAS/CP cases). Where possible, the prevalence estimates should be extracted by age group, gender and setting. Any discrepancies over the systematic review or extracted data were resolved by consensus through group discussion. For studies where censoring age groups were reported (e.g., younger than 40 years or older than 60 years), we imputed the missing age band by taking the same width as other complete age groups in the same study. The midpoint of the age range reported or the reported mean or median age was adopted for analysis. For studies where the year of investigation was not reported, we imputed the investigation year by subtracting three years from the publication year. This was done for six studies based on the average time lag between investigation and publication in the other 31 studies with available information (Supplementary Table 2).

2.1. Search strategy and selection criteria This systematic review and meta-analysis is in accordance with the Preferred Reporting Items for Systematic reviews and MetaAnalysis (PRISMA) guidelines [26]. The PRISMA checklist for this study is available in Supplementary Data. Two authors (S. Jin and J. Wang) independently performed a comprehensive systematic review of literature that was published from 1st January 1990 to 15th April 2017, and reported the prevalence of CAS or CP in China. The period from 1990 onwards was chosen to capture literature in the last three decades, as well as to mask a rather recent prevalence of CAS or CP in China. The literature search was limited to three Chinese and three English bibliographic databases, namely, China National Knowledge Infrastructure (CNKI), Wanfang Data, VIP Database for Chinese Technical Periodical (VIP), PubMed, Embase and Medline. No language restrictions were placed. The search terms included “prevalence” or “rate” or “epidemiology”, in combination with “carotid atherosclerosis or carotid plaque” and “China or Chinese”, in the forms of controlled vocabulary (i.e., medical subject headings) or free words. The detailed search strategies customized for each database are listed in Supplementary Table 1. Reference lists in potentially relevant articles were hand-searched to locate any additional studies. All records were independently screened by two authors (S. Jin and J. Wang) for relevance. To be included in this systematic review and meta-analysis, studies had to be population-based that

2.3. Statistical analysis In this study, the prevalence rates of CAS and CP were estimated separately. In the data extraction process, age-, gender- or settingstratified prevalence estimates were generally available in a single study, to take this hierarchical data structure into account, a multilevel mixed-effects meta-regression was adopted [27,28]. The variance of prevalence rates was stabilized by the logit transformation [29]. Given that:

prevalence ¼ p ¼ ðCAS=CP casesÞ=ðnumber of participantsÞ Then, the prevalence estimates were stabilized by using the logit link,

logitðpÞ ¼ lnðp=½1  pÞ ¼ lnðoddsÞ ¼ a þ b1 *x1 þ b2 *x2 þ … To assess the associations of cluster-level variables (age, gender, setting and investigation year) and the prevalence of CAS/CP, a univariable meta-regression was conducted, followed by a multivariable meta-regression (Supplementary Table 3). Age and investigation year were treated as continuous variables, whereas gender and setting were as categorical variables. The study identification number was used as a random effect (ui ). Ultimately, age and gender were found to be significantly associated with the prevalence of CAS/CP. Therefore,

P. Song et al. / Atherosclerosis 276 (2018) 67e73

logitðpÞ ¼ a þ b1 *Age þ b2 *Gender þ ui then,

prevalence ¼ p ¼ eðaþb1 *Ageþb2 *Genderþui Þ

.  1 þ eðaþb1 *Ageþb2 *Genderþui Þ

By using the final model, the age- and gender-specific prevalence of CAS/CP was estimated. To generate the national number of people with CAS/CP, the estimated age- and gender-specific prevalence rates were applied to the corresponding age- and gender-specific demographic data, which were obtained from the United Nations Population Division (UNPD) [30]. The midpoint of each age category was adopted. This was done for an age range of 30e79 years and the years 2000, 2010 and 2020 respectively. A total of seven studies additionally provided information on risk factors for CAS, based on multivariate logistic regression. Because of the diversity of definitions, only four risk factors, including male gender, current smoking, hypertension and diabetes, were with at least three informative data points and included for subsequent random-effects (DerSimonian and Laird method) meta-analyses [27]. The prevalence of those four risk factors in

69

China was then obtained from the 2010 China Non-communicable and Chronic Disease surveillance (Supplementary Table 4). Based on the distribution of those four risk factors in China, the national number of people with CAS was distributed into different regions (East, Central and West) and settings (urban and rural) in 2010 [27,31e33]. For CP, the meta-analysis of major risk factors and subnational estimation of cases were not conducted due to the limited availability of relevant data. Statistical analyses were performed using STATA version 14.0 (STATA Corporation, College Station, Texas, USA) and R version 3.3.0 (R Foundation for Statistical Computing, Vienna, Austria). The map was produced using ArcMap version 10.1 (Environmental Systems Research Institute, Redlands, CA). A p value of less than 0.05 was regarded to be statistically significant. 3. Results 3.1. Description of the included studies A total of 2124 records were identified from our initial search. After exclusion of 822 duplicates, 1302 unique records were retained for title and abstract screening. Then, 400 records were reviewed for eligibility in full-text. Finally, 37 articles met the inclusion criteria, of which 26 articles provided data on CAS

Fig. 1. PRISMA flowchart of study selection process.

70

P. Song et al. / Atherosclerosis 276 (2018) 67e73

Fig. 2. Estimated age- and gender-specific prevalence of CAS and CP in China, with 95% confidence interval.

prevalence and 20 on CP prevalence. Briefly, the 26 studies on CAS included a total of 67,316 participants (31,393 CAS cases) and the 20 studies on CP included 61,203 participants (20714 CP cases). The full list of the 37 included studies is shown in Supplementary Table 5. The PRISMA flowchart showing the study selection process is in Fig. 1. All the 37 studies, published from 2004 to 2017, were all cross-sectional in design. The geographic distribution of the included studies is shown in Supplementary Fig.1. The detailed characteristics of the included studies can be found in Supplementary Table 6.

3.2. Age- and gender-specific prevalence of CAS and CP The final multivariable meta-regression models for estimating the prevalence of CAS and CP are shown below:

lnðpCAS =½1  pCAS Þ ¼ 6:171 þ 0:092*Age þ 0:897*GenderMale þ ui lnðpCP =½1  pCP Þ ¼ 6:134 þ 0:087*Age þ 0:423*GenderMale þ ui

p is the prevalence of CAS/CP GenderMale ¼ 1 for males and 0 for females ui ¼ variance of the study-level random effect Based on the final models, the age-and gender-specific prevalence of CAS and CP was generated. As shown in Fig. 2 and Table 1, the prevalence of CAS and CP was higher in males than in females and both increased with advanced age. In males, the prevalence of CAS increased from 8.89% (95% CI: 5.64e13.73) in individuals aged 30e34 years to 86.01% (95% CI: 79.44e90.73) in those aged 75e79 years. In females, the prevalence of CAS ranged from 3.83% (95% CI: 2.38e6.10) in the 30e34 years age group to 71.49% (95% CI: 61.18e79.96) in the 75e79 years age group. For CP, the prevalence rates ranged from 5.11% (95% CI: 3.52e7.37) in males aged 30e34 years to 73.13% (95% CI: 64.94e79.99) in males aged 75e79 years. The prevalence of CP in females was lower than that in males, which ranged from 3.41% (95% CI: 2.33e4.95) in those aged 30e34 years to 64.06% (95% CI: 54.80e72.37) in those aged 75e79 years. 3.3. The national number of people with CAS and CP, 2000e2020 By applying the estimated prevalence of CAS and CP to the

where: Table 1 Gender-specific prevalence of CAS and CP in China, by age group. Age group

30e34 35e39 40e44 45e49 50e54 55e59 60e64 65e69 70e74 75e79

years years years years years years years years years years

Prevalence of CAS (%, 95% CI)

Prevalence of CP (%, 95% CI)

Male

Female

Male

Female

8.89 (5.64e13.73) 13.39 (8.75e19.95) 19.68 (13.31e28.11) 27.97 (19.69e38.07) 38.09 (28.09e49.21) 49.37 (38.31e60.49) 60.71 (49.61e70.81) 71.00 (60.89e79.39) 79.51 (71.06e85.98) 86.01 (79.44e90.73)

3.83 (2.38e6.10) 5.93 (3.76e9.23) 9.08 (5.89e13.76) 13.67 (9.09e20.05) 20.06 (13.74e28.33) 28.45 (20.20e38.44) 38.66 (28.64e49.73) 49.96 (38.83e61.10) 61.28 (50.04e71.43) 71.49 (61.18e79.96)

5.11 (3.52e7.37) 7.69 (5.37e10.89) 11.41 (8.11e15.83) 16.61 (12.04e22.46) 23.54 (17.50e30.89) 32.26 (24.71e40.85) 42.40 (33.65e51.67) 53.23 (43.89e62.36) 63.77 (54.64e72.00) 73.13 (64.94e79.99)

3.41 (2.33e4.95) 5.17 (3.58e7.41) 7.78 (5.46e10.96) 11.54 (8.23e15.95) 16.78 (12.20e22.64) 23.77 (17.69e31.15) 32.53 (24.92e41.18) 42.71 (33.86e52.04) 53.54 (44.09e62.75) 64.06 (54.80e72.37)

CP

þ1.28 4.23 þ13.02 þ35.23 þ99.28 þ114.62 þ83.25 þ109.13 þ82.25 þ73.36 þ72.19 þ1.59 4.16 þ12.97 þ35.18 þ99.05 þ113.82 þ82.68 þ108.92 þ82.11 þ73.77 þ70.07 5.53 (3.80e8.00) 6.56 (4.57e9.34) 9.20 (6.50e12.84) 16.62 (11.97e22.67) 24.77 (18.25e32.88) 28.03 (21.22e36.01) 28.77 (22.49e35.64) 35.28 (28.59e42.08) 26.59 (22.36e30.56) 18.48 (16.12e20.54) 199.83 (155.87e250.55)

2020 2010

4.14 (2.85e5.99) 7.73 (5.38e11.00) 12.09 (8.55e16.88) 14.66 (10.56e20.00) 16.60 (12.23e22.03) 23.36 (17.68e30.01) 21.34 (16.69e26.44) 19.44 (15.76e23.17) 19.07 (16.05e21.90) 15.39 (13.42e17.11) 153.82 (119.16e194.53) 5.46 (3.75e7.90) 6.85 (4.77e9.75) 8.14 (5.76e11.37) 12.29 (8.86e16.77) 12.43 (9.16e16.49) 13.06 (9.89e16.78) 15.70 (12.28e19.44) 16.87 (13.67e20.11) 14.59 (12.27e16.76) 10.66 (9.28e11.85) 116.05 (89.70e147.22)

2000

8.31 (5.24e12.96) 9.91 (6.42e14.96) 13.85 (9.25e20.14) 24.67 (17.06e34.41) 35.79 (25.76e47.70) 39.00 (29.34e49.55) 38.18 (30.08e46.30) 44.48 (36.65e51.66) 31.87 (27.38e35.67) 21.20 (18.89e23.00) 267.25 (206.08e336.35)

2020 2010

6.21 (3.92e9.69) 11.66 (7.56e17.61) 18.17 (12.13e26.44) 21.77 (15.05e30.36) 24.01 (17.28e31.98) 32.51 (24.46e41.30) 28.33 (22.32e34.35) 24.54 (20.25e28.47) 22.89 (19.69e25.59) 17.64 (15.72e19.15) 207.73 (158.38e264.95)

2000

8.18 (5.16e12.76) 10.34 (6.70e15.61) 12.26 (8.19e17.84) 18.25 (12.62e25.46) 17.98 (12.94e23.95) 18.24 (13.73e23.14) 20.90 (16.49e25.30) 21.29 (17.57e24.71) 17.50 (15.04e19.57) 12.20 (10.85e13.26) 157.14 (119.3e201.61) 30e34 years 35e39 years 40e44 years 45e49 years 50e54 years 55e59 years 60e64 years 65e69 years 70e74 years 75e79 years Total (30e79 years)

People living with CP (million)

This study, benefiting from a comprehensive search strategy and strict eligibility criteria, has allowed us to estimate the prevalence of CAS and CP in general Chinese population in a metaregression manner. To the best of our knowledge, this is the first systematic analysis to provide the prevalence estimates of CAS and CP in China based on published studies. The large geographic scale of the included study sites ensured the representativeness of our results at the national level. As revealed in this study, the prevalence of CAS and CP increased steadily with advanced age for both genders, but males consistently had a higher prevalence of CAS and CP than females. We have also estimated and projected the number of CAS and CP cases at the national level from 2000 to 2020. Findings from this study lead us to conclude that CAS and CP are considerable health problems in China, especially in rural areas. To ensure our ability to synthesize multiple data sources, the definitions of CAS and CP were “standardized” in this study. CAS was defined as a cIMT of 1.0 mm or greater. The overall CAS prevalence was revealed as 27.22% in 2010, which was much lower than that reported in the CSPP study (36.2%) [25]. Although the participants in the CSPP study were without carotid endarterectomy, carotid stenting, stroke or coronary heart disease, their rather older age structure (40 years) might contribute to a

People living with CAS (million)

4. Discussion

Age group

The effects of four risk factors on CAS were evaluated by using random-effects meta-analysis (Supplementary Table 7). Compared to females, males had a higher risk of developing CAS, with a meta-OR of 1.51 (95% CI: 1.34e1.69). In addition, hypertension, diabetes and current smoking were all significant risk factors for CAS, with meta-ORs of 1.82 (95% CI: 1.64e2.02), 1.66 (95% CI: 1.41e1.96) and 1.62 (95% CI: 1.40e1.87) respectively. According to different distributions of above-mentioned risk factors in different regions and settings, the national number of people with CAS in 2010 was distributed at the subnational level (Table 3 and Supplementary Fig. 2). The distribution of CAS cases across China was substantially uneven, with the majority (>70%) in rural areas. Among the three regions, the East owed the most CAS cases (88.66 million [95% CI: 67.55e113.15]) while the West the least (52.38 million [95% CI: 40.00e66.73]).

Table 2 Estimated number of people living with CAS and CP in China in the years 2000, 2010 and 2020, and the rate of change from 2000 to 2020, by age group.

3.4. Risk factors for CAS and subnational number of CAS cases in 2010

Rate of change (%, 2000e2020)

national demographic data, the estimated numbers of people with CAS were 157.14 million (95% CI: 119.3e201.61) in 2000, 207.73 million (95% CI: 158.38e264.95) in 2010 and 267.25 million (95% CI:206.08e336.35) in 2020, indicating a relative increase rate of 70% from 2000 to 2020. For CP, the total number of affected people is expected to increase by 72% within the same time frame - from 116.05 million (95% CI:89.70e147.22) in 2000 to 199.83 million (95% CI:155.87e250.55) in 2020. The great increases in the number of people living with CAS/CP will occur in the 55e59 years and 65e69 years age groups, where the relative rates of change are projected to be more than 100%. By 2020, 75% of the individuals with CAS (or with cIMT 1.0 mm) will have some degree of CP (Table 2). At the national level, the overall prevalence of CAS in people aged 30e79 years was 24.93% (95% CI: 18.93e31.99) in 2000 and 27.22% (95% CI: 20.75e34.71) in 2010, and is projected to increase to 30.07% (95% CI: 23.19e37.85) in 2020. The overall prevalence of CP in those aged 30e79 years was 18.41% (95% CI: 14.23e23.36) and 20.15% (95% CI: 15.61e25.49) in the years 2000 and 2010 respectively, and will be 22.49% (95% CI: 17.54e28.19) in 2020.

71

CAS

P. Song et al. / Atherosclerosis 276 (2018) 67e73

1.54 (0.97e2.40) 3.19 (2.06e4.81) 4.64 (3.10e6.75) 5.28 (3.65e7.37) 5.51 (3.97e7.34) 7.91 (5.95e10.06) 7.12 (5.61e8.64) 6.79 (5.60e7.88) 6.13 (5.28e6.86) 4.28 (3.81e4.64) 52.38 (40.00e66.73) 2.79 (1.76e4.35) 4.78 (3.10e7.21) 7.62 (5.09e11.09) 9.39 (6.49e13.09) 10.96 (7.89e14.59) 14.11 (10.62e17.92) 11.90 (9.38e14.42) 9.81 (8.10e11.38) 9.43 (8.11e10.54) 7.87 (7.01e8.55) 88.66 (67.55e113.15) 1.89 (1.19e2.94) 3.67 (2.38e5.54) 5.90 (3.94e8.59) 7.11 (4.91e9.92) 7.59 (5.46e10.11) 10.51 (7.91e13.35) 9.31 (7.34e11.30) 7.89 (6.51e9.16) 7.31 (6.28e8.17) 5.51 (4.91e5.98) 66.69 (50.83e85.06) 1.15 (0.73e1.79) 2.40 (1.56e3.63) 3.57 (2.39e5.20) 4.06 (2.81e5.66) 4.27 (3.07e5.68) 6.36 (4.78e8.08) 5.79 (4.56e7.02) 5.55 (4.58e6.44) 4.90 (4.22e5.48) 3.43 (3.06e3.73) 41.49 (31.75e52.72) 1.41 (0.89e2.21) 2.59 (1.68e3.91) 4.59 (3.06e6.67) 5.67 (3.92e7.91) 6.71 (4.83e8.94) 9.06 (6.82e11.50) 7.82 (6.17e9.48) 6.53 (5.39e7.57) 6.15 (5.30e6.87) 5.16 (4.60e5.61) 55.69 (42.66e70.65) 1.37 (0.86e2.13) 2.71 (1.76e4.09) 4.47 (2.98e6.50) 5.32 (3.67e7.42) 5.67 (4.08e7.56) 8.18 (6.16e10.40) 7.37 (5.81e8.94) 6.25 (5.16e7.25) 5.68 (4.89e6.35) 4.35 (3.87e4.72) 51.36 (39.24e65.35) 0.39 (0.24e0.60) 0.78 (0.51e1.18) 1.06 (0.71e1.55) 1.22 (0.85e1.70) 1.24 (0.89e1.65) 1.55 (1.17e1.98) 1.33 (1.05e1.62) 1.24 (1.02e1.44) 1.23 (1.06e1.37) 0.84 (0.75e0.91) 10.89 (8.25e14.01)

East East

1.37 (0.87e2.14) 2.19 (1.42e3.31) 3.04 (2.03e4.41) 3.72 (2.58e5.19) 4.25 (3.06e5.66) 5.05 (3.80e6.42) 4.07 (3.21e4.94) 3.29 (2.71e3.82) 3.27 (2.81e3.66) 2.71 (2.41e2.94) 32.97 (24.89e42.50)

Overall

Central West East

Rural

Central Central

West Urban Age group

Table 3 Estimated age-specific number of people living with CAS in China in 2010, by region and setting.

0.52 (0.33e0.81) 0.96 (0.62e1.45) 1.44 (0.96e2.09) 1.79 (1.24e2.50) 1.92 (1.38e2.55) 2.33 (1.75e2.96) 1.94 (1.53e2.36) 1.64 (1.35e1.91) 1.63 (1.40e1.82) 1.16 (1.04e1.26) 15.33 (11.59e19.71)

West

P. Song et al. / Atherosclerosis 276 (2018) 67e73

30e34 years 35e39 years 40e44 years 45e49 years 50e54 years 55e59 years 60e64 years 65e69 years 70e74 years 75e79 years Total (30e79 years)

72

higher prevalence. In accordance with the Mannheim Carotid Intima-Media Thickness and Plaque Consensus, CP was defined as a focal structure that encroaches into the arterial lumen of 0.5 mm or 50% than the surrounding wall or a cIMT of 1.5 mm or greater [18]. At the national level, the overall prevalence of CP was estimated to be 20.15% in 2010, which was much higher than that in the asymptomatic population as reported in the CSPP study (13.9%). In comparison with the prevalence of CP presented in China Kadoorie Biobank (CKB), a large cross-sectional investigation covering multiple geographic locations across China, our prevalence estimate of CP was relatively low (20.15% vs. 30.9%) [19]. Given that the prevalence of CP increases with advanced age, one explanation for this difference might be a younger age structure in our study than that in the CKB study (30e79 years vs. 40e89 years). In line with many previous studies, this study confirms that CAS and CP are highly age-related [19,22,25]. This vascular ageing is an adaptive process and has been suggested to be accelerated by cardiovascular risk factors [7,34]. Largely due to the demographic ageing in China, the number of people affected by CAS and CP is expected to increase by more than 70% by 2020, when around 267 million people will be affected by CAS and 200 million will have CP. This increasing threat to the whole society and health care system should draw more attention, more preventative strategies and implementations should be taken in the foreseen future by stakeholders. In addition, the gender difference of CAS and CP prevalence has also been confirmed in this study, with males having a higher risk of CAS and CP than females. This pattern persisted across all age ranges and has been suggested to be related with gender differences of exposure to risk factors for CAS and CP, and the physiological and anatomical differences between genders might also contribute [22,25,35,36]. Another key feature of this study was the meta-analysis of major risk factors for CAS. Except for advanced age and male gender, current smoking, hypertension and diabetes have all been additionally recognized as risk factors for CAS in our study. The effects of those factors have been widely explored in previous investigations [37e39]. Our study demonstrated the first effort in exploring major risk factors for CAS in the Chinese population, however, due to the diversity of definitions of risk factors across studies, only four variable (male gender, current smoking, hypertension and diabetes) could be included in our meta-analysis. With new data coming in the foreseen future, the effects of other factors, such as obesity and dyslipidemia, should also be assessed, and the meta-analysis of risk factors for CAS should be updated on a regular basis. In this study, we distributed the national number of people with CAS to the subnational level, based on the distribution of major factors (age, gender, current smoking, hypertension and diabetes) in different regions and settings across China. Noteworthy, the CAS cases were concentrated in rural areas (>70%) in 2010. This urban-rural gap, as a combined result of different demographic structures and uneven exposures to risk factors, suggests that more efforts to combat the epidemic of CVDs are especially needed in rural settings. We acknowledge that our study is not free from limitations. First, a considerable heterogeneity existed in the integration of included studies, despite our efforts of applying strict inclusion criteria and pre-setting “standardized” case definitions. Second, many potential correlates of CAS or CP, e.g., dyslipidaemia, were unable to be extracted from the included studies, partly due to limited data sources and variable definitions of correlates. Third, given we adopted the multilevel mixed-effects meta-regression to generate prevalence estimation, the risk of bias across and within studies was not assessed. Fourth, our national estimation of CAS

P. Song et al. / Atherosclerosis 276 (2018) 67e73

and CP cases was based on the assumption that age- and genderspecific prevalence rates would remain the same across 2000 to 2010, which might overestimate or underestimate the overall CAS and CP cases to some extent. Moreover, our subnational estimation of CAS cases was actually driven by the difference of age and gender structure and distribution of current smoking, hypertension and diabetes and their effects on CAS, which cannot fully reflect the reality. In conclusion, the prevalence of CAS was 27.22% and that of CP was 20.15% in Chinese people aged 30e79 years in 2010, translating to 207.73 million and 153.82 million affected individuals respectively. With demographic ageing, the numbers of people affected by CAS and CP will increase to 267.25 million and 199.83 million by 2020. Advanced age and male gender were associated with higher risks of CAS and CP. In addition, current smoking, hypertension and diabetes were found to be risk factors for CAS. More than 70% of the CAS cases were in rural China in 2010. More studies are still needed to explore the risk factors of CAS and CP. The huge burden of CAS and CP in China calls for efforts on effective preventative health strategies and early-detection of CVDs. Conflicts of interest The authors declared they do not have anything to disclose regarding conflict of interest with respect to this manuscript. Author contributions PS, LA: study conception and design; SJ, JW: literature search; MW, XC: data extraction; PS, WX, YZ: data analysis and interpretation of data; PS: first draft. All authors provided critical revision of the article and final approval of the version to publish. Appendix A. Supplementary data Supplementary data related to this article can be found at https://doi.org/10.1016/j.atherosclerosis.2018.07.020. References [1] S. Mendis, et al., Global Atlas on Cardiovascular Disease Prevention and Control, World Health Organization, Geneva, 2011. [2] G.A. Roth, et al., Global, regional, and national burden of cardiovascular diseases for 10 causes, 1990 to 2015, J. Am. Coll. Cardiol. 70 (1) (2017) 1e25. [3] Organization, W.H, Global Status Report on Noncommunicable Diseases 2014, World Health Organization, Geneva, 2014. [4] T.A. Gaziano, et al., Growing epidemic of coronary heart disease in low-and middle-income countries, Curr. Probl. Cardiol. 35 (2) (2010) 72e115. [5] C. Weiwei, et al., Outline of the report on cardiovascular diseases in China, 2014, Eur. Heart J. Suppl. 18 (suppl_F) (2016) F2eF11. [6] C. Wei-Wei, et al., China cardiovascular diseases report 2015: a summary, J. Geriatr. Cardiol.: JGC 14 (1) (2017) 1. [7] A.J. Lusis, Atherosclerosis. Nat 407 (6801) (2000) 233e241. [8] A. Kitamura, et al., Carotid intima-media thickness and plaque characteristics as a risk factor for stroke in Japanese elderly men, Stroke 35 (12) (2004) 2788e2794. [9] T.F. Whayne Jr., Prevention of carotid artery atherosclerosis: what is the evidence? Angiology 68 (8) (2017) 661e668. [10] M.W. Lorenz, et al., Prediction of clinical cardiovascular events with carotid intima-media thickness: a systematic review and meta-analysis, Circulation 115 (4) (2007) 459e467. [11] M.W. Lorenz, et al., Carotid intima-media thickening indicates a higher vascular risk across a wide age range: prospective data from the Carotid Atherosclerosis Progression Study (CAPS), Stroke 37 (1) (2006) 87e92.

73

[12] L.E. Chambless, et al., Carotid wall thickness is predictive of incident clinical stroke: the Atherosclerosis Risk in Communities (ARIC) study, Am. J. Epidemiol. 151 (5) (2000) 478e487. [13] M. Hollander, et al., Comparison between measures of atherosclerosis and risk of stroke: the Rotterdam Study, Stroke 34 (10) (2003) 2367e2372. [14] F. Hobbs, et al., 2016 European Guidelines on cardiovascular disease prevention in clinical practice, Eur. Heart J. 37 (29) (2016) 2315e2381. [15] N. Katsiki, et al., Hyperuricaemia and non-alcoholic fatty liver disease (NAFLD): a relationship with implications for vascular risk? Curr. Vasc. Pharmacol. 9 (6) (2011) 698e705. [16] H.-T. Nguyen-Thanh, B.S. Benzaquen, Screening for subclinical coronary artery disease measuring carotid intima media thickness, Am. J. Cardiol. 104 (10) (2009) 1383e1388. [17] J.H. Stein, et al., Use of carotid ultrasound to identify subclinical vascular disease and evaluate cardiovascular disease risk: a consensus statement from the American Society of Echocardiography Carotid Intima-Media Thickness Task Force endorsed by the Society for Vascular Medicine, J. Am. Soc. Echocardiogr. 21 (2) (2008) 93e111. [18] P.-J. Touboul, et al., Mannheim carotid intima-media thickness and plaque consensus (2004e2006e2011), Cerebrovasc. Dis. 34 (4) (2012) 290e296. [19] R. Clarke, et al., Burden of carotid artery atherosclerosis in Chinese adults: implications for future risk of cardiovascular diseases, Eur. J. Prev. Cardiol. 24 (6) (2017) 647e656. [20] Organization, W.H, Global Action Plan for the Prevention and Control of Noncommunicable Diseases 2013-2020, World Health Organization, Geneva, 2013. [21] G. Alleyne, et al., Embedding non-communicable diseases in the post-2015 development agenda, Lancet 381 (9866) (2013) 566e574. [22] M. de Weerd, et al., Prevalence of asymptomatic carotid artery stenosis according to age and sex: systematic review and metaregression analysis, Stroke 40 (4) (2009) 1105e1113. [23] M. de Weerd, et al., Prevalence of asymptomatic carotid artery stenosis in the general population: an individual participant data meta-analysis, Stroke 41 (6) (2010) 1294e1297. [24] C.B. Rockman, et al., The prevalence of carotid artery stenosis varies significantly by race, J. Vasc. Surg. 57 (2) (2013) 327e337. [25] X. Wang, et al., Carotid atherosclerosis detected by ultrasonography: a national cross-sectional study, J. Am. Heart Assoc. 7 (8) (2018), e008701. [26] D. Moher, et al., Preferred reporting items for systematic reviews and metaanalyses: the PRISMA statement, PLoS Med. 6 (7) (2009), e1000097. [27] F.G.R. Fowkes, et al., Comparison of global estimates of prevalence and risk factors for peripheral artery disease in 2000 and 2010: a systematic review and analysis, Lancet 382 (9901) (2013) 1329e1340. [28] D. Fanelli, R. Costas, J.P. Ioannidis, Meta-assessment of bias in science, Proc. Natl. Acad. Sci. Unit. States Am. 114 (14) (2017) 3714e3719. [29] J.J. Barendregt, et al., Meta-analysis of prevalence, J. Epidemiol. Community Health 67 (11) (2013) 974e978. [30] United Nations, Department of economic and social affairs, population division. World population prospects, the 2017 revision, Available from, https:// esa.un.org/unpd/wpp/, 2017. (Accessed 8 December 2017). [31] I. Rudan, et al., Global estimate of the incidence of clinical pneumonia among children under five years of age, Bull. World Health Organ. 82 (12) (2004) 895e903. [32] National Health and Family Planning Commission of the People's Republic of China, China Health and Family Planning Statistical Yearbook 2014, China union medical university press, 2014. [33] Chinese Center for Disease Control and Prevention and Center for Chronic and Non-communicable Disease Control and Prevention, Report on Chronic Disease Risk Factor Surveillance in China (2010), Military Medical Science Press, Beijing, 2012. [34] S.A. Peters, D.E. Grobbee, M.L. Bots, Carotid intimaemedia thickness: a suitable alternative for cardiovascular risk as outcome? Eur. J. Cardiovasc. Prev. Rehabil. 18 (2) (2011) 167e174. [35] U.G. Schulz, P.M. Rothwell, Sex differences in carotid bifurcation anatomy and the distribution of atherosclerotic plaque, Stroke 32 (7) (2001) 1525e1531. [36] G. Le Gal, et al., Hormone replacement therapy use is associated with a lower occurrence of carotid atherosclerotic plaques but not with intima-media thickness progression among postmenopausal women, Atherosclerosis 166 (1) (2003) 163e170. [37] C. Zhan, et al., Prevalence and risk factors of carotid plaque among middleaged and elderly adults in rural Tianjin, China, Sci. Rep. 6 (2016) 23870. [38] R. Sturlaugsdottir, et al., Prevalence and determinants of carotid plaque in the cross-sectional REFINE-Reykjavik study, BMJ open 6 (11) (2016), e012457. [39] S.Y. Woo, et al., Prevalence and risk factors for atherosclerotic carotid stenosis and plaque: a population-based screening study, Medicine 96 (4) (2017).