Prevalence of dyslipidaemia in patients treated with lipid-lowering agents in China: Results of the DYSlipidemia International Study (DYSIS)

Accepted Manuscript Prevalence of dyslipidaemia in patients treated with lipid-lowering agents in China: results of the DYSlipidemia International Study (DYSIS) Shuiping Zhao , MD Yongjun Wang , MD Yiming Mu , MD Bilian Yu , MD Ping Ye , MD Xiaowei Yan , MD Zhanquan Li , MD Yidong Wei , MD Baishaili M. Ambegaonakr , PhD Dayi Hu , MD PII:

S0021-9150(14)01166-6

DOI:

10.1016/j.atherosclerosis.2014.05.916

Reference:

ATH 13541

To appear in:

Atherosclerosis

Received Date: 15 April 2014 Accepted Date: 2 May 2014

Please cite this article as: Zhao S, Wang Y, Mu Y, Yu B, Ye P, Yan X, Li Z, Wei Y, Ambegaonakr BM, Hu D, , Prevalence of dyslipidaemia in patients treated with lipid-lowering agents in China: results of the DYSlipidemia International Study (DYSIS), Atherosclerosis (2014), doi: 10.1016/ j.atherosclerosis.2014.05.916. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

Prevalence of dyslipidaemia in patients treated with lipid-lowering agents in China: results of the DYSlipidemia International Study (DYSIS)

RI PT

Authors: Prof. ShuipingZhaoa, MD, Prof. YongjunWangb, MD, Prof. YimingMuc, MD, Dr. BilianYua, MD, Prof. Ping Yed, MD, Prof. XiaoweiYane, MD, Prof.

ZhanquanLif, MD, Prof. YidongWeig, MD, Dr. Baishaili M. Ambegaonakrh, PhD,

SC

Prof. Dayi Hui, MD, on behalf of the DYSIS-China Study Investigators.

a.Department of Cardiology, Second Xiangya Hospital, Central South University No.139, People Street (M.), Changsha 410011, China

M AN U

b. Department of Neurology, Beijing Tiantan Hospital, Capital Medical University No. 6, Tiantan Xi Li, Dongcheng District, Beijing 100050, China c. Department of Endocrinology, Chinese PLA General Hospital No.28, Fuxing Rd, Haidian District, Beijing 100853, China

d. Department of Geriatric Cardiology, Chinese PLA General Hospital No.28, Fuxing Rd, Haidian District, Beijing 100853, China

e. Department of Cardiology, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College

TE D

No.1, Shuai Fu Yuan, Dongcheng District, Beijing 100730, China

f. Department of Cardiology, The People’s Hospital of Liaoning Province No. 33, Wen Yi Rd., Shenhe District, Shenyang 110016, China g. Department of Cardiology, Tenth People’s Hospital of Tongji University No. 301, Yanchang Rd. (M), Shanghai 200072, China h. Merck& Co., Inc.; Whitehouse Station, NJ, USA

EP

i. Department of Cardiology, Peking University People’s Hospital

AC C

No.11, Xi Zhi Men Nan Da Jie, Xicheng District, Beijing 100044, China

Correspondence to:

Dr. Dayi Hu at Department of Cardiology, Peking University People’s Hospital, No.11, Xi Zhi Men Nan Da Jie, Xicheng District, Beijing 100044 China, or at [email protected]. Tel.:+86 10 8832 4726

1

ACCEPTED MANUSCRIPT

Abstract Background Despite clear guideline recommendations, there is a paucity of data regarding the

RI PT

prevalence and type of persistent lipid profile abnormalities in patients on stable lipid-lowering therapy in China. Methods

This cross-sectional trial included 25,697 patients treated with lipid-lowering agents

SC

from 122 centres between April 2012 and October 2012; all underwent clinical

examination and had their latest fasting lipid profiles while on lipid-lowering therapy recorded. Logistic regression was performed to assess predictors for lipid

M AN U

abnormalities classified according to current Chinese guidelines. Findings

Overall, 29·1% of patients had no lipid abnormalities, and 38·5% of patients did not achieve the therapeutic goal for low-density lipoprotein cholesterol (LDL-C), either as a single lipid anomaly or associated with low high-density lipoprotein cholesterol

TE D

(HDL-C), elevated triglycerides, or both. Subjects with low risk were more likely than those with very high and high risk to be at target LDL-C levels. Furthermore, 10·4% of very high-risk patients and 11·1% of high-risk patients who attained the LDL-C goal failed to attain non-HDL-C goals. Diabetes was shown to be a strong predictor of

EP

failure in attaining non-HDL-C and both goals (OR 3·03; 3·22, 95% CI 2·58–3·55; 2·73–3·79, respectively).

AC C

Conclusion

Although great improvements have been made over the past decade, the large majority of very high-risk and high-risk patients treated with lipid-lowing agents still had one or more manifestations of dyslipidaemia. Further clinical evidence is needed to clarify whether adding other lipid-lowering agents to a statin will be associated with additional cardiovascular risk reduction. Keywords cardiovascular diseases dyslipidaemia cholesterol

statin

non-HDL cholesterol 2

lipid-lowering agents

LDL

ACCEPTED MANUSCRIPT

Introduction Today, cardiovascular diseases (CVD) have become the leading cause of death in China. According to a national prospective cohort study from 1991 to 2000, 43·8% of deaths in Chinese adults aged ≥40 years were attributable to heart disease and stroke.1

RI PT

Dyslipidaemia is one of the most important modifiable risk factors for CVD.2 It has been shown recently that serum total cholesterol (TC) and low-density lipoprotein (LDL) cholesterol levels are high and increasing in the Chinese population.3 Although

SC

the use of lipid-lowering agents, particularly statins, has increased considerately during the last decades, a substantial number of clinical events still exist with the global epidemics of metabolic syndrome, obesity, and type 2 diabetes mellitus (DM).

M AN U

Dyslipidaemia secondary to these conditions is not limited to elevated TC or LDL cholesterol levels, but is often combined with reduced levels of high-density lipoprotein (HDL) cholesterol and/or elevated triglycerides (TGs),4,5 which are also important risk factors for CVD.6,7

TE D

Effective treatment targeting LDL cholesterol by statins substantially reduces cardiovascular risk: reducing LDL cholesterol by 1 mmol/L (39 mg/dL) leads to a 21% reduction in risk for coronary events.8 However, even in patients with optimal LDL cholesterol levels, a high residual risk of atherothrombotic events remains.9 Thus,

EP

beyond statins for LDL cholesterol reduction, combined therapy may be necessary in order to achieve target levels for HDL cholesterol or TGs. It is noted in the National

AC C

Cholesterol Education program (NCEP ATP III) that therapeutic intervention targeting low HDL-C and high TGs should be initiated in high-risk individuals with additional lipid abnormalities.10 In addition, the 2011 ESC/EAS guidelines of dyslipidaemia recommended that for patients with combined hyperlipidaemias, DM, or metabolic syndrome, a secondary lipid goal of non-HDL cholesterol, which is easily calculated from TC minus HDL cholesterol, should be considered.11

However, little is known about the prevalence and type of persistent lipid profile abnormalities in patients treated with lipid-lowering agents in China, and how their 3

ACCEPTED MANUSCRIPT

lipid profile is related to patient characteristics. This study, the Dyslipidemia International Study of China (DYSIS-China), has as its objective to estimate the prevalence of persistent dyslipidaemia in patients treated with lipid-lowering agents,

RI PT

and to investigate predictors for non-achievement of various recommended lipid profiles.

Methods

SC

Study design and patient population

DYSIS is a string of worldwide epidemiological multicentre cross-sectional studies. DYSIS-China included 25,697 patients from 122 centres in China between April 2012

M AN U

and October 2012. It was purely observational, as the only available data were documented, and treatment or assessment of patients was not changed by participation in the study. The investigators were internists, cardiologists, endocrinologists, geriatricians, and neurologists. Consecutive outpatients were eligible for study inclusion if they were 45 years old or older, currently treated with any kind of

TE D

lipid-lowering drugs, had a documented fasting lipid profile during the previous 6 months performed after drug therapy for at least 3 months. The only exclusion criterion was active participation in a clinical study. All subjects gave informed consent before entering the study, and the study protocol was approved by the Ethics

EP

Committee of each clinic centre.

AC C

Data collection by the physicians All patient data were collected from clinical examination and medical charts from a single outpatient visit over a 6-month recruitment period. Data on demographics (gender, age, and nationality), cardiovascular risk profile, cardiovascular history, and lipid-lowering therapy were recorded.

Cardiovascular risk factors were documented at one visit. Diabetes mellitus was defined as previous diagnosis of DM or patient currently on antidiabetic or insulin therapy. Elevated fasting plasma glucose was defined as serum glucose between 6·1 4

ACCEPTED MANUSCRIPT

and 7·0 mmol/L from the latest available laboratory test. Metabolic syndrome was recorded according to the definition of the National Cholesterol Education Program Adult Treatment Panel III (NCEP ATP III).12 Obesity was defined as a body-mass

RI PT

index (BMI) of more than 28 kg/m2. Definition of hypertension was treatment of previously diagnosed hypertension. First grade family history of premature CVD was

considered if a first-degree relative (parents, brothers or sisters) suffered from myocardial infarction or sudden death (<55 years in men and <65 years in women).

SC

Sedentary lifestyle was determined if the patient did not conduct usual physical activity (ie, a minimum of walking 20–30 min on 3–4 days a week or equivalent). Alcohol consumption was reported if the patient drank liquor ≥50 mL per day, red

M AN U

wine ≥150 mL per day, and beer ≥500 mL per day. Cardiovascular history including definite coronary heart disease (CHD), ischaemic cerebrovascular disease, peripheral arterial disease (PAD), and symptomatic chronic heart failure (NYHA II-IV) were also recorded.

TE D

The documentation of chronic medication focused on statins and other lipid-lowering therapies, including cholesterol absorption inhibitors, fibrates, nicotinic acid, and xuezhikang, which is a traditional Chinese medicine extracted from red yeast rice. The name and daily dose of the lipid-lowering drugs taken by the patient at the time

EP

of visit as well as during the past 1 year were noted. Furthermore, antihypertensive,

AC C

antidiabetic, and antiplatelet drugs were recorded.

Risk classification and treatment goals Classification of patient risk and definition of therapeutic goals for TC and LDL cholesterol were based on the 2007 Chinese guidelines for the management of dyslipidaemias.13 Patients were classified into low (10-year risk score of ischaemic CVD less than 5%), moderate (10-year risk score 5% to 10%), and high-risk (having CHD or other atherosclerotic vascular disease, diabetes, or 10-year risk score 10% to 15%) groups. Very high-risk patients were considered to be those with CHD or ischaemic stroke plus diabetes, or acute coronary syndrome. LDL cholesterol 5

ACCEPTED MANUSCRIPT

treatment goals are as follows: <4·1 mmol/L (160 mg/dL), <3·4 mmol/L (130 mg/dL), <2·6 mmol/L (100 mg/dL), and <2·0 mmol/L (80 mg/dL) for low-, moderate-, highand very high-risk patients, respectively. TC treatment goals are <6·2 mmol/L (240

RI PT

mg/dL), <5·2 mmol/L (200 mg/dL), <4·1 mmol/L (160 mg/dL), and <3·1 mmol/L (120 mg/dL) for low-, moderate-, high- and very high-risk patients respectively. The goals for non-HDL cholesterol are 0·78 mmol/L (30 mg/dL) above the LDL

SC

cholesterol goal for each level of risk.

Statistical analysis

Categorical variables are presented as frequencies and percentages and continuous

M AN U

variables as means±standard deviation (SD). Multiple logistic regression analysis was performed to determine independent predictors of lipid abnormalities. The variables included in the model were: age ≥65 years, 20–40 versus 10 mg/day simvastatin equivalent, female, 80 versus 10 mg/day simvastatin equivalent, current smoker, sedentary lifestyle, drinking, BMI ≥28 kg/m2, hypertension, DM, and cerebrovascular

TE D

disease. Backward elimination (α=0·05) was performed. Statistical significance was accepted at the two-sided 0·05 level and all confidence intervals were computed at the 95% level. Statistical analyses were performed with SAS version 9.2 (SAS Institute

EP

Inc., Cary, NC, USA).

Role of the funding source

AC C

The study was designed by the members of steering committee in collaboration with sponsor. Data analysis and writing of the report wereconducted by an independent contract research organization. The corresponding author and the steering committee had full access to all data in the study and had fi nal responsibility for the decision to submit for publication.

Results Patient characteristics and risk categories For 380 (1·48%) of the included patients, lipid parameters were inappropriate or 6

ACCEPTED MANUSCRIPT

missing. These patients were not included in the lipid analyses. Data from 25,317 consecutive outpatients from 27 provinces were analysed. The geographical distribution of participants was well balanced between the northeast (18·0%), north

RI PT

(17·6%), east (16·8%), middle of south (16·1%), southwest (16·0%), and northwest (15·5%) regions of China. Half of the subjects (50·8%) came from third-level hospitals.

SC

Patient characteristics, risk categories and lipid parameters are summarised in Table 1.

The mean patient age was 65·4±10·49 years and 51·3% were men. Comorbidities were common; at least one comorbid condition was seen in most patients (89%),

M AN U

including hypertension (65·8%), DM (34·7%), CHD (37·2%), and cerebrovascular disease (16·9%). When applying the categorisation criteria of the Chinese guidelines for the management of dyslipidaemias, the great majority of patients (71·1%) fell into the very high- or high-risk category.

TE D

Lipid-lowering treatment

Almost all the patients were on monotherapy (98·0%). Of the patients with monotherapy, statins were reported in 88·9% of patients. Atorvastatin (36·6%) and simvastatin (36·6%) were both the most frequently prescribed statins followed by

EP

rosuvastatin (8·5%), pravastatin (3·3%), fluvastatin (2·5%), lovastatin (0·7%), and pitavastatin (0·6%). Only 11·1% of patients received other lipid-lowering treatment:

AC C

ezetimibe (0·1%), fibrates (5·1%), nicotinic acid (0·1%), and xuezhikang (5·0%) (Table 2).

Figure 1 shows the patients’ statin regimen classified according to potency for all the patients or the very high- and high-risk patients. The majority of patients received statin dose potency in the intermediate range (equivalent to simvastatin 20 or 40 mg/day). The LDL-C goal attainment is 64·2%, 58·0%, 60·4%, 63·3%, 62·8%, and 59·4% for statin dose potency 1 to 6, respectively. 7

ACCEPTED MANUSCRIPT

Lipid abnormalities The lipid profile stratified according to the 2007 Chinese guidelines for the management of dyslipidaemias is shown in Table 3. Overall, 29·1% of patients had no

RI PT

lipid abnormalities, the majority of patients (51·2%) did not have a TC at goal and 38·5% of patients did not have a LDL-C at goal. LDL-C not at goal＋low HDL-C

and/or elevated TGs were found in 33·4% of patients, and LDL-C at goal＋low HDL-C and/or elevated TGs in 14·2% of patients (Figure 2A). Of note, the TC and

SC

LDL-C goals were not achieved in 84·1% and 60·3% of very high-risk category

patients, and 57·9% and 45·2% of high-risk category patients, respectively. In 12·2% of very high-risk category patients, no lipid abnormalities were found in only 16·7%

M AN U

of patients, LDL-C not at goal＋low HDL-C and/or elevated TGs were found in 57·0% of patients, and LDL-C at goal＋low HDL-C and/or elevated TGs in 17·2% of patients (Figure 2B). Corresponding numbers in the high-risk category patients were: no lipid abnormalities in 26·8% of patients, LDL-C not at goal＋low HDL-C and/or elevated TGs were found in 38·5% of patients, and LDL-C at goal＋low HDL-C

TE D

and/or elevated TGs in 14·4% of patients (Figure 2C).

In Table 4 lipid profiles are stratified according to a history of diabetes and/or CHD. A majority (65·2%) of patients with diabetes alone did not achieve TC targets and

EP

more than half missed LDL-C targets (51·9%). Patients with CHD alone had better control rates of both TC and LDL-C than patients with diabetes alone. Relative to

AC C

patients with diabetes and no CHD, those with both diabetes and CHD were less likely to be at their TC and LDL-C goals, and also less likely to have desirable HDL-C levels.

We also calculated non-HDL-C attainment. Overall, 35·9% of patients did not attain their non-HDL-C goal. When the rates of non-HDL-C not at goal were evaluated by risk level, the rate was highest in the very high-risk category (52%) as compared with the high-risk (41·2%), moderate-risk (31%) and low-risk category (10·4%). The highest rates of failure in attaining both LDL-C and non-HDL-C goals simultaneously 8

ACCEPTED MANUSCRIPT

were found in the very high-risk category (47·9%). Importantly, 10·4% of very high-risk patients and 11·1% of high-risk patients who attained the LDL-C goal failed to attain non-HDL-C goals (Table 3). In addition, in patients with diabetes (without

RI PT

CHD), the rate for non-HDL-C not at goal was 48·6% compared with 32·0% in those with CHD only, and among patients attaining LDL-C success, 13·7% failed to attain

their non-HDL-C goal. More than half (52·0%) of patients with both diabetes and

SC

CHD failed to attain their non-HDL-C goal (Table 4).

Predictors for lipid abnormalities

In the multivariate logistic regression models, a number of variables were identified

M AN U

which independently were associated with lipid abnormalities (Table 5). Factors significantly associated with LDL-C treatment failure were: 20–40 versus 10 mg/day simvastatin equivalent (OR 1·51, 95% CI 1·30–1·76), female (OR 1·35, 95% CI 1·15–1·58), current smoker (OR 1·39, 95% CI 1·09–1·78), diabetes (OR 3·24, 95% CI

TE D

2·76–3·81), and cerebrovascular disease (OR 2·20, 95% CI 1·77–2·74).

Table 6 shows the OR estimates for predictors of non-HDL-C and both LDL-C and non-HDL-C failure. Importantly, diabetes was shown to be a strong predictor of failure in attaining non-HDL-C and both goals (OR 3·03; 3·22, 95% CI 2·58–3·55;

AC C

Discussion

EP

2·73–3·79, respectively).

DYSIS-China provides an up-to-date view of lipid-lowering therapy across different levels of hospital. The results of this large-scale observational study suggest that overall more than one third of patients did not reach the LDL-C and non-HDL-C targets recommended by the Chinese guidelines. Meanwhile, despite stable lipid-lowering therapy, the large majority of very high-risk and high-risk patients still had one or more manifestations of dyslipidaemia. These findings are supported by the recently published China Intensive Lipid Lowering with Statins in Acute Coronary Syndrome (CHILLAS) trial in which treatment with atorvastatin 20 or 40 mg for 2 9

ACCEPTED MANUSCRIPT

years resulted in only half reaching the LDL-C goal of <2·0 mmol/L.14

However, our results indicated that great improvements have been made in LDL-C

RI PT

goal attainment over the past decade. In a 2000 survey in China,15 only 16·6% of those with established atherosclerotic disease such as CHD, stroke and peripheral

artery diseases had an LDL cholesterol level of <100 mg/dL (<2·6 mmol/L). The less “potent” statins such as pravastatin, fluvastatin and xuezhikang, generally resulting in a smaller decline in LDL-C, were widely used. Similarly, in the second multicentre

SC

survey of dyslipidaemia management in China,16 which enrolled patients between January 2004 and February 2006, the goal attainment rates of LDL-C according to

M AN U

NCEP ATP III 2004 were 31% and 22% for patients at high and very high risk, respectively. Among 78·4% of patients treated with statins, only 35% achieved the LDL-C goal. The results of DYSIS in 12 European countries and Canada were reported recently,17,18 in which 48·2% of 22,063 statin-treated patients did not achieve the therapeutic goal for LDL-C, either as a single lipid anomaly or associated with

TE D

low HDL-C, elevated triglycerides, or both. Remarkable is the fact that, despite the discrepancies in intervention target level and risk stratification, the overall LDL-C goal attainment in DYSIS-China is higher than that in European countries. The improvement of LDL-C goal attainment may largely be due to the widespread use of

EP

more “potent” statins. The great majority of patients received atorvastatin, simvastatin, or rosuvastatin treatment. These results turn out contrary to the prospective

AC C

epidemiological survey by Yusuf and colleagues19 which showed only 2·0% of individuals with cardiovascular disease took statins in China.

The observation that subjects with low risk were more likely than those with very high and high risk to be at target LDL-C levels are inconsistent with those observed in European countries,17,20 which suggests better achievement of LDL-C therapeutic goal was found among patients at high cardiovascular risk. The reasons underlying this apparent

discrepancy

are

likely

multifactorial,

with

both

patient-

and

physician-associated factors. Very few patients with very high or high risk were on 10

ACCEPTED MANUSCRIPT

high-dose statins or receiving combination therapy because of suboptimal patient adherence towards typical and increasing dosages of statins or the perceived complexity and cost of combinational therapy. There is also a significant gap between clinicians’ perceptions and actual achievement rates of LDL-C due to the lack of

RI PT

recognition on guidelines.21 In addition, despite the fact that diabetes is one of the

major controllable risk factors for cardiovascular disease, and the wide dissemination

of guideline-recommended target LDL-C levels, clinical inertia was apparent in this

SC

cohort of patients who were already at high risk and on statin therapy. In the present study, amongst the patients with very high and high risk, those with diabetes had

or more intense regimen is needed.

M AN U

markedly lower LDL-C goal attainment than those without diabetes, indicating a new

A few studies have evaluated the attainability of the aggressive LDL-C treatment goal of <70 mg/dL.14,22,23 Kennedy et al.22 reported that it would be extremely difficult to reach an LDL-C goal in patients with diabetes and CVD in the United States. They

TE D

estimated that 25% of patients would require more than two lipid-lowering drugs at maximal doses to attain this goal. Furthermore, It has previously been noted that an algorithm-based statin uptitration/ezetimibe combination regimen was useful to increase LDL-C lowering where statin monotherapy had not achieved target lipid

EP

values.24,25 In the present study, the median daily doses of atorvastatin 10–20 mg or simvastatin 20–40 mg, which were lower than those used in western clinical outcome

AC C

studies, reflects the usual prescribing practice in China. However, it is totally unexpected that the LDL-C goal attainments of higher statin doses equivalent to 80–160 mg/day simvastatin were not superior to that of lower doses equivalent to 20–40 mg/day simvastatin, indicating that further upward dose titration of statins did not bring an additional increment in goal attainment in the Chinese population. These findings highlight the potential necessity to add a second drug rather than further titrate the statin doses upward. Although the value of ezetimibe in reducing adverse cardiovascular outcomes remains unproven, it seems reasonable to employ the combination of a statin with ezetimibe, which was rarely prescribed in the present 11

ACCEPTED MANUSCRIPT

study.

In contrast to other studies that focused only on LDL-C target attainment,26,27

RI PT

DYSIS-China, however, took a broader perspective. Our study also examined other lipid fractions and found LDL-C was frequently raised, in combination with low HDL-C and/or elevated triglycerides (TGs), particularly in those with very high risk. Despite this, very few patients were receiving treatment for other lipid parameters.

SC

Similar findings were described in other DYSIS studies.17,18,20,28 Additionally, many patients who reached their LDL-C goal had persistently elevated non-HDL-C levels, which is particularly relevant given the worldwide increase in the incidence of obesity

M AN U

and DM.29 Indeed, the presence of DM was the most important determinant of failure to attain the non-HDL-C goal. The results above suggest a future need for multi-targeted treatment strategies that could address the full range of lipid abnormalities.

TE D

Our present study enrolled a nationally representative sample of 25,697 patients from 27 provinces and 122 hospitals across China, providing new information regarding the “real world” management in the Chinese population. However, it still has several limitations that should be addressed. First, it was an observational cross-sectional trial

EP

that did not assess long-term outcomes. A prospective follow-up study is required to evaluate medical treatment and attainment in relation to mortality in patients treated

AC C

with lipid-lowering agents. Second, lipid parameters were not measured in a central core laboratory. Finally, given that current lipid-lowering agents use was a patient eligibility criterion, we may have overestimated the target attainment of all lipid parameters, especially in high-risk patients.

Contributors DH conceived and initiated the study, supervised its conduct and data analysis and had primary responsibility for writing the report. SZ, YW and YM are the co-principal investigators of Cardiology, Neurology and Endocrinology respectively, and contributed to 12

ACCEPTED MANUSCRIPT

the conception, design and implementation of the study.BY wrote the report. PY, XY, ZL, YW contributed to conception, design and assembly of data. BMA contributed to conception

RI PT

and design. All the authors reviewed the report and provided critical inputs for its revision.

DYSIS-China investigators

Bu Ai Jia Er, Cao Guiwen, Chen Buxing, Chen Hong*, Chen Jin, Chen Jing, Chen

Liefeng, Chen Min, Chen Qiong, Chen Shaoliang, Chen Tielong, Chen Xiaofei, Chen

Xiaohong, Chen You, Cheng Guoli, Cheng Mei, Cong Hongliang, Cui Qin, Cui Shiwei, Cui

SC

Yong, Dai Shudong, Dai Henghua, Deng Xiaomei, Di Yirong, Ding Xiaoyan, Dong Birong*, Dong Yumei, Dong Yugang*, Du Ping, Duan Lei, Fan Limei, Fang Ningyuan, Fei Lixia, Feng

M AN U

Lie, Ge Jun, GeWeihong, Guo Hongmin, Guo Minxia, Han Qinghua, He Fengchang, Hu Dayi*, Hu Lingzhi, Hu Xueqiang, Hu Yaojun, Hu Yiming, Hu Zhiping, Hua Fei, Hua Qi*, Huang Dejia*, Huang Gewen, Huang Hongman, Huang Liming, Huang Qiong, Huang Ruowen, Huang Taifu, Jiang Bin, Jiao Kai, Jin Hui, Jin Huigen, KuangJinsong, Li Bao, Li Chengjiang, Li Hongjuan, Li Jun(Xinjiang), Li Jun(Jiangsu), Li Nanfang*, Li Qifu, Li Qiang,

TE D

Li Xin, Li Xueyou, Li Yanbing*, Li Yanping, Li Yansheng*, Li Yong*, Li Yuling, Li Zhanquan*, Li Zhengfang, Liang Li, Liang Yongxue, Liang Zerong, Liao Yuhua*, Liu Fan, Liu Hong, Liu Hui, Liu Minling, Liu Qiang, Liu Qingsong, Liu Shaokui, Liu Weidong, Liu Xueping, Lu Xinjian, Luo Benyan, Luo Shaonian, Luo Suxin, Lv Hong, LvYun, Ma Aiqun*,

EP

Ma Jianhua, Ma Qiang, Ma Yan, Ma Changsheng, Miao Yide, Mu Yiming*, NieXiaoli, NiuXiaoyuan, Pan Hongtao, Pan Mingkang, Peng Qiaoqing, Qian Huifen, Qu Qiumin, Qu

AC C

Lijie, Ren Liqun, Shao Jingshan, Se Qiang, Huang Jing, Shen Xiuyun, Shi Yongquan*, Si Liangyi, Song Zhi, Song Zhiyuan, Sun Yufeng, Tan Chunyan, TengXiaochun, Tian Haoming, Tian Wenhua, Tong Qinwei, TuQiuyun, Wang Keying, Wang Aihong, Wang Chaohui, Wang Chunning, Wang Dezhao, Wang Guixia, Wang Hanqiao, Wang Jianan*, Wang Jianjun, Wang Lan, Wang Xiaoming, Wang Yaping, Wang Yangwei, Wang Yongjun*, Wei Meifang, Wei Yidong*, Wu Hongyun, Wu Chun, Wu Dongmei, Wu Jiang,Wu Jun, Wu Xiaolin, Wu Zonggui*, XiGuangxia, Xiang Yi, Xiao Qian, Xing Xiaoping, Xing Yulong, Xu Anding*, XueYuanming, Yan Chuanzhu, Yan Tao, Yan Xiaowei*, Yang Gangyi, Yang Jian, Yang Wangpingm Yang Xiaosu, Yang Xinchun, Yang Yifang, Yang Yu, Yao Mingyu, Ye Min, Ye 13

ACCEPTED MANUSCRIPT

Ping*, Yu Bo, Yu Jiangyi, Yu Jinming*, Yu Yan, Zeng Ling, ZengLongyi, ZengXiaoyun, Zhang Baorong, Zhang Bei, Zhang Chaoxin, Zhang Xuelian, Zhang Dadong, Zhang Dongping, Zhang Fuchun, Zhang Hong, Zhang Huifang, Zhang Liping, Zhang Liyang, Zhang Rufu, Zhang Saidan, Zhang Weijuan, Zhao Dong*, Zhao Gang*, Zhao Shuiping*, Zhao Xiuxin,

Zou Xiangping * Denotes Steering Committee Member.

Conflicts of interest

M AN U

We declare that we have no conflicts of interest.

SC

RI PT

Zheng Qiangsun, Zheng Yang*, Zhou Xiaohui, Zhou Yali, Zhou Yujie*, Zhu Yi, Zhu Yulan,

Acknowledgements

The study was funded by research grant from Merck sharp &Dohme (China) Co., Ltd.

References

He J, Gu D, Wu X, et al. Major causes of death among men and women in China.

TE D

1

N Engl J Med 2005; 353: 1124 –34. 2

Yusuf S, Hawken S, Ounpuu S, et al. Effect of potentially modifiable risk factors

associated with myocardial infarction in 52 countries (the INTERHEART study):

3

EP

case-control study. Lancet 2004; 364: 937–52. Yang W, Xiao J, Yang Z, et al. Serum lipids and lipoproteins in Chinese men and

AC C

women. Circulation 2012; 125: 2212–21. 4

American Diabetes Association (ADA). Dyslipidemia management in adults with

diabetes. Diabetes Care 2004; 27(90001): S68–71

5

Snow V, Aronson MD, Hornbake ER, Mottur-Pilson C, Weiss KB. Lipid control

in the management of type 2 diabetes mellitus: a clinical practice guideline from the American College of Physicians. Ann Intern Med 2004; 140: 644–9. 6

Gordon T, Castelli WP, Hjortland MC, Kannel WB, Dawber TR. High density

lipoprotein as a protective factor against coronary heart disease. The Framingham Study. Am J Med 1977; 62: 707–14. 14

ACCEPTED MANUSCRIPT

7

Sarwar N, Danesh J, Eiriksdottir G, et al. Triglycerides and the risk of coronary

heart disease: 10,158 incident cases among 262,525 participants in 29 Western prospective studies. Circulation 2007; 115: 450–8. Baigent C, Keech A, Kearney PM, et al. Efficacy and safety of

RI PT

8

cholesterol-lowering treatment: prospective meta-analysis of data from 90,056 participants in 14 randomised trials of statins. Lancet 2005; 366: 1267–1278. 9

Fruchart JC, Sacks FM, Hermans MP, et al. The Residual Risk Reduction

SC

Initiative: a call to action to reduce residual vascular risk in dyslipidaemic patient. Diab Vasc Dis Res 2008; 5: 319–35.

10 Expert Panel on Detection E, and Treatment of High Blood Cholesterol in Adults.

M AN U

Executive Summary of the Third Report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III). JAMA 2001; 285: 2486–97. 11 European Association for Cardiovascular Prevention & Rehabilitation, Reiner Z, Catapano AL, et al. ESC/EAS Guidelines for the management of dyslipidaemias: the

TE D

Task Force for the management of dyslipidaemias of the European Society of Cardiology (ESC) and the European Atherosclerosis Society (EAS). Eur Heart J 2011; 32: 1769–818.

12 National Cholesterol Education Program (NCEP) Expert Panel on Detection,

EP

Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III). Third Report of the National Cholesterol Education Program (NCEP)

AC C

Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III) Final Report. Circulation 2002; 106: 3143–421. 13 Hu DY, Ding RJ. Guidelines for management of adult dyslipidemia in China.

Zhonghua Xin Xue Guan Bing Za Zhi 2007; 35: 390–410.

14 Yu BL, Zhao SP, Peng DQ, Huo Y, Hu L. A comparison of non-HDL and LDL

cholesterol goal attainment in the CHILLAS trial. Int J Cardiol 2013 9;168:4340-2. 15 Current Status in Clinical Control of Hypercholesterolemia in China Research Group.

A

multi-center

study

of

current 15

status

on

clinical

control

of

ACCEPTED MANUSCRIPT

hypercholesterolemia in China: success rate and related factors. Zhonghua Xin Xue Guan Bing Za Zhi 2002; 30: 109–114. 16 Wu YF. The second multi-center survey of dyslipidemia management in China:

35: 420–7.

RI PT

goal attainment rate and related factors. Zhonghua Xin Xue Guan Bing Za Zhi 2007;

17 Gitt AK, Drexel H, Feely J, et al. Persistent lipid abnormalities in statin-treated

patients and predictors of LDL-cholesterol goal achievement in clinical practice in

SC

Europe and Canada. Eur J Prev Cardiol 2012; 19: 221–30.

18 Goodman SG, Langer A, Bastien NR, et al. Prevalence of dyslipidemia in statin-treated patients in Canada: results of the DYSlipidemia International Study

M AN U

(DYSIS). Can J Cardiol 2010; 26: e330–5.

19 Yusuf S, Islam S, Chow CK, et al. Use of secondary prevention drugs for cardiovascular disease in the community in high-income, middle-income, and low-income countries (the PURE Study): a prospective epidemiological survey. Lancet 2011; 378: 1231–43.

TE D

20 Gitt AK, Jünger C, Smolka W, Bestehorn K. Prevalence and overlap of different lipid abnormalities in statin-treated patients at high cardiovascular risk in clinical practice in Germany. Clin Res Cardiol 2010; 99: 723–33. 21 Hwang JY, Jung CH, Lee WJ, et al. Low density lipoprotein cholesterol target

EP

goal attainment rate and physician perceptions about target goal achievement in Korean patients with diabetes. Diabet Metab J 2011; 35: 628–36.

AC C

22 Kennedy AG, MacLean CD, Littenberg B, Ades PA, Pinckney RG. The challenge of achieving national cholesterol goals in patients with diabetes. Diabet Care 2005; 28: 1029–34. 23 Yan AT, Yan RT, Tan M, et al. Vascular Protection (VP) and Guidelines Oriented Approach to Lipid Lowering (GOALL) Registries Investigators. Contemporary management of dyslipidemia in high-risk patients: targets still not met. Am J Med 2006; 119: 676–83.

16

ACCEPTED MANUSCRIPT

24 Teoh H, Mendelsohn AA, Goodman SG, et al. Usefulness of statin-ezetimibe combination to reduce the care gap in dyslipidemia management in patients with a high risk of atherosclerotic disease. Am J Cardiol 2009; 104: 798–804.

RI PT

25 Maron DJ, Hartigan PM, Neff DR, et al. Impact of adding ezetimibe to statin to achieve low-density lipoprotein cholesterol goal (from the Clinical Outcomes

Utilizing Revascularization and Aggressive Drug Evaluation [COURAGE] trial). Am J Cardiol 2013;111:1557–62.

SC

26 Chin CW, Gao F, Le T, Tan R. Lipid goal attainment and prescription behavior in asian patients with acute coronary syndromes: experience from a tertiary hospital. Clin Med Insights Cardiol 2013; 7: 51–7.

M AN U

27 Gao F, Zhou YJ, Hu da Y, et al. Contemporary management and attainment of cholesterol targets for patients with dyslipidemia in China. PLoS One 2013; 8: e47681.

28 Leiter LA, Lundman P, da Silva PM, et al. Persistent lipid abnormalities in statin-treated patients with diabetes mellitus in Europe and Canada: results of the

TE D

Dyslipidaemia International Study. Diabet Med 2011; 28: 1343–51. 29 Robinson JG, Want S, Smith BJ, Jacobson TA. Meta-analysis of the relationship between non-high-density lipoprotein cholesterol reduction and coronary heart disease.

AC C

EP

J Am Coll Cardiol 2009; 53: 316–22.

17

ACCEPTED MANUSCRIPT

Table 1: Patient characteristics, risk categories, and lipid parameters All patients (N=25,317) Age (years)

RI PT

65·4±10·49

≥65 years

(13089) 51·7%

Male

(12975) 51·3%

Family history of premature CVDa

(2294) 9·1%

(3143) 12·4%

SC

Current smoker Drinking

(2309) 9·1%

Sedentary lifestyleb

(4997) 19·7%

M AN U

Hypertension Systolic BP (mmHg)c Diastolic BP (mmHg)c DM Haemoglobin A1c (%)d

<6·5% CHD Cerebrovascular disease

TE D

<7·0%

(16650) 65·8% 130·9±15·63 78·2±9·72 (8775) 34·7% 7·3±1·89 (3290) 55·3% (2320) 39·0% (9420) 37·2% (4281) 16·9% (263) 1·0%

Metabolic syndrome (ATP III)

(4523) 17·6%

Waist circumference (cm)e

87·0±11·20

AC C

EP

Peripheral arterial disease

BMI (kg/m2)f

24·7±3·27

BMI ≥28 kg/m2

(3469) 13·7%

Risk level (%) Very-high

(3092) 12·2%

High

(14916) 58·9%

Moderate

(2782) 11·0%

Low

(4527) 17·9%

18

ACCEPTED MANUSCRIPT

Lipid profile (mmol/L) 4·59±1·21

LDL-C

2·63±0·97

HDL-C

1·28±0·37

RI PT

TC

TG

1·89±1·45

Non-HDL-C

3·31±1·15

CVD, cardiovascular disease; BP, blood pressure; DM, diabetes mellitus; CHD, coronary heart disease; BMI, body mass index.

SC

Data presented as mean±SD or percentage. aData on 25,312 patients were available. bData on 25,315 patients were available. cData on 25,311 patients were available. dData on 5,950

M AN U

patients were available. eData on 25,204 patients were available. fData on 25,308 patients

AC C

EP

TE D

were available.

19

ACCEPTED MANUSCRIPT

Table 2: Lipid-lowering therapy All patients (N=25,317) (516)2%

Monotherapy

(24801)98%

RI PT

Combination

Statins

(22039)88·9% (9086)36·6%

Atorvastatin

(9080)36·6%

Rosuvastatin

(2107)8·5%

Pravastatin

(810)3·3%

Fluvastatin

(632)2·5%

Lovastatin

(176)0·7%

Pitavastatin

(148)0·7%

Other lipid-lowering agents Ezetimibe

(2762)11·1% (23)0·1%

(1257)5·1%

Nicotinic acid

AC C

EP

TE D

Fibrates

Xuezhikang

M AN U

SC

Simvastatin

20

(20)0·1% (1246)5·0%

ACCEPTED MANUSCRIPT

Table 3: Lipid abnormalities according to Chinese guideline risk levels in patients with complete lipid profiles Very high risk

High risk

Moderate risk

Low risk

TC not at goal

12960 (51·2%)

2601 (84·1%)

8640 (57·9%)

1175 (42·2%)

544 (12·0%)

LDL-C not at goal

9746 (38·5%)

1866 (60·3%)

6742 (45·2%)

741 (26·6%)

397 (8·8%)

TG >1·7 mmol/L

10594 (41·8%)

1237 (40·0%)

5925 (39·7%)

1320 (47·4%)

2112 (46·7%)

HDL-C <1·0 mmol/L (male)

8082 (31·9%)

1263 (40·8%)

4945 (33·2%)

681 (24·5%)

1193 (26·4%)

Non-HDL-C not at goal

9081 (35·9%)

1607 (52·0%)

Both LDL-C and non-HDL-C

7523 (29·7%)

1480 (47·9%)

6142 (41·2%)

862 (31·0%)

470 (10·4%)

5238 (35·1%)

551 (19·8%)

254 (5·6%)

AC C

EP

TE D

M AN U

not at goal

SC

or <1·2 mmol/L (female)

RI PT

All patients

21

ACCEPTED MANUSCRIPT

Table 4:

Lipid abnormalities according to diagnosis of diabetes mellitus (DM) and/or

coronary heart disease (CHD) All CHD

DM without

CHD without

CHD

DM

CHD with DM

RI PT

All DM

6308 (71·9%)

5716 (60·7%)

3707 (65·2%)

3115 (49·2%)

2601 (84·1%)

LDL-C not at goal

4815 (54·9%)

4174 (44·3%)

2949 (51·9%)

2308 (36·5%)

1866 (60·3%)

TG >1·7 mmol/L

3834 (43·7%)

3434 (36·5%)

2597 (45·7%)

2197 (34·7%)

1237 (40·0%)

HDL-C <1·0 mmol/L (male) or 3280 (37·4%)

3325 (35·3%)

2017 (35·5%)

2062 (32·6%)

1263 (40·8%)

<1·2 mmol/L (female) 4368 (49·8%)

3629 (38·5%)

Both LDL-C and non-HDL-C

3832 (43·7%)

3167 (33·6%)

AC C

EP

TE D

not at goal

2761 (48·6%)

2022 (32·0%)

1607 (52·0%)

2352 (41·4%)

1687 (26·7%)

1480 (47·9%)

M AN U

Non-HDL-C not at goal

SC

TC not at goal

22

ACCEPTED MANUSCRIPT

LDL-C not at goal

Elevated TG

OR (95% CI)

p value

OR (95% CI)

p value

1·514 (1·300–1·763)

0·0000

1·554 (1·218–1·983)

0·0004

Female

1·346 (1·147–1·581)

0·0003

0·403 (0·333–0·487)

Current smoker

1·394 (1·093–1·778)

0·0075

Drinking

n.s.

BMI ≥28 kg/m2 (obesity)

LDL-C not at goal and low HDL-C and elevated TG p value

OR (95% CI)

p value

n.s.

n.s.

n.s.

n.s.

0·0000

n.s.

n.s.

n.s.

n.s.

n.s.

n.s.

n.s.

n.s.

n.s.

n.s.

n.s.

0·614 (0·438–0·863)

0·0049

n.s.

n.s.

n.s.

n.s.

n.s.

n.s.

1·442 (1·110–1·873)

0·0062

n.s.

n.s.

n.s.

n.s.

Diabetes mellitus

3·238 (2·755–3·807)

0·0000

1·675 (1·387–2·022)

0·0000

2·495 (1·781–3·496)

0·0000

2·495 (1·781–3·496)

0·0000

Cerebrovascular disease

2·205 (1·773–2·742)

0·0000

n.s.

n.s.

n.s.

n.s.

n.s.

n.s.

20–40 versus 10 mg/day

AC C

EP

TE D

M AN U

simvastatin equivalent

23

OR (95% CI)

SC

Variables

Low HDL-C

RI PT

Table 5: Independent predictors for LDL-C, HDL-C, and triglyceride (TG) abnormalities in multivariable logistic regression models

ACCEPTED MANUSCRIPT

Table 6: Independent predictors associated with failure of non-HDL-C or both LDL-C and non-HDL-C goals after multivariate regression analyses Non-HDL-C not at goal OR (95% CI)

p value

1·184 (1·017–1·380)

0·0300

Female

1·634 (1·389–1·922)

0·0000

Current smoker

1·351 (1·056–1·727)

0·0165

Diabetes mellitus

3·026 (2·578–3·552)

Cerebrovascular disease

1·837 (1·481–2·279)

0·0031

1·530 (1·292–1·813)

0·0000

1·397 (1·083–1·803)

0·0101

0·0000

3·220 (2·734–3·792)

0·0000

0·0000

2·200 (1·769–2·737)

0·0000

M AN U TE D EP AC C 24

p value

1·273 (1·085–1·493)

SC

Age ≥65 years

OR (95% CI)

RI PT

Variables

Both LDL-C and non-HDL-C at goal

ACCEPTED MANUSCRIPT

Figure legends Figure 1

Statin dose potency. Potency 1 is equivalent to simvastatin 5 mg/day;

potency 2 is equivalent to simvastatin 10 mg/day; potency 3 is equivalent to

RI PT

simvastatin 20 mg/day; potency 4 is equivalent to simvastatin 40 mg/day; potency 5 is equivalent to simvastatin 80 mg/day; potency 6 is equivalent to simvastatin ≥160 mg/day.

SC

Figure 2: Distribution of single and multiple combined lipid abnormalities. (A) In all

AC C

EP

TE D

M AN U

patients; (B) in very high-risk patients; (C) in high-risk patients.

25

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT