Small dense low-density lipoprotein is a risk for coronary artery disease in an urban Japanese cohort: The Suita study

Transcription

1 Small dense low-density lipoprotein is a risk for coronary artery disease in an urban Japanese cohort: The Suita study Hidenori Arai 1, Yoshihiro Kokubo 2, Makoto Watanabe 2, Tatsuya Sawamura 3, Tomonori Okamura 4, and Yoshihiro Miyamato 2 1 Department of Human Health Sciences, Kyoto University Graduate School of Medicine 2 Department of Preventive Cardiology, 3 Department of Vascular Physiology, National Cerebral and Cardiovascular Center 4 Department of Preventive Medicine and Public Health, School of Medicine, Keio University

2 Abstract Background and aim: Increased plasma low-density lipoprotein (LDL) cholesterol (LDL-C) concentrations have been shown to be a significant risk factor for coronary artery disease (CAD). Clinical evidence also indicated that small dense LDL (sd-ldl) particles are more atherogenic than large buoyant LDL particles. Because there was no easy assay to measure the amount of sd-ldl, no study has addressed the association between sd-ldl and cardiovascular disease. Therefore, we examined the association between sd-ldl cholesterol (sd-ldl-c) and CAD using a new assay kit for sd-ldl-c in an urban Japanese cohort. Methods: The Suita study is an 11.7-year prospective study in an urban population aged without history of myocardial infarction or stroke. Direct LDL-C and sd-ldl-c were measured in samples from 2034 participants (968 men and 1066 women) by a kit based on the precipitation method with filtration. We calculated the multivariable-adjusted hazard ratios (HRs) of sd-ldl-c and sd-ldl/ldl ratio for CAD using a proportional hazards regression model after adjusting for age, sex, hypertension, diabetes, use of lipid lowering agent, body mass index (BMI), current smoking, and alcohol drinking. Results: We divided men and women to 4 groups according to the quartiles of sd-ldl-c. Levels of sd-ldl-c were positively associated with total cholesterol, triglyceride, BMI, blood pressure, and were negatively correlated with HDL-cholesterol both in men and women. Levels of sd-ldl-c were positively correlated with age only in women. After adjusting age and sex, increasing quartiles of sd-ldl-c were associated with increased incidence of CAD (p=0.001). The HR of the 4th quartile was 3.35 (95% confidence intervals [95% CI]: ). In multivariable adjustment, the HR of the 4th quartile was almost similar. Sex-specific analysis showed that the association between sd-ldl-c and CAD was statistically significant in men, but not in women. We also found that sd- LDL/LDL ratio was a significant risk for CAD after age and sex- and multivariable-adjustment (p=0.001, 0.005, respectively). Furthermore, higher sd-ldl/ldl ratios were significantly associated with CAD even after LDL-C adjustment. Conclusions: We demonstrated that sd-ldl-c and sd-ldl/ldl are significantly associated with the development of CAD in Japanese without history of myocardial infarction or stroke. Our data indicate that sd-ldl-c is an independent and new risk factor for CAD and should be incorporated for a risk assessment. However, a larger study should be done to show the role of sd-ldl on CAD in women.

3 Background Compared to large, buoyant LDL, small dense LDL (sd-ldl) particles exhibit a prolonged plasma residence time, increased penetration into the arterial wall, lower affinity for the LDL receptor, and increased susceptibility to oxidation. Thus, sd-ldl particles possess an elevated atherogenic potential. However, few prospective studies have been done to address the role of sd-ldl in the development of cardiovascular disease.

4 Aims Address whether sd-ldl cholesterol can predict the development of CAD in a large cohort study of Japan, the Suita study.

5 Methods The Suita study, a cohort study for CVD of urban residents was established in men and women aged years had a baseline survey at the National Cerebral and Cardiovascular Center between September 1989 and March For the purposes of this study, we used archived plasma samples that had been frozen at 80 o C for the assessment of direct LDL-C and sd-ldl-c by automated standardized enzymatic analysis on a Hitachi 911 automated analyzer. The kits used for these tests (LDL-C and sd-ldl-c) were provided by Denka Seiken, Tokyo, Japan.

6 Principle of the homogeneous assay for sd LDL-C TRL lb-ldl sd-ldl HDL R1 Surfactant A Sphingomyelinase Surfactant B Surfactant A React and Eliminate React and Eliminate Protect React and Eliminate Cholesterol Cholesterol Cholesterol Cholesterol Esterase Cholesterol Oxidase Catalase Cholesterol Esterase Cholesterol Oxidase Catalase 2H 2 O + O 2 2H 2 O + O 2 2H 2 O + O 2 R2 Purple-red color Measurement H 2 O 2 Cholesterol Esterase Cholesterol Oxidase peroxidase Cholesterol sd-ldl

7 Results

8 1. Baseline Characteristic of Cardiovascular Risk Factors According to Small-dense LDL Cholesterol Quartiles in Men Men Small-dense LDL Cholesterol Q1 Q2 Q3 Q4 Number of subjects Small-dence LDL, range (mean), mg/dl (21.1) (32.7) (45.3) (67.3) P value for Trend Age, year 60.9± ± ± ± Body mass index, kg/m^2 21.5± ± ± ±2.7 <0.001 TC, mg/dl 170±25 189±24 199±25 220±27 <0.001 HDL-C, mg/dl 60±15 57±14 51±11 48±11 <0.001 LDL, mg/dl 86±20 111±21 124±23 140±26 <0.001 Large-LDL, mg/dl 65±17 78±21 79±22 72±24 <0.001 Sd-LDL/LDL ratio 0.25± ± ± ±0.11 <0.001 Blood pressure category, % Optimal blood pressure Normal blood pressure High-normal blood pressure Hypertension grade I-III Antilipidemic drug use, % Diabetes, % Current Smoking, % Current Drinking, %

9 2. Baseline Characteristic of Cardiovascular Risk Factors According to Small-dense LDL Cholesterol Quartiles in Women Women Small-dense LDL Cholesterol Q1 Q2 Q3 Q4 Number of subjects Small-dence LDL, range (mean), mg/dl (18.7) (28.6) (38.5) (59.7) P value for Trend Age, year 51.7± ± ± ±9.1 <0.001 Body mass index, kg/m^2 21.0± ± ± ±2.8 <0.001 TC, mg/dl 175±23 200±22 216±25 234±32 <0.001 HDL-C, mg/dl 67±13 64±12 60±13 54±12 <0.001 LDL, mg/dl 83±17 109±17 130±18 153±30 <0.001 Large-LDL, mg/dl 64±14 81±15 92±17 93±25 <0.001 Sd-LDL/LDL ratio 0.23± ± ± ±0.08 <0.001 Blood pressure category, % <0.001 Optimal blood pressure Normal blood pressure High-normal blood pressure Hypertension grade I-III Antilipidemic drug use, % Diabetes, % <0.001 Current Smoking, % Current Drinking, %

10 3. Age- and Multivariate-adjusted Hazard Ratios for the incidence of CVD according to small dense LDL cholesterol quartiles Small dense LDL Cholesterol Q1 (Lower) Q2 Q3 Q4 (Higher) P for trend Men and women Person-years 5,593 5,773 5,597 5,739 Cardiovascular disease Case Age-adjusted ( ) 1.11 ( ) 1.64 ( ) Multivariable-adjusted ( ) 1.08 ( ) 1.60 ( ) Stroke Case Age-adjusted ( ) 0.80 ( ) 1.21 ( ) Multivariable-adjusted ( ) 0.79 ( ) 1.19 ( ) Cerebral infarction Case Age-adjusted ( ) 1.14 ( ) 1.74 ( ) Multivariable-adjusted ( ) 1.16 ( ) 1.85 ( ) Coronary heart disease Case Age-adjusted ( ) 2.26 ( ) 3.35 ( ) Multivariable-adjusted ( ) 2.17 ( ) 3.26 ( ) 0.003

11 4. Age- and Multivariate-adjusted Hazard Ratios for the incidence of CVD according to small dense LDL cholesterol quartiles in men Small dense LDL Cholesterol Q1 (Lower) Q2 Q3 Q4 (Higher) P for trend Men, range Person-years 1,990 2,488 2,457 3,306 Cardiovascular disease Case Age-adjusted ( ) 1.28 ( ) 1.62 ( ) Multivariable-adjusted ( ) 1.32 ( ) 1.85 ( ) Stroke Case Age-adjusted ( ) 0.99 ( ) 1.05 ( ) Multivariable-adjusted ( ) 1.11 ( ) 1.22 ( ) Cerebral infarction Case Age-adjusted ( ) 1.13 ( ) 1.30 ( ) Multivariable-adjusted ( ) 1.18 ( ) 1.43 ( ) Coronary heart disease Case Age-adjusted ( ) 2.28 ( ) 3.50 ( ) Multivariable-adjusted ( ) 2.16 ( ) 3.96 ( ) 0.004

12 5. Age- and Multivariate-adjusted Hazard Ratios for the incidence of CVD according to small dense LDL cholesterol quartiles in women Small dense LDL Cholesterol Q1 (Lower) Q2 Q3 Q4 (Higher) P for trend Women, range Person-years 3,603 3,285 3,140 2,432 Cardiovascular disease Case Age-adjusted ( ) 0.91 ( ) 1.77 ( ) Multivariable-adjusted ( ) 0.80 ( ) 1.41 ( ) Stroke Case Age-adjusted ( ) 0.61 ( ) 1.57 ( ) Multivariable-adjusted ( ) 0.57 ( ) 1.32 ( ) Cerebral infarction Case Age-adjusted ( ) 1.31 ( ) 3.91 ( ) Multivariable-adjusted Coronary heart disease Case Age-adjusted ( ) 2.25 ( ) 2.82 ( ) Multivariable-adjusted

13 6. Age- and Multivariate-adjusted Hazard Ratios for the incidence of CVD according to large dense LDL cholesterol quartiles Large buoyant LDL Q1 (Lower) Q2 Q3 Q4 (Higher) P for trend Men and women Person-years 5,707 5,828 6,053 5,934 Cardiovascular disease Case Age-adjusted ( ) 0.70 ( ) 1.00 ( ) Multivariable-adjusted ( ) 0.73 ( ) 1.01 ( ) Stroke Case Age-adjusted ( ) 0.61 ( ) 0.82 ( ) Multivariable-adjusted ( ) 0.68 ( ) 0.88 ( ) Cerebral infarction Case Age-adjusted ( ) 0.83 ( ) 0.79 ( ) Multivariable-adjusted ( ) 0.91 ( ) 0.83 ( ) Coronary heart disease Case Age-adjusted ( ) 0.83 ( ) 1.31 ( ) Multivariable-adjusted ( ) 0.81 ( ) 1.20 ( ) 0.801

14 7. Age- and Multivariate-adjusted Hazard Ratios for the incidence of CVD according to sdldl/ldl ratio quartiles Ratio of small dense LDL and LDL Cholesterol Q1 (Lower) Q2 Q3 Q4 (Higher) P for trend Men and women Person-years 5,627 5,735 5,646 5,625 Cardiovascular disease Case Age-adjusted ( ) 1.58 ( ) 2.82 ( ) <0.001 Multivariable-adjusted ( ) 1.59 ( ) 2.51 ( ) Stroke Case Age-adjusted ( ) 1.82 ( ) 2.88 ( ) Multivariable-adjusted ( ) 1.81 ( ) 2.46 ( ) Cerebral infarction Case Age-adjusted ( ) 3.02 ( ) 2.96 ( ) Multivariable-adjusted ( ) 2.93 ( ) 2.70 ( ) Coronary heart disease Case Age-adjusted ( ) 1.30 ( ) 2.86 ( ) Multivariable-adjusted ( ) 1.31 ( ) 2.76 ( ) 0.005

15 8. Age- and Multivariate-adjusted Hazard Ratios for the incidence of CVD according to sdldl/ldl ratio quartiles in men Ratio of small dense LDL and LDL Cholesterol Q1 (Lower) Q2 Q3 Q4 (Higher) P for trend Men Person-years 1,885 2,015 2,573 3,768 Cardiovascular disease Case Age-adjusted ( ) 1.46 ( ) 2.28 ( ) Multivariable-adjusted ( ) 1.64 ( ) 2.26 ( ) Stroke Case Age-adjusted ( ) 1.45 ( ) 1.67 ( ) Multivariable-adjusted ( ) 1.64 ( ) 1.64 ( ) Cerebral infarction Case Age-adjusted ( ) 2.06 ( ) 1.53 ( ) Multivariable-adjusted ( ) 2.23 ( ) 1.50 ( ) Coronary heart disease Case Age-adjusted ( ) 1.63 ( ) 4.01 ( ) Multivariable-adjusted ( ) 1.73 ( ) 4.38 ( ) 0.019

16 9. Age- and Multivariate-adjusted Hazard Ratios for the incidence of CVD according to sdldl/ldl ratio quartiles in women Ratio of small dense LDL and LDL Cholesterol Q1 (Lower) Q2 Q3 Q4 (Higher) P for trend Women Person-years 3,742 3,720 3,073 1,857 Cardiovascular disease Case Age-adjusted ( ) 1.71 ( ) 4.18 ( ) Multivariable-adjusted ( ) 1.74 ( ) 3.60 ( ) Stroke Case Age-adjusted ( ) 3.02 ( ) 9.04 ( ) Multivariable-adjusted Cerebral infarction Case Age-adjusted Multivariable-adjusted Coronary heart disease Case Age-adjusted ( ) 1.19 ( ) 2.23 ( ) Multivariable-adjusted ( ) 1.19 ( ) 1.58 ( ) 0.206

17 10. Age- and Multivariate-adjusted Hazard Ratios for the incidence of CVD according to LDL cholesterol quartiles LDL Cholesterol Q1 (Lower) Q2 Q3 Q4 (Higher) P for trend Men and women Person-years 5,591 5,675 5,715 5,684 Cardiovascular disease Case Age-adjusted ( ) 0.82 ( ) 1.34 ( ) Multivariable-adjusted ( ) 0.91 ( ) 1.30 ( ) Stroke Case Age-adjusted ( ) 0.62 ( ) 1.08 ( ) Multivariable-adjusted ( ) 0.71 ( ) 1.12 ( ) Cerebral infarction Case Age-adjusted ( ) 0.96 ( ) 1.11 ( ) Multivariable-adjusted ( ) 1.08 ( ) 1.15 ( ) Coronary heart disease Case Age-adjusted ( ) 1.28 ( ) 1.99 ( ) Multivariable-adjusted ( ) 1.41 ( ) 1.76 ( ) 0.107

18 11. Age- and Multivariate-adjusted Hazard Ratios for the incidence of CVD according to LDL cholesterol quartiles in men LDL Cholesterol Q1 (Lower) Q2 Q3 Q4 (Higher) P for trend Men Person-years 2,561 2,694 2,641 2,377 Cardiovascular disease Case Age-adjusted ( ) 1.08 ( ) 1.66 ( ) Multivariable-adjusted ( ) 1.26 ( ) 1.83 ( ) Stroke Case Age-adjusted ( ) 0.89 ( ) 1.11 ( ) Multivariable-adjusted ( ) 1.17 ( ) 1.42 ( ) Cerebral infarction Case Age-adjusted ( ) 1.10 ( ) 1.08 ( ) Multivariable-adjusted ( ) 1.30 ( ) 1.15 ( ) Coronary heart disease Case Age-adjusted ( ) 1.53 ( ) 2.98 ( ) Multivariable-adjusted ( ) 1.73 ( ) 2.96 ( ) 0.017

19 Summary This study showed an increased risk of CAD attributed to elevated sd-ldl-c concentrations in a Japanese population without previous history of CVD. Sd-LDL-C/LDL-C is also a significant predictor for CAD both in men and women.

20 Conclusions In this large urban cohort study conducted in Japan, we demonstrated that sd-ldl-c and sd-ldl-c/ldl-c are significantly associated with the development of CAD, providing evidence of sd-ldl-c as an important biomarker to predict CAD. A large intervention study is required to address the target goal of sd- LDL-C.