EURO PEAN SO CIETY O F CARDIOLOGY ®
Original scientific paper
Plasma homocysteine and coronary artery calcification in Korean men Byung Jin Kim, Bum Soo Kim and Jin Ho Kang
European Journal of Preventive Cardiology 2015, Vol. 22(4) 478–485 ! The European Society of Cardiology 2014 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav DOI: 10.1177/2047487314522136 ejpc.sagepub.com
Abstract Background: Homocysteine is an independent risk factor for atherosclerosis, plausibly through induction of endothelial dysfunction and smooth muscle cell proliferation. Under oxidative stress and inflammatory stimuli, vascular smooth muscle cells may undergo osteogenic differentiation, which leads to coronary artery calcification. This study evaluated the association between plasma homocysteine and coronary artery calcification. Design and methods: Coronary artery calcium scores (CACSs) and plasma homocysteine concentrations were measured in 21,235 men (42 6.5 years) who participated in the Kangbuk Samsung Health Study between 2010 and 2011. Subjects were grouped by quartile of plasma homocysteine. Results: The prevalence of coronary artery calcification (CAC) among the 21,235 men was 13.5%. In the first to fourth homocysteine quartiles, CAC(þ) prevalence rates were 12.1%, 12.6%, 13.9%, and 15.3%, respectively. The CAC(þ) group had unfavorable cardiometabolic and lipid profiles. In multivariate regression analysis after adjusting for variables with a univariate relationship (p < 0.20), the odds ratio (OR) for the presence of CAC was higher for the highest homocysteine quartile than for the lowest quartile group (OR (95% confidence interval (CI)), 1.275 (1.027, 1.583)), and presence of CAC was positively associated with quartile of homocysteine (p for trend ¼ 0.009). Moreover, absolute plasma homocysteine concentration was positively and significantly related to presence of CAC and to CACS, respectively (OR (95% CI) 1.399 (1.089, 1.796): standardized b ¼ 0.040, p < 0.001). Conclusions: This study shows that plasma homocysteine is independently related to coronary artery calcification in Korean men, suggesting that plasma homocysteine concentration may serve as a marker for CAC.
Keywords Homocysteine, coronary artery calcification, coronary calcium, cardiovascular risk Received 29 August 2013; accepted 11 January 2014
Introduction Although global cardiovascular risk assessments based on traditional risk factors have been widely used for the purpose of adequate treatment according to risk status and risk stratification, it has been reported that both the presence and severity of coronary artery calcification correlate with atherosclerotic plaque burden, and the total amount of coronary artery calcium provides information about future coronary artery disease events over and above the information provided by traditional risk factors.1–4 Thus, coronary artery calcium measurement has been widely used to detect early coronary heart disease. Although the precise mechanisms of the development of coronary artery calcification are not fully established, osteogenic differentiation in vascular smooth muscle cells may play an
important role, particularly in conjunction with oxidative stress.5–8 Homocysteine (Hcy) may independently influence atherosclerosis through induction of endothelial dysfunction and smooth muscle cell proliferation.9–11 These findings suggest a role for Hcy in coronary artery calcification; however, the few studies done to assess this relationship yield conflicting results.12–17 Division of Cardiology, Department of Internal Medicine, Kangbuk Samsung Hospital, Sungkyunkwan University School of Medicine, Seoul, Republic of Korea Corresponding author: Byung Jin Kim, 108 Pyung-dong, Jongro-gu, Seoul, 110-746 Republic of Korea. Email: [email protected]
Downloaded from cpr.sagepub.com at NORTH CAROLINA STATE UNIV on April 7, 2015
Kim et al.
479
Moreover, no study has yet tested the association between Hcy and coronary artery calcification in an Asian population. The present study was performed to evaluate the relationship between Hcy and coronary artery calcification in a group of Korean men, and to evaluate this association according to categories of traditional cardiovascular risk.
Methods Study population As participants in the Kangbuk Samsung Health Study, 26,231 Korean adults underwent coronary computed tomography to measure coronary artery calcium between 2010 and 2011. The Kangbuk Samsung Health Study is a retrospective cohort study of individuals who visited Kangbuk Samsung Health Promotion Center between 2002 and 2011. All participants in the study provided written informed consent. Of the 26,231 individuals who were screened, 4996 subjects were excluded from this study for the following reasons: 4793 subjects did not measure plasma Hcy; 203 subjects were women. The final analysis comprised 21,235 men.
Anthropometric and laboratory measurements Medical and medication histories, alcohol consumption (three times per week), smoking status (non-smokers versus ex and current smokers) and physical activity (three times per week) were assessed using standard questionnaires. Height and weight were measured using automatic instruments with participants wearing light clothing and no shoes. Body mass index (BMI) was calculated as weight in kilograms divided by height in meters squared. Waist circumference was measured at the mid-level between the lowest rib and the iliac crest with the subject standing and breathing normally. Systolic and diastolic blood pressures (SBP and DBP respectively) were measured by a trained nurse using a mercury sphygmomanometer on the right arm of subjects in a seated position after they had rested for five minutes or longer. Hypertension was defined as SBP 140 mmHg or DBP 90 mmHg or current use of medication for hypertension. Diabetes mellitus (DM) was defined as a fasting blood glucose of 126 mg/dl or greater, hemoglobin A1c (HbA1c) of 6.5% (48 mmol/mol) or greater, or current use of medication for diabetes. A morning blood sample was collected after at least 12 hours of fasting. Serum glucose, serum total cholesterol, triglycerides, low density lipoprotein (LDL) cholesterol, and high density lipoprotein (HDL) cholesterol concentrations were measured using an autoanalyzer
(Advia 1650 Autoanalyzer, Bayer Diagnostics, Leverkusen, Germany). Apolipoprotein A1 (Apo A1) and B (Apo B) measurements were performed with rate nephelometery (IMMAGE System, Beckman Coulter, CA, USA) and the intra- and inter-assay coefficients of variation were less than 2.5% for Apo A1 and Apo B. Fasting insulin concentration was measured by immunoradiometric assay (Biosource, Nivelles, Belgium) and HbA1c was measured using an immunoturbidimetric assay with a Cobra Integra 800 automatic analyzer (Roche Diagnostics, Basel, Switzerland) with a reference value range of 4.4% (25 mmol/mol) to 6.4% (46 mmol/mol). Serum creatinine and uric acid levels were determined using the Jaffe reaction method (Advia 1650 kit, Bayer Corp., PA, US) and the uricase EMST method, respectively. High sensitivity C-reactive protein (hs-CRP) concentration was measured using particle-enhanced immunonephelometry (Behring Nephelometer II, Dade Behring, Marburg, Germany), with a lower limit of detection of 0.175 mg/l after a 1:20 sample dilution for hs-CRP. The glomerular filtration rate (GFR) was calculated using the Modification of Diet in Renal Disease (MDRD) equation.18 Plasma total Hcy was assayed using a fluorescence polarization immunoassay (FPIA) with IMx Analyzers (Axsym Abbott Inc., Abbott Park, IL, USA) in two laboratories in an identical manner. The intra-assay coefficients of variance at the two laboratories were 4% and 2.6%, respectively, at a level of 7 mmol/l and 2% and 3.4% at a level of 12 mmol/l. The interlaboratory correlation was r ¼ 0.63.
Measurements of coronary artery calcium score Coronary artery calcium (CAC) was assessed by a nonenhanced 64-slice MDCT scanner (Lightspeed VCT XTE-64 slice, GE Healthcare, Tokyo, Japan). Image acquisition was electrocardiographically triggered at 70–75% of the R-R interval with prospective ECGgating. In the scanning protocol, section collimation was 16 mm 2.5 mm, rotation time 400 ms, tube voltage 120 kV, tube current 31 mAS (310 mA 0.1 s) and tomographic slice thickness 2.5 mm. Coronary artery calcification was defined as a hyperattenuated lesion above a threshold of 130 HU with an area of three or more adjacent pixels. Coronary artery calcium scores (CACSs) were calculated using the original Agatston method.19
Statistical analysis Data are expressed as mean standard deviation (SD) or median and interquartile range for continuous variables and as percentages (%) for categorical variables. Among the variables, serum triglyceride, hs-CRP,
Downloaded from cpr.sagepub.com at NORTH CAROLINA STATE UNIV on April 7, 2015
480
European Journal of Preventive Cardiology 22(4)
fasting insulin, homeostasis model assessment-insulin resistance (HOMA-IR) and Hcy concentrations were log-transformed for analysis to correct for skewed distributions. Values in the tables are expressed as untransformed data for ease of interpretation. Individuals were classified into two groups according to the presence or absence of coronary artery calcium, group I with CACS ¼ 0 and group II with CACS > 0. Also, they were categorized into quartiles according to plasma total Hcy concentration: 8.9, 8.91–10.40, 10.41–12.20, and > 12.20 mmol/l. The age-adjusted comparisons of characteristics between the two CACS groups were assessed using general linear model. The differences in characteristics among the quartile groups of Hcy levels were tested using one-way ANOVA or the Chi-square test. Multivariate regression analysis was performed to evaluate the effect of Hcy quartile or absolute plasma Hcy level on the presence of CAC after adjusting for variables with a univariate relationship (p < 0.20); these included age, lifestyle factors (alcohol and smoking), BMI, SBP, fasting glucose and insulin, HbA1c, HOMA-IR, blood urea nitrogen, MDRD-GFR, total cholesterol, triglyceride, HDL cholesterol, LDL cholesterol, hs-CRP and Framingham 10-year risk score (FRS). Subgroup analyses were performed according to the presence or absence of diabetes, or three cardiovascular risk groups (low risk, < 10%; intermediate risk, 10–20%; and high risk > 20% or diabetes) defined by FRS. Statistical analyses were performed using PASW version 18.0 (SPSS Inc., Chicago, IL, USA), at the 5% significance level. The Institutional Review Board of Kangbuk Samsung Hospital approved this research protocol.
Results The prevalence of CAC( þ ) was 13.5% in the entire study group of 21,235 men (mean age, 42 6.5 years), 31.5% in diabetic individuals, and 9.6%, 20.9% and 27.5%, respectively, in the low-, intermediate- and high-risk groups assembled according to the FRS categories. The comparisons of characteristics between group I (CACS ¼ 0) and group II (CACS > 0) according to the presence or absence of CAC are presented in Table 1. Individuals in group II were older, with greater BMI, more unfavorable cardiometabolic and lipid profiles, and higher blood pressure and FRS. Those in group II also had higher prevalence of alcohol consumption, ex and current smokers, diabetes and hypertension. The comparative analyses among the Hcy quartile groups showed that both the CACS and the prevalence of group II (CACS > 0) increased with increasing Hcy
quartile (CACS ¼ 6.8 52.2, 8.1 70.5, 9.9 61.4 and 15.2 134.0 in the first to fourth Hcy quartiles, respectively, p < 0.001: and CACS > 0 prevalence was 12.1%, 12.6%, 13.9% and 15.3%, respectively, p < 0.001) (Table 2). Cardiometabolic and lipid profiles, BMI, waist circumference, blood pressure and FRS differed significantly among the Hcy quartile groups. In the multivariate logistic regression analysis, the odds ratio (OR) for presence of CAC in the highest as compared with the lowest Hcy quartile group was 1.251 ((95% confidence interval (95% CI)), (1.010– 1.550)) and OR also increased significantly with increasing Hcy quartile (p ¼ 0.030 for trend) (Table 3). The increase in absolute serum Hcy level was independently and positively associated with OR for presence of CAC in the multivariate logistic regression model (1.347 (1.050–1.728), p ¼ 0.019). Among variables included in the multivariate logistic regression analysis, age (1.145 (1.129–1.161)), MDRD-GFR (1.019 (1.013–1.024)), HbA1c (1.808 (1.507–2.169)), BMI (1.058 (1.027– 1.091)), SBP (1.009 (1.002–1.015)), alcohol consumption (1.329 (1.127–1.568)) and ex and current smokers (1.281 (1.075, 1.527)) showed significant associations with presence of CAC. In the multivariate logistic regression model substituting apo A1 and apo B for total cholesterol, triglycerides, HDL cholesterol, and LDL cholesterol; waist circumference for BMI; and diastolic BP for systolic BP, all gave ORs for presence of CAC with respect to absolute Hcy level and Hcy quartile that were similar to those in the above original multivariate model (data not shown). Subgroup analyses according to the three Framingham 10-year risk groups revealed an association of absolute Hcy level with presence of CAC only in the high-risk groups (ORs (95% CI), 1.161 (0.836–1.613), 1.571 (0.910–2.714) and 2.061 (1.131–3.753) in the low-, intermediate-, and high-risk groups, respectively) (Table 4). Increasing quartile of Hcy was positively associated with presence of CAC only in the intermediateand high-risk groups (p for trend ¼ 0.790, 0.037 and 0.010 in the low-, intermediate- and high-risk groups). In the analysis according to presence or absence of diabetes, plasma Hcy was significantly associated with presence of CAC in only the DM(–) group (1.314 (1.002–1.722) in DM(–) group versus 1.568 (0.787– 3.124) in the DM(þ) group) (Table 4).
Discussion This large observational study showed an independent association of plasma Hcy with the presence and extent of CAC. In subgroup analysis, the association was significant in individuals at intermediate or high risk, and in non-diabetics.
Downloaded from cpr.sagepub.com at NORTH CAROLINA STATE UNIV on April 7, 2015
Kim et al.
481
Table 1. Clinical characteristics according to the absence or presence of coronary artery calcium.
Age, years Glucose, mmol/l Blood urea nitrogen, mmol/l Creatinine, mmol/l MDRD-GFR, ml/min Uric acid, mmol/l Total cholesterol, mmol/l Triglyceride, mmol/l HDL cholesterol, mmol/l LDL cholesterol, mmol/l Apolipoprotein B, g/l Apolipoprotein A1, g/l HbA1c, % (mmol/mol) Insulin, pmol/l HOMA-IR Homocysteine, mmol/l hs-CRP, mg/l Body mass index, kg/m2 Waist circumference, cm Systolic blood pressure, mmHg Diastolic blood pressure, mmHg CACS FRP FRS Alcohol, 3 times/week, n (%) Smoking, ex and current, n (%) Exercise, 3 times/week, n (%) Diabetes, n (%) Hypertension, n (%)
CACS ¼ 0 N ¼ 18,375
CACS > 0 N ¼ 2860
p value
41.2 6.0 5.4 0.9 4.8 1.1 87.5 11.5 91.1 13.9 362.8 71.4 5.31 0.91 1.41 (1.01, 2.02) 1.34 0.31 3.35 0.82 0.96 0.23 1.35 0.21 5.70 0.52 (39 5.7) 36.1 (24.3, 52.1) 1.24 (0.82, 1.84) 10.4 (8.9, 12.1) 0.6 (0.3, 1.1) 24.8 2.9 86.3 7.5 118 11.7 76 9.0 0 6.8 4.7 4.8 4.9 2559/10,679 (24.0) 6676/10,988 (60.8) 1436/10,669 (13.5) 924/18,375 (5.0) 2453/18,374 (13.4)
46.9 6.9 5.8 1.3 5.0 1.3 86.6 16.8 90.2 13.3 362.8 77.3 5.48 1.00 1.58 (1.12, 2.26) 1.30 0.31 3.50 0.90 1.01 0.23 1.33 0.21 5.94 0.80 (41 8.7) 38.2 (25.7, 57.6) 1.37 (0.89, 2.16) 10.6 (9.1, 12.5) 0.6 (0.4, 1.3) 25.4 2.9 88.0 7.6 120 12.6 78 9.7 19 (5, 60) 9.8 3.6 8.0 6.1 466/1415 (32.9) 1028/1459 (70.5) 270/14,156 (19.1) 421/2860 (14.7) 743/2859 (26.0)
Original scientific paper
Plasma homocysteine and coronary artery calcification in Korean men Byung Jin Kim, Bum Soo Kim and Jin Ho Kang
European Journal of Preventive Cardiology 2015, Vol. 22(4) 478–485 ! The European Society of Cardiology 2014 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav DOI: 10.1177/2047487314522136 ejpc.sagepub.com
Abstract Background: Homocysteine is an independent risk factor for atherosclerosis, plausibly through induction of endothelial dysfunction and smooth muscle cell proliferation. Under oxidative stress and inflammatory stimuli, vascular smooth muscle cells may undergo osteogenic differentiation, which leads to coronary artery calcification. This study evaluated the association between plasma homocysteine and coronary artery calcification. Design and methods: Coronary artery calcium scores (CACSs) and plasma homocysteine concentrations were measured in 21,235 men (42 6.5 years) who participated in the Kangbuk Samsung Health Study between 2010 and 2011. Subjects were grouped by quartile of plasma homocysteine. Results: The prevalence of coronary artery calcification (CAC) among the 21,235 men was 13.5%. In the first to fourth homocysteine quartiles, CAC(þ) prevalence rates were 12.1%, 12.6%, 13.9%, and 15.3%, respectively. The CAC(þ) group had unfavorable cardiometabolic and lipid profiles. In multivariate regression analysis after adjusting for variables with a univariate relationship (p < 0.20), the odds ratio (OR) for the presence of CAC was higher for the highest homocysteine quartile than for the lowest quartile group (OR (95% confidence interval (CI)), 1.275 (1.027, 1.583)), and presence of CAC was positively associated with quartile of homocysteine (p for trend ¼ 0.009). Moreover, absolute plasma homocysteine concentration was positively and significantly related to presence of CAC and to CACS, respectively (OR (95% CI) 1.399 (1.089, 1.796): standardized b ¼ 0.040, p < 0.001). Conclusions: This study shows that plasma homocysteine is independently related to coronary artery calcification in Korean men, suggesting that plasma homocysteine concentration may serve as a marker for CAC.
Keywords Homocysteine, coronary artery calcification, coronary calcium, cardiovascular risk Received 29 August 2013; accepted 11 January 2014
Introduction Although global cardiovascular risk assessments based on traditional risk factors have been widely used for the purpose of adequate treatment according to risk status and risk stratification, it has been reported that both the presence and severity of coronary artery calcification correlate with atherosclerotic plaque burden, and the total amount of coronary artery calcium provides information about future coronary artery disease events over and above the information provided by traditional risk factors.1–4 Thus, coronary artery calcium measurement has been widely used to detect early coronary heart disease. Although the precise mechanisms of the development of coronary artery calcification are not fully established, osteogenic differentiation in vascular smooth muscle cells may play an
important role, particularly in conjunction with oxidative stress.5–8 Homocysteine (Hcy) may independently influence atherosclerosis through induction of endothelial dysfunction and smooth muscle cell proliferation.9–11 These findings suggest a role for Hcy in coronary artery calcification; however, the few studies done to assess this relationship yield conflicting results.12–17 Division of Cardiology, Department of Internal Medicine, Kangbuk Samsung Hospital, Sungkyunkwan University School of Medicine, Seoul, Republic of Korea Corresponding author: Byung Jin Kim, 108 Pyung-dong, Jongro-gu, Seoul, 110-746 Republic of Korea. Email: [email protected]
Downloaded from cpr.sagepub.com at NORTH CAROLINA STATE UNIV on April 7, 2015
Kim et al.
479
Moreover, no study has yet tested the association between Hcy and coronary artery calcification in an Asian population. The present study was performed to evaluate the relationship between Hcy and coronary artery calcification in a group of Korean men, and to evaluate this association according to categories of traditional cardiovascular risk.
Methods Study population As participants in the Kangbuk Samsung Health Study, 26,231 Korean adults underwent coronary computed tomography to measure coronary artery calcium between 2010 and 2011. The Kangbuk Samsung Health Study is a retrospective cohort study of individuals who visited Kangbuk Samsung Health Promotion Center between 2002 and 2011. All participants in the study provided written informed consent. Of the 26,231 individuals who were screened, 4996 subjects were excluded from this study for the following reasons: 4793 subjects did not measure plasma Hcy; 203 subjects were women. The final analysis comprised 21,235 men.
Anthropometric and laboratory measurements Medical and medication histories, alcohol consumption (three times per week), smoking status (non-smokers versus ex and current smokers) and physical activity (three times per week) were assessed using standard questionnaires. Height and weight were measured using automatic instruments with participants wearing light clothing and no shoes. Body mass index (BMI) was calculated as weight in kilograms divided by height in meters squared. Waist circumference was measured at the mid-level between the lowest rib and the iliac crest with the subject standing and breathing normally. Systolic and diastolic blood pressures (SBP and DBP respectively) were measured by a trained nurse using a mercury sphygmomanometer on the right arm of subjects in a seated position after they had rested for five minutes or longer. Hypertension was defined as SBP 140 mmHg or DBP 90 mmHg or current use of medication for hypertension. Diabetes mellitus (DM) was defined as a fasting blood glucose of 126 mg/dl or greater, hemoglobin A1c (HbA1c) of 6.5% (48 mmol/mol) or greater, or current use of medication for diabetes. A morning blood sample was collected after at least 12 hours of fasting. Serum glucose, serum total cholesterol, triglycerides, low density lipoprotein (LDL) cholesterol, and high density lipoprotein (HDL) cholesterol concentrations were measured using an autoanalyzer
(Advia 1650 Autoanalyzer, Bayer Diagnostics, Leverkusen, Germany). Apolipoprotein A1 (Apo A1) and B (Apo B) measurements were performed with rate nephelometery (IMMAGE System, Beckman Coulter, CA, USA) and the intra- and inter-assay coefficients of variation were less than 2.5% for Apo A1 and Apo B. Fasting insulin concentration was measured by immunoradiometric assay (Biosource, Nivelles, Belgium) and HbA1c was measured using an immunoturbidimetric assay with a Cobra Integra 800 automatic analyzer (Roche Diagnostics, Basel, Switzerland) with a reference value range of 4.4% (25 mmol/mol) to 6.4% (46 mmol/mol). Serum creatinine and uric acid levels were determined using the Jaffe reaction method (Advia 1650 kit, Bayer Corp., PA, US) and the uricase EMST method, respectively. High sensitivity C-reactive protein (hs-CRP) concentration was measured using particle-enhanced immunonephelometry (Behring Nephelometer II, Dade Behring, Marburg, Germany), with a lower limit of detection of 0.175 mg/l after a 1:20 sample dilution for hs-CRP. The glomerular filtration rate (GFR) was calculated using the Modification of Diet in Renal Disease (MDRD) equation.18 Plasma total Hcy was assayed using a fluorescence polarization immunoassay (FPIA) with IMx Analyzers (Axsym Abbott Inc., Abbott Park, IL, USA) in two laboratories in an identical manner. The intra-assay coefficients of variance at the two laboratories were 4% and 2.6%, respectively, at a level of 7 mmol/l and 2% and 3.4% at a level of 12 mmol/l. The interlaboratory correlation was r ¼ 0.63.
Measurements of coronary artery calcium score Coronary artery calcium (CAC) was assessed by a nonenhanced 64-slice MDCT scanner (Lightspeed VCT XTE-64 slice, GE Healthcare, Tokyo, Japan). Image acquisition was electrocardiographically triggered at 70–75% of the R-R interval with prospective ECGgating. In the scanning protocol, section collimation was 16 mm 2.5 mm, rotation time 400 ms, tube voltage 120 kV, tube current 31 mAS (310 mA 0.1 s) and tomographic slice thickness 2.5 mm. Coronary artery calcification was defined as a hyperattenuated lesion above a threshold of 130 HU with an area of three or more adjacent pixels. Coronary artery calcium scores (CACSs) were calculated using the original Agatston method.19
Statistical analysis Data are expressed as mean standard deviation (SD) or median and interquartile range for continuous variables and as percentages (%) for categorical variables. Among the variables, serum triglyceride, hs-CRP,
Downloaded from cpr.sagepub.com at NORTH CAROLINA STATE UNIV on April 7, 2015
480
European Journal of Preventive Cardiology 22(4)
fasting insulin, homeostasis model assessment-insulin resistance (HOMA-IR) and Hcy concentrations were log-transformed for analysis to correct for skewed distributions. Values in the tables are expressed as untransformed data for ease of interpretation. Individuals were classified into two groups according to the presence or absence of coronary artery calcium, group I with CACS ¼ 0 and group II with CACS > 0. Also, they were categorized into quartiles according to plasma total Hcy concentration: 8.9, 8.91–10.40, 10.41–12.20, and > 12.20 mmol/l. The age-adjusted comparisons of characteristics between the two CACS groups were assessed using general linear model. The differences in characteristics among the quartile groups of Hcy levels were tested using one-way ANOVA or the Chi-square test. Multivariate regression analysis was performed to evaluate the effect of Hcy quartile or absolute plasma Hcy level on the presence of CAC after adjusting for variables with a univariate relationship (p < 0.20); these included age, lifestyle factors (alcohol and smoking), BMI, SBP, fasting glucose and insulin, HbA1c, HOMA-IR, blood urea nitrogen, MDRD-GFR, total cholesterol, triglyceride, HDL cholesterol, LDL cholesterol, hs-CRP and Framingham 10-year risk score (FRS). Subgroup analyses were performed according to the presence or absence of diabetes, or three cardiovascular risk groups (low risk, < 10%; intermediate risk, 10–20%; and high risk > 20% or diabetes) defined by FRS. Statistical analyses were performed using PASW version 18.0 (SPSS Inc., Chicago, IL, USA), at the 5% significance level. The Institutional Review Board of Kangbuk Samsung Hospital approved this research protocol.
Results The prevalence of CAC( þ ) was 13.5% in the entire study group of 21,235 men (mean age, 42 6.5 years), 31.5% in diabetic individuals, and 9.6%, 20.9% and 27.5%, respectively, in the low-, intermediate- and high-risk groups assembled according to the FRS categories. The comparisons of characteristics between group I (CACS ¼ 0) and group II (CACS > 0) according to the presence or absence of CAC are presented in Table 1. Individuals in group II were older, with greater BMI, more unfavorable cardiometabolic and lipid profiles, and higher blood pressure and FRS. Those in group II also had higher prevalence of alcohol consumption, ex and current smokers, diabetes and hypertension. The comparative analyses among the Hcy quartile groups showed that both the CACS and the prevalence of group II (CACS > 0) increased with increasing Hcy
quartile (CACS ¼ 6.8 52.2, 8.1 70.5, 9.9 61.4 and 15.2 134.0 in the first to fourth Hcy quartiles, respectively, p < 0.001: and CACS > 0 prevalence was 12.1%, 12.6%, 13.9% and 15.3%, respectively, p < 0.001) (Table 2). Cardiometabolic and lipid profiles, BMI, waist circumference, blood pressure and FRS differed significantly among the Hcy quartile groups. In the multivariate logistic regression analysis, the odds ratio (OR) for presence of CAC in the highest as compared with the lowest Hcy quartile group was 1.251 ((95% confidence interval (95% CI)), (1.010– 1.550)) and OR also increased significantly with increasing Hcy quartile (p ¼ 0.030 for trend) (Table 3). The increase in absolute serum Hcy level was independently and positively associated with OR for presence of CAC in the multivariate logistic regression model (1.347 (1.050–1.728), p ¼ 0.019). Among variables included in the multivariate logistic regression analysis, age (1.145 (1.129–1.161)), MDRD-GFR (1.019 (1.013–1.024)), HbA1c (1.808 (1.507–2.169)), BMI (1.058 (1.027– 1.091)), SBP (1.009 (1.002–1.015)), alcohol consumption (1.329 (1.127–1.568)) and ex and current smokers (1.281 (1.075, 1.527)) showed significant associations with presence of CAC. In the multivariate logistic regression model substituting apo A1 and apo B for total cholesterol, triglycerides, HDL cholesterol, and LDL cholesterol; waist circumference for BMI; and diastolic BP for systolic BP, all gave ORs for presence of CAC with respect to absolute Hcy level and Hcy quartile that were similar to those in the above original multivariate model (data not shown). Subgroup analyses according to the three Framingham 10-year risk groups revealed an association of absolute Hcy level with presence of CAC only in the high-risk groups (ORs (95% CI), 1.161 (0.836–1.613), 1.571 (0.910–2.714) and 2.061 (1.131–3.753) in the low-, intermediate-, and high-risk groups, respectively) (Table 4). Increasing quartile of Hcy was positively associated with presence of CAC only in the intermediateand high-risk groups (p for trend ¼ 0.790, 0.037 and 0.010 in the low-, intermediate- and high-risk groups). In the analysis according to presence or absence of diabetes, plasma Hcy was significantly associated with presence of CAC in only the DM(–) group (1.314 (1.002–1.722) in DM(–) group versus 1.568 (0.787– 3.124) in the DM(þ) group) (Table 4).
Discussion This large observational study showed an independent association of plasma Hcy with the presence and extent of CAC. In subgroup analysis, the association was significant in individuals at intermediate or high risk, and in non-diabetics.
Downloaded from cpr.sagepub.com at NORTH CAROLINA STATE UNIV on April 7, 2015
Kim et al.
481
Table 1. Clinical characteristics according to the absence or presence of coronary artery calcium.
Age, years Glucose, mmol/l Blood urea nitrogen, mmol/l Creatinine, mmol/l MDRD-GFR, ml/min Uric acid, mmol/l Total cholesterol, mmol/l Triglyceride, mmol/l HDL cholesterol, mmol/l LDL cholesterol, mmol/l Apolipoprotein B, g/l Apolipoprotein A1, g/l HbA1c, % (mmol/mol) Insulin, pmol/l HOMA-IR Homocysteine, mmol/l hs-CRP, mg/l Body mass index, kg/m2 Waist circumference, cm Systolic blood pressure, mmHg Diastolic blood pressure, mmHg CACS FRP FRS Alcohol, 3 times/week, n (%) Smoking, ex and current, n (%) Exercise, 3 times/week, n (%) Diabetes, n (%) Hypertension, n (%)
CACS ¼ 0 N ¼ 18,375
CACS > 0 N ¼ 2860
p value
41.2 6.0 5.4 0.9 4.8 1.1 87.5 11.5 91.1 13.9 362.8 71.4 5.31 0.91 1.41 (1.01, 2.02) 1.34 0.31 3.35 0.82 0.96 0.23 1.35 0.21 5.70 0.52 (39 5.7) 36.1 (24.3, 52.1) 1.24 (0.82, 1.84) 10.4 (8.9, 12.1) 0.6 (0.3, 1.1) 24.8 2.9 86.3 7.5 118 11.7 76 9.0 0 6.8 4.7 4.8 4.9 2559/10,679 (24.0) 6676/10,988 (60.8) 1436/10,669 (13.5) 924/18,375 (5.0) 2453/18,374 (13.4)
46.9 6.9 5.8 1.3 5.0 1.3 86.6 16.8 90.2 13.3 362.8 77.3 5.48 1.00 1.58 (1.12, 2.26) 1.30 0.31 3.50 0.90 1.01 0.23 1.33 0.21 5.94 0.80 (41 8.7) 38.2 (25.7, 57.6) 1.37 (0.89, 2.16) 10.6 (9.1, 12.5) 0.6 (0.4, 1.3) 25.4 2.9 88.0 7.6 120 12.6 78 9.7 19 (5, 60) 9.8 3.6 8.0 6.1 466/1415 (32.9) 1028/1459 (70.5) 270/14,156 (19.1) 421/2860 (14.7) 743/2859 (26.0)
