Original research 433
Combined use of apolipoprotein B/apolipoprotein A1 ratio and non-high-density lipoprotein cholesterol before routine clinical lipid measurement in predicting coronary heart disease Liting Pan, Guoping Lu and Zhenyue Chen Background and aims Our aim was to examine whether the combined use of apolipoprotein B (apoB)/ apolipoprotein A1 (apoA1) and non-high-density lipoprotein cholesterol (non-HDL-C) was useful before routine clinical lipid measurement in predicting coronary heart disease (CHD).
those in the bottom quartile. Patients with combined high levels of apoB/apoA1 and non-HDL-C (N = 92, 79.31%) had the highest risk of CHD. The combined use of apoB/apoA1 ratio and non-HDL-C (0.762; 95% CI 0.677–0.847) showed greater receiver operating characteristics area than its individual components or other lipid profiles.
Patients and methods In total, 826 patients were enrolled and they were classified into a CHD group (532 cases) and a normal group (294 cases) according to the results of coronary angiography. Laboratory data including fasting lipid profile were obtained after an overnight fast. Serum apoB/apoA1 ratio and non-HDL-C were calculated. Logistic regression was applied to estimate the cross-sectional association between the apoB/apoA1 ratio, non-HDL-C, and CHD. Receiver operating characteristics curve analysis was used to determine the value of apoB/apoA1 ratio and non-HDL-C in the diagnosis of CHD.
Conclusion The combination of apoB/apoA1 and non-HDL-C had even greater predictive value than its individual components or other lipid profiles. Coron Artery c 2014 Wolters Kluwer Health | Lippincott Dis 25:433–438 Williams & Wilkins.
Results The associations with an increased risk of CHD were much stronger for the apoB/apoA1 ratio [odds ratio (OR) = 8.941, 95% confidence interval (CI) 4.363–18.323] than for non-HDL-C (OR = 1.373, 95% CI 1.163–1.622). The patients in the top quartile of the apoB/apoA1 distribution had an OR of 7.321 (95% CI 3.891–13.771) compared with
Background Dyslipidemia plays an important role in the development of coronary heart disease (CHD). Low-density lipoprotein cholesterol (LDL-C) is a well-established risk factor for CHD and a primary target for lipid-lowering therapy for the prevention and treatment of cardiovascular disease [1]. However, several large, prospective epidemiological studies, including the AMORIS [2] and INTERHEART study [3], have shown that the apolipoprotein B (apoB)/ apolipoprotein A1 (apoA1) ratio, which reflects the cholesterol balance between potentially atherogenic and antiatherogenic lipoprotein particles, is a better predictor of cardiovascular disease than any of the traditional cholesterol indices [4,5]. In addition, there is growing evidence that non-high-density lipoprotein cholesterol [nonHDL-C; calculated as the difference between total cholesterol (TC) and HDL-C [6]], which reflects the TC content of atherogenic lipoproteins, is superior to LDL-C as a predictor of future cardiovascular events [7–10]. Although some studies have attempted to c 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins 0954-6928
Coronary Artery Disease 2014, 25:433–438 Keywords: apolipoprotein A1, apolipoprotein B, coronary heart disease, non-high-density lipoprotein cholesterol Department of Cardiology, Ruijin Hospital, Shanghai Jiaotong University School of Medicine, Shanghai, China Correspondence to Zhenyue Chen, MD, PhD, Department of Cardiology, Ruijin Hospital, Shanghai Jiaotong University School of Medicine, No.197, Ruijin 2nd Road, Shanghai 200025, China Tel: + 86 216 437 0045; fax: + 86 216 437 2614; e-mail: [email protected] Received 18 November 2013 Revised 30 January 2014 Accepted 2 February 2014
compare the superiority of LDL-C, apoB/apoA1, and nonHDL-C in predicting CHD, it still remains controversial [11,12]. Furthermore, none of the previous studies evaluated the predictive capacity of the combined use of apoB/apoA1 and non-HDL-C. Therefore, we carried out a retrospective study to examine whether the combined use of apoB/A1 and non-HDL-C is useful before routine clinical lipid measurement in predicting CHD.
Patients and methods Study population
We consecutively enrolled 826 patients (including 520 men and 306 women, age range 33–94 years) admitted to the Department of Cardiology, Shanghai Ruijin Hospital, for clinically indicated coronary angiography between 2008 and 2012. Patients with ongoing systemic inflammatory diseases, renal or hepatic dysfunction, significant valvular disease, myocarditis, cardiomyopathies, and malignancy were excluded from the study. The patients DOI: 10.1097/MCA.0000000000000100
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
434
Coronary Artery Disease 2014, Vol 25 No 5
were divided into two groups according to their coronary angiography results: 532 patients were classified into the CHD group (lesion Z 50% of at least one major branch) and the rest of the 294 patients with normal coronary arteries were included in the normal group. For further investigation, we also divided all patients into four groups according to the quartile of apoB or apoB/apoA1 ratio, and non-HDL-C quartile groups. ApoB (g/l): group A, men < 0.74, women < 0.74; group B, men 0.74 r apoB < 0.90, women 0.74 r apoB < 0.90; group C, men 0.90 r apoB < 1.06, women 0.90 r apoB < 1.06; group D, apoB men Z 1.06, women Z 1.06. ApoB/apoA1: group A, men < 0.62, women < 0.56; group B, men 0.62 r apoB/apoA1 < 0.76, women 0.56 r apoB/ apoA1 < 0.72; group C, men 0.76 r apoB/apoA1 < 0.96, women 0.72 r apoB/apoA1 < 0.89; Group D, men Z 0.96, women Z 0.89. Non-HDL-C: group A, men < 2.49, women < 2.67; group B, men 2.49 r non-HDL-C < 3.17, women 2.67 r non-HDLC < 3.36; group C, men 3.17 r non-HDL-C < 3.89, women 3.36 r non-HDL-C < 4.10; group D, men Z 3.89, women Z 4.10. Baseline investigation
Detailed information such as demographics, symptoms, cardiovascular risk factors, history of hypertension, diabetes mellitus, dyslipidemia, other cardiovascular diseases, and medication use was obtained at baseline. Height, weight, and vital signs were measured and BMI was calculated. Ultrasonic cardiography was performed using the Siements Acuson Sequoia C512 equipment (Siemens Medical Solutions USA Inc., Mountain View, California, USA).
Statistical analysis
Continuous parametric variables were expressed as mean±SD values. Parametric variables were compared between groups using the Student t-test and analysis of variance. Nonparametric variables were compared between groups using Dunnett’s T3 method. Categorical variables were presented as number and proportion [(n (%)] and differences in the study were tested using the w2-test. Multivariate logistic regression analyses (backward: LR) were used to identify the predictive value of independent factors for the presence of CHD. Receiver operating characteristics (ROC) curve analyses were also used to quantify the predictive value of independent parameters for the presence of CHD. All of the reported P values were two-tailed, and those less than 0.05 were considered to be statistically significant. All statistical analyses were carried out using the software package SPSS 13.0 (SPSS Inc., Chicago, Illinois, USA) for Windows.
Results Baseline characteristics
The mean age of the study group was 64.14±10.94 years, and 62.95% of the patients were men. Table 1 summarizes the baseline characteristics of the patients. There were more male patients in the CHD group than in the normal group (P < 0.001). Patients in the CHD group were older, and had higher BMI and systolic blood pressure than those in the normal group (P < 0.01) and they were more likely to have a history of smoking, hypertension, hypercholesterolemia, hypercholesterolemia, and diabetes (P < 0.001). There were no significant differences between the two groups in terms of diastolic blood pressure and heart rate (P > 0.05). Patients in the CHD group had higher non-HDL-C (P < 0.05), apoB (P < 0.05), apoB/apoA1 (P < 0.001), and lower HDL-C (P < 0.001) and apoA1 (P < 0.001). No significant differences were found between the two groups in TG, TC, and LDL-C.
Laboratory methodology
All venous blood samples drawn at 6:00 a.m. from all patients after an overnight fast were immediately refrigerated and transported to the department of Clinical Laboratory of Ruijin Hospital for measurements of TC, LDL-C, HDL-C, triglycerides (TGs), apoB, apoA1, fasting blood glucose, fasting insulin, ultrasensitive C-reactive protein, CK-MB, cTnI, and pro-BNP. The apoB/apoA1 ratio and non-HDL-C were then calculated. Coronary angiography
Coronary angiography was performed using the Judkins technique by INNOVA 20000 equipment (GE Healthcare, Waukesha, Wisconsin, USA), and all procedures followed the standard American College of Cardiology/ American Heart Association guidelines for coronary angiography. The coronary angiograms were evaluated by two experienced interventional cardiologists who were unaware of the patients’ biochemical status.
Association of the apolipoprotein B/apolipoprotein A1 ratio and non-high-density lipoprotein cholesterol with coronary heart disease
In the multivariate logistic regression analysis, apoB, apoB/apoA1 ratio, and non-HDL-C were all strongly associated with an increased risk of CHD after adjusting for sex, age, BMI, history of hypertension, diabetes, smoking, systolic blood pressure, diastolic blood pressure, and LDL-C. Furthermore, it also indicated that the associations were much stronger for apoB than for the apoB/apoA1 ratio or non-HDL-C [apoB: odds ratio (OR) = 11.285, 95% confidence interval (CI) 4.513–28.220; apoB/ apoA1: OR = 8.941, 95% CI 4.363–18.323; non-HDL-C: OR = 1.373, 95% CI 1.163–1.622; Table 2]. After adjusting for sex, age, BMI, history of hypertension, diabetes, smoking, systolic blood pressure, diastolic blood pressure, and LDL-C, patients in the top quartile of the apoB/apoA1 distribution had an OR of 7.321
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Combined use of apoB/apoA1 Pan et al. 435
Patient characteristics: coronary heart disease group and normal group
Table 1
Male [n (%)] Age (years) BMI (kg/m2) Systolic blood pressure (mmHg) Diastolic blood pressure (mmHg) Heart rate (bpm) Smoking [n (%)] Hypertension [n (%)] Hypercholesterolemia [n (%)] Diabetes [n (%)] TG (mmol/l) TC (mmol/l) HDL-C (mmol/l) LDL-C (mmol/l) Non-HDL-C (mmol/l) ApoA1 (g/l) ApoB (g/l) ApoB/apoA1
Normal group (n = 294)
CHD group (n = 532)
P value
153 (52.04) 60.88±9.80 24.29±2.96 126.56±17.46
367 (68.99) 65.94±11.14 25.06±3.39 130.29±20.18
0.000*** 0.000*** 0.006** 0.008**
77.16±9.54
76.46±11.16
0.368
76.09±11.83 90 (30.61) 180 (61.22) 31 (10.54)
75.64±10.86 248 (46.62) 372 (69.93) 86 (16.17)
0.105 0.000*** 0.000*** 0.000***
66 (22.45) 1.78±1.17 4.44±0.98 1.21±0.33 2.66±0.83 3.22±0.93 1.30±0.33 0.88±0.22 0.71±0.23
249 (46.81) 1.73±1.15 4.46±1.08 1.09±0.30 2.74±0.92 3.37±1.05 1.17±0.31 0.93±0.25 0.84±0.29
0.000*** 0.505 0.741 0.000*** 0.205 0.049* 0.000*** 0.011* 0.000***
apoA1, apolipoprotein A1; apoB, apolipoprotein B; CHD, coronary heart disease; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; TC, total cholesterol; TG, triglyceride. Compared with the normal group: *P < 0.05; **P < 0.01; ***P < 0.001.
Table 3 Logistic regression analysis for the association of the quartile groups of apolipoprotein B/apolipoprotein A1 with coronary heart disease 95% CI Regression coefficient A B C D Constant
0.526 1.463 1.991 – 2.159
Standard error
P value
0.254 0.296 0.322 1.337
0.000 0.038 0.000 0.000 0.106
OR
Lower Upper
1 1.693 1.029 2.784 4.317 2.419 7.705 7.321 3.891 13.771 0.115
CI, confidence interval; OR, odds ratio.
Table 4 Logistic regression analysis for the association of the quartile groups of apolipoprotein B with coronary heart disease 95% CI
A B C D Constant
Regression coefficient
Standard error
P value
0.958 0.972 1.412 – 0.694
0.275 0.275 0.288 1.136
0.000 0.000 0.000 0.000 0.541
OR
Lower Upper
1 2.606 1.521 4.465 2.642 1.540 4.534 4.105 2.332 7.225 0.500
CI, confidence interval; OR, odds ratio.
Multivariate logistic regression analysis for the association of the apolipoprotein B/apolipoprotein A1 ratio and non-high-density lipoprotein cholesterol with coronary heart disease
Table 2
Table 5 Logistic regression analysis for the association of the quartile groups of non-high-density lipoprotein cholesterol with coronary heart disease 95% CI
95% CI
ApoB ApoA1 ApoB/apoA1 Non-HDL-C
Regression coefficient
Standard error
P value
OR
Lower
Upper
2.423 – 1.836 2.191 0.317
0.468 0.334 0.366 0.085
0.000 0.000 0.000 0.000
11.285 0.159 8.941 1.373
4.513 0.083 4.363 1.163
28.220 0.307 18.323 1.622
apoA1, apolipoprotein A1; apoB, apolipoprotein B; CI, confidence interval; HDL-C, high-density lipoprotein cholesterol; OR, odds ratio.
(95% CI 3.891–13.771; Table 3) compared with those in the bottom quartile, and patients in the top quartile of apoB had an OR of 4.105 (95% CI 2.332–7.225; Table 4) compared with those in the bottom quartile, whereas patients in the top quartile of non-HDL-C had an OR of 2.427 (95% CI 1.407–4.187; Table 5) compared with those in the bottom quartile.
Combined use of apolipoprotein B/apolipoprotein A1 and non-high-density lipoprotein cholesterol in predicting coronary heart disease
In our study, the top quartile of the apoB/apoA1 group was defined as the high apoB/apoA1 ratio group (apoB/ apoA1: Z 0.96 in men and Z 0.89 in women) and the top quartile of the non-HDL-C group was defined as the high non-HDL-C group (non-HDL-C: Z 3.89 mmol/l in men and Z 4.10 mmol/l in women). All of the participants were further divided into four groups on the basis of the results of apoB/apoA1 and non-HDL-C as follows:
Regression coefficient A B C D Constant
0.292 0.554 0.887 – 1.550
Standard error
P value
0.262 0.277 0.278 1.254
0.012 0.265 0.046 0.001 0.216
OR
Lower Upper
1 1.339 0.802 2.236 1.740 1.011 2.994 2.427 1.407 4.187 0.212
CI, confidence interval; OR, odds ratio.
group A, low apoB/apoA1 ratio/low non-HDL-C; group B, low apoB/apoA1 ratio/high non-HDL-C; group C, high apoB/apoA1 ratio/low non-HDL-C; and group D, high apoB/apoA1 ratio/high non-HDL-C. The incidence rate of CHD in the high apoB/apoA1 group was much higher than that in the low apoB/apoA1 group (78.85 vs. 59.55%, P < 0.001). In addition, compared with group 1, patients in groups 3 and 4 had a higher incidence rate of CHD (group 3 or 4 vs. group 1, P < 0.001; group 3 or 4 vs. group 2, P < 0.01). Patients with the combined high levels of apoB/apoA1 and non-HDL-C (group 4) had the highest risk of CHD (79.31%; Table 6). Apolipoprotein B/apolipoprotein A1, non-high-density lipoprotein cholesterol, and the receiver operating characteristics curve of coronary heart disease
We used the area under the curve (AUC) of ROC curve to determine the predictive values of various lipid markers.
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Coronary Artery Disease 2014, Vol 25 No 5
Table 6 Combined use of apolipoprotein B/apolipoprotein A1 and non-high-density lipoprotein cholesterol in predicting coronary heart disease Low apoB/apoA1 ratio (N = 618) Group 1 Low nonHDL-C (N = 524)
Fig. 1
1.0
High apoB/apoA1 ratio (N = 208)
Group 2 High nonHDL-C (N = 94)
Group 3 Low nonHDL-C (N = 92)
0.8
Group 4 High nonHDL-C (N = 116)
Incidence rate of 311 (59.35) 57 (60.64) 72 (78.26)***## 92 (79.31)***## CHD [N (%)] apoA1, apolipoprotein A1; apoB, apolipoprotein B; CHD, coronary heart disease; non-HDL-C, non-high-density lipoprotein cholesterol. Compared with group 1: ***P < 0.001. Compared with group 2: ##P < 0.01.
Sensitivity
436
0.6
0.4
0.2 Area under the receiver operating characteristics curve for coronary heart disease Table 7
95% CI
Combined use of the apoB/apoA1 ratio and non-HDL-C ApoB/apoA1 Non-HDL-C ApoA1 ApoB TG TC HDL-C LDL-C
0.0 0.0
Area under the ROC curve
Standard error
P value
Lower
Upper
0.762
0.043
0.000
0.677
0.847
0.757 0.619 0.307 0.663 0.525 0.586 0.378 0.614
0.043 0.055 0.053 0.051 0.057 0.056 0.052 0.057
0.000 0.046 0.001 0.006 0.668 0.147 0.039 0.055
0.674 0.511 0.203 0.563 0.415 0.475 0.277 0.502
0.841 0.727 0.410 0.763 0.636 0.697 0.479 0.726
apoA1, apolipoprotein A1; apoB, apolipoprotein B; CI, confidence interval; LDL-C, low-density lipoprotein cholesterol; non-HDL-C, non-high-density lipoprotein cholesterol; ROC, receiver operating characteristics; TC, total cholesterol; TG, triglyceride.
The combined use of the apoB/apoA1 ratio and nonHDL-C showed a greater ROC area than the apoB/apoA1 ratio, apoB, non-HDL-C, or other lipid profiles (Table 7 and Fig. 1).
Discussion In the present study, higher levels of apoB, apoB/apoA1 ratio, and non-HDL-C were all strongly and independently associated with an increased risk of CHD after adjusting for sex, age, BMI, hypertension, diabetes, smoking, blood pressure, and other lipid profiles. These profiles appear to be much stronger predictors of the risk of CHD than other routine lipid profiles, which are consistent with previous studies [7]. We also found that after adjustment for matching factors, the relative risk of CHD in the highest quartile compared with the lowest quartile was 7.321 (95% CI 3.891–13.771) for apoB/apoA1 and 4.105 (95% CI 2.332–7.225) for apoB, and 2.427 (95% CI 1.407–4.187) for non-HDL-C, which was the similar to Chien et al.’s [13] findings. Through ROC analysis, our study also showed that an elevated apoB/apoA1 ratio
0.2
0.4 0.6 1 − Specificity
0.8
1.0
Combined use of apoB/apoA1 and non-HDL-C
apoB
Non-HDL-C
Reference line
apoB/apoA1
Receiver operating characteristics curve of serum apolipoprotein B (apoB), apoB/apolipoprotein A1 (apoA1), non-high-density lipoprotein cholesterol (non-HDL-C), and combined factors for the diagnosis of coronary heart disease.
(AUC = 0.757) is more useful in predicting a greater risk of CHD than apoB (AUC = 0.663), non-HDL-C (AUC = 0.619), or other traditional lipid parameters. Chien and colleagues also found that the ROC area of apoB/apoA1 (AUC = 0.66) for the diagnosis of CHD was greater than apoB (AUC = 0.63), non-HDL-C (AUC = 0.60), LDL-C (AUC = 0.60), or other lipid profiles. In the current study, we could reproduce this finding, but we also compared the predictive value of the combined use of apoB/apoA1 and non-HDL-C. As expected, the area of combined use was greater than its individual components or other lipid profiles, suggesting that the combination of apoB/apoA1 and non-HDL-C was a stronger predictor for the risk of CHD. However, on comparison of the four subgroups, our study also showed that high ratios of apoB/ apoA1 in combination with high levels of non-HDL-C may further enhance their ability to predict CHD. ApoB is essential for the binding of LDL-C particles, allowing cells to internalize LDL-C and thus absorb cholesterol, leading to cardiovascular disease more easily, so the level of apoB represents the number of atherogenic particles. The European guidelines on Cardiovascular Disease Prevention in Clinical Practice
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Combined use of apoB/apoA1 Pan et al. 437
(version 2012) [14] suggested that apoB seemed to be a risk marker similar to LDL-C and a better index of the adequacy of LDL-lowering therapy. The guideline recommended that it should be less than 80 mg/dl and less than 100 mg/dl for patients with very high or high risk of cardiovascular disease, respectively; whereas, nonHDL-C comprises all the cholesterol contained in LDL, very low-density lipoprotein (VLDL) and intermediate-density lipoprotein particles. It can be calculated by simply subtracting HDL-C from TC. Thus, it is the cholesterol in the atherogenic particles and has been recommended as a target especially in patients with high nonfasting TG levels by the National Cholesterol Education Program guidelines (NCEP ATP III 2002) [15]. An elevated apoB has been shown to have the best predictive ability for measurable quantities of the atherogenic small, dense LDL particles, which can penetrate into the arterial wall and become deposited under the vascular endothelium more easily than larger but looser LDL particles because of their small size [16]. Therefore, some studies suggested that the atherogenic risk resulting from lipoproteins does not relate primarily to their cholesterol content, but rather their size and number [17–19]. This may be the reason why the associations with an increased risk of CHD were much stronger for apoB than for non-HDL-C, which are consistent with previous studies [7,11,12]. As mentioned above, apoB represents the number of atherogenic particles, whereas levels of apoA1 are strongly correlated with those of HDL-C, and expression of apoA1 may be largely responsible for determining the plasma level of HDL-C. It has long been known that a low level of HDL-C is a powerful predictor of increased cardiovascular risk. Although recent prospective research failed to prove that increasing the level of HDL-C could reduce the incidence rate of CHD, many epidemiological studies have found that patients with high levels of HDL-C have a lower incidence rate of CHD [20,21]. Therefore, the apoB/apoA1 ratio mirrors the balance between proatherogenic and antiatherogenic lipoprotein particles, which is why apoB/apoA1 was superior to nonHDL-C or other traditional lipid profiles in predicting the risk of CHD as we found in the present study. However, in the multivariate logistic regression, we found that apoB was more strongly associated with an increased risk of CHD than the apoB/apoA1 ratio after adjustment (apoB: OR = 11.285, 95% CI 4.513–28.220; apoB/apoA1: OR = 8.941, 95% CI 4.363–18.323), which was not in agreement with The INTERHEART study [3], which showed that the apoB/apoA1 ratio had a stronger association with the risk of myocardial infarction (OR = 1.59, 95% CI 1.52–1.64) than apoB (OR = 1.32, 95% CI 1.28–1.36). The reason for this discrepancy could be that they mainly focused on patients with acute myocardial infarction rather than all of the CHD patients. Furthermore, the ethnicity and the scale of the study may have
also affected the results. However, no previous studies have investigated the predictive value of the combination of apoB/apoA1 and non-HDL-C. Our study found that the combination of apoB/apoA1 and non-HDL-C had greater predictive value than its individual components or other lipid profiles. TG-rich lipoproteins such as chylomicron emulsion or VLDL would exchange their contents with cholesterol-rich LDL, and during this procedure not only sd-LDL but also cholesterol-rich VLDL residues would be produced, which are also proatherogenic. NonHDL-C is the mass of cholesterol in the atherogenic particles and the secondary lipid-lowering target in patients with elevated serum TGs (NCEP ATP III). Elevated non-HDL-C represents a high concentration of cholesterol in TG-enriched lipoproteins and their particles, whereas the level of LDL-C remains normal. Whereas the apoB/apoA1 ratio represents the balance of proatherogenic and antiatherogenic lipoprotein particles, naturally, the combined use of these two profiles have a higher predictive value for the risk of CHD. Limitations
Our study, however, has some limitations. First, our study mainly focused on the CHD patients whose lesion of coronary artery was 50% or more in at least one major branch according to the diagnostic criteria from the guideline of American College of Cardiology/American Heart Association. Although increasingly more studies [22–26] have reported that mild coronary stenosis may have higher potential of getting vulnerable plaque in its coronary arteries, which easily leading to acute coronary syndrome, we were also aware of the importance of patients with mild coronary stenosis, and relevant studies are ongoing in our work group. However, in this study, we have only focused on the CHD group. Second, the patients were not from a general population, but rather visitors to our hospital, which may have led to a selection bias. Therefore, this study may not be an accurate reflection on the entire population in China. Because of the small sample size, our ability to investigate the predictive power of the ratio and non-HDL-C was limited. Therefore, further research on a large scale and with more participants is required. Conclusion
We clearly show that apoB, apoB/apoA1 ratio, and nonHDL-C were superior to other routine lipid profiles in predicting the risk of CHD, and the ratio was the best predictor among them. However, the combination of apoB/apoA1 and non-HDL-C had even greater predictive value than its individual components or other lipid profiles.
Acknowledgements This work was supported by the National Natural Science Foundation of China (81370331), the Shanghai Municipal
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
438
Coronary Artery Disease 2014, Vol 25 No 5
Science and Technology Commission (12140903402), and the Young Medical Talents Training Program of Shanghai Municipal Health Bureau (XYQ2011011) to Z. Chen, and Shanghai Municipal Science and Technology Commission (10JC1410502) to G. Lu.
12
Conflicts of interest
15
13
14
There are no conflicts of interest.
References 1
Expert Panel. 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) final report. Circulation 2002; 106:3143–3421. 2 Walldius G, Jungner I, Holme I, Aastveit AH, Kolar W, Steiner E. High apolipoprotein B, low apolipoprotein A-I, and improvement in the prediction of fatal myocardial infarction (AMORIS study): a prospective study. Lancet 2001; 358:2026–2033. 3 Yusuf S, Hawken S, Ounpuu S, Dans T, Avezum A, Lanas F, et al. Effect of potentially modifiable risk factors associated with myocardial infarction in 52 countries (the INTERHEART study): case–control study. Lancet 2004; 364:937–952. 4 Walldius G, Jungner I. The apoB/apoA-I ratio: a strong, new risk factor for cardiovascular disease and a target for lipid-lowering therapy – a review of the evidence. J Intern Med 2006; 259:493–519. 5 Thompson A, Danesh J. Associations between apolipoprotein B, apolipoprotein AI, the apolipoprotein B/AI ratio and coronary heart disease: a literature-based meta-analysis of prospective studies. J Intern Med 2006; 259:481–492. 6 Miller M, Ginsberg HN, Schaefer EJ. Relative atherogenicity and predictive value of non-high-density lipoprotein cholesterol for coronary heart disease. Am J Cardiol 2008; 101:1003–1008. 7 Robinson JG, Wang S, Smith BJ, Jacobson TA. Meta-analysis of the relationship between non-high-density lipoprotein cholesterol reduction and coronary heart disease risk. J Am Coll Cardiol 2009; 53:316–322. 8 Rosenson RS. Management of non-high-density lipoprotein abnormalities. Atherosclerosis 2009; 207:328–335. 9 Lu W, Resnick HE, Jablonski KA, Jones KL, Jain AK, Howard WJ, et al. Non-HDL cholesterol as a predictor of cardiovascular disease in type 2 diabetes: the Strong Heart study. Diabetes Care 2003; 26:16–23. 10 Jiang R, Schulze MB, Li T, Rifai N, Stampfer MJ, Rimm EB, Hu FB. Non-HDL cholesterol and apolipoprotein B predict cardiovascular disease events among men with type 2 diabetes. Diabetes Care 2004; 27:1991–1997. 11 Denke MA. Weighing in before the fight: low-density lipoprotein cholesterol and non-high-density lipoprotein cholesterol versus apolipoprotein B as the best predictor for coronary heart disease and the best measure of therapy. Circulation 2005; 112:3368–3370.
16
17 18
19
20
21
22
23
24
25
26
Sniderman AD. Apolipoprotein B versus non-high-density lipoprotein cholesterol: and the winner is y. Circulation 2005; 112:3366–3367. Chien KL, Hsu HC, Su TC, Chen MF, Lee YT, Hu FB. Apolipoprotein B and non-high density lipoprotein cholesterol and the risk of coronary heart disease in Chinese. J Lipid Res 2007; 48:2499–2505. Perk J, De Backer G, Gohlke H, Graham I, Reiner Z, Verschuren WM, et al. European Guidelines on cardiovascular disease prevention in clinical practice (version 2012). Eur Heart J 2012; 33:1635–1701. National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, andTreatment of High Blood Cholesterol in Adults (Adult Treatment Panel III). 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) final report. Circulation 2002; 106:3143–3421. Tsimihodimos V, Gazi I, Kostara C, Tselepis AD, Elisaf M. Plasma lipoproteins and triacylglycerol are predictors of small, dense LDL particles. Lipids 2007; 42:403–409. Sacks FM. The apolipoprotein story. Atheroscler Suppl 2006; 7:23–27. Pischon T, Girman CJ, Sacks FM, Rifai N, Stampfer MJ, Rimm EB. Non-highdensity lipoprotein cholesterol and apolipoprotein B in the prediction of coronary heart disease in men. Circulation 2005; 112:3375–3383. Sniderman A, Williams K, Haffner S, Sattar N. Insights from apoB: from better diagnosis and therapy to the Medusa hypothesis. Atheroscler Suppl 2004; 5:19–24. Schwartz GG, et al. dal-OUTCOMES Investigators. Effects of dalcetrapib in patients with a recent acute coronary syndrome. N Engl J Med 2012; 367:2089–2099. Al-Hijji M, et al. Effect of equivalent on-treatment apolipoprotein levels on outcomes (from the AIM-HIGH and HPS2-THRIVE). Am J Cardiol 2013; 112:1697–1700. Virmani R, Kolodgie FD, Burke AP, Finn AV, Gold HK, Tulenko TN, et al. Atherosclerotic plaque progression and vulnerability to rupture: angiogenesis as a source of intraplaque hemorrhage. Arterioscler Thromb Vasc Biol 2005; 25:2054–2061. Calvert PA, Obaid DR, O’Sullivan M, Shapiro LM, McNab D, Densem CG, et al. Association between IVUS findings and adverse outcomes in patients with coronary artery disease: the VIVA (VH-IVUS in Vulnerable Atherosclerosis) study. JACC Cardiovasc Imaging 2011; 4:894–901. Hong MK, Mintz GS, Lee CW, Lee JW, Park JH, Park DW, et al. A three-vessel virtual histology intravascular ultrasound analysis of frequency and distribution of thin-cap fibroatheromas in patients with acute coronary syndrome or stable angina pectoris. Am J Cardiol 2008; 101: 568–572. Barlis P, Serruys PW, Gonzalo N, van der Giessen WJ, de Jaegere PJ, Regar E. Assessment of culprit and remote coronary narrowings using optical coherence tomography with long-term outcomes. Am J Cardiol 2008; 102:391–395. Halvorsen B, Otterdal K, Dahl TB, Skjelland M, Gullestad L, Øie E, Aukrust P. Atherosclerotic plaque stability – what determines the fate of a plaque? Prog Cardiovasc Dis 2008; 51:183–194.
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Combined use of apolipoprotein B/apolipoprotein A1 ratio and non-high-density lipoprotein cholesterol before routine clinical lipid measurement in predicting coronary heart disease Liting Pan, Guoping Lu and Zhenyue Chen Background and aims Our aim was to examine whether the combined use of apolipoprotein B (apoB)/ apolipoprotein A1 (apoA1) and non-high-density lipoprotein cholesterol (non-HDL-C) was useful before routine clinical lipid measurement in predicting coronary heart disease (CHD).
those in the bottom quartile. Patients with combined high levels of apoB/apoA1 and non-HDL-C (N = 92, 79.31%) had the highest risk of CHD. The combined use of apoB/apoA1 ratio and non-HDL-C (0.762; 95% CI 0.677–0.847) showed greater receiver operating characteristics area than its individual components or other lipid profiles.
Patients and methods In total, 826 patients were enrolled and they were classified into a CHD group (532 cases) and a normal group (294 cases) according to the results of coronary angiography. Laboratory data including fasting lipid profile were obtained after an overnight fast. Serum apoB/apoA1 ratio and non-HDL-C were calculated. Logistic regression was applied to estimate the cross-sectional association between the apoB/apoA1 ratio, non-HDL-C, and CHD. Receiver operating characteristics curve analysis was used to determine the value of apoB/apoA1 ratio and non-HDL-C in the diagnosis of CHD.
Conclusion The combination of apoB/apoA1 and non-HDL-C had even greater predictive value than its individual components or other lipid profiles. Coron Artery c 2014 Wolters Kluwer Health | Lippincott Dis 25:433–438 Williams & Wilkins.
Results The associations with an increased risk of CHD were much stronger for the apoB/apoA1 ratio [odds ratio (OR) = 8.941, 95% confidence interval (CI) 4.363–18.323] than for non-HDL-C (OR = 1.373, 95% CI 1.163–1.622). The patients in the top quartile of the apoB/apoA1 distribution had an OR of 7.321 (95% CI 3.891–13.771) compared with
Background Dyslipidemia plays an important role in the development of coronary heart disease (CHD). Low-density lipoprotein cholesterol (LDL-C) is a well-established risk factor for CHD and a primary target for lipid-lowering therapy for the prevention and treatment of cardiovascular disease [1]. However, several large, prospective epidemiological studies, including the AMORIS [2] and INTERHEART study [3], have shown that the apolipoprotein B (apoB)/ apolipoprotein A1 (apoA1) ratio, which reflects the cholesterol balance between potentially atherogenic and antiatherogenic lipoprotein particles, is a better predictor of cardiovascular disease than any of the traditional cholesterol indices [4,5]. In addition, there is growing evidence that non-high-density lipoprotein cholesterol [nonHDL-C; calculated as the difference between total cholesterol (TC) and HDL-C [6]], which reflects the TC content of atherogenic lipoproteins, is superior to LDL-C as a predictor of future cardiovascular events [7–10]. Although some studies have attempted to c 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins 0954-6928
Coronary Artery Disease 2014, 25:433–438 Keywords: apolipoprotein A1, apolipoprotein B, coronary heart disease, non-high-density lipoprotein cholesterol Department of Cardiology, Ruijin Hospital, Shanghai Jiaotong University School of Medicine, Shanghai, China Correspondence to Zhenyue Chen, MD, PhD, Department of Cardiology, Ruijin Hospital, Shanghai Jiaotong University School of Medicine, No.197, Ruijin 2nd Road, Shanghai 200025, China Tel: + 86 216 437 0045; fax: + 86 216 437 2614; e-mail: [email protected] Received 18 November 2013 Revised 30 January 2014 Accepted 2 February 2014
compare the superiority of LDL-C, apoB/apoA1, and nonHDL-C in predicting CHD, it still remains controversial [11,12]. Furthermore, none of the previous studies evaluated the predictive capacity of the combined use of apoB/apoA1 and non-HDL-C. Therefore, we carried out a retrospective study to examine whether the combined use of apoB/A1 and non-HDL-C is useful before routine clinical lipid measurement in predicting CHD.
Patients and methods Study population
We consecutively enrolled 826 patients (including 520 men and 306 women, age range 33–94 years) admitted to the Department of Cardiology, Shanghai Ruijin Hospital, for clinically indicated coronary angiography between 2008 and 2012. Patients with ongoing systemic inflammatory diseases, renal or hepatic dysfunction, significant valvular disease, myocarditis, cardiomyopathies, and malignancy were excluded from the study. The patients DOI: 10.1097/MCA.0000000000000100
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
434
Coronary Artery Disease 2014, Vol 25 No 5
were divided into two groups according to their coronary angiography results: 532 patients were classified into the CHD group (lesion Z 50% of at least one major branch) and the rest of the 294 patients with normal coronary arteries were included in the normal group. For further investigation, we also divided all patients into four groups according to the quartile of apoB or apoB/apoA1 ratio, and non-HDL-C quartile groups. ApoB (g/l): group A, men < 0.74, women < 0.74; group B, men 0.74 r apoB < 0.90, women 0.74 r apoB < 0.90; group C, men 0.90 r apoB < 1.06, women 0.90 r apoB < 1.06; group D, apoB men Z 1.06, women Z 1.06. ApoB/apoA1: group A, men < 0.62, women < 0.56; group B, men 0.62 r apoB/apoA1 < 0.76, women 0.56 r apoB/ apoA1 < 0.72; group C, men 0.76 r apoB/apoA1 < 0.96, women 0.72 r apoB/apoA1 < 0.89; Group D, men Z 0.96, women Z 0.89. Non-HDL-C: group A, men < 2.49, women < 2.67; group B, men 2.49 r non-HDL-C < 3.17, women 2.67 r non-HDLC < 3.36; group C, men 3.17 r non-HDL-C < 3.89, women 3.36 r non-HDL-C < 4.10; group D, men Z 3.89, women Z 4.10. Baseline investigation
Detailed information such as demographics, symptoms, cardiovascular risk factors, history of hypertension, diabetes mellitus, dyslipidemia, other cardiovascular diseases, and medication use was obtained at baseline. Height, weight, and vital signs were measured and BMI was calculated. Ultrasonic cardiography was performed using the Siements Acuson Sequoia C512 equipment (Siemens Medical Solutions USA Inc., Mountain View, California, USA).
Statistical analysis
Continuous parametric variables were expressed as mean±SD values. Parametric variables were compared between groups using the Student t-test and analysis of variance. Nonparametric variables were compared between groups using Dunnett’s T3 method. Categorical variables were presented as number and proportion [(n (%)] and differences in the study were tested using the w2-test. Multivariate logistic regression analyses (backward: LR) were used to identify the predictive value of independent factors for the presence of CHD. Receiver operating characteristics (ROC) curve analyses were also used to quantify the predictive value of independent parameters for the presence of CHD. All of the reported P values were two-tailed, and those less than 0.05 were considered to be statistically significant. All statistical analyses were carried out using the software package SPSS 13.0 (SPSS Inc., Chicago, Illinois, USA) for Windows.
Results Baseline characteristics
The mean age of the study group was 64.14±10.94 years, and 62.95% of the patients were men. Table 1 summarizes the baseline characteristics of the patients. There were more male patients in the CHD group than in the normal group (P < 0.001). Patients in the CHD group were older, and had higher BMI and systolic blood pressure than those in the normal group (P < 0.01) and they were more likely to have a history of smoking, hypertension, hypercholesterolemia, hypercholesterolemia, and diabetes (P < 0.001). There were no significant differences between the two groups in terms of diastolic blood pressure and heart rate (P > 0.05). Patients in the CHD group had higher non-HDL-C (P < 0.05), apoB (P < 0.05), apoB/apoA1 (P < 0.001), and lower HDL-C (P < 0.001) and apoA1 (P < 0.001). No significant differences were found between the two groups in TG, TC, and LDL-C.
Laboratory methodology
All venous blood samples drawn at 6:00 a.m. from all patients after an overnight fast were immediately refrigerated and transported to the department of Clinical Laboratory of Ruijin Hospital for measurements of TC, LDL-C, HDL-C, triglycerides (TGs), apoB, apoA1, fasting blood glucose, fasting insulin, ultrasensitive C-reactive protein, CK-MB, cTnI, and pro-BNP. The apoB/apoA1 ratio and non-HDL-C were then calculated. Coronary angiography
Coronary angiography was performed using the Judkins technique by INNOVA 20000 equipment (GE Healthcare, Waukesha, Wisconsin, USA), and all procedures followed the standard American College of Cardiology/ American Heart Association guidelines for coronary angiography. The coronary angiograms were evaluated by two experienced interventional cardiologists who were unaware of the patients’ biochemical status.
Association of the apolipoprotein B/apolipoprotein A1 ratio and non-high-density lipoprotein cholesterol with coronary heart disease
In the multivariate logistic regression analysis, apoB, apoB/apoA1 ratio, and non-HDL-C were all strongly associated with an increased risk of CHD after adjusting for sex, age, BMI, history of hypertension, diabetes, smoking, systolic blood pressure, diastolic blood pressure, and LDL-C. Furthermore, it also indicated that the associations were much stronger for apoB than for the apoB/apoA1 ratio or non-HDL-C [apoB: odds ratio (OR) = 11.285, 95% confidence interval (CI) 4.513–28.220; apoB/ apoA1: OR = 8.941, 95% CI 4.363–18.323; non-HDL-C: OR = 1.373, 95% CI 1.163–1.622; Table 2]. After adjusting for sex, age, BMI, history of hypertension, diabetes, smoking, systolic blood pressure, diastolic blood pressure, and LDL-C, patients in the top quartile of the apoB/apoA1 distribution had an OR of 7.321
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Combined use of apoB/apoA1 Pan et al. 435
Patient characteristics: coronary heart disease group and normal group
Table 1
Male [n (%)] Age (years) BMI (kg/m2) Systolic blood pressure (mmHg) Diastolic blood pressure (mmHg) Heart rate (bpm) Smoking [n (%)] Hypertension [n (%)] Hypercholesterolemia [n (%)] Diabetes [n (%)] TG (mmol/l) TC (mmol/l) HDL-C (mmol/l) LDL-C (mmol/l) Non-HDL-C (mmol/l) ApoA1 (g/l) ApoB (g/l) ApoB/apoA1
Normal group (n = 294)
CHD group (n = 532)
P value
153 (52.04) 60.88±9.80 24.29±2.96 126.56±17.46
367 (68.99) 65.94±11.14 25.06±3.39 130.29±20.18
0.000*** 0.000*** 0.006** 0.008**
77.16±9.54
76.46±11.16
0.368
76.09±11.83 90 (30.61) 180 (61.22) 31 (10.54)
75.64±10.86 248 (46.62) 372 (69.93) 86 (16.17)
0.105 0.000*** 0.000*** 0.000***
66 (22.45) 1.78±1.17 4.44±0.98 1.21±0.33 2.66±0.83 3.22±0.93 1.30±0.33 0.88±0.22 0.71±0.23
249 (46.81) 1.73±1.15 4.46±1.08 1.09±0.30 2.74±0.92 3.37±1.05 1.17±0.31 0.93±0.25 0.84±0.29
0.000*** 0.505 0.741 0.000*** 0.205 0.049* 0.000*** 0.011* 0.000***
apoA1, apolipoprotein A1; apoB, apolipoprotein B; CHD, coronary heart disease; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; TC, total cholesterol; TG, triglyceride. Compared with the normal group: *P < 0.05; **P < 0.01; ***P < 0.001.
Table 3 Logistic regression analysis for the association of the quartile groups of apolipoprotein B/apolipoprotein A1 with coronary heart disease 95% CI Regression coefficient A B C D Constant
0.526 1.463 1.991 – 2.159
Standard error
P value
0.254 0.296 0.322 1.337
0.000 0.038 0.000 0.000 0.106
OR
Lower Upper
1 1.693 1.029 2.784 4.317 2.419 7.705 7.321 3.891 13.771 0.115
CI, confidence interval; OR, odds ratio.
Table 4 Logistic regression analysis for the association of the quartile groups of apolipoprotein B with coronary heart disease 95% CI
A B C D Constant
Regression coefficient
Standard error
P value
0.958 0.972 1.412 – 0.694
0.275 0.275 0.288 1.136
0.000 0.000 0.000 0.000 0.541
OR
Lower Upper
1 2.606 1.521 4.465 2.642 1.540 4.534 4.105 2.332 7.225 0.500
CI, confidence interval; OR, odds ratio.
Multivariate logistic regression analysis for the association of the apolipoprotein B/apolipoprotein A1 ratio and non-high-density lipoprotein cholesterol with coronary heart disease
Table 2
Table 5 Logistic regression analysis for the association of the quartile groups of non-high-density lipoprotein cholesterol with coronary heart disease 95% CI
95% CI
ApoB ApoA1 ApoB/apoA1 Non-HDL-C
Regression coefficient
Standard error
P value
OR
Lower
Upper
2.423 – 1.836 2.191 0.317
0.468 0.334 0.366 0.085
0.000 0.000 0.000 0.000
11.285 0.159 8.941 1.373
4.513 0.083 4.363 1.163
28.220 0.307 18.323 1.622
apoA1, apolipoprotein A1; apoB, apolipoprotein B; CI, confidence interval; HDL-C, high-density lipoprotein cholesterol; OR, odds ratio.
(95% CI 3.891–13.771; Table 3) compared with those in the bottom quartile, and patients in the top quartile of apoB had an OR of 4.105 (95% CI 2.332–7.225; Table 4) compared with those in the bottom quartile, whereas patients in the top quartile of non-HDL-C had an OR of 2.427 (95% CI 1.407–4.187; Table 5) compared with those in the bottom quartile.
Combined use of apolipoprotein B/apolipoprotein A1 and non-high-density lipoprotein cholesterol in predicting coronary heart disease
In our study, the top quartile of the apoB/apoA1 group was defined as the high apoB/apoA1 ratio group (apoB/ apoA1: Z 0.96 in men and Z 0.89 in women) and the top quartile of the non-HDL-C group was defined as the high non-HDL-C group (non-HDL-C: Z 3.89 mmol/l in men and Z 4.10 mmol/l in women). All of the participants were further divided into four groups on the basis of the results of apoB/apoA1 and non-HDL-C as follows:
Regression coefficient A B C D Constant
0.292 0.554 0.887 – 1.550
Standard error
P value
0.262 0.277 0.278 1.254
0.012 0.265 0.046 0.001 0.216
OR
Lower Upper
1 1.339 0.802 2.236 1.740 1.011 2.994 2.427 1.407 4.187 0.212
CI, confidence interval; OR, odds ratio.
group A, low apoB/apoA1 ratio/low non-HDL-C; group B, low apoB/apoA1 ratio/high non-HDL-C; group C, high apoB/apoA1 ratio/low non-HDL-C; and group D, high apoB/apoA1 ratio/high non-HDL-C. The incidence rate of CHD in the high apoB/apoA1 group was much higher than that in the low apoB/apoA1 group (78.85 vs. 59.55%, P < 0.001). In addition, compared with group 1, patients in groups 3 and 4 had a higher incidence rate of CHD (group 3 or 4 vs. group 1, P < 0.001; group 3 or 4 vs. group 2, P < 0.01). Patients with the combined high levels of apoB/apoA1 and non-HDL-C (group 4) had the highest risk of CHD (79.31%; Table 6). Apolipoprotein B/apolipoprotein A1, non-high-density lipoprotein cholesterol, and the receiver operating characteristics curve of coronary heart disease
We used the area under the curve (AUC) of ROC curve to determine the predictive values of various lipid markers.
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Coronary Artery Disease 2014, Vol 25 No 5
Table 6 Combined use of apolipoprotein B/apolipoprotein A1 and non-high-density lipoprotein cholesterol in predicting coronary heart disease Low apoB/apoA1 ratio (N = 618) Group 1 Low nonHDL-C (N = 524)
Fig. 1
1.0
High apoB/apoA1 ratio (N = 208)
Group 2 High nonHDL-C (N = 94)
Group 3 Low nonHDL-C (N = 92)
0.8
Group 4 High nonHDL-C (N = 116)
Incidence rate of 311 (59.35) 57 (60.64) 72 (78.26)***## 92 (79.31)***## CHD [N (%)] apoA1, apolipoprotein A1; apoB, apolipoprotein B; CHD, coronary heart disease; non-HDL-C, non-high-density lipoprotein cholesterol. Compared with group 1: ***P < 0.001. Compared with group 2: ##P < 0.01.
Sensitivity
436
0.6
0.4
0.2 Area under the receiver operating characteristics curve for coronary heart disease Table 7
95% CI
Combined use of the apoB/apoA1 ratio and non-HDL-C ApoB/apoA1 Non-HDL-C ApoA1 ApoB TG TC HDL-C LDL-C
0.0 0.0
Area under the ROC curve
Standard error
P value
Lower
Upper
0.762
0.043
0.000
0.677
0.847
0.757 0.619 0.307 0.663 0.525 0.586 0.378 0.614
0.043 0.055 0.053 0.051 0.057 0.056 0.052 0.057
0.000 0.046 0.001 0.006 0.668 0.147 0.039 0.055
0.674 0.511 0.203 0.563 0.415 0.475 0.277 0.502
0.841 0.727 0.410 0.763 0.636 0.697 0.479 0.726
apoA1, apolipoprotein A1; apoB, apolipoprotein B; CI, confidence interval; LDL-C, low-density lipoprotein cholesterol; non-HDL-C, non-high-density lipoprotein cholesterol; ROC, receiver operating characteristics; TC, total cholesterol; TG, triglyceride.
The combined use of the apoB/apoA1 ratio and nonHDL-C showed a greater ROC area than the apoB/apoA1 ratio, apoB, non-HDL-C, or other lipid profiles (Table 7 and Fig. 1).
Discussion In the present study, higher levels of apoB, apoB/apoA1 ratio, and non-HDL-C were all strongly and independently associated with an increased risk of CHD after adjusting for sex, age, BMI, hypertension, diabetes, smoking, blood pressure, and other lipid profiles. These profiles appear to be much stronger predictors of the risk of CHD than other routine lipid profiles, which are consistent with previous studies [7]. We also found that after adjustment for matching factors, the relative risk of CHD in the highest quartile compared with the lowest quartile was 7.321 (95% CI 3.891–13.771) for apoB/apoA1 and 4.105 (95% CI 2.332–7.225) for apoB, and 2.427 (95% CI 1.407–4.187) for non-HDL-C, which was the similar to Chien et al.’s [13] findings. Through ROC analysis, our study also showed that an elevated apoB/apoA1 ratio
0.2
0.4 0.6 1 − Specificity
0.8
1.0
Combined use of apoB/apoA1 and non-HDL-C
apoB
Non-HDL-C
Reference line
apoB/apoA1
Receiver operating characteristics curve of serum apolipoprotein B (apoB), apoB/apolipoprotein A1 (apoA1), non-high-density lipoprotein cholesterol (non-HDL-C), and combined factors for the diagnosis of coronary heart disease.
(AUC = 0.757) is more useful in predicting a greater risk of CHD than apoB (AUC = 0.663), non-HDL-C (AUC = 0.619), or other traditional lipid parameters. Chien and colleagues also found that the ROC area of apoB/apoA1 (AUC = 0.66) for the diagnosis of CHD was greater than apoB (AUC = 0.63), non-HDL-C (AUC = 0.60), LDL-C (AUC = 0.60), or other lipid profiles. In the current study, we could reproduce this finding, but we also compared the predictive value of the combined use of apoB/apoA1 and non-HDL-C. As expected, the area of combined use was greater than its individual components or other lipid profiles, suggesting that the combination of apoB/apoA1 and non-HDL-C was a stronger predictor for the risk of CHD. However, on comparison of the four subgroups, our study also showed that high ratios of apoB/ apoA1 in combination with high levels of non-HDL-C may further enhance their ability to predict CHD. ApoB is essential for the binding of LDL-C particles, allowing cells to internalize LDL-C and thus absorb cholesterol, leading to cardiovascular disease more easily, so the level of apoB represents the number of atherogenic particles. The European guidelines on Cardiovascular Disease Prevention in Clinical Practice
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Combined use of apoB/apoA1 Pan et al. 437
(version 2012) [14] suggested that apoB seemed to be a risk marker similar to LDL-C and a better index of the adequacy of LDL-lowering therapy. The guideline recommended that it should be less than 80 mg/dl and less than 100 mg/dl for patients with very high or high risk of cardiovascular disease, respectively; whereas, nonHDL-C comprises all the cholesterol contained in LDL, very low-density lipoprotein (VLDL) and intermediate-density lipoprotein particles. It can be calculated by simply subtracting HDL-C from TC. Thus, it is the cholesterol in the atherogenic particles and has been recommended as a target especially in patients with high nonfasting TG levels by the National Cholesterol Education Program guidelines (NCEP ATP III 2002) [15]. An elevated apoB has been shown to have the best predictive ability for measurable quantities of the atherogenic small, dense LDL particles, which can penetrate into the arterial wall and become deposited under the vascular endothelium more easily than larger but looser LDL particles because of their small size [16]. Therefore, some studies suggested that the atherogenic risk resulting from lipoproteins does not relate primarily to their cholesterol content, but rather their size and number [17–19]. This may be the reason why the associations with an increased risk of CHD were much stronger for apoB than for non-HDL-C, which are consistent with previous studies [7,11,12]. As mentioned above, apoB represents the number of atherogenic particles, whereas levels of apoA1 are strongly correlated with those of HDL-C, and expression of apoA1 may be largely responsible for determining the plasma level of HDL-C. It has long been known that a low level of HDL-C is a powerful predictor of increased cardiovascular risk. Although recent prospective research failed to prove that increasing the level of HDL-C could reduce the incidence rate of CHD, many epidemiological studies have found that patients with high levels of HDL-C have a lower incidence rate of CHD [20,21]. Therefore, the apoB/apoA1 ratio mirrors the balance between proatherogenic and antiatherogenic lipoprotein particles, which is why apoB/apoA1 was superior to nonHDL-C or other traditional lipid profiles in predicting the risk of CHD as we found in the present study. However, in the multivariate logistic regression, we found that apoB was more strongly associated with an increased risk of CHD than the apoB/apoA1 ratio after adjustment (apoB: OR = 11.285, 95% CI 4.513–28.220; apoB/apoA1: OR = 8.941, 95% CI 4.363–18.323), which was not in agreement with The INTERHEART study [3], which showed that the apoB/apoA1 ratio had a stronger association with the risk of myocardial infarction (OR = 1.59, 95% CI 1.52–1.64) than apoB (OR = 1.32, 95% CI 1.28–1.36). The reason for this discrepancy could be that they mainly focused on patients with acute myocardial infarction rather than all of the CHD patients. Furthermore, the ethnicity and the scale of the study may have
also affected the results. However, no previous studies have investigated the predictive value of the combination of apoB/apoA1 and non-HDL-C. Our study found that the combination of apoB/apoA1 and non-HDL-C had greater predictive value than its individual components or other lipid profiles. TG-rich lipoproteins such as chylomicron emulsion or VLDL would exchange their contents with cholesterol-rich LDL, and during this procedure not only sd-LDL but also cholesterol-rich VLDL residues would be produced, which are also proatherogenic. NonHDL-C is the mass of cholesterol in the atherogenic particles and the secondary lipid-lowering target in patients with elevated serum TGs (NCEP ATP III). Elevated non-HDL-C represents a high concentration of cholesterol in TG-enriched lipoproteins and their particles, whereas the level of LDL-C remains normal. Whereas the apoB/apoA1 ratio represents the balance of proatherogenic and antiatherogenic lipoprotein particles, naturally, the combined use of these two profiles have a higher predictive value for the risk of CHD. Limitations
Our study, however, has some limitations. First, our study mainly focused on the CHD patients whose lesion of coronary artery was 50% or more in at least one major branch according to the diagnostic criteria from the guideline of American College of Cardiology/American Heart Association. Although increasingly more studies [22–26] have reported that mild coronary stenosis may have higher potential of getting vulnerable plaque in its coronary arteries, which easily leading to acute coronary syndrome, we were also aware of the importance of patients with mild coronary stenosis, and relevant studies are ongoing in our work group. However, in this study, we have only focused on the CHD group. Second, the patients were not from a general population, but rather visitors to our hospital, which may have led to a selection bias. Therefore, this study may not be an accurate reflection on the entire population in China. Because of the small sample size, our ability to investigate the predictive power of the ratio and non-HDL-C was limited. Therefore, further research on a large scale and with more participants is required. Conclusion
We clearly show that apoB, apoB/apoA1 ratio, and nonHDL-C were superior to other routine lipid profiles in predicting the risk of CHD, and the ratio was the best predictor among them. However, the combination of apoB/apoA1 and non-HDL-C had even greater predictive value than its individual components or other lipid profiles.
Acknowledgements This work was supported by the National Natural Science Foundation of China (81370331), the Shanghai Municipal
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
438
Coronary Artery Disease 2014, Vol 25 No 5
Science and Technology Commission (12140903402), and the Young Medical Talents Training Program of Shanghai Municipal Health Bureau (XYQ2011011) to Z. Chen, and Shanghai Municipal Science and Technology Commission (10JC1410502) to G. Lu.
12
Conflicts of interest
15
13
14
There are no conflicts of interest.
References 1
Expert Panel. 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) final report. Circulation 2002; 106:3143–3421. 2 Walldius G, Jungner I, Holme I, Aastveit AH, Kolar W, Steiner E. High apolipoprotein B, low apolipoprotein A-I, and improvement in the prediction of fatal myocardial infarction (AMORIS study): a prospective study. Lancet 2001; 358:2026–2033. 3 Yusuf S, Hawken S, Ounpuu S, Dans T, Avezum A, Lanas F, et al. Effect of potentially modifiable risk factors associated with myocardial infarction in 52 countries (the INTERHEART study): case–control study. Lancet 2004; 364:937–952. 4 Walldius G, Jungner I. The apoB/apoA-I ratio: a strong, new risk factor for cardiovascular disease and a target for lipid-lowering therapy – a review of the evidence. J Intern Med 2006; 259:493–519. 5 Thompson A, Danesh J. Associations between apolipoprotein B, apolipoprotein AI, the apolipoprotein B/AI ratio and coronary heart disease: a literature-based meta-analysis of prospective studies. J Intern Med 2006; 259:481–492. 6 Miller M, Ginsberg HN, Schaefer EJ. Relative atherogenicity and predictive value of non-high-density lipoprotein cholesterol for coronary heart disease. Am J Cardiol 2008; 101:1003–1008. 7 Robinson JG, Wang S, Smith BJ, Jacobson TA. Meta-analysis of the relationship between non-high-density lipoprotein cholesterol reduction and coronary heart disease risk. J Am Coll Cardiol 2009; 53:316–322. 8 Rosenson RS. Management of non-high-density lipoprotein abnormalities. Atherosclerosis 2009; 207:328–335. 9 Lu W, Resnick HE, Jablonski KA, Jones KL, Jain AK, Howard WJ, et al. Non-HDL cholesterol as a predictor of cardiovascular disease in type 2 diabetes: the Strong Heart study. Diabetes Care 2003; 26:16–23. 10 Jiang R, Schulze MB, Li T, Rifai N, Stampfer MJ, Rimm EB, Hu FB. Non-HDL cholesterol and apolipoprotein B predict cardiovascular disease events among men with type 2 diabetes. Diabetes Care 2004; 27:1991–1997. 11 Denke MA. Weighing in before the fight: low-density lipoprotein cholesterol and non-high-density lipoprotein cholesterol versus apolipoprotein B as the best predictor for coronary heart disease and the best measure of therapy. Circulation 2005; 112:3368–3370.
16
17 18
19
20
21
22
23
24
25
26
Sniderman AD. Apolipoprotein B versus non-high-density lipoprotein cholesterol: and the winner is y. Circulation 2005; 112:3366–3367. Chien KL, Hsu HC, Su TC, Chen MF, Lee YT, Hu FB. Apolipoprotein B and non-high density lipoprotein cholesterol and the risk of coronary heart disease in Chinese. J Lipid Res 2007; 48:2499–2505. Perk J, De Backer G, Gohlke H, Graham I, Reiner Z, Verschuren WM, et al. European Guidelines on cardiovascular disease prevention in clinical practice (version 2012). Eur Heart J 2012; 33:1635–1701. National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, andTreatment of High Blood Cholesterol in Adults (Adult Treatment Panel III). 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) final report. Circulation 2002; 106:3143–3421. Tsimihodimos V, Gazi I, Kostara C, Tselepis AD, Elisaf M. Plasma lipoproteins and triacylglycerol are predictors of small, dense LDL particles. Lipids 2007; 42:403–409. Sacks FM. The apolipoprotein story. Atheroscler Suppl 2006; 7:23–27. Pischon T, Girman CJ, Sacks FM, Rifai N, Stampfer MJ, Rimm EB. Non-highdensity lipoprotein cholesterol and apolipoprotein B in the prediction of coronary heart disease in men. Circulation 2005; 112:3375–3383. Sniderman A, Williams K, Haffner S, Sattar N. Insights from apoB: from better diagnosis and therapy to the Medusa hypothesis. Atheroscler Suppl 2004; 5:19–24. Schwartz GG, et al. dal-OUTCOMES Investigators. Effects of dalcetrapib in patients with a recent acute coronary syndrome. N Engl J Med 2012; 367:2089–2099. Al-Hijji M, et al. Effect of equivalent on-treatment apolipoprotein levels on outcomes (from the AIM-HIGH and HPS2-THRIVE). Am J Cardiol 2013; 112:1697–1700. Virmani R, Kolodgie FD, Burke AP, Finn AV, Gold HK, Tulenko TN, et al. Atherosclerotic plaque progression and vulnerability to rupture: angiogenesis as a source of intraplaque hemorrhage. Arterioscler Thromb Vasc Biol 2005; 25:2054–2061. Calvert PA, Obaid DR, O’Sullivan M, Shapiro LM, McNab D, Densem CG, et al. Association between IVUS findings and adverse outcomes in patients with coronary artery disease: the VIVA (VH-IVUS in Vulnerable Atherosclerosis) study. JACC Cardiovasc Imaging 2011; 4:894–901. Hong MK, Mintz GS, Lee CW, Lee JW, Park JH, Park DW, et al. A three-vessel virtual histology intravascular ultrasound analysis of frequency and distribution of thin-cap fibroatheromas in patients with acute coronary syndrome or stable angina pectoris. Am J Cardiol 2008; 101: 568–572. Barlis P, Serruys PW, Gonzalo N, van der Giessen WJ, de Jaegere PJ, Regar E. Assessment of culprit and remote coronary narrowings using optical coherence tomography with long-term outcomes. Am J Cardiol 2008; 102:391–395. Halvorsen B, Otterdal K, Dahl TB, Skjelland M, Gullestad L, Øie E, Aukrust P. Atherosclerotic plaque stability – what determines the fate of a plaque? Prog Cardiovasc Dis 2008; 51:183–194.
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
