Abstract
Objective
Obesity is associated with cardiovascular disease (CVD) and CVD mortality. However, previous reports showed a paradoxical protective effect in patients with known CVD referred as “obesity paradox”. Therefore, the aim of the present study was to investigate the association of body mass index (BMI) with coronary artery calcification (CAC) in a large outpatient cardiac CT cohort.
Methods
4.079 patients who underwent cardiac CT between December 2007–May 2014 were analyzed. BMI and clinical risk factors (current smoking, diabetes mellitus type 2, family history, systolic blood pressure, lipid spectrum) were assessed. Missing values were imputed using multiple imputation. CAC extent was categorized as absent (0), mild (>0–100), moderate (>100–400) and severe (>400).
Results
Multivariable multinomial logistic regression analysis, including all risk factors as independent variables, showed no association between BMI and CAC. Using absence of calcification as reference category, the odds ratios per unit increase in BMI were 1.01 for mild; 1.02 for moderate; and 1.00 for severe CAC (p‐values ≥0.103).
Conclusions
No statistically significant association was observed between BMI and CAC after adjustment for other risk factors.
Keywords: cardiovascular disease, computed tomography, coronary artery disease (CAD), obesity
1. INTRODUCTION
Obesity is recognized as a well‐established risk factor for cardiovascular disease (CVD) and concerns an increasing problem of the public health worldwide. According to the National, Heart, Lung, and Blood Institute criteria, obesity is defined as a body mass index (BMI) equal or greater than 30 kg/m2 which concerns the most widely accepted and used obesity index. 1 The European Society of Cardiology guidelines on the prevention of CVD lists obesity as a modifiable cardiovascular risk factor. 2 Despite the clear associations between obesity and cardiovascular mortality, it is a matter of debate whether obesity is actually an independent risk factor, since obesity is strongly associated with insulin resistance, type 2 diabetes mellitus, inflammation, dyslipidaemia and hypertension, known as metabolic syndrome. 3 , 4 , 5 , 6 There is even a subset of persons with obesity which are considered as “metabolically healthy” due to the fact that these patients have optimal regulated risk factors for CVD and whereby no increase of cardiovascular mortality is observed when compared to healthy subjects. 7 , 8 Additionally, it is indicated that obesity is associated with CAC and cardiovascular risk particularly in young and middle‐aged individual's. 9 , 10 Previous studies on various patients groups with established CVD reported a paradoxical protective effect of obesity which known as the “obesity paradox”. 11 , 12 , 13 In accordance with this phenomenon, the presence and extent of coronary artery disease (CAD) and especially major adverse cardiovascular events have been inversely associated with BMI. 14 , 15 , 16 A series of reports have suggested that a “U‐shaped” relationship exists between coronary artery calcification (CAC) and obesity indices including BMI. 15 , 16 , 17 Despite these observations, the majority of previous studies were performed in patients with established CVD, relatively small study populations, subjects with limited age and extreme categories of BMI. In addition, these study populations are not a good reflection of the daily outpatient clinical practice.
The extent of CAC is associated with significant luminal stenosis and concerns a well‐established predictor for major adverse cardiac events. 18 , 19 , 20 Therefore, the aim of the present study was to investigate the association between BMI and presence and extent of CAC in a large single‐center outpatient coronary CT cohort.
2. METHODS
2.1. Study population
5738 consecutive patients from the cardiology outpatient clinic with a low‐to‐intermediate pretest probability for obstructive CAD, referred for cardiac CT in the period between December 2007 and May 2014 as part of their diagnostic work‐up, were selected.
Included were patients who underwent a non‐contrast enhanced CT scan to determine the Agatston score (AS). 21 Excluded from the analysis were patients with known history of revascularization (percutaneous coronary intervention (n = 61) or coronary artery bypass grafting (n = 57)), unreliable CAC measurement due to a pacemaker or implantable cardioverter‐defibrillator (n = 14) and missing data on length, weight and classical cardiovascular risk factors (n = 1.527). The final analysis was based on 4.079 patients (Figure 1). Missing values according to current smoking (n = 229), type 2 diabetes mellitus (n = 169), family history (n = 315), systolic blood pressure (n = 240), total cholesterol (n = 676), HDL cholesterol (n = 790) and LDL (n = 803) were imputed using multiple imputation. Complete data were available for 2857 patients (70%).
FIGURE 1.
Flowchart of the study design
This study was approved by the Institutional Review Board (IRB) and Ethics Committee of Maastricht University Medical Center (METC 15‐4‐119). Written informed consent was waived because data were retrospectively analyzed anonymously in accordance with the IRB guidelines. This study complies with the ethical principles of the Declaration of Helsinki.
2.2. CAC assessment
A non‐contrast gated ECG triggered scan was performed to determine CAC using the Agatston method. 21 An attenuation threshold of 130 Hounsfield units (HU) was used to identify calcifications in the main coronary branches. All individual calcifications were manually picked, summed and expressed as the AS. The AS was independently measured by an experienced radiologist and cardiologist. In case of disagreement, consensus was reached by reviewing the findings jointly.
From December 2007 until June 2010, a 64‐slice multi‐detector CT‐scanner (Brilliance 64; Philips Healthcare, Best, The Netherlands; n = 1.735) was used to measure CAC. Data acquisition parameters were a slice collimation of 64 × 0.625 mm, a gantry rotation time of 420 ms, scan time of 0.4 s, tube voltage of 120 kV and slice thickness of 3 mm. CAC was measured using dedicated calcium scoring software (Heartbeat‐CS, EBW, Philips Healthcare, Best, The Netherlands).
From June 2010 onward, a second generation dual‐source CT‐scanner was used (Somatom Definition Flash, Siemens Medical Solutions, Forchheim, Germany; n = 2.344) to measure CAC. Data acquisition parameters were: pitch 3.4, slice collimation 2 × 128 × 0.6 mm, gantry rotation time 280 ms, tube voltage 120 kV, tube current 100–150 reference mAs and slice thickness of 3 mm; reconstruction was performed with a B35f kernel. CAC was calculated using dedicated software (Syngo.Via Calcium scoring, Siemens Healthcare, Forchheim, Germany).
The presence and extent of CAC was categorized as absent (Agatston score = 0), mild (Agatston score >0–100), moderate (Agatston score > 100–400) and severe (Agatston score > 400).
3. BMI
BMI was divided into clinically relevant categories on the basis of National Heart, Lung and Blood Institute criteria 1 : underweight (BMI <18.5 kg/m2), normal weight (BMI 18.5–24.9 kg/m2), overweight (BMI 25.0–29.9 kg/m2), class I obesity (BMI 30.0–34.9 kg/m2), class II obesity (BMI 35.0–39.9 kg/m2), and class III extreme obesity (BMI ≥ 40.0 kg/m2). Additionally, patients with class I, II and III obesity, were combined into one category as having obesity.
3.1. Cardiovascular risk factors
Cardiovascular risk factors were collected prior to cardiac CT. Diabetes mellitus type 2 was defined as fasting glucose levels of ≥7 mmol/L or treatment with either diet intervention, oral glucose lowering agent or insulin; smoking was defined as current smoking. A positive family history was defined as having a first‐degree relative with a history of myocardial infarction or sudden cardiac death before the age of sixty. The systolic blood pressure, total cholesterol, HDL and LDL cholesterol were assessed at the outpatient cardiology clinic prior to cardiac CT.
Missing values according to the cardiovascular risk factors were imputed using multiple imputation.
3.2. Statistical analysis
The distribution of continuous variables were described by means and standard deviation (in case of normal distributions) and as median with interquartile range (IQR) (in case of skewed distributions). Differences between CAC categories with respect to continuous variables were tested for statistical significance using one‐way analysis of variance (ANOVA) test since all continuous variables were normally distributed. Differences in categorical variables were tested with the Chi‐square test.
Multinomial logistic regression analysis was conducted with four CAC categories as dependent variable: absent (AS = 0), mild (AS > 0–100), moderate (AS > 100–400) and severe CAC (AS > 400). Absence of calcification (AS = 0) was used as the reference category. To evaluate the association between BMI and CAC extent, both univariate and multivariable regression analyses were performed. The multivariate model included BMI, and conventional risk factors for CVD such as age, male gender, diabetes mellitus type 2, active smoking, systolic blood pressure, total cholesterol, LDL and HDL cholesterol. The multinomial logistic regression analysis provides odds ratios with 95% confidence intervals for each independent variable for three comparisons: mild CAC versus no CAC, moderate CAC versus no CAC and severe CAC versus no CAC. Data were analyzed using SPSS version 24.0 (SPSS Inc., Chicago, IL, USA). Only two‐sided p‐values were used and a p‐value ≤0.05 was considered statistically significant.
4. RESULTS
4.1. Study population
The mean age of the total study population was 56 ± 11 years and mean BMI was 27.0 ± 4.7 kg/m2. 52% of the population were male; 23% were current smokers and 8% had diabetes mellitus type 2. Mean systolic blood pressure was 144 ± 21 mmHg. A total of 1.809 patients (44%) had an AS of zero (no CAC); 1.171 (29%) had an AS between 0 and 100 (mild CAC); 580 (14%) had an AS between 100 and 400 (moderate CAC) and 519 (13%) had an AS > 400 (severe CAC). Between the CAC categories there were significant differences with respect to age, male gender, diabetes mellitus type 2, systolic blood pressure, total cholesterol, LDL and HDL (all p‐values <0.001). Mean BMI values were similar across the different CAC groups (p‐value = 0.083). The baseline characteristics of the total study population according the CAC categories are further described in Table 1.
TABLE 1.
Baseline characteristics of the study population according to the presence and extent of CAC
| Total n = 4.079 | No CAC | Mild CAC | Moderate CAC | Severe CAC | p‐value | |
|---|---|---|---|---|---|---|
| AS = 0 n = 1.809 | AS > 0–100 n = 1.171 | AS > 100–400 n = 580 | AS > 400 n = 519 | |||
| Age, years | 56 ± 11 | 51 ± 11 | 58 ± 10 | 62 ± 9 | 65 ± 9 | <0.001 |
| Male gender, % | 2.109 (52) | 767 (42) | 631 (54) | 348 (60) | 363 (70) | <0.001 |
| BMI, kg/m2 | 27.0 ± 4.7 | 26.8 ± 4.9 | 27.1 ± 4.4 | 27.4 ± 4.7 | 27.1 ± 4.6 | 0.083 |
| Current smoking, % | 932 (23) | 412 (23) | 274 (23) | 126 (22) | 120 (23) | 0.865 |
| Diabetes mellitus, % | 340 (8) | 94 (2) | 103 (9) | 68 (12) | 75 (15) | <0.001 |
| Positive family history, % | 1.462 (36) | 654 (36) | 425 (36) | 200 (34) | 183 (35) | 0.812 |
| Systolic blood pressure, mmHg | 144 ± 21 | 141 ± 19 | 145 ± 21 | 148 ± 21 | 149 ± 22 | <0.001 |
| Total cholesterol, mmol/L | 5.5 ± 1.2 | 5.5 ± 1.1 | 5.5 ± 1.2 | 5.5 ± 1.2 | 5.1 ± 1.2 | <0.001 |
| LDL cholesterol, mmol/L | 3.4 ± 1.1 | 3.4 ± 1.0 | 3.5 ± 1.1 | 3.4 ± 1.1 | 3.1 ± 1.1 | <0.001 |
| HDL cholesterol, mmol/L | 1.4 ± 0.4 | 1.4 ± 0.5 | 1.3 ± 0.4 | 1.4 ± 0.4 | 1.3 ± 0.4 | <0.001 |
Abbreviations: AS, Agatston score; BMI, body mass index; LDL, low‐density lipoprotein; HDL, high‐density lipoprotein.
4.2. Association of BMI with CAC
Figure 2 visualizes the relationship of median AS with BMI within quartiles of age. Below the age of 56 years, the median AS is zero within all BMI categories. Within the age category >56–64 years, there is a trend toward higher median AS with increasing BMI. Median AS is highest in the BMI category <18.5 kg/m2 years with age >64 years. However, this should be interpreted with caution since this group only contains 10 patients.
FIGURE 2.
Median Agatston score according to body mass index (BMI) and age categories
Log transformation of the dependent variable AS did not result in a normal distribution due to a large proportion of patients with a zero value for AS. Therefore, a classification into four categories was used to perform multinomial logistic regression analysis. No CAC (AS = 0) was used as baseline category and was compared with three alternative categories: mild (AS > 0–100), moderate (AS > 100–400) and severe CAC (AS > 400). Table 2 shows the results from a univariate multinomial logistic regression analysis (without adjustment for other risk factors) and presents the odds ratio per unit increase in BMI with 95% CI for each alternative category when compared with the baseline category (AS = 0). BMI showed only a significant positive association in the comparison of moderate CAC with no CAC (OR: 1.02, 95% CI 1.01–1.05, p = 0.012). No significant association was observed for BMI when comparing mild CAC with no CAC (OR: 1.01, 95% CI 0.99–1.03, p = 0.175) and severe CAC with no CAC (OR: 1.01, 95% CI 0.99–1.03, p = 0.328) (Table 2, model 1).
TABLE 2.
Multinomial logistic regression analysis of factors associated with odds AS
| Mild CAC | Moderate CAC | Severe CAC | |
|---|---|---|---|
| AS > 0–100 | AS > 100–400 | AS > 400 | |
| OR (95% CI) p‐value | OR (95% CI) p‐value | OR (95% CI) p‐value | |
| AS = 0 | 1.00 (reference) | 1.00 (reference) | 1.00 (reference) |
| Univariate model | |||
| BMI, kg/m 2 | 1.01 (0.99–1.03) 0.175 | 1.02 (1.01–1.05) 0.012 | 1.01 (0.99–1.03) 0.328 |
| Multivariate model | |||
| BMI, kg/m2 | 1.01 (0.98–1.02) 0.952 | 1.02 (1.00–1.05) 0.103 | 1.00 (0.97–1.02) 0.774 |
| Age, years | 1.09 (1.08–1.10) <0.001 | 1.15 (1.14–1.17) <0.001 | 1.21 (1.19–1.23) <0.001 |
| Male gender, yes versus no | 2.22 (1.85–2.63) <0.001 | 3.85 (3.03–4.76) <0.001 | 6.67 (5.00–8.33) <0.001 |
| Diabetes mellitus, yes versus no | 1.69 (1.20–2.38) 0.002 | 2.22 (1.52–3.23) <0.001 | 2.50 (1.67–3.85) <0.001 |
| Family history, yes versus no | 1.20 (1.01–1.43) 0.040 | 1.28 (1.02–1.61) 0.031 | 1.59 (1.23–2.04) <0.001 |
| Current smoking, yes versus no | 1.52 (1.25–1.85) <0.001 | 1.96 (1.49–2.56) <0.001 | 2.56 (1.92–3.45) <0.001 |
| Systolic blood pressure, mmol/L | 1.00 (1.00–1.01) 0.041 | 1.01 (1.00–1.02) 0.001 | 1.01 (1.00–1.01) 0.005 |
| Total cholesterol, mmol/L | 1.04 (0.84–1.29) 0.732 | 0.97 (0.75–1.26) 0.809 | 0.94 (0.70–1.26) 0.670 |
| LDL cholesterol, mmol/L | 0.99 (0.79–1.24) 0.908 | 1.05 (0.79–1.34) 0.764 | 0.88 (0.64–1.22) 0.443 |
| HDL cholesterol, mmol/L | 0.81 (0.64–1.04) 0.097 | 0.91 (0.66–1.24) 0.534 | 0.63 (0.43–0.92) 0.018 |
Table 2 also displays the results from a multivariable multinomial logistic regression analysis, including BMI, and conventional risk factors for CVD such as age, male gender, diabetes mellitus type 2, active smoking, systolic blood pressure, total cholesterol, LDL and HDL cholesterol. BMI did not independently predict the extent of CAC: mild (OR: 1.01, 95% CI 0.98–1.02, p = 0.952, moderate (OR: 1.02, 95% CI 1.00–1.05, p = 0.103), and severe (OR: 1.00, 95% CI 0.97–1.02, p = 0.774) (Table 2, model 2).
The classical risk factors: age, male gender, diabetes mellitus type 2, family history, current smoking and systolic blood pressure independently predicted CAC (all p‐values ≤0.041) showing increasing odds ratios across CAC extent.
5. DISCUSSION
Within the present study, the association between BMI and CAC was investigated in a large single‐center outpatient coronary CT cohort. The main finding of this study was that no significant independent association was observed between BMI and presence and extent of CAC. Only age, male gender and classical cardiovascular risk factors, including diabetes mellitus type 2, family history, current smoking and systolic blood pressure showed an independent positive association with the extent of CAC. Based upon the findings of the present study, it might be reconsidered whether a high BMI should serve as an independent risk factor for coronary atherosclerosis in patients without known established CVD.
Previously, several studies have evaluated the association between body morphology and CAC. Allison et al studied a healthy population with a mean age of 57 and BMI of 27 kg/m2 where they described higher BMI as a significant predictor of CAC. 22 Comparable findings were reported in the Multi‐Ethnic study of Atherosclerosis (MESA) and the CARDIA study after adjustment for risk factors. 6 , 10 , 23 The Dallas Heart study also evaluated the association between obesity and prevalent atherosclerosis within a large population‐based study. They observed weak, although positive associations between BMI and prevalent CAC in contrast to strong associations for waist‐to‐hip ratio (WHR) and prevalent CAC. 24 Fujiyoshi et al also demonstrated a significant positive association of BMI and CAC within a multi‐ethnic study cohort independent of classical risk factors. 9 However, their study population existed of unselected male subjects with a relatively young age (40–49). A significant positive association between BMI and CAC was not observed in this study. A possible explanation for the lack of significant findings is the use of multinomial logistic regression analysis with AS classified into four categories as dependent variable, whereas other studies used binary logistic regression with presence or absence of CAC as dependent variable. 9 , 10 , 23
Comparison of several categories according to extent of CAC with a reference category without any calcification has the advantage that the strength of the association with risk factors can be evaluated according to extent of CAC. However, the sample size per category is smaller and thereby the power to detect weak or moderate associations decreases. Since the AS is a non‐normally distributed variable due to a large amount of zero variables, where log transformation cannot correct for, the possibilities for additional statistical analysis are limited.
There is also a subset of persons with obesity which are considered as “metabolically healthy” due to the fact that these patients have optimally regulated risk factors for CVD and whereby no increase of CV mortality is observed when compared to healthy subjects. 7 Since metabolically healthy obese individuals represent 10%–45% of the adult obese population, one could state that the persons with obesity within the present study population are especially patients with “metabolically healthy” obesity since cardiac CT is recommended as an alternative to stress imaging techniques in stable chest pain patients with a low to intermediate pre‐test probability for obstructive CAD. 8 , 25
Recent studies have described an “obesity paradox” indicating that after CVD has been diagnosed, mildly obese and persons with obesity have a similar or even decreased mortality risk compared to patients with normal weight. 15 , 16 This U‐shaped association has also been observed for body size and CAD. In 2012, Kovacic et al studied a large group of patients who underwent percutaneous coronary intervention and identified an inverse correlation between BMI and index lesion calcification. 15 Dangas et al also reported the inverse relationship between body size and CAC in patients with obstructive CAD. 16 With regard to the previous studies that observed a U‐shaped association of BMI with mortality as well as with CAD, no clear underlying pathophysiological mechanisms have been established which could explain this possible complex relationship. Interesting is the fact that besides the “obesity paradox” there are also reports of a “calcification paradox”, whereby reduced bone mineral density is associated with increased vascular calcification. It is predicted that persons with increased BMI have less osteoporosis and an inverse relationship between BMI and CAC may apply. 20 , 26 , 27 Within the present, which particularly included subjects without a known history of obstructive CAD, no protective effect of high BMI was observed. It could be assumed that this explains the controversial finding, since the previous studies reporting an inverse association between body size and CAC are particularly performed in patients with known CVD or obstructive CAD. Kim et al have discussed that one must be careful when interpreting the “obesity paradox” because of the fact that this phenomenon has been investigated in cohorts with established CVD where there is a difference in risk factors for recurrent coronary heart disease in normal weight or underweight patients in contrast to persons with obesity. 28 Selection of subjects in whom the disease is already present or has previously occurred can result in so‐called “index event bias” explaining paradoxical findings in medical research. Namely, due to conditioning on the disease or event, dependency between risk factors occurs despite these risk factors are independently associated within the general population. 29 , 30
Within the present study cohort of patients with a low‐to intermediate risk for obstructive CAD, no significant association was observed between BMI and CAC in contrast to age, male gender, diabetes mellitus type 2, family history, current smoking and systolic blood pressure. This could influence daily clinical practice, since it has to be reconsidered whether a high BMI should serve as an independent risk factor for coronary atherosclerosis in patients without known established CVD and low‐to‐intermediate risk for obstructive CAD.
The present study has some limitations that have to be mentioned. Firstly, the patient samples in the lowest and highest BMI categories were small compared to the other groups and therefore could lack statistical power. It is reasonable that especially in case of a very high BMI the physician might avoid a CT‐scan as diagnostic tool, because of the effects of obesity on image quality. However, this study reflects a cardiology outpatient clinic population since we included a large representative sample from a single‐center outpatient CT cohort. Secondly, only BMI was available within the present study population, whereby one could argue that BMI lacks discriminatory power to differentiate between body fat, metabolic state of this body fat and lean mass and thus to diagnose obesity. However, BMI is the most widely accepted obesity index and still used by the World Health Organization to define obesity. The European Society of Cardiology guidelines on CVD prevention also recommend BMI as the obesity index to predict CVD risk in routine practice. 2
6. CONCLUSION
Within the present study, no significant association was observed between BMI and CAC after adjustment for other risk factors. Therefore, it needs to be reconsidered whether a high BMI should concern as an independent risk factor for presence and extent of CAC in outpatient low‐to‐intermediate risk patients.
CONFLICT OF INTEREST
The authors declared no conflict of interest.
ACKNOWLEDGMENT
Sibel Altintas, Samanta van Workum, Madeleine Kok, Ivo A.P.G. Joosen, Mathijs O. Versteylen participated in the collection and analysis of data and helped draft the manuscript. Patricia J. Nelemans, Joachim E. Wildberger, Harry J.G.M. Crijns, Marco Das, Bas L.J.H. Kietselaer participated in the analysis of data and helped draft the manuscript. All authors read and approved the manuscript. The authors received no financial support for the research, authorship, and/or publication of this article.
Altintas S, van Workum S, Kok M, et al. BMI is not independently associated with coronary artery calcification in a large single‐center CT cohort. Obes Sci Pract. 2023;9(2):172‐178. 10.1002/osp4.636
REFERENCES
- 1. Clinical guidelines on the identification, evaluation, and treatment of overweight and obesity in adults‐‐the evidence report. National institutes of health. Obes Res 1998;6 (Suppl 2): 51s‐209s. [PubMed] [Google Scholar]
- 2. Piepoli MF, Hoes AW, Agewall S, et al. 2016 European guidelines on cardiovascular disease prevention in clinical practice: the sixth joint task force of the European society of cardiology and other societies on cardiovascular disease prevention in clinical practice (constituted by representatives of 10 societies and by invited experts) developed with the special contribution of the European association for cardiovascular prevention & rehabilitation (EACPR). Eur Heart J. 2016;37(29):2315‐2381. 10.1093/eurheartj/ehw106 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Sehested TS, Hansen TW, Olsen MH, et al. Measures of overweight and obesity and risk of cardiovascular disease: a population‐based study. Eur J Cardiovasc Prev Rehabil Official J Eur Soc Cardiol Work Groups Epidemiol Prev Cardiac Rehabilitation Exerc Physiology. 2010;17(4):486‐490. 10.1097/hjr.0b013e3283373f63 [DOI] [PubMed] [Google Scholar]
- 4. Twig G, Yaniv G, Levine H, et al. Body‐mass index in 2.3 million adolescents and cardiovascular death in adulthood. N. Engl J Med. 2016;374(25):2430‐2440. 10.1056/nejmoa1503840 [DOI] [PubMed] [Google Scholar]
- 5. Despres JP, Lemieux I. Abdominal obesity and metabolic syndrome. Nature. 2006;444(7121):881‐887. 10.1038/nature05488 [DOI] [PubMed] [Google Scholar]
- 6. Burke GL, Bertoni AG, Shea S, et al. The impact of obesity on cardiovascular disease risk factors and subclinical vascular disease: the Multi‐Ethnic Study of Atherosclerosis. Archives Intern Med. 2008;168(9):928‐935. 10.1001/archinte.168.9.928 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Meigs JB, Wilson PW, Fox CS, et al. Body mass index, metabolic syndrome, and risk of type 2 diabetes or cardiovascular disease. J Clin Endocrinol metabolism. 2006;91(8):2906‐2912. 10.1210/jc.2006-0594 [DOI] [PubMed] [Google Scholar]
- 8. Goncalves CG, Glade MJ, Meguid MM. Metabolically healthy obese individuals: key protective factors. Nutr. 2016;32(1):14‐20. 10.1016/j.nut.2015.07.010 [DOI] [PubMed] [Google Scholar]
- 9. Fujiyoshi A, Sekikawa A, Shin C, et al, A cross‐sectional association of obesity with coronary calcium among Japanese, Koreans, Japanese Americans, and U.S. whites. European heart journal cardiovascular Imaging. 2013;14(9):921‐927. 10.1093/ehjci/jet080 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Lee CD, Jacobs DR, Jr. , Schreiner PJ, Iribarren C, Hankinson A. Abdominal obesity and coronary artery calcification in young adults: the coronary artery risk development in young adults (CARDIA) study. Am J Clin Nutr. 2007;86(1):48‐54. 10.1093/ajcn/86.1.48 [DOI] [PubMed] [Google Scholar]
- 11. Artham SM, Lavie CJ, Milani RV, Ventura HO. Obesity and hypertension, heart failure, and coronary heart disease‐risk factor, paradox, and recommendations for weight loss. Ochsner J. 2009;9:124‐132. [PMC free article] [PubMed] [Google Scholar]
- 12. Romero‐Corral A, Montori VM, Somers VK, et al. Association of bodyweight with total mortality and with cardiovascular events in coronary artery disease: a systematic review of cohort studies. Lancet (London, England). 2006;368(9536):666‐678. 10.1016/s0140-6736(06)69251-9 [DOI] [PubMed] [Google Scholar]
- 13. Kaneko H, Yajima J, Oikawa Y, et al. Obesity paradox in Japanese patients after percutaneous coronary intervention: an observation cohort study. J Cardiol. 2013;62(1):18‐24. 10.1016/j.jjcc.2013.02.009 [DOI] [PubMed] [Google Scholar]
- 14. Yusuf S, Hawken S, Ounpuu S, et al. Obesity and the risk of myocardial infarction in 27, 000 participants from 52 countries: a case‐control study. Lancet (London, England). 2005;366(9497):1640‐1649. 10.1016/s0140-6736(05)67663-5 [DOI] [PubMed] [Google Scholar]
- 15. Kovacic JC, Lee P, Baber U, et al. Inverse relationship between body mass index and coronary artery calcification in patients with clinically significant coronary lesions. Atherosclerosis. 2012;221(1):176‐182. 10.1016/j.atherosclerosis.2011.11.020 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Dangas GD, Maehara A, Evrard SM, et al. Coronary artery calcification is inversely related to body morphology in patients with significant coronary artery disease: a three‐dimensional intravascular ultrasound study. Eur Heart J Cardiovascular Imaging. 2014;15(2):201‐209. 10.1093/ehjci/jet139 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Shaffer JR, Kammerer CM, Rainwater DL, et al. Decreased bone mineral density is correlated with increased subclinical atherosclerosis in older, but not younger, Mexican American women and men: the San Antonio Family Osteoporosis Study. Calcif Tissue Int. 2007;81(6):430‐441. 10.1007/s00223-007-9079-0 [DOI] [PubMed] [Google Scholar]
- 18. Hou ZH, Lu B, Gao Y, et al. Prognostic value of coronary CT angiography and calcium score for major adverse cardiac events in outpatients. JACC Cardiovasc Imaging. 2012;5(10):990‐999. 10.1016/j.jcmg.2012.06.006 [DOI] [PubMed] [Google Scholar]
- 19. Detrano R, Guerci AD, Carr JJ, et al. Coronary calcium as a predictor of coronary events in four racial or ethnic groups. N. Engl J Med. 2008;358(13):1336‐1345. 10.1056/nejmoa072100 [DOI] [PubMed] [Google Scholar]
- 20. Budoff MJ, Shaw LJ, Liu ST, et al. Long‐term prognosis associated with coronary calcification: observations from a registry of 25, 253 patients. J Am Coll Cardiol. 2007;49(18):1860‐1870. 10.1016/j.jacc.2006.10.079 [DOI] [PubMed] [Google Scholar]
- 21. Agatston AS, Janowitz WR, Hildner FJ, Zusmer NR, Viamonte M, Jr. , Detrano R. Quantification of coronary artery calcium using ultrafast computed tomography. J Am Coll Cardiol. 1990;15(4):827‐832. 10.1016/0735-1097(90)90282-t [DOI] [PubMed] [Google Scholar]
- 22. Allison MA, Michael Wright C. Body morphology differentially predicts coronary calcium. Int J Obes Relat Metab Disord J Int Assoc Study Obes. 2004;28(3):396‐401. 10.1038/sj.ijo.0802571 [DOI] [PubMed] [Google Scholar]
- 23. Kronmal RA, McClelland RL, Detrano R, et al. Risk factors for the progression of coronary artery calcification in asymptomatic subjects: results from the Multi‐Ethnic Study of Atherosclerosis (MESA). Circulation. 2007;115(21):2722‐2730. 10.1161/circulationaha.106.674143 [DOI] [PubMed] [Google Scholar]
- 24. See R, Abdullah SM, McGuire DK, et al. The association of differing measures of overweight and obesity with prevalent atherosclerosis: the Dallas Heart Study. J Am Coll Cardiol. 2007;50(8):752‐759. 10.1016/j.jacc.2007.04.066 [DOI] [PubMed] [Google Scholar]
- 25. Montalescot G, Sechtem U, Achenbach S, et al. ESC guidelines on the management of stable coronary artery disease: the Task Force on the management of stable coronary artery disease of the European Society of Cardiology. Eur Heart J. 2013;34:2949‐3003. [DOI] [PubMed] [Google Scholar]
- 26. Persy V, D'Haese P. Vascular calcification and bone disease: the calcification paradox. Trends Mol Med. 2009;15(9):405‐416. 10.1016/j.molmed.2009.07.001 [DOI] [PubMed] [Google Scholar]
- 27. Sage AP, Tintut Y, Demer LL. Regulatory mechanisms in vascular calcification. Nat Rev Cardiol. 2010;7(9):528‐536. 10.1038/nrcardio.2010.115 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Kim SH, Despres JP, Koh KK. Obesity and cardiovascular disease: friend or foe? Eur Heart J. 2015;37(48):3560‐3568. 10.1093/eurheartj/ehv509 [DOI] [PubMed] [Google Scholar]
- 29. Dahabreh IJ, Kent DM. Index event bias as an explanation for the paradoxes of recurrence risk research. JAMA. 2011;305(8):822‐823. 10.1001/jama.2011.163 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Smits LJ, van Kuijk SM, Leffers P, Peeters LL, Prins MH, Sep SJ. Index event bias‐a numerical example. J Clin Epidemiol. 2013;66(2):192‐196. 10.1016/j.jclinepi.2012.06.023 [DOI] [PubMed] [Google Scholar]