Association of Breast Arterial Calcifications,Metabolic Syndrome,and the 10-Year Coronary Heart Disease Risk: A Cross-Sectional Case-Control Study

Abstract

Objective:

Breast arterial calcifications (BACs) are common but underreported findings on mammography. The purpose of this study was to examine the relationship between BACs, metabolic syndrome (MS), and coronary heart disease (CHD) risk.

Research Design and Methods:

This was a case-control study. A total of 101 women with BACs and 101 age-matched controls without BACs were included in the analysis. BACs were calculated using the mammography of both breasts, and then the total BACs were summed. MS was defined by using criteria from the American Heart Association/National Heart, Lung, and Blood Institute (AHA/NHLBI). The 10-year estimated CHD relative risk (10-year CHD risk) was calculated based on the Framingham Scores. Metabolic characteristics were compared between the BAC positive group and the BAC negative group.

Results:

Diabetes (P=0.039) and MS (P=0.043) were more frequently observed in the BAC positive group than in the BAC negative group. The only independent factor was MS (P=0.04). The 10-year CHD risk was higher in the BAC positive group than in the negative group (P=0.007). Furthermore, the more extensive the BACs, the greater increase in the 10-year CHD risk (r=0.167, P=0.018).

Conclusions:

Findings of BACs on mammogramy are associated with an increased risk of MS and the calculated risk of CHD in women older than 40 years of age.

Mammography as a screening test for the early detection of breast cancer is recommended for all women beginning at the ages of 40 to 50, because some breast cancers in the early stage have been characterized by certain identifying masses and/or clustered calcifications and microcalcifications.^1
–3 However, breast arterial calcifications (BACs), commonly seen on mammography, have not been given attention since they are typically benign findings. Arterial calcifications have been associated with atherosclerotic cardiovascular disease. Atherosclerotic breast artery may be clinically an important sign of vascular risk. Previous studies have highlighted the potential of BACs as a valuable tool in identifying women with dysmetabolic conditions^4,5 and/or a risk of coronary artery disease (CAD).^6
–8 Baum's study investigated the prevalence of BACs in a cohort of diabetic women and reported that the incidence of BAC in the diabetic group was higher than that in the control group.⁴ Other studies showed that the presence of BACs on routine screening mammography is associated with the presence of hypertension and CAD.^9,10 In addition, BACs were suggested to be a marker for increased risk of cardiovascular morbidity and mortality.^8,11,12 However, previous studies have had some limitations, such as relatively poor resolution of mammography and unadjusted confounding factors such as old age and obesity.^7,10

CAD is a leading cause of mortality and morbidity worldwide, and its incidence is increasing.¹³ Therefore, early detection and prediction of CAD are very important factors. Metabolic syndrome (MS) as a predictor value of CAD represents a constellation of physiological risk factors that include increased waist circumference, blood pressure elevation, low high-density lipoprotein cholesterol (HDL-C), high triglycerides, and hyperglycemia.^14,15 The Framingham risk score is another tool that can be used to evaluate CAD. This multivariable statistical model can be conveniently used to estimate the 10-year coronary heart disease (CHD) risk, given age, sex, blood pressure, current smoking status, total cholesterol (TC), HDL-C, and blood glucose levels or history of diabetes.^16,17 Although both MS and the 10-year CHD risk are well-known predictors of CAD, little is known about the prevalence of MS and the estimated 10-year CHD risk in women with BAC. As there is a growing interest in BAC, knowledge of the clinical and metabolic features of BAC on mammography can provide valuable information.

We hypothesized that a prevalence of MS and the 10-year CHD risk are different in women with and without BAC after matching for age. Therefore, the aims of this cross-sectional study were to assess the prevalence of BACs on digital mammography and to investigate whether BACs are related to a prevalence of MS and the 10-year CHD risk after adjusting for confounding factors in women.

Research Design and Methods

Participants

This was a cross-sectional, case-control study. The study was conducted in the Health Promotion Center at Pusan National University Hospital, from June 2008 to April 2009. A total of 995 women older than 40 years of age underwent screening digital mammography and biochemical tests as part of an annual health check-up. After screening, 2 patients were excluded because of drug abuse, 34 were excluded because they were diagnosed with angina (n=10), cardiovascular disease (n=5), cerebral vascular disease (n=5), or cancer (n=14). This yielded a first sample of 959 subjects. Among these subjects, 101 women with BACs (BAC positive group) and 101 age-matched controls without BACs (BAC negative group) were included in the analysis. Written informed consent was obtained from each subject before enrollment in this study. The study was approved by the institutional review board at the Medical Research Institute, Pusan National University, and was performed in accordance with the principles of the Declaration of Helsinki.

Measurements

Height and weight were measured by an automatic height-weight scale to the nearest 0.1 cm and 0.1 kg, respectively, and the body mass index (BMI) was calculated as weight (kg) divided by the square of height (m²). The waist circumference measurements were taken at the end of normal expiration and to the nearest 0.1 cm, measuring from the narrowest point between the lower borders of the rib cage and iliac crest. The percentages of body fat and total fat mass were measured by bioelectric impedance analysis (Inbody 3.0; Biospace, Seoul, Korea). Blood pressures were detected by an automated noninvasive blood pressure device (BP-203RV II; Colin Corp., Aichi, Japan).

Blood samples were collected from the antecubital vein to determine serum concentrations TC, triglyceride, HDL-C, low-density lipoprotein cholesterol (LDL-C), glucose, homocysteine, and high-sensitivity C-reactive protein (hs-CRP) from each participant after a 12-hour fast. These parameters are risk factors for either atherosclerosis or cardiovascular disease. All biochemical analyses were carried out within 2 hours of blood sampling. The serum biochemical tests were conducted by Modular DPE (Roche co., Ltd, Basel, Switzerland) and commercially available kits. Fasting insulin was determined using a radioimmunoassay (Diagnostic Product Corporation, Los Angeles, California) with antibody-coated tubes. The mean intra- and interassay CVs were 4.2 and 6.3%, respectively. HOMA-IR (homeostasis assessment of insulin resistance) {fasting insulin (μIU/mL)×fasting glucose (mg/dL)/405}¹⁸ and QUICKI (quantitative insulin-sensitivity check index) {1/ [(log fasting insulin (μIU/mL)×log fasting glucose (mg/dL)]}¹⁹ were used for estimating insulin sensitivity. An experienced dietitian calculated the quantities of mean calcium and alcohol consumption based on a semiquantitative food frequency questionnaire.²⁰

All digital mammography (mediolateral oblique and craniocaudal views of both breasts) were obtained with a MAMMOMAT Novation^DR (Siemens AG co., Ltd, Berlin, Germany). Each mammography was analyzed for the presence of BACs with the criteria of Kemmeren et al. ⁶ by one radiologist who was blinded to the participants' history. BACs were respectively counted on each mammography of the right and the left; then the total BACs were calculated by summing the counts. If two branches originated from the same source, they were considered as one (Fig. 1).

FIG. 1.

Typical breast arterial calcifications (A and B) in mammogram.

Definition of metabolic syndrome

The 2005 American Heart Association/National Heart, Lung, and Blood Institute (AHA/NHLBI) criteria¹⁴ were used for the diagnosis of MS. The criteria recommend a lower waist circumference cutoff for the Asian population than the Caucasian population because Asians are predisposed to metabolic abnormalities at a lower waist circumference. Therefore, we defined central obesity as a waist circumference ≥85 cm in females, in accordance with the suggestion of the Korean Society for the Study of Obesity.²¹

The 10-year CHD risk

The 10-year estimated CHD relative risk was calculated based on the Framingham Scores for participants, aged 30 to 74 years.¹⁶ This estimate was used for each participant to determine the probability of having a heart attack or dying of CHD during a 10-year period based on the categories of age, sex, smoking status, total cholesterol, and systolic blood pressure.

Statistical analysis

The prevalence of coronary heart disease from a previous study²² was used to develop the statistical validity of this study. To test the hypothesis that the prevalence of MS would be at least 15% greater in cases than in controls, a sample size of 75 subjects per group would provide 80% power to detect this difference with an α of 5%. The D'Agostino-Pearson test, paired t-test, and Wilcoxon signed rank test were applied to comparisons for baseline anthropometric, nutritional, and metabolic characteristics and the calculated 10-year CHD risk (%) between the BAC positive group and the BAC negative group. The McNemar test was used to compare the prevalence of present illness, each component of MS, and the metabolic syndrome between groups, using AHA/NHLBI criteria. Conditional logistic regression analysis was used to assess independent associations between BACs and present illness, each component of MS, and MS by AHA/NHLBI criteria. A P-value less than 0.05 was deemed statistically significant. SPSS 12.0 for Windows (SPSS, Chicago, Illinois) was used for all statistical analyses.

Results

Characteristics of study subjects

The demographic, anthropometric, and dietary characteristics of the study subjects are shown in Table 1. The mean age of all participants was 58.9 years, with a range of 41 to 78 years, and the mean BMI was 23.9 kg/m², with a range of 18.9 to 30.8 kg/m². The mean waist circumference was 81.0 cm in all participants. There were no significant differences in baseline BMI, percent body fat, waist circumference, calcium intake, and alcohol intake between the BACs positive group and the BACs negative group (Table 1). Also, the number of current smokers did not differ between the BAC positive group and the BAC negative group (5% vs. 3%, P>0.05).

Table 1.

Anthropometrics and Nutritional Characteristics of the Study Participants

Variables	BACs (n=101)	Control (n=101)	P-value
Age (years)	58.9±7.0	58.9±7.0	1.000
BMI (kg/m²)	23.7±3.0	24.1±2.3	0.125
Percent body fat (%)	31.4±4.5	32.0±3.4	0.170
Waist circumference (cm)	81.0±7.8	81.1±6.8	0.629
Calcium intake (mg/day)	1088.7 (363.2–1930.5)	1041.7 (564.0–1884.0)	0.261
Alcohol intake (kcal/day)	0.0 (0.0–1429.0)	0.0 (0.0–357.0)	0.137

Data are mean±SD or median (range) or number (%).

P-value by paired t-test or Wilcoxon signed rank test.

BACs, breast arterial calcifications.

BACs, the 10-year CHD risk (%), and metabolic characteristics

The total number of BACs in the right and left breast at mammography and the percentage of participants were as follows: BAC=1, 59.4%, BAC=2, 31%, BAC=3, 2%, BAC=4, 4%, and BAC≥5, 4%. The 10-year CHD risk (%) was remarkably higher in the BAC positive group than in the BAC negative group (P=0.007, Table 2). However, triglyceride, HDL-C, systolic blood pressure, diastolic blood pressure, glucose, HOMA-IR, QUICKI, hs-CRP, and homocysteine were not significantly different between groups (Table 2). A conditional logistic regression analysis was performed to determine the relationship of the 10-year CHD risk and the number of BACs. Based on a count of BAC 0, the odds ratio for above 10% of 10-year CHD risk increased with each count increment in BACs (Table 3). Also, gradation of BAC was significantly related to MS (r=0.149, P=0.035) and the 10-year CHD risk (r=0.167, P=0.018).

Table 2.

Metabolic Characteristics and the Calculated 10-Year CHD Risk of the Study Participants

Variables	BACs (n=101)	Control (n=101)	P-value
Triglyceride (mg/dL)	112.0 (40.0–436.0)	96.0 (32.0–483.0)	0.135
HDL cholesterol (mmHg)	52.0 (26.0–88.0)	53.0 (21.0–104.0)	0.452
Systolic BP (mmHg)	122.0±18.6	122.0±8.0	0.994
Diastolic BP (mmHg)	75.0±11.7	75.2±10.3	0.913
Fasting glucose (mg/dL)	92.0 (70.0–231.0)	91.0 (71.0–139.0)	0.185
HOMA-IR	1.0 (0.1–23.4)	1.0 (0.2–4.0)	0.272
QUICKI	0.8 (–3.1–4.5)	0.8 (–9.6–3.7)	0.290
Hs-CRP (mg/L)	0.1 (0.0–3.8)	0.1 (0.0–1.0)	0.134
Homocysteine (μ mol/L)	6.0 (1.0–54.0)	1.0 (0.2–4.0)	0.073
The 10-year CHD risk (%)^*	7.0 (1.0–32.0)	6.0 (1.0–24.0)	0.007

Coronary heart disease 10-year risk.¹⁶

Data are mean±SD or median (range).

HOMA-IR=[fasting insulin (μIU/mL)×fasting glucose (mg/dL)]/405.

QUICKI=1/[log fasting insulin (μIU/mL)×log fasting glucose (mg/dL)].

P-value by paired t-test or Wilcoxon signed rank test.

BP, blood pressure; HOMA-IR, homeostasis assessment of insulin resistance; QUICKI, quantitative insulin-sensitivity check index; Hs-CRP, high sensitivity C-reactive protein; CHD, coronary heart disease.

Table 3.

Odds Ratio and 95% Confidence Interval for Above 10% of the 10-year CHD Risk ^* to the Increasing Number of Breast Arterial Calcification

Variables	No. of subjects	Odds ratio	95% Confidence intervalI	P-value
Counts of BACs
BAC 0	101	1.00
BAC 1	60	1.14	0.63–2.10	0.78
BAC 2	31	2.17	1.06–4.42	0.03
BAC ≥3	10	4.90	1.38–17.38	0.01

Coronary heart disease 10-year risk.¹⁶

P-value by conditional logistic regression.

BACs, co-morbidity, and the metabolic syndrome

Diabetes (P=0.039) and MS by AHA/NHLBI criteria (P=0.043) were more frequently observed in the BAC positive group than in the BAC negative group (Table 4).

Table 4.

Present Illness, Metabolic Syndrome and Criteria Prevalence

Variables	BACs (n=101)	Control (n=101)
Present illness
Diabetes	10.9	3.0^*
Hypertension	30.7	27.7
Dyslipidemia	5.9	7.9
Each component of metabolic syndrome
Increased waist circumference (≥85 cm)	29.7	27.7
Raised blood pressure (≥130/85 mmHg)^†	46.5	46.5
Raised fasting blood glucose (≥100 mg/dL)^‡	31.7	26.7
Raised triglycerides (≥150 mg/dL)	31.7	27.7
Low HDL cholesterol (<50 mg/dL)	46.5	47.5
AHA/NHLBI criteria	39.6	26.7^§

P=0.039; ^§ P=0.043 by McNemar test.

†

47% subjects in the BAC group were taking a medicine compared with 59% in the control group. ^‡29% subjects in the BAC group were taking a medicine compared with 11% in the control group.

Data are expressed as percent.

AHA/NHLBI, American Heart Association/National Heart, Lung, and Blood.

Independent factor of BACs

MS by AHA/NHLBI criteria was the only independent factor relating to the presence of BACs as a result of conditional logistic regression analyses. Women with BACs were found to have a significantly higher risk of MS than women without BACs (odds ratio [OR], 2.79; 95% confidence interval [CI-, 1.03–7.50, Table 5).

Table 5.

Co-Morbidities and Conditions Associated with Breast Arterial Calcifications

Variables	Odds ratio	95% CI	P-value
Present illness
Diabetes			0.35
No	1.00
Yes	1.55	0.62–3.87
Hypertension			0.54
No	1.00
Yes	1.24	0.62–2.44
Dyslipidemia			0.74
No	1.00
Yes	1.19	0.44–3.22
Each component of metabolic syndrome
Raised blood pressure			0.33
No	1.00
Yes	0.71	0.35–1.42
Raised fasting blood glucose			0.36
No	1.00
Yes	0.73	0.38–1.43
Raised triglycerides			0.44
No	1.00
Yes	0.77	0.40–1.48
Low HDL cholesterol			0.22
No	1.00
Yes	0.71	0.41–1.23
AHA/NHLBI criteria			0.04
No	1.00
Yes	2.79	1.03–7.50

P-value by conditional logistic regression.

AHA/NHLBI, American Heart Association/National Heart, Lung, and Blood.

Conclusions

BACs are calcium deposits found in the media layer of peripheral arterioles. This finding on mammography was overlooked until Baum et al. ⁴ suggested that vascular calcification on mammography may clinically imply unsuspected diabetes mellitus because women with BACs were more likely to have type 2 diabetes than women without BACs (OR=14.8, 95% CI=6.6–33.1). In this cross-sectional case-control study, using tools of MS defined by AHA/NHLBI criteria and the calculated 10-year CHD risk, we observed that BACs are associated with an increased prevalence and a significant risk of CHD. As a result, the prevalence of BACs among 959 women was 9.5% (101 women), a finding similar to previous results documenting a range from 3 to 17.5%.^4,7,10,11 Among the metabolic factors, we found that BACs were only associated with diabetes and MS by the AHA/NHLBI definition. In addition, after matching for age, the counts of BACs were related to HOMA-IR (homeostasis assessment of insulin resistance) (r=0.41, P<0.001; result not presented) as well as MS and the 10-year CHD risk. These results may suggest a possible association between BACs and insulin resistance. However, in conditional logistic regression analyses, diabetes mellitus was not an independent factor. Abdominal obesity, high fasting glucose, and hypertriglyceridemia, which are the components of MS, were more prevalent in the BAC positive group, although there was no statistically significant difference. MS by the AHA/NHLBI definition was the only dependent risk factor in the results of the conditional logistic regression analyses. Those data collectively suggest that BACs are not related to a specific disease or component of metabolic syndrome but may instead be related to the summation of various metabolic components. This study's strength is that, for the first time, MS as a risk factor of CHD was considered with BACs.

In the BAC positive group, the 10-year CHD risk is significantly higher than that of the BAC negative group (P=0.007). The largest published cohort study on BACs and arteriosclerotic-related disease outcomes was performed among 12,761 women of the Kaiser Permanente Northern California Region.⁸ This study reported significant relationships between BACs and hypertension, stroke, and myocardial infarction. As a result, screening mammography as a tool of early detection of cardiovascular disease was suggested. The present study also suggests that BACs could be a predictive factor of coronary heart disease.^8,11 On the other hand, the prevalence of BACs was similar to that obtained in the Israeli report on 187 women with coronary artery disease and 132 women with a normal coronary angiogram.²² However, in that study, the prevalence of BACs in the group of women with coronary artery disease and those with a normal coronary angiogram was 43.9% and 37.1%, respectively. Those results are much higher than those in previous reports, and so other confounding factors may be related. Recently, a prospective study was conducted on BACs and cardiac catheterization.²⁴The authors of this study did not observe a correlation between BACs and coronary angiography-detected CHD. However, they found a significant association between BACs and cardiovascular events (myocardial infarction, transient ischemic attack, and stroke). The reason may be that the origin of CHD in women is more complex, and possible mechanisms include microvascular and endothelial dysfunction. Interestingly, in the present study, as the BAC counts rose, the 10-year CHD risk increased (Table 3).

Our study found no significant difference between BACs and hypertension, diabetes, and hypercholesterolemia, which are the risk factors of coronary heart disease. In spite of previous studies,^5,6 our findings support the fact that the prevalence of CHD was increased not by a separate risk component, but by a summation of risk factors.

Since women who have BACs are generally older and age is a significant factor in BAC, ^4,10 controls were matched to cases by age in the present study. This should be considered an important strength of our study. Another strength of our study was our use of the up-to-date full-field digital mammographic system which provides exceptional image quality. However, this study also had some limitations. Our study was limited by its cross-sectional study design and the relatively small size of our study sample. Although we were able to determine a relationship between BAC and CHD, the precedent cause remained uncertain. Further investigation with a cohort study is warranted. Another limitation is the possibility that there may have been a co-related effect from the medication used to treat the participants' preexisting diseases, such as diabetes, hypertension, and hypercholesterolemia. This may influence the relationship between BACs and the risk factors of MS and the 10-year CHD risk. Another of our concerns was selection bias among controls. Because population-based sampling is not practicable in our setting, we used individuals participating in the health screening procedures.

In conclusion, BACs were found to be related to MS and the 10-year CHD risk. Prospective studies are needed to further assess whether BACs can be considered an effective screening tool for CHD in women over 40 years of age.

Footnotes

Acknowledgments

Authors' contribution: M.J.B. analyzed and interpreted data and wrote the first draft of the article. S.Y.L. helped conceptualize the study idea, interpreted data, and contributed to the drafts of the article. J.J.M. contributed to the data analyses, interpreted data, and reviewed the drafts of the article. Y.J.K., J.G.L., D.W.J., Y.H.Y., Y.H.C., and E.J.C. contributed to the interpretation of data and reviewed drafts of the article. K.S.C. guided the process of data analysis and interpretation. All authors were involved in the review and editing of the article. None of the other coauthors had a personal or financial conflict of interest.

Author Disclosure Statement

No competing financial interests exist.

References

Bartow

, Pathak

, Mettler

. Radiographic microcalcification and parenchymal patterns as indicators of histologic “high-risk” benign breast disease. Cancer, 1990; 66:1721–1725.

Feig

, D'Orsi

, Hendrick

et al. American College of Radiology guidelines for breast cancer screening. Am J Roentgenol, 1998; 171:29–33.

Galkin

, Frasca

, Feig

et al. Non-calcified breast particles: A possible new marker of breast cancer. Invest Rad, 1982; 17:119–122.

Baum

, Comstock

, Joseph

. Intramammary arterial calcifications associated with diabetes. Radiology, 1980; 136:61–62.

Cetin

, Cetin

, Tamer

et al. Breast arterial calcifications associated with diabetes and hypertension. J Diabetes Complications, 2004; 18:363–366.

Kemmeren

, van Noord

, Beijerinck

et al. Arterial calcification found on breast cancer screening mammograms and cardiovascular mortality in women: The DOM Project. Doorlopend Onderzoek Morbiditeit en Mortaliteit. Am J Epidemiol, 1998; 147:333–341.

Crystal

, Crystal

, Leor

et al. Breast artery calcium on routine mammography as a potential marker for increased risk of cardiovascular disease. Am J Cardiol, 2000; 86:216–217.

Iribarren

, Go

, Tolstykh

et al. Breast vascular calcification and risk of coronary heart disease, stroke, and heart failure. J Womens Health (Larchmont), 2004; 13:381–389.

Maas

, van der Schouw

, Beijerinck

, Deurenberg

, Mali

, van der Graaf

. Arterial calcifications seen on mammograms: Cardiovascular risk factors, pregnancy, and lactation. Radiology, 2006; 240:33–38.

10.

Van Noord

PAH

, Beijerinck

, Kemmeren

et al. Mammograms may convey more than breast cancer risk: Breast arterial calcification and arteriosclerotic related diseases in women of the DOM cohort. Eur J Cancer Prev, 1996; 5:483–487.

11.

Kataoka

, Warren

, Luben

et al.

How predictive is breast arterial calcification of cardiovascular disease and risk factors when found at screening mammography?

Am J Roentgenol, 2006; 187:73–80.

12.

Rotter

, Schnatz

, Currier

et al. Breast arterial calcifications (BACs) found on screening mammography and their association with cardiovascular disease. Menopause, 2008; 15:276–281.

13.

Lopez

, Mathers

, Ezzati

et al. Global and regional burden of disease and risk factors, 2001: Systematic analysis of population health data. Lancet, 2001; 367:1747–1757.

14.

Grundy

, Cleeman

, Daniels

et al. Diagnosis and management of the metabolic syndrome: An American Heart Association/National Heart, Lung, and Blood Institute Scientific Statement. Circulation, 2005; 112:2735–2752.

15.

Zimmet

, Magliano

, Matsuzawa

et al. The metabolic syndrome: A global public health problem and a new definition. J Atheroscler Thromb, 2005; 12:295–300.

16.

Wilson

, D'Agostino

, Levy

et al. Prediction of Coronary Heart Disease Using Risk Factor Categories. Circulation, 1998; 97:1837–1847.

17.

D'Agostino

Sr , Vasan

, Pencina

et al. General Ccrdiovascular risk profile for use in primary care: The Framingham Heart Study. Circulation, 2008; 117:743–753.

18.

Matthews

, Hosker

, Rudenski

et al. Homeostasis model assessment: Insulin resistance and beta-cell function from fasting plasma glucose and insulin concentration in man. Diabetologia, 1985; 28:412–419.

19.

Katz

, Nambi

, Mather

et al. Quantitative insulin sensitivity check Index: A simple, accurate method for assessing insulin sensitivity in humans. J Clin Endocrinol Metab, 2000; 85:2402–2410.

20.

Hong

, Choi

, Lee

et al. Development and validation of a semi-quantitative food frequency questionnaire to assess diets of Korean type 2 diabetic patients. Korean Diabetes J, 2010; 34:32–39.

21.

Lee

, Park

, Kim

et al. Appropriate waist circumference cutoff points for central obesity in Korean adults. Diabetes Res Clin Pract, 2007; 75:72–80.

22.

Schnatz

, Marakovits

, O'Sullivan

. The association of breast arterial calcification and coronary heart disease. Obstet Gynecol, 2011; 117:233–241.

23.

Henkin

, Abu-Ful

, Shai

et al. Lack of association between breast artery calcification seen on mammography and coronary artery disease on angiography. J Med Screen, 2003; 10:139–142.

24.

Zgheib

, Buchbinder

, Abi Rafeh

et al. Breast arterial calcifications on mammograms do not predict coronary heart disease at coronary angiography. Radiology, 2010; 254:367–373.