Association of Breast Arterial Calcification With Stroke and Angiographically Proven Coronary Artery Disease

A Meta-analysis

Xuezhi Jiang, MD, FACOG, NCMP; Maureen Clark, BS; Rupali K. Singh, MD; Alex Juhn, BA; Peter F. Schnatz, DO, FACOG, FACP, NCMP


Menopause. 2015;22(2):136-143. 

In This Article



PubMed, Google Scholar,, and Ovid were searched up to October 2013 using the following search terms: "breast arterial calcification," "breast vascular calcification," "coronary artery disease," "coronary heart disease," "cardiovascular disease," "myocardial infarction," "angina pectoris," "abnormal coronary angiography," and "stroke." In addition, we manually searched the reference lists of relevant review articles and meta-analyses but did not identify additional articles.

Study Selection

The literature search was conducted independently by two authors to identify the studies that met the following inclusion criteria: (1) written in English; (2) quantitatively assessed the correlation between BAC and CVD (CAD and/or stroke); and (3) assured that CAD (defined as at least one stenosis of 50% or more in any major coronary artery) was diagnosed via coronary angiography. Disagreements between the two authors regarding inclusion and exclusion criteria were resolved by a senior author.

For this meta-analysis to be valid, we focused on universal outcomes using objective diagnostic criteria. Some studies analyzed CAD diagnosed by clinical presentation; some studies analyzed multiple outcomes separately (including myocardial infarction, angina, abnormal angiogram, and coronary artery bypass graft) but did not include the total number of women with CAD because one woman can have more than one of the aforementioned outcomes. A clinical diagnosis of CAD does not necessarily indicate an angiographically proven CAD; thus, we based our inclusion criteria for CAD on whether or not it was diagnosed by coronary angiography.

The methodology conformed to MOOSE (meta-analysis of observational studies in epidemiology) guidelines.[15] This meta-analysis was conducted using SAS version 9.2. The primary outcomes assessed were angiographically diagnosed CAD and stroke. Heterogeneity of estimated effects across studies was assessed by Cochran's Q test and I2 statistics to determine the suitability of the studies to be pooled for the meta-analysis. Between-study heterogeneity was considered significant for P < 0.10.[16] Odds ratios (ORs) for individual studies, along with pooled ORs and 95% CIs, were calculated and reported by forest plot. The relatively small number of studies in this meta-analysis makes it difficult to determine publication bias by visual inspection of a funnel plot; hence, a normal quantile plot was used for the assessment of potential publication bias. Most data points falling at or near a straight oblique line and within the 95% CI bands in a normal quantile plot would suggest no evidence of severe publication bias.[17]