Mammography as a screening tool for peripheral vascular disease.

BACKGROUND

Peripheral vascular disease (PVD) is a progressive and debilitating disease often diagnosed only when patients become symptomatic. Currently there are no widespread screening tests available for the early detection of PVD. Patients with diabetes and coronary artery disease are known to have a higher incidence of PVD. Prior studies have indicated that benign vascular calcifications seen on routine screening mammogram are more prevalent in women with diabetes and coronary artery disease. The same association has not been shown for women with PVD. The purpose of this study was to identify an association between benign vascular calcifications identified on mammography and PVD. If such an association exists then screening mammography, already widely used as a screening tool for breast cancer, may identify women at high risk for PVD.

METHODS

To determine the incidence of vascular calcifications in our general screening population we prospectively evaluated consecutive routine screening mammograms for the presence of benign vascular calcifications. We then identified a population of women with PVD by using a computerized database maintained by the Division of Vascular Surgery. The population of women identified with PVD was categorized further to identify those women who had received routine screening mammogram within our hospital system. These mammograms were reviewed retrospectively with particular attention to the presence of vascular calcifications. All mammograms were reviewed by our dedicated mammographers. Statistical analysis of the study group using the chi-square test was performed to determine the association of PVD and mammographic vascular calcifications.

RESULTS

Prospective evaluation of 645 women undergoing consecutive routine screening mammography identified 123 (19%) with benign vascular calcifications. By using our PVD computerized database between the years 2002 and 2004 we identified 763 women with PVD. Of this group only 121 (15%) had undergone a routine screening mammogram in our hospital system. These mammograms were reviewed retrospectively by the same group evaluating the screening mammograms. On evaluation vascular calcifications were identified in 42% (51 of 121) of these women with PVD. Statistical analysis was performed using the chi-square test, odds ratio, and relative risk. A highly significant association was identified between PVD and the presence of mammographic vascular calcifications (P > .001). With a confidence interval of 95% the presence of benign vascular microcalcifications on routine screening mammogram identifies a significant risk for PVD with an odds ratio of 3.06. We showed through our analysis that women with calcifications are 2.19 times more likely to have PVD if microcalcifications are present. By using vascular calcifications identified on screening mammography as a means to identify women with PVD the sensitivity and specificity are 42% and 80%, respectively. The positive predictive value and the negative predictive value are 29% and 88%, respectively, with an accuracy of 75%.

CONCLUSIONS

This initial study indicates that the presence of vascular calcifications identified on routine screening mammogram is significantly higher in women with PVD and the lack of vascular calcifications on screening mammography correlates well with a negative history of PVD. We have identified a significant association with PVD and vascular calcifications in our patient population. Further studies are indicated to determine if screening mammography may become a widespread inexpensive screening tool to identify women at risk for PVD. Additional studies are underway at our institution to evaluate the association of PVD, diabetes and coronary artery disease, and vascular calcifications identified on routine screening mammography.