Prognostic Utility of Stress Testing and Cardiac Biomarkers in Menopausal Women at Low to Intermediate Risk for Coronary ARTery Disease (SMART Study): 5-Year Outcome

Background: In women with low to intermediate risk of coronary artery disease (CAD), prognostic detection strategies have been controversial. We present the follow-up data of the SMART trial in peri/postmenopausal women at low to intermediate risk of CAD. Objectives: To determine the value of contrast stress echocardiography (CSE), stress electrocardiogram (sECG), and serum biomarkers for prediction of cardiovascular events (CE) in peri/postmenopausal women at low to intermediate risk of CAD. Materials and Methods: From January 2004 to August 2007, 400 peri/postmenopausal women were prospectively enrolled. All women had detailed risk factor assessment, and underwent simultaneous CSE (Definity^®, Lantheus Medical Imaging) and sECG. Laboratories included brain natriuretic peptide (BNP), atrial natriuretic peptide, endothelin, and high sensitivity C-reactive protein. Wall motion score index was based on a 16-segment model. Abnormal CSE was defined as new or worsening wall motion abnormality at stress, while abnormal sECG was ≥1 mm horizontal/downsloping ST segment depression/elevation (80 mseconds duration). Self-reported outcome data were collected from a mailed Women's Heart Clinic Questionnaire. CE outcomes included all-cause mortality, nonfatal myocardial infarction (MI), heart failure, chest pain hospitalization or development of typical angina (CP), and revascularization (REVASC). Adjusted Cox proportional hazard ratios (HR; 95% confidence intervals) were reported. Results: A total of 366 women (54.4 ± 5.5 years, Framingham risk 6.5% ± 4.4%) completed simultaneous CSE and sECG. Forty-two (11.5%) had abnormal CSE, while sECG was abnormal in 22 (6%) women. Follow-up (4.4 ± 1.2 years) was available in 315/366 (86%) women (78% exercise-CSE, 22% dobutamine-CSE). In those who completed follow-up, CSE was abnormal in 33 women (10.5%) and sECG was abnormal in 21 (6.7%). In 33 women with abnormal CSE, sECG was abnormal in 7 (21.2%) and normal in 26 (79%), p = 0.0004. CE occurred in 27 (8.6%) women: 8 all-cause mortality, 2 nonfatal MI, 13 CP, and 4 REVASC. CE occurred in 21% versus 7% of women with abnormal versus normal CSE, p = 0.014 and 38% versus 6% of women with abnormal versus normal sECG, p < 0.0001. Rest BNP was higher in women with CE versus those without (p = 0.018). Abnormal sECG and abnormal CSE were associated with CE, while only abnormal sECG was an independent predictor of CE (adjusted HR 10.3 [1.9-61.4], p = 0.007). Of the laboratory results, only BNP was associated with CE (adjusted HR 2.9 [1.1-7.3], p = 0.028). Conclusions: sECG and rest BNP were independent predictors of subsequent CE within 5 years in peri/postmenopausal women at low to intermediate risk of CAD.