Author + information
- Di Zhao, PhD,
- Eliseo Guallar, MD, DrPH and
- Erin D. Michos, MD, MHS∗ ()
- ↵∗Ciccarone Center for the Prevention of Heart Disease, Division of Cardiology, Department of Medicine, Johns Hopkins University School of Medicine, Blalock 524-B, 600 N. Wolfe Street, Baltimore, Maryland 21287
We thank Dr. Georgiopoulos and colleagues for their comments regarding our recent paper (1). As suggested, we examined the association between free androgen index (FAI) levels, calculated as the ratio of total testosterone (T) to sex hormone–binding globulin (SHBG) levels, and incident cardiovascular disease (CVD) among postmenopausal women in the MESA (Multi-Ethnic Study of Atherosclerosis) cohort. We found that a higher FAI was associated with CVD and coronary heart disease (CHD) after adjustment for demographic factors (Model 1), but the associations were attenuated and no longer statistically significant in models further adjusted for lifestyle and other CVD risk factors. The fully adjusted hazard ratios (95% confidence intervals) for CVD, CHD, and heart failure associated with 1 SD higher log-transformed FAI were 1.10 (0.96 to 1.26), 1.09 (0.92 to 1.31), and 1.05 (0.83 to 1.32), respectively (Table 1, Model 3).
In our original paper (1), we evaluated the association of free T and bioavailable T with CVD events. Free T was calculated as the fraction of total T unbounded to SHBG or albumin, and bioavailable T as free T plus T weakly bound to albumin (2). Similar to FAI, free T and bioavailable T are derived from total T and SHBG, and are markers of biologically active T. As expected, the associations between FAI and CVD risk were consistent with those of free T and bioavailable T. Similarly to FAI, we found that bioavailable T was statistically significantly associated with CVD and CHD only in Model 1 (1).
Although free T and bioavailable T are more reliable markers of androgen status than FAI, some studies may not collect the data required for their estimation (3). Data on free T and bioavailable T are also often unavailable in clinical settings. In these circumstances, the FAI can be used as a marker of bioactive T, with potentially broad applications. With respect to CVD risk, however, our MESA findings show that total T may be a more useful biomarker than markers of bioavailable T. Furthermore, the total T/estradiol ratio may be a simple index that summarizes the contribution of T and estradiol in a single variable for risk discrimination in postmenopausal women.
Please note: The authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- 2018 American College of Cardiology Foundation
- Zhao D.,
- Guallar E.,
- Ouyang P.,
- et al.
- Morris P.D.,
- Malkin C.J.,
- Channer K.S.,
- Jones T.H.