Characterizing vascular and hormonal changes in women across the life span: a cross-sectional analysis

Abstract

Vascular dysfunction, marked by lower endothelial function and increased aortic stiffness, is a nontraditional risk factor that precedes the development of cardiovascular disease (CVD). However, the age at which these changes in vascular function occur in women and the degree to which reproductive hormones mediate these changes has not been characterized. Women free from major disease were enrolled across the adult life span (aged 18-70 yr, n = 140). Endothelial function was assessed as flow-mediated dilation (FMD) of the brachial artery during reactive hyperemia using duplex ultrasound and expressed as percent dilation. Aortic stiffness was measured by carotid-femoral pulse wave velocity (cfPWV). Blood samples were obtained to quantify reproductive hormone concentration. Regression models determined age-related breakpoints and mediating factors between age and vascular outcomes. FMD declined with age with a breakpoint and steeper decline occurring at 47 yr of age. Thereafter, age was independently associated with lower FMD (B = -0.13, P < 0.001). cfPWV was relatively stable until a breakpoint at age 48, and age was independently associated with higher cfPWV thereafter (B = 0.10, P < 0.001). Path analysis revealed that the association between age and FMD was partially mediated by follicle-stimulating hormone (ab_ind = 0.051, P = 0.01) and progesterone (ab_ind = 0.513, P < 0.001) but not estradiol (ab_ind = -0.004, P = 0.08). No mediation was present for cfPWV. Age was associated with endothelial dysfunction and aortic stiffness in women beginning at 47 and 48 yr old, respectively, 3 to 4 yr before the average age of menopause. The association between age and endothelial dysfunction was explained in part by elevations in follicle-stimulating hormone and progesterone, but not declining estradiol.NEW & NOTEWORTHY We demonstrate that the age at which endothelial function declines and aortic stiffness increases in healthy women is 47 and 48, respectively. The inflection point in flow-mediated dilation (FMD) is 6 yr earlier than previously reported, and the association between age and FMD was mediated by follicle-stimulating hormone (FSH) and progesterone (P₄) but not estradiol (E₂).

Keywords: aging; arterial stiffness; endothelial function; estrogen; sex hormones.

Graphical abstract

Figure 1.

Segmented linear regressions showing association between age and vascular function in women aged 18–70 yr. A: flow-mediated dilation (%FMD; n = 140). B: carotid-femoral pulse wave velocity (cfPWV; n = 99). C: systolic blood pressure (SBP; n = 140). D: diastolic blood pressure (DBP; n = 140). n, number of subjects. Data were analyzed using a one-way analysis of variance and post hoc pairwise comparisons were done using Tukey’s HSD.

Figure 2.

Segmented linear regressions showing the association between age and reproductive hormone concentrations (log transformed) in 126 women aged 18–70 yr. A–E: estradiol (E₂; A) progesterone (P₄; B), follicle-stimulating hormones (FSHs; C), luteinizing hormone (LH; D), and total testosterone (E). n, number of subjects. Data were analyzed using a one-way analysis of variance and post hoc pairwise comparisons were done using Tukey’s HSD.

Figure 3.

Path analysis evaluating the degree to which reproductive hormone concentrations mediated the association between age and flow-mediated dilation (FMD). Arrows denote significant direct effects. Significant mediation effects (FSH, a₁b₁ = 0.051; P₄, a₂b₂ = 0.513) are presented in bold. FSHs, follicle-stimulating hormones; P₄, progesterone; E₂, estradiol.