Association of serum anti-mullerian hormone levels with vascular function may differ by historyof conception after in vitro fertilization: pilot study

Low ovarian reserve, as measured by serum anti-Mullerian hormone (AMH), and the use of in vitro fertilization (IVF) are both associated with increased risk for cardiovascular disease (CVD) in women; however, the mechanisms underlying this association remain unclear. Low AMH is associated with increased carotid intima-media thickening (cIMT) among women with chronic kidney disease and HIV. Carotid artery stiffness and carotid intima-media thickness (cIMT) are subclinical CVD risk factors that increase prior to the development of overt CVD and therefore can be used to interrogate the early pathogenesis of CVD. Accordingly, we evaluated the association between AMH and biomarkers of atherosclerosis (e.g., carotid stiffness and cIMT) in a pilot sample of healthy, recreationally active premenopausal women. We hypothesized that lower AMH would be associated with subclinical biomarkers of atherosclerosis, particularly among women with prior IVF. Premenopausal women (30-45 years) were recruited based on a prior unassisted (n=18) or IVF pregnancy (n=13) with a live birth within the past 1-5 years and excluded for use of vascular- or lipid-altering medications. All underwent testing following an overnight fast. Common carotid artery ß-stiffness and cIMT were measured in the supine position after 10 minutes of quiet rest using carotid ultrasonography (Vivid I, GE Healthcare). AMH was measured in duplicate from serum using an enzyme-linked immunosorbent assay (Ansh Labs). Associations between carotid stiffness and cIMT were modeled by AMH using linear regression; multivariable models adjusted for covariates (age, body mass index [BMI], IVF status) and interactions between AMH and IVF. Age (mean±SD for IVF and unassisted pregnancies: 38±4 vs. 37±3 years, p=0.34), BMI (27.4±5.7 vs. 25.7±5.6 kg/m2, p=0.41), seated systolic (107±11 vs. 109±8 mmHg, p=0.48) and diastolic blood pressure (68±10 vs. 71±8 mmHg, p=0.37), and AMH (2.5±2.0 vs 3.5±2.7, p=0.29) did not differ between groups based on mode of conception. In unadjusted models, carotid stiffness was not associated with AMH (b=0.06 [-0.25, 0.36], p=0.70) or IVF (b=0.27 [-1.24, 1.79], p=0.71). However, the association between carotid stiffness and AMH differed marginally by conception type (p=0.07), where the direction of the association for the IVF group (b=-0.38 [-0.92, 0.17], p=.19) and unassisted group (b=0.24 [-0.11, 0.6], p=0.17) were opposite one another. The interaction between AMH and IVF on carotid stiffness no longer differed after adjustment for age and BMI (p=0.24). cIMT was not associated with AMH (p=0.44) or IVF (p=0.80) in unadjusted models; this remained unchanged after accounting for mode of conception (p=0.89). AMH was not associated with carotid stiffness or cIMT in healthy premenopausal women in this pilot study. There was a trend for the association between AMH and carotid stiffness to differ by conception type, but larger studies are needed.