Sex, pregnancy and aortic disease in Marfan syndrome

BACKGROUND: Sex-related differences as well as the adverse effect of pregnancy on aortic disease outcome are well-established phenomena in humans with Marfan syndrome (MFS). The underlying mechanisms of these observations are largely unknown. OBJECTIVES: In an initial (pilot) step we aimed to confirm the differences between male and female MFS patients as well as between females with and without previous pregnancy. We then sought to evaluate whether these findings are recapitulated in a pre-clinical model and performed in-depth cardiovascular phenotyping of mutant male and both nulliparous and multiparous female Marfan mice. The effect of 17β-estradiol on fibrillin-1 protein synthesis was compared in vitro using human aortic smooth muscle cells and fibroblasts. RESULTS: Our small retrospective study of aortic dimensions in a cohort of 10 men and 20 women with MFS (10 pregnant and 10 non-pregnant) confirmed that aortic root growth was significantly increased in the pregnant group compared to the non-pregnant group (0.64mm/year vs. 0.12mm/year, p = 0.018). Male MFS patients had significantly larger aortic root diameters compared to the non-pregnant and pregnant females at baseline and follow-up (p = 0.002 and p = 0.007, respectively), but no significant increase in aortic root growth was observed compared to the females after follow-up (p = 0.559 and p = 0.352). In the GT-8/+ MFS mouse model, multiparous female Marfan mice showed increased aortic diameters when compared to nulliparous females. Aortic dilatation in multiparous females was comparable to Marfan male mice. Moreover, increased aortic diameters were associated with more severe fragmentation of the elastic lamellae. In addition, 17β-estradiol was found to promote fibrillin-1 production by human aortic smooth muscle cells. CONCLUSIONS: Pregnancy-related changes influence aortic disease severity in otherwise protected female MFS mice and patients. There may be a role for estrogen in the female sex protective effect.