Objectives Firstly, to compare pregnancy outcomes and cardiac complications in women with: 1. either mechanical or bioprosthetic valves at the mitral site; 2. mechanical valves treated with warfarin or subcutaneous heparin. Secondly, to determine pregnancy and cardiac outcomes in women with aortic homograft valves.
Design Historical cohort study.
Setting Greenlane Hospital, Auckland, New Zealand.
Population Young women (n= 255) who had valve replacements between 1972 and 1992. Seventy-nine women underwent 147 pregnancies.
Main outcome measures Pregnancy loss, cardiac complications.
Results Pregnancy loss occurred in 59% of pregnancies with mitral mechanical valves (n= 50) and 7% with mitral bioprosthetic valves (n= 33) (RR 8.20,95% CI2.10–31.93). Pregnancy loss rate was 70% in pregnancies treated with warfarin, compared with 25% for those switched from warfarin to heparin (RR 2.81, 95% CI 1.03–7.73). All heparin-associated losses occurred in the first trimester, whereas there were four stillbirths with warfarin. Cardiac complications occurred in 10 pregnancies (20%) in the women with mitral mechanical valves and four (13%) with mitral bioprosthetic valves (RR 1.55, 95% CI 0.53–4.52). All four thromboembolic complications with mechanical valves occurred in the 14 women treated with heparin throughout pregnancy. Structural valve deterioration occurred in four pregnancies (10%) with mitral bioprosthetic valves. No cardiac complications or known pregnancy losses occurred with aortic homograft valves (n= 41).
Conclusion The high pregnancy loss rate in women with mitral mechanical valves was associated with warfarin throughout pregnancy, whereas the thromboembolic cardiac complications were associated with heparin. Pregnancy outcome was very good in women with bioprosthetic and homograft valves.