Purpose: Low-dose warfarin throughout pregnancy (≤5 mg/day) was suggested to be safe in patients with mechanical heart valve prosthesis (MHVP). Being only shown in selected small low-risk groups, we aimed to evaluate this regimen in high-risk cases with mitral ± aortic MHVP.
Methods: Group A: 100 pregnant patients with mitral MHVP (+aortic; 24%) were randomized in a 3:1 ratio, to either receive low-dose warfarin (≤5 mg/day) or phenindione (≤100 mg/day) throughout pregnancy, respectively. Group B: 100 females in the child-bearing period with mitral MHVP (+aortic; 19%) were randomized similarly and followed the same dosage limits. Both groups targeted an INR of 2.5–3.5. Patients failing to achieve target were initially shifted to the other oral anticoagulant (OA), then finally excluded on heparin. Aspirin (100 mg/day) was supplemented in the 2nd trimester/4th follow-up month (Group B).
Results: Group A: 20 patients (26.7%) were shifted to phenindione and none to warfarin (P=0.001). In comparison, only 4 patients in Group B were shifted to phenindione (5.3%; P<0.001) and none to warfarin. Despite overall lower INR achieved (2.26 ± 0.3 vs. 2.61 ± 0.31; P<0.001), pregnant patients received higher warfarin (4.16 ± 0.88 vs. 3.19 ± 1.1 mg; P<0.001) or phenindione doses (82.28 ± 23.95 vs. 64.28 ± 24.77 mg; P=0.007); compared to Group B. We had no patient mortality. Maternal complications were 6 minor bleedings and 2 threatened abortions. We had no fetal embryopathy. Th e recorded 9 spontaneous abortions and 2 stillbirths were unrelated to type of OA but to the higher dose (P>0.05) and lower INR achieved (2.08 ± 0.22 vs. 2.28 ± 0.31; P = 0.039).
Conclusions: Despite the need for high OA dosage, pregnancy was associated with achieved low INR levels, making low-dose warfarin insufficient to maintain target INR in high-risk patients. Phenindione may offer an effective and probably safe alternative.