Top

Introduction and background:

The risk of pregnancy in patients with mechanical valve prosthesis (VP) has traditionally been attributed to the use of oral anticoagulants necessary to maintain the integrity of the mechanical prosthesis. The impact of residual valvular lesions has not been a point of interest in these patients, as it is assumed that these were corrected at surgery. Whereas Mitral and Aortic valve replacements have proven to be successful and have stood the test of time, Tricuspid valve replacements have become unpopular due to the high incidence of thrombosis and malfunction (1-7) . Tricuspid valve repair can be done in several ways, by either a ring or an annuloplasty (5-7). However in rheumatic affection of the Tricuspid valve, other problems arise: the valve may not be solely incompetent, it may be stenotic as well; the incompetence may not be solely due to annular dilatation, actual fibrosis and shortening of the cusps may occur rendering repair more difficult. Finally, there have been reports about late occurrence of Tricuspid incompetence after Mitral valve replacement, that had not been present at the time of surgery (8) This is most probably due to progression of fibrosis in the cusps affected by rheumatic fever. The patients suffering from Tricuspid valve disease (TVD) after valve surgery are in a unique situation where the hemodynamic burden is purely due to the Tricuspid valve. This problem is of importance in Egypt, where rheumatic valvular heart disease is still common , affecting mainly the young population. As a consequence, a large portion of Egyptian females with VP are in the childbearing age. Those who do not suffer from TVD are prone to develop it with time. In this study we tried to determine the effect of TVD on the outcome of pregnancy in Egyptian females with VP in terms of maternal as well as fetal morbidity and mortality.

Patients and Methods: 

Three hundred and forty eight (348) female patients were followed up in the Cairo University prosthetic valve clinic from 1986 to 1999. Of these, 70 patients became pregnant with a total of 108 pregnancies. Five patients were lost to follow up and three pregnancies are still ongoing. These were excluded from the analysis, leaving 67 patients with 100 pregnancies. All patients had mechanical prosthesis, 43 in the Mitral position, 12 in the Aortic position, and 12 had a double valve replacement. Fourteen patients with 23 pregnancies suffered from TVD, predominantly incompetence, which was clinically apparent and severe enough to necessitate diuretic use in form of Alductone 100-150 mg and/or Frusemide 40mg daily. The pregnancies were divided into two groups : Group A , in which the mother suffered from Tricuspid incompetence, and Group B, in which the mother did not have any clinically manifest TVD

Results:

Fetal outcome:
Of 100 pregnancies, 56 (56%) resulted in a healthy baby. Fetal loss occurred in 44%, and was due to abortion in the 1^st trimester ( 28%), intrauterine fetal death (4%), stillbirth (3 %), neonatal death due to meconium aspiration (1%), prematurity (2%), RH incompatibility (2%) and preterm death of the mother (4%). In 23 of the 100 pregnancies, the mother suffered from Tricuspid incompetence severe enough to necessitate diuretic use. These pregnancies did not fare well, 17 of them ending in an abortion or intrauterine fetal death. When comparing the results to those of patients not suffering from Tricuspid incompetence, there was a statistically significant difference.(p = 0.001) see table 1
Maternal Outcome
Of the 100 pregnancies followed up, 15 developed complications and 5 ended fatally. The maternal complications included valve malfunction (thrombosis in 8 patients and infective endocarditis in 1 patient), postpartum hemorrhage (5 patients) , bleeding during pregnancy due to placenta brevia (1 patient), toxemia of pregnancy (1 patient) and Cesarean section was needed in 2 patients.
We postulated that Tricuspid incompetence may aggravate the tendency to post partum hemorrhage, and increase the likelihood of infective endocarditis, however the presence of TVD did not have any impact on the maternal outcome of pregnancy. (see table 1) The most common and most serious complication was acute valvular malfunction. In all cases of acute valvular obstruction, the patient had received heparin in the 1^st trimester with the exception of a woman who developed infective endocarditis. Five of these patients died and 4 needed reoperation, which was performed successfully. This led us to examine the effect of type anticoagulation on the outcome of pregnancy (see table 2).

Table 1: Effect of TVD on the outcome of pregnancy

Anticoagulation during the first trimester
Of the 100 pregnancies in patients with VP, 66 were put on a strict regimen of subcutaneous heparin, 5000 IU every eight hours or 10 000 IU every 12 hours. As most of these patients were from remote rural areas, a proper follow up of aPTT was not possible. In the other 34 pregnancies, the mother continued oral anticoagulant (warfarin) intake, so as to adjust the prothrombin time to 2.5 of the control and later, when INR became available, to adjust the INR to 2 - 2.5 . Of the 66 pregnancies initially put on heparin, 38 (57.5%) resulted in a healthy baby versus 18 out of 34 (52.9%) in the group on Warfarin. The difference was statistically insignificant (see table 2). As mentioned before, maternal outcome was affected by the type of anticoagulation used.