ASSOCIATION OF PREGNANCY COMPLICATIONS AND CHOICE OF IMMUNOSUPPRESSANT IN LIVER TRANSPLANT PATIENTS¹

Abstract

Background. 

The purpose of this study was to identify factors associated with antenatal complications for an ongoing series of pregnant women who have undergone orthotopic liver transplantation.

Methods. 

We reviewed Magee-Womens Hospital records from 14 pregnancies in 13 women in whom a liver had been transplanted before pregnancy. We collected and analyzed data regarding the primary liver disease, allograft status, liver function at conception and during pregnancy, immunosuppressive medications, associated medical conditions, time from transplant to conception, cytomegalovirus serostatus, and maternal and fetal outcome.

Results. 

Seven patients had evidence of renal dysfunction (creatinine, 1.3-2.0 mg/dl), five of whom also were hypertensive at their first prenatal visit. The complications of preeclampsia, worsening hypertension, and small for gestational age occurred only in women with renal dysfunction at conception. Renal dysfunction was more often associated with cyclosporine than tacrolimus use.

Conclusions. 

Renal dysfunction is the primary determinant of adverse pregnancy outcome in liver transplant recipients. Immunosuppression with cyclosporine during pregnancy was more often associated with antenatal complications than with the use of tacrolimus.

Although experience with pregnancy after liver transplantation is limited, prior reports have described an increased risk of preeclampsia, premature rupture of membranes, preterm delivery, small for gestational age(SGA^*) neonates, infection, anemia, and first trimester abortion ^(1-3). To determine whether certain subgroups of women were at particular risk for pregnancy complications, we reviewed our ongoing experience at the University of Pittsburgh's Magee-Womens Hospital in managing these pregnancies.

We reviewed hospital records from all women who have delivered at Magee-Womens Hospital and who had previously undergone orthotopic liver transplantation. We collected and analyzed data regarding the primary liver disease, liver and renal function at conception, allograft status and function during pregnancy, immunosuppressant medications, associated medical conditions, transplant to conception interval, cytomegalovirus serostatus, and obstetric and fetal outcome.

Preeclampsia was defined as new onset hypertension (140/90 mmHg), an increase in systolic blood pressure of 30 mmHg, or an increase in diastolic blood pressure of 15 mmHg and 300 mg of protein in a 24-hr urine collection. Small for gestational age was defined as a birthweight less than the 10% for gestational age. We defined renal dysfunction as a serum creatinine level≥1.3 mg/dl.

Because the metabolism and clearance of both tacrolimus and cyclosporine change throughout pregnancy, blood levels are frequently monitored and dosages changed to maintain a therapeutic range. Our patients took divided daily doses of either cyclosporine or tacrolimus to achieve target plasma levels in the ranges 250-1000 ng/ml for cyclosporine and 5-20 ng/ml for tacrolimus. The doses of cyclosporine ranged from 100 to 800 mg twice daily, and the doses of tacrolimus ranged from 2 to 10 mg twice daily.

We analyzed the data with SPSS for Windows (Chicago, IL). The chi-square test with Yates' correction and the Fisher's exact test were used as appropriate. A P-value >0.05 was considered significant.

Since 1981, our group has cared for 14 pregnancies in 13 women who had undergone orthotopic liver transplantation. The indication for liver transplantation included a variety of congenital and acquired conditions(Table 1). None of the women had known renal disease at the time of transplant. At conception, all of the women were taking one to three of the following immunosuppressant medications: prednisone (13, 14), azathioprine (4, 14), cyclosporine (8, 14) and tacrolimus (5, 14). No patient received cyclosporine and tacrolimus simultaneously. Several of the women had associated medical conditions that were present at their first prenatal visit (Table 2). The mean maternal age at conception was 25.9 years (range, 19-42 years), and the majority (64%) of the patients were primiparous. There were no spontaneous abortions, fetal deaths, or congenital malformations. One woman had an elective termination of pregnancy at 18 weeks, and thus her pregnancy is excluded from the description of maternal and fetal outcomes.

Antepartum complications that developed included worsening hypertension in two (15%), preeclampsia in four (31%), SGA in one (8%), gestational diabetes in one (8%), anemia in four (31%), and preterm premature rupture of membranes(PPROM) in four (31%). Except for PPROM, anemia, and gestational diabetes, all antepartum complications occurred exclusively in women with a baseline creatinine level greater than or equal to 1.3 mg/dl. In our patients, renal dysfunction (creatinine, ≥1.3 mg/dl) was more commonly associated with use of cyclosporine ^{(6, 7)} than tacrolimus^{(1, 7)} (P=0.03).

Antepartum complications were not more common in women with other underlying conditions such as abnormal liver enzymes or preexisting anemia(P=NS). Although corticosteroid use in pregnancy has previously been associated with preterm premature rupture of membranes, presumably by altering vaginal flora or weakening of membranes, we were unable to test this association as all but one patient was taking prednisone as part of their immunosuppressant regimen.

Nine patients (69%) delivered before 37 weeks' gestation. Preterm delivery was primarily associated with preterm premature rupture of membranes or preeclampsia. Neonatal survival was 77% (10, 13). The three neonatal deaths that occurred were all associated with well documented cytomegalovirus infection ⁽⁵⁾. Two of the patients had recently undergone transplant (3 weeks and 2 months before conception), and the third was receiving high doses of immunosuppressants for chronic allograft rejection⁽⁵⁾.

We and other investigators have previously observed that pregnancy after liver transplantation is associated with an increased risk of worsening hypertension, preeclampsia, preterm premature rupture of membranes, anemia, intrauterine growth restriction, and preterm cesarean delivery^{(1-3, 6)}. However, the factor(s) responsible for the increased incidence of these adverse outcomes has not been clarified.

The increased risk of preeclampsia in transplant recipients has been repeatedly observed and confirmed in our series of patients^{(1-3, 6)}. In our series, this complication was only observed in women with renal dysfunction (creatinine,≥1.3 mg/dl), a well recognized risk factor for preeclampsia. Interestingly, all of these women were also on cyclosporine. Of the five women in our series on tacrolimus, one had renal dysfunction and none developed preeclampsia. Other groups have made similar observations. Tacrolimus seems to cause less arterial hypertension in transplant recipients overall and a lower incidence of preeclampsia in pregnant transplant recipients than cyclosporine^(7-9). The association of preeclampsia with prepregnancy renal dysfunction was also observed by Pruvot et al.⁽¹⁰⁾ in their study of seven pregnancies in five liver transplant recipients. Therefore, we speculate that renal dysfunction, presumably caused by long term exposure to cyclosporine, is the factor that predisposes transplant recipients to preeclampsia. The complications of SGA and worsening hypertension in our series also occurred only in women with renal dysfunction, a risk factor for these complications.

Alternatively, the immunosuppressants themselves may contribute to the observed increased risk of preeclampsia. Both tacrolimus and cyclosporine decrease endogenous nitric oxide production ⁽¹¹⁾. If nitric oxide is a mediator of smooth muscle relaxation, interference with its production may enhance hypersensitivity to vasoconstrictors and therefore predispose to preeclampsia.

In summary, women with liver allografts seem to have a higher risk of preeclampsia, worsening hypertension, PPROM, anemia, SGA, preterm delivery, and cesarean section than the normal obstetric population. We observed that renal dysfunction, a direct consequence of immunosuppressant medications, confers an increased risk of worsening hypertension, preeclampsia, and SGA and that women with normal renal function may not be at increased risk. Renal dysfunction was more often associated with cyclosporine than tacrolimus use in our series of patients. Although experience with tacrolimus in pregnancy is still limited ⁽⁸⁾, we observed no adverse fetal effects and believe that, if possible, this may be the immunosuppressant of choice for reproductive age women. We propose that a woman's future childbearing plans might be considered in the choice of a primary immunosuppressant agent after liver transplantation.