PREGNANCY OUTCOMES in Maternal Transplant Recipients

Fertility and contraception

A major concern of patients is whether or not they can reproduce after receiving a transplanted organ. Several studies document the rapid return of fertility after transplantation (3). The hypothalamic-pituitary access is suppressed in patients with ESRD, but gonadal suppression appears to be reversible, with reports of pregnancy occurring within months of successful transplantation (4). It is not known whether fertility is restored to age-appropriate “normal” levels after transplantation, because only scattered reports are available (of assisted reproduction) in transplant recipients. Men with ESRD have several defects in spermatogenesis that may be reversible (5, 6), but isolated deficits in ovarian function have not been documented to our knowledge. Another concern is whether immunosuppressive medications impair fertility. At this time it does not appear that immunosuppressive medications directly impair female fertility, although sirolimus (rapamycin) (7) is clearly associated with male infertility, and men wishing to father a child should therefore not take sirolimus.

Given that fertility is rapidly restored after transplantation, the patient and her partner need to be counseled about pregnancy prevention early in the process of pretransplant workup. Optimal contraception is a decision to be made between the patient and her gynecologist, inasmuch as there are no contraindications to the use of any contraceptive method. The options to consider include sterilization of either the transplant recipient or her male partner, and whether the patient wishes to have irreversible contraception or reversible contraception. If the patient wishes to have reversible contraception, the choices include intrauterine devices (IUDs) (based on either copper or progestin), progesterone-containing systemic contraceptives such as depot medroxyprogesterone acetate, progestin implant, progestin-only pills, estrogen-containing contraceptives such as birth control pills, contraceptive patches, and a vaginal ring (8, 9). Generally it is not optimal to use barrier methods alone because of their risk for nonuse, although they may provide protection against transmissible diseases. Some methods previously considered to be ill-advised in women who have undergone transplantation are being reconsidered (8). IUDs had previously been thought to increase the risk of uterine infection and to not be as effective, because it was reasoned that their efficacy was dependent on an intact immune system. Newer types of IUDs appear to be more effective and not complicated by increased infection risk (8).

Timing of pregnancy

The first few months after transplantation are complicated by multiple medical challenges, including the frequent adjustment of immunosuppressive medications, the concomitant use of teratogenic medications (such as valganciclovir), and early rejection episodes. It has thus been recommended that women considering pregnancy wait until graft function and immunosuppression are stable; this usually occurs by a year after successful transplantation (10, 11). An earlier recommendation had been to wait 2 years, but now that the wait-list time for allografts is lengthy and women are reaching transplantation at older ages, this suggestion has been changed (12, 13). Recent reports suggest that if a woman has good stable allograft function, with no episodes of rejection for 6 months, and is not required to take fetotoxic medications, she could consider pregnancy earlier than a year after transplantation (14).

Risks of graft loss or rejection

Whether or not pregnancy will increase the risk of graft loss is a concern that must be discussed with potential transplant recipients. Registries in the United States, the United Kingdom, and Europe, as well as case reports from around the world, have confirmed that the risk of graft loss is probably low if the patient has good graft function at the onset of pregnancy (14). Creatinine level is a poor indicator of graft function, but unfortunately we are not aware of any studies of GFR changes after pregnancy in transplant recipients. Therefore, the definition of good graft function includes a stable creatinine level (≤1.5 mg/dL), the absence of significant proteinuria (≤500 mg/24 hours), and no graft rejection within 6 months (10). Given these parameters, the risk of graft loss associated with pregnancy is not different from that in the nonpregnant transplant recipient (15, 16). Likewise, the risk of rejection is probably low as well, as long as there has not been evidence of poorly suppressed immunoreactivity (e.g., recent graft rejection) (14).

Immunosuppressive medications and pregnancy

Adequate immunosuppression must be maintained during pregnancy because drug levels vary widely throughout gestation. The mother is not immunosuppressed by her pregnancy, contrary to some folk beliefs, and therefore requires maintenance of adequate immunosuppression (17). Blood levels of immunosuppressive drugs should be monitored frequently during the pregnancy (14). At our transplant centers we see the patient bimonthly, checking calcineurin inhibitor and creatinine blood levels at each visit, and we continue this frequency for at least 2 months after delivery. The frequency of monitoring requires that the patient be willing to comply with close follow-up, and this requirement should be discussed with the patient before she becomes pregnant.

Maternal risks associated with pregnancy

Other risks to consider for the maternal transplant recipient are worsening hypertension and pre-eclampsia. Hypertension is common in transplant recipients and often worsens during the pregnancy (18, 19). Generally, the recommendations are to keep the pregnant transplant recipient normotensive if possible, which differs from the advice given to pregnant patients with chronic kidney disease (20). Angiotensin-converting enzyme inhibitors and angiotensin receptor blockers are contraindicated because of their fetotoxic potential (21), and therefore maternal transplant recipients are generally prescribed methyldopa or labetalol for the treatment of hypertension (20). Other acceptable agents include nifedipine, hydralazine, and thiazide diuretics (20).

Pre-eclampsia is also commonly diagnosed in pregnant transplant recipients. In the registries, pre-eclampsia was diagnosed in over 30 percent of pregnancies, in contrast to a 5 percent occurrence of that diagnosis in the general population (14). Pre-eclampsia occurs more frequently in patients with chronic kidney disease than in the general population as well (20). It is not known whether there is an independent effect of immunosuppressive medications on the placenta that contributes to the high risk of pre-eclampsia, but data now show that pre-eclampsia is also common in recipients of heart, lung, and liver transplants, who presumably do not have significantly impaired renal function (22).

The accurate diagnosis of pre-eclampsia is difficult in transplant recipients because surrogate markers occur frequently in patients with impaired renal function. Transplant patients often are hypertensive, have proteinuria, and have increased uric acid levels (because of calcineurin inhibitors). Pregnancy may be associated with worsening proteinuria, hyperuricemia, and hypertension, thereby mimicking pre-eclampsia. Furthermore, small changes in serum creatinine levels may hide more serious changes in GFR because of the natural hyperfiltration of pregnancy. Newer markers of pre-eclampsia, such as s-flt and soluble endoglin, have not been validated in the renal transplant population (20).

There are many other medical considerations in the maternal transplant recipient, including gestational diabetes, anemia, and infections such as urinary tract infections (23–27). It is recommended that maternal kidney transplant recipients be screened every trimester for gestational diabetes (28). Other comorbidities, such as urinary tract infections, are quite common in renal transplant patients, and therefore frequent screening is mandatory (25, 28). Several other infections need to be considered in the maternal transplant recipient; the reader is referred to an earlier review for details (14).

Many infants are delivered by caesarean section (29). However, the presence of the transplanted kidney in the false pelvis does not interfere with vaginal delivery (28, 30). Thus, unless there is an obstetric reason to indicate caesarean delivery, vaginal delivery is preferred (10).

Fetal risks of pregnancy in transplant recipients

There are potential risks to the developing fetus that should be discussed with the maternal transplant recipient and her partner. We believe that a frank discussion of these concerns should be conducted long before pregnancy occurs so that the future parents are prepared for the possibility of adverse outcomes, including premature delivery, intrauterine growth retardation (IUGR), and long-term developmental problems.

Data from all three registries have demonstrated an extremely high risk for premature delivery (2). Premature delivery is defined as any delivery occurring earlier than 37 weeks. Premature delivery has been documented in recipients of all solid organs but occurs in about 50 percent of renal transplant pregnancies (2). Among the consequences of premature delivery are increased risk of learning disabilities and neurocognitive deficits (31). There is also a very high risk for IUGR, suggesting a primary pathologic process involving the placenta. IUGR occurs in approximately 20 percent of deliveries and is associated with comorbidities including hypertension, diabetes mellitus, neurologic abnormalities, and developmental delay (14).

Interestingly, gross congenital abnormalities are not common in infants exposed in utero to immunosuppressive medications, with the exception of mycophenolate mofetil (2, 32–34). Recent data have shown a pattern of congenital abnormalities in infants exposed in utero to mycophenolate mofetil, and the Food and Drug Administration has changed its labeling to category D. It is therefore recommended that women considering pregnancy cease taking any mycophenolate drug (CellCept or Myfortic) at least 6 weeks before attempting pregnancy (14). Whether to add azathioprine to the patient’s drug regimen is something to consider. We do this at our transplant centers to be sure that adequate immunosuppression is maintained. At this time there are insufficient data about the safety of sirolimus or everolimus, and therefore we have also recommended a change in these medications 6 weeks before pregnancy is attempted.

Although obvious congenital malformations are rare, whether less obvious abnormalities are induced by in utero exposure to immunosuppressive medications is not known. All immunosuppressive medications cross the maternal–fetal barrier, although there are important differences in the delivery of active metabolites to the developing fetus (14). For instance, prednisone easily crosses the placental circulation, the placenta metabolizes prednisone, and therefore the fetal dosing is diminished. Likewise, azathioprine crosses the maternal–fetal barrier, but active metabolites are not present in the fetus because of the lack of a fetal enzyme to metabolize 6-mercaptopurine.

Calcineurin inhibitors easily pass through the maternal–fetal interface, and active metabolites have been reported in the fetal circulation (14). In fact, serum levels of cyclosporine have been reported in newborns at levels bioequivalent to that of the mother. Therefore, it appears that the developing fetus is likely exposed to calcineurin inhibitors throughout gestation. There is substantial evidence from animal models that in utero exposure to cyclosporine and tacrolimus induces autoimmunity by interfering with the negative selection of autoreactive T cells in the developing thymus. Whether the same phenomena occur in human infants is not known.

There is limited information on the neurocognitive or immunologic development of the human fetus exposed to immunosuppressive medications, and well-designed studies are needed. The National Transplantation Pregnancy Registry has tried to follow up children after delivery to determine whether more subtle defects are associated with fetal exposure to immunosuppressive medications. In the data from the National Transplantation Pregnancy Registry there was noted to be a 27 percent incidence of learning disabilities in school-age children exposed to immunosuppressive medications. Recently another report has suggested that this was associated with premature birth (35).

Breastfeeding

Many patients inquire about the possibility of breastfeeding their infants. Unfortunately, there are few data from which to derive recommendations for or against breastfeeding. The immunosuppressive levels in breast milk vary widely, and the pharmacokinetics and pharmacodynamics of immunosuppressant secretion in breast milk have not been defined (36–38). Large controlled studies that evaluate breast milk concentrations of immunosuppressant medications have not been performed, to our knowledge. The mother should be informed that it is unknown whether the risks of further exposure of her infant to immunosuppression outweigh the benefits of breastfeeding.

Conclusion

The first woman to become pregnant after a kidney transplant died this year at the age of 76. Fifty-three years since the report of her first pregnancy and many thousands of pregnancies later, it is clear that pregnancies in transplant patients can occur successfully if kidney function is good and proteinuria is minimal, without a negative impact on the allograft. These pregnancies are high risk. Pregnancy in renal transplant recipients must be approached with counseling both before and after transplantation, and with close follow-up, for the prevention and management of medical and obstetric complications.