Post-Transplant Pregnancy and Contraception

Abstract

History

The first pregnancy in a transplant recipient occurred >60 years ago. Edith Helm, a 21-year-old woman with kidney failure, received a living-donor kidney transplant from her identical twin on May 24, 1956. She was the third person to receive a successful kidney transplant, the first woman to receive a successful kidney transplant, and the first to become pregnant. Her son, born in March 1958, was full term, 3300 g, and delivered by caesarian section. Mrs. Helm went on to have a daughter 2 years later, also delivered by caesarian section (1). When she died at the age of 76 in 2011, she was the world’s longest-surviving transplant recipient. Her transplanted kidney was functioning at the time of her death, illustrating that pregnancy need not interfere with graft longevity (2). Mrs. Helm’s pregnancies taught us that fertility could be restored in women who had kidney failure, that a growing fetus and enlarging uterus would not injure the transplant, and that women with kidney transplants could have multiple pregnancies. However, this initial experience is not generalizable. Mrs. Helm’s fetuses avoided immunosuppression exposure, and her transplanted kidney was protected from immunologic attack and immunosuppression nephrotoxicity. Pregnancy in the setting of immunosuppression is a very different situation.

Pregnancy outcomes in individuals who have not been transplanted focus on the offspring and the mothers. In women who have received a transplant, the considerations include what happens to the allografts. We have learned about these results from the Transplant Pregnancy Registry International (TPR), a voluntary registry founded by Dr. Vincent Armenti in 1991 as the National Transplant Pregnancy Registry (3).

Reproductive Counseling

Reproductive counseling ideally begins before transplant and is repeated in the immediate post-transplant period and at follow-up visits during childbearing years. Counseling should include assessment of a woman’s desire to achieve or avoid pregnancy, discussion of fertility, education regarding pregnancy risks and ideal timing of pregnancy, and explanation of contraceptive options. Women with kidney failure who desire motherhood have options, including pregnancy during dialysis or post-transplant, surrogate pregnancy, and adoption. There are considerations affecting risks for mother, baby, and graft that are specific for each woman. Outcomes are affected by maternal age, genetic causes of kidney failure, sensitization, maternal comorbidities and life expectancy, infection history, medications, social support, and medical resources available. For women desiring pregnancy, the general recommendations are to be at least 1 year post-transplant, have stable allograft function with no proteinuria, no recent episodes of rejection or infection, and well-controlled medical conditions (hypertension, diabetes) to achieve optimal pregnancy outcomes (4,5).

Fertility

Dysregulation of the hypothalamic-pituitary-gonadal axis causes low levels of sex hormones, oligo- or amenorrhea, reduced vaginal lubrication, difficulty achieving orgasm, loss of libido, and infertility in women with kidney failure. Of all women on dialysis, >90% have irregular or absent menstrual cycles. Restoration of fertility and improved sexual function can occur within the first year post-transplant, with return of ovulation and menses as early as the first month after transplantation (6,7).

Pregnancy rates are low on dialysis, potentially distorting women’s perception of fertility post-transplant. In a US study of recipients of solid organ transplants, nearly half of women aged 19–49 years were unaware that pregnancy was possible after transplant (8). In a Norwegian study of 118 recipients of kidney transplants of reproductive age, 37% reported not receiving contraceptive counseling early post-transplant, 84% were sexually active, and 78% had no intention of pregnancy (9). Similarly, in a Polish study including recipients of kidney transplants, only 34% reported receiving effective post-transplantation contraception counseling. Nearly half of these patients used condoms (a method with a high failure rate) for birth control. Contraceptive counseling was associated with significantly greater use of effective contraception (10). Unplanned pregnancy rates in recipients of kidney transplants ranging from 49% to 93% have been reported (11).

Attitudes regarding reproductive counseling were recently studied in candidates for liver transplant (n=14), recipients of liver transplants (n=60), and their health care providers (n=43) (12). Family planning was a high priority in 86% of the study group, with preference for in-person discussions with the providers. Of the health care providers, 96% voiced an interest in additional reproductive education; misconceptions about the safety of estrogen and intrauterine devices (IUD) were observed in 53% and 42% of providers, respectively. In the immediate post-transplant period when fertility may become restored quickly, females of reproductive age may not realize the importance of accessing reproductive health services. Transplant team members must become adept at reproductive and contraceptive counseling, or ensure early referral to an obstetrician or primary care provider comfortable with reproductive counseling and contraception in medically complex women. In addition to providing accurate information regarding medical risks and pregnancy outcomes, incorporating the wishes, values, and patient-level acceptance of pregnancy risks are vital in shared decision making.

Infertility and In Vitro Fertilization

Infertility rates after kidney transplant are poorly described, and reports of pregnancy achieved by in vitro fertilization (IVF) in recipients of transplant are limited. In a single-center report, eight of 13 recipients of transplants who had infertility for an average of 2 years achieved 11 pregnancies, with a live birth rate per procedure of 25% (13). In a population-based retrospective study crosslinking the IVF registry with the Medical Birth Registry in Sweden, pregnancy outcomes were compared in recipients of kidney transplants undergoing IVF with those experiencing spontaneous conception. Although small numbers of pregnancies occurred with IVF, preterm births, low birth weight, and preeclampsia were comparable (14).

Immunosuppression Considerations

Recipients of transplants require education on the potential teratogenic effects of immunosuppression. Women desiring pregnancy should review medications with their provider several months before attempting conception. Due to exclusion of pregnant females from immunosuppression trials, safety data are limited to animal studies and epidemiologic data from transplant pregnancy registries and case reports. Prednisone, azathioprine, and calcineurin inhibitors are generally considered safe, whereas mycophenolate mofetil and mycophenolic acid (mycophenolate products) are contraindicated. There are limited data on safety and outcomes with the mammalian target of rapamycin inhibitors, belatacept, basiliximab, anti-thymocyte globulin, and rituximab (4–7).

The use of mycophenolate products in female recipients of transplants is associated with spontaneous abortion and fetal malformations. Spontaneous abortions occur in nearly half of pregnancies conceived on these agents, and up to 26% of infants exposed in utero have birth defects, including microtia, cleft palate, and esophageal, cardiac, and kidney abnormalities (7,15,16). Women should be counseled to discontinue mycophenolate products at least 6 weeks preconception; when this is done, the observed birth defect rates are similar to the general population. Due to the incidence and severity of birth defects, all women of childbearing potential must be advised to use either an IUD or two alternative contraceptive agents while taking mycophenolate products. The US Food and Drug Administration–mandated Mycophenolate Risk Evaluation and Mitigation Strategy program collects outcome data on pregnancies conceived on or within 6 weeks of stopping mycophenolate products. Azathioprine is commonly substituted for mycophenolate products in women attempting conception. Although teratogenic in animal models, the human fetal liver does not convert azathioprine from its inactive to active form. Substantial observational data support azathioprine safety in pregnancy (7).

Corticosteroids freely cross the placenta, where 90% is metabolized to inactive forms, resulting in low fetal exposure. Fetal exposure to calcineurin inhibitors is greater, at approximately 70% of maternal tacrolimus and between 37% and 64% of maternal cyclosporine concentration (7,17). Despite this, initial concerns of higher risks of birth defects have not been supported by larger studies. Additionally, obstetric complications, including prematurity and low birth weight, in patients on calcineurin inhibitors are similar to other immunosuppressive regimens, and may relate to underlying comorbidities (7).

Recipients of kidney transplants warrant intensified clinical and laboratory monitoring to include assessment of BP, kidney function, proteinuria, blood glucose, urine culture, and calcineurin inhibitor trough levels. Follow-up is recommended every 2–4 weeks during the first and second trimesters, and every 1–2 weeks thereafter (5). Increased activity of the drug-metabolizing enzyme cytochrome P4503A, increased maternal blood volume, and decreased albumin and hemoglobin concentrations during pregnancy can result in lower whole-blood calcineurin-inhibitor concentrations, with relatively unchanged unbound (active) drug concentrations. Clinically, tacrolimus concentrations are measured in whole blood; dose increases required to maintain therapeutic levels may result in elevated unbound concentrations and possible toxicity in women with significant hypoalbuminemia or anemia. Trough concentrations should be followed postpartum, particularly for those in whom tacrolimus dose increases were made during pregnancy (17).

Other potential fetotoxic drugs, including valganciclovir hydrochloride and angiotensin-converting enzyme inhibitors, should be discontinued before or at time of pregnancy confirmation.

Pregnancy Risks to the Allograft

Although pregnancy was historically considered an immunosuppressed condition, our understanding has evolved because studies depict it as a modified and active state in which rejections can occur (18–20). Monitoring of allograft function is complicated by the hyperfiltration of pregnancy, which decreases creatinine and can mask a decline in GFR (21). Serum creatinine should decrease by 4–6 weeks gestation, remain stable during the second trimester, and increase to near prepregnancy values in the third trimester. Failure to decrease in the first trimester, or an increase in serum creatinine above prepregnancy baseline, is concerning and should prompt investigation, including ultrasound, measurement of proteinuria, donor-specific antibodies, and possible allograft biopsy. With the recent introduction of new immune-monitoring tools, the difficulty of assessing graft injury may lessen in the future. Data from the TPR registry for conception years 1967–2016 indicate that biopsy-proven acute rejection rates in pregnant recipients with transplant rejection are approximately 0.9%–1% postpartum. A recent meta-analysis demonstrated that acute rejection rates were similar in pregnant and nonpregnant recipients of transplants (18).

According to TPR data, 6% of kidney grafts are lost within 2 years of pregnancy. A 2009 Australian study compared 120 transplanted women after first live birth with control patients who were nonpregnant, nulliparous, and had received a transplant. History of delivering a live birth was not associated with a higher 20-year risk for graft loss (22). Mohammadi et al. (23) examined 56 pregnancies in 35 women who had received a transplant. They stratified the patients by kidney function: those with serum creatinine >140 μmol/L had a higher likelihood of worsening kidney function and allograft loss. A Norwegian registry study compared outcomes in recipients of transplants with and without pregnancies. The pregnancy group had better graft survival. However, this group was younger, had shorter dialysis time, more living donors, less antihypertensives, better HLA matches, and lower ischemic times (24). These studies demonstrate that graft outcomes depend on graft function before pregnancy.

Risks to the Baby

In contrast to Edith Helm’s full-term babies, most babies born to mothers with kidney transplants are born early (average, 35.9±3.4 months): 51% are born earlier than 37 weeks, compared with 10% in the general population, and 21% are born before 34 weeks. Consequently, average birth weights are lower in this group than in babies born to mothers who have not received a transplant: 10% are reported to have a very low birth weight (<1500 g), compared with 1% in the general population (25).

Small studies on the children of recipients of kidney transplants have not demonstrated a higher incidence of intellectual impairment or abnormalities in neurologic development, beyond what would be expected for their gestational age (26–28). Stillbirths and early perinatal deaths (<24 hours after birth) occur more frequently than in the nontransplant population (29). Cytomegalovirus, a common post-transplant viral infection, can cause congenital malformations or congenital liver disease in 10%–15% of infected pregnancies, with risk being highest during early pregnancy. The incidence of congenital cytomegalovirus in infants born to recipients of transplants is not well described. Planning conception at least 1 year post-transplant decreases this risk (6).

Risks to the Mother

Recipients of kidney transplants have a markedly higher rate of preeclampsia than the general population, with 30% in TPR reports compared with 2%–5% in the general population. Preeclampsia in recipients of kidney transplants has been associated with chronic hypertension and prior history of preeclampsia (29,30), and has serious consequences with higher postpartum serum creatinine levels and higher risks of preterm delivery, caesarean section, and small-for-gestational-age babies (30).

For recipients of transplants who are pregnant, cesarean section is the most common form of delivery; indications should only be obstetric. In the TPR registry, 53% of recipients were identified as delivering by caesarean section compared with 32% in the general population (25).

Breastfeeding

Breastfeeding has many benefits (31). For mothers on immunosuppression, it also has the potential unwanted side effect of exposing the infant to immunosuppression (7). Corticosteroid use has been deemed safe with infant exposure of 0.35%–0.58% of maternal prednisone dose. For cyclosporine, it has been estimated that the infant would receive at most 2% of the mother’s weight-adjusted dose. Cyclosporine is generally not detectable in the infant’s blood, although rarely has it been (32). For tacrolimus, estimates of the ingested dose are 0.06%–0.5% of maternal weight-adjusted dose (33). Blood levels of tacrolimus in bottle- and breastfed infants are comparable (34). Data are lacking for breast milk concentration and infant exposure to mycophenolate products and mammalian target of rapamycin inhibitors. Without dismissing the potential significance of immunosuppression exposure, decisions regarding whether to breastfeed should consider that the infant’s exposure to immunosuppression with breastfeeding is lower than in utero exposure.

A 2002 survey of transplant providers uncovered that 67% of providers advised female recipients of transplants to avoid nursing (35). In contrast, the 2003 American Society of Transplantation (AST) consensus conference issued the opinion that “breast feeding need not be viewed as absolutely contraindicated” (4). This professional proclamation lagged behind the one made in the court of public opinion, or perhaps reflected it, because women with kidney transplants were more frequently breastfeeding, with rates increasing since 1995 (36).

Special Considerations

Chronic hypertension, presence of anti-phospholipid antibodies, gestational hypertension, and obesity are risk factors for preeclampsia in the general public (37). Although not specifically studied in recipients of transplants, these associations likely affect recipients of kidney transplants and contribute to the elevated incidence of preeclampsia.

Limited information has been published on the safety and outcomes of pregnancy in recipients of transplants who are sensitized. The higher number of individuals who are sensitized and are successfully transplanted makes knowing these outcomes important. One may speculate that switching from mycophenolate to azathioprine in a patient who is sensitized may disproportionately pose a rejection risk. Ajaimy et al. (38) described pregnancies in 11 recipients of kidney transplants, eight of whom were considered sensitized. The sensitized group had worse pregnancy outcomes, including one stillbirth and two second trimester miscarriages. Three women who were sensitized developed preeclampsia. Babies born to women who were sensitized were more likely to be born preterm. Three of the eight patients who were sensitized developed antibody-mediated rejection within a year of delivery, resulting in graft loss. Because the study was retrospective, noncontrolled, and from a single center, it is not definitive; rather, it serves as an indicator that risks to the mother, baby, and graft are higher in pregnancies of recipients who are sensitized.

Fatherhood after Kidney Transplantation

Male infertility in kidney failure is common and multifactorial. Dysregulation in the hypothalamic-pituitary-gonadal axis leads to low testosterone and hypogonadism in >50% of men with kidney failure. Additionally, young males may have congenital or hereditary kidney conditions associated with impaired fertility. For example, autosomal dominant polycystic kidney disease is associated with asthenozoospermia, prune belly syndrome is characterized by undescended testicles and infertility, and posterior urethral valves are associated with erectile dysfunction. Men with kidney failure commonly experience impaired spermatogenesis, lower semen volume, and reduced sperm motility and viability, with severity correlating with dialysis duration. Additionally, erectile dysfunction is reported in over half of the kidney failure population, resulting from underlying conditions such as diabetes, medications, poor body image, anxiety, and depression. As a result of infertility, spontaneous pregnancies fathered by men with kidney failure occur at significantly lower rates than in the general population (39).

Transplant surgery poses reproductive risks because the retroperitoneal exposure can result in damage to the spermatic cord structures, including the vas deferens, and testicular blood supply. However, successful transplant can result in overall improved male fertility and paternity rates for many. In one series, hypogonadism resolved in over half of patients within the first post-transplant year, with normalization of testosterone levels as early as 3 months after transplant (40). Improved sperm morphology and motility has additionally been reported in a subset of patients. Post-transplant improvement of erectile dysfunction can lead to higher paternity rates. For male recipients of transplants with persistent oligoasthenozoospermia, successful paternity-assisted reproduction with intracytoplasmic sperm injection has been reported (39).

In contrast to female recipients of kidney transplants, data from the TPR illustrate that outcomes of the offspring fathered by kidney transplant recipients on mycophenolate at the time of conception have been comparable with those in the general public in terms of gestational age, weight, newborn complications, and incidence of birth defects (25,41). Confirming these findings, data from the Norwegian Renal Registry show that children fathered by men taking mycophenolate products, unlike offspring exposed in utero, are not at higher risk for malformations (42).

Registry data also suggest that paternal exposure to corticosteroids, calcineurin inhibitors, and azathioprine does not cause higher risk of obstetric complications or birth defects. Sirolimus may cause lower sperm counts, dysmotility, and reduced spontaneous pregnancy rates (43). Male recipients of transplants seeking paternity should be counseled regarding the potential effects of sirolimus on fertility. There are limited data on pregnancy rates and outcomes with fathers receiving basiliximab, anti-thymocyte globulin, and belatacept at the time of conception.

Ethics and Realities of Post-Transplant Parenthood

In 2006, McKay et al. (35) published results of a survey that questioned transplant surgeons and nephrologists about transplant pregnancy management. One survey item asked whether respondents advised recipients of transplants to avoid pregnancy. Of all respondents, 82% affirmed they did give such advice—although for most, it was for a limited time period, and 15% advised patients to completely avoid pregnancy. Hopefully, at approximately 15 years and thousands of births later, we have moved past that recommendation. Generalizations about pregnancy for all recipients of kidney transplants are no longer useful. Decisions should be made on the basis of known risks and the risk tolerance of the individual. It is important to think about the realities of life and survival with a kidney transplant, as illustrated in a retrospective Dutch study of 42 patients who had been transplanted between 1971 and 2016 (44). Their median transplant to first delivery time was 6.5 years, with median follow-up of 12.5 years. Graft survival in these women was better than the rest of their transplant cohort. Nevertheless, five (12%) patients died 1–20 years after childbirth. They did not see their children reach adulthood. This rate is higher than the general Dutch population, in which 4% of children lose a parent before adulthood. Furthermore, 40% of the women who had received a transplant were back on dialysis before their child attended elementary school. Ross (45) spoke to the issue of expected life span of recipients of transplants, arguing that, although there are never guarantees that any prospective parents will remain healthy until their children reach adulthood, “the greater likelihood of a lower maternal life expectancy is morally relevant in that physicians have an obligation to encourage women to consider their reproductive decisions both from their own perspective and from that of the child-to-be.”

Contraception

Consensus recommendations for contraceptive use in recipients of solid organ transplants were published by the AST in 2005 (4). The Centers for Disease Control and Prevention (CDC) recommendations for contraception use in recipients of solid organ transplants were published in 2016 (46). Additional information on the safety and efficacy of contraceptive options in recipients of transplants, most notably the IUD, has been published since. Currently available contraceptive options, relative efficacy, and prescribing considerations are listed in Table 1.

The CDC contraception recommendations were separated by stable versus “complicated” graft function. Complicated graft function was defined as acute or chronic graft failure or rejection, without denoting a threshold of kidney impairment. CDC recommendations additionally exist for common transplant-associated conditions, including diabetes, hypertension, lupus, and prior deep venous thrombosis. All hormonal methods are categorized as safe in women with stable graft function. However, combined hormonal contraception is not recommended in women with complicated graft function, uncontrolled hypertension, history of stroke, thrombosis, or hypercoagulable state (46).

The copper and levonorgestrel IUD are designated safety grade 3 (theoretic or proven risks usually outweigh the advantages) for initiation of therapy in complicated graft function, but may remain in place (safety grade 2) if graft dysfunction develops after placement. Early concerns of IUD failure and the theoretic risk of pelvic inflammatory disease in recipients of transplants have not been supported by observational studies (47,48). Advantages of the IUD (including low failure rate, ease of use, and lack of immunosuppressive drug interactions and systemic side effects), along with restoration of fertility upon removal, have resulted in recommendations for its use among many transplant professionals (6,7,11).

Successful use of hysteroscopic sterilization in recipients of transplants is limited to case reports and can be confirmed by hysterosalpingography 3 months postprocedure (49). Use of emergency contraception (EC) in recipients of transplants has rarely been reported; in one study, 16% of 118 women reported EC use. The specific agents used and pregnancy rates after EC were not reported (9).

Conclusions

The thousands of successful pregnancies that have followed the birth of Edith Helm’s son >60 years ago make clear that pregnancy after kidney transplantation can occur safely and without negative effects on the mother, baby, or the kidney transplant. Nevertheless, it is high risk and safest when planned so that immunosuppression can be modified and risks are well understood.

Disclosures

M.A. Josephson reports serving as a scientific advisor or member of American Society of Nephrology, American Society of Transplantation, National Kidney Foundation, and Mycophenolate Pregnancy Exposure Registry Advisory Committee; receiving honoraria from American Society of Nephrology Highlights and Board Review Course and Update; receiving research funding from Bucksbaum Institute and Gift of Hope; having ownership interest in Seagen; having consultancy agreements with UBC Pharmaceutical Support Services for the Mycophenolate Pregnancy Registry and Labcorp; and being employed by the University of Chicago. C.L. Klein reports serving on the advisory board for CareDx and speakers bureau for CareDx, Sanofi, and Veloxis; having consultancy agreements with, and receiving honoraria from, CareDx, Sanofi, and Veloxis; serving as a scientific advisor or member of LifeLink Board of Governors; being employed by Piedmont Transplant Institute; and having patents and inventions with UpToDate.

Funding

None.

Footnotes

References