Pregnancy and liver transplantation
Sammy Saab, MD, MPH, AGAF, UCLA Medical Center, Pfleger Liver Institute, University of California Los Angeles David Geffen School of Medicine at UCLA, 200 Medical Plaza, Suite 214, Los Angeles, CA 90095, USA
Tel: +310 206 6705
Fax: +310 206 4197
Since the first pregnancy in a transplant recipient in 1958, pregnancy in recipients of solid organ transplants has become increasingly common. Although previously considered a hazardous event, data collected over the last 50 years demonstrate that despite an increased risk of maternal and fetal complications, pregnancy in transplant recipients can have a successful outcome. As of 2006, there were over 3000 female liver transplant recipients of childbearing age in the USA. Two hundred and two pregnancies and 205 outcomes were reported in 121 liver transplant recipients in the National Transplantation Pregnancy Registry. Children born to female liver recipients have a greater risk of prematurity and low birth weight than the general population, but no malformation patterns have been observed. Mothers are more likely to experience pregnancy-induced hypertension, pre-eclampsia and caesarian section, but overall mortality is not worse. Rates of acute rejection and graft loss are similar to nonpregnant liver recipients. The optimal timing of conception post-transplant is controversial, but current recommendations suggest waiting for at least 1 year after transplantation. Choice of contraception is also debatable, although barrier methods have traditionally been preferred. Many medications used for post-transplant immunosuppression have potential effects during pregnancy and breast-feeding. The risks and benefits of each medication should be reviewed with patients contemplating pregnancy, and regimens should be tailored accordingly.
Since the first pregnancy occurring in a kidney transplant recipient in 1958 (1), pregnancy in post-transplant patients has become increasingly common. Although previously considered a hazardous event, data collected over the last 50 years demonstrate that despite an increased risk of maternal and fetal complications, pregnancy in transplant recipients can have a successful outcome.
As of 2006, there were over 3000 female liver transplant recipients of childbearing age (between 18 and 34 years) in the USA (2). As of January 2006, 202 pregnancies and 205 outcomes have been reported in 121 female liver transplant recipients in the National Transplantation Pregnancy Registry (NTPR) (3). The mean time from transplant to conception was 4.5 years. There was a greater risk of prematurity (35 vs 11.0–12.7%) and low birth weight (34 vs 8.2%) in children born to female liver transplant recipients than in the general population, but no malformation patterns were observed (3). Maternal complications were also more common, with higher rates of pregnancy-induced hypertension (34 vs 4–10%), pre-eclampsia (22 vs 6–8%) and caesarian section (35 vs 20–25%) than the general population, but there are no data suggesting that overall maternal mortality is worse. Seven percent of liver recipient pregnancies were complicated by acute rejection and 8% had graft loss within 2 years of delivery, not significantly different from nonpregnant liver transplant recipients (3).
Timing of pregnancy
The optimal timing of conception is a matter of debate, but experts generally recommend waiting for at least 1 year post-transplantation. At this time, graft function is more likely to be optimal on lower maintenance doses of immunosuppressants, the risk of acute rejection is minimized, the risk of opportunistic infection is lower and other medical conditions such as diabetes and hypertension tend to be better controlled (4). A recent study of renal transplant recipients found that 11 patients who conceived ≤1 year after transplantation had similar maternal and fetal complications as 63 patients who conceived >1 year after transplant (5).
Several studies demonstrate that outcomes may be further improved the longer conception is postponed after transplantation. In a single-centre study in England with 71 pregnancies in 45 post-liver transplant women, pregnancies that occurred within 1 year after transplantation had an increased incidence of prematurity (58 vs 29%), acute cellular rejection (33 vs 14%) and caesarean section (50 vs 24%) compared with those that occurred later than 1 year (6). Findings in a retrospective review of 38 pregnancies in 29 liver recipients were similar. In this study, the interval between transplantation and pregnancy was shorter in the group that had spontaneous and therapeutic abortions (24.4±24.3 months) than in the group that had live births (47.8±28.7 months) (7). Of the pregnancies that occurred <12 months post-transplant, 86% (6 of 7) ended in first trimester abortion. In contrast, in pregnancies occurring >24 months post-transplant, 22% (5 of 23) ended in abortions and 78% (18 of 23) resulted in live births (7).
The length of time patients should wait from transplant to conception is controversial. In general, pregnancy at least 1 year post-transplantation appears safe, but rates of maternal and fetal complications may improve the longer conception is postponed.
The preferred method of contraception in post-transplant patients is also debatable. There is consensus, however, that it should begin soon after transplantation, as menstruation can occur as early as 1 month post-transplant (8).
Barrier methods have traditionally been preferred, as they do not have any inherent systemic effects, but some reports argue that high noncompliance and failure rates may make this suboptimal (4, 9).
Intrauterine devices (IUDs) have generally not been recommended, as immunosuppressive medications theoretically decrease their efficacy and increase the risk of infection. However, this may not be accurate, as the majority of immunosuppressants used in the post-transplant setting interfere with T-cell function, while IUDs rely on macrophages for their mechanism of action (9). Additionally, the use of IUDs is effective and accepted in HIV-infected women, who are immunosuppressed, similar to post-transplant patients (9). Furthermore, in the nontransplant population, the risk of infection with IUDs is negligible. The highest risk occurs during the first 20 days after insertion, with studies showing varied risk ranging from 1 to 10 per 1000 women. After the first month, infection with IUDs is rare, and when it does occur, it is most often caused by a new sexually transmitted disease. Further studies are needed, however, to better determine the safety and efficacy of IUDs in the post-transplant setting.
Data regarding the use of hormonal contraception in post-transplantation patients are limited. A recent study suggested that low-dose hormonal contraceptives may be well tolerated and effective if used when graft function is stable and hypertension is well controlled (10).
The risks and benefits of each potential contraceptive method should be reviewed with every patient, and an individualized decision should be made. Regardless of the ultimate method chosen, contraception is important in all transplant recipients of childbearing age, and should be discussed with patients before transplantation (Table 1).
Table 1. Contraception after liver transplantation*
|Barriers||No contraindication in transplant patients|
No drug interactions
Higher failure rates
|Recommended, particularly early after transplantation|
| Male condom|| ||15% failure rate|| |
| Female condom|| ||21% failure rate|| |
| Cervical cap|| ||Require fitting by healthcare|
| Diaphragm|| ||16% failure rate|| |
|Intrauterine device||Long lasting|
Low failure rate
No drug interactions
|Theoretical risk of|
ineffectiveness and infection
|Hormonal contraception||Lower failure rates||No protection against STDs|
Theoretical decreased efficacy
|Basis on risk/benefits of individual patient|
Hypertension should be well controlled
| Progestin only|
| Oral||5% failure rate||Metabolism through liver|| |
| DPMA, intramuscular||2% failure rate||Irregular bleeding amenorrhoea|| |
|No liver metabolism||Delayed return to fertility|| |
| ||Decreased bone mineral density|| |
| Etonorgestrel implant||<1% failure rate|
No decrease in bone
No liver metabolism
|Similar to DMPA|| |
| Oestrogen/Progestin||Less irregular bleeding||May interfere with|
| Patch|| ||Conflicting data regarding risk of|
| Vaginal ring||No liver metabolism|| || |
As in other solid organ recipients, fetal loss, prematurity and low birth weight are more common in pregnancies in female liver recipients than in the general population (3, 11). The 2005 NTPR reported 5% therapeutic abortions, 19% spontaneous abortions, 2% still births and 74% live births among 205 outcomes in 121 liver transplant recipients (3). Of the 151 live births, there were 35% premature births, 34% with low birth weight (<2500 g) and 30% with newborn complications (3). As the newborn complications were self-reported, it remains unclear what the most common ones are. Rates of preterm delivery have been reported to be as high as 55% (12), most commonly a result of pre-eclampsia, hypertension and premature rupture of membranes (13). Rates of preterm delivery and low birth weight are lower in liver transplant recipients than in kidney transplant recipients. Kidney transplant recipients are more likely to experience hypertension and pre-eclampsia (14, 15).
No malformation patterns associated with the post-transplant state have been identified. Fetal malformation rate among 285 pregnancies in liver recipients was 3%, which is comparable to the general population fetal malformation rate of 2–3% (13).
It has been well demonstrated that liver recipients are at a higher risk for hypertension (34 vs 4–10%) and pre-eclampsia (22 vs 6–8%) during pregnancy than the general population (3). The incidence of pregnancy-induced or worsening hypertension may be associated with the type of immunosuppressive regimen, the highest risk occurring with cyclosporine, followed by tacrolimus and then corticosteroids (6). These complications have been shown to occur more frequently in women with renal dysfunction (creatinine ≥1.3 mg/dl) and/or those on cyclosporine therapy (6, 16), as seen in the renal transplant literature. The risk of gestational diabetes does not appear to be higher in liver recipients and is comparable to that of the general population (3).
Pregnancy does not appear to be associated with an overall higher risk for hepatic allograft dysfunction or rejection than in nonpregnant liver recipients, except in women who conceive within 6 months of transplant (16). Of the 121 liver recipients in the 2005 NTPR report, 7% developed acute rejection during pregnancy and 8% suffered graft loss within 2 years of delivery (3). Experience with renal transplant patients suggests that immunosuppression monitoring is essential particularly during pregnancy to prevent graft dysfunction (17).
Modes of delivery
Vaginal delivery is possible in liver recipients, but caesarean section should be performed when indicated for obstetrical complications. The 2005 NTPR reports that of 151 live births in 121 liver recipients, 35% underwent caesarean section (3). The rate of caesarean section in the general population is approximately 20–25%. Reasons for caesarean section in liver recipients are mostly obstetrical complications such as pre-eclampsia, fetal distress, breech presentation, failed vaginal birth after caesarean and previous caesarean delivery (6). The higher rate of caesarean section among liver transplant recipients is probably related to the higher rate of pre-eclampsia and preterm delivery rather than the actual liver transplantation itself.
With an increased number of antirejection medications available today, combination therapy allows adequate immunosuppression to prevent graft rejection with smaller doses of each individual drug, thereby decreasing potential toxic effects. In the US, the majority of pregnant liver recipients have been on cyclosporine and prednisone with or without azathioprine (3) (Table 2).
Table 2. Immunosuppressants used in liver transplant recipients
|4% Malformation rate|
Premature rupture of
|Yes||(7, 18, 19)|
GI side effects
|7% Malformation rate|
Immune deficiency Infection
|3–5% Malformation rate||No||(7, 16)|
|6% Malformation rate|
GI side effects
45% 1st trimester
|No||(3, 30, 31)|
|OKT3||C||Flu-like symptoms||Unknown||Unknown|| |
|Antithymocyte globulin (ATG)||C||Flu-like symptoms||Unknown||Unknown|| |
Corticosteroids, classified by the Food and Drug Administration (FDA) as Pregnancy Category B drugs, have not been shown to affect fertility adversely (18). The effects of corticosteroids during pregnancy on the mother include exacerbation of gestational diabetes and hypertension in addition to the same adverse effects in nonpregnant patients. Corticosteroids have been associated with premature rupture of membranes, possibly by weakening membranes or altering vaginal flora, as well as adrenal insufficiency in newborns (19). The effect of high-dose pulse steroid therapy on the fetus is less clear.
Only a small percentage of corticosteroid administered to the mother reaches the fetus. The incidences of fetal adrenal suppression and infection are low in those exposed to corticosteroids, although monitoring should be performed routinely. Although a small amount of glucocorticoids are present in breast milk, breast-feeding is considered safe in mothers taking corticosteroids according to the American Academy of Pediatrics (AAP) (19).
An FDA Pregnancy Category D medication, azathioprine, suppresses cell-mediated immunity and alters antibody production by inhibiting purine metabolism. Azathioprine has been reported to cause chromosome breakage, and thus should be discontinued in males and females considering having a child (20).
Once pregnancy is achieved, however, one can consider continuation of azathioprine treatment. Although the drug crosses the placenta, the embryo lacks the enzyme to convert azathioprine and 6-mercaptopurine to active metabolites (21). Studies in humans suggest that the risk of teratogenicity is minimal to small, although data are limited and large, well-controlled studies are lacking (22). Earlier studies of women on azathioprine during pregnancy have reported fetal anaemia, thrombocytopaenia, lymphopaenia, immune deficiency and infection (23, 24). Recent studies have demonstrated that azathioprine treatment during pregnancy is associated with lower birth weight, lower gestational age and prematurity (25).
Breast-feeding is not recommended during azathioprine treatment, as azathioprine and its metabolites enter breast milk at low levels.
Cyclosporine, FDA Pregnancy Category C, is a calcineurin inhibitor that interferes with T-lymphocyte activity. Immunosuppression with cyclosporine during pregnancy is more often associated with maternal complications such as renal dysfunction and hypertension than corticosteroids or tacrolimus (16). Clinical data and literature reviews do not indicate an increased risk of congenital malformations; however, there is a moderate risk for fetal growth restriction.
Breast-feeding while on cyclosporine is not recommended according to the AAP because of the possible immunosuppressive effects on the infant. The level of cyclosporine detected in the infant varies widely regardless of the dose that the mother is taking.
Tacrolimus, FDA Pregnancy Category C, is a calcineurin inhibitor that is more potent than cyclosporine. In a prospective study of 37 post-liver transplant gravid women on tacrolimus, lower rates of pregnancy-induced hypertension and pre-eclampsia were seen in patients on tacrolimus than in those on cyclosporine (26). In a literature review of 83 solid organ recipients (66% liver and 27% kidney) treated with tacrolimus during pregnancy, the incidence of fetal malformations was 5.9% (4/68 life births) and there was no consistent pattern of malformation (27). Preterm delivery and low birth weight rates are comparable to other forms of immunosuppression (26).
Tacrolimus is secreted in breast milk. There are no adequate studies determining the safety of breast-feeding while on tacrolimus. A few case reports suggest that tacrolimus therapy may be compatible with breast-feeding (28, 29).
Mycophenolate mofetil or mycophenolic acid
Mycophenolate mofetil (MMF) was changed from FDA Pregnancy Category C to D in October 2007, based on post-marketing data from the NTPR and the Roche worldwide adverse reporting system. MMF is associated with increased risk of first trimester pregnancy loss as well as increased risk of congenital malformations. Of the 24 female transplant recipients (three liver recipients) with 33 pregnancy outcomes who were exposed to MMF reported in the NTPR, there were 15 (45%) spontaneous abortions. Of the 18 live births, four (22%) had structural malformations (3). Post-marketing data collected between 1995 and 2007 found that 25 of 77 (33%) women exposed to MMF during pregnancy had spontaneous abortions and 14 (18%) had an offspring with malformations. Of the malformations, six of the 14 (43%) had ear malformations. Other congenital malformations reported in offspring exposed to MMF in utero include cleft lip and palate, micrognathia, hypertelorism and anomalies of the eyes, distal limbs, heart, oesophagus and kidney (30, 31). MMF is secreted into breast milk; therefore, breast-feeding is contraindicated in mothers on MMF.
All women of childbearing age should receive contraceptive counselling and should have a negative pregnancy test before starting MMF therapy. Patients should use two methods of contraception, beginning at least 4 weeks before initiation of MMF, and continue the use of contraception during treatment and for at least 6 weeks after discontinuing MMF. MMF reduces blood levels of oestrogen and progestin and may theoretically decrease the efficacy of hormonal contraception.
OKT3 and antithymocyte globulin
OKT3, a monoclonal antibody, and antithymocyte globulin (ATG), a rabbit-derived antibody against human T cells, are both FDA Pregnancy Category C drugs used to treat acute graft rejection. Animal reproductive studies have not been conducted, and it is unknown whether OKT3 and ATG use in pregnancy can cause harm to the fetus. It is also unknown whether OKT3 and ATG are secreted into breast milk.
Male liver transplant recipients and fertility
Although the mechanism and details are not clearly understood, males with end-stage liver disease have hypothalamic–pituitary–testicular dysfunction resulting in decreased levels of testosterone, free testosterone and luteinizing hormone (32). Before liver transplantation, men often suffer from functional impotence as well as azoospermia (32). After successful liver transplantation, levels of testosterone, free testosterone and luteinizing hormone return to normal levels and sperm have normal density, motility and form in the majority of patients (32). Outcomes of pregnancies fathered by male transplant recipients are similar to that of the general population (3).
Pregnancy outcomes in female liver transplant recipients are similar to nontransplant women. Physicians should be knowledgeable regarding the timing of pregnancy, methods of contraception and the safety and adverse effects of immunosuppressants as they relate to pregnancy. The risk and benefits of continuing a recipients' immunosuppression strategy should be carefully discussed.