Transplantation affords recipients the potential for a full life and, for some, parenthood. Female transplant recipients must continue to take immunosuppression during pregnancy and breast-feeding. This article reviews case and series reports regarding breast-feeding in those taking transplant medications. Avoidance of breast-feeding has been the customary advice because of the potential adverse effects of immunosuppressive exposure on the infant. Subsequent studies have demonstrated that not all medication exposure translates to risk for the infant, that the exposure in utero is greater than via breast milk and that no lingering effects due to breast-feeding have been found to date in infants who were breast-fed while their mothers were taking prednisone, azathioprine, cyclosporine, and/or tacrolimus. Thus, except for those medications where clinical information is inadequate (mycophenolic acid products, sirolimus, everolimus, and belatacept), the recommendation for transplant recipients regarding breast-feeding has evolved into one that is cautiously optimistic.
Introduction
Breast-feeding and the use of human milk confer unique nutritional and non-nutritional benefits to the infant and the mother. The American Academy of Pediatrics (AAP) recommends exclusive breast-feeding for about 6 months, followed by continued breast-feeding as complementary foods are introduced, and continuation of breast-feeding for 1 year or longer as mutually desired by the mother and the infant . In the past decade, there has been a modest increase in the rate of “any breast-feeding” at 3 and 6 months of age, and data from the National Transplantation Pregnancy Registry (NTPR) regarding breast-feeding among solid-organ transplant recipients mirror that increase .
Breast-feeding can benefit the infant by decreasing the risks of various infections, for example, otitis media, respiratory infections, gastrointestinal (GI) infections , early allergies , celiac disease , and childhood inflammatory bowel disease (IBD) , as well as improving neurodevelopmental outcomes . Over the past two decades, the benefits of breast-feeding low-birth-weight and preterm infants have been demonstrated in controlled clinical trials in the general population . The AAP has incorporated these findings into their guidelines, which state that the benefits are such that all preterm infants should receive human milk . Feeding preterm infants breast milk is also associated with a decrease in the incidence of necrotizing enterocolitis, and it has been found that extremely preterm infants receiving the greatest proportion of human milk in the neonatal intensive care unit (NICU) had significantly greater scores for mental, motor, and behavior ratings at ages 18 and 30 months . As low-birth-weight and preterm delivery can occur in 30–80% of live births in female transplant recipients depending on the transplanted organ , breast-feeding or administering human milk to these infants could be an effective therapeutic intervention. This potentially could reduce the risk of sepsis and necrotizing enterocolitis in the perinatal period, support growth, and improve neurodevelopment outcomes. Moreover, there are health benefits to the mothers who breast-feed, such as decreased postpartum blood loss, rapid involution of the uterus, and increased child spacing due to lactational amenorrhea .
The current AAP guidelines note that some drugs are not excreted into human milk in clinically significant amounts, and detecting the presence of a drug in human milk does not always imply a risk to the infant. It is recommended that physicians consider multiple factors when counseling those who must take medications while breast-feeding, including the need for the drug by the mother, the potential effects of the drug on milk production, the amount of the drug excreted into human milk, the extent of oral absorption by the breast-feeding infant, and the potential adverse effects on the breast-feeding infant, noting that the most critical time for adverse events associated with drug exposure via lactation is within the first 2 months .
Breast milk composition
Human milk composition varies according to the sequential phase during which it is secreted: colostrum, transitional milk, and mature milk. Colostrum, the first milk, is produced prepartum through days 1–5 post partum and contains approximately 58 Kcal/dL compared to mature milk with 70 Kcal/dL. Colostrum also has the highest concentration of immunoglobulins and a higher protein content. Transitional milk may appear as soon as 12 h post partum and can continue for up to 2 weeks. The concentrations of lactose, fat and total calories increase in transitional milk while the concentrations of immunoglobulins and proteins, decrease. Mature milk, appearing as early as 3 days post partum and becoming predominant by day 9, contains the lowest concentration of immune factors.
Breast milk is considered an immune promoter that nourishes, protects, and enhances the growth and development of the infant’s GI tract . Secretory immunoglobulin A (IgA) is the primary antibody found in human milk, although other immunoglobulins (IgG, IgM, and IgE) are found in small decreasing amounts. The cells found in colostrum and breast milk are predominantly macrophages with some polymorphonuclear leukocytes and lymphocytes . Milk prolactin acts as a regulator of neonatal immune system development; thus, breast milk constituents may serve as neonatal immunodevelopment agents . Milk cytokines are thought to prevent autoimmunity and anti-inflammatory effects in the upper respiratory and GI tracts and stimulate host defenses and the development of the mucosal immune system of the GI tract.
Breast milk composition
Human milk composition varies according to the sequential phase during which it is secreted: colostrum, transitional milk, and mature milk. Colostrum, the first milk, is produced prepartum through days 1–5 post partum and contains approximately 58 Kcal/dL compared to mature milk with 70 Kcal/dL. Colostrum also has the highest concentration of immunoglobulins and a higher protein content. Transitional milk may appear as soon as 12 h post partum and can continue for up to 2 weeks. The concentrations of lactose, fat and total calories increase in transitional milk while the concentrations of immunoglobulins and proteins, decrease. Mature milk, appearing as early as 3 days post partum and becoming predominant by day 9, contains the lowest concentration of immune factors.
Breast milk is considered an immune promoter that nourishes, protects, and enhances the growth and development of the infant’s GI tract . Secretory immunoglobulin A (IgA) is the primary antibody found in human milk, although other immunoglobulins (IgG, IgM, and IgE) are found in small decreasing amounts. The cells found in colostrum and breast milk are predominantly macrophages with some polymorphonuclear leukocytes and lymphocytes . Milk prolactin acts as a regulator of neonatal immune system development; thus, breast milk constituents may serve as neonatal immunodevelopment agents . Milk cytokines are thought to prevent autoimmunity and anti-inflammatory effects in the upper respiratory and GI tracts and stimulate host defenses and the development of the mucosal immune system of the GI tract.
Immunosuppression during breast-feeding
According to the AAP, a drug’s chemical properties, such as low maternal serum protein binding, lack of ionization, small molecular weight, low volume of distribution, and high lipid solubility, facilitate drug excretion into human milk. A drug with a longer half-life is more likely to accumulate in human milk and the infant will more easily absorb a drug with higher oral bioavailability . Although minimizing medication exposure in utero would appear to be the safest for the infant, discontinuation of immunosuppressive therapy during pregnancy is not an option for transplant recipients. The AAP notes that deciding to breast-feed while continuing treatment with an agent for which in utero exposure also has occurred is different from initiating a new therapy while breast-feeding .
In order to better counsel new mothers regarding the risks associated with breast-feeding on a given medication, in 2008, the US Food and Drug Administration (FDA) proposed a revision to the drug labeling regulations regarding lactation. The final revision is expected to include three subsections: risk summary, clinical considerations, and supporting data. While the AAP guidelines contain specific recommendations to avoid breast-feeding while taking some medications (e.g., psychotropic agents, narcotics, amphetamines, chemotherapy agents, ergotamines, and statins), clinicians are referred to the LACTMED database for available information to help them make recommendations to female transplant recipients considering breast-feeding while taking immunosuppressants .
Transplant recipients must take immunosuppressive medication to preserve organ function. In an effort to improve both efficacy and decrease side effects, the different immunosuppressive medications are used in various combinations. Usually, recipients are maintained on a calcineurin inhibitor plus an antiproliferative agent with or without prednisone. One of the most common posttransplantation regimens currently prescribed is tacrolimus, mycophenolate mofetil, and prednisone.
The following sections highlight the current understanding regarding the peripartum risks associated with maintenance immunosuppressive agents.
Corticosteroids
Since the first transplants were performed in the early 1960s, prednisone has been a cornerstone of maintenance immunosuppressive regimens. Common dosages range from 5 to 10 mg daily for transplant recipients. The placental metabolism of prednisolone and methylprednisolone results in <10% of maternal dose exposure to the fetus . Similarly, exposure from breast milk remains at most 0.1% of total maternal dose of prednisolone, which is also <10% of the endogenous corticosteroid production by an infant . Published data regarding the effects of corticosteroids on lactation and the breast-fed infant indicates that a maternal dose of up to 20 mg/day results in low detectable levels in breast milk and is not expected to cause any adverse effects on the infant. Corticosteroids are routinely administered to both posttransplant and non-transplant mothers at the risk of premature delivery to accelerate fetal lung maturation. In a comprehensive review of animal studies regarding the effects of antenatal corticosteroid exposure on lung and brain function, and on growth restriction, both beneficial and adverse effects were found. Repeated doses of antenatal steroids may have beneficial effects on lung function, but may have adverse effects on brain function and fetal growth . Human studies have been conducted to assess long-term outcomes in the offspring of mothers who were administered antenatal corticosteroids. One study revealed no adverse effects on growth, sensorineural function, cognition, or lung function 14 years after the administration of a single dose of antenatal betamethasone . A comparative 30-year follow-up study conducted in conjunction with the Auckland Steroid Trial demonstrated no differences in lung function or the incidence of asthma between those whose mothers had been given recommended doses of betamethasone versus those whose mothers received placebo . Studies of the effects of betamethasone administration on lactation were conducted by Henderson et al. who showed that antenatal corticosteroid administration between 28 and 34 weeks of gestation resulted in greater milk volume if delivery occurred 0–2 days after steroid treatment, while women who delivered 3–9 days after betamethasone administration expressed the least milk volume . Although lactation may be temporarily affected by high doses of corticosteroids , no adverse effects on the infants have been reported with exposure to corticosteroids during breast-feeding .
Azathioprine
From 1962 until the early 1980s, azathioprine (AzA) in combination with corticosteroids was the immunosuppressive standard of care and has been widely used among pregnant female transplant recipients. AzA is a prodrug that is metabolized to 6-mercaptopurine (6-MP). This metabolite is a potent inhibitor of purine de novo synthesis, and as a result a potent inhibitor of lymphocyte proliferation . There is a narrow therapeutic index for AzA and the potential for toxicity resulting in myelosuppression. AzA is used as an adjunctive drug and prescribed for transplant recipients at dosages from 0.5 to 1 mg/kg/day . It has been shown that serum 6-MP levels peak within 1 h of oral dose and may be detectable for up to 7 h .
Several case series have examined breast-feeding while on AzA (reviewed below) and, as with prednisone, the amount of AzA found in breast milk and infant serum is negligible. No long-term adverse effects due to breast-feeding have been noted in the children. Overall, breast-feeding while taking AzA is now considered acceptable by most experts.
Gardiner et al. studied the serum level of active metabolites of AzA in breast-fed infants and their mothers . Samples were taken at least 3 months post partum to ensure that AzA metabolite concentrations in the infant were a consequence of breast-feeding rather than in utero exposure. One mother was a kidney transplant recipient. AzA doses ranged from 1.2 to 2.1 mg/kg/day. All of the mothers were found to have appropriate amounts of active metabolites, but there were no detectable levels in the serum of the infants.
Sau et al. reported 10 mothers who were taking AzA as part of immunosuppression for various conditions (systemic lupus erythematosus (seven), kidney transplantation (two), and Crohn’s disease (one)) . The AzA doses ranged from 75 to 150 mg daily. The authors examined immediate postpartum levels of the active metabolites of AzA in the breast milk and serum of mothers and the breast-fed neonates. Most of the mothers provided samples of breast milk on days 3–4, 7–10, and 28 post delivery, with samples taken both before AzA dose and periodically for the next 12–18 h. Of the 31 breast milk samples, 6-MP was detected in two from one mother on day 28, at a level well below the concentration for a therapeutic effect. None of the neonates exhibited clinical or hematological signs of immunosuppression.
Christensen et al. reported on eight breast-feeding women who were taking AzA for IBD, with a daily dose ranging from 75 to 200 mg . The mothers were breast-feeding infants ranging from 1.5 to 7 months of age between five and eight times each day. The investigators measured trough plasma levels of 6-MP and collected samples of breast milk and serum 6-MP 30 and 60 min after the AzA dose and hourly for the next 5 h. Most peak 6-MP serum levels occurred within the first hour and all peaks were reached within approximately 3 h after AzA ingestion. Breast milk had significantly lower concentrations of 6-MP with a slightly longer delay in peak levels. The infant’s highest possible exposure equated to <1% of the maternal dose.
The long-term effects of AzA exposure during breast-feeding were assessed by Angelberger et al. who surveyed mothers on AzA ( n = 11) for IBD for 3–6 months after delivery and compared them to mothers who did not take AzA ( n = 12) . The average daily dose was 150 mg. The average breast-feeding period for mothers on AzA was 6 months (range 1–18 months) and 8 months for those not on AzA (range 3.5–23 months). The average age of children at the time of interview was 3.3 years (0.6–6 years) among those who were breast-fed by mothers on AzA ( n = 15) and 4.7 years in the control group ( n = 15). No significant difference was found when comparing the two groups for the number of hospitalizations or the rate of infections. To date, this is the longest published follow-up study of children exposed to AzA while breast-feeding.
Cyclosporine
In the 1980s, cyclosporine (CsA) was introduced and quickly became the primary immunosuppressant of choice due to lower acute rejection and improved early graft survival. CsA has a wide range of bioavailability due to differences in intestinal reabsorption, which has been mostly addressed with the microemulsion formulation. It is mostly protein bound and primarily metabolized by the liver through the cytochrome P450 system, with a half-life of 6.4–8.7 h . The usual dosage range of CsA is 2–10 mg/kg/day in two divided doses, and in contrast to corticosteroids and AzA, dosing is adjusted by measuring trough blood levels.
Early studies discouraged breast-feeding by mothers on CsA as it was excreted in colostrum and breast milk . In1983, Lewis et al. reported one of the first cases of successful pregnancy in a kidney transplant recipient who was treated with CsA. Breast milk was sampled on days 2–4 post delivery and noted to have levels ranging from 101 to 263 mcg/L. It was recommended that mothers on CsA avoid breast-feeding . Similarly, Flechner et al., in 1985 reported a term pregnancy in a mother with a kidney transplant treated with CsA. The CsA was present in the fetal circulation during gestation at similar concentrations to those in the mother. CSA was present in the maternal breast milk levels . Behrens et al. measured CsA levels in simultaneously obtained maternal and infant blood and breast milk and found that CsA levels in breast milk were 15–90% higher than in maternal blood, although the infant would absorb <5% of a normal CsA dose . Both authors confirmed Lewis’ recommendation that mothers avoid breast-feeding while taking CsA .
Several years later, however, Thiru et al. confirmed that fetuses are exposed to maternal blood concentrations of CsA during gestation, but found that the absorption of CsA during breast-feeding resulted in minimal CsA blood levels in the infant . As a result, this group concluded that when advising transplant recipients regarding breast-feeding, the advantages of breast-feeding should be balanced with the minimal risk of exposure to CsA in breast milk.
Nyberg et al. reported on seven transplant recipient mothers (five kidney and two kidney/pancreas) who breast-fed their infants while taking CsA and measured levels in both the blood of mothers and babies and in breast milk . The average ingestion of CsA through breast-feeding was <300 mcg/day (<0.1 mg/kg/day), and the serum CsA levels were undetectable in all infants. The authors noted that the therapeutic dose for transplanted children is higher than the adult dose, and that the greater in utero exposure during a transplant recipient’s pregnancy is not considered harmful to the developing fetus. Thus, they concluded that transplant recipients could be permitted to breast-feed while taking CsA. Additional studies have confirmed these findings .
While there is a lack of information about long-term outcomes, there is no evidence of immediate harm to infants who are breast-fed while their mother is taking CsA.
Tacrolimus
Tacrolimus is a macrolide immunosuppressant that was introduced in clinical transplantation in 1989 and is widely used to prevent rejection after solid-organ transplantation. The tacrolimus mechanism of action is similar to CsA although more potent, and the usual oral dosage range is 0.05–0.2 mg/kg/day, in two divided doses. The pharmacokinetics of tacrolimus is a complex process, which is further complicated in pregnancy .
As with early CsA studies, at first mothers taking tacrolimus were also discouraged from breast-feeding. Jain et al. looked extensively at tacrolimus exposure in the placenta, cord blood, infant serum, and breast milk samples from 21 liver transplant recipients taking tacrolimus during 27 pregnancies . While the study focused on gestational and peripartum effects, 10 colostrum samples were collected from six mothers. The breast milk samples had a median 50% concentration of tacrolimus compared to that in the mothers’ serum. The average follow-up of the 25 surviving infants (two had died due to extreme prematurity) was 39 months. The newborn had a 36% incidence of transient perinatal hyperkalemia and mild reversible renal impairment. Although these issues may have been normal findings resulting from reduced maternal kidney function and incomplete maturation of newborn kidneys, the authors concluded that mothers taking tacrolimus who breast-feed their infants may be subjecting them to additional risk.
Later studies suggested that tacrolimus may be compatible with breast-feeding. French et al. reported a case of a liver transplant recipient from whom tacrolimus levels were obtained in breast milk and maternal samples . The measurement of tacrolimus levels in exclusively breast-fed infant revealed that the average total daily ingestion of tacrolimus would be 0.06% of the mother’s weight-adjusted dose and, given the low bioavailability of tacrolimus, the infant would receive a maximum of 0.02% of the mother’s weight-adjusted dose. At 2.5 months, no physical or social developmental delays were noted in the infant.
Gardiner et al. measured maternal serum and breast milk levels of a kidney recipient on a daily dose of 4 mg of tacrolimus who exclusively breast-fed her infant . Maternal serum tacrolimus levels peaked within the first 2 h; however, breast milk concentration remained relatively stable, although it was higher than that reported earlier. The average daily ingestion through breast milk was determined to be about 0.5% of the maternal weight-adjusted dose, or <0.2% of the recommended pediatric dose for liver or kidney transplantation. Thus, the authors agreed with French et al. that breast-feeding while on tacrolimus may be safe.
Gouraud et al. reported on five organ transplant recipient mothers on tacrolimus, whose infants’ serum tacrolimus levels were measured within 2–3 weeks post partum while they breast-fed . The average daily dose among the mothers was 9.6 mg/day. There was no detectable tacrolimus level in any of the infants’ serum.
Bramham et al. reported 14 mothers (15 infants) who were taking tacrolimus during pregnancy and lactation . Twelve infants were exclusively breast-fed and three were bottle-fed. Serum tacrolimus levels from mothers and infants, as well as tacrolimus levels in cord blood and breast milk, were sampled between days 0 and 72. The tacrolimus levels in infant serum samples progressively declined by approximately 15% per day. Eight breast-fed infants went on to have undetectable tacrolimus levels by median day 14, while two bottle-fed infants had undetectable levels at days 8 and 20. The breast-fed infants did not have higher tacrolimus levels than the bottle-fed infants. The estimated absorption from the breast milk was equivalent to 0.23% of the weight-adjusted maternal dose and similar to other reports . It was noted that lower-birth-weight infants had higher tacrolimus levels, suggesting that preterm or low-birth-weight infants may take longer to metabolize tacrolimus, and monitoring levels in these infants may be beneficial. However, breast-feeding did not appear to affect the normal decline in infant tacrolimus levels post partum and that breast-feeding on tacrolimus should not be discouraged .
In general, tacrolimus is now considered acceptable to use during breast-feeding . However, long-term follow-up of infants breast-fed with exposure to tacrolimus is warranted.
Mycophenolic acid products
Two oral forms of MPA are available, the mofetil ester and the enteric-coated mycophenolate sodium. Both formulations undergo hydrolysis to form the active metabolite mycophenolic acid (MPA). Blood levels do not correlate with efficacy or toxicity . In animal studies, MPA was found to be present in the breast milk of rats. There is no published information about the measurements of MPA in human breast milk or the risks of breast-feeding while taking MPA. Because it is expected that MPA would also be secreted in human breast milk, the product labeling advises against breast-feeding while taking MPA due to the lack of information. The high rate of spontaneous abortion and a specific pattern of birth defects (microtia and other facial malformations) found when women take MPA during the first trimester mandated the recommendation that females of childbearing potential use contraception while taking MPA and discontinue MPA 6 weeks prior to conception being added to product labeling in 2008. It is unlikely that much additional data will become available regarding breast-feeding on MPA as health-care providers heed this warning and recipients discontinue MPA and/or switch to AzA before they conceive and often remain off MPA into the peripartum period. Seven infants born between 2002 and 2008 to transplant recipients (five kidney and two heart) participating in the NTPR reported breast-feeding their infants while being maintained on MPA ( Table 1 ). Two infants had birth defects (hereditary corpus callosum thinning: n = 1; AV canal defect, duodenal atresia, and tetralogy of Fallot: n = 1), but to date none have reported adverse effects due to breast-feeding.
