Objectives: Pregnancy after kidney transplant has a high risk for maternal and fetal complications; however, it can be successful if patients are properly selected. Here, we studied outcomes and com-plications of pregnancies in kidney transplant recipients who received calcineurin inhibitor-based immunosuppression.
Materials and Methods: In this case control study, we reviewed patients who became pregnant between 2004 and 2017. For this analysis, each pregnancy was considered an event. We divided pregnancies into 2 groups according to calcineurin inhibitor-based main-tenance immunosuppression: group 1 (49 pregnancies) received cyclosporine, and group 2 (33 pregnancies) received tacrolimus. Patients also received steroids and azathioprine. Patients had regular antenatal follow-up at the Hamed Alessa Organ Transplant Center (Kuwait) and in the maternity hospital (monthly until month 7 and then weekly until delivery).
Results: Of 750 female kidney transplant recipients within childbearing potential, there were 82 preg-nancies (10.9%) in 49 recipients (6.5%). Seventy-eight pregnancies were planned, and 4 pregnancies occurred while women were using contraception. There was 1 triple pregnancy, 5 double, and 76 single pregnancies. Two women had preeclampsia as maternal com-plication, 2 had uncontrolled hypertension, and 7 developed graft dysfunction. Forty-seven women (57.3%) had caesarean section, and the remaining had vaginal deliveries. Of 89 babies, 86 were viable (1 intrauterine fetal death and 2 abortions). Eight babies were delivered prematurely with low birth weight, and 2 needed incubators. Mean serum creatinine levels were 97.9 ± 24, 109 ± 38, 100 ± 39, 120 ± 46, and 115 ± 57 µmol/L at baseline, first, second, and third trimesters, and postpartum, respectively. Twelve patients showed high panel reactive antibodies but without donor-specific antibodies.
Conclusions: Posttransplant pregnancy can be successful in most renal allograft recipients, but the increased risk of fetal and maternal complications, including low birth weight, spontaneous abortus, and preeclampsia, should be considered.
Key words : Immunosuppression, Kidney transplantation, Preeclampsia, Posttransplant pregnancy
End-stage renal disease affects fertility negatively through hypothalamic-pituitary suppression, resulting in abnormal menstruation in most premenopausal women.1,2 Successful kidney transplant can lead to restoration of fertility, as early normalization of pituitary-gonadal hormone levels can occur from 3 to 4 months after transplant. In addition, both men and women have reported improved libido and sexual functions.3
Female kidney transplant recipients regain fertility early after renal transplant, allowing women to get pregnant.4 The renal allograft can adapt to the physiologic changes of pregnancy with an increase in creatinine clearance of approximately 30% in the first trimester, which is sustained with a small decrease in the second trimester and a return to prepregnancy levels during the third trimester.5 Allograft recipients also have a higher 24-hour protein excretion compared with healthy women, which increases throughout pregnancy, becomes threefold higher by the third trimester (regularly exceeding 500 mg vs 200 mg in healthy women), and returns to prepregnancy levels at 3 months post-partum. Proteinuria in pregnancy should never be attributed to normal pregnancy-related changes, and common comorbidities like urinary tract infection and preeclampsia should be ruled out.6
Pregnancy after kidney transplant has been considered as having high risk of maternal and fetal complications, but it can be successful if patients are properly selected. Since 1954, over 14 000 women have given birth after solid-organ transplant.4 It is difficult to distinguish preeclampsia from hypertension in renal transplant recipients because of the frequent increase in blood pressure after 20 weeks in previously normotensive women and hyperfiltration-related worsening of preexisting proteinuria. Hyperuricemia becomes a less reliable marker for diagnosing preeclampsia because renal transplant recipients are usually on calcineurin inhibitors (CNIs), which can also increase uric acid levels.7 In addition, sudden worsening of hypertension and a marked increase in proteinuria are also noted in acute rejection, which further makes the diagnosis of preeclampsia challenging. Hypertension during pregnancy increases the risk of preterm delivery, intrauterine growth retardation, and graft loss.
The rate of live births in allograft recipients is comparable to the general population and ranges from 71% to 79%.8 The incidence of preterm delivery has been reported to be as high as 40% to 60% versus 5% to 15% in the general population and occurs mostly due to maternal or fetal compromise rather than spontaneous preterm labor.9
Management of immunosuppression in pregnant renal transplant recipients is important due to the concern for teratogenicity risk and potential adverse effects. All immunosuppressive drugs cross the maternal-fetal circulation and have been detected in variable degrees in the fetal circulation.10 The Food and Drug Administration has categorized drugs for pregnancy safety as follows: A (no human risk), B (animal studies showing risk but no evidence of human risk), C (human risk not ruled out), D (evidence of human risk), and X (absolutely contraindicated). Most drugs fall into category C, where risk and benefits have to be weighed. The commonly used immunosuppressive drugs used in renal transplant recipients and their pregnancy information are summarized in Table 1.11 Mycophenolate mofetil should be discontinued at least 6 weeks before pregnancy, as it has been shown in case reports and registry data to cause fetal malformations, with ear, mouth, finger, and ocular/organ involvement.12 Although there are some concerns regarding azathioprine use, with fetal abnormalities shown in animals, most practitioners consider this medication as a safe alternative, based on its long record of use in transplant recipients. Both azathioprine and corticosteroids have little to no evidence of teratogenic risk. Although CNIs can cross the placenta and have been previously shown to have immunosuppressive properties in fetal blood, the US registry data have shown no pattern of congenital anomalies associated with these agents, and both cyclosporine and tacrolimus are usually continued during pregnancy.13 Data on the safety profile of sirolimus and everolimus are limited, and both are currently classified as category C and should be stopped before conception, as per the KDIGO guidelines.14
In this study, we evaluated outcomes of preg-nancies in kidney recipients who were using CNI-based immunosuppression, focusing on fetal and maternal complications.
Materials and Methods
This was a single-center case control study of a cohort of kidney transplant recipients who were followed at the Hamed Alessa Organ Transplant Center of Kuwait. The study design was reviewed and approved by the local Ethics Committee. Eligible patients were renal transplant recipients who became pregnant during the period between 2004 and 2017. All patients were followed at our center in collaboration with a maternity hospital in the Sabah area in Kuwait. Patients were determined to be pregnant by positive β-human chorionic gonadotropin test during, which was then confirmed by ultraso-nography. We focused on patient, fetal, and graft outcomes. All pregnancy data were collected from medical records maintained by the institution. For this analysis, each pregnancy was considered as an event.
All patients were maintained on steroids and azathioprine in addition to CNIs. We divided the pregnancies into 2 comparable groups according to type of CNI used as maintenance immunosup-pression: group 1 (49 pregnancies) had cyclosporine-based maintenance immunosuppression, whereas group 2 (33 pregnancies) had tacrolimus-based maintenance immunosuppression. All patients had regular antenatal follow-up in our center and in the maternity hospital (monthly until month 7 and then weekly until delivery).
All patients were clinically evaluated with special focus on infection and preeclampsia (body weight, blood pressure, edema of lower limbs). Laboratory assessments were conducted at each visit for kidney graft function, complete blood count, trough level of CNIs, serum electrolytes, urine analysis and culture, and 24-hour urine protein. Radiologic assessments were undertaken by antenatal sonography. Fetal growth was monitored to diagnose intrauterine growth retardation and uteroplacental insufﬁciency at the earliest. For patients with planned pregnancies, the immunosuppression protocol consisted of induction with antilymphocyte antibody with 5 daily doses of antithymocyte globulin at a dose of 1 mg/kg (rabbit-derived thymoglobulin; Sanofi Aventis, Inc., Bridgewater, NJ, USA) or 2 doses of interleukin 2 receptor blocker (basiliximab; Novartis, Inc., Basel, Switzerland) based on immunologic risk stratification. Maintenance therapy consisted of prednisone, mycophenolate mofetil, and CNIs. Sirolimus was given to patients who were stable with low immunologic risk and when conversion from a CNI was applicable. Acute cellular rejection was treated with intravenous pulse steroids (methylprednisolone at 1 g daily for 3 d) and/or thymoglobulin (1 mg/kg daily for 7-10 d) for steroid-resistant rejection. Antibody-mediated rejection was treated with plasma exchange, 2 g/kg intravenous immunoglobulin, and rituximab. All rejection episodes were diagnosed by biopsies and treated according to Banff criteria. All female kidney transplant recipients were maintained on azathioprine for at least 6 weeks before pregnancy.
Statistical analyses were performed with SPSS software (SPSS: An IBM Company, version 20, IBM Corporation, and Armonk, NY, USA). Means were compared using paired sample t test, independent sample t test, analysis of variance, chi-square test, and Fisher exact test as appropriate. Results are shown as means and standard deviation, with differences considered significant at P ≤ .05.
Of 750 female kidney transplant recipients within their childbearing period at our center, there were 82 pregnancies (10.9%) among 49 kidney allograft recipients (6.5%). Seventy-eight pregnancies were planned, and 4 women became pregnant while they were using contraception. There was 1 triple pregnancy, 5 double, and 76 single pregnancies. Two women had preeclampsia as maternal complication, 2 had uncontrolled hypertension, and 7 developed graft dysfunction during their pregnancies. Regarding mode of delivery, 47 women (57.3%) had cesarean section, with the remaining having vaginal deliveries. The total number of babies was 89 (86 viable, 1 intrauterine fetal death, and 2 abortions). Eight babies were delivered prematurely with low birth weight, and 2 needed incubators. Forty-two infants were males and 44 were females, with mean birth weight of 2.44 ± 0.6 kg.
Forty-nine women were maintained on a cyclosporine-based immunosuppressive regimen (group 1), and 33 were maintained on a tacrolimus-based regimen (group 2). The 2 groups matched regarding demographic data (pretransplant dialysis type, donor type, induction immunosuppression, and pretransplant comorbidities; Table 2). Twelve patients showed high panel reactive antibodies but without donor-specific antibodies. The number of patients with slow graft function immediately after transplant was significantly higher in group 2 (6 patients, 18.1%) than in group 1 (1 patient, 2.1%) (P = .03).
Mean serum creatinine levels were 97.9 ± 24, 109 ± 38, 100 ± 39, 120 ± 46, and 115 ± 57 μmol/L at baseline, first, second, and third trimesters, and postpartum, respectively. However, long-term graft function was significantly better in group 2 patients (P = .001).
Fetal outcomes in women maintained on tacro-limus were worse (2 abortions [6.1%] and 1 intrauterine fetal death [3.3%]); however, these results were not significant (P = .16; Table 3). The mean fetal body weight was comparable in both groups (2.5 ± 0.7 vs 2.57 ± 0.6; P = .97). Urinary tract infection in 1 patient was easily controlled by oral antibiotics.
Female kidney transplant recipients can regain fertility early after kidney transplant, enabling women to get pregnant. Pregnancy after kidney transplant has been considered as having a high risk for maternal and fetal complications; however, it can be successful if the patient is properly selected.
From 750 kidney transplant patients within their childbearing period, there were 82 pregnancies (10.9%) among 49 kidney allograft recipients (6.5%) at our center. Hold and associates reported that optimal contraception is important to initiate before transplant in women of childbearing age.15 Four patients in our cohort became pregnant while using contraceptions. This could be explained by the failure rate of different methods of contraception as reported by Kenny and colleagues, who showed that con-traceptions should be used before transplant because women could become pregnant during the peri-transplant period.16 The optimal timing of pregnancy depends somewhat on the circumstances of the transplant recipient. Historically, the recom-mendation has been to wait 2 years after successful transplant.17 This recommendation has been replaced by the American Society of Transplantation Consensus Opinion, which has stated that, as long as graft function is optimal (defined as a serum creatinine < 1.5 mg/dL, with < 500 mg/24 h protein excretion), no concurrent fetotoxic infections are present, the patient is not using teratogenic or fetotoxic medications, and immunosuppressive dosing is stable at maintenance levels, the patient can safely proceed with the pregnancy.18 We have adopted this policy at our center, with education sessions delivered to all female transplant patients within their childbearing period.
Coscia and associates reported that the incidence of preeclampsia in renal transplant recipients ranges between 24% and 38%, with a 6-fold higher risk, compared with an incidence of 4% to 5% in the general population.9 In our study, 2 women had preeclampsia and 2 had uncontrolled hypertension. The lower prevalence of preeclampsia and uncontrolled hyper-tension among our cohort could be due to the low-risk factors before conception (according to the American Society of Transplantation Consensus Opinion), with patients having stable graft function, minimal or no proteinuria, and none or 1 antihypertensive agent at the start of pregnancy. Moreover, we closely monitor our patients for potential high-risk pregnancies, and all patients are maintained on aspirin.
Sibanda and colleagues reported that hyper-tension was common in kidney transplant recipients with a reported incidence of 52% to 69%.8 Duley and associates19 documented that antihypertensives should be initiated if blood pressure is consistently higher than 140/90 mm Hg. Alpha-methyldopa and hydralazine are the traditional agents that have been used safely for controlling blood pressure during pregnancy. Other antihypertensive agents that are safe during pregnancy include beta-blockers and calcium channel blockers. Angiotensinogen-con-verting enzyme inhibitors are contraindicated due to their association with pulmonary hypoplasia and oligo-hydramnios in fetus. Low-dose aspirin can reduce the risk of preeclampsia in high-risk patients and should be given to all renal transplant recipients.19
The lower prevalence of hypertension in our cohort (4 of 82 patients) could be explained by the lower number of antihypertensive agents used by our patients before conception. All women with hypertension received alpha-methyldopa alone (n = 2) or in combination with a calcium channel blocker (n = 2).
In our study, 11 women developed graft dys-function (5 in group 1 and 6 in group 2; P > .05) during their pregnancies. Twelve patients showed high panel reactive antibodies but without donor-specific antibodies. Stratta and colleagues20 reported that pregnancy is a state of immunologic tolerance associated with immune-depressant activity of lymphocytes, which creates tolerance to the fetus and which may benefit the renal allograft. However, there is a possibility that the antigenic stimulus provided by the fetus may trigger graft rejection as well. In addition, acute rejection may be higher in the postpartum period due to return to normal immunosurveillance status.20 As shown previously, the rate of allograft rejection is not increased during pregnancy or 3 months postpartum and varies between 1% and 14.5%, which is comparable to nonpregnant transplant recipients.9,21 We reported 2 resistant rejections that resulted in graft failure; in these patients, the pregnancies were terminated fruitfully (1 patient was on dialysis support until labor). The remaining graft losses were due to nonimmunologic causes.
The risk of requiring cesarean section in renal transplant recipients is higher than in the general population, with a reported incidence of 43% to 64%.9,22 Our higher prevalence of cesarean section (57.3%) matched that reported by Bramham and associates, who showed that the likelihood of cesarean section in renal transplant recipients from a UK transplant registry was 5-fold higher and was twice as common compared with the general population (64% vs 24%), with most because of fetal distress and 3% due to the presence of the renal allograft.22 In our cohort, concern for the fetus was the main cause of cesarean delivery.
In our study, there were 86 viable births of 89 babies, with 8 babies delivered prematurely with low birth weight (2 needing incubators). This result matched observations from Coscia and associates9 and Sibanda and colleagues8 who reported that the rate of live births in allograft recipients is comparable to the general population and ranges from 71% to 79%. In addition, they reported that the incidence of preterm delivery was as high as 40% to 60% versus 5% to 15% in the general population and occurs mostly because of maternal or fetal compromise rather than spontaneous preterm labor.9 High serum creatinine ≥ 1.7 mg/dL and the presence of maternal hypertension can predispose to preterm delivery.8 In addition, renal transplant patients have a high incidence of preterm birth (52% to 53%), low birth weight (42% to 46%), and intrauterine growth retar-dation (30% to 50%).9,20,23 Renal allograft recipients have a 13-fold higher risk of preterm deliveries, 12-fold higher risk of low-birth-weight babies, and 5-fold higher risk of small for gestation babies compared with the general population as reported by Bramham and colleagues.22 The mean gestational age for newborns is 35.6 weeks with mean birth weight of 2420 g.24 The miscarriage rate ranges from 11% to 26% versus 8% to 9% in the general population; however, there is no higher risk of perinatal mortality in the absence of risk factors of hypertension, proteinuria, and impaired allograft dysfunction.8,9,22
Long-term graft function was significantly better in our group 2 (tacrolimus) patients (P = .001) (Figure 1). Such an observation could be explained by the better immunosuppressive effects of tacrolimus and less nephrotoxicity. Stoumpos and associates reported that there was no difference in estimated glomerular filtration rate at 1, 5, and 10 years in pregnant women compared with nonpregnant controls, and pregnancy was not associated with worse 10-year allograft or 20-year patient survival.25
Posttransplant pregnancy could be successful in most renal allograft recipients but can have increased risk of fetal and maternal complications, including low birth weight, spontaneous abortion, and preeclampsia. However, with meticulous follow-up, outcomes can be successful without significant adverse effects. Tacrolimus may be beneficial to long-term graft survival.
Volume : 17
Issue : 1
Pages : 99 - 104
DOI : 10.6002/ect.MESOT2018.O42
From the 1Nephrology Department, Hamed Al-Essa Organ Transplant Center, Sabah
Area, Kuwait; the 2Department of Dialysis and Transplantation, Urology
Nephrology Center, Mansoura University, Mansoura, Egypt; and the 3Department of
Internal Medicine, Faculty of Medicine Ain Shams University, Egypt
Acknowledgements: The authors want to acknowledge sisters in the outpatient clinic of Hamed Alessa Organ Transplant Center, especially sisters Sherin Farag, Bancy Baby, and Sijy Paul for their great efforts during follow-up and data collection of our patients. The authors have not sources of funding or conflicts of interest to declare.
Corresponding author: Ayman Maher Nagib, Urology and Nephrology Center, Mansoura University, Mansoura, Egypt
Phone: +96 560354347
Table 1. Common Immunosuppressive Drugs Used in Transplantation
Table 2. Demographic Characteristics of Study Patients
Table 3. Maternal and Fetal Characteristics of Pregnant Study Patients
Figure 1. Graft Function as Represented by Serum Creatinine Levels at Different Time Intervals