Introduction

Women with end-stage chronic renal failure have low fertility [1]. Recent advances in surgical techniques and immunosuppression have improved not only the survival, but also the quality of life of organ transplant recipients [2]. Following renal transplantation, normal menstrual function is restored within 1–20 months [3], with about 2–5% of women of child-bearing age experiencing a pregnancy, unintended in half of them [4, 5].

However, pregnancy in renal transplant recipients is not without risks [6]. The potential conflict between fetus, mother and allograft function needs to be carefully considered [6].

Pregnant women with renal disease are at an increased risk of worsened proteinuria and development of hypertension and/or preeclampsia. These maternal complications may lead to a higher incidence of intrauterine growth restriction and preterm delivery [7].

Although more than 50% of the newborns have no neonatal problems, fetal risks include intrauterine growth restriction (IUGR), intrauterine death, preterm delivery, preterm premature rupture of membranes (P-PROM), toxic effects of immunosuppressant drugs and fetal infections due to cytomegalovirus and hepatitis B virus [8].

Since the first description of a successful pregnancy in a renal transplant recipient by Murray in (1963), there have been significant changes in immunosuppressor therapies and graft outcomes [6].

The impact of pregnancy on allografts is still controversial [8], with some studies suggesting that pregnancy does not adversely affect kidney function [2, 3] and others pointing out just the opposite [9]. Moreover, the changes in plasma volume occurring during pregnancy may predispose to acute rejection episodes, jeopardizing long-term allograft survival [10].

The most important factors affecting pregnancy outcomes seem to be the length of time from transplantation to conception (women desiring a pregnancy should wait at least 2 years to conceive) and a stable renal function (with creatinine levels preferably lower than 1.5 mg/dl) [11–13]. Other recommended criteria include the absence of acute rejection episodes, a systemic arterial blood pressure of 140/90 mmHg (or less, while adhering to prescribed drug regimens), less than 500 mg of proteinuria per day, immunosuppressant drugs lowered to maintenance levels and a normal allograft ultrasound [14].

The objective of this study was to evaluate the risk factors affecting obstetric, perinatal and short-term graft outcomes in women who go through pregnancy following a successful renal transplantation.

Methods

This study was made in a single tertiary center; all the pregnant women were submitted to a thorough follow-up conducted by an obstetrician and a nephrologist. Appointments were conducted every 2–3 weeks during the first and second trimesters of pregnancy and weekly in the third.

From January 1989 to January 2007, 34 pregnancies were carried out in 28 renal recipients followed in our institution. Pregnancy outcome was gathered by chart in an in-patient record.

The outcomes of pregnancy, including the percentage of live births, miscarriages, therapeutic abortions and intrauterine deaths were determined by retrospective review.

The following maternal details were obtained: age at time of conception, time interval between pregnancy and transplant, presence of hypertension or other co-morbid conditions, type of immunosuppressor and creatinine clearance (1 month before and 1 month after pregnancy) and creatinine plasma values before, during and 1 year after pregnancy.

Creatinine clearance was calculated using the Cockcroft–Gault formula.

Mean values, standard deviations and proportions with paired Student T-test were used to assess the evolution of plasma creatinine levels and clearance of creatinine (in SPSS 15.0 software).

Maternal and obstetric complications of pregnancy were reviewed, namely new onset hypertension, gestational diabetes mellitus (GDM), pre-eclampsia (PEC), urinary tract infections (UTI) and incidence of cesarean section (C-section).

A 50 g, 1 h glucose test was performed to screen for GDM; abnormal findings (≥140 mg/dl) were followed by the 100 g, 3 h glucose tolerance test. GDM was defined as ≥2 abnormal glucose concentration values on the 3 h test at 24–28 weeks or 32–34 weeks.

PEC was defined as pregnancy-related hypertension (≥140/90 mmHg) developing after 20 weeks of gestation in previously normotensive women associated with a proteinuria of 300 mg/dl, or greater, irrespective of the presence or absence of edema.

Superimposed PEC was defined using the same criteria in previously chronic hypertensive women.

Details of the following neonatal outcomes were recorded: preterm delivery, intrauterine growth restriction (IUGR), birth weight, congenital abnormalities and infant survival.

Preterm delivery was defined when delivery occurred before 37 completed weeks of gestation.

By the ACOG definition, IUGR occurs when a fetus fails to reach his potential growth. It was diagnosed by the obstetrician by estimations of fetal growth patterns (an abdominal circumference less than 10th percentile or an estimated fetal weight below the 10th percentile for that gestacional age).

Small for gestational age (SGA) was defined as a birth weight below 10% of normal for the gestational age; low-birth weight (LBW) as a birth weight between 1,500 and 2,500 g and very low-birth weight (VLBW) when it was less than 1,500 g.

Phone and written inquiries were performed to evaluate the age and possible medical or intellectual problems of the 27 children born, with a response rate of 100%.

Results

We identified 34 pregnancies in 28 renal recipients, 6 of them with two pregnancies. The patients mean age was 27 ± 5.1 year (range 18–37) at the time of conception.

All of them were using an immunosuppressor protocol consisting of various combinations of prednisone, cyclosporine, azathioprine, tacrolimus and mycophenolate mofetil. During pregnancy, mycophenolate mofetil was replaced by azathioprine.

Pre-pregnancy characteristics of the patients are shown in Table 1.

Table 1 Pregnancy characteristics of the renal transplant recipients
Full size table

Of the 34 pregnancies, there were 27 live deliveries (preterm or term), 2 therapeutic abortions, 3 spontaneous abortions and 2 stillbirths. The therapeutic abortions were conducted in the case of a pregnancy that occurred within a month of transplantation and in the presence of a high grade cervical intraepithelial neoplasia. The spontaneous abortions occurred at 6, 8 and 19 weeks of gestation.

There were two stillbirths: one at 34 weeks of gestational age (GA) in a patient with a 9-month pos-transplantation period, with hypertension since the 32nd week; the other at 35 weeks GA was an intrapartum death in a renal recipient with recurrent UTI. Autopsy studies results revealed placental abruption in the first case and the existence of two cervical umbilical cord encirclements in the second case.

Among the 29 deliveries, there were 10 spontaneous vaginal deliveries (34.5%) and 19 C-sections (65.5%). The majority of C-sections were performed for obstetric reasons (pre-labor elective deliveries in 31.6% or 6/19).

The mean gestational age at delivery was 35.7 ± 2.2 weeks (range 31–39) and the prematurity rate was 59.3% (16/27).

The mean birth weight of the offspring was 2,465 ± 538 g (range 1,300–3,530); there was only one VLBW newborn (3.7%) and 13/27 LBW babies (48.1%).

There were three IUGR (11.1%) and four SGA (14.8%) babies; there was no neonatal death.

No congenital abnormalities were noted and of the 27 babies born, none showed renal disease.

Follow-up evaluation showed that all children were having a normal height and weight progress at a mean age of 106.7 ± 49.6 months (range 4–198). In the group of children older than 6 years, school results were consistently good (19/27).

Maternal complications consisted of ten urinary tract infections (29.4%); three GDM (8.8%), two controlled by diet but one needing insulin therapy; and three cases (8.8%) of anemia needing erythropoietin.

Pre-pregnancy hypertension was found in 41% (14/34) of pregnancies; among the remaining ones, 12% (4/34) had new-onset hypertension, with half the cases presenting in the last trimester, and the other half after delivery. Pre-eclampsia developed in two patients, superimposed on underlying chronic hypertension.

There were three cases of allograft rejection (8.8%): one acute failure during pregnancy caused by therapeutic non-compliance and two post-pregnancy rejections. There was also one case (2.9%) of allograft failure after pregnancy.

There was a statistically significant decrease of plasma creatinine during pregnancy (P < 0.001), with a significant rise after delivery (P < 0.001), but no difference between pre- and post-pregnancy values (P = 0.5).

A statistically significant increase in creatinine clearance was also observed 1 month after delivery (P = 0.02).

The outcomes in the three pregnancies conceived within a year of transplantation were poor: one stillbirth at 34 weeks of GA, one therapeutic abortion and one live child whose mother suffered an allograft rejection after pregnancy.

Discussion

Although kidney transplantation improves the reproductive prognosis for patients with renal failure, [15]. Nevertheless, renal transplant recipients may be optimistic about a successful outcome to their pregnancy if they meet the criteria for considering pregnancy, as set out in the European Best Practice Guidelines IV [14].

It has also been argued that higher pregnancy rate is more probable in present settings due to transplant selection criteria favoring young patients with primary renal disease and without end organ damage to other systems [16].

Our percentage of live births (79.4%) is similar to that reported in the literature, probably due to the strict application of the guidelines referred. The total miscarriage rate of 5/34 (14.7%) is similar to the one reported in the literature [17–22] and to the general population.

The prevalence of pre-conception hypertension (41%) is lower in our study than that in other series (63–77%) [17]. Maternal hypertension is one of the most important predictors of low birth weight and preterm delivery [18, 19].

The literature suggests that IUGR occurs in 9–40% of pregnancies [20, 21]; in our series it was the most frequent fetal complication, responsible for 13.8% of cases.

No congenital abnormality was detected in this study. Nevertheless, more information about the intrauterine effects and neonatal consequences of maternal immunosuppression are needed.

There were three cases of gestational diabetes mellitus (GDM) in our series, one of them requiring insulin therapy. It is reported that abnormal glucose tolerance test results occur in up to 13.8% of pregnant transplant recipients, whose immunosuppressive therapy consists of corticosteroids [16, 23, 24]. However, not only corticosteroids but also nearly all immunosuppressive drugs are associated with glucose intolerance [2].

Preeclampsia is diagnosed in approximately one-third of pregnant renal transplant recipients [8, 22, 25]. In this study, it occurred in 6.7% (2/29) of cases, a lower rate that might be explained due to the small number of cases.

The rate of preterm delivery was of 59.3%, higher than that in most series published [7]; the explanation for this difference is difficult due to the small number of cases.

Crowe et al. [26] reported that creatinine levels did not change significantly during pregnancy in renal transplant recipients as long as renal function was stable before pregnancy [24–32]. Moreover, pregnancy does not seem to have adverse effects on graft survival [32]. Our study showed similar results.

The risk of an acute allograft rejection during pregnancy is about 9% [24, 25], but rejection is unusual in women who waited 2 years to conceive [27, 32]. By contrast, chronic rejection with a progressive sub clinical course may be a problem in up to 15% of pregnant graft recipients [27].

In our group, there was one acute rejection caused by therapeutic non-compliance; two allograft rejections (12 and 63 months after delivery, respectively) and one allograft failure. Among the two rejections after delivery, one occurred 74 months after transplant in a patient whose initial creatinine level was 1.8 mg/dl, had chronic hypertension and delivered a 31 weeks GA preterm baby; the other occurred 63 months after delivery in a patient who got pregnant 8 months after her renal transplant and who suffered from chronic hypertension; a preterm IUGR baby was born at 33 weeks GA. Graft failure occurred during pregnancy in a patient with a creatinine value of 7.5 mg/dl during pregnancy, needing dialysis.

Despite its pelvic location, the transplanted kidney rarely produces dystocia and is not injured during vaginal delivery [2], so, cesarean rates of 27–92% (66% in our study) are probably not justified by obstetric necessity alone [11, 13, 24, 25]; the high C-section rate is possibly more related to the need of scheduling delivery or due to fetal/maternal complications.

Conclusion

This series results are in agreement with those of other studies. Even though pregnancy does not seem to adversely affect short-term renal allograft function, as long as the guidelines are strictly followed, risks of obstetric and perinatal complications seem to be increased. We cannot draw a conclusion with regard to the incidence of fetal malformations in relation to certain drug regimen due to the small amount of cases. Further studies of long term graft function and pediatric follow-up are needed. Therefore, it is our opinion that pregnancy, delivery and follow-up should take place in tertiary centers.