Introduction

Preeclampsia, occurring in 2–8% of pregnancies, is associated with severe adverse maternal and neonatal outcomes. Among others, preeclampsia is associated with maternal multi-organ failure [1,2,3], progression to eclampsia, a life-threatening event [4], intra-uterine growth restriction [5], and preterm deliveries [6]. Rising evidence further associates preeclampsia with future maternal and offspring health. Epidemiologic studies have shown a possible association between preeclampsia and long-term maternal mortality, subsequent risk of cancer [7] and metabolic morbidity [8], as well as with increased risk of morbidity and mortality for the adult offspring [9, 10]. Despite years of research invested in the study of preeclampsia and many theories that have been proposed, there is still poor understanding of the etiology; thus, efforts have been directed towards identification of populations at risk and towards studying various prophylactic measures.

Prophylactic use of low-dose aspirin is currently being investigated as a promising treatment for prevention of preeclampsia. A recent meta-analysis including five randomized controlled trials concluded that low-dose aspirin administrated at or before 16 weeks of gestation reduces mainly the risk of early preeclampsia [11]. This potential preventive treatment reinforces the importance of identifying specific groups that are at risk for development of preeclampsia.

Oocyte donation (OD), originally offered to women with premature ovarian failure, has now become an integral part of assisted reproductive care. Donated oocytes are currently used by women with many reproductive disorders and commonly by women in later reproductive years [12]. Over the age of 45, OD is the most applicable option for a woman desiring to achieve motherhood. Since women in this age group are at an increased risk of hypertensive complications during pregnancy [13, 14], we aimed to evaluate the risk of preeclampsia in singleton pregnancies achieved by OD as compared to natural pregnancies after the age of 45.

Materials and methods

A historical prospective study of singleton live births of parturients ≥ 45 years of age at four major hospitals in Jerusalem, Israel was conducted. For the purpose of the study, the population was divided by the mode of conception into two groups: OD and natural conception. A comparison ratio of 1:2 was used between the study group and the control group. Women who conceived by in-vitro fertilization (IVF) without OD (i.e., autologous treatment) were excluded from the study. Data were collected during the years 1995–2012. The oocyte donors ranged in age from 19 to 33 years. All donors were examined by a physician and were proven to be generally healthy, and with no ongoing medical treatment or disability.

Data on demographic and obstetric characteristics, as well as neonatal data, were retrieved from the delivery room management software. Data concerning fertility treatments were further confirmed through the infertility outpatient clinic of each participating hospital. The study protocol was approved on January 1st 2013 by the institutional review board, approval number 0064-13-HMO.

Clinical characteristics that were collected and analyzed included maternal age, gravidity and parity, mode of conception, antepartum course, and neonatal weight.

The main outcome variable in this study was preeclampsia. The main exposure variable was oocyte donation pregnancies. Secondary outcomes included pregnancy-induced hypertension and Small for Gestational Age (SGA).

Diagnosis of preeclampsia and pregnancy-induced hypertension was made according to the criteria of the International Society for the Study of Hypertension in Pregnancy [15]. Pregnancy-induced hypertension was defined by systolic blood pressure ≥ 140 mmHg and/or diastolic blood pressure ≥ 90 mmHg on at least two occasions, 4 h apart, developing after 20 weeks’ gestation in previously normotensive women in the absence of proteinuria. Preeclampsia was defined by systolic blood pressure ≥ 140 mmHg and/or diastolic blood pressure ≥ 90 mmHg with proteinuria of ≥ 300 mg in 24 h, or two readings of at least ++ on dipstick analysis of midstream or catheter urine specimens if no 24-h collection was available. Birth weight percentiles were determined from the gender-specific charts of Poon et al. [16] and SGA was defined as a birth weight < 10th percentile.

Statistical analysis

Women who received OD and delivered were compared to those who conceived naturally. Univariate comparisons used the t test or the χ2 test, as appropriate. Multivariate logistic regression models were performed to examine the association between the use of OD and preeclampsia, pregnancy-induced hypertension, and SGA, controlling for possible confounders: maternal age and parity. A P value < 0.05 was considered statistically significant. We report odds ratios (OR), 95% confidence interval, and two-sided P values. The statistical software package SPSS 23.0 (SPSS Inc., Chicago, IL, USA) was used for all data analyses.

Results

During the study period, 135 women ≥ 45 years of age conceived by OD and carried and delivered a singleton pregnancy. As mentioned earlier, a 1:2 comparison ratio was used to compare these conceptions to randomly chosen 270 women who conceived spontaneously during this period and had their most recent delivery after the age of 45 years.

Data on demographic and obstetric characteristics, as well as information regarding background diseases, are presented in Table 1. The mean maternal age of oocyte recipients was 2 years older than natural conception controls (P < 0.01). Mean gravidity and parity for the OD and natural conception groups were 2.6 vs. 11.1 and 0.5 vs. 8.2, respectively. Whereas only three of the women in the natural conceptions group were nulliparous, 68.9% of women in the OD group were nulliparous. OD conceptions were more likely to have gestational diabetes mellitus than natural conceptions. There was no significant difference between the groups in rates of chronic hypertension.

Table 1 Maternal demographic characteristics of study population
Full size table

Three out of 270 (1.1%) naturally conceived pregnancies and 17 out of 135 (12.6%) OD pregnancies were complicated with preeclampsia. After controlling for parity and age at delivery, OD pregnancies were found to be associated with a 12-fold increased risk for preeclampsia (95% CI 2.69–52.73, P < 0.001) (Table 2). Controlling also for chronic hypertension did not change the results. Since primiparity is associated with an increased risk of preeclampsia by itself, we performed an additional risk analysis within the group of the OD pregnancies, comparing between primiparous and multiparous women. The rate of preeclampsia in nulliparous and multiparous pregnancies following OD was 16.7 and 10.8%, respectively (P = 0.40), and remained non-significant also after controlling for age. In addition, we found no significant risk of preeclampsia related to the maternal age in this group after the age of 45 years (data not shown; age analyzed as a continuous variable or dichotomized by the median).

Table 2 Study outcomes by mode of conception
Full size table

While there were no cases of early preeclampsia (< 34 gestation weeks) in the natural conception group, there were five cases of early preeclampsia in the OD group. Rates of pregnancy-induced hypertension of naturally conceived and oocyte donation pregnancies were 5.5 and 9.6%, respectively (P = 0.22). This difference remained non-significant also after controlling for age at delivery and parity.

Mean gestational week at delivery was 37.4 weeks (range 25-42) for the OD group and 39.4 weeks (range 32–43) for the natural conception group. Data regarding mode of delivery was available for 88.1% of patients. Rates of caesarean section for the OD and the naturally conception groups were 58.5 and 21.9%, respectively.

The rate of SGA (birth weight under 10th percentile for the specific gestational age) was higher for the OD group as compared to natural conception pregnancies (23.7 vs. 13.3%). However, a multivariate analysis, controlling for age, parity, hypertensive disorders, gestational and non-gestational diabetes mellitus, and gestational age, revealed a non-significant association.

Discussion

The present study shows that after the age of 45, OD pregnancies are associated with a 12-fold increased risk of preeclampsia as compared to pregnancies following natural conception. Previous studies reported an elevated risk of various obstetrical adverse outcomes, including preeclampsia, for women conceiving by OD [17]. However, in most cases, the risk was assessed as compared to autologous oocyte in-vitro fertilization (IVF) pregnancies rather than to natural conception pregnancies, and multiple pregnancies were not excluded.

Our study and control groups are unique in their composition. The study group is unique in the fact that it includes multiparous mothers who conceived via OD. Women conceiving naturally at extremely advanced maternal age characterize the control group. This group of women encourages natural conception, discourages contraceptive use, and views fecundity as a blessing; therefore, they continue to challenge their reproductive system until menopause. Those unique populations enabled us to isolate the effect of the donated oocyte.

Natural conceptions were chosen in our study as the control group due to their biological neutrality. The main drawback of choosing natural conceptions as the reference group is the possible selection bias due to their medical background that enabled them to extend their fertility up to this age. Choosing IVF autologous pregnancies as the reference group also has its limitations, including their medical background, the hormones used for ovarian stimulation and the fact that in our country, IVF is not allowed after the age of 45. Hence, we sought to choose natural conceptions as the control group and controlling for possible confounders.

We excluded multiple gestation pregnancies from our study. Multiple gestations are related to distinct maternal physiologic adaptation mechanisms as well as to increased risk for specific pregnancy complications, particularly preeclampsia [14]. Including multiple gestations, even if controlling for them, can bias the results and they were, therefore, excluded.

Previous studies assessing adverse obstetric outcomes in elderly parturients reported rates of preeclampsia of 2.7–28.4% (Table 3) [13, 18,19,20,21,22,23,24,25,26,27,28,29,30]. Of note, as the mother’s age rose, the rate of preeclampsia increased. However, most of these studies included multiple gestations and did not assess different modes of conception.

Table 3 Summary of the previous studies of preeclampsia in advanced aged gravidas
Full size table

Studies assessing adverse obstetric outcomes in pregnancies achieved by OD demonstrated a high risk of preeclampsia in this group (Table 4) [17, 31,32,33,34]. Yet, these studies compared the results to pregnancies achieved by IVF and did not limit their analysis to a specific range of age or to singleton pregnancies.

Table 4 Summary of the previous studies of preeclampsia in pregnancies following oocyte donation
Full size table

The pathophysiology of preeclampsia likely involves both maternal and fetal/placental factors. Some researchers have postulated shallow trophoblast invasion and defective placental vasculature early in pregnancy, resulting in placental underperfusion, hypoxia, and ischemia, which then lead to release of antiangiogenic factors into the maternal circulation that alter maternal systemic endothelial function, leading to hypertension and other multisystemic manifestations of the disease [35, 36].

In addition, it is speculated that immunologic factors may contribute substantially to the underlying defective placentation in terms of rejection of the fetoplacental unit as an allograft or aberrant immunological response directed against tissue of organ-specific antigens associated with the placenta [37, 38]. This hypothesis is supported by higher rates of preeclampsia in cases in which there was no prior exposure to paternal/fetal antigens, such as in nulliparous gestations [14] and in conceptions following sperm donation [39]. Based on a similar rational process, in OD pregnancies, the foreign oocyte may play a role in the immunologic cascade and contribute to an abnormal placentation, thus accounting for the markedly higher rate of preeclampsia in oocyte donation pregnancies, as compared to conceptions with an autologous oocyte. Our study may support this hypothesis, since trophoblasts of pregnancies following OD carry 100% “foreign” tissue for the mother host as compared with only 50% in autologous oocyte IVF and naturally conceived pregnancies. Our findings of non-significant, different rates of preeclampsia among different parity and age groups of OD parturients lend further support to this theory, predisposing for significantly higher rates of preeclampsia when conception is through a foreign oocyte.

In our study, we found no significant association between the mode of conception and SGA. Prior studies have shown an association between oocyte donation and birth weight [31]. However, this may be due to a definition bias, since the outcome reported in some studies was low birth weight (≤ 2500 g), rather than the specific birth weight percentile in accordance with the gestational age. We decided not to use low birth weight as it does not reflect any biological mechanism due to its disconnection to gestational age. Our findings of a significant association of oocyte donation conceptions with preeclampsia but not with SGA may reflect the dominance of the immunologic preeclampsia pathway, rather the vascular theories.

In a recent meta-analysis, a comparison between elective delivery and expectant management for preeclampsia was presented. For women with preeclampsia beyond 34 weeks, gestation elective delivery was shown to decrease the incidence of complications, severe hypertension and need for antihypertensive medical therapy. Before 34 weeks gestation, elective delivery was shown to decrease the risk of placental abruption in women suffering from severe preeclampsia, but this reduction may come at the cost of increased risk of neonatal complications [40].

The role of the male partner in preeclampsia and SGA is yet to be fully understood [41]. We did not have access to data regarding the male partner and whether all pregnancies were achieved with the same partner or not. However, the control group is comprised mainly of multiparous women who are likely to have a more religious background. In those communities, the divorce rate is extremely low, and hence, it can be assumed that most of those women had conceived by the same male partner. We were not able to retrieve data regarding Body Mass Index and gestational weight gain. We recognize this as a limitation of our study. However, as the risk of preeclampsia in the OD group was increased 12-fold as compared to the natural conception group, it is unlikely that this data would have had a significant impact on the results.

OD is an effective method of achieving pregnancy for women of advanced aged who seek motherhood. In the United States, from 2000 to 2010, there was an increase in the number of donor oocyte cycles from 10,801 to 18,306 per year [42]. This significant growth should encourage clinicians to become familiar with the specific risk factors of this unique population.

During the last decade, guidelines on routine antenatal care were published by national and international bodies. Some of those have recommended that a woman’s risk for preeclampsia should be evaluated early in pregnancy. There are now certain guidelines that advocate in favor of daily low-dose aspirin for women who are at high risk [43,44,45]. One of the new ways in which clinicians address antenatal care consists of an integrated approach in which at 11–13 weeks, maternal characteristics and history, along with biophysical and biochemical markers, are used to evaluate the risk of developing various complications later in pregnancy [46]. In the context of preeclampsia, the main goal of such a combined approach would be to detect the patients who could benefit from prophylactic interventions.

Our study showing a significant increased risk for preeclampsia in this group of women may encourage clinicians to implement this approach and perform a first-trimester screening test for preeclampsia as part of the routine follow-up in those pregnancies.