Analysis of pronuclear zygote configurations in 459 clinical pregnancies obtained with assisted reproductive technique procedures

One of the most important problems in In Vitro Fertilization (IVF) treatments is the selection of the best embryos for transfer, a crucial point to maintain high performance in terms of pregnancy rate, reducing at the same time the risk of multiple pregnancy. Nowadays, the quality evaluation and the selection of in vitro obtained embryos are made on the basis of morphological parameters involving embryo development, uniformity of blastomeres, percentage of fragmentation, cytoplasmic irregularities, rate of cleavage, blastomeres multinucleation and other visible features [1‐5]. These assessments are non-invasive for the embryo development, but on the other hand can not provide any information about embryonic chromosomal arrangement, one of the most relevant aspects of human reproduction, both in vivo and in vitro. In fact, alterations of chromosomic copy number (aneuploidies) are common in human oocytes and embryos and seem to be mostly implicated in the first-trimester abortions, a complication affecting 50%-70% of all spontaneous conceptions [6]. As described, autosomal trisomies and sex chromosome monosomies, followed by polyploidy and structural rearrangements show a global range from 50% to 80% in first-trimester miscarriages [7, 8].

An important contribution in the evaluation of embryo quality seems to come from the pronuclear and nucleolar characteristics, proposed as an indicator of embryo development and chromosomal complement in human fertilized oocytes [9‐15]. Nucleoli are the sites of the synthesis of pre-RNA, and ribosomal RNA (rRNA) is necessary for the translational process whereby the embryonic genome becomes fully activated [16]. Despite the high number of studies conducted, to date there are conflicting data about the clinical efficacy of zygote-score: in fact, recent data show that scoring system based on pronuclear morphology seems to provide a good criterion to select embryos for transfer when combined with embryo morphology evaluation on Days 2 and 3 [14] other authors concluded that late parameters (such as the cell number and embryo grade) have a better prognostic value than zygote score when selecting embryos for transfer [17‐19].

Finally, to overcome definitely the problem of aneuploidy during IVF, preimplantation genetic diagnosis (PGD) and preimplantation genetic screening (PGS) have been proposed as alternative approaches for embryo selection based on chromosomal arrangement [20], but their efficacy is still debatable [21]. In fact, although several studies have reported an increase in implantation rates and take home baby rates following PGD [22‐25] other have failed to show any positive effect of this technique [26‐30]. Moreover, one report showed that PGS can have a detrimental effect on pregnancy outcomes [31].

Since several studies seem to confirm that aneuploidy in the human oocytes and embryos tend to increase with the advancing maternal age exceeding 50% by the age 40 years old [32, 33], the aim of this study was correlate the pronuclear and nucleolar characteristics with the maternal age and the outcome of pregnancy.

In the present study we analyzed the pronuclear and nucleolar morphology, the polar body alignment and the zygote configuration in 459 clinical pregnancies obtained by IVF (202 clinical pregnancies) and ICSI (257 clinical pregnancies).

As showed in Table 1, we did not found statistically significant differences in terms of IVF and ICSI distribution in our population, retrieved oocytes per cycle, inseminated oocytes, 2 pronucleate zygotes, and implantation rate between patients ≤32 years old, 33-37 years old and 38-41 years old. On the contrary, we found a statistically significant increase of fertilization rate in patients 33-37 years old in comparison to patients ≤ 32 years old (84.0% vs. 78.4%, O.R. 0.692;C.I. 95% 0.487-0.984) and an increase in cleaved and transferred embryos in patients 33-37 years old in comparison to patients ≤32 years old (p < 0.05).

The assessment of pronuclear morphology (Table 2) showed a statistically significant increase in clinical pregnancies in "A" configuration in 38-41 years old patients in comparison to ≤ 32 years old patients (79.4% vs.. 72.0%, O.R. 0.667;C.I. 95% 0.456-0.976) and in comparison to 33-37 years old patients (79.4% vs.. 69.8%, O.R. 0.600;C.I. 95% 0.434-0.828). The same trend was observed in ongoing pregnancies/deliveries (Table 2).

Moreover, the pronuclear morphology analysis showed an opposite trend in "B" configuration (Table 2): in fact, we observed a statistically significant decrease in clinical pregnancies in "B" configuration in 38-41 years old patients in comparison to ≤ 32 years old patients (15.0% vs. 23.0%, O.R. 1.698;C.I. 95% 1.118-2.578) and in comparison to 33-37 years old patients (15.0% vs. 26.0%, O.R. 1.996;C.I. 95% 2.398-2.849). Similarly, we found the same trend in ongoing pregnancies/deliveries (Table 2).

We did not observe any statistically significant differences in "C", "D" and "E" configurations related to maternal age both in clinical pregnancies and in ongoing pregnancies/deliveries (Table 2).

In nucleolar morphology analysis, we found no statistically significant differences in all considered configurations (1, 2, 3, 4) related to maternal age both in clinical pregnancies and in ongoing pregnancies/deliveries (Table 3).

In Table 4 are reported the results of polar body alignment analysis. We found a statistically significant increase of "β" configuration in 33-37 years old patients compared with 38-41 years old patients both in clinical pregnancies (31.1% vs. 37.9%, O.R. 0.740;C.I. 95% 0.554-0.989), and in ongoing pregnancies/deliveries. On the contrary, in "α" and "γ" configurations we found no statistically significant differences related to maternal age in clinical pregnancies and in ongoing pregnancies/deliveries (Table 4).

Moreover, we found no statistically significant differences correlating P1 (A1α, A2β and A3β) and P2 (pool of all the other configurations) configurations with embryo cleavage and embryo quality in clinical pregnancies (Table 5). We observed a similar trend in ongoing pregnancies/deliveries (data not showed). Finally, we found no statistically significant differences in P1 and P2 configurations related to maternal age in clinical pregnancies. Similarly, we found the same result in ongoing pregnancies/deliveries (Figure 3).

One of the main items of IVF treatments is maintain an high performance in terms of pregnancy rate reducing at same time the multiple gestations; the selection of the best embryo/embryos to transfer is a key step to achieve this objective. Nowadays, the embryo selection is mainly performed on the basis of visible morphological parameters [2‐5], but this assessment can not provide any information about embryonic aneuploidies, a problem increasing with the advancing maternal age [32, 33], and probably implicated in the most of the first-trimester spontaneous miscarriages [8]. The evaluation of the pronuclear and nucleolar characteristics (zygote-score) has been proposed as an indicator of embryo development and chromosomal complement in human fertilized oocytes [13‐19]. The study of human zygote seem to provide important information about embryonic chromosomal arrangements, even if, to date, there is no definitive scientific evidence about its clinical efficacy [17, 18]. Recent data seem to consider pronuclear evaluation as a good criterion when combined with embryo morphology evaluation on Days 2 and 3 [14], other authors conclude that late parameters (number of blastomeres and embryo grade) have a better prognostic value than zygote score [19].

In the present study, we have evaluated the clinical significance of zygote-score related to maternal age in patients submitted to ART cycles successfully, obtaining a clinical pregnancy. We observed that all parameters analyzed (pronuclear morphology, nucleolar morphology, polar body alignment and zygote configuration) were generally uniformly distributed in patients ≤32 years old, 33-37 years old and 38-41 years old, showing only few differences related to maternal ages.

Studying the pronuclear morphology, we observed a statistically significant increase in "A" configuration, with a concomitant decrease in "B" configuration, in 38-41 years old in comparison to ≤37 years old patients, both in clinical pregnancies and in ongoing pregnancies/deliveries groups. Due to our unique population characteristics (all patients with clinical pregnancy and maternal age subdivision), it is difficult to compare our results with others present in literature, nevertheless these results appear to corroborate with what has been reported in literature [9]. In agreement with previously reported data [19], our analysis of nucleolar morphology showed no statistically significant differences in relation to maternal ages, both in clinical pregnancy and in ongoing pregnancy/deliveries groups. This result is in contrast with recently published data showing a correlation between pronuclear morphology and maternal ages [15]. We believe that the discrepancy could be related to the different sample size and to the different maternal age between our patients and those analyzed by Maille and coworkers. Moreover, in contrast with Maille and coworkers, we included in the study population clinical pregnancies obtained both with IVF and ICSI. Again, more recent data seem to confirm the poor clinical significance of nucleolar morphology during embryo selection, reporting no correlations between this parameter and implantation rate [19].

Finally, studying the last zygote parameter - polar body alignment - related to maternal age, our data showed a statistically significant decrease of "β" configuration in patients 38-41 years old, both in clinical pregnancies and in ongoing pregnancies/deliveries group. We can speculate that polar body alignment, usually evaluated during zygote-score assessment, not appear to have a prognostic value in terms of embryo viability and pregnancy rate. Than, it can not be considered a good tool to select the best embryo to transfer.

This assumption seems to be confirmed by the evaluation of the last parameter considered in our analysis: the whole zygote configuration (pronuclear and nucleolar morphologies plus polar body alignment). In particular, the total absence of statistically significant differences between zygote configurations P1 and P2 grouped by maternal ages, probably provide an additional evidence about the limited importance of zygote-score in assisted reproduction outcomes, as proposed by our previously data and confirmed by most recently published results [18, 19]. Our results showed any statistical significance in term of embryo cleavage and embryo quality related to P1 or P2 configurations both in clinical pregnancies and in ongoing pregnancies/deliveries.

In conclusion, the results of our study confirm that zygote-score assessment have a limited clinical significance in the choice of the best embryos to transfer during IVF treatments. Probably, as proposed by other authors [19, 40] and suggested in this study, the evaluation of embryo quality performed on the basis of number of blastomeres, embryo morphological characteristics and grade is more predictive than zygote-score.