| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
EXTENSIVE CLINICAL EXPERIENCE |
Instituto Valenciano de Infertilidad (S.R.S., C.T., E.B., V.S., C.S., J.R., A.P.), University of Valencia, 46015, Valencia, Spain; and Hospital Universitario Dr. Peset (A.P.), 46017, Valencia, Spain
Address all correspondence and requests for reprints to: Antonio Pellicer, Instituto Valenciano de Infertilidad, Plaza de la Policía Local, 3, 46015, Valencia, Spain. E-mail: apellicer{at}ivi.es.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Setting: The present work was developed in a private infertility clinic.
Patients or Other Participants: We retrospectively evaluate the results of 3089 OD cycles that ended up in a d 3 embryo transfer. Severe male factor infertility was an exclusion criterion.
Main Outcome Measures: The impact of patients’ age on pregnancy, implantation, and miscarriage rates and obstetric outcome is analyzed, as is the relevance of endometrial thickness, serum estradiol levels, and duration of exogenous estrogen therapy to said rates.
Results: Pregnancy and implantation rates are significantly reduced and miscarriage rate is significantly increased from 45 yr of age onward. Concerning obstetric outcome, incidences of hypertension, proteinuria, premature rupture of membranes, second- and third-trimester hemorrhage, and preterm delivery are higher and mean birth weight is lower in this age group. With regard to endometrial preparation, estrogen therapy lasting more than 7 wk is associated with reduced PR and IR (P = 0.01 and P = 0.02, respectively).
Conclusions: The results of OD cycles and obstetric outcome are significantly worse when recipients are 45 yr of age or older. Concerning endometrial preparation, results are significantly worse when estrogen therapy lasts more than 7 wk.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Oocyte donors
All donors were included in our OD program after being thoroughly informed and having fulfilled our inclusion criteria. Subjects were aged between 18 and 35 yr old, and we had access to their complete medical history, which considered current or past exposure to radiation or hazardous chemical substances, iv drug use, and reproductive history. All subjects were shown to be normal in a physical and gynecological examination, had no family history of hereditary or chromosomal diseases, had a normal karyotype, and tested negative in a screening for sexually transmitted diseases. An economic compensation stipulated by the Spanish government is ascribed to the oocyte donor with the aim to mitigate the physical burden, job restrictions, and the number of times the donor must come to the clinic during an ovarian stimulation cycle.
All patients underwent a long protocol of down-regulation with daily doses of a GnRH agonist (GnRH-a). Transvaginal ultrasound (US) was performed to ascertain ovarian quiescence on the first 3 d of menses, and controlled ovarian stimulation was then initiated. In short, the starting dose varied from 150 to 300 U/d of FSH and/or human menopausal gonadotropin for the first 2–5 d, according to age, body mass index, and response to previous ovarian stimulations. The dose was then adjusted according to ovarian response, which was monitored through serum E2 levels and ultrasound every 2–3 d. Stimulation was performed until leading follicles had a mean diameter of more than 18 mm. Human chorionic gonadotrophin was then administered, and ovarian puncture was performed an additional 36 h later.
Anonymous donors were matched with their recipient(s) according to phenotype and blood group.
Oocyte recipients
Oocyte recipients entered our OD program due to one of the following diagnosis: premature ovarian failure/menopause, failure to achieve pregnancy after at least three cycles of assisted reproduction techniques, genetic or chromosomal disorders, low response to controlled ovarian hyperstimulation, or recurrent miscarriage.
All oocyte recipients were undergoing hormone replacement therapy (HRT), as described previously (15). In patients with ovarian function, a depot GnRH-a was administered in the midluteal phase of their cycle. HRT was initiated on d 1–3 of the following cycle, and doses of estradiol valerate (EV) (Progynova; Schering Spain, Madrid, Spain) were increased as follows: 2 mg/d for the first 8 d of treatment, 4 mg/d for the following 3 d, and 6 mg/d until a pregnancy test was performed after embryo transfer or until vaginal spotting (before embryo transfer) was identified, in which case the cycle would have been cancelled and would not have been included in the study. On the 15th or 16th days of HRT, a transvaginal ultrasound was performed to measure endometrial thickness, and serum E2 levels were tested. Recipients without ovarian function were submitted to the same endometrial preparation protocol, with the exception of the administration of the depot GnRH-a. Our OD program differs from those of other centers in that the recipients’ endometrial state is maintained for a number of weeks (up to 10, with a mean of 5.5) due to the continued administration of EV, not having been ascribed a specific oocyte donor at the beginning of endometrial preparation. This approach allows us to significantly shorten the interval between the recipients’ first consultation and embryo transfer (mean duration < 12 wk). As a result of this strategy, it has become very important for us to ascertain whether or not the duration of EV therapy, endometrial thickness, and serum E2 levels affect the outcome of cycles.
Micronized progesterone (P) (800 mg/d, vaginally) (Progeffik; Effik Laboratories, Madrid, Spain) was initiated the day after OD, and embryos were transferred on d 3 of embryo cleavage.
Embryos were classified according to cell number, symmetry, and degree of fragmentation (16). Serum ß-human chorionic gonadotropin was measured 14 d after oocyte retrieval. Clinical pregnancy was confirmed 1–2 wk later if the existence of a gestational sac was confirmed by transvaginal US. PR is the percentage of patients undergoing embryo transfer that were shown to have one or more gestational sacs on US evaluation. IR was obtained by dividing the number of gestational sacs seen in US by the number of replaced embryos. The number of embryos to be replaced was decided based on the following criteria: embryo quality, patient’s age, the outcomes of previous assisted reproduction treatments, reproductive history, and the presence of uterine malformations. MR was defined as the percentage of pregnancies that terminated before the completion of the 20th week of gestation after previous US detection of the embryo’s heartbeat.
Complete pregnancy follow-up of patients treated in our clinic was performed by their private obstetricians in the majority of cases. We documented the obstetric outcomes of singleton pregnancies followed up by us based on incidence of preterm delivery, cesarean delivery, birth weight, hypertension, proteinuria, gestational diabetes, obstetric cholestasis, second and third trimester hemorrhage, preterm rupture of membranes, and fetal anomaly.
Statistical analysis
The effect of age on PR, IR, and MR was evaluated by means of univariate regression analysis (PR and MR) and linear regression analysis (IR). In the case of no significant association being established, the nonparametric 
2 test was used to compare rates in four age groups (women under 40, 40–44 yr olds, 45–49 yr olds, and women aged 50 or more). The mean number of replaced embryos, mean number of good-quality embryos replaced per patient in each group, and duration of E2 therapy in each age group were compared by means of the ANOVA. The Bonferroni correction was applied to multiple comparisons.
The incidences of the aforementioned obstetric outcomes were compared in the different age groups using the 2 test. In the case of birth weight, mean values were compared by using the Student’s t test.
A multivariate analysis was performed to assess the possible impact of the following parameters on PR, IR, and MR: duration of EV therapy for endometrial preparation; endometrial thickness and serum E2 levels (both evaluated on d 15–16 of HRT); patient age; and number of good-quality embryos replaced (either isolated or in correlation with each other). In all cases, significance was considered to be P < 0.05.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
). Mean ± SD number of replaced embryos was 2.8 ± 0.7. Global PR, IR, and MR were 49.0, 20.2, and 17.6%, respectively.
|
The univariate and linear regression analyses of the association between age and PR, IR, and MR showed that age was not significantly relevant to any of these outcomes (r2 = 0.001 in all three cases; P was nonsignificant).
Comparisons of PR, IR, and MR in the four age groups showed a statistically significant heterogeneity in IR and MR distribution (P = 0.01 and P = 0.02, respectively) (Table 2). Although multiple comparisons (using the Bonferroni correction) of each pair of age groups did not detect statistically significant differences, PR and IR were shown to gradually decrease from 45 yr of age onward, whereas MR gradually increased with age. Patients were then divided into two age groups based on the cutoff point of 45 yr of age, and the cycle outcome of the two groups was compared by means of the 2 test: PR and IR were significantly lower, and MR was significantly higher in older women (P = 0.045, P = 0.02, and P = 0.03, respectively) (Table 3). The mean number of replaced embryos, mean number of good-quality embryos transferred, and number of days of E2 therapy were virtually the same in each of the four age groups and in the two groups under or over 45 (Tables 2 and 3, respectively).
|
|
Obstetric outcome was also compared in the two age groups defined in Table 3
(<45 yr vs. 
45 yr). Older recipients showed a significantly increased frequency of preterm delivery and lower mean gestational age and weight at delivery (P < 0.0001, P = 0.006, and P = 0.01, respectively). There was a significantly higher incidence of hypertension and proteinuria, second and third trimester hemorrhage, and premature rupture of membranes in this age group (Table 4).
|
Multivariate analysis did not detect statistically significant associations between recipients’ age, serum E2 levels, or endometrial thickness and PR, IR, or MR. Conversely, duration of E2 therapy was significantly related to PR and IR, and the number of good-quality embryos replaced was significantly associated with PR (Table 5).
|
). The 2 test comparison of PR and IR in the 3089 cycles when separated into two groups along this cutoff point (7 wk of E2 oral therapy) showed that both were significantly reduced (P = 0.01 and P = 0.02, respectively) when therapy continued for longer (Fig. 1).
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Age vs. PR, IR, and MR
The logistic and linear regression analysis failed to demonstrate a statistically significant global relationship between age and PR, IR, or MR. Comparisons of the four age groups defined showed that PR, IR, and MR were poorer from 45 yr of age onwards, but statistical analysis merely confirmed that IR and MR were not evenly distributed (Table 2). Comparisons of women aged 44 or less with those aged 45 or more demonstrated that PR, IR, and MR really were significantly worse among the older women (Table 3). Notwithstanding, these patients showed a highly acceptable PR (44.4%).
In 2002, the multivariate retrospective evaluation of the experience of OD cycles of the United States drew conclusions very similar to our own with respect to cycle prognosis based on recipient age: PR and IR did not vary until the "late 40s" and fell significantly from this time on (even more clearly from 50 yr onward). Also, the increase in MR was close to significant (17). This report enjoyed the advantage of an enormous number of cases (>17,000) and the obvious disadvantage of any analysis of data originating in many different centers: in other words, a lack of control over the protocols used and documentation of information. Interestingly, similar conclusions have been reached using varying methodological approaches.
The influence of age on the results of OD cycles has been described previously (10, 11, 18). In one series, significantly decreased PR was reported but without any reference to IR (undoubtedly the most accurate variable in the evaluation of uterine receptiveness) (10). In another, lower IRs were observed in older subjects, but infertile or aged (up to 40 yr) women were accepted as oocyte donors and the quality of replaced embryos was not ascertained (11), which raises doubts about ascribing cycle results exclusively to uterine factor. Nonsignificantly (10) and a significantly (18) increased MR has been described in women of advanced age, a MR close to 50% having been observed in the latter study (18). The P dose prescribed in that case was 100 mg/d im, much lower than the daily dose we have been using in recent years (800 mg vaginally). Although the need of supraphysiological doses of P to achieve adequate endometrial luteinization has been described in older women (19, 20), we do not believe that the P supplementation protocol used was responsible for the impressive MR observed. This could be a chance finding due to small sample size. However, if age really does determine endometrial histological and ultrastructural changes that affect the implantation process, it is logical to assume that compromised implantation is associated with increased MR. The statistical confirmation of this, however, demands a large group of older women, which possibly explains why an increased MR observed with advancing age did not reach statistical significance in some series (10, 12).
Some studies have failed to detect any impact of age on OD cycle results (12, 13, 14). None of said studies explored the association between age and IR. Similar PR, cumulative PR, and delivery rates were reported in different age groups. Small sample size with a low statistical power and the cutoff age applied may have contributed to the results obtained.
The effect of age on endometrial receptiveness has also been evaluated by studies in which oocytes from the same donor were shared by recipients from different age groups. The advantage of this methodology is the homogeneity of oocyte quality in both groups. The disadvantage is the small sample size. In this framework, results have also been conflicting. Markedly reduced PR and IR have been reported in older women (47.3 vs. 24.5 and 24.5 vs. 14.9%, respectively) in a series in which mean ± SD age in each group was not determined (21). It would be of interest to identify the age difference that led to such an impressive discrepancy in PR and IR. Once again, the quality of replaced embryos was not a controlled variable. Another group performed a very similar study, and no difference was observed in PR, IR, and MR between groups (22). Mean ± SD age in each group was 31.8 ± 3.8 and 45.5 ± 3.2 yr, respectively, and mean number of embryos transferred was identical in both groups, with no reference to embryo quality.
Age vs. obstetric outcome
Pregnancies obtained with OD have long been reported to be associated with increased frequencies of negative obstetric outcomes (such as hypertension and cesarean section) and neonatal complications (preterm delivery and low birth weight) (reviewed by Söderström-Anttila in Ref. 23). Factors believed to be responsible for these findings are an increased incidence of multiple pregnancies, advanced age in patients, and immune disorders specific to OD gestations. The particular role of each of these factors may be difficult to establish. Here, we have observed in singleton pregnancies that women aged 45 or more experience increased frequencies of some negative endpoints (Table 4). Preterm birth was significantly more common in this age group, which determined lower mean gestational age and weight at birth. Factors that may have influenced this neonatal outcome are hypertension and proteinuria, premature rupture of membranes, and second and third trimester hemorrhage. Age has already been reported as an independent risk factor for these complications and others, such as gestational diabetes (not significantly increased in our group of older pregnant women, perhaps because of its size) and thrombophlebitis (24, 25). Publications on this issue differ with regard to the age at which obstetric risks increase but generally specify some point between 40 and 50 yr. Because our older group was almost entirely composed of women between 45 and 50 yr old (10 of 12 pregnancies), we would argue that an effect is clearly observed before the age of 50. Trophoblastic invasion, uterine blood flow, and/or compliance are certainly somewhat compromised in these patients. Obviously, the functioning of other organs and systems (i.e. renal function and glucose metabolism) is also relevant to obstetric outcome.
The multivariate analysis
Recently, cycles of donations of oocytes that were shared by recipients who experienced discordant outcomes (pregnant and nonpregnant women) were analyzed so that factors that could have influenced outcome could be compared between both groups (26). No difference was found between groups in terms of recipients’ age, serum E2 levels, and endometrial thickness on d 15–16 of estrogen therapy, number and quality of embryos transferred, and indications for OD.
The results of this study raise once more the issue of the predictive value of parameters habitually used to ascertain endometrial receptiveness. The multivariate regression analyses we performed allowed us to simultaneously evaluate the manner in which patient age, endometrial thickness, serum E2, duration of estrogen therapy for endometrial preparation, and the number of good-quality embryos replaced (independent variables) affect PR, IR, and MR (dependent variables). The results of these analyses demonstrated that the number of days of E2 therapy correlated significantly with PR and IR and that PR was also considerably affected by the number of good-quality embryos replaced.
Embryo quality has been described as having a linear relationship with IR (27). As expected, the number of good-quality embryos replaced also correlates linearly with PR (28). Because in our study this variable was only included in the multivariate analysis to monitor its interference with endometrial preparation variables, we have not explored further the study of its relationship with PR.
The need for flexibility in the duration of estrogen treatment to coordinate donor and recipient preparations has raised the question of the possible impact of duration of estrogen therapy on success rates in OD cycles. As with almost all other aspects of endometrial preparation for OD cycles, there is a lack of unanimity between the results of studies evaluating said factor. Some authors report a significant drop in PR after 35 (29), 19 (30), or even 11 d (31) of E2 administration, whereas others have not detected any association between the duration of this therapy and cycle outcome (15, 32). Obviously, long periods of E2 use are associated with a higher risk of spotting, a circumstance that requires cycles to be cancelled. Nevertheless, our data strongly indicate the following: 1) PR and IR do not fall until E2 has been performed for 6 wk; and 2) even in the absence of spotting, E2 administration for 7 wk or longer is consistently associated with a diminished success rate. In the present series, it is important to point out that no recipient received treatment with E2 for less than 10 d. We deliberately avoid such short E2 administration because of the increased risk of miscarriage with which it is associated (32, 33).
Endometrial thickness has long been considered an important prognostic factor for embryo transfer in OD cycles as a correlation between endometrial thickness (measured by scan) and cycle result has been reported (14, 34, 35). In some centers, P supplementation is not initiated if the recipient’s endometrial line is thinner than 9 mm or thicker than 12 mm (21). Our data do not support such a significant predictive value for this parameter. No specific endometrial thickness determined a significant change in cycle prognosis. Even a thin endometrium (<7 mm) had good PR and IR, without an increased MR. These findings confirm data published previously by ourselves and other authors (36, 37).
Theoretically, serum E2 levels throughout endometrial preparation for OD should ideally be as close as possible to those seen in natural cycles. Indeed, that was the parameter adhered to in the first attempts at OD (38). However, this approach dictated a complicated protocol, with "ups and downs" in E2 daily dose. With the goal of simplifying treatments, protocols of progressively increasing doses and the maintenance of the maximum dose achieved were proposed. Naturally, concerns were raised about the effect on cycles of such an "artificial" pattern of E2 use. Our data demonstrate that neither supraphysiological nor low (<100 pg/ml) serum E2 levels have a significant impact on OD results. This finding confirms previously published data (14, 31).
In conclusion, there is no significant linear relationship between oocyte recipients’ age and PR, IR, or MR. PR and IR are reduced and MR increased from 45 yr of age onward. Obstetric outcome also deteriorates in women of this age: incidences of hypertension, proteinuria, premature rupture of membranes, second and third trimester hemorrhage, and preterm delivery are higher and mean birth weight is lower. The confirmation of the loss of uterine receptiveness with age is of importance and should be taken into account not only in the context of OD but also in the emerging field of ovarian tissue cryopreservation in delayed childbearing. With respect to endometrial preparation, serum E2 levels and endometrial thickness have no significant influence on OD cycle results, but estrogen therapy of more than 7 wk is associated with reduced PR and IR.
|
Footnotes |
|---|
Abbreviations: E2, Estradiol; EV, estradiol valerate; GnRH-a, GnRH agonist; HRT, hormone replacement therapy; IR, implantation rate; MR, miscarriage rate; OD, oocyte donation; P, progesterone; PR, pregnancy rate; US, ultrasound.
Received November 16, 2004.
Accepted March 23, 2005.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  A.L. Waylen, M. Metwally, G.L. Jones, A.J. Wilkinson, and W.L. Ledger Effects of cigarette smoking upon clinical outcomes of assisted reproduction: a meta-analysis Hum. Reprod. Update, January 1, 2009; 15(1): 31 - 44. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. A. Horcajadas, E. Goyri, M. A. Higon, J. A. Martinez-Conejero, P. Gambadauro, G. Garcia, M. Meseguer, C. Simon, and A. Pellicer Endometrial Receptivity and Implantation Are Not Affected by the Presence of Uterine Intramural Leiomyomas: A Clinical and Functional Genomics Analysis J. Clin. Endocrinol. Metab., September 1, 2008; 93(9): 3490 - 3498. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Vidal, J. Roa, L. Pinilla, A. Pellicer, and M. Tena-Sempere Maternal serum ghrelin levels in early IVF pregnancies: lack of prognostic value for viable pregnancy and altered post-prandial responses Hum. Reprod., April 1, 2008; 23(4): 958 - 963. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. Vernaeve, D. Bodri, M. Colodron, R. Vidal, M. Durban, and O. Coll Endometrial receptivity after oocyte donation in recipients with a history of chemotherapy and/or radiotherapy Hum. Reprod., November 1, 2007; 22(11): 2863 - 2867. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Alvarez, L. Marti-Bonmati, E. Novella-Maestre, R. Sanz, R. Gomez, M. Fernandez-Sanchez, C. Simon, and A. Pellicer Dopamine Agonist Cabergoline Reduces Hemoconcentration and Ascites in Hyperstimulated Women Undergoing Assisted Reproduction J. Clin. Endocrinol. Metab., August 1, 2007; 92(8): 2931 - 2937. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Pau, I. Alonso-Muriel, R. Gomez, E. Novella, A. Ruiz, J. A. Garcia-Velasco, C. Simon, and A. Pellicer Plasma levels of soluble vascular endothelial growth factor receptor-1 may determine the onset of early and late ovarian hyperstimulation syndrome Hum. Reprod., June 1, 2006; 21(6): 1453 - 1460. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M.A.B. Melo, M. Meseguer, N. Garrido, E. Bosch, A. Pellicer, and J. Remohi The significance of premature luteinization in an oocyte-donation programme Hum. Reprod., June 1, 2006; 21(6): 1503 - 1507. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
