Arch Gynecol Obstet DOI 10.1007/s00404-014-3332-3
GYNECOLOGIC ENDOCRINOLOGY AND REPRODUCTIVE MEDICINE
Predictive value of anti-mu¨llerian hormone, follicle-stimulating hormone and antral follicle count on the outcome of ovarian stimulation in women following GnRH-antagonist protocol for IVF/ET E. Tsakos • A. Tolikas • Angelos Daniilidis B. Asimakopoulos
•
Received: 25 February 2014 / Accepted: 17 June 2014 Ó Springer-Verlag Berlin Heidelberg 2014
Abstract Aim To compare the efficacy of three methods: the levels of anti-mu¨llerian hormone (AMH), the levels of follicle stimulating hormone (FSH) and the antral follicle count (AFC), for the prediction of the number of retrieved mature oocytes and the number of generated embryos by intracytoplasmic sperm injection (ICSI) in women stimulated with a GnRH-antagonist protocol. Materials and methods 105 women were enrolled in the study. At the day 2 of a preceding cycle, AFC was performed and FSH and AMH were measured in serum by immunoenzymatic assays. All women were stimulated with a GnRH-antagonist protocol and ovulation was induced with human chorionic gonadotropin. ICSI was performed in all retrieved mature oocytes. Embryo transfers were performed at days 2–3. According to the oocytes retrieved, patients were categorized as poor (\4), normal (4–12) or high responders ([12). Results AFC and the levels of baseline FSH and AMH were significantly different among poor, normal and high responders. The number of oocytes as well as the number of embryos was negatively correlated with baseline FSH and positively correlated with baseline AMH and AFC, whereas AFC showed the strongest correlation. Stepwise
E. Tsakos A. Tolikas B. Asimakopoulos Embryoclinic, Thessalonı´ki, Greece A. Daniilidis (&) School of Medicine, Aristotle University of Thessaloniki, Greece, 9 Smirnis, Evosmos, 56224 Thessalonı´ki, Greece e-mail: [email protected] B. Asimakopoulos Laboratory of Physiology, School of Medicine, Democritus University of Thrace, Alexandroupolis, Greece
regression analysis indicated AFC and baseline AMH as the most significant parameters for the prediction of the number of oocytes; for the prediction of the number of embryos, the most significant parameter was AFC. Conclusions AFC, baseline AMH and baseline FSH are good predictors for the outcome of ovarian stimulation in GnRH-antagonist cycles. However, AFC appears to have the best predictive value. Keywords Antral follicle count FSH Anti-mu¨llerian hormone Ovarian response GnRH antagonist
Introduction Fertility potential of women is dependent on the quantity and the quality of the ovarian reserve that declines with age. The estimation of the ovarian reserve is of crucial importance for the fertility potential of women. It is one of the prerequisites for women undergoing in vitro fertilization (IVF) treatment. The ovarian reserve constitutes by the size of the follicular pool that is established during fetal life. Around the 16th week of fetal life approximately seven millions of follicles exist in both ovaries. Each primordial follicle consists of an oocyte surrounded by somatic cells. At birth, around one million of follicles are present and at menarche only 300–400,000 remain [1]. During the female reproductive life, waves of follicles leave the primordial pool to undergo further development. The last includes proliferation and differentiation of somatic cells, increase of the follicular size, completion of the first meiotic division and introduction to the second meiotic division of the oocyte that stops at the stage of metaphase. The vast majority of follicles fail to complete development undergoing atresia.
123
Arch Gynecol Obstet
In each menstrual cycle, usually only one follicle is selected to become the dominant one which completes the developmental cycle and thus resulting to ovulation. As this process continues throughout reproductive life, the follicle pool is finally exhausted leading to menopause [1]. This is the classical view of a finite follicle pool. Although it has been challenged by recent findings that, at least in mouse, germline stem cells maintain oogenesis throughout adulthood [2], the gradual depletion of the follicle pool during female reproductive life is well established. Various methods of assessing the ovarian reserve have been used in clinical practice [3]. Among them, the measurement of baseline follicle stimulating hormone (FSH) levels, the assessment of the number of antral follicles (AFC) by ultrasonography and the estimation of antimu¨llerian hormone (AMH) levels are probably the more frequently used worldwide. FSH secreted by the gonadotrope cells of anterior pituitary stimulates ovarian granulosa cells to express aromatase. Thus, granulosa cells of early antral follicles convert androgens to estrogens. In turn, estrogens decrease the secretion of FSH by gonadotrope pituitary cells via a negative feedback system. As the follicle pool declines, estradiol (E2) blood levels decline too and consequently blood FSH levels rise [1]. The AFC by ultrasonography at the early follicular phase reliably reflects the ovarian reserve [3]. AMH is a dimeric glycoprotein first studied for its role in differentiation of gonads. It belongs to the TGFb family and in females during adulthood is produced by granulosa cells of primary and secondary follicles which are not yet influenced by FSH. Its production decreases in the larger follicles and also decreases as the follicular pool diminishes [4]. The ovarian reserve is a determinative factor for the number of retrieved oocytes in a stimulated cycle for IVF. Moreover, the quality of the ovarian reserve also influences the quality of retrieved oocytes and therefore the number and quality of embryos generated in IVF. This study was designed to find out the best predictor of the number of retrieved oocytes and the number of embryos generated by intracytoplasmic sperm injection in women undergoing ovarian stimulation with the GnRHantagonist multiple dose protocol. The parameters that were evaluated for their predictive value were serum FSH at day 2 (baseline FSH), AFC before stimulation, AMH serum levels at days 2 and 5 and AMH intrafollicular levels.
Materials and methods This study included prospectively 105 women attending our intracytoplasmic sperm injection (ICSI) program from October 2009 to January 2013.
123
The inclusion criteria were (1) age 25–45 years, (2) regular menstruation, (3) the absence of serious endocrinological disorders, (4) the presence of both ovaries, and (5) BMI 19–30. Approval was obtained from the Bioethics Committee of the Aristotle’s University of Thessaloniki and all patients signed a consent form. One or 2 months before starting ovarian stimulation, on day 2 of a spontaneous menstrual cycle, patients underwent a transvaginal ultrasound scan by two different operators using the same ultrasound machine (Philips HD-3; Koninkijke Philips Electronics N.V.) for the estimation of AFC by measuring all visible follicles 2–5 mm in diameter in both ovaries. On the same day, a peripheral blood sample was obtained for the measurement of FSH and AMH levels. After extracting sera, the blood samples were stored at -20 °C until assayed. For pituitary suppression the GnRH-antagonist protocol with Ganirelix (OrgalutranÒ N. V. Organon, Oss, The Netherlands) or Cetrorelix (CetrotideÒMerck Serono Europe Ltd, London, UK) from day 6 of the controlled ovarian stimulation cycle was used. Recombinant FSH (Gonal-FÒ, Merck Serono Europe Ltd, London, UK or PuregonÒ N. V. Organon, Oss, The Netherlands) or urinary purified FSH was used for ovarian stimulation. The starting dose was chosen on the basis of age, FSH, BMI and experience from previous cycles. FSH doses were further adjusted according to ultrasound findings and E2 measurements during the stimulation monitoring. On day 5 of stimulation, peripheral blood was obtained for E2 and LH measurements while extra sample was also obtained for AMH levels. The last was stored, after serum extraction, as described earlier. When at least two follicles reached a diameter C18 mm, 10,000 IU of hCG (PregnylÒ N. V. Organon, Oss, The Netherlands or OvitrelleÒ, Merck Serono Europe Ltd, London, UK) was administered and 35–36 h later oocyte retrieval was performed under light sedation. Follicular fluid from the first mature follicle aspirated was collected before using any flushing for oocyte recovery. The follicular fluid was then centrifuged at 3,000 cycles/min and stored at -20 °C until assayed. If there was an oocyte in this first follicle it was isolated in a separate Petri with culture medium and if fertilized the ensuing embryo was cultured separately. Intracytoplasmic sperm injection (ICSI) was performed in all cases. The embryos were cultured in 37 °C, 5 % CO2 and humified atmosphere. Each day, the embryos were scored by two different senior embryologists according to the number of blastomeres and standard morphological criteria described before [5]. Embryo transfers were performed on day 2 or 3 under ultrasound guidance. The best-quality embryos using standard embryological criteria were transferred. In patients \40-year old 1–3 embryos were transferred and in patients
Arch Gynecol Obstet
[40-year old no more than 4 embryos were transferred according to the Greek law. Luteal phase was supported with micronized progesterone (UtrogestanÒ, Laboratoires Besins Int., Paris, France). Pregnancy was confirmed by a serum b-hCG pregnancy test performed 14 days after oocyte collection and repeated 2 days later if positive. Clinical (ongoing) pregnancy was defined as a viable intrauterine pregnancy (positive heartbeat) on transvaginal ultrasound scan performed between 6 and 7 weeks of pregnancy. Only clinical pregnancies were included in the analysis. Intraplasmic FSH was assessed using an immunoenzymatic assay (Immuno 1, Technicon, Bayer Corp., Tarrytown, NY, USA). The sensitivity of the assay was 0.1 IU/l. E2 and AMH levels were measured with an ELISA commercially available kit (Diagnostic Systems Laboratories, Webster, TX, USA). The sensitivity was 0.025 ng/ml. The intra- and inter-assay variation of the assay was \7 %. The main outcome measures were the number of large follicles at the day of oocyte retrieval, the number of metaphase II (M II) oocytes, the number of embryos generated by ICSI and clinical pregnancy. Poor ovarian response was defined as \4 oocytes retrieved and high response as [12 oocytes retrieved. Data were analysed with the Statistical Package for the Social Science (SPSS Inc., Chicago USA version 7). The level of significance for all statistical tests was 0.05.
Results At the day of follicular puncture, there were no mature oocytes in four patients. Depending on the number of oocytes retrieved, 35 patients were characterized as poor
responders, 62 had a normal response whereas 8 had an excessive response. Embryo transfers were performed at day 2 or day 3 post ICSI. In two patients embryo transfer was not performed due to lack of fertilization. Thirty-four patients (32.4 %) had a viable intrauterine pregnancy. Demographic data and parameters related to the outcome of ovarian stimulation and ICSI are presented in Table 1. There were statistically significant differences among poor, normal and high responders regarding the baseline FSH and AMH levels, peak E2 levels, number of follicles and number of embryos (Table 1). As expected, the levels of AMH at day 5 were similar to the baseline AMH levels in all groups of patients. The number of oocytes as well as the number of embryos was significantly negatively correlated with baseline FSH and positively correlated with baseline AMH, AMH at day 5 and AFC. The strongest correlation was found with AFC (Table 2). Receiver-operation curve (ROC) analysis was performed for the groups of poor and high responders including baseline FSH, baseline AMH and AFC. The results of ROC analysis confirmed the prognostic value of these parameters in both groups regarding the number of retrieved oocytes (Table 3). Stepwise regression analysis showed that the most significant parameters for the prediction of the number of retrieved oocytes were AFC and baseline AMH giving the following model with r2 = 0.453 (p \ 0.001): number of oocytes = -1.965 ? 1.026 AFC ? 0.704 baseline AMH. For the prediction of the number of embryos, stepwise regression analysis showed that the most significant parameter was AFC giving the following model with r2 = 0.696 (p \ 0.001): number of embryos = 0.538 AFC
Table 1 Demographic data and outcome of ovarian stimulation and ICSI in 105 cycles Total (n = 105)
Poor responders (n = 35)
Normal responders (n = 62)
High responders (n = 8)
p value
Age
36.64 ± 3.68
37.34 ± 4.14
36.40 ± 3.38
35.38 ± 3.66
0.292
BMI
23.29 ± 2.34
23.06 ± 2.24
23.39 ± 2.33
23.50 ± 3.02
0.775
Total FSH administered (IU)
2,613.45 ± 965.62
2,882.86 ± 1,066.64
2,518.35 ± 871.03
2,171.88 ± 1,024.56
0.081
Days of ovarian stimulation Baseline FSH (IU/l)
10.02 ± 1.49 8.24 ± 4.31
9.63 ± 1.96 9.71 ± 4.95
10.21 ± 1.15 7.71 ± 3.92
10.25 ± 1.28 5.94 ± 2.06
0.164 0.025
Baseline AMH (ng/ml)
2.45 ± 1.25
2.07 ± 0.96
2.53 ± 1.19
3.45 ± 2.11
0.012
AFC
5.60 ± 2.38
3.69 ± 1.61
6.32 ± 2.07
8.38 ± 1.51
\0.001
3.31 ± 2.73
0.027
AMH at day 5 (ng/ml)
2.35 ± 1.38
1.95 ± 0.89
1,480.13 ± 1,059.66
797.38 ± 452.83
Number of large follicles at the day of oocyte retrieval
6.88 ± 4.27
3.31 ± 2.03
7.87 ± 3.05
14.75 ± 5.12
\0.001
Number of oocytes
5.50 ± 4.17
1.63 ± 1.00
6.35 ± 1.86
15.88 ± 4.42
\0.001
Number of embryos
3.10 ± 2.41
1.00 ± 0.70
3.95 ± 1.88
5.75 ± 4.03
\0.001
Peak E2 (pg/ml)
2.45 ± 1.31 1,617.93 ± 714.64
3,399.25 ± 2,127.95 \0.001
Comparison among poor, normal and high responders was performed with ANOVA. Values are presented as mean ± standard deviation
123
Arch Gynecol Obstet Table 2 Spearman correlation coefficient among the number of retrieved oocytes, number of embryos and baseline FSH, AMH blood levels and AFC Baseline FSH
Baseline AMH
AMH at day 5
AFC
Number of oocytes
-0.831**
0.259**
0.196*
0.696**
Number of embryos
-0.266**
0.206*
0.229*
0.491**
** p \ 0.01, * p \ 0.05
Table 3 ROC analysis for the evaluation of prognostic value of baseline FSH, baseline AMH and AFC on the number of retrieved oocytes Area under the curve (AUC)
Significance (p)
Embryo quality
n
AMH (ng/ml)
NE
25
1.88 ± 1.20
Bad
12
2.67 ± 1.92
8
2.70 ± 3.79 2.28 ± 2.05
Medium Fair
27
Excellent
27
3.01 ± 1.81
Total
99
2.46 ± 1.99
AMH values are given as mean ± standard deviation. NE no embryos available due to the absence of oocytes in follicular fluids or fertilization failure. There were no statistical significant differences
Asymptotic 95 % CI Lower bound
Upper bound
Poor responders AFC
0.858
\0.001
0.785
0.930
Baseline FSH
0.671
0.004
0.563
0.780
Baseline AMH
0.634
0.026
0.523
0.745
AFC
0.862
0.001
0.772
0.952
Baseline FSH
0.724
0.036
0.565
0.882
Baseline AMH
0.664
0.125
0.465
0.863
High responders
In 99 cases, the fluid of the first punctured follicle was collected without using medium for flushing. In the follicular fluids collected, AMH was measured. The intrafollicular level of AMH was 2.46 ± 1.99 ng/ml. Twenty-eight follicular fluids were collected from poor responders with AMH concentration of 1.77 ± 1.48 ng/ml; 65 follicular fluids were collected from normal responders with AMH concentration of 2.60 ± 2.01 ng/ml and 6 were collected from high responders with AMH concentration of 1.82 ± 1.45 ng/ml. The differences between the three groups were not at a statistical significant level (p = 0.080). In 74 out of 99 follicular fluids oocytes were found, 62 of them were fertilized after ICSI and developed into embryos. The quality of embryos based on morphological criteria and assessed at day 3 was classified as ‘‘bad’’, ‘‘medium’’, ‘‘fair’’ or ‘‘excellent’’. In Table 4, the intrafollicular concentrations of AMH corresponding to these embryos are presented. Although the intrafollicular concentrations of AMH in the group of ‘‘excellent’’ embryos were higher, there were not statistically significant differences between the groups.
Comments The results of this study indicate that all the investigated parameters are useful predictors of the number of retrieved
123
Table 4 Intrafollicular concentrations of AMH corresponding to embryos
mature oocytes in women stimulated with GnRH-antagonist protocol. Among the prognostic markers investigated AFC showed the strongest correlation with the number of oocytes retrieved and the number of embryos generated (Table 2). Particularly, on the prediction of poor and high responders, ROC analysis indicated that AFC had the best performance (Table 3). Moreover, stepwise regression analysis showed that AFC and baseline AMH can successfully predict the number of retrieved oocytes whereas AFC alone can successfully predict the number of embryos. It is worth noting that previous studies have shown that AFC and AMH have a similar predictive value [6–14] whereas others have suggested AFC as a better predictor than AMH [15, 16]. The blood levels of AMH at day 5 did not significantly differ than baseline AMH levels and they did not either offer a better predictive value despite previous reports [6]. Besides, serum AMH levels have shown to be stable during the menstrual cycle [4]. Thus, to our opinion the measurement of AMH at mid-stimulation has no meaning if AMH was measured at the beginning of the cycle. In the present study, the relation of intrafollicular AMH levels with the embryo quality was also investigated. According to our results, the embryo quality is independent of intrafollicular AMH levels. Similar observations have also reported by Fanchin et al. [17]. To our opinion, AFC, baseline AMH serum levels and baseline FSH serum levels all appear as good and useful predictors for the outcome of ovarian stimulation in GnRHantagonist stimulated cycles. The better predictive value of AFC makes it a useful tool in counseling patients and individualizing the dose of exogenous gonadotropins during ovarian stimulation. Conflict of interest All authors declare that they have participated to this study and approved the final version submitted to your journal. Also all authors declare that there is no conflict of interest.
Arch Gynecol Obstet
References 1. Kolibianakis EM, Griesinger G, Tarlatzis BC (2012) GnRHantagonists in ovarian stimulation for IVF. In: Gardner DK, Weissman A, howles CM, Shoham Z (eds) Textbook for Assisted Reproductive Techniques, 4th edn, vol 2, clinical perspectives. Informa healthcare, pp 124–131 2. Johnson J, Canning J, Kaneko T, Pu JK, Tilly JL (2004) Germline stem cells and follicular renewal in the postnatal mammalian ovary. Nature 428(6979):145–150 3. Fauser BC, Diedrich K, Devroey P, Evian Annual Reproduction Workshop Group 2007 (2008) Predictors of ovarian response: progress towards individualized treatment in ovulation induction and ovarian stimulation. Hum Reprod Update 14(1):1–14 4. Tsepelidis S, Devreker F, Demeestere I, Flahaut A, Gervy Ch, Englert Y (2007) Stable serum levels of anti-Mu¨llerian hormone during the menstrual cycle: a prospective study in normo-ovulatory women. Hum Reprod 22:1837–1840 5. Baczkowski T, Kurzawa R, Glabowski W (2004) Methods of embryo scoring in in vitro fertilization. Reprod Biol 4(1):5–22 6. Penarrubia J, Fa´beques F, Manau D, Creus M, Casals G, Casamitjana R et al (2005) Basal and stimulation day 5 anti-mullerian hormone serum concentrations as predictors of ovarian response and pregnancy in assisted reproductive technology cycles stimulated with gonadotropin-releasing hormone agonist—gonadotropin treatment. Hum Reprod 20(4):915–922 7. Van Rooij IA, Broekmans FJ, te Velde ER, Fauser BC, Bancsi LF, Jong FH et al (2002) Serum anti-mullerian hormone levels: a novel measure of ovarian reserve. Hum Reprod 17:3065–3071 8. Muttukrishna S, McGarrigle H, Wakim R, Khadum I, Ranieri DM, Serhal P (2005) Antral follicle count, anti-mullerian hormone and inhibin B: predictors of ovarian response in assisted reproductive technology? BJOG 112:1384–1390 9. Elgindy EA, El-Haieg DO, El-Sebaey A (2008) Anti-mullerian hormone: correlation of early follicular, ovulatory and midluteal levels with ovarian response and cycle outcome in intracytoplasmic sperm injection patients. Fertil Steril 89:1670–1676 10. Lekamge DN, Barry M, Kolo M, Lane M, Gilchrist RB, Tremellen KP (2007) Anti-mullerian hormone as a predictor of IVF outcome. Reprod Biomed Online 14:602–610
11. Broer SL, Mol BW, Hendriks D, Broekmans FJ (2009) The role of antimullerian hormone in prediction of outcome after IVF: comparison with the antral follicle count. Fertil Steril 91(3):705–714 12. Jayaprakasan K, Campbell B, Hopkisson J, Johnson I, RaineFenning N (2010) A prospective, comparative analysis of antiMullerian hormone, inhibin-B, and three-dimensional ultrasound determinants of ovarian reserve in the prediction of poor response to controlled ovarian stimulation. Fertil Steril 93(3):855–864 13. Broer SL, Dolleman M, Opmeer BC, Fauser BC, Mol BW, Broekmans FJ (2011) AMH and AFC as predictors of excessive response in controlled ovarian hyperstimulation: a meta-analysis. Hum Reprod Update 17(1):46–54 14. Broer SL, van Disseldorp J, Broeze KA, Dolleman M, Opmeer BC, Bossuyt P et al (2013) Added value of ovarian reserve testing on patient characteristics in the prediction of ovarian response and ongoing pregnancy: an individual patient data approach. Hum Reprod Update 19(1):26–36 15. Eldar-Geva T, Ben-Chetrit A, Spitz IM, Rabinowitz R, Markowitz E, Mimoni T et al (2005) Dynamic assays of inhibin B, antimullerian hormone and estradiol following FSH stimulation and ovarian ultrasonography as predictors of IVF outcome. Hum Reprod 20:3178–3183 16. Kwee J, Schats R, McDonnell J, Themmen A, de Jong F, Lambalk C (2008) Evaluation of anti-mu¨llerian hormone as a test for the prediction of ovarian reserve. Fertil Steril 90:737–743 17. Fanchin R, Mendez Lozano DH, Frydman N, Gougeon A, di Clemente N, Frydman R et al (2007) Anti-mu¨llerian hormone concentrations in the follicular fluid of the preovulatory follicle are predictive of the implantation potential of the ensuing embryo obtained by in vitro fertilization. J Clin Endocrinol Metab 92(5):1796–1802
123
GYNECOLOGIC ENDOCRINOLOGY AND REPRODUCTIVE MEDICINE
Predictive value of anti-mu¨llerian hormone, follicle-stimulating hormone and antral follicle count on the outcome of ovarian stimulation in women following GnRH-antagonist protocol for IVF/ET E. Tsakos • A. Tolikas • Angelos Daniilidis B. Asimakopoulos
•
Received: 25 February 2014 / Accepted: 17 June 2014 Ó Springer-Verlag Berlin Heidelberg 2014
Abstract Aim To compare the efficacy of three methods: the levels of anti-mu¨llerian hormone (AMH), the levels of follicle stimulating hormone (FSH) and the antral follicle count (AFC), for the prediction of the number of retrieved mature oocytes and the number of generated embryos by intracytoplasmic sperm injection (ICSI) in women stimulated with a GnRH-antagonist protocol. Materials and methods 105 women were enrolled in the study. At the day 2 of a preceding cycle, AFC was performed and FSH and AMH were measured in serum by immunoenzymatic assays. All women were stimulated with a GnRH-antagonist protocol and ovulation was induced with human chorionic gonadotropin. ICSI was performed in all retrieved mature oocytes. Embryo transfers were performed at days 2–3. According to the oocytes retrieved, patients were categorized as poor (\4), normal (4–12) or high responders ([12). Results AFC and the levels of baseline FSH and AMH were significantly different among poor, normal and high responders. The number of oocytes as well as the number of embryos was negatively correlated with baseline FSH and positively correlated with baseline AMH and AFC, whereas AFC showed the strongest correlation. Stepwise
E. Tsakos A. Tolikas B. Asimakopoulos Embryoclinic, Thessalonı´ki, Greece A. Daniilidis (&) School of Medicine, Aristotle University of Thessaloniki, Greece, 9 Smirnis, Evosmos, 56224 Thessalonı´ki, Greece e-mail: [email protected] B. Asimakopoulos Laboratory of Physiology, School of Medicine, Democritus University of Thrace, Alexandroupolis, Greece
regression analysis indicated AFC and baseline AMH as the most significant parameters for the prediction of the number of oocytes; for the prediction of the number of embryos, the most significant parameter was AFC. Conclusions AFC, baseline AMH and baseline FSH are good predictors for the outcome of ovarian stimulation in GnRH-antagonist cycles. However, AFC appears to have the best predictive value. Keywords Antral follicle count FSH Anti-mu¨llerian hormone Ovarian response GnRH antagonist
Introduction Fertility potential of women is dependent on the quantity and the quality of the ovarian reserve that declines with age. The estimation of the ovarian reserve is of crucial importance for the fertility potential of women. It is one of the prerequisites for women undergoing in vitro fertilization (IVF) treatment. The ovarian reserve constitutes by the size of the follicular pool that is established during fetal life. Around the 16th week of fetal life approximately seven millions of follicles exist in both ovaries. Each primordial follicle consists of an oocyte surrounded by somatic cells. At birth, around one million of follicles are present and at menarche only 300–400,000 remain [1]. During the female reproductive life, waves of follicles leave the primordial pool to undergo further development. The last includes proliferation and differentiation of somatic cells, increase of the follicular size, completion of the first meiotic division and introduction to the second meiotic division of the oocyte that stops at the stage of metaphase. The vast majority of follicles fail to complete development undergoing atresia.
123
Arch Gynecol Obstet
In each menstrual cycle, usually only one follicle is selected to become the dominant one which completes the developmental cycle and thus resulting to ovulation. As this process continues throughout reproductive life, the follicle pool is finally exhausted leading to menopause [1]. This is the classical view of a finite follicle pool. Although it has been challenged by recent findings that, at least in mouse, germline stem cells maintain oogenesis throughout adulthood [2], the gradual depletion of the follicle pool during female reproductive life is well established. Various methods of assessing the ovarian reserve have been used in clinical practice [3]. Among them, the measurement of baseline follicle stimulating hormone (FSH) levels, the assessment of the number of antral follicles (AFC) by ultrasonography and the estimation of antimu¨llerian hormone (AMH) levels are probably the more frequently used worldwide. FSH secreted by the gonadotrope cells of anterior pituitary stimulates ovarian granulosa cells to express aromatase. Thus, granulosa cells of early antral follicles convert androgens to estrogens. In turn, estrogens decrease the secretion of FSH by gonadotrope pituitary cells via a negative feedback system. As the follicle pool declines, estradiol (E2) blood levels decline too and consequently blood FSH levels rise [1]. The AFC by ultrasonography at the early follicular phase reliably reflects the ovarian reserve [3]. AMH is a dimeric glycoprotein first studied for its role in differentiation of gonads. It belongs to the TGFb family and in females during adulthood is produced by granulosa cells of primary and secondary follicles which are not yet influenced by FSH. Its production decreases in the larger follicles and also decreases as the follicular pool diminishes [4]. The ovarian reserve is a determinative factor for the number of retrieved oocytes in a stimulated cycle for IVF. Moreover, the quality of the ovarian reserve also influences the quality of retrieved oocytes and therefore the number and quality of embryos generated in IVF. This study was designed to find out the best predictor of the number of retrieved oocytes and the number of embryos generated by intracytoplasmic sperm injection in women undergoing ovarian stimulation with the GnRHantagonist multiple dose protocol. The parameters that were evaluated for their predictive value were serum FSH at day 2 (baseline FSH), AFC before stimulation, AMH serum levels at days 2 and 5 and AMH intrafollicular levels.
Materials and methods This study included prospectively 105 women attending our intracytoplasmic sperm injection (ICSI) program from October 2009 to January 2013.
123
The inclusion criteria were (1) age 25–45 years, (2) regular menstruation, (3) the absence of serious endocrinological disorders, (4) the presence of both ovaries, and (5) BMI 19–30. Approval was obtained from the Bioethics Committee of the Aristotle’s University of Thessaloniki and all patients signed a consent form. One or 2 months before starting ovarian stimulation, on day 2 of a spontaneous menstrual cycle, patients underwent a transvaginal ultrasound scan by two different operators using the same ultrasound machine (Philips HD-3; Koninkijke Philips Electronics N.V.) for the estimation of AFC by measuring all visible follicles 2–5 mm in diameter in both ovaries. On the same day, a peripheral blood sample was obtained for the measurement of FSH and AMH levels. After extracting sera, the blood samples were stored at -20 °C until assayed. For pituitary suppression the GnRH-antagonist protocol with Ganirelix (OrgalutranÒ N. V. Organon, Oss, The Netherlands) or Cetrorelix (CetrotideÒMerck Serono Europe Ltd, London, UK) from day 6 of the controlled ovarian stimulation cycle was used. Recombinant FSH (Gonal-FÒ, Merck Serono Europe Ltd, London, UK or PuregonÒ N. V. Organon, Oss, The Netherlands) or urinary purified FSH was used for ovarian stimulation. The starting dose was chosen on the basis of age, FSH, BMI and experience from previous cycles. FSH doses were further adjusted according to ultrasound findings and E2 measurements during the stimulation monitoring. On day 5 of stimulation, peripheral blood was obtained for E2 and LH measurements while extra sample was also obtained for AMH levels. The last was stored, after serum extraction, as described earlier. When at least two follicles reached a diameter C18 mm, 10,000 IU of hCG (PregnylÒ N. V. Organon, Oss, The Netherlands or OvitrelleÒ, Merck Serono Europe Ltd, London, UK) was administered and 35–36 h later oocyte retrieval was performed under light sedation. Follicular fluid from the first mature follicle aspirated was collected before using any flushing for oocyte recovery. The follicular fluid was then centrifuged at 3,000 cycles/min and stored at -20 °C until assayed. If there was an oocyte in this first follicle it was isolated in a separate Petri with culture medium and if fertilized the ensuing embryo was cultured separately. Intracytoplasmic sperm injection (ICSI) was performed in all cases. The embryos were cultured in 37 °C, 5 % CO2 and humified atmosphere. Each day, the embryos were scored by two different senior embryologists according to the number of blastomeres and standard morphological criteria described before [5]. Embryo transfers were performed on day 2 or 3 under ultrasound guidance. The best-quality embryos using standard embryological criteria were transferred. In patients \40-year old 1–3 embryos were transferred and in patients
Arch Gynecol Obstet
[40-year old no more than 4 embryos were transferred according to the Greek law. Luteal phase was supported with micronized progesterone (UtrogestanÒ, Laboratoires Besins Int., Paris, France). Pregnancy was confirmed by a serum b-hCG pregnancy test performed 14 days after oocyte collection and repeated 2 days later if positive. Clinical (ongoing) pregnancy was defined as a viable intrauterine pregnancy (positive heartbeat) on transvaginal ultrasound scan performed between 6 and 7 weeks of pregnancy. Only clinical pregnancies were included in the analysis. Intraplasmic FSH was assessed using an immunoenzymatic assay (Immuno 1, Technicon, Bayer Corp., Tarrytown, NY, USA). The sensitivity of the assay was 0.1 IU/l. E2 and AMH levels were measured with an ELISA commercially available kit (Diagnostic Systems Laboratories, Webster, TX, USA). The sensitivity was 0.025 ng/ml. The intra- and inter-assay variation of the assay was \7 %. The main outcome measures were the number of large follicles at the day of oocyte retrieval, the number of metaphase II (M II) oocytes, the number of embryos generated by ICSI and clinical pregnancy. Poor ovarian response was defined as \4 oocytes retrieved and high response as [12 oocytes retrieved. Data were analysed with the Statistical Package for the Social Science (SPSS Inc., Chicago USA version 7). The level of significance for all statistical tests was 0.05.
Results At the day of follicular puncture, there were no mature oocytes in four patients. Depending on the number of oocytes retrieved, 35 patients were characterized as poor
responders, 62 had a normal response whereas 8 had an excessive response. Embryo transfers were performed at day 2 or day 3 post ICSI. In two patients embryo transfer was not performed due to lack of fertilization. Thirty-four patients (32.4 %) had a viable intrauterine pregnancy. Demographic data and parameters related to the outcome of ovarian stimulation and ICSI are presented in Table 1. There were statistically significant differences among poor, normal and high responders regarding the baseline FSH and AMH levels, peak E2 levels, number of follicles and number of embryos (Table 1). As expected, the levels of AMH at day 5 were similar to the baseline AMH levels in all groups of patients. The number of oocytes as well as the number of embryos was significantly negatively correlated with baseline FSH and positively correlated with baseline AMH, AMH at day 5 and AFC. The strongest correlation was found with AFC (Table 2). Receiver-operation curve (ROC) analysis was performed for the groups of poor and high responders including baseline FSH, baseline AMH and AFC. The results of ROC analysis confirmed the prognostic value of these parameters in both groups regarding the number of retrieved oocytes (Table 3). Stepwise regression analysis showed that the most significant parameters for the prediction of the number of retrieved oocytes were AFC and baseline AMH giving the following model with r2 = 0.453 (p \ 0.001): number of oocytes = -1.965 ? 1.026 AFC ? 0.704 baseline AMH. For the prediction of the number of embryos, stepwise regression analysis showed that the most significant parameter was AFC giving the following model with r2 = 0.696 (p \ 0.001): number of embryos = 0.538 AFC
Table 1 Demographic data and outcome of ovarian stimulation and ICSI in 105 cycles Total (n = 105)
Poor responders (n = 35)
Normal responders (n = 62)
High responders (n = 8)
p value
Age
36.64 ± 3.68
37.34 ± 4.14
36.40 ± 3.38
35.38 ± 3.66
0.292
BMI
23.29 ± 2.34
23.06 ± 2.24
23.39 ± 2.33
23.50 ± 3.02
0.775
Total FSH administered (IU)
2,613.45 ± 965.62
2,882.86 ± 1,066.64
2,518.35 ± 871.03
2,171.88 ± 1,024.56
0.081
Days of ovarian stimulation Baseline FSH (IU/l)
10.02 ± 1.49 8.24 ± 4.31
9.63 ± 1.96 9.71 ± 4.95
10.21 ± 1.15 7.71 ± 3.92
10.25 ± 1.28 5.94 ± 2.06
0.164 0.025
Baseline AMH (ng/ml)
2.45 ± 1.25
2.07 ± 0.96
2.53 ± 1.19
3.45 ± 2.11
0.012
AFC
5.60 ± 2.38
3.69 ± 1.61
6.32 ± 2.07
8.38 ± 1.51
\0.001
3.31 ± 2.73
0.027
AMH at day 5 (ng/ml)
2.35 ± 1.38
1.95 ± 0.89
1,480.13 ± 1,059.66
797.38 ± 452.83
Number of large follicles at the day of oocyte retrieval
6.88 ± 4.27
3.31 ± 2.03
7.87 ± 3.05
14.75 ± 5.12
\0.001
Number of oocytes
5.50 ± 4.17
1.63 ± 1.00
6.35 ± 1.86
15.88 ± 4.42
\0.001
Number of embryos
3.10 ± 2.41
1.00 ± 0.70
3.95 ± 1.88
5.75 ± 4.03
\0.001
Peak E2 (pg/ml)
2.45 ± 1.31 1,617.93 ± 714.64
3,399.25 ± 2,127.95 \0.001
Comparison among poor, normal and high responders was performed with ANOVA. Values are presented as mean ± standard deviation
123
Arch Gynecol Obstet Table 2 Spearman correlation coefficient among the number of retrieved oocytes, number of embryos and baseline FSH, AMH blood levels and AFC Baseline FSH
Baseline AMH
AMH at day 5
AFC
Number of oocytes
-0.831**
0.259**
0.196*
0.696**
Number of embryos
-0.266**
0.206*
0.229*
0.491**
** p \ 0.01, * p \ 0.05
Table 3 ROC analysis for the evaluation of prognostic value of baseline FSH, baseline AMH and AFC on the number of retrieved oocytes Area under the curve (AUC)
Significance (p)
Embryo quality
n
AMH (ng/ml)
NE
25
1.88 ± 1.20
Bad
12
2.67 ± 1.92
8
2.70 ± 3.79 2.28 ± 2.05
Medium Fair
27
Excellent
27
3.01 ± 1.81
Total
99
2.46 ± 1.99
AMH values are given as mean ± standard deviation. NE no embryos available due to the absence of oocytes in follicular fluids or fertilization failure. There were no statistical significant differences
Asymptotic 95 % CI Lower bound
Upper bound
Poor responders AFC
0.858
\0.001
0.785
0.930
Baseline FSH
0.671
0.004
0.563
0.780
Baseline AMH
0.634
0.026
0.523
0.745
AFC
0.862
0.001
0.772
0.952
Baseline FSH
0.724
0.036
0.565
0.882
Baseline AMH
0.664
0.125
0.465
0.863
High responders
In 99 cases, the fluid of the first punctured follicle was collected without using medium for flushing. In the follicular fluids collected, AMH was measured. The intrafollicular level of AMH was 2.46 ± 1.99 ng/ml. Twenty-eight follicular fluids were collected from poor responders with AMH concentration of 1.77 ± 1.48 ng/ml; 65 follicular fluids were collected from normal responders with AMH concentration of 2.60 ± 2.01 ng/ml and 6 were collected from high responders with AMH concentration of 1.82 ± 1.45 ng/ml. The differences between the three groups were not at a statistical significant level (p = 0.080). In 74 out of 99 follicular fluids oocytes were found, 62 of them were fertilized after ICSI and developed into embryos. The quality of embryos based on morphological criteria and assessed at day 3 was classified as ‘‘bad’’, ‘‘medium’’, ‘‘fair’’ or ‘‘excellent’’. In Table 4, the intrafollicular concentrations of AMH corresponding to these embryos are presented. Although the intrafollicular concentrations of AMH in the group of ‘‘excellent’’ embryos were higher, there were not statistically significant differences between the groups.
Comments The results of this study indicate that all the investigated parameters are useful predictors of the number of retrieved
123
Table 4 Intrafollicular concentrations of AMH corresponding to embryos
mature oocytes in women stimulated with GnRH-antagonist protocol. Among the prognostic markers investigated AFC showed the strongest correlation with the number of oocytes retrieved and the number of embryos generated (Table 2). Particularly, on the prediction of poor and high responders, ROC analysis indicated that AFC had the best performance (Table 3). Moreover, stepwise regression analysis showed that AFC and baseline AMH can successfully predict the number of retrieved oocytes whereas AFC alone can successfully predict the number of embryos. It is worth noting that previous studies have shown that AFC and AMH have a similar predictive value [6–14] whereas others have suggested AFC as a better predictor than AMH [15, 16]. The blood levels of AMH at day 5 did not significantly differ than baseline AMH levels and they did not either offer a better predictive value despite previous reports [6]. Besides, serum AMH levels have shown to be stable during the menstrual cycle [4]. Thus, to our opinion the measurement of AMH at mid-stimulation has no meaning if AMH was measured at the beginning of the cycle. In the present study, the relation of intrafollicular AMH levels with the embryo quality was also investigated. According to our results, the embryo quality is independent of intrafollicular AMH levels. Similar observations have also reported by Fanchin et al. [17]. To our opinion, AFC, baseline AMH serum levels and baseline FSH serum levels all appear as good and useful predictors for the outcome of ovarian stimulation in GnRHantagonist stimulated cycles. The better predictive value of AFC makes it a useful tool in counseling patients and individualizing the dose of exogenous gonadotropins during ovarian stimulation. Conflict of interest All authors declare that they have participated to this study and approved the final version submitted to your journal. Also all authors declare that there is no conflict of interest.
Arch Gynecol Obstet
References 1. Kolibianakis EM, Griesinger G, Tarlatzis BC (2012) GnRHantagonists in ovarian stimulation for IVF. In: Gardner DK, Weissman A, howles CM, Shoham Z (eds) Textbook for Assisted Reproductive Techniques, 4th edn, vol 2, clinical perspectives. Informa healthcare, pp 124–131 2. Johnson J, Canning J, Kaneko T, Pu JK, Tilly JL (2004) Germline stem cells and follicular renewal in the postnatal mammalian ovary. Nature 428(6979):145–150 3. Fauser BC, Diedrich K, Devroey P, Evian Annual Reproduction Workshop Group 2007 (2008) Predictors of ovarian response: progress towards individualized treatment in ovulation induction and ovarian stimulation. Hum Reprod Update 14(1):1–14 4. Tsepelidis S, Devreker F, Demeestere I, Flahaut A, Gervy Ch, Englert Y (2007) Stable serum levels of anti-Mu¨llerian hormone during the menstrual cycle: a prospective study in normo-ovulatory women. Hum Reprod 22:1837–1840 5. Baczkowski T, Kurzawa R, Glabowski W (2004) Methods of embryo scoring in in vitro fertilization. Reprod Biol 4(1):5–22 6. Penarrubia J, Fa´beques F, Manau D, Creus M, Casals G, Casamitjana R et al (2005) Basal and stimulation day 5 anti-mullerian hormone serum concentrations as predictors of ovarian response and pregnancy in assisted reproductive technology cycles stimulated with gonadotropin-releasing hormone agonist—gonadotropin treatment. Hum Reprod 20(4):915–922 7. Van Rooij IA, Broekmans FJ, te Velde ER, Fauser BC, Bancsi LF, Jong FH et al (2002) Serum anti-mullerian hormone levels: a novel measure of ovarian reserve. Hum Reprod 17:3065–3071 8. Muttukrishna S, McGarrigle H, Wakim R, Khadum I, Ranieri DM, Serhal P (2005) Antral follicle count, anti-mullerian hormone and inhibin B: predictors of ovarian response in assisted reproductive technology? BJOG 112:1384–1390 9. Elgindy EA, El-Haieg DO, El-Sebaey A (2008) Anti-mullerian hormone: correlation of early follicular, ovulatory and midluteal levels with ovarian response and cycle outcome in intracytoplasmic sperm injection patients. Fertil Steril 89:1670–1676 10. Lekamge DN, Barry M, Kolo M, Lane M, Gilchrist RB, Tremellen KP (2007) Anti-mullerian hormone as a predictor of IVF outcome. Reprod Biomed Online 14:602–610
11. Broer SL, Mol BW, Hendriks D, Broekmans FJ (2009) The role of antimullerian hormone in prediction of outcome after IVF: comparison with the antral follicle count. Fertil Steril 91(3):705–714 12. Jayaprakasan K, Campbell B, Hopkisson J, Johnson I, RaineFenning N (2010) A prospective, comparative analysis of antiMullerian hormone, inhibin-B, and three-dimensional ultrasound determinants of ovarian reserve in the prediction of poor response to controlled ovarian stimulation. Fertil Steril 93(3):855–864 13. Broer SL, Dolleman M, Opmeer BC, Fauser BC, Mol BW, Broekmans FJ (2011) AMH and AFC as predictors of excessive response in controlled ovarian hyperstimulation: a meta-analysis. Hum Reprod Update 17(1):46–54 14. Broer SL, van Disseldorp J, Broeze KA, Dolleman M, Opmeer BC, Bossuyt P et al (2013) Added value of ovarian reserve testing on patient characteristics in the prediction of ovarian response and ongoing pregnancy: an individual patient data approach. Hum Reprod Update 19(1):26–36 15. Eldar-Geva T, Ben-Chetrit A, Spitz IM, Rabinowitz R, Markowitz E, Mimoni T et al (2005) Dynamic assays of inhibin B, antimullerian hormone and estradiol following FSH stimulation and ovarian ultrasonography as predictors of IVF outcome. Hum Reprod 20:3178–3183 16. Kwee J, Schats R, McDonnell J, Themmen A, de Jong F, Lambalk C (2008) Evaluation of anti-mu¨llerian hormone as a test for the prediction of ovarian reserve. Fertil Steril 90:737–743 17. Fanchin R, Mendez Lozano DH, Frydman N, Gougeon A, di Clemente N, Frydman R et al (2007) Anti-mu¨llerian hormone concentrations in the follicular fluid of the preovulatory follicle are predictive of the implantation potential of the ensuing embryo obtained by in vitro fertilization. J Clin Endocrinol Metab 92(5):1796–1802
123
