Child, Tim J. MD, MRCOG; Phillips, Simon J. BSc; Abdul‐Jalil, Ahmad Kamal MS; Gulekli, Bulent MD; Tan, Seang Lin MD, FRCOG
Conventional in vitro fertilization (IVF) treatment requires gonadotropin ovarian stimulation to increase the numbers of mature oocytes retrieved, embryos available for transfer, and, consequently, pregnancy rates. However, there are numerous disadvantages with gonadotropin stimulation including high drug costs, need for daily injections and frequent monitoring, and potential side effects such as abdominal bloating, breast tenderness, mood swings, and nausea. More importantly, ovarian stimulation is associated with a potential risk of ovarian hyperstimulation syndrome. Women who have polycystic ovaries visualized on ultrasound examination, regardless of whether there are clinical (amenorrhea, oligomenorrhea, hirsutism) or endocrine (raised luteinizing hormone [LH] and androgen levels, elevated LH/follicle‐stimulating hormone [FSH] ratio) features of polycystic ovarian syndrome (PCOS), are at particularly high risk of developing ovarian hyperstimulation syndrome.1 Finally, there is concern over a possible link between ovarian stimulation and the long‐term risk of ovarian cancer.2 Consequently, there is an increasing desire for less aggressive ovarian stimulation using lower amounts of gonadotropins or, ideally, dispensing with ovarian stimulation altogether.3
In vitro maturation (IVM) of immature oocytes retrieved from unstimulated ovaries is a reproductive technology of increasing interest. In IVM, immature oocytes are retrieved transvaginally from 2‐ to 8‐mm diameter antral follicles within unstimulated ovaries and matured in vitro for 24–48 hours.4 Mature oocytes are then fertilized and multiple embryos transferred to the uterus 2–3 days later. Although IVM success rates have in the past been relatively low, recent studies have reported much higher pregnancy rates, particularly for women with polycystic ovaries.4–8 However, data are required comparing the relative success of IVM or IVF treatment for infertile women with polycystic ovaries. Because around a quarter of women attending assisted conception clinics have polycystic ovaries,9,10 such data are vital if patients are to have reliable information on which to base their choice of treatment. The aim of this study was to compare the numbers of mature oocytes and embryos produced and the rates of embryo implantation, pregnancy, live birth, and complications between women with polycystic ovaries undergoing IVM and those undergoing IVF treatment.
MATERIALS AND METHODS
We identified all IVM and IVF treatment cycles performed at the McGill Reproductive Center between February 1998 and November 2000 from our unit database for this case‐control study. Cases were defined as IVM cycles in which the patient had a diagnosis of polycystic ovaries. There were a total of 169 IVM cycles performed during the study period, of which 107 were for women with polycystic ovaries. An ovary was considered polycystic if there were ten or more small cysts or follicles visualized that were arranged around the periphery or scattered through an enlarged ovarian stroma on transvaginal ultrasound scan performed during the early follicular phase of the menstrual cycle.
The control for each IVM cycle was defined as the preceding IVF cycle for which the patient was within the same age group (younger than 30, 30–34, 35–39, and older than 39), had the same infertility diagnosis, and had polycystic ovaries on ultrasound examination. All patient files were then manually checked to confirm the polycystic ovaries diagnosis and to assess whether there had been complications such as ovarian hyperstimulation syndrome, pelvic infection, or bleeding during or after the treatment cycle.
During the study period women with polycystic ovaries who required assisted conception were offered, after full explanation, the option of either IVM or IVF treatment. The patient commenced the treatment of her choice after having given written informed consent. The characteristics of women choosing IVM or IVF could have been different. However, by using a case‐control study design we were able to match patients for two important baseline variables that may affect outcome—namely, age and diagnosis.
Women with amenorrhea received vaginal progesterone (Prometrium; Schering, Pointe‐Claire, Quebec, Canada), 300 mg once daily for 10 days, to induce withdrawal bleeding. All women underwent a baseline ultrasound scan on days 2–4 of menstrual bleeding, whether spontaneous or induced, to ensure that no ovarian cysts were present. No ovarian stimulation was used during the treatment cycle. Transvaginal ultrasound scans were repeated on the day of human chorionic gonadotropin (hCG) administration (2 days before immature oocyte retrieval) to exclude the development of a dominant follicle. All patients received 10,000 IU of hCG (Profasi; Serono, Oakville, Ontario, Canada) subcutaneously 36 hours before oocyte retrieval. We previously demonstrated, in a randomized controlled trial, that hCG priming increases both the percentage and rate of immature oocyte maturation.6 All follicles had to be less than 10 mm in diameter on the day of hCG to proceed to oocyte retrieval, which was performed between days 9 and 11 of the cycle for women with ovulatory cycles and days 9 and 14 for anovulatory patients. Data suggest that the presence of a dominant follicle at the time of immature oocyte retrieval is deleterious to outcome in IVM.11,12
Transvaginal ultrasound–guided oocyte collection was performed using a specially designed 17‐gauge single‐lumen aspiration needle (K‐OPS‐1235‐Wood; Cook IVF, Eight Mile Plains, Queensland, Australia) with a reduced aspiration pressure of 7.5 kPa. Aspiration of all small follicles was performed using either spinal anesthesia or intravenous fentanyl and midazolam with or without a paracervical block of 10 mL of 1% lidocaine. Follicular flushing was not performed.
The oocyte IVM technique used in this study has been reported in detail previously.6 Oocytes were collected in culture tubes containing warm 0.9% saline with 2‐IU/mL heparin. The oocytes were evaluated for the presence or absence of a germinal vesicle in the cytoplasm of the oocyte, and the immature oocytes were then transferred into maturation medium for culture. All oocyte‐handling procedures were conducted on warm stages and plates at 37C.
The immature oocytes were incubated in culture dishes containing 1 mL of maturation medium, TC‐199 medium supplemented with 20% heat‐inactivated maternal serum, 0.25‐mmol/L pyruvic acid (Sigma Chemical Co., St. Louis, MO), 50‐mg/mL penicillin, 75‐mg/mL streptomycin, and 75‐mIU/mL FSH and LH (Humegon; Organon, Scarborough, Ontario, Canada) at 37C in an atmosphere of 5% CO2 and 95% air with high humidity. After culture, the maturity of the oocytes was determined under the microscope at 24 h and 48 h. Mature oocytes were determined by the presence of first polar body extrusion. Oocytes that were mature at the time of checking were denuded of cumulus cells ready for intracytoplasmic sperm injection. A single spermatozoon was injected into each metaphase II oocyte. After intracytoplasmic sperm injection, each oocyte was transferred into a 20‐μL droplet of 1.2 medium G (Vitrolife, Göteborg, Sweden). Fertilization was assessed 18 h after intracytoplasmic sperm injection for the appearance of two distinct pronuclei and two polar bodies.
Embryos were transferred on day 2 or 3 after intracytoplasmic sperm injection. Because the oocytes were not matured and inseminated at the same time after maturation in culture, the developmental stages of embryos at the time of embryo transfer often varied. Before transfer, all embryos for each patient were pooled and selected for transfer based on standard embryological criteria such as cleavage stage and morphological quality.13
For endometrial preparation, patients received estradiol valerate (Estrace; Roberts Pharmaceutical, Mississauga, Ontario, Canada), starting on the day of oocyte retrieval, depending on the endometrial thickness on that day. If the endometrial thickness was less than 6 mm, a 10‐mg dose was given, and if it was greater than 6 mm, a 6‐mg dose was administered. If the endometrial thickness was less than 7 mm on the day of embryo transfer, the recommendation was cryopreservation of all embryos for replacement in a later cycle. Luteal support was provided by 200 mg of intravaginal progesterone (Prometrium) twice daily starting on the day of intracytoplasmic sperm injection and continued, along with estradiol, until 12 weeks' gestation.
Women underwent a long gonadotropin‐releasing hormone (GnRH) agonist protocol with a follicular phase start. Patients were pretreated with either the oral contraceptive pill for 2 weeks before GnRH agonist14 or oral progesterone (Norlutate; Pfizer, New York, NY) for 5 days from the first day of menstruation.15 Ovarian stimulation with urinary or recombinant FSH began after pituitary suppression was confirmed. An appropriate daily FSH dose was selected based on patient age, early follicular phase serum FSH concentration, and any previous response to ovarian stimulation. Human chorionic gonadotropin was administered when there were at least three follicles of at least 18 mm average diameter. Transvaginal ultrasound–guided oocyte collection was performed 36 hours later. Intracytoplasmic sperm injection was performed as required, and cleaving embryos replaced in the uterus 2–4 days later. Patients in both the IVM and the IVF groups were offered the option to cryopreserve any remaining best‐quality embryos. Luteal support was provided with progesterone pessaries (Prometrium), 200 mg twice daily, until 12 weeks' gestation.
For the purposes of this study, patients with ovarian hyperstimulation syndrome were classified into three groups: mild, moderate, and severe.16 Using this classification, mild ovarian hyperstimulation syndrome includes abdominal distension and discomfort, enlarged ovaries, and/or nausea, vomiting, or diarrhea. Moderate ovarian hyperstimulation syndrome is diagnosed when there is, in addition, ultrasonographic evidence of ascites. Severe ovarian hyperstimulation syndrome is diagnosed when there are clinical ascites, hydrothorax, and/or biochemical or hematological abnormalities.
A pregnancy test was considered positive when a serum hCG taken 16 days after oocyte fertilization had a concentration of greater than 25 IU/L. The implantation rate was calculated by dividing the total number of gestation sacs present on transvaginal ultrasound 4 weeks after oocyte retrieval by the total number of embryos transferred within a treatment group.
Statistical comparisons between categoric data were performed using the χ2 test. Because none of the oocyte and embryo outcome variables displayed a normal distribution (Kolmogorov‐Smirnov test), the non‐parametric Mann‐Whitney U test was used to analyze differences between unpaired data. All P values quoted are two sided, and values below .05 were taken to indicate statistical significance. Analyses were performed using the SPSS statistical package (SPSS Inc., Chicago, IL).
One hundred seven IVM and 107 IVF cycles were identified as cases and controls, respectively. There were 83 women in the IVM group and 81 in the IVF. There were no differences in mean age, previous number of IVF cycles undertaken, or infertility diagnosis between the treatment groups, indicating satisfactory matching for these variables.
Two cycles in each treatment group failed to reach embryo transfer. Causes were failure to retrieve oocytes in one IVM cycle and maturation failure of the four retrieved immature oocytes in the other. There was no transfer in two IVF cycles because of 1) cross‐contamination with Candida albicans from the patient to the embryo culture system and 2) a high estradiol level on the day of hCG administration requiring all embryos to be frozen for transfer in a later cycle because of concern over the risk of ovarian hyperstimulation syndrome.
One thousand one hundred two viable immature oocytes were collected in the IVM group. By 48 hours of culture 835 oocytes had matured to metaphase II (maturation rate 76%). After intracytoplasmic sperm injection of the in vitro–matured oocytes, the fertilization rate was 78% and the cleavage rate was 74%. The fertilization and cleavage rates of IVF oocytes were 78% and 72%, respectively (nonsignificantly different from IVM values).
There were significantly fewer oocytes collected, metaphase II oocytes, fertilized oocytes, and cleaving embryos in the IVM group (P < .01) (Table 1). A total of 336 and 284 embryos were transferred to the uterine cavity in the IVM and IVF groups, respectively. Consequently, the mean number transferred in the IVM group (mean ± SD 3.2 ± 0.9, range 1–5) was significantly greater than that in the IVF group (2.7 ± 0.8, range 1–6) (P < .01), though the median number transferred was three in both groups.
In vitro maturation cycles appeared less likely to result in pregnancy, clinical pregnancy, and live birth than IVF cycles, though the 95% confidence intervals for these outcomes just crossed unity, confirming nonsignificance. The implantation rate was significantly lower in the IVM group (9.5%) than in the IVF (17.1%) (P < .01). There were seven multiple live births in the IVM group (41.2%) and 10 in the IVF (37.0%) (nonsignificant), including one set of triplets in each.
There were 12 cases (11.2%) of moderate or severe ovarian hyperstimulation syndrome in the IVF group as defined using the classification of Golan.16 Seven of these (6.5%) were moderate ovarian hyperstimulation syndrome, and five (4.7%) severe. One patient required admission for 1 night for intravenous hydration, and another underwent transvaginal drainage of ascites, performed as an outpatient procedure. As was expected, there were no cases of ovarian hyperstimulation syndrome in the IVM group (P < .01).
There were no cases in either group of hemorrhage or pelvic infection requiring antibiotic treatment. One patient in each group was diagnosed with ectopic pregnancy. Both women were treated successfully with systemic methotrexate.
The results suggest that for women with polycystic ovaries who require assisted conception treatment, IVM is a promising alternative to stimulated IVF. Although the numbers of oocytes and embryos are significantly less, on average, at least five cleaving embryos were available per oocyte retrieval through IVM treatment. These embryos were produced without the need for prolonged, expensive, intrusive, and potentially dangerous gonadotropin ovarian stimulation. The rates of fertilization and embryo cleavage were similar whether or not the oocytes were matured in vitro or in vivo.
We found the implantation rate of embryos derived from in vitro–matured oocytes to be significantly lower than that of IVF embryos. From our data it is not possible to determine whether the lower implantation rate of IVM embryos is a result of reduced oocyte potential, reduced endometrial receptivity, or a combination of both factors.
Explanations for reduced in vitro–matured oocyte developmental potential could include suboptimal culture conditions or defective oocytes due to inadequate cytoplasmic maturation.17 Both nuclear and cytoplasmic maturation, which involve a complex cascade of events, need to be closely integrated to ensure developmental competence. It is possible that, in in vitro–matured oocytes, nuclear maturation may be complete, as evidenced by extrusion of the first polar body, whereas cytoplasmic maturation is incomplete. Previous data suggest that, compared with in vivo–matured oocytes, those matured in vitro have a reduced embryo development rate, with increased blockage of cleavage at the zygote stage.18 Our data do not support this finding. However, we did not attempt to culture IVM embryos to the blastocyst stage during the study period, and so cannot comment on developmental blockage beyond day 3 of embryo culture. In the IVF study group blastocysts were transferred in 11 cycles.
We have recently shown that the pregnancy rate in IVM treatment is related to uterine factors such as endometrial thickness on the day of embryo transfer (Child TJ, Gulekli B, Sylvestre C, Tan SL, unpublished data). This supports data from the majority of IVF studies.19 In IVM treatment, endometrial priming using oral estrogen is commenced from the day of oocyte retrieval, because administration from earlier in the cycle may be deleterious to outcome.11 Because follicular puncture for oocyte retrieval is performed before dominant follicle development, the endometrium is exposed to low levels of estradiol until priming is commenced. Consequently, the endometrium is exposed to raised serum estradiol concentrations for only a few days before embryo transfer. This contrasts with the situation during standard IVF where supraphysiologic estradiol concentrations exist for 10 days or more before embryo transfer. If the lower implantation rate in IVM is indeed partly related to endometrial receptivity, then improvements in IVM endometrial priming regimens through pharmacological or other methods may lead to improved outcomes.
We routinely administered 10,000 IU of hCG 36 hours before immature oocyte retrieval to all women undergoing IVM. This is based on data from a trial that we performed in which women were randomized to be primed with 10,000 IU of hCG before the retrieval or not to be primed.6 In no cycles were mature metaphase II oocytes retrieved. However, hCG priming significantly increased the numbers of oocytes matured at 24 hours (78.2% versus 4.9%) and 48 hours (85.2% versus 68.0%) of culture. This study was based on the observation that immature oocytes retrieved from small follicles (less than 10 mm diameter) in women undergoing ovarian stimulation respond to hCG. More recently we have found that similarly high rates of oocyte maturation are obtained when hCG priming is used in women undergoing IVM who have normal ovaries, ovulatory polycystic ovaries, or anovulatory PCOS.4
Maturation media containing FSH significantly increase fertilization and early embryo development and consequently are routinely used in published IVM studies.7 We have achieved good results using Humegon, containing 75 mIU/mL each of FSH and LH, in the culture media.6 Extremely low fertilization rates are usually obtained after standard insemination of in vitro–matured oocytes, suggesting that intracytoplasmic sperm injection is the best option, even when the sperm parameters are not impaired.20 Qualitative changes, including zona hardening, occur in the zona pellucida during oocyte maturation in vitro and may reduce the fertilization rates using conventional IVF.21 We therefore routinely used intracytoplasmic sperm injection for fertilization of in vitro–matured oocytes.
In our age‐matched population the rates of pregnancy and clinical pregnancy per cycle with IVM treatment were 26.2% and 21.5%, respectively, compared with 38.3% and 33.7% for IVF. Similar proportions of cycles with positive serum pregnancy tests in both treatment groups continued on to live birth or an ongoing pregnancy (IVM 61%, IVF 68% [nonsignificant]). Although success rates appear lower, IVM treatment costs less than IVF because less monitoring, both hormonal and ultrasonographic, is needed and no FSH is required. On average, 31.4 ± 11.1 ampoules of FSH were administered per IVF cycle. During an IVM cycle only five clinic visits are needed—namely, two ultrasound scans, the oocyte collection, embryo transfer, and serum pregnancy test—resulting in greater treatment simplicity and less disruption to patients' lives.
The potential risks of ovarian stimulation, particularly for women with polycystic ovaries, are illustrated by the fact that 12 of the IVF cycles (11.2%) resulted in moderate or severe ovarian hyperstimulation syndrome as per the classification of Golan et al.16 This rate of ovarian hyperstimulation syndrome at first seems particularly high. However, it should be recognized that all of the women had polycystic ovaries and were young, with an average age of 33.1 years. Both of these factors significantly increase the risk of developing ovarian hyperstimulation syndrome. Indeed, two previous studies examining the outcome of IVF in women with polycystic ovaries reported rates of moderate or severe ovarian hyperstimulation syndrome that were very similar to those in the present study: 10.5%1 and 10.3%.22 Although five women in the current study were classified as having severe ovarian hyperstimulation syndrome, only one required hospital admission, and another had drainage of her ascites as an outpatient procedure.
From 107 IVM cycles, 25 infants were born (seven multiple live births), and from an equal number of IVF cycles, 39 infants (ten multiple live births) were delivered. There was one set of triplets born in each treatment group. There were no reported cases of congenital abnormality in the infants. It is vital that data be collected on infants born after IVM treatment to allow full assessment of this new technology. We are currently engaged, in collaboration with pediatricians, in a prospective trial assessing the normality and neurodevelopmental development of IVM children conceived in our center. Results will be reported in due course, but at present IVM infants appear no different than those conceived through IVF.
1. MacDougall MJ, Tan SL, Balen A, Jacobs HS. A controlled study comparing patients with and without polycystic ovaries undergoing in-vitro fertilization. Hum Reprod 1993;8:233–7.
2. Whittemore AS. The risk of ovarian cancer after treatment for infertility. N Engl J Med 1994;331:805–6.
3. Edwards RG, Lobo RA, Bouchard P. Why delay the obvious need for milder forms of ovarian stimulation? Hum Reprod 1997;12:399–401.
4. Child TJ, Abdul-Jalil AK, Gulekli B, Tan SL. In-vitro maturation of immature oocytes from infertile women with normal ovaries, polycystic ovaries, or polycystic ovarian syndrome. Fertil Steril 2001;76:936–42.
5. Chian RC, Gulekli B, Buckett WM, Tan SL. Priming with human chorionic gonadotropin before retrieval of immature oocytes in women with infertility due to the polycystic ovary syndrome. N Engl J Med 1999;341:1624–6.
6. Chian RC, Buckett WM, Tulandi T, Tan SL. Prospective randomized study of human chorionic gonadotrophin priming before immature oocyte retrieval from unstimulated women with polycystic ovarian syndrome. Hum Reprod 2000;15:165–70.
7. Cha KY, Han SY, Chung HM, Choi DH, Lim JM, Lee WS, et al. Pregnancies and deliveries after in vitro maturation culture followed by in vitro fertilization and embryo transfer without stimulation in women with polycystic ovary syndrome. Fertil Steril 2000;73:978–83.
8. Mikkelsen AL, Smith S, Lindenberg S. Impact of oestradiol and inhibin A concentrations on pregnancy rate in in-vitro oocyte maturation. Hum Reprod 2000;15:1685–90.
9. Buckett WM, Bouzayen R, Watkin KL, Tulandi T, Tan SL. Ovarian stromal echogenicity in women with normal and polycystic ovaries. Hum Reprod 1999;14:618–21.
10. Kousta E, White DM, Cela E, McCarthy MI, Franks S. The prevalence of polycystic ovaries in women with infertility. Hum Reprod 1999;14:2720–3.
11. Russell JB. Immature oocyte retrieval combined with in-vitro oocyte maturation. Hum Reprod 1998;13 Suppl 3:63–70.
12. Cobo AC, Requena A, Neuspiller F, Aragon M, Mercader A, Navarro J, et al. Maturation in vitro of human oocytes from unstimulated cycles: Selection of the optimal day for ovum retrieval based on follicular size. Hum Reprod 1999; 14:1864–8.
13. Steer CV, Mills CL, Tan SL, Campbell S, Edwards RG. The cumulative embryo score: A predictive embryo scoring technique to select the optimal number of embryos to transfer in an in-vitro fertilization and embryo transfer programme. Hum Reprod 1992;7:117–9.
14. Biljan MM, Mahutte NG, Dean N, Hemmings R, Bissonnette F, Tan SL. Effects of pretreatment with an oral contraceptive on the time required to achieve pituitary suppression with gonadotropin-releasing hormone analogues and on subsequent implantation and pregnancy rates. Fertil Steril 1998;70:1063–9.
15. Engmann L, Maconochie N, Bekir J, Tan SL. Progestogen therapy during pituitary desensitization with gonadotropin-releasing hormone agonist prevents functional ovarian cyst formation: A prospective, randomized study. Am J Obstet Gynecol 1999;181:576–82.
16. Golan A, Ron-El R, Herman A, Soffer Y, Weinraub Z, Caspi E. Ovarian hyperstimulation syndrome: An update review. Obstet Gynecol Surv 1989;44:430–40.
17. Moor RM, Dai Y, Lee C, Fulka J Jr. Oocyte maturation and embryonic failure. Hum Reprod Update 1998;4:223–36.
18. Barnes FL, Kausche A, Tiglias J, Wood C, Wilton L, Trounson A. Production of embryos from in vitro-matured primary human oocytes. Fertil Steril 1996;65:1151–6.
19. Friedler S, Schenker JG, Herman A, Lewin A. The role of ultrasonography in the evaluation of endometrial receptivity following assisted reproductive treatments: A critical review. Hum Reprod Update 1996;2:323–35.
20. Hwang JL, Lin YH, Tsai YL. In vitro maturation and fertilization of immature oocytes: A comparative study of fertilization techniques. J Assist Reprod Genet 2000;17:39–43.
21. Choi TS, Mori M, Kohmoto K, Shoda Y. Beneficial effect of serum on the fertilizability of mouse oocytes matured in vitro. J Reprod Fertil 1987;79:565–8.
22. Engmann L, Maconochie N, Sladkevicius P, Bekir J, Campbell S, Tan SL. The outcome of in-vitro fertilization treatment in women with sonographic evidence of polycystic ovarian morphology. Hum Reprod 1999;14:167–71.