Immature oocyte retrieval: lessons fromunstimulated IVF cycles

Melvin H. Thornton, M.D., Mary M. Francis, B.S., and Richard J. Paulson, M.D.*

Division of Reproductive Endocrinology and Infertility, Department of Obstetrics and Gynecology, Universityof Southern California School of Medicine, Los Angeles, California

Objective: To describe retrieval of immature oocytes during unstimulated IVF and assess the in vitro maturation andfertilization rates.

Design: Retrospective analysis.

Setting: The USC program for assisted reproduction.

Patient(s): Spontaneously ovulatory women with predominantly pelvic factor as their principal cause of infertility, underthe age of 40, and no male factor.

Intervention(s): HCG administration in mid-cycle, aspiration of all visible follicles, in vitro maturation of immatureoocytes in culture media versus 50% follicular fluid in media, IVF, and embryo transfer.

Main Outcome Measure(s): Rates of in vitro maturation, fertilization, and implantation after embryo transfer.

Result(s): A total of 101 immature oocytes were obtained during 59 follicle aspirations. Thirty percent of prophase Ioocytesmatured to metaphase II in vitro compared with 44% of metaphase I oocytes. Fertilization rates for matured prophase Ioocytes were 62% for those cultured in standard culture media (controls) and 87% for follicular fluid culture media. Twopregnancies resulted from the transfer of embryos derived from immature oocytes when no other embryos were transferred.

Conclusion(s): Immature oocytes may be retrieved successfully during the mid-cycle aspiration of the dominant follicle inunstimulated IVF cycles. Maturation of immature oocytes in vitro with follicular fluid results in similar maturation andfertilization rates as for control incubation. Immature oocytes thus retrieved contribute to the overall pregnancy success ofunstimulated IVF cycles. It may be better to retrieve immature oocytes during unstimulated cycles than during the follicularphase of natural cycles. (Fertil Sterilt 1998;70:647–50. ©1998 by American Society for Reproductive Medicine.)

Key Words: Immature oocytes, unstimulated IVF, infertility, maturation, pregnancy

The first successful human pregnancy fromIVF (1) was achieved with eggs from naturalovulatory cycles, and the first successful seriesof patients conceived after IVF (2) with eggsretrieved from natural ovulatory cycles. Cur-rently, most IVF programs use controlled ovar-ian hyperstimulation (COH) during the follic-ular phase of the cycle. Among the advantagesof COH are the increased numbers of oocytes(3) and thus the increased number of embryosthat are available for embryo transfer and theenhanced cycle control that is achievable, es-pecially with the concurrent use of the GnRHanalogs (4). However, COH increases the costof the cycle and is associated with decreasedendometrial receptivity (5).

Recently, there has been a rekindling ofinterest in IVF in the unstimulated cycle (6–9),induced partly by the simplicity of transvaginalfollicle aspiration along with improvements in

laboratory techniques that make the likelihoodof achieving fertilization and embryo cleavagehigh enough to justify this type of approach.

The advantage of the unstimulated cycle isits simplicity and decreased cost. In addition,the endometrium is spared the adverse effectsassociated with COH. The disadvantage of un-stimulated IVF cycles lie in the decreased num-ber of oocytes available, which leads to a de-crease in the number of embryos available fortransfer, thus decreasing the per cycle preg-nancy rate. However, a life-table analysis ofour data suggested that per cycle rates do notdecline during successive cycles and thus adecrease in the per cycle success rate may beoffset when more cycles are performed (10).Thus, unstimulated IVF appears to be a clini-cally viable alternative to stimulated cycles inwomen who are under the age of 40 and whosepartners do not have male factor infertility.

Received November 26,1997; revised andaccepted May 28, 1998.Reprint requests: RichardJ. Paulson, M.D.,Department of Obstetricsand Gynecology, Women’sand Children’s Hospital,Room 1M2, 1240 NorthMission Road, LosAngeles, California 90033(FAX: 323-226-2850).

FERTILITY AND STERILITY tVOL. 70, NO. 4, OCTOBER 1998Copyright ©1998 American Society for Reproductive MedicinePublished by Elsevier Science Inc.Printed on acid-free paper in U.S.A.

0015-0282/98/$19.00PII S0015-0282(98)00274-X

647

The retrieval of immature oocytes during the follicularphase of natural cycles with subsequent maturation andfertilization in vitro has been demonstrated (11). We notedthat in the course of follicle aspiration during IVF in un-stimulated cycles in which a single dominant follicle isstimulated with hCG during the mid-cycle, secondary im-mature oocytes are obtained frequently (10). These second-ary oocytes may participate in pregnancy success of thesecycles (10, 12). Therefore, given the pregnancy potential ofimmature oocytes, it seemed plausible that the unstimulatedIVF model may represent the optimal setting for their re-trieval. The purpose of this study was to analyze the resultsof the in vitro maturation of immature secondary oocytesobtained at the time of follicle aspiration during unstimu-lated IVF cycles.

MATERIALS AND METHODS

We analyzed 123 consecutive unstimulated IVF cyclesperformed at our institution during the years 1991–1994. Theclinical outcome of most of these cycles has been describedpreviously (10). The protocol for unstimulated IVF wasapproved by the Institutional Review Board of the CaliforniaHospital Medical Center, Los Angeles, California.

In all instances, spontaneous ovulatory status was docu-mented by a combination of clinical parameters (folliculardevelopment by ultrasound (US) examination, mid-cycle LHsecretion by urinary assays, and/or determination of mid-luteal ovulatory serum P concentrations). Serial transvaginalultrasonograms were obtained and serum E2 levels weredetermined for all patients 4 days before the expected day ofovulation. Concentrations of E2 were measured by a com-mercially available extraction assay (Pantex, Santa Monica,CA).

Ovulation was triggered with hCG (10,000 IU) given IMwhen follicle maturity was achieved, as described previously(12). If an LH surge was detected in urine collected beforethe hCG injection, the cycle was canceled, otherwise follicleaspiration was scheduled for 34–36 hours after hCG admin-istration.

Aspirations were performed in an outpatient setting underIV conscious sedation. After the aspiration of the dominantfollicle, an attempt was made to aspirate all visible second-ary follicles. These were 12 mm or less in diameter in allinstances. Because several follicles were aspirated during asingle pass with the needle, the exact size of follicles yield-ing immature oocytes could not be determined. A standard(long-beveled) 17-gauge aspiration needle (Cook Ob/Gyn,Spencer, IN) was used for all aspirations.

Oocyte maturity was estimated by morphologic assess-ment of the corona-cumulus complex using high magnifica-tion on an inverted microscope. All oocytes from dominantfollicles were inseminated 4–6 hours after aspiration. Im-mature oocytes were divided into two categories by matu-

rity: those that were judged to be metaphase I and those witha visible germinal vesicle (prophase I). Metaphase I oocyteswere inseminated at the same time as the dominant oocytes,whereas prophase I oocytes were kept in culture.

Prophase I oocytes were further divided into two groups:the study group was cultured in 50% follicular fluid asdescribed by Cha et al. (13), whereas the control group waskept in standard culture medium alone. After 24 hours inculture, prophase I oocytes were mechanically separatedfrom the corona-cumulus complex and observed every 24hours for up to 5 days. When a polar body was observed, theoocyte was inseminated in standard fashion.

Fertilization was defined as the presence of two or morepronuclei and the attainment of oocyte maturity was definedas the presence of a polar body. Oocytes which fertilizedwithin 48 hours of follicle aspiration were transferred duringthe unstimulated IVF cycle. Oocytes fertilizing later than thistime were judged to be out of synchrony with the window ofimplantation and were cryopreserved for transfer duringsubsequent cycles.

All cleaving embryos were transferred by the transcervi-cal route in standard fashion in a volume of 10–15mL.Progesterone supplementation was given in the form ofvaginal P suppositories (50 mg twice daily) beginning on theday of embryo transfer. Pregnancies were confirmed withrising b-hCG concentrations and the presence of an intra-uterine sac documented by US examination. Statistical sig-nificance was determined byx2 analysis.

RESULTS

In 59 of the 123 cases (48%), a total of 101 immatureoocytes were obtained (mean 0.7, range, 1–8). Morphologicassessment of the corona-cumulus complex revealed 55cases to be prophase I and 56 cases to be metaphase I.Twenty-five of the prophase I oocytes were kept in standardculture medium (controls), whereas 30 were matured with50% follicular fluid from the dominant oocyte.

Of the 25 prophase I control cycles, 8 matured (32%) and5 fertilized (62%), with one polypronuclear fertilization(20%). Of the 30 prophase I cultured in follicular fluid, 9matured (30%) and 7 fertilized (77%,P 5 not significant[NS] versus control oocytes) with two polypronuclear fertil-izations (28%,P 5 NS versus control oocytes) (Table 1). Ofthe 56 metaphase I oocytes, 20 fertilized (35%), of whichtwo were polypronuclear (10%). Another five oocytes werenoted to have polar bodies, implying that 25 (44%) of the 56matured to metaphase II within the first 24 hours afteraspiration. Overall, 17 of the 55 prophase I oocytes maturedin vitro (30%) compared with 25 of the 56 metaphase Ioocytes matured in vitro (44%) (P 5 0.19).

The contribution of immature oocytes to implantationcould only be calculated in cases in which embryos resulting

648 Thornton et al. Immature oocytes from unstimulated IVF Vol. 70, No. 4, October 1998

from the dominant oocytes were not transferred. Only onesuch transfer took place during a fresh cycle because ofpolypronuclear fertilization of the dominant oocyte (becauseall of the dominant oocytes had fertilized), and it resulted inone live birth. Nine immature oocytes were fertilized 72hours after aspiration and were cryopreserved at the twopronuclear stage. Six such embryo transfers resulted in onelive birth (17%).

DISCUSSION

Recently, there has been a renewed interest in immatureoocyte retrieval. Trounson et al. (11) described the recoveryof immature oocytes during the follicular phase of sponta-neous ovulatory cycles and during anovulatory cycles inpatients with polycystic ovaries. These oocytes have beenshown to be capable of reaching nuclear maturity after36–48 hours of culture in media containing gonadotropins.Recently, it has been suggested that estrogen stimulation ofthe endometrium may enhance the success of immatureoocyte retrieval (14).

However, these investigators performed retrievals in theearly follicular phase and did not pursue immature oocytesafter a dominant follicle was selected (14). After the domi-nant follicle has been selected in a spontaneous ovulatorycycle, secondary follicles are generally thought to be atreticwith only the dominant follicle possessing the appropriatehormonal milieu necessary to achieve oocyte maturity andovulation (15). Whereas immature oocytes are commonlyretrieved during stimulated cycles (16), no prior series hasaddressed the issue of immature oocytes retrieved at the timeof ovulation in spontaneous cycles.

Unstimulated IVF cycles differ from purely natural cyclesin that hCG is added in the mid-cycle to facilitate schedulingof oocyte retrieval. Therefore, all of the immature oocytes inthe present study were retrieved from secondary follicleswhich were exposed to hCG for 34–36 hours. Our findingssuggest that whereas secondary follicles may be atretic, the

oocytes within them have not necessarily undergone demiseor apoptosis and may be matured in vitro. Furthermore, onlyabout one-half of the immature oocytes that were obtainedcontained germinal vesicles. However, nuclear maturity ofthe oocyte takes place only after the onset of LH stimulationof the follicle, with a subsequent decrease in intra-follicularoocyte maturation inhibitor (17).

Our results suggest that some secondary follicles developsufficient follicle maturity to respond to the LH surge byinducing (at least partially) oocyte maturity. Conversely, theobservation that some of the oocytes had germinal vesiclessuggests that the number of LH receptors, or the ability torespond to LH-like stimulation, was nonetheless limited.

We evaluated the usefulness of the addition of 50%follicular fluid from the dominant follicle to the media con-taining oocytes with germinal vesicles. This follicular fluid isexpected to contain high steroid levels and hCG in additionto any other putative intrafollicular factors which mightenhance oocyte maturity. This approach was used by previ-ous investigators who either used 50% follicular fluid (13) oradded gonadotropins to culture media (11). However, ourdata showed no difference in maturity attainment. Therefore,follicular fluid may not be useful in the enhancement ofoocyte maturity of germinal vesicle oocytes obtained inunstimulated cycles.

We performed follicle aspirations of secondary follicleswith a standard needle and standard suction pressure used forstimulated cycles. This is in contrast with the method de-scribed by Trounson et al. (11), who used a needle with ashort bevel and lower suction pressure. It is tempting tospeculate that the number of immature oocytes retrievedduring unstimulated IVF could be increased with the use ofsuch specialized equipment.

For immature oocyte retrieval, the principal advantage ofunstimulated IVF is the retrieval of the dominant oocyte andnatural endometrial priming, which obviates the need forexogenous endometrial priming (14). Therefore, at least oneembryo is likely to be obtained in at least 85% of cases (10).Conversely, oocytes that mature after a delay of more than24 hours, probably cannot be transferred fresh because of theluteinization of the endometrium and consequent dyssyn-chrony with the resulting embryo (18). We believe thatsuch embryos probably are best cryopreserved and trans-ferred in a subsequent cycle as described in the presentstudy.

In summary, immature oocytes can be retrieved success-fully during the mid-cycle aspiration of a dominant oocyteduring unstimulated IVF. Maturation of these immature oo-cytes in vitro with follicular fluid results in similar matura-tion and fertilization rates as those for control incubationwith standard culture medium. The immature oocytes re-trieved contribute to the pregnancy success of unstimulatedIVF cycles. Finally, it may be better to retrieve the immature

T A B L E 1

In vitro maturation and fertilization of immature oocytesobtained during unstimulated IVF cycles.

Morphologicassessment

No. ofoocytes

Maturationin vitro Fertilization

Polypronuclearfertilization

Prophase I oocytes(controls) 25 8 (32%) 5 (62%) 1 (20%)

(20% of n)Prophase I oocytes

(follicular fluid) 30 9 (30%) 7 (77%) 2 (28%)(23% of n)

Metaphase I oocytes 56 25 (44%) 20 (80%) 2 (10%)(35% of n)

FERTILITY & STERILITY t 649

oocytes during unstimulated cycles than during the follicularphase of natural cycle because the dominant oocyte repre-sents at least one mature oocyte during each aspiration.

References1. Steptoe PC, Edwards RG. Successful birth after IVF. Lancet 1978;II:

366.2. Edwards RG, Steptoe PC, Purdy JM. Establishing full-term human

pregnancies using cleaving embryos in vitro. Br J Obstet Gynaecol1980;87:737–56.

3. Paulson RJ, Marrs RP. Ovulation stimulation and monitoring for in-vitro fertilization. Curr Probl Obstet Gynecol Fertil 1986;9:497–526.

4. Meldrum DR, Wisot A, Hamilton F, Gutlay AL, Kempton WF, HuynhD. Routine pituitary suppression with leuprolide before ovarian stimu-lation for oocyte retrieval. Fertil Steril 1989;51:455–9.

5. Paulson RJ, Sauer MV, Lobo RA. Embryo implantation after human invitro fertilization: importance of endometrial receptivity. Fertil Steril1990;53:870–4.

6. Ranoux C, Foulot H, Dubuisson JB, Rambaud D, Aubriot FX, Poirot C,et al. Returning to spontaneous cycles in in vitro fertilization. J In VitroFert Embryo Transf 1988;5:304–5.

7. Paulson RJ, Sauer MV, Francis MM, Macaso TM, Lobo RA. In vitrofertilization in unstimulated cycles: a clinical trial utilizing hCG intiming of follicle aspiration. Obstet Gynecol 1990;76:788–91.

8. Patton PE. IVF in the unstimulated cycle. Contemp Obstet Gynecol1992;36:67–74.

9. Fahy UM, Cahill DJ, Wardle PG, Hull MGR. In-vitro fertilization incompletely natural cycles. Hum Reprod 1995;10:572–5.

10. Paulson RJ, Sauer MV, Francis MM, Macaso TM, Lobo RA. In vitro

fertilization in unstimulated cycles: the University of Southern Califor-nia experience. Fertil Steril 1992;57:290–3.

11. Trounson A, Wood C, Kausche A. In vitro maturation and the fertili-zation and developmental competence of oocytes recovered from un-treated polycystic ovarian patients. Fertil Steril 1994;62:353–62.

12. Paulson RJ, Sauer MV, Francis MM, Macaso TM, Lobo RA. Factorsaffecting pregnancy success of human in-vitro fertilization in unstimu-lated cycles. Hum Reprod 1994;9:1571–5.

13. Cha KY, Koo JJ, Choi DH, Hans SY, Yoon TK. Pregnancy after invitro fertilization of human follicular oocytes collected from nonstimu-lated cycles, their culture in vitro and their transfer in a donor oocyteprogram. Fertil Steril 1991;55:109–13.

14. Russell JB, Knezevich KM, Fabian KF, Dickson JA. Unstimulatedimmature oocyte retrieval: early versus midfollicular endometrial prim-ing. Fertil Steril 1997;67:616–20.

15. Hodgen GD. The dominant ovarian follicle. In: Wallach EE, KempersRD, editors. Modern trends in infertility and conception control. Chi-cago: Year Book Medical Publishers, Inc., 1985;3:26–45.

16. Veeck LL, Wortham JW Jr, Witmyer J, Sandow BA, Acosta AA,Garcia JE, et al. Maturation and fertilization of morphologically imma-ture human oocytes in a program of in vitro fertilization. Fertil Steril1983;39:594–602.

17. Paulson RJ. Oocytes from development to fertilization. In: Lobo RA,Mishell DR Jr, Paulson RJ, Shoupe D, editors. Infertility, contraceptionand reproductive endocrinology. 4th ed. Cambridge: Blackwell Scien-tific Publications, 1997:174–82.

18. Paulson RJ, Sauer MV, Lobo RA. Potential enhancement of endome-trial receptivity in cycles utilizing controlled ovarian hyperstimulationwith antiprogestins: a hypothesis. Fertil Steril 1997;67:321–5.

650 Thornton et al. Immature oocytes from unstimulated IVF Vol. 70, No. 4, October 1998