The predictive value of idiopathic failure to fertilize on the first in vitro fertilization attempt*

Transcription

1 FERTILITY AND STERILITY Copyright <> 1991 The American Fertility Society Printed on acid-free paper in U.S.A. The predictive value of idiopathic failure to fertilize on the first in vitro fertilization attempt* David Molloy, M.D.t Keith Harrison, M.App.Sc. Terry Breen, B.Sc. John Hennessey, M.D. Queensland Fertility Group, Watkins Medical Centre, Brisbane, Queensland, Australia Objective: To investigate the subsequent performance of patients with idiopathic fertilization failure on the first in vitro fertilization (IVF) cycle. Design: A retrospective study of 2,322 consecutive patients undergoing their initial IVF cycle. Setting: Advanced infertility treatment in an IVF/general infertility clinic. Patients: In 5 years, 94 couples with unexplained failed oocyte fertilization had 270 cycles of treatment. Each couple's performance was tracked through subsequent cycles of treatment. Interventions: In vitro fertilization with husband and donor sperm. Main Outcome Measure(s): Investigated retrospectively after 5 years of data collection. Results: Sixty-five couples of the original94 had a second IVF attempt. Fifty of these successfully fertilized oocytes with husband's sperm and 4 with donor sperm. Nineteen of the 65 couples who continued treatment achieved a pregnancy, and only one couple had continuing fertilization failure. Conclusions: The prognosis in the study group was surprisingly favorable despite the initial failed IVF treatment cycle. Fertil Steril 56:285, 1991 In vitro fertilization (IVF) units may accept that 10% to 20% of couples with successful oocyte retrieval will not have an embryo transfer (ET) because of failed fertilization. 1 2 This is often because of an etiologic factor such as oligospermia, and completely inexplicable fertilization failure is relatively rare. When this does occur in a first IVF treatment cycle, it can have devastating psychological effects by seriously questioning a couple's ability to ever conceive a child. A review of our unit's results over the previous 5 years suggests that true idiopathic fertilization failure is indeed uncommon. This study Received September 12, 1990; revised and accepted April 5, * Presented at the 9th Annual Meeting, Fertility Society of Australia, Perth, Australia, September 26 to 29, t Reprint requests: David Molloy, M.D., Queensland Fertility Group, 1st Floor, Watkins Medical Centre, 225 Wickham Terrace, Brisbane, Queensland, Australia examines the outcome of subsequent treatment cycles to provide an assessment of reproductive potential that can be used in the management of patients with a similar outcome in the first IVF cycle. MATERIALS AND METHODS From January 1985 to January 1990, 2,322 patients had initial cycles of IVF treatment. Three hundred seventy-eight patients had first treatment cycles in which oocyte fertilization failed to occur (16%). Of these, 94 patients (25%) had fertilization failure in which no specific oocyte or sperm parameter could be identified as the reason. All these patients had at least 2 years of infertility with their current partner. They had been extensively investigated with laparoscopy, endometrial sampling, tests of ovulation, postcoital testing, and, in some cases, hysterosalpingography. Previous semen analysis was normal as defined by the World Molloy et al. IVF idiopathic fertilization failure 285

2 Table 1 Age Versus Oocyte Number on First Attempt Oocyte number No. of cases Mean age (y) Median age (y) Age range (y) 26 to to to 44 Health Organization criteria 3 as was the semen sample presented for use in the IVF cycle. Couples with first cycle fertilization failure and a past history of abnormal semen parameters or antisperm autoantibodies were excluded from the study. Patients were also excluded if the semen sample used for the treatment cycle had any abnormalities, even if prior analyses were normal. Etiologic factors were therefore restricted to tubal damage (53 couples), endometriosis (16), and idiopathic infertility (28). Three patients had multiple etiologies with endometriosis and tubal problems. The ages of the patients are presented in Table 1. All were hyperstimulated using clomiphene citrate 100 mg/d for 5 days followed by human menopausal gonadotropin injections until ovulation. Hormonal monitoring with estradiol (E 2), luteinizing hormone (LH), and progesterone was used to define oocyte maturation and release. First cycle oocyte retrieval was LH surge-timed in 83 cases and human chorionic gonadotropin-timed in 11 cases. An average of three oocytes/patient was retrieved. Laboratory IVF procedures were as previously described. 4 5 At least three oocyte examinations were performed on sequential days to establish failed fertilization. All batches of media involved were mouse embryo tested. Cancellation criteria in our program depend on a variety of factors. Obviously, poor response to stimulation as defined by E 2 or ultrasound (US) monitoring is most important, but this is modified by such factors as the age of the patient, previous gonadotropin response in non-ivf cycles, and factors of patient convenience. Our single ET pregnancy rate (PR) per cycle is 10%, so the prospect of retrieving less than three oocytes is not necessarily an indication for cancellation. Patients were counseled after the first failed IVF cycle and offered further treatment. Diagnosis of pregnancy in subsequent treatment cycles was defined on clinical grounds by US or tissue diagnosis. No biochemical pregnancies were included in the unit's statistics. RESULTS Ninety-four couples had a total of 270 cycles of IVF including the first failed cycle. Only 64 of the original94 patients sought further treatment (68%), and 54 (84%) of these successfully fertilized at least one oocyte on the next cycle. Four of these cases however, used donor sperm in the second cycle eithe; for reinsemination after the first failed oocyte check (3 cases) or a primary split insemination of oocytes between husband and donor semen. Therefore, 50 of 64 couples (78%) successfully fertilized oocytes in their second treatment cycle using only their own oocytes and sperm. Fertilization rate data are presented in Table 2 according to the number of oocytes retrieved on the first IVF attempt. Fourteen couples therefore had two sequential cycles with failed IVF fertilization using husbands' sperm, but in 4 of these cases donor sperm was used as well to prove oocyte fertilization. Of the 10 couples left, 5 had no further treatment, 3 had successful fertilization with husbands' sperm on a subsequent cycle, 1 fertilized with donor sperm, and 1 failed to fertilize on two further IVF cycles. One couple successfully fertilized an oocyte with husband's sperm in the second treatment cycle, but the low fertilization rate achieved with husband's sperm with poor embryo quality prompted a change to donor sperm in the third treatment cycle. Six couples therefore eventually opted for donor sperm, and three pregnancies occurred in this group. A total of 23 IVF pregnancies was achieved in 19 patients. Twenty of these pregnancies resulted from fertilization with husband's sperm. The clinical PR per patient for those who persisted with treatment Table 2 Oocyte Fertilization Rates No. of patients in first attempt No. of patients in second attempt No. in attempt 2 who failed to fertilize with husband's sperm Mean oocyte numbers in subsequent cycles Oocyte fertilization rate in all attempts (%) Oocyte fertilization rate excluding first attempt(%) Oocyte number in first attempt Four couples fertilized oocytes on the second attempt with donor sperm. 286 Molloy et al. IVF idiopathic fertilization failure Fertility and Sterility

3 after the first failed cycle was 30% overall and 28% for those who used only husband's sperm for subsequent treatment. The live birth rate per patient with two or more treatment cycles was 23%. Data relating to pregnancies are presented in Table 3. DISCUSSION Idiopathic first cycle fertilization failure is relatively uncommon and only represents 4% of all initial IVF cycles in the period surveyed. However, these 94 cases did account for 25% (94/378) of all couples who failed to fertilize on their primary treatment cycle. Defective sperm parameters or assay-proven antisperm autoantibodies probably explained failed fertilization in the remaining 284 couples, although some patients had obvious oocyte abnormalities, usually associated with low numbers of oocytes retrieved. Couples entering their first IVF cycle had a 16% (378/2322) chance of failed fertilization in that cycle, which may be relatively high compared with other IVF programs. Our unit, however, has an active policy of encouraging couples with a diagnosed semen problem to use IVF as a method of testing sperm fertilizing capacity. Also, since 1986, many couples with a good prognosis because of normal sperm parameters and fallopian tubes have been transferred from IVF to the gamete intrafallopian transfer program. This has led to a higher first cycle fertilization failure rate as the proportion of male infertility on the IVF program increased. Unexpected failure to fertilize would appear to affect future treatment decisions by the infertile couple. These patients were less likely to proceed with future IVF treatment, and only 68% had a second treatment cycle despite an apparently good prognosis. This is much lower than the 83% repeat cycle treatment rate for patients who successfully fertilized oocytes in their first cycle but failed to become pregnant. Fertilization failure also prompted six couples to use donor sperm, either in addition to or instead of husband's sperm in subsequent cycles. Three of these patients delivered a baby, which supports the proposition that defective sperm probably caused fertilization failure in the original cycles as suggested by other authors. 6 Results in this paper have been categorized on the basis of oocyte numbers retrieved in the initial cycle. Patients with only one or two oocytes retrieved fell into the low responder group, although the oocytes were graded average or better on microscopic Table 3 Clinical Pregnancies Oocyte number in first attempt No. of patients in first attempt No. of patients in subsequent attempts Total pregnancies Pregnancies-patient Pregnancies-husband's sperm Pregnancies-donor sperm 3 Pregnancy rate/couple continuing(%) Pregnancy rate/cycle (exclude first cycle) (%) Pregnancy rate/cycle (all cycles) (%) examination. However, some degree of oocyte abnormality is likely in this group. Previous investigation 7-9 on unfertilized oocytes produced by hyperstimulation have shown evidence of chromosomal and structural defects. Patients with low oocyte numbers were 48% of the total cases analyzed. This was disproportionate to the total number of patients in the IVF program to have only one or two oocytes retrieved (21% ). Therefore, patients with low oocyte numbers may have an increased risk of failed fertilization because of stimulation problems, submicroscopic oocyte abnormalities, or even random chance. The mean oocyte numbers in subsequent cycles for this group were significantly increased to 2.8 (median of 3) for patients with one initial oocyte retrieved and 3.8 (median of 4) for those with two oocytes on the primary cycle. This suggests an improvement in the stimulation protocol applied to those patients in subsequent cycles. The proportion of LH surge-timed oocyte retrievals is unusually high in our program and may have contributed in three cases to the low oocyte number group. Mistimed LH surge retrievals can be associated with failed oocyte retrieval or poor oocyte numbers that may fail to fertilize because of inadequate oocyte maturation. However, there were only three cases in which later cycles with hcg-timed retrieval showed a major increase in oocyte number. In two of these cases, the surge timing appeared correct on the available data, but more oocytes could have been expected on the monitoring results. This situation of fewer than expected oocytes can also occur with hcg-timed oocyte retrievals. It would seem therefore that this study targeted an increased Molloy et al. IVF idiopathic fertilization failure 287

4 number of poor responders to hyperstimulation. This is supported by the mean increase in oocyte numbers in subsequent cycles. Although mean oocyte numbers did increase, this was more because of an improvement in hyperstimulation than a change from surge timing. If the first (and failed) cycle is excluded, the oocyte fertilization rate is surprisingly constant across all three groups for subsequent cycles (51% to 66%). The total fertilization rate when all oocytes were included is lower in the greater than three oocyte group. The range of oocyte numbers aspirated in this group was 3 to 14 (median of 4). The cases with relatively high oocyte numbers in this group deflated the fertilization rate when the first cycle was included in the statistics. The fertilization rate for all three groups in subsequent cycles is consistent with that of the whole IVF program over the time frame studied. The PR is encouraging considering the poor start this group of patients made in their IVF treatments. Nineteen (30%) achieved a pregnancy if they persisted with IVF for up to four more attempts. However, 3 of these pregnancies occurred after use of donor sperm. Sixteen (28%) of 58 couples who persisted with husband's sperm were successful in subsequent cycles. These PRs were comparable for all the IVF program (31% per patient) over the same time frame. The per cycle PR overall was 8.50%, but 13% if the first cycle was excluded. The subsequent per cycle PR was therefore comparable with that of the IVF program. The subsequent PR was best in the single oocyte group and may have reflected the improvement in hyperstimulation in following cycles. In this group, 5 of the 15 patients who persisted with treatment became pregnant, all resulting in live births. Overall, only four abortions and no ectopics occurred in the pregnancy group. The reasons for fertilization failure at IVF are complex and numerous. Sperm defects are an obvious and important cause, and evidence exists 10 that even careful semen analysis may miss an occult sperm factor that contributes to fertilization failure. In our own program, couples with identified sperm problems contributed to 69% of first cycle fertilization failure. Conversely, semen analysis is still an unreliable guide to IVF prognosis with unexpected fertilization and pregnancies occurring in patients with severe oligospermia or asthenospermia. Low oocyte numbers and possible oocyte abnormalities may also be contributing factors. The random chance of failed fertilization is difficult to predict. It may be quite unlikely that normal sperm will necessarily fertilize a structurally and chromosomally perfect oocyte with a 100% chance of success. In the presence of low oocyte numbers, it is possible that fertilization failure is truly idiopathic, but this is less likely with large numbers of oocytes. Another possible cause for failed fertilization is an aberration in the IVF laboratory because of poor media or culture conditions. Every competent laboratory is aware of this and analyzes its data and checks its procedures carefully to prevent this occurrence. During the 5-year period surveyed, there were at least 2 clinical pregnancies per week from each media batch. In all, 1,146 clinical pregnancies occurred. Fertilization in other patients having IVF at the same time under the same culture conditions also occurred, enabling us to match them against the cases included in this study as a cross-check of culture conditions. Mouse embryo testing was also routine. This would suggest that laboratory conditions were an unlikely cause of this idiopathic and unpredicted fertilization failure. Couples who have unpredicted fertilization failure on their primary IVF attempt can therefore be given a good prognosis of future fertilization and a hopeful prognosis of pregnancy. Future treatment cycles are clearly justified because the second cycle failure rate is quite low and will help define couples who may need other avenues of assistance including donor sperm. Although their PR is slightly lower than the total IVF program, they have a reasonable prospect of delivering a child with adequate subsequent treatment. Acknowledgments. We thank Douglas Keeping, M.D., Warren De Ambrosis, M.D., John Hynes, M.D., and Gordon Kilvert, M.D., of the Queensland Fertility Group for access to their records and patients. REFERENCES 1. Trounson A, Wood C: In vitro fertilization results, at Monash University, Queen Victoria and Epworth Medical Centres. J In Vitro Fert Embryo Transfer 1:42, Lancaster P: IVF and GIFT pregnancies. Sydney, Australia and New Zealand 1987 National Perinatal Statistics Unit, World Health Organization: WHO Laboratory Manual for the Examination of Human Semen and Semen-Cervical Mucus Interaction. 2nd edition. Cambridge, The Press Syndicate of the University of Cambridge, 1987, p 3 4. Cummins JM, Breen TM, Harrison KL, Shaw JM, Wilson LM, Hennessey JF: A formula for scoring human embryo growth rates in in vitro fertilization: its value in predicting 288 Molloy et al. IVF idiopathic fertilization failure Fertility and Sterility

5 pregnancy and in comparison with visual estimates of embryo quality. J In Vitro Fert Embryo Transfer 3:284, Shaw JM, Harrison KL, Wilson LM, Breen TM, Shaw G, Cummins JM, Hennessey JFM: Results using medium supplemented with either fresh or frozen stored serum in human in vitro fertilization. J In Vitro Fert Embryo Transfer 4:5, Deschacht J, Devoey P, Camus M, Khan I, Smitz J, Staessen C, van Waesberghe L, Wisanta A, van Steirteghem AC: Invitro fertilization with husband and donor sperm in patients with previous fertilization failures using husband sperm. Hum Reprod 3:105, Gwatkin RB, Conaver JC, Collins RL, Quigley MM: Failed fertilization in human in-vitro fertilization analyzed with the deoxyribonucleic aid-specific fluorochrome Hoechst Am J Obstet Gynecol 160:31, Peters MH, Geraedts JP, Dumoulin JC, Evers JL, Bras M, Kamips FM, Menhoere PP: Cytogenetic analysis of in-vitro fertilization (IVF) failures. Hum Genet 81:367, Bongso A, Chye NS, Ratnam S, Sathananthan H, Wong PC: Chromosome anomalies in human oocytes failing to fertilize after insemination in-vitro. Hum Reprod 3:645, Oehninger S, Acosta AA, Kruger T, Veeck LL, Flood J, Jones HW: Failure of fertilization in in-vitro fertilization: the "occult" male factor. J In Vitro Fert Embryo Transfer 5:181, 1988 Molloy et al. IVF idiopathic fertilization failure 289