- Email: [email protected]
Reinsemination by intracytoplasmic sperm injection of 1-day-old oocytes after complete conventional fertilization failure
Vol. 68, No.3,September
FERTILITY AND STERILI~ Copyright
o 1997 American
Published
by Elsevier
Society for Reproductive
Medicine
Printed
on acid-free
paper
1997
in U. S. A.
Science Inc.
Reinsemination by intracytoplasmic sperm injection of 1 -day-old _ oocytes after complete c&v-entional fertilization failure
Paula C. Morton, B.S. Cynthia S. Yoder, B.S. Michael J. Tucker, Ph.D.* Reproductive
Biology Associates,
Graham Wright, B.Sc. Wendy D. W. Brockman, B.A. Hilton I. Kort, M.D. Atlanta,
Georgia
Objective: To analyze the efficacy of intracytoplasmic sperm injection (ICSI) for “rescue” of failed conventional insemination cycles. Design: Retrospective clinical study. Setting: Private infertility clinic. Patient(s): Fifty-four couples with non-male factor infertility undergoing routine IVF within our clinic. Intervention(s): Twenty- to 24-hour-old unfertilized mature oocytes from conventional IVF cycles in which fertilization failure was complete were reinseminated by ICSI to attempt late fertilization and rescue of the otherwise failed M? cycle. Main Outcome Measure(s): Late fertilization, pregnancy, and embryonic implantation. Result(s): Intracytoplasmic sperm injection reinsemination of 489 unfertilized mature oocytes caused degeneration in 50 (10.2%) oocytes, gave rise to normal fertilization in 215 (44.0%) of the injected oocytes, and allowed ET in 48 of the 54 cycles in which initial complete fertilization failure had occurred. One hundred sixty-four (76.3%) of 215 late-fertilized oocytes either were used for fresh transfer or were frozen-stored. Eight viable pregnancies resulted, yielding a 14.8% pregnancy rate per initiated cycle. Conclusion(s): Barring potential concerns regarding the chromosomal normality of embryos arising from reinsemination, our results suggest that ICSI is a relatively successful means of rescuing conventional IVF cycles in which fertilization fails completely. (Fertil Sterile 1997;68:488-91. 0 1997 by American Society for Reproductive Medicine.) Key Words: Fertilization failure, reinsemination, ICSI, embryonic implantation, pregnancy
In non-male factor infertility IW cases, complete fertilization failure still may occur because of a variety of factors, including oocyte quality and cryptic spermatozoa1 defects. Such an event does not always occur consistently in a couple undergoing IVF, but regardless of frequency, it can be psychologically devastating for such a couple to undertake IVF’and not achieve ET because of unexpected conventional fertilization failure. Consequently, reinsemination of unfertilized oocytes in such cases can be carried out to try to achieve some late fertilization, that is, to “rescue” what otherwise would be a completely failed attempt at pregnancy. ReceivedDecember6, 1996; revisedand acceptedMay 2, 1997. * Reprintrequests:MichaelJ. Tucker,Ph.D.,ReproductiveBiology Associates, 5505 Peachtree, Dunwoody Road, Suite 400, Atlanta, 488
Georgia 30342 (FAX: 404-256-1912).
Conventional reinsemination traditionally has yielded poor outcomes in terms of both fertilization and embryonic implantation (l-3). More recently, however, micromanipulation has improved the potential to achieve late fertilization after reinsemination, though not necessarily pregnancy (4-6). Extreme forms of male factor infertility are obvious cases that justify the use of ICSI as the means of insemination in IW. However, there exist few other clear rules to establish when to choose ICSI over conventional insemination (7). Consistent, repeated failed conventional fertilization failure can be good reason to apply ICSI, but such failure is difficult to anticipate. If ICSI replaced conventional IVF insemination completely, any concerns over when ICSI should be applied would be obviated, although the question has been raised whether ICSI is justifiable when it might generate higher numbers 0015-0282/97/$17.00 PI1 s0015-0282(97)00223-9
of abnormal offspring and also cost more. If ICSI reinsemination was an effective means to rescue failed conventional IVF cycles, then automatic application of ICSI would not be needed; ICSI would be used only in those questionable cases in which conventional insemination initially has failed. In an attempt to determine the clinical efficacy of this strategy, we retrospectively reviewed all cases in the past 3.5 years in which ICSI reinsemination was used after total conventional fertilization failure occurred. MATElRIALS AND METHODS
From January 1993 to July 1996,54 couples who experienced complete fertilization failure after initial conventional insemination consented to the use of reinsemination by ICSI. Women underwent ovarian stimulation using either a long- (n = 32) or short- (n = 16) protocol leuprolide acetate regime coupled with FSH. Over six cycles, clomiphene citrate was given alone with minimal monitoring. Ovulation was initiated using hCG, 10,000 IU, in all cases, with ultrasound-guided oocyte retrieval performed 34 to 36 hours after hCG administration. In all cases there was no significant marker of overall oocyte quality that seemed awry. Primary causes of infertility in all cases excluded male factor but incorporated a wide range of female diagnoses, including tubal factor, endometriosis, pelvic adhesions, and ovulatory dysfunction. Women were between 28 and 43 years old (mean t SD, 35.3 ? 4.3 years). A total of 536 oocytes were collected initially, none of which became fertilized after conventional insemination. All semen samples were processed through 90% Percoll columns followed by three washes in Earle’s balanced salt solution (EBSS) with 6% Plasmatein (Alpha Therapeutic Corp., Los Angeles, CA). The final sperm suspension was adjusted to 5 to 20 x lo6 sperm/ml and was left at room temperature (20 to 23°C) until 1 hour before insemination, when the sample was placed into a 37”C, 5% COz in air, humidified atmosphere. Conventional insemination was carried out in microdroplets of EBSS + 6% Plasmatein under mineral oil. The sperm concentration was adjusted to a concentration between 100,000 and 3OO,OOO/mL, depending on the percentage of progressive motility. Sixteen to 18 hours later, oocytes were cleaned carefully to remove any remaining cumulus and coronal cells and were checked for the presence of pronuclei (PN) on an inverted microscope fitted with Nomarski differential interference contrast optics (Olympus IMT-2 or IX-70; Olympus, Tokyo, Japan). Three related features of fertilization were assessed: Vol. 68, No. 3, September
1997
presence of PN (with nucleoli apparent), cytoplasmic contraction, and extrusion of a second polar body. If none of these features were present, the oocyte was deemed unfertilized and was washed free of sperm. All oocytes possessed either one polar body or fragmented polar bodies, and these oocytes were considered unfertilized once it was clear that there were no other related features of fertilization. Sperm suspensions prepared from the previous day of oocyte retrieval were stored at room temperature overnight. Sperm for reinsemination was taken from this stock suspension. As soon as total conventional fertilization failure was established with all oocytes, reinsemination by ICSI was undertaken as described previously (4). Therefore, this procedure was carried out between 20 and 24 hours after the initial oocyte retrieval to minimize the effects of gamete aging. In all cases, some sperm motility remained, and such sperm were selected preferentially for injection into the unfertilized oocytes. Some unfertilized oocytes (n = 47) were rejected for reinsemination on the grounds that they either were significantly dysmorphic or were immature. After ICSI, all reinseminated oocytes were placed into EBSS + 15% maternal serum and were checked for signs of fertilization between 6 and 10 hours after insemination and again between 20 and 24 hours. All fertilized zygotes were checked carefully for number and regular size of PN and normal presence of nucleoli within the pronucleus. Culture was continued until 72 to 78 hours after original oocyte retrieval, at which time up to four cleaving ‘late-fertilized” embryos were replaced. Transfer of two- to four-cell embryos was the norm because of the delayed fertilization. However, in some cases in which PN appeared within 6 to 10 hours after insemination, it was common to have fastercleaving embryos (e.g., 6 to 8 cells at 76 hours). All embryos from ICSI reinsemination underwent assisted hatching before transfer by drilling a loto 12-pm-diameter hole in the zona pellucida with acidified Tyrode’s medium (pH 2.4). This was performed on the theory that easier blastocyst hatching for these delayed embryos would help compensate for their developmental tardiness. Embryos in excess of those needed for transfer were cryopreserved using 1,2-propanediol and sucrose as described preAll transfers were uterine, and pregnanviously (1). cies currently either are delivered or have advanced to 16 weeks’ gestation and beyond. RESULTS
Fifty-four couples who underwent IVF in which inexplicable fertilization failure occurred after conMorton
et al.
ICSI reinsemination
489
Table 1 Intracytoplasmic Sperm Injection Reinsemination for Rescue of Completely Failed Conventional IVF Cycles No. of patient cycles No. of unfertilized reinseminated oocytes No. of intact oocytes after ICSI No. of 2PN oocytes No. of 3PN oocytes No. of embryos transferred No. of embryos cryopreserved Total no. of cleaved embryos* No. of ETs Average no. of embryos per transfer No. of viable pregnancies per cycle Embryonic implantation rate
54 489 439 (89.9) 215 (44.0) 28 (5.7) 140 164 (2746.3) 48 (88.9) 2.9 8 (14.8) 15/140 (10.7)
Note: Values in parentheses are percentages.
* Value in parentheses is the percentage of normally fertilized reinjected oocytes.
ventional insemination underwent reinsemination by ICSI of all mature oocytes of reasonable quality. The data from these 54 cycles are shown in Table 1. Of 536 available oocytes, 489 were chosen for ICSI reinsemination. Of the reinjected oocytes, 10.2% (n = 50) degenerated after injection, 215 became fertilized normally (44.0% of injected oocytes), and 28 had 3PN (5.7%). One hundred sixty-four of these zygotes (76.3% of all zygotes) developed adequately for cryopreservation or uterine transfer. Five cycles had either no fertilization or failure of embryonic development after normal fertilization after ICSI reinsemination. In 49 cycles (91%), late fertilization was achieved and ET was accomplished in 48 of these cycles (88.9%). In one case, two late embryos were frozen after a failed transfer that was due to stenosis and bleeding of the cervix. Eight pregnancies resulted from the 48 rescue cycles transferred (16.6% per ET), one of which originally was a triplet implantation and was reduced selectively to a twin, now delivered. There were two other triplet pregnancies and one twin. Thus, there were 15 term or close-to-term infants, resulting in a rate of embryonic implantation of 10.7% (15 infants/ 140 embryos transferred). In three cases, embryos were frozen (n = 24) in addition to those used for transfer, and all three patients became pregnant. To date, no transfer of any frozen embryos from this study has taken place. All couples involved in these cases were strongly advised to undergo prenatal fetal screening, and to date, all offspring are ostensibly normal. DISCUSSION
Conventional fertilization failure may arise through simple inability of the sperm to penetrate the oocyte vestments, a failure that could be a consequence of defective fusogenic properties on the part 490
Morton et al.
ZCSZ reinsemination
of either gamete. Such an occurrence probably is corrected easily by use of ICSI, which bypasses most steps of gamete fusion. Alternatively, conventional fertilization failure may be due to more fundamental inadequacies of the sperm or oocyte, for example, the inability of the sperm head to decondense or failure of oocyte activation. Such barriers to fertilization may not be overcome easily with the use of ICSI until more is understood about the biology of ICSI fertilization in humans. Presumably, both extremes of such fertilization failure were represented within the patient cycles considered here. Perhaps this might account for the lower fertilization outcome overall, including the complete fertilization failure even after ICSI reinsemination in some cases. There was no preponderance of a single type of ovarian stimulation or female diagnosis of infertility related to this group of subjects in whom conventional fertilization failed. The proportions of each group represented are as would be expected based on the numbers of women in each stimulation group or diagnosis type. The fertilization rate of the l-day-old oocytes was adequate, albeit lower than the average fertilization outcome from ICSI on fresh oocytes from our regular ICSI program (58%). The 3PN zygotes appeared at approximately three times the normal rate after ICSI on fresh eggs (5.7% versus 1.9%). This may be a consequence of oocyte aging and may be a gross indicator of the more subtle chromosomal defects that may occur in late-fertilized oocytes, such as haploidy, aneuploidy, or triploidy. Karyotyping of 10 embryos from ICSI reinsemination in another study (8) showed five to be abnormal; also in this study, a high rate of 3PN fertilization was found. This was suggestive of oocyte aging problems causing specific failure of second polar body extrusion and other spindle and microfilamentous disruption, which might lead to chromosomal anomalies. Reinsemination by ISCI of unfertilized oocytes in this study (81, however, took place 1 or 2 days after initial oocyte collection, and in any event, no ICSI reinsemination took place before 22 hours. Usually the oocytes were considerably more aged than in our study, and although late fertilization and development were seen, no embryos were used for transfer. Another study (Minguez Y et al., abstract) obtained late fertilization with ICSI, with a relatively high 3PN rate, but achieved no pregnancies after 10 transfers. Lundin et al. (5) achieved two pregnancies from transfer of late-fertilized ICSI embryos when rescuing 57 cycles of failed or near-failed conventional IVF. They believed that sperm aging was a problem and concluded that fresh sperm were needed the day after oocyte collection to produce not Fertility
and Sterility@
only fertilization but viable pregnancies. Intracytoplasmic sperm injection was performed between 25 and 26 hours after oocyte retrieval. This group also found a higher-than-expected rate of 3PN zygotes. In our study, ICSI was applied earlier than in other studies to reduce the effects of oocyte aging. Sperm were stored overnight at room temperature, having spent only 1 hour at 37°C before their initial use on the day of oocyte collection. One criticism of such early reinsemination is that initial fertilization might be missed and ICSI might be carried out on a normal zygote (9). It is doubtful that fertilization would be missed during the intricate micromanipulative procedure. However, if fertilization were missed, the reinjected zygote might become an arrested 3PN. A zygote that cleaved without 2PN being seen would not have been used for transfer in the present study. The potential for generating higher-than-expected numbers of chromosomally abnormal embryos with this late-fix ICSI must be considered in light of the known high numbers of aneuploid embryos that are generated from older women by IVF (10) and that are still transferred. Clearly, older women are strongly advised to undergo prenatal screening and consider the potential need for therapeutic abortion. Likewise, we always have counseled couples to undergo a prenatal screen when late-fertilized ICSI embryos are used. In addition, it is possible to screen the embryos before transfer by fluorescent in situ hydridization to reduce considerably the risk of aneuploidy. In one case of reinsemination in a woman undergoing IVF for X-linked-disease screening of her embryos, we generated two extra late-fertilized embryos by ICSI. Both were biopsied: one was aneuploid, and the other, which was transferred, had a normal 46,xX karyotype, as judged by fivechromosome fluorescent in situ hybridization. Although ICSI reinsemination of failed fertilization lst-day ICSI oocytes has been shown to cause high rates of mosaicism (Morton PC et al., abstract), it appears that ICSI rescue of failed conventional fertilization oocytes is feasible and moderately successful. Until chromosomal screening of all late-fertilized embryos becomes routine, the use of this reinsemination technique appears acceptable, assuming prenatal screening is advised. In addition, the more successful the reinsemination ICSI is in terms of fertilization, the more likely it is that the embryos will implant. Recently, we carried out a cycle of ICSI
Vol. 68, No. 3, September 1997
reinsemination after only 1 of 20 oocytes fertilized conventionally. The reinsemination was carried out on 15 of the unfertilized eggs, of which 7 fertilized normally. We transferred four embryos, only one of which was fertilized on the day after oocyte retrieval. Four late-fertilized embryos were cryopreserved, and the other three were transferred with the one timely fertilized embryo. Unfortunately, all four embryos implanted, one having split into a monozygotic twin. At 31 weeks, the patient was delivered of two boys and three girls, all apparently healthy. These results, in combination with the high rate of multiple implantation seen in the present study, emphasize that the usual concerns about numbers of embryos for transfer should apply to ICSI reinsemination embryos as well, to help avoid a high multiple gestation rate. Acknowledgments. The authors gratefully acknowledge the clinical skills and assistance of Carlene Elsner, M.D., Lisa Hasty, M.D., David Keenan, M.D., Joe Massey, M.D., Dorothy MitchellLeef, M.D., Daniel Shapiro, M.D., and Andrew Toledo, M.D.
REFERENCES 1. Tucker M, Elsner C, Kort H, Massey J, Mitchell-Leef D, Toledo A. Poor implantation of cryopreserved reinseminationfertilized human embryos. Fertil Steril 1991;56:1111-6. 2. Boldt J, Howe AM, Butler WJ, McDonough PG, Padilla SL. The value of oocyte reinsemination in human in vitro fertilization. Fertil Steril 1987;48:617-23. 3. Pampiglione JS, Mills C, Campbell S, Steer C, Kingsland C, Mason BA. The clinical outcome of reinsemination of human oocytes fertilized in vitro. Fertil Steril 1990;53:306-10. 4. Tucker MJ, Morton PC, Wright G, Ingargiola PE, Jones AE, Sweitzer CL. Factors affecting success with intracytoplasmic sperm injection. Reprod Fertil Dev 1995;7:229-36. 5. Lundin K, Sjiigren A, Hamberger L. Reinsemination of oneday-old oocytes by use of intracytoplasmic sperm injection. Fertil Steril 1996;66:118-21. 6. Tsirigotis M, Redgment C, Craft I. Late intracytoplasmic sperm injection (ICSI) in in vitro fertilization (IVF) cycles. Hum Reprod 1994;9:1359. 7. Oehninger S, Franken D, Kruger T. Approaching the next millenium: how should we manage andrology diagnosis in the intracytoplasmic sperm injection era? Fertil Steril 1997;67: 434-6. a. Nagy ZP, Staessen C, Liu J, Joris H, Devroey P, Van Steirteghem AC. Prospective, auto-controlled study on reinsemination of failed-fertilized oocytes by intracytoplasmic sperm injection. Fertil Steril 1995;64:1130-5. 9. Van Blerkom J, Davis PW, Merriam J. A retrospective analysis of unfertilized and presumed parthenogenetically activated human oocytes demonstrates a high frequency of sperm penetration. Hum Reprod 1994;9:2381-8. 10. Munne S, Alikani M, Tomkin G, Grifo J, Cohen J. Embryo morphology, developmental rates and maternal age are correlated with chromosomal abnormalities. Fertil Sterill995; 64: 382-91.
Morton
et al.
ICSI reinsemination
491
FERTILITY AND STERILI~ Copyright
o 1997 American
Published
by Elsevier
Society for Reproductive
Medicine
Printed
on acid-free
paper
1997
in U. S. A.
Science Inc.
Reinsemination by intracytoplasmic sperm injection of 1 -day-old _ oocytes after complete c&v-entional fertilization failure
Paula C. Morton, B.S. Cynthia S. Yoder, B.S. Michael J. Tucker, Ph.D.* Reproductive
Biology Associates,
Graham Wright, B.Sc. Wendy D. W. Brockman, B.A. Hilton I. Kort, M.D. Atlanta,
Georgia
Objective: To analyze the efficacy of intracytoplasmic sperm injection (ICSI) for “rescue” of failed conventional insemination cycles. Design: Retrospective clinical study. Setting: Private infertility clinic. Patient(s): Fifty-four couples with non-male factor infertility undergoing routine IVF within our clinic. Intervention(s): Twenty- to 24-hour-old unfertilized mature oocytes from conventional IVF cycles in which fertilization failure was complete were reinseminated by ICSI to attempt late fertilization and rescue of the otherwise failed M? cycle. Main Outcome Measure(s): Late fertilization, pregnancy, and embryonic implantation. Result(s): Intracytoplasmic sperm injection reinsemination of 489 unfertilized mature oocytes caused degeneration in 50 (10.2%) oocytes, gave rise to normal fertilization in 215 (44.0%) of the injected oocytes, and allowed ET in 48 of the 54 cycles in which initial complete fertilization failure had occurred. One hundred sixty-four (76.3%) of 215 late-fertilized oocytes either were used for fresh transfer or were frozen-stored. Eight viable pregnancies resulted, yielding a 14.8% pregnancy rate per initiated cycle. Conclusion(s): Barring potential concerns regarding the chromosomal normality of embryos arising from reinsemination, our results suggest that ICSI is a relatively successful means of rescuing conventional IVF cycles in which fertilization fails completely. (Fertil Sterile 1997;68:488-91. 0 1997 by American Society for Reproductive Medicine.) Key Words: Fertilization failure, reinsemination, ICSI, embryonic implantation, pregnancy
In non-male factor infertility IW cases, complete fertilization failure still may occur because of a variety of factors, including oocyte quality and cryptic spermatozoa1 defects. Such an event does not always occur consistently in a couple undergoing IVF, but regardless of frequency, it can be psychologically devastating for such a couple to undertake IVF’and not achieve ET because of unexpected conventional fertilization failure. Consequently, reinsemination of unfertilized oocytes in such cases can be carried out to try to achieve some late fertilization, that is, to “rescue” what otherwise would be a completely failed attempt at pregnancy. ReceivedDecember6, 1996; revisedand acceptedMay 2, 1997. * Reprintrequests:MichaelJ. Tucker,Ph.D.,ReproductiveBiology Associates, 5505 Peachtree, Dunwoody Road, Suite 400, Atlanta, 488
Georgia 30342 (FAX: 404-256-1912).
Conventional reinsemination traditionally has yielded poor outcomes in terms of both fertilization and embryonic implantation (l-3). More recently, however, micromanipulation has improved the potential to achieve late fertilization after reinsemination, though not necessarily pregnancy (4-6). Extreme forms of male factor infertility are obvious cases that justify the use of ICSI as the means of insemination in IW. However, there exist few other clear rules to establish when to choose ICSI over conventional insemination (7). Consistent, repeated failed conventional fertilization failure can be good reason to apply ICSI, but such failure is difficult to anticipate. If ICSI replaced conventional IVF insemination completely, any concerns over when ICSI should be applied would be obviated, although the question has been raised whether ICSI is justifiable when it might generate higher numbers 0015-0282/97/$17.00 PI1 s0015-0282(97)00223-9
of abnormal offspring and also cost more. If ICSI reinsemination was an effective means to rescue failed conventional IVF cycles, then automatic application of ICSI would not be needed; ICSI would be used only in those questionable cases in which conventional insemination initially has failed. In an attempt to determine the clinical efficacy of this strategy, we retrospectively reviewed all cases in the past 3.5 years in which ICSI reinsemination was used after total conventional fertilization failure occurred. MATElRIALS AND METHODS
From January 1993 to July 1996,54 couples who experienced complete fertilization failure after initial conventional insemination consented to the use of reinsemination by ICSI. Women underwent ovarian stimulation using either a long- (n = 32) or short- (n = 16) protocol leuprolide acetate regime coupled with FSH. Over six cycles, clomiphene citrate was given alone with minimal monitoring. Ovulation was initiated using hCG, 10,000 IU, in all cases, with ultrasound-guided oocyte retrieval performed 34 to 36 hours after hCG administration. In all cases there was no significant marker of overall oocyte quality that seemed awry. Primary causes of infertility in all cases excluded male factor but incorporated a wide range of female diagnoses, including tubal factor, endometriosis, pelvic adhesions, and ovulatory dysfunction. Women were between 28 and 43 years old (mean t SD, 35.3 ? 4.3 years). A total of 536 oocytes were collected initially, none of which became fertilized after conventional insemination. All semen samples were processed through 90% Percoll columns followed by three washes in Earle’s balanced salt solution (EBSS) with 6% Plasmatein (Alpha Therapeutic Corp., Los Angeles, CA). The final sperm suspension was adjusted to 5 to 20 x lo6 sperm/ml and was left at room temperature (20 to 23°C) until 1 hour before insemination, when the sample was placed into a 37”C, 5% COz in air, humidified atmosphere. Conventional insemination was carried out in microdroplets of EBSS + 6% Plasmatein under mineral oil. The sperm concentration was adjusted to a concentration between 100,000 and 3OO,OOO/mL, depending on the percentage of progressive motility. Sixteen to 18 hours later, oocytes were cleaned carefully to remove any remaining cumulus and coronal cells and were checked for the presence of pronuclei (PN) on an inverted microscope fitted with Nomarski differential interference contrast optics (Olympus IMT-2 or IX-70; Olympus, Tokyo, Japan). Three related features of fertilization were assessed: Vol. 68, No. 3, September
1997
presence of PN (with nucleoli apparent), cytoplasmic contraction, and extrusion of a second polar body. If none of these features were present, the oocyte was deemed unfertilized and was washed free of sperm. All oocytes possessed either one polar body or fragmented polar bodies, and these oocytes were considered unfertilized once it was clear that there were no other related features of fertilization. Sperm suspensions prepared from the previous day of oocyte retrieval were stored at room temperature overnight. Sperm for reinsemination was taken from this stock suspension. As soon as total conventional fertilization failure was established with all oocytes, reinsemination by ICSI was undertaken as described previously (4). Therefore, this procedure was carried out between 20 and 24 hours after the initial oocyte retrieval to minimize the effects of gamete aging. In all cases, some sperm motility remained, and such sperm were selected preferentially for injection into the unfertilized oocytes. Some unfertilized oocytes (n = 47) were rejected for reinsemination on the grounds that they either were significantly dysmorphic or were immature. After ICSI, all reinseminated oocytes were placed into EBSS + 15% maternal serum and were checked for signs of fertilization between 6 and 10 hours after insemination and again between 20 and 24 hours. All fertilized zygotes were checked carefully for number and regular size of PN and normal presence of nucleoli within the pronucleus. Culture was continued until 72 to 78 hours after original oocyte retrieval, at which time up to four cleaving ‘late-fertilized” embryos were replaced. Transfer of two- to four-cell embryos was the norm because of the delayed fertilization. However, in some cases in which PN appeared within 6 to 10 hours after insemination, it was common to have fastercleaving embryos (e.g., 6 to 8 cells at 76 hours). All embryos from ICSI reinsemination underwent assisted hatching before transfer by drilling a loto 12-pm-diameter hole in the zona pellucida with acidified Tyrode’s medium (pH 2.4). This was performed on the theory that easier blastocyst hatching for these delayed embryos would help compensate for their developmental tardiness. Embryos in excess of those needed for transfer were cryopreserved using 1,2-propanediol and sucrose as described preAll transfers were uterine, and pregnanviously (1). cies currently either are delivered or have advanced to 16 weeks’ gestation and beyond. RESULTS
Fifty-four couples who underwent IVF in which inexplicable fertilization failure occurred after conMorton
et al.
ICSI reinsemination
489
Table 1 Intracytoplasmic Sperm Injection Reinsemination for Rescue of Completely Failed Conventional IVF Cycles No. of patient cycles No. of unfertilized reinseminated oocytes No. of intact oocytes after ICSI No. of 2PN oocytes No. of 3PN oocytes No. of embryos transferred No. of embryos cryopreserved Total no. of cleaved embryos* No. of ETs Average no. of embryos per transfer No. of viable pregnancies per cycle Embryonic implantation rate
54 489 439 (89.9) 215 (44.0) 28 (5.7) 140 164 (2746.3) 48 (88.9) 2.9 8 (14.8) 15/140 (10.7)
Note: Values in parentheses are percentages.
* Value in parentheses is the percentage of normally fertilized reinjected oocytes.
ventional insemination underwent reinsemination by ICSI of all mature oocytes of reasonable quality. The data from these 54 cycles are shown in Table 1. Of 536 available oocytes, 489 were chosen for ICSI reinsemination. Of the reinjected oocytes, 10.2% (n = 50) degenerated after injection, 215 became fertilized normally (44.0% of injected oocytes), and 28 had 3PN (5.7%). One hundred sixty-four of these zygotes (76.3% of all zygotes) developed adequately for cryopreservation or uterine transfer. Five cycles had either no fertilization or failure of embryonic development after normal fertilization after ICSI reinsemination. In 49 cycles (91%), late fertilization was achieved and ET was accomplished in 48 of these cycles (88.9%). In one case, two late embryos were frozen after a failed transfer that was due to stenosis and bleeding of the cervix. Eight pregnancies resulted from the 48 rescue cycles transferred (16.6% per ET), one of which originally was a triplet implantation and was reduced selectively to a twin, now delivered. There were two other triplet pregnancies and one twin. Thus, there were 15 term or close-to-term infants, resulting in a rate of embryonic implantation of 10.7% (15 infants/ 140 embryos transferred). In three cases, embryos were frozen (n = 24) in addition to those used for transfer, and all three patients became pregnant. To date, no transfer of any frozen embryos from this study has taken place. All couples involved in these cases were strongly advised to undergo prenatal fetal screening, and to date, all offspring are ostensibly normal. DISCUSSION
Conventional fertilization failure may arise through simple inability of the sperm to penetrate the oocyte vestments, a failure that could be a consequence of defective fusogenic properties on the part 490
Morton et al.
ZCSZ reinsemination
of either gamete. Such an occurrence probably is corrected easily by use of ICSI, which bypasses most steps of gamete fusion. Alternatively, conventional fertilization failure may be due to more fundamental inadequacies of the sperm or oocyte, for example, the inability of the sperm head to decondense or failure of oocyte activation. Such barriers to fertilization may not be overcome easily with the use of ICSI until more is understood about the biology of ICSI fertilization in humans. Presumably, both extremes of such fertilization failure were represented within the patient cycles considered here. Perhaps this might account for the lower fertilization outcome overall, including the complete fertilization failure even after ICSI reinsemination in some cases. There was no preponderance of a single type of ovarian stimulation or female diagnosis of infertility related to this group of subjects in whom conventional fertilization failed. The proportions of each group represented are as would be expected based on the numbers of women in each stimulation group or diagnosis type. The fertilization rate of the l-day-old oocytes was adequate, albeit lower than the average fertilization outcome from ICSI on fresh oocytes from our regular ICSI program (58%). The 3PN zygotes appeared at approximately three times the normal rate after ICSI on fresh eggs (5.7% versus 1.9%). This may be a consequence of oocyte aging and may be a gross indicator of the more subtle chromosomal defects that may occur in late-fertilized oocytes, such as haploidy, aneuploidy, or triploidy. Karyotyping of 10 embryos from ICSI reinsemination in another study (8) showed five to be abnormal; also in this study, a high rate of 3PN fertilization was found. This was suggestive of oocyte aging problems causing specific failure of second polar body extrusion and other spindle and microfilamentous disruption, which might lead to chromosomal anomalies. Reinsemination by ISCI of unfertilized oocytes in this study (81, however, took place 1 or 2 days after initial oocyte collection, and in any event, no ICSI reinsemination took place before 22 hours. Usually the oocytes were considerably more aged than in our study, and although late fertilization and development were seen, no embryos were used for transfer. Another study (Minguez Y et al., abstract) obtained late fertilization with ICSI, with a relatively high 3PN rate, but achieved no pregnancies after 10 transfers. Lundin et al. (5) achieved two pregnancies from transfer of late-fertilized ICSI embryos when rescuing 57 cycles of failed or near-failed conventional IVF. They believed that sperm aging was a problem and concluded that fresh sperm were needed the day after oocyte collection to produce not Fertility
and Sterility@
only fertilization but viable pregnancies. Intracytoplasmic sperm injection was performed between 25 and 26 hours after oocyte retrieval. This group also found a higher-than-expected rate of 3PN zygotes. In our study, ICSI was applied earlier than in other studies to reduce the effects of oocyte aging. Sperm were stored overnight at room temperature, having spent only 1 hour at 37°C before their initial use on the day of oocyte collection. One criticism of such early reinsemination is that initial fertilization might be missed and ICSI might be carried out on a normal zygote (9). It is doubtful that fertilization would be missed during the intricate micromanipulative procedure. However, if fertilization were missed, the reinjected zygote might become an arrested 3PN. A zygote that cleaved without 2PN being seen would not have been used for transfer in the present study. The potential for generating higher-than-expected numbers of chromosomally abnormal embryos with this late-fix ICSI must be considered in light of the known high numbers of aneuploid embryos that are generated from older women by IVF (10) and that are still transferred. Clearly, older women are strongly advised to undergo prenatal screening and consider the potential need for therapeutic abortion. Likewise, we always have counseled couples to undergo a prenatal screen when late-fertilized ICSI embryos are used. In addition, it is possible to screen the embryos before transfer by fluorescent in situ hydridization to reduce considerably the risk of aneuploidy. In one case of reinsemination in a woman undergoing IVF for X-linked-disease screening of her embryos, we generated two extra late-fertilized embryos by ICSI. Both were biopsied: one was aneuploid, and the other, which was transferred, had a normal 46,xX karyotype, as judged by fivechromosome fluorescent in situ hybridization. Although ICSI reinsemination of failed fertilization lst-day ICSI oocytes has been shown to cause high rates of mosaicism (Morton PC et al., abstract), it appears that ICSI rescue of failed conventional fertilization oocytes is feasible and moderately successful. Until chromosomal screening of all late-fertilized embryos becomes routine, the use of this reinsemination technique appears acceptable, assuming prenatal screening is advised. In addition, the more successful the reinsemination ICSI is in terms of fertilization, the more likely it is that the embryos will implant. Recently, we carried out a cycle of ICSI
Vol. 68, No. 3, September 1997
reinsemination after only 1 of 20 oocytes fertilized conventionally. The reinsemination was carried out on 15 of the unfertilized eggs, of which 7 fertilized normally. We transferred four embryos, only one of which was fertilized on the day after oocyte retrieval. Four late-fertilized embryos were cryopreserved, and the other three were transferred with the one timely fertilized embryo. Unfortunately, all four embryos implanted, one having split into a monozygotic twin. At 31 weeks, the patient was delivered of two boys and three girls, all apparently healthy. These results, in combination with the high rate of multiple implantation seen in the present study, emphasize that the usual concerns about numbers of embryos for transfer should apply to ICSI reinsemination embryos as well, to help avoid a high multiple gestation rate. Acknowledgments. The authors gratefully acknowledge the clinical skills and assistance of Carlene Elsner, M.D., Lisa Hasty, M.D., David Keenan, M.D., Joe Massey, M.D., Dorothy MitchellLeef, M.D., Daniel Shapiro, M.D., and Andrew Toledo, M.D.
REFERENCES 1. Tucker M, Elsner C, Kort H, Massey J, Mitchell-Leef D, Toledo A. Poor implantation of cryopreserved reinseminationfertilized human embryos. Fertil Steril 1991;56:1111-6. 2. Boldt J, Howe AM, Butler WJ, McDonough PG, Padilla SL. The value of oocyte reinsemination in human in vitro fertilization. Fertil Steril 1987;48:617-23. 3. Pampiglione JS, Mills C, Campbell S, Steer C, Kingsland C, Mason BA. The clinical outcome of reinsemination of human oocytes fertilized in vitro. Fertil Steril 1990;53:306-10. 4. Tucker MJ, Morton PC, Wright G, Ingargiola PE, Jones AE, Sweitzer CL. Factors affecting success with intracytoplasmic sperm injection. Reprod Fertil Dev 1995;7:229-36. 5. Lundin K, Sjiigren A, Hamberger L. Reinsemination of oneday-old oocytes by use of intracytoplasmic sperm injection. Fertil Steril 1996;66:118-21. 6. Tsirigotis M, Redgment C, Craft I. Late intracytoplasmic sperm injection (ICSI) in in vitro fertilization (IVF) cycles. Hum Reprod 1994;9:1359. 7. Oehninger S, Franken D, Kruger T. Approaching the next millenium: how should we manage andrology diagnosis in the intracytoplasmic sperm injection era? Fertil Steril 1997;67: 434-6. a. Nagy ZP, Staessen C, Liu J, Joris H, Devroey P, Van Steirteghem AC. Prospective, auto-controlled study on reinsemination of failed-fertilized oocytes by intracytoplasmic sperm injection. Fertil Steril 1995;64:1130-5. 9. Van Blerkom J, Davis PW, Merriam J. A retrospective analysis of unfertilized and presumed parthenogenetically activated human oocytes demonstrates a high frequency of sperm penetration. Hum Reprod 1994;9:2381-8. 10. Munne S, Alikani M, Tomkin G, Grifo J, Cohen J. Embryo morphology, developmental rates and maternal age are correlated with chromosomal abnormalities. Fertil Sterill995; 64: 382-91.
Morton
et al.
ICSI reinsemination
491