The role of cleavage rates, number of embryos trans- ferred, and some other variables in pregnancy outcome in 222 human embryo transfers was studied.
Journal o f in Vitro Fertilization and Embryo Transfer, Vol. 4, No. 4~ 1987
The Impact of Embryo Quality and Quantity on Implantation and the Establishment of Viable Pregnancies PAUL CLAMAN, 1-3 D. R A N D A L L ARMANT, 2 M A C H E L L E M. SEIBEL, 2 TZONG-AN WANG, ~,4 SELWYN E OSKOWITZ, 2 and MELVIN L. TAYMOR2
Submitted: June l, 1986 Accepted: April 21, 1987 (North American Editorial Office)
INTRODUCTION
The role o f cleavage rates, number o f embryos transferred, and some other variables in pregnancy outcome in 222 human embryo transfers was studied. Pregnancy rates were significantly higher in a group of 117 patients receiving at least one embryo that had reached the fourcell stage at 40 hr postinsemination (26% total pregnancies~transfer and 18% ongoing pregnancies~transfer) than in the 105 patients receiving embryos developing at a slower rate (7% total pregnancies~transfer and 2% ongoing pregnancies~transfer). Of the 23 ongoing or term pregnancies produced, 21 came from transfers of at least one embryo that had reached four-cell stage by 40 hr postinsemination. Pregnancy rates were unaffected in either the fast-cleaving or the slower-cleaving embryos by culturing in vitro for an additional 24 hr. The presence of anucleate cell fragments also had no effect on pregnancy rates. Pregnancy rates increased progressively with the transfer o f more embryos per transfer. These results suggest that procedures to improve ovulation induction and in vitro embryo culture technique may better the success o f in vitro fertilization by providing a high number o f rapidly cleaving embryos for transfer.
Most attempts to establish human pregnancies by in vitro fertilization and embryo transfer (IVF-ET) appear to fail after transfer. Other than technical problems with transfer technique there are two possible causes of implantation failure; the condition of the endometrium and the quality of the embryos. To determine the relative importance of both embryo quality and quantity, the number of embryos transferred and the role of cleavage rate in pregnancy outcome in 222 patients who had embryos transferred were studied.
MATERIALS AND METHODS Ovulation induction was accomplished using a protocol of sequential or combined clomiphene citrate and gonadotropin therapy (1). Oocyte harvest was carried out 34-36 hr after human chorionic gonadotropin (hCG), 5000 u ira, or 22-24 hr after the start of the luteinizing hormone (LH) surge detected by serum L H r a d i o i m m u n o a s s a y (RIA) (Pantex, Santa Monica, CA). Follicle puncture was done laparoscopically or by ultrasound-guided needle aspiration. Heparin-containing Ham's F-10 or Delbecco's phosphate buffer was used to irrigate the follicles. After harvest, oocytes were transferred into a modified Ham's F-10 medium supplemented with 10% patient serum. Semen was prepared by a swim-up technique (2), and insemination was performed 4-8 hr postharvest. The occurrence of fertilization was assessed by the presence of two distinct pronuclei 16-20 hr postinsemination. Unfertilized oocytes were rein-
KEY WORDS: in vitro fertilization-embryo transfer; embryo quality; embryo quantity; cleavage rate; pregnancy rates.
University of Ottawa, Ottawa, Ontario, Canada.
2 Dana Biomedical Research Labortories, Department of Obstetrics and Gynecology,Beth Israel Hospital and Harvard Medical School, Boston, Massachusetts. 3 To whom correspondenceshouldbe addressedat Universityof Ottawa, Ottawa Civic Hospital, Department of Obstetrics & Gynecology, 1053 Carling Avenue, Ottawa, Ontario, Canada KIY 4E9. 4 Visitingfellowfrom Chung-ShanMedicalCollege,Taiwan. 0740-7"769/87/0800-0218505.00/09 1987 Plenum PublishingCorporation
2 18
219
EMBRYO QUALITY AND PREGNANCY RATES
seminated and examined for the presence of pronuclei 24 hr later. Oocytes with three pronuclei were not transferred. Fertilized oocytes were transferred to fresh medium containing 20% patient serum for further culture. Forty to forty-four hours after successful insemination (i.e., 24 hr after the observation of two pronuclei) the embryos were reexamined. They were then transferred or left in culture for another 24 hr, at which time they were again examined and transferred via an open-ended transcervical Teflon catheter. The presence of the beta subunit of human chorionic gonadotropin by RIA (Coming Medical, Medfield, MA) (>10 mlU/ml) in the recipient's serum 2 weeks after transfer was considered diagnostic of pregnancy. Continuation of pregnancy for at least 14 weeks posttransfer was considered an ongoing pregnancy. Chi-square analysis was used to detect statistical significance. If the expected values in any of the cells of the 2 X 2 contingency tables were less than 5, a Fisher's exact test was used.
RESULTS From November 1983 through December 1985, 259 patients had successful oocyte retrieval. Of these, 222 patients had at least one egg fertilize and go on to cleave (86% fertilization rate per retrieval). Two hundred twenty-two embryo transfers were performed, with one to five embryos per transfer. Of these 222 transfers, 118 included at least one embryo that required reinsemination for fertilization. Of all 222 transfers, 37 (17%) recipients became pregnant and 23 (10%) continued to have ongoing or term pregnancies. Only 4 of these 37 pregnancies resulted from the 118 transfers which included at least one reinseminated ovum. In reviewing our data we observed a number of variables that had an effect on pregnancy rates.
Embryo Cleavage Rates Embryo transfers done with at least one embryo at the four-cell stage at 40 hr postinsemination (N = 117) resulted in pregnancy rates of 26%, while cases using only slower-cleaving embryos (N = 105) resulted in pregnancy rates of 7% (P < 0.0002, • (Table I). Ongoing pregnancy rates were 18 and 2% per transfer for the faster- and slower-cleaving embryos, respectively (P < 0.0001, • Of those
who became pregnant the loss rate resulting from the transfer of faster (N = 30)- and slower (N = 7)-cleaving e m b r y o s was 30 and 71% per pregnancy, respectively (P < 0.05, Fisher exact test). Expresed another way, 21 of our 23 ongoing pregnancies came from transfers using at least one embryo that had reached the four-cell stage by 40 hr postinsemination (Table I). Number of Embryos Transferred Total and ongoing pregnancy rates increased as more embryos were transferredmup to four embryos per case (Tables II and III). Regardless of whether one, two, three, four, or five embryos were transferred per case, the total and ongoing pregnancy rates were higher if at least one embryo transferred was on the faster cleavage schedule. Pregnancy rates were low when transferring only slower-cleaving embryos no matter how many embryos were transferred. Although there appears to be higher pregnancy rates with the transfer of higher numbers of only slowly cleaving embryos, our experience transferring five slowly cleaving embryos was limited (N = 7) (Tables II and III). The number of embryos transferred did not appear to affect the risk of multiple births. We observed only four twin pregnancies obtained with the transfer of two (one twin), three (two twins), and five (one twin) embryos and no other multiple pregnancies. We have also considered the effect of the number of embryos transferred, focusing only on embryos cleaving at the faster rate. When the number of faster-cleaving embryos transferred in each case was examined, we found that the total and ongoing pregnancy rates increased with the number of faster-cleaving embryos transferred (Fig. l). Time in Culture One hundred four transfers were performed at 40-44 hr and 118 transfers at 6 4 - 6 8 hr postinsemination with similar pregnancy rates of 19 and 14%, respectively (P > 0.05, • with Yate's correction). When transferring only more slowly cleaving embryos, we obtained 6 pregnancies (1 I%) from the 55 cases transferred at 4 0 - 4 4 hr and 1 pregnancy (2%) from the 50 transfers done at 64-68 hr (P > 0.05, Fisher exact test). Although there appeared to be more pregnancies when transferring slowly cleaving embryos earlier, a statistical difference in pregnancy rates was not demonstrated.
Journal of in Vitro Fertilization and Embryo Transfer, Vol. 4, No. 4, 1987
220
CLAMAN, ARMANT, SEIBEL, WANG, OSKOWITZ, AND TAYMOR
Table I. Effect of Cleavage Rate on Pregnancy rates in IVF-ET Maximum cleavage at 40 hr postinsemination
Total pregnant (%/transfer)
Total transfers 105
All embryos 4 cells Total
7 (7%)
117 222
Fragmentation Of our 222 embryo transfers, 52 involved transferring only embryos that contained anucleate cell fragments. When transferring only fragmented embryos 1 I% of the patients became pregnant, while 19% of the patients receiving at least one unfragmented embryo became pregnant. There was no statistical difference between the two groups (P = 0.911, X2 with Yate's correction). Of pregnant patients receiving only fragmented embryos, 50% had ongoing pregnancies, and if receiving at least one unfragmented embryo 66% of the pregnancies were ongoing (P > 0.05, Fisher exact test).
DISCUSSION Observations by others of preimplantation embryos grown in vitro suggest that there is a rate of c l e a v a g e that is c o n s i d e r e d normal ( 3 - 6 ) . A number of workers have noticed a trend to higher pregnancy rates with the transfer of faster-cleaving e m b r y o s (7-10), while one group claimed that cleavage rates were unimportant in establishing pregnancies (11). Only one of these reports (10) attempted to establish a connection between the developmental schedule maintained by preimplantao Table II. Transfers with at Least One Embryo Cleaved to the Four-Cell Stage 40 hr Postinsemination
N embryo transfers N pregnancies % pregnant N ongoing/term pregnancies % ongoing/term pregnant
3
4
5
21 3 14%
24 5 21%
20 4 20%
41 15 37%
I1 3 27%
117 30 26%
2
2
4
12
1
21
20%
29%
9%
18%
8%
2 (2%)
5 (71%)
21 (18%) 23 (10.4%)
9 (30%) 14 (38%)
P < 0.0002
P < 0.0001
P < 0.05
tion embryos and pregnancy rates. These workers reported that 50% of embryos resulting in pregnancies reached the four-cell stage by 32-40 hr after insemination (10). Furthermore, no one has addressed the question of whether the higher pregnancy rate resulting from the transfer of a higher number of embryos per case (6,12) is actually the result of transferring more embryos that are on a faster cleavage schedule (13). We have found significantly (P < 0.0002) higher total (26%) and ongoing (18%) pregnancy rates when transferring at least one embryo that had reached the four-cell stage by 40-44 hr postinsemination than when transferring only more slowly cleaving embryos (7% total and 2% ongoing pregnancies per transfer). The tendency toward increased total and ongoing pregnancy rates with at least one faster-cleaving embryo remains no matter how many embryos were transferred per case (Tables II and III). Furthermore, our data suggest that oocytes not fertilizing after a first attempt may not be healthy. Only rare pregnancies result from the transfer of embryos even if only one ovum in the transferred cohort requires reinsemination. Some groups have observed a general tendency toward higher p r e g n a n c y rates with increased numbers of embryos transferred per case for up to three and some for up to four embryos (6,12). We Table IlL All Embryos Transferred at Less than the Four-Cell
Stage 40 hr Postinsemination N total transferred embryos/case
2
10%
Pregnancy losses (%/pregnancy)
30 (26%) 37 (16.7%)
N total transferred embryos/case I
Ongoing pregnancies (%/transfer)
Total N embryo transfers N pregnancies % pregnant N ongoing/term pregnancies % ongoing/term pregnant
1
2
3
4
5
16 0 0%
40 2 5%
23 2 9%
19 1 5%
7 2 29%
105 7 7%
0
i
0
0
1
2
0%
3%
0%
0%
14%
Total
2%
Journal of in Vitro Fertilization and Embryo Tranafer, Vol. 4, No. 4, 1987
221
EMBRYO QUALITY AND PREGNANCY RATES
45"
3S" M %Total Pregnant 30 . . . . . % 0 %
Pregnant/ 25. Transfer 20.
:. ..........
15"
.............
.......................s-
10"
o o N=105
i
i
N=69
N=32
i N=11
i N=5
No. Fast Cleaving Embryos Transferred
Fig. 1. Percentage pregnancies with faster embryos: percentage total (top, solid line) and ongoing (bottom, dashed line) pregnancies per transfer considering only faster-cleaving embryos.
observed a trend to higher total and ongoing pregnancy rates per transfer with more embryos transferred up to four (Tables II and III). When only more slowly cleaving embryos are transferred, no dramatic trend toward increased pregnancies with increased numbers of embryos transferred was observed. Although Table III suggests higher pregnancy rates with increasing numbers of slowcleaving embryos, it is difficult to evaluate the significance of this trend, as only 7 pregnancies in 105 of these transfers were observed. Furthermore, we found that the total and ongoing pregnancy rates were proportional to the number of faster-cleaving embryos transferred, regardless of the total number of embryos transferred (Fig. I). These results suggest that although many embryos do not reach the four-cell stage by 40-44 hr of in vitro culture, it is only the faster-cleaving embryos that have a good potential for establishing pregnancies. The developmental schedule maintained by faster-cleaving embryos may reflect the temporal profile of human embryos developing in vivo. There has also been discussion in the literature as to the best time after fertilization to transfer the embryos. Some suggest that early transfer is detrimental (13). More recent data suggest that early transfer may be beneficial (14) even at the pronucleate stage (15). Our data are similar to those of Trounson and co-workers (9,16) in that we did not find any difference in pregnancy rates when we performed embryo transfer at 40-44 or 64-68 hr postinsemination. Moreover, the time of transfer did not change pregnancy rates significantly even for those embryos that were on the slower cleavage schedule. One recent paper reports higher pregnancy rates
in patients receiving unfragmented embryos (17). Our data agree with others (6,7,10,11,16,18,19) in showing pregnancy rates to be unaffected when transferring only fragmented embryos. The human embryo appears able to dispose of these anucleate cytoplasmic fragments. First, it appears, then, that the cleavage rate of preimplantation embryos is an important independent variable in the success of IVF-ET. Second, transferring increasing quantities of embryos up to four, and, if slowly cleaving, up to five, may increase pregnancy rates. This finding has an impact on the availability of " e x c e s s " embryos for study or donor embryo programs. Third, embryos can be cultured in vitro for 40 or 64 hr postinsemination before being transferred with impunity. Finally, the presence of anucleate cell fragments in the embryo does not imply nonviability. These results suggest that procedures to improve ovulation induction and in vitro embryo culture technique may improve the success of IVF-ET by providing a high number of rapidly cleaving embryos for transfer. Furthermore, by studying cases where only faster-cleaving embryos are transferred, it should be possible to improve experiments designed to assess the impact of endometrial quality on the success of IVF-ET.
REFERENCES I. Taymor ML, Seibel MM, Oskowitz SP, Smith DM, Lee G: In vitro fertilization and embryo transfer: An individualized approach to ovulation induction. J Vitro Fert Embryo Transfer 1984;2:162-168 2. Alper MM, Lee GS, Seibel MM, Smith D, Oskowitz SP, Ransil BJ, Taymor ML: The relationship of semen parameters to fertilization in patients participating in a program of in vitro fertilization. J Vitro Fert Embryo Transfer 1985;2:217-223 3. Lauritsen JG: Results of in vitro fertilization and culture of human ova. In Follicular Maturation and Ovulation, Proceedings of the 14th Reinier De Graaf Symposium, Nijmegen, Aug 20-22, 1981, R Rolland, EV Van Hall, SG Hillier, KP McNatty, J Schoemaker (eds). Princeton, N J, Excerpta Medica, pp 346-351 4. Sundstrom P, Nilsson O, Liedholm: Cleavage rate and morphology of early human embryos obtained after artificial fertilization and culture. A c t a O b s t e t G y n e c o l Scand 1981;60:109-120 5. Edwards RG: Culture and storage of preimplantation embryos. In Conception in the Human Female. RG Edwards (ed). New York, Academic Press, 1980, pp 685-689 6. Byrd W, Wolf DP: Oogenesis, fertilization and early devel-
Journal o f in Vitro Fertilization and E m b ~ o Transfer, Vol. 4, No. 4, 1987
222
7. 8.
9.
10.
11.
12.
13.
CLAMAN~ ARM.ANT,SEIBEL, WANG, OSKOWITZ,AND TAYMOR
opment. In Human in Vitro Fertilization and Embryo Transfer, DP Wolf, MM Quigley (eds). New York: Plenum Press, 1984, pp 213-273 Edwards RG: Test-tube babies, 1981. Nature 1981;293:253256 Humeau: Criteres d'evaluation de la qualite des embryons obtenus par FIV et de leur aptitude a l'implantation. Ann Biol Clin 1985;43:23-26 Trounson AO, Mohr LR, Wood C, Leeton JF: Effect of delayed insemination on in-vitro fertilization, culture and transfer of human embryos. J Reprod Fen 1982;64:285-294 Mohr LR, Trounson AO, Leeton JF, Wood C: Evaluation of normal and abnormal human embryo development during procedures in vitro. In Fertilization of the Human Egg in Vitro, HM Beier, HR Lindner (eds). New York, SpringerVerlag, 1983, pp 211-221 Wortham JEW Jr, Veeck LL, Witmyer J, Jones HW: Vital initiation of pregnancy (VIP) using human menopausal gonadotropin and human chorionic gofiadotropin ovulation induction: Phase 1--1981. Fertil Steril 1983;39:785-793 Testart J, Belaisch-Allart J, Hazout A, Guillet-Rosso F, Frydman R: Actual methods and results in an IVF and ET program. In Gamete Quality and Fertility Regulation, Proceedings of the 5th Reinier de Graaf Symposium, Nijmegen, The Netherlands, Aug 23-24, 1984, R Rolland, MJ Heineman, SG Hillier, H Vetoer (eds). New York, Excerpta Medica, Elsevier, 1985, pp 271-287 Adams CE: Consequences of accelerated ovum transport,
14.
15.
16.
17.
18.
19.
including a re-evaluation of the Este's operation. J Reprod Fert 1979;55:239-246 Kreitmann O, Hodgen GD: Retarded cleavage rates of p r e i m p l a n t a t i o n monkey embryos in vitro. JAMA 1981;246:627- 629 Feichtinger W, Kemeter P: Early stage embryo transfer--A possible reason for a highly reduced abortion rate. In Gamete Quality and Fertility Regulation, Proceedings of the 5th Reinier de Graaf Symposium, Nijmegen, The Netherlands, Aug 23-24, 1984, R Rolland, MJ Heineman, SG Hitlier, H Vetoer, (eds). New York, Excerpta Medica, Elsevier, 1985, pp 283-288 Trounson A: Factors controlling normal embryo development and implantation of human oocytes fertilized in vitro. In Fertilization of the Human Egg in Vitro, HM Beier, HR Lindner (eds). New York, Springer-Verlag, 1983, pp 235-250 Dor J, Rudak E, Mashiach S, Nebel L, Serf DM, Goldman B: Periovulatory 17B-estradiol changes and embryo morphologic features in conception and nonconceptional cyctes after human in vitro fertilization. Fertii Steril 1986;45:63-68 Kileen D, Moore NW: The morphological appearance and development of sheep ova fertilized by surgical insemination. J Reprod Fert 1971;24:63-70 Lopata A: Factors influencing the growth of human preimplantation embryos in vitro. In Human Conception in Vitro, RG Edwards, JM Purdy, (eds). New York, Academic Press, 1982, pp 207-226
Journal of in Vitro Fertilization and Embryo Transfer, Vol. 4, No. 4, 1987
