Prediction of the rates of fertilization, cleavage, and pregnancy success by cumulus-coronal morphology in an in vitro fertilization program

Transcription

1 FERTILITY AND STERILITY VOL. 72, NO. 3, SEPTEMBER 1999 Copyright 1999 American Society for Reproductive Medicine Published by Elsevier Science Inc. Printed on acid-free paper in U.S.A. Prediction of the rates of fertilization, cleavage, and pregnancy success by cumulus-coronal morphology in an in vitro fertilization program Siu T. Ng, M.D., Tai-Ho Chang, M.D., and T-C. Jackson Wu, M.D., Ph.D. Department of Obstetrics and Gynecology, University of California-Los Angeles School of Medicine, Los Angeles, California Objective: To examine the relation between the grading of cumulus-coronal morphology at oocyte retrieval and the rates of fertilization, cleavage, and pregnancy success in IVF-ET cycles. Design: Retrospective study. Setting: University-affiliated medical center. Patient(s): Infertile women who underwent IVF-ET treatment. Intervention(s): None. Main Outcome Measure(s): Fertilization and cleavage of the oocytes and the pregnancy outcome. Result(s): Mature grade 3 cumulus-oocyte complexes (COCs) constituted the highest percentage among all grades and had a higher fertilization rate than COCs of other grades (77% versus 65%, 43%, and 28% for grades 2, 1, and 4, respectively). The cleavage and polyspermy rates did not correlate with cumulus-coronal morphology grading. The pregnancy rate was higher in cycles with 50% grade 3 COCs than in cycles with 50% grade 3 COCs (32% versus 16%). In cycles with 80% grade 3 COCs, the pregnancy rate was 57%. The correlation between the percentage of grade 3 COCs and the pregnancy rate was independent of patient age and the number of COCs retrieved. Conclusion(s): The cumulus-coronal morphology grade correlates with the fertilization rate but not with the cleavage or polyspermy rate. In vitro fertilization cycles that have a greater percentage of grade 3 COCs have an increased chance of resulting in pregnancy. The cumulus-coronal morphology grade predicts pregnancy success in IVF-ET cycles. (Fertil Steril 1999;72: by American Society for Reproductive Medicine.) Key Words: Cumulus-coronal morphology, predictive value, pregnancy outcome, in vitro fertilization Received November 6, 1998; revised and accepted April 5, Reprint requests: T-C. Jackson Wu, M.D., Ph.D., Department of Obstetrics and Gynecology, University of California-Los Angeles School of Medicine, Le Conte Avenue, Los Angeles, California (FAX: ) /99/$20.00 PII S (99) Cumulus-coronal cell morphology (CCM) undergoes distinctive changes during the preovulatory period (1). These changes, including the dispersion (a decrease in cell density) and mucification of the cumulus mass, are due to the accumulation of hyaluronic acid in the matrix and the disappearance of gap junctions among cumulus-coronal cells (2). Testart et al. (3) showed that the cell density of the cumulus mass decreased to about one tenth of its original value on the achievement of meiotic maturation and that this decrease continued until ovulation. In naturally ovulatory cycles, the pattern of changes in CCM closely parallels the meiotic maturation of the oocyte. However, in hyperstimulated ovarian cycles, varying degrees of asynchrony exist between CCM and nuclear maturity (4, 5). It has been suggested that such asynchrony might be due either to a different sensitivity of the cumulus-coronal cells and the oocyte nuclei to gonadotropin-induced maturation or to other intrafollicular factor(s) (5, 6). On the other hand, evidence suggests that cumulus cells play an important role in cytoplasmic maturation, a critical element in achieving successful fertilization, pronucleus formation, and subsequent development (7 9). Despite the reported asynchrony between CCM and nuclear maturity of the oocyte, many IVF programs continue to use CCM to approximate oocyte maturity because of its simplicity 412

2 and the desire to preserve cumulus for sperm activation. It is prudent to assess the status of oocyte maturity to determine the appropriate period of incubation before sperm insemination in IVF cycles (10, 11). Less mature oocytes may require a longer preincubation period than more mature oocytes. To date, the impact of CCM grading on the outcomes of an IVF program has not been fully evaluated. The purpose of this study was to examine the relation between CCM and the rates of normal fertilization, polyspermy, cleavage, and pregnancy success in an IVF program. MATERIALS AND METHODS Patients and Protocols A total of 138 cycles in 116 couples who underwent IVF-ET for a variety of infertility problems between 1989 and 1992 were included in this study. Thirteen couples with male factor infertility were excluded from the study. Causes of infertility included tubal factors (87%), endometriosis (6%), and unexplained infertility (7%). The female partners were between 22 and 44 years old (average, 35.8 years old). Institutional review board approval was not required for this study because it was retrospective and required only chart review. Controlled ovarian hyperstimulation was achieved with the standard long protocol (89%) or the short protocol (11%). In brief, patients who underwent the long protocol began receiving daily SC injections of 1 mg of a GnRH agonist (Lupron; TAP Pharmaceuticals, Deerfield, IL) on day 21 of the cycle preceding IVF. The dose of the GnRH agonist was decreased to 0.5 mg daily when ovarian suppression was achieved (i.e., serum 17 -E 2 levels were 20 pg/ml) and no follicle measured 5 mm in diameter. Patients who underwent the short protocol received daily SC injections of 1 mg of a GnRH agonist starting on day 3 of the menstrual cycle. Human menopausal gonadotropin (Pergonal; Serono Laboratories, Inc., Norwell, MA) and/or FSH (Metrodin; Serono Laboratories, Inc.) was used to stimulate multiple follicular development. The doses of gonadotropins were adjusted according to the degree of follicular growth as monitored by serial ultrasound examinations and serum 17 -E 2 measurements. When more than two dominant follicles reached 18 mm in the largest dimension, hcg (10,000 U, Profasi; Serono Laboratories, Inc.) was given intramuscularly to induce final follicular and oocyte maturation. Ultrasound-guided follicular aspiration was performed 35 hours after hcg administration. As soon as the cumulus-oocyte complexes (COCs) were obtained, they were graded according to their CCM by the following criteria: grade 1 absent to sparse cumulus cells and 1 3 layers of corona cells; grade 2 dense cumulus cells and tightly packed corona cells; grade 3 expanded, fluffy cumulus cells and expanded corona cells; TABLE 1 Distribution of cumulus-coronal morphology grades in 1,517 cumulus-oocyte complexes from 138 IVF cycles. Cumulus-coronal morphology grade and grade 4 expanded, scanty cumulus cells and expanded, often partially lost corona cells (12). The COCs were preincubated for 6 hours before insemination (50,000 motile spermatozoa per COC). Sixteen hours after insemination, the COCs were examined for the presence of pronuclei. The normally fertilized zygotes were cultured for an additional hours and examined for cleavage. Two days after retrieval of the oocytes, embryos were selected for intrauterine transfer. Outcome Measures The term normal fertilization was assigned to oocytes with 2 pronuclei and polyspermy was designated in oocytes with 3 pronuclei. The cleavage rate equaled the number of embryos with 2 cells divided by the number of normally fertilized zygotes. Pregnancy success was defined by a positive test result for serum hcg days after ET. STATISTICAL ANALYSIS Statistical analysis was performed using the 2 test, Fisher s exact test, logistic regression analysis, or linear regression analysis where appropriate. P.05 was considered statistically significant. RESULTS No. (%) of cumulus-oocyte complexes 1 14 (1) (37) (60) 4 32 (2) Total 1,517 (100) A total of 1,517 COCs from 138 IVF cycles were obtained and included in this study. The distribution of COCs according to the CCM grade is summarized in Table 1. There was a higher percentage of grade 3 COCs (60%) than of grade 1, 2, or 4 COCs (1%, 37%, and 2%, respectively). No statistically significant differences in the CCM distribution pattern were detected between the long and short ovarian hyperstimulation protocols. Impact of CCM Grade on the Rates of Normal Fertilization, Polyspermy, and Cleavage The comparisons of normal fertilization, polyspermy, and cleavage rates among the various CCM groups were ana- FERTILITY & STERILITY 413

3 FIGURE 1 The grading of CCM at oocyte retrieval. Representative photographs of grades 1 (A),2(B),3(C), and 4 (D) CCM. Grade 1 COCs display corona radiata cells and sparse cumulus cells. Grade 2 COCs display tightly packed corona cells and dense cumulus cells. Grade 3 mature COCs display expanded corona cells and expanded fluffy cumulus cells. Grade 4 postimmature COCs display dark, degenerated corona cells and expanded, scanty cumulus cells. lyzed by the 2 and Fisher s exact tests. A total of 1,080 oocytes achieved normal fertilization (71%), whereas 76 oocytes attained polyspermy (5%). The rates of normal fertilization, polyspermy, and cleavage for the four CCM groups are summarized in Figure 1. The normal fertilization rates for grade 1, 2, 3, and 4 COCs are 43%, 65%, 77%, and 28%, respectively. Grade 3 COCs underwent normal fertilization at a significantly higher rate than the other groups (P.05, 2 test and Fisher s exact test). The rates of polyspermy for grade 1, 2, 3, and 4 COCs were 14%, 4%, 6%, and 9%, respectively. Although the polyspermy rate (14%) was highest in the grade 1 COCs, it did not differ significantly from the other groups (P.05). The overall cleavage rate was 91%. The cleavage rates for grade 1, 2, 3, and 4 COCs are 83%, 91%, 91%, and 89%, respectively. Likewise, the differences in the cleavage rates among the four CCM groups did not reach statistical significance (P.05). Correlation of CCM Grade With Pregnancy Rate In this study, there were 35 pregnancies from 138 cycles (25%). Twenty-six of the 35 pregnancies (74%) resulted from cycles with 50% grade 3 COCs. A positive correlation was noted between the pregnancy rate and the percentage of grade 3 COCs in a given IVF cycle (Fig. 2). In cycles with 50% grade 3 COCs (82 cycles), the pregnancy rate FIGURE 2 The rates of normal fertilization, polyspermy, and cleavage among the four CCM grades. *P.05 compared with the other CCM groups. grade 1; _ grade 2; ^ grade 3; and grade Ng et al. Cumulus-coronal morphology and IVF Vol. 72, No. 3, September 1999

4 FIGURE 3 The percentage of pregnancies per cycle versus the percentage of grade 3 COCs per cycle. The bars represent the SE. *P.05. number of COCs per cycle, and patient age. The results are presented in Table 2. These results indicate that when the effects of patient age and the total number of COCs per cycle are controlled for, the percentage of grade 3 COCs per cycle is a statistically significant predictor of pregnancy success (P.03), whereas both patient age and the total number of COCs per cycle are not (P.66 and P.23, respectively). Similarly, patient age and the total number of COCs per cycle did not correlate significantly with the percentage of grade 3 COCs in a given IVF cycle by linear regression analysis (P.61, r and P.35, r 2.007, respectively). DISCUSSION was 32%, significantly higher than the rate of 16% in cycles with 50% grade 3 COCs (56 cycles; P.05, 2 test). In cycles with 80% grade 3 COCs, the pregnancy rate was 57%. Figure 3 presents the distribution of pregnancy outcomes in the different IVF cycles according to the percentage of grade 3 COCs and patient age. A positive pregnancy outcome was likely to be associated with a high percentage of grade 3 COCs per cycle, independent of patient age (Fig. 4). To analyze this finding further and to control for other possible confounding factors, we performed logistic regression analysis to examine the relation between the pregnancy rate and the following parameters: the percentage of grade 3 COCs per cycle, the total Whether it occurs in vivo or in vitro, fertilization follows the same series of events: [1] sperm capacitation, [2] penetration through the cumulus mass, [3] binding to the zona pellucida, [4] induction of the acrosomal reaction, [5] fusion of the gamete membranes, [6] pronucleus formation, and [7] fusion of the male and female pronuclei. Thus, many factors may influence the process of fertilization. The results of this study show that a mature CCM grade (grade 3) during the periovulatory period predicts a higher rate of normal fertilization in an IVF cycle. Using a similar grading system, other investigators (6, 13) also found better fertilization rates with mature COCs. A maturing oocyte has an intricate relation with its surrounding cumulus-coronal cells, which interdigitate with the oocyte membrane through cytoplasmic processes (14) and transport nutrients to the oocyte through gap junctions (15). FIGURE 4 The distribution of pregnancy outcome according to the percentage of grade 3 COCs per cycle and patient age. not pregnant; F pregnant. FERTILITY & STERILITY 415

5 TABLE 2 Regression coefficients from logistic regression analysis with occurrence of pregnancy as the dependent variable. Independent variable Coefficient SE P value Percentage of grade 3 cumulus-oocyte complexes per cycle Patient age No. of cumulus-oocyte complexes per cycle On the other hand, the oocyte may exert a paracrine effect on the ability of the cumulus-coronal cells to expand or mucify. Using a murine model, Vanderhyden et al. (16) found that a cumulus expansion-enabling factor (CEEF) was secreted by the oocyte in response to the administration of FSH. Growing, meiotically incompetent oocytes did not produce detectable amounts of CEEF, but fully grown, meiosisarrested maturing oocytes and metaphase II oocytes did (16, 17). Further, the ability of oocytes to secrete CEEF and the capacity of cumulus cells to respond to FSH and CEEF are temporally correlated with the acquisition of oocyte competence to undergo germinal vesicle breakdown. Our recent findings that mammalian oocytes express both estrogen and activin receptor genes suggest that the increased production of estrogen and activin from the growing follicle may exert a paracrine effect on the oocyte maturation process (18 20). These reciprocal effects, exerted by the oocyte and its investing cumulus-coronal cells on each other s development, provide an explanation for the validity of using CCM as an estimator of oocyte maturity. Both nuclear and cytoplasmic maturation play important roles in achieving successful fertilization and subsequent development (21, 22). Available data suggest that CCM reflects the status of cytoplasmic maturity (7 9). Although ovarian hyperstimulation may cause a high degree of discrepancy between CCM and nuclear maturity (4, 5), its effects on the relation between CCM and cytoplasmic maturity are unclear. In addition to demonstrating a positive relation between CCM and the rate of normal fertilization, the present study showed that CCM correlates with the pregnancy success rate in a given IVF cycle. An IVF cycle with 50% grade 3 COCs had twice the pregnancy rate of an IVF cycle with 50% grade 3 COCs (32% versus 16%). The pregnancy rate for cycles with 80% grade 3 COCs was 57%. Clinically, this information is helpful in predicting the pregnancy outcome in an IVF program. This is the first report on the relation between CCM and pregnancy success. The exact mechanisms that enable grade 3 COCs to achieve higher than normal fertilization rates and pregnancy success rates are unknown. Both events of fertilization and pregnancy are good indications of optimal cytoplasmic maturity of the oocytes (23, 24). Therefore, it is possible that grade 3 COCs have achieved better cytoplasmic maturation, leading to better fertilization and pregnancy success rates. In addition, our results suggest that controlled ovarian hyperstimulation does not disturb the close relation between CCM and the cytoplasmic maturity of the oocyte. The blockage to polyspermy is the consequence of the cortical reaction induced by membrane depolarization immediately after sperm penetration. It involves the fusion of the oocyte and the cortical granule membranes, and the release of hydrolytic enzymes into the perivitelline space. Despite this complex mechanism, about 6% of the oocytes fertilized in vitro display three pronuclei, and four fifths of these have been fertilized by two spermatozoa (25). It has been shown that oocyte immaturity or postmaturity is associated with higher rates of polyspermy (13, 26 30). It is postulated that in immature oocytes, the cortical granules are still migrating to the vicinity of the plasma membrane; therefore, the cortical reaction fails to occur after the penetration of the first spermatozoon. However, in vitro maturation of the oocytes before fertilization significantly reduced the rate of polyspermy (10, 28). In contrast, Veeck et al. (31) reported no increase in the incidence of polyspermy in immature human oocytes (8.1%) compared with mature oocytes (8.8%). In our study, the polyspermy rate was the highest in the immature grade 1 COCs (14%). Nonetheless, the polyspermy rates did not differ significantly among the four CCM groups. This may be due in part to the relatively small sample sizes in the grade 1 and grade 4 groups. Approximately 32 hours after insemination, as many as 98% of the fertilized eggs underwent cleavage, leading to two-cell embryos (32). In the present study, the cleavage rate was calculated from the total number of normally fertilized zygotes. No statistically significant difference in cleavage rates was detected among the four CCM groups. On the other hand, in 1989, Bar-Ami et al. (6) found a significantly higher cleavage rate in mature COCs compared with immature, intermediate, and atretic COCs. However, both studies had relatively small numbers of COCs in the immature and atretic groups; therefore, further studies are needed to delineate the relation between CCM and cleavage rates. In conclusion, the present study demonstrated that CCM grading correlates significantly and positively with normal fertilization rates and may serve as a useful predictor for pregnancy success in an IVF cycle. Further studies are needed to clarify the exact relation between CCM grading and the rates of polyspermy and cleavage. References 1. Steptoe PC, Edwards RG. Laparoscopic recovery of preovulatory human oocytes after priming of ovaries with gonadotrophins. Lancet 1970;1: Ng et al. Cumulus-coronal morphology and IVF Vol. 72, No. 3, September 1999

6 2. Dekel N, Kraicer PF. Induction in vitro of mucification of rat cumulus oophorus by gonadotrophins and adenosine 3, 5 -monophosphate. Endocrinology 1978;102: Testart J, Thebault A, Frydman R. Timing of ovulation and oocyte maturity after LH surge in women. In: Hafez ESE, Semm K, eds. In vitro fertilization and embryo transfer. Lancaster, England: MTP Environment Ltd., 1982: Testart J, Frydman R, De Mouzon J, Lassalle B, Belaisch JC. A study of factors affecting the success of human fertilization in vitro. I. Influence of ovarian stimulation upon the number and condition of oocytes collected. Biol Reprod 1983;28: Laufer N, Tarlatzis BC, DeCherney AH, Masters JT, Haseltine FP, MacLusky N, et al. Asynchrony between human cumulus-corona cell complex and oocyte maturation after menopausal gonadotropin treatment for in vitro fertilization. Fertil Steril 1984;42: Bar-Ami S, Gitay-Goren H, Brandes J. Different morphological and steroidogenic patterns in oocyte/cumulus-corona cell complexes aspirated at in vitro fertilization. Biol Reprod 1989;41: Motlik J, Fulka J. Fertilization of rabbit oocytes co-cultured with granulosa cells. J Reprod Fertil 1981;63: Vanderhyden BC, Armstrong DT. Role of cumulus cells and serum on the in vitro maturation, fertilization, and subsequent development of rat oocytes. Biol Reprod 1989;40: Das K, Stout LE, Hensleigh HC, Tagatz GE, Phipps WR, Leung BS. Direct positive effect of epidermal growth factor on the cytoplasmic maturation of mouse and human oocytes. Fertil Steril 1991;55: Trounson AO, Mohr LR, Wood C, Leeton JF. Effect of delayed insemination on in-vitro fertilization, culture and transfer of human embryos. J Reprod Fertil 1982;64: Veeck LL. Morphological estimation of mature oocytes and their preparation for insemination. In: Jones HW, Jones GS, Hodgen GD, Rosenwaks Z, eds. In vitro fertilization. Baltimore: Williams & Wilkins, 1986: Wolf DP. Oocyte quality and fertilization. In: Wolf DP, ed. In vitro fertilization and embryo transfer. New York: Plenum Press, 1988: Khan I, Staessen C, Van den Abbeel E, Camus M, Wisanto A, Smitz J, et al. Time of insemination and its effect on in-vitro fertilization, cleavage and pregnancy rates in GnRH agonist/hmg-stimulated cycles. Hum Reprod 1989;4: Zamboni L. Fine morphology of the follicle wall and follicle celloocyte association. Biol Reprod 1974;10: Donahue RP, Stern A. Follicular cell support of oocyte maturation: production of pyruvate in vitro. J Reprod Fertil 1968;18: Vanderhyden BC, Caron PJ, Buccione R, Eppig JJ. Developmental pattern of the secretion of cumulus expansion-enabling factor by mouse oocytes and the role of oocytes in promoting granulosa cell differentiation. Dev Biol 1990;140: Eppig JJ, Peters AH, Telfer EE, Wigglesworth K. Production of cumulus expansion enabling factor by mouse oocytes grown in vitro: preliminary characterization of the factor. Mol Reprod Dev 1993;34: Wu TCJ, Wang L, Wan YJJ. Expression of estrogen receptor gene in mouse oocyte and during embryogenesis. Mol Reprod Dev 1992;33: Wu TCJ, Wang L, Jih MH, Wan YJY. Activin receptor type II gene expression is induced during embryonic development and differentiation of murine F-9 embryonal carcinoma cells. Dev Growth Differ 1993;35: Wu TCJ, Wang L, Wan YJY. Detection of estrogen receptor messenger ribonucleic acid in human oocytes and cumulus-oocyte complexes using reverse transcriptase-polymerase chain reaction. Fertil Steril 1993;59: Wassarman PM. The mammalian ovum. In: Knobil E, Neil J, eds. The physiology of reproduction. New York: Raven Press Ltd., 1988: Perreault SD. Chromatin remodeling in mammalian zygotes. Mutat Res 1992;296: Tarin JJ, Pellicer A. Oocyte maturation in human in vitro fertilization programs. Ann Acad Med Singapore 1992;21: Morgan PM, Warikoo PK, Bavister BD. In vitro maturation of ovarian oocytes from unstimulated rhesus monkeys: assessment of cytoplasmic maturity by embryonic development after in vitro fertilization. Biol Reprod 1991;45: Plachot M, Veiga A, Montagut J, de Grouchy J, Calderon G, Lepretre S, et al. Are clinical and biological IVF parameters correlated with chromosomal disorders in early life: a multicentric study. Hum Reprod 1988;3: Acosta AA, Jones GS, Garcia JE, Sandow B, Veeck L, Mantzavinos T. Correlation of human menopausal gonadotropin/human chorionic gonadotropin stimulation and oocyte quality in an in vitro fertilization program. Fertil Steril 1984;41: Rudak E, Dor J, Mashiach S, Nebel L, Goldman B. Chromosome analysis of human oocytes and embryos fertilized in vitro. Ann NY Acad Sci 1985;442: van der Ven HH, Al-Hasani S, Diedrich K, Hamerich U, Lehmann F, Krebs D. Polyspermy in in vitro fertilization of human oocytes: frequency and possible causes. Ann NY Acad Sci 1985;442: Angell RR, Templeton AA, Aitken RJ. Chromosome studies in human in vitro fertilization. Hum Genet 1986;72: Badenas J, Santaló J, Calafell JM, Estop AM, Egozcue J. Effect of the degree of maturation of mouse oocytes at fertilization: a source of chromosome imbalance. Gamete Res 1989;24: Veeck LL, Wortham JWE, Witmyer J, Sandow BA, Acosta AA, Garcia JE, et al. Maturation and fertilization of morphologically immature human oocytes in a program of in vitro fertilization. Fertil Steril 1983;39: Plachot M, Mandelbaum J. Oocyte maturation, fertilization and embryonic growth in vitro. Br Med Bull 1990;46: FERTILITY & STERILITY 417