REPRODUCTIVE BIOLOGY. The Jones Institute for Reproductive Medicine, Eastern Virginia Medical School, Norfolk, Virginia

Transcription

1 FERTILITY AND STERILITY VOL. 72, NO. 4, OCTOBER 1999 Copyright 1999 American Society for Reproductive Medicine Published by Elsevier Science Inc. Printed on acid-free paper in U.S.A. REPRODUCTIVE BIOLOGY Perifollicular blood flow Doppler indices, but not follicular po 2, pco 2, or ph, predict oocyte developmental competence in in vitro fertilization Suzanne Huey, M.D., Alfred Abuhamad, M.D.,* Gerardo Barroso, M.D., Ming-I Hsu, M.D., Paul Kolm, Ph.D., Jacob Mayer, Ph.D., and Sergio Oehninger, M.D. The Jones Institute for Reproductive Medicine, Eastern Virginia Medical School, Norfolk, Virginia Received December 29, 1998; revised and accepted April 26, Presented in part at the 54th Annual Meeting of the American Society for Reproductive Medicine, San Francisco, California, October 4 9, Reprint requests: Sergio Oehninger, M.D., The Jones Institute for Reproductive Medicine, Department of Obstetrics and Gynecology, 601 Colley Avenue, Norfolk, Virginia, (FAX: ; * Division of Maternal-Fetal Medicine, Department of Obstetrics and Gynecology. Department of Biostatistics. Department of Obstetrics and Gynecology /99/$20.00 PII S (99) Objective: To assess the relationships among perifollicular blood flow; follicular fluid po 2, pco 2, and ph; oocyte developmental capacity; preimplantation embryo quality. Design: Prospective study. Setting: Academic, tertiary care institution. Patient(s): Unselected, gonadotropin-stimulated IVF cycles. Intervention(s): Color, pulsed Doppler analysis of perifollicular blood flow, and follicular po 2, pco 2, and ph determinations of randomly designated, mapped ovarian follicles. Main Outcome Measure(s): Fertilization and day 3 embryo cleavage and morphology. Result(s): Perifollicular vascularity indices were significantly and negatively correlated with day 3 embryo cleavage. Pulsatility index and S-D ratio also were significantly and negatively correlated with follicular po 2. The same correlation was found between resistance index and the fertilization rate of preovulatory oocytes. No relationship existed between follicular metabolic analysis and fertilization or embryo quality. The resistance index had a sensitivity of 0.57 and a specificity of 0.71 for the prediction of advanced embryo cleavage status. Conclusion(s): Results confirm and extend previous reports demonstrating that color, pulsed Doppler ultrasound analysis of individual preovulatory follicles during IVF therapy may provide an indirect index of the developmental competence of the corresponding oocyte. Although these methods may provide means to select embryos for transfer with the highest implantation potential, the moderate predictive power showed so far may limit their clinical applicability. (Fertil Steril 1999;72: by American Society for Reproductive Medicine.) Key Words: Blood flow, Doppler analysis, embryo cleavage, follicular fluid, IVF, oocyte development, ph, po 2, predictive value The ultimate goal in IVF is to achieve the transfer of high-quality embryos to the uterine cavity, thereby providing the infertile couple with maximal chances of conception. Intense research is being performed to identify a prognostic biomarker(s) of embryo viability and implantation (1) that would allow for the selection of those embryos demonstrating highest developmental competence to be transferred or cryopreserved. Ovarian folliculogenesis, supported by a changing endocrine-paracrine milieu, requires dynamic interactions among the maturing oocyte, the nurturing granulosa cells, and factors present in the follicular fluid. The onset of follicle growth involves physiological, genetic, and metabolic changes in the oocyte as it prepares for the completion of meiosis, the prevention of polyspermy, and the initial stages of embryogenesis after fertilization. Both an adequate oocyte quality and quantity are needed for IVF. The number of recruited growing follicles depends on an inherent ovarian reserve, related to the woman s chronological age and basal serum FSH, LH, E 2, and inhibin B profiles. Oocyte quality, on 707

2 the other hand, is a direct consequence of metabolic processes, chromosomal-spindle apparatus intactness allowing for a fertilizable status, and an adequate genetic control of the mechanisms leading to early embryo growth and differentiation and to inhibition of programmed cell death. Each embryo has a unique developmental potential, and only a relatively small proportion of cleavage stage embryos is competent to implant after IVF and develop through gestation (2, 3). Developmentally lethal defects occur in the female gamete before insemination. Mature oocytes are known to contain numerical chromosomal disorders (aneuploidism) and cytoplasmic structural defects that predispose the fertilized oocyte to developmental failure (3 5). Embryo implantation potential can be correlated with morphological grading and cleaving status, but the overall predictive value of such embryo features is relatively low (6). Embryos showing fragmentation or other morphological defects and retarded growth may also have a high incidence of chromosomal anomalies (7). All these abnormalities may be the result of altered intrafollicular conditions during preovulatory oocyte maturation occurring either naturally or during controlled ovarian hyperstimulation. The extent to which aneuploidies detectable in mature human oocytes are a consequence of chromosomal defects that occur before the arrest of meiosis at the prophase I stage is unknown (4, 5). Clearly, poor culture conditions, including use of suboptimal culture media, may generate or be additive to preexisting defects. Unfortunately, no single factor(s) secreted into the circulation or present in the follicular fluid has been demonstrated to provide definitive prediction of the developmental competence of the oocyte-embryo. Recent reports have addressed this issue by analyzing follicular fluid biochemistry (content of dissolved O 2, growth factors, and ph), granulosa cell behavior in vitro (presence of metabolic products and regulatory proteins), and perifollicular blood flow and by correlating those results with various oocyte-embryo developmental capacities under in vitro conditions and after uterine ET (3 5, 8 12). We have followed these lines of investigation with the ultimate goal to identify noninvasive methods for the selection of follicles-oocytes-embryos with highest developmental potential. For this purpose, we performed a prospective study designed to assess further the relationships among perifollicular blood flow, follicular po 2, pco 2, and ph, and oocyte developmental competence and preimplantation embryo quality. MATERIALS AND METHODS Patients A total of 16 patients undergoing IVF treatment between February and August 1998 were recruited for the study, chosen randomly based on day of the week when follicular aspiration was performed. The institutional review board of Eastern Virginia Medical School approved the study. Patients heard and read detailed explanations of the proposed study before recruitment and gave written consent. The following characteristics describe the patient population included in this study: [1] maternal age ranging from 21 to 39 years; [2] infertility was tubal-peritoneal in origin (n 10) or associated with a male factor (n 6) showing moderate to severe oligoasthenoteratozoospermia but with a total motile sperm fraction recovered after swim-up of more than 1 million spermatozoa; [3] an attempt was made to include follicles of various diameters at the time of Doppler analysis to follow oocyte development until ET or cryopreservation; and [4] all patients underwent controlled ovarian hyperstimulation with a combination of a GnRH agonist (leuprolide acetate, Lupron; Tap Pharmaceuticals, Abbott Park, IL) using a long protocol and recombinant FSH (Gonal-F; Serono Laboratories, Inc., Randolph, MA). Lupron was commenced in the midluteal phase of the preceding cycle at the dose of 0.5 mg/d, reduced to 0.25 mg/d at the 1st day of menses and continued until hcg administration. Follicle-stimulating hormone was initiated on day 3 of the down-regulated cycle at a dose of three to four ampules (75 IU/ampule) daily and then adjusted in an individualized fashion using a step-down protocol. When at least three follicles 16 mm were seen on ultrasound, 10,000 IU hcg were administered intramuscularly followed by transvaginal follicular aspiration performed hours thereafter. On the day before oocyte aspiration (the morning after hcg), the patients underwent an evaluation of perifollicular blood flow measured by transvaginal, color and pulsed Doppler ultrasound (Sequoia 512; Acuson, Mountain View, CA) in two or three randomly designated follicles per ovary per patient. Each ovary was imaged in an axial and longitudinal view by means of real-time endovaginal ultrasound with frequencies ranging from 4 to 8 MHz. Color flow was used to localize perifollicular flow surrounding the selected follicles ensuring that signals assigned to one follicle were not close to another. The pulsed Doppler range gate was placed over the vessel of interest (arterial side), and the recorded velocity waveforms were used for spectral analysis (13). The following Doppler indices were calculated with the software included in the ultrasonography equipment: pulsatility index (PI), a measure of the systolicdiastolic differential of the velocity pulse (PI peak systolic frequency shift [S] end diastolic frequency shift [D]/ temporal mean frequency shift over one cardiac cycle [A]); resistance index (RI), an angle-independent measure of the pulsatility (RI S D/S); and S-D ratio, a simpler index of the pulsatility (14). The peak flow velocity (PVmax) and mean follicular diameter were also recorded. The follicles were mapped in longitudinal and transverse planes by the surgeon who would perform the aspiration and a second observer, and subjective cartoons and a hard copy 708 Huey et al. Perifollicular blood flow and embryo quality Vol. 72, No. 4, October 1999

3 (photograph) of the location of the follicles in each ovary were made. Because of possible changes in ovarian architecture with each aspiration, only three follicles were studied per ovary to ensure accuracy of localization and consistency of data. The blood flow indices were measured consistently by the same ultrasound operator and surgeon. At oocyte retrieval, all study follicles were identified with use of the hard copy of the previous day s ultrasound and the observer s cartoon without difficulty. These follicles were the first ones to be aspirated from each ovary using the standard aspirating needle suction pump ( 100 mm Hg) system routinely used in our program. The content of each individual follicle was aspirated initially into a vacuum-sealed tube devoid of culture medium. This initial gentle aspiration was limited in terms of time and needle manipulation inside the follicle to minimize contamination with blood. After disconnecting the tube, a sample of follicular fluid (usually clear in appearance or minimally contaminated with blood, giving a clear amber color) was withdrawn immediately by capillary action into a sterile 0.8-mm-inside diameter capillary tube (approximate volume aspirated, 32 L). Within 45 seconds of collection, the follicular fluid was processed for ph, po 2, and pco 2 determinations. At the same time, with the aspirating needle remaining in the follicle, a repeat aspiration was performed into a second vacuum-sealed tube containing Dulbecco s phosphate-buffered saline (GIBCO Laboratories, Grand Island, NY) to ensure oocyte collection. The metabolic analysis of the follicular fluid was performed by placing the aspirated sample into a 37 C EC8 cartridge, which was run through a portable clinical analyzer after ensuring quality control with the electronic simulator (i-stat Corporation, Princeton, NJ). Integrity of cartridges was verified before commencement of each study followed by calibration as recommended by the manufacturer. This equipment analyzes po 2 amperometrically and pco 2 and ph by ion-selective electrode potentiometry. The intraassay coefficient of variation for po 2 and pco 2 measurements was 5%, whereas for ph determination it was 0.1%. In preliminary studies, it was determined that under these conditions, po 2 measurements were unaffected up to 2 minutes after aspiration (the manufacturer recommends reading before 3 minutes). Each follicular fluid aspirated as part of the study (as well as the other nonstudy follicles recovered per patient) was analyzed immediately by the embryologist for oocyte identification and classification of maturational status. Oocytes were cultured individually (to properly follow developmental potential until ET or cryopreservation) in 3-mL culture dishes in Ham s F-10 medium (GIBCO Laboratories) supplemented with 7.5% serum synthetic substitute (Irvine Scientific, Santa Ana, CA). Each oocyte was processed individually with laboratory personnel blinded to blood flow indices or metabolic analysis. Oocyte insemination was performed in a standard fashion or through intracytoplasmic sperm injection (ICSI), depending on the diagnosis of male infertility, following techniques described elsewhere (15). After verification of fertilization at approximately hours after insemination, pronuclear embryos were transferred to growth medium (G1 medium) with 0.5 mg/ml human serum albumin (Irvine Scientific) until transfer or cryopreservation in the morning of day 3. Immediately before transfer embryo cleavage stage and morphology scoring (using Veeck s criteria) (16) were recorded. Embryos were either transferred to the uterine cavity with use of a Wallace s catheter (Simcare Ltd., West Sussex, England) or were cryopreserved with 1,2-propranediol. Statistical Analysis The correlation between Doppler indices, metabolic analysis, peak serum E 2 levels, fertilization, and embryo quality (cleavage and morphology) was calculated by analysis of covariance. This analysis takes into account the fact that multiple measures were obtained from each patient. Variance component analysis was used to assess the percent of variance within patients in relation to the total variance for each variable. The predictive value of blood flow parameters for embryo cleavage status and morphology grading was calculated by receiving operating characteristics (ROC) curves. Data are presented as means SD; P.05 was considered statistically significant. RESULTS The overall IVF results of the population studied were as follows: the normal (diploid) fertilization rate was 69.8% (148 fertilized of 212 mature oocytes inseminated); these rates were similar for standard insemination (68.6% [83 of 121]) and ICSI (71.4% [65 of 91]). Because multiple embryos were transferred per attempt and the transferred embryos originated from oocytes aspirated from study and nonstudy follicles, relationships with implantation rates could not be evaluated. Table 1 shows females age, peak serum E 2, follicle diameter, Doppler indices, and metabolic analysis of the 80 follicles studied in the 16 patients. The within-patient variance was low for the RI (5.1%), PI (7.8%), S-D ( 1.0%), and pco 2 (4.6%), but was higher for the po 2 (57.6%). The follicular po 2 varied between 64 and 167 mm Hg, equivalent to 1.3% 3.6% dissolved O 2. Of the 80 study follicles aspirated, 79 oocytes were recovered (73 metaphase II, 3 metaphase I, and 3 prophase I oocytes). The 76 mature oocytes were inseminated or injected (metaphase II oocytes 5 hours after aspiration and metaphase I oocytes 5 hours after extrusion of the first polar body). Fifty-five oocytes had normal, diploid fertilization (2pn), 9 showed abnormal fertilization (4 1pn and 5 3pn), and 12 had no fertilization (0pn). For the 55 embryos assessed on day 3, the average number of blastomeres (cleavage status) was FERTILITY & STERILITY 709

4 TABLE 1 Overall data of the patient and follicle population analyzed. Variable Mean SD Range Age (y) Peak serum E 2 (pg/ml) Follicle diameter (mm) RI PI S-D PVmax (m/s) po 2 (mm Hg) pco 2 (mm Hg) ph Note: PI pulsatility index; PV peak flow velocity; RI resistance index; S-D systolic-diastolic ratio (range, 1 9), and the average morphology score was (range, 1 5). There were no significant differences in embryo quality characteristics (cleavage or morphology) comparing standard IVF insemination with ICSI (data not shown). The results of the correlation analysis between blood flow Doppler indices, metabolic parameters and preimplantation embryo quality are shown in Table 2. All Doppler indices (RI, PI, and S-D) were significantly and negatively correlated with embryo cleavage (this relationship was independent of whether IVF or ICSI had been performed). However, no correlation existed between Doppler analysis results and embryo morphology. In addition, no significant associations were found between the metabolic parameters (po 2, pco 2, or ph) and embryo cleavage or morphology. TABLE 2 Correlation between blood flow and metabolic parameters and embryo quality. Embryo cleavage Embryo morphology r P r P Blood flow RI NS PI NS S-D NS PVmax.06 NS.14 NS Metabolic analysis po 2.11 NS.08 NS pco 2.21 NS.06 NS ph.01 NS.09 NS Note: PI pulsatility index; PV peak flow velocity; RI resistance index; NS not significant. TABLE 3 Correlation between Doppler indices and other variables. Variable RI PI S-D r P r P r P Age (y).04 NS.02 NS.05 NS Peak serum E 2.17 NS.12 NS.06 NS Follicle diameter.06 NS.07 NS.06 NS po pco ph.09 NS.07 NS.06 NS Fertilization NS Note: PI pulsatility index; RI resistance index; S-D systolic-diastolic ratio. Table 3 presents results of the correlation between Doppler indices and other variables examined. Of importance, the Doppler indices PI and S-D showed a significant and negative correlation with follicular po 2 (borderline significance for RI), whereas RI and S-D had a significant and positive correlation with follicular pco 2 (borderline significance for PI). No relationship could be demonstrated for follicular ph. In addition, a significant and negative correlation was found between RI and oocyte diploid fertilization (borderline significance for S-D). Follicular po 2 had a positive and significant correlation with peak serum E 2 levels (r.35, P.009); no correlation was found between po 2 and follicular diameter or fertilization. Because a significant correlation was found between perifollicular blood flow and embryo cleaving status, further analyses were conducted to determine the predictive power of Doppler analysis for embryo quality. Categorizing embryo quality as good or poor ( 8 or 8 blastomeres on day 3), ROC analysis demonstrated that RI, PI, and S-D were moderate predictors of embryo cleavage (Fig. 1). The area under the ROC curve was 0.65, 0.64 and 0.65 for RI, PI, and S-D, respectively, demonstrating a similar predictive value for the three indices. For an RI 0.52 (mean value found in our study), the predictive statistics indicated a specificity of 0.71, a sensitivity of 0.57, a positive predictive value of 0.79, and a negative predictive value of On the other hand, as expected, ROC analysis failed to demonstrate any significant relationship between Doppler indices and embryo morphology scores (area under the curves essentially identical to the 45 line). DISCUSSION We have prospectively analyzed individual ovarian follicle vascularity and microenvironment and assessed the developmental capacity of the corresponding oocyte-embryo. 710 Huey et al. Perifollicular blood flow and embryo quality Vol. 72, No. 4, October 1999

5 FIGURE 1 Receiving operating characteristics curve analysis: predictive value of perifollicular Doppler indices for embryo cleavage. This study confirmed that perifollicular color Doppler analysis performed at a late stage of controlled ovarian hyperstimulation provides an indirect index of oocyte-embryo development in IVF (3 5, 8, 9). Our results extend previous reports by demonstrating the value of not only RI but also PI and S-D ratio for the prediction of fertilization and embryo cleavage competence under in vitro conditions. Doppler indices are markers of downstream impedance to blood flow. In the ovarian perifollicular region, lower Doppler indices are indicative of poor blood flow in the microvasculature that provides oxygen and protein-rich blood to the developing oocyte and cumulus-corona complex (17). Perifollicular blood flow Doppler indices were significantly and negatively correlated with the cleavage status of day 3 embryos. We have demonstrated a significant relationship between follicular blood flow, its po 2 (or content of percent dissolved oxygen) and pco 2, and the capacity of the corresponding oocyte-embryo to develop successfully. Such competence was evident in terms of achievement of normal (diploid) fertilization and the capacity to undergo timely cleavage up to day 3 of culture. Adjacent follicles of the same size may have different oxygen tensions, not readily recognized by routine ultrasound (3). Color Doppler, on the other hand, provides an index of vascularity that also correlates well with measures of metabolic activity such as po 2 and pco 2. However, we could not demonstrate a direct relationship between follicular metabolic analysis (po 2, pco 2,orpH) and oocyte fertilization or embryo quality. Oxygenation appears to be a significant factor in adequate oocyte spindle formation, chromosomal aggregation, oocyte maturity and fertilization, and, ultimately, implantation of the subsequent embryo (3, 18). Although we found no significant correlation with follicular oxygenation on the day before oocyte retrieval, a specifically timed hypoxic event with its associated changes in CO 2 and ph could potentially result in lethal or sublethal (albeit dysfunctional) consequences. Others have reported an association between follicular fluid hypoxia and preimplantation embryo quality (3 5). The cumulus cells maintain high adenosine triphosphate and oxygen requirements for completion of meiosis (19). With growing follicular size, the partial pressure of oxygen falls, as does the ph, demonstrating an insufficient increase in blood flow during follicular maturation (20). In a large sample size study, oocytes from follicles with dissolved oxygen contents 3% tended to result in better fertilization rates and more advanced (6-to-8-cell stage) cleaving embryos before transfer (3, 4). Moreover, oocytes with cytoplasmic defects, disorganized chromosomes, and cleavage stage embryos with multinucleated blastomeres seemed to be derived predominantly from severely hypoxic follicles having a reduced vascularity and lower concentrations of vascular endothelial growth factor (VEGF) (3 5). Nevertheless, other investigators have claimed that elevated follicular fluid VEGF concentrations may be associated with poor conception rates after IVF (12). Perifollicular vascularity was significantly associated with embryo cleavage but not with embryo morphology scores. Although the number of blastomeres and morphological appearance of embryos are used routinely to judge embryo quality in IVF programs, their predictive power for successful implantation is relatively poor (6). Among preimplantation embryo characteristics, we found that day 3 cleavage status is the most predictive of pregnancy (6). This relationship underscores the results herein obtained with perifollicular Doppler indices. Theoretically, Doppler-assessed blood flow indices of individual follicles (a follicle-specific attribute) could be used to select oocytes-embryos with superior implantation potential (3). Notwithstanding the robust association demonstrated, the moderate predictive power observed in our studies by ROC analysis may limit their clinical daily applicability. Nevertheless, the combination of this information with results of other biomarkers, such as angiogenic or vasoregulatory factors, or other modulatory molecules present in the follicular fluid or synthesized early by the zygote-embryo, may enhance our capacity to select the most viable embryos. Also, their relationships with in vitro blastocyst development may prove to be more predictive. Whether perifollicular Doppler indices and/or follicular po 2 will ultimately predict pregnancy remains to be determined by additional prospective clinical trials (3, 4, 8, 9, 21 23). More studies are needed to define the correlation between perifollicular blood flow and follicular homeostasis and to determine how changes in the follicular milieu (hypoxiaacidosis or paracrine-autocrine factors) affect oocyte fertilizability and early embryo development. FERTILITY & STERILITY 711

6 In conclusion, color Doppler ultrasound analysis of individual preovulatory follicles during IVF therapy offers an indirect index of the developmental competence of the corresponding oocyte. Although these methods may provide noninvasive means to select embryos for transfer with the highest implantation potential, their moderate predictive power (at present) may limit their clinical applicability. Acknowledgment: The authors thank Dr. David Gardner for providing the S1 growth medium. References 1. Gardner D, Lane M. Culture and selection of viable blastocysts: a feasible proposition for human IVF? Hum Reprod Update 1997;3: Acosta A, Oehninger S, Morshedi M, Swanson R, Scott R, Irianni F. Implantation potential of each pre-embryo in multiple pregnancies obtained by in vitro fertilization seems to be different. Fertil Steril 1988;50: Van Blerkom J, Antczak M, Schrader R. The developmental potential of the human oocyte is related to the dissolved oxygen content of follicular fluid: association with vascular endothelial growth factor levels and perifollicular blood flow characteristics. Hum Reprod 1997; 12: Van Blerkom J. Can the developmental competence of early human embryos be predicted effectively in the clinical IVF laboratory? Hum Reprod 1997;12: Van Blerkom J. Epigenetic influences on oocyte developmental competence: perifollicular vascularity and intrafollicular oxygen. J Assist Reprod Genet 1998;15: Hsu M, Mayer J, Aronshon M, Lanzendorf S, Muasher S, Oehninger S. Embryo implantation in IVF and ICSI: impact of the number of embryos transferred, morphology grade and cleavage status. Fertil Steril 1999;72: Kligman I, Benadive C, Alikani M, Munne S. The presence of multinucleated blastomeres in human embryos is correlated with chromosomal abnormalities. Hum Reprod 1996;11: Nargund G, Bourne T, Doyle P, Parsons J, Cheng W, Campbell S, et al. Associations between ultrasound indices of follicular blood flow, oocyte recovery and preimplantation embryo quality. Hum Reprod 1996; 11: Chui D, Pugh N, Waler S, Shaw R. Follicular vascularity-the predictive value of transvaginal Doppler ultrasonography in an in vitro fertilization programme: a preliminary study. Hum Reprod 1997;12: Nayudu PL, Lopata A, Jones G, Gook D, Bourne H, Sheather S, et al. An analysis of human oocytes and follicles from stimulated cycles: oocyte morphology and associated follicular fluid characteristics. Hum Reprod 1989;4: Barrionuevo M, Rao P, Graham LI, Yeko T. Follicular fluid nitric oxide level and its impact on oocyte potential. Assist Reprod Rev 1997;7: Friedman C, Seifer D, Kennard E, Arbogast L, Alak B, Danforth D. Elevated level of follicular fluid vascular endothelial growth factor is a marker of diminished pregnancy potential. Fertil Steril 1998;70: Abuhamad A, Mari G, Bogdan D, Evans A III. Doppler flow velocimetry of the splenic artery in the human fetus: is it a marker of chronic hypoxia? Am J Obstet Gynecol 1995;172: Maulik D. Principles of Doppler signal processing and hemodynamic analysis. In: Copel JA, Reed KL, eds. Doppler ultrasound in obstetrics and gynecology. New York: Raven Press, 1995; Oehninger S, Veeck L, Lanzendorf S, Maloney M, Toner J, Muasher S. Intracytoplasmic sperm injection (ICSI): achievement of high pregnancy rates in couples with severe male factor infertility is dependent upon female and not male factor. Fertil Steril 1995;64: Veeck LL. Atlas of the human oocyte and early conceptus. Vol. 2. Baltimore: Williams and Wilkins, 1991: Gordon J, Shifren J, Foulk R, Taylor R, Jaffe R. Angiogenesis in the human female reproductive tract. Obstet Gynecol Surv 1995;50: Gaulden M. Maternal age effect: the enigma of Down syndrome and other trisomic conditions. Mutat Res 1992;296: Crisp T. Organization of the ovarian follicle and events in its biology: oogenesis, ovulation or atresia. Mutat Res 1992;296: Fischer B, Kunzel W, Gips H. Oxygen tension in follicular fluid falls with follicle maturation. Eur J Obstet Gynecol Reprod Biol 1992;43: Balakier H, Stronell R. Color Doppler assessment of folliculogenesis in in vitro fertilization patients. Fertil Steril 1994;62: Weiner Z, Thaler I, Levron J, Lewit N, Itskovitz-Eldor J. Assessment of ovarian and uterine blood flow by transvaginal color Doppler in ovarian-stimulated women: correlation with the number of follicles and steroid hormone levels. Fertil Steril 1993;59: Tekay A, Martikainen H, Jouppile P. Blood flow changes in uterine and ovarian vasculature, and predictive value of transvaginal pulsed colour Doppler ultrasonography in an in-vitro fertilization programme. Hum Reprod 1995;10: Huey et al. Perifollicular blood flow and embryo quality Vol. 72, No. 4, October 1999