Three-dimensional ultrasonographic ovarian measurements and in vitro fertilization outcome are related to age

Transcription

1 FERTILITY AND STERILITY VOL. 79, NO. 1, JANUARY 2003 Copyright 2003 American Society for Reproductive Medicine Published by Elsevier Science Inc. Printed on acid-free paper in U.S.A. Three-dimensional ultrasonographic ovarian measurements and in vitro fertilization outcome are related to age Sanja Kupesic, M.D., Ph.D., Asim Kurjak, M.D., Ph.D., Drazena Bjelos, M.D., and Sanja Vujisic, Ph.D. Department of Obstetrics and Gynecology, Medical School University of Zagreb, Sveti Duh Hospital, Zagreb, Croatia Objective: To evaluate whether the number of ovarian antral follicles, ovarian volume, and ovarian stromal blood flow change with age and to prospectively analyze whether three-dimensional ultrasonographic measurements predict ovarian response and IVF outcome. Design: Prospective analysis. Setting: Assisted reproductive unit. Patient(s): Fifty-six consecutive women 22 to 43 years of age with normal basal serum FSH concentrations who were undergoing their first IVF cycle. Main Outcome Measure(seconds): Number of ovarian antral follicles, ovarian volume, and ovarian stromal flow index were determined by three-dimensional and power Doppler ultrasonography. Pretreatment measurements were compared with number of recovered oocytes, fertilization rates, and pregnancy rates. Result(s): As patient age increased, significant trends in ovarian volume, number of follicles, and stromal vascularity decreased. Three-dimensional ovarian measurements and fertilization rates differed significantly among age groups. For each age group, a higher number of antral follicles, greater ovarian volume, and favorable ovarian stromal vascularity was associated with higher number of retrieved oocytes and increased pregnancy rates. Conclusion(s): Increasing patient age is associated with poor ovarian response, as represented by smaller ovarian volume, lower antral follicle count, and poor stromal vascularity. Three-dimensional power Doppler ultrasonography can help to individualize IVF in patients regardless of age. (Fertil Steril 2003;79: by American Society for Reproductive Medicine.) Key Words: Three-dimensional ultrasonography, age, IVF-ET, antral follicle number, ovarian volume, ovarian stromal vascularity Received February 13, 2002; revised and accepted May 22, Reprint requests: Sanja Kupesic, M.D., Ph.D., Department of Obstetrics and Gynecology, Medical School University of Zagreb, Sveti Duh 64, Zagreb, Croatia (FAX: ; sanja.kupesic@ zg.tel.hr) /03/$30.00 PII S (02) Fertility is known to decline in women after the age of 30 years, and striking decreases are observed after 35 years of age. Fecundity is almost completely lost by 45 years of age (1 3). This age-related decline in fertility results from several factors that contribute to overall reproductive failure, including inadequate endometrium and poor oocyte quality. Older women retain uterine receptivity to embryo implantation (3); women of advanced reproductive age can become pregnant after IVF-ET of oocytes donated by younger women (1, 2). Satisfactory pregnancy rates in older women having oocyte donation indicates that fertility is lost with age in women because of a decline in oocyte quality and ovarian aging rather than uterine aging (2, 4). Although several indirect biomarkers have been used to assess ovarian reserve (i.e., day 3 serum FSH levels, day 3 serum estradiol levels, and the clomiphene citrate challenge test), there are currently no reliable, direct, noninvasive tests to examine the follicular status of the human ovary (5, 6). Methods of predicting ovarian responsiveness before ovulation induction would be helpful in counseling patients, predicting treatment outcome, and tailoring the gonadotropins dose (6 9). Currently, chronologic female age (10) and serum FSH level in the early follicular phase 190

2 (11) are the most useful clinical variables for predicting ovarian reserve. However, ovarian reserve does not always directly correlate with chronologic age, and a normal basal FSH level does not necessarily predict optimal response to controlled ovarian hyperstimulation (12). The ovarian antral follicle count as assessed by using transvaginal ultrasonography correlates well with increasing patient age (7, 8). Some women with small ovarian volume on ultrasonography demonstrate poor responses to controlled ovarian hyperstimulation in assisted reproductive programs (9). Doppler assessment of ovarian stromal blood flow in the early follicular phase of spontaneous cycles has been related to ovarian follicular response (13), and measurement of ovarian stromal blood flow after pituitary suppression can predict ovarian responsiveness and the outcome of IVF treatment in patients with normal early follicular phase FSH levels (14). Therefore, ultrasonography may provide an estimate of reproductive potential to help predict the patient s prospects for pregnancy. Since some clinical studies suggest that small ovarian volume, reduced stromal area, and decreased number of antral follicles predict a poor response to controlled ovarian hyperstimulation (6 9), we previously studied whether ovarian antral follicle count, ovarian volume, stromal area, and ovarian stromal blood flow predict ovarian response and IVF outcome (15, 16). We concluded that total number of antral follicles was the best predictive measure of favorable IVF outcome, followed by ovarian stromal flow index, peak E 2 level on the day of hcg administration, total ovarian volume, total ovarian stromal area, and age. In the current study, we used three-dimensional ultrasonography to evaluate the association between antral follicle count, ovarian volume, and ovarian stromal blood flow after pituitary suppression and age. We also correlated threedimensional ultrasonographic measurements with ovulation induction parameters (number of oocytes) and cycle outcome (fertilization and pregnancy rates). MATERIALS AND METHODS We included 56 consecutive patients enrolled in a treatment cycle for IVF-ET between May 1, 2000, and March 31, The local ethics committee approved the study protocol, and all patients gave written informed consent. Institutional review board approval was also obtained. Inclusion criteria were [1] an early follicular phase FSH level 10 miu/ml; [2] presence of both ovaries and lack of ovarian cysts; [3] no history of partial or complete surgical resection of the ovary; [4] no history or signs of ovarian endometriosis or uterine leiomyoma; [5] no hormonal or ultrasonographic features of polycystic ovaries; and [6] no steroidal hormone treatment within 6 months of ultrasonographic evaluation. In addition, all patients were undergoing their first IVF cycle, and all were nonsmokers. Patients received a standard regimen of GnRH agonist. Triptorelin (Decapeptyl; Ferring, Kiel, Germany) was administered s.c. in the mid-luteal phase at a daily dose of 100 g. After confirmation of pituitary down-regulation (no ovarian cysts 1 cm, endometrial thickness 5 mm, and E 2 level 50 pg/ml), transvaginal ultrasonography was performed by using an electronic 7.5-MHz transducer with three-dimensional capability (Combison 530D; Kretztechnique, Zipf, Austria). Three-dimensional ultrasonography was performed exclusively at our department by two experienced sonographers. Three rotatable perpendicular planes were simultaneously displayed. The ability to store three-dimensional data on a hard disk permitted a relatively short examination time (average of 2 to 5 minutes). At analysis, three-dimensional planes could then be rotated and manipulated to obtain the largest and most accurate volume dimensions. Detailed analysis of antral follicle number and ovarian volume was performed after the patient had completed her examination and required 10 to 20 minutes. Ovarian volume was obtained by manually outlining serial parallel sections of the ovary and was calculated by using a trapezoid formula. At least 10 serial slices were taken from each ovary to outline the volume measurement. Each ovarian volume was recorded and then summed to obtain the total ovarian volume. Data were stored digitally on a hard disk (SyQuest; Technology Inc., Fremont, CA) for later analysis. After B-mode analysis, power Doppler imaging was initiated with the volume mode. To decrease acquisition time, the volume of the color box and sweep angle were reduced. The color frame rate was adjusted so that color density and color quality were as low as possible, to obtain a good image. The pulsed repetition frequency was as high as possible to enable display of targeted flow velocity. The spatial peak temporal average intensity was approximately 80 mw/cm 2. Wall filters (50 Hz) were used to eliminate low frequency signals. The average patient examination time for threedimensional power Doppler ultrasonography was 3 minutes. With use of the lowest line density, the average acquisition time was 22 seconds (range, 12 to 32 seconds). At the end of each examination, combined color and gray scale rendering was performed and quantitative analysis of blood flow in the outlined area was achieved by using the color histogram mode. The flow index reflects the intensity of blood flow and was calculated automatically by the built-in computer program as follows: flow index weighted color values/color values. Because no significant differences were noted between the two sides for ovarian stromal blood flow in any patient, the data were calculated together and the average measurement of both sides in an individual patient was used for statistical analysis. The subsequent later analysis of the power Doppler reformatted sections lasted 5 to 10 minutes. FERTILITY & STERILITY 191

3 TABLE 1 Relationship between total antral follicle count, total ovarian volume, mean stromal vascularity, fertilization rate, and pregnancy rate and female age. Age group No. of patients Total antral follicle number Total ovarian volume (cm 3 ) Mean ovarian stromal vascularity (flow index) Fertilization rate (%) Pregnancy rate (%) 30 y (3.00) [7, 16] (1.70) [8.5, 12.7] (0.60) [11.6, 13.4] (18.25) [57, 92] y (1.75) [6, 14] 8.30 (1.20) [7.0, 10.2] (0.475) [11.9, 13.4] (11.42) [71, 100] y (3.00) [2, 9] 6.85 (0.85) [5.8, 8.5] (1.87) [9.6, 12.8] (9.65) [60, 83] y (4.50) [2, 10] 7.20 (1.70) [6.0, 9.0] (0.95) [10.0, 12.4] (11.05) [50, 75] 23.1 P value.001 a.001 a.001 a.001 a.055 (NS) b Note: Data are the median (interquartile range) [minimum, maximum]. a Kruskal Wallis test. b 2 test. The standard starting dose of FSH was 2 to 4 ampoules (150 to 300 IU of FSH), depending on patient age and basal serum FSH concentration. Follicular growth was monitored by using serial ultrasonography, and the dose of FSH was adjusted according to follicular response. When the average diameter of the three leading follicles was at least 18 mm each on ultrasonography, 10,000 IU of hcg was administered as a single injection. Transvaginal ultrasonography directed oocyte retrieval was performed approximately 36 hours after hcg administration. Embryo transfer took place 5 days after oocyte retrieval in the blastocyst stage. Progesterone pessaries (400 mg b.i.d.) (Utrogestan; Laboratories Besins International, Paris, France) were given as luteal support, starting on the day of ET and continuing until 16 days thereafter. Pregnancy was defined as the occurrence of a positive -HCG ( 10 IU) value on day 12 after embryo transfer and a second higher value 2 days later. Only pregnancies reaching -HCG values 100 IU were considered for evaluation. The outcome variables of total antral follicle number, total ovarian volume, and mean FI of the ovarian stromal blood flow were correlated to number of oocytes retrieved and number of clinical pregnancies. Statistical Analysis Data are expressed as medians with interquartile ranges and minimum and maximum values. The Kruskal Wallis test was used to compare total antral follicle count, total ovarian volume, ovarian stromal vascularity, and fertilization rates among age groups. The 2 test was used to compare pregnancy rates among age groups. Comparisons among age, number of oocytes retrieved, and fertilization rate with various total antral follicle counts, total ovarian volumes, and extent of ovarian stromal vascularity were made by using the Kruskal Wallis test. Comparison between pregnancy rates and number of total antral follicles, ovarian volume, and ovarian stromal vascularity was performed by using the 2 test. P.01 was considered statistically significant. The Spearman correlation was used to compare age with antral follicle count, total ovarian volume, and ovarian stromal flow index. Correlation was considered significant at the.01 level (two-tailed). RESULTS Fifty-six women with normal basal serum FSH concentrations were enrolled. Indications for assisted reproduction were tubal factors (n 28), male infertility (n 20), and mixed causes (n 8). The average patient age was 33.5 years (range, 22 to 43 years). Age distribution did not differ between groups. All initiated cycles reached oocyte retrieval, and 22 clinical pregnancies (39.3%) occurred. The clinical pregnancy rate was similar in the three infertility groups. Table 1 shows the relationship of three-dimensional ultrasonographic ovarian measurements to age. Younger women had greater ovarian volume, more total antral follicles, and higher stromal intraovarian vascularity as measured by the ovarian flow index. In contrast, values for these three variables decreased as age increased. We found a statistically significant difference among age groups in total antral follicle count, total ovarian volume, mean ovarian stromal vascularity, and fertilization rate but no significant difference in overall pregnancy rate. Table 2 shows trends of total antral follicle count, total ovarian volume, and mean ovarian stromal vascularity in relation to age and outcome measures (number of oocytes retrieved, fertilization rates, and pregnancy rates). Higher total antral follicle count, greater total ovarian volume, and favorable mean ovarian stromal vascularity (flow index of 11 to 13) correlated with a higher number of retrieved oocytes and improved pregnancy rates. Older patients generally had fewer antral follicles, smaller ovarian volume, and poor ovarian stromal vascularity. Older 192 Kupesic et al. 3-D ovarian measurements, IVF, and age Vol. 79, No. 1, January 2003

4 TABLE 2 Total antral follicle count, total ovarian volume, and mean ovarian stromal vascularity in relation to age, number of oocytes retrieved, fertilization rate, and pregnancy rate. Predictor No. of Pregnancy patients Mean age of the patients a No. of oocytes a Fertilization rate (%) a rate (%) b No. of total antral follicles c (3.00) [39, 43] 4.00 c (1.00) [3, 5] d (16.60) [50, 75] 0.0 d c (5.00) [25, 42] 8.00 c (2.50) [5, 12] d (22.95) [57, 100] 37.9 d c (2.50) [22, 32] c (2.50) [12, 17] d (16.55) [57, 100] 68.8 d Total ovarian volume (cm 3 ) c (3.00) [37, 43] 4.00 c (2.00) [3, 5] d (16.60) [50, 80] 0.0 e c (7.50) [22, 42] 9.00 c (4.00) [4, 15] d (19.55) [57, 100] 45.9 e c (1.50) [25, 31] c (2.50) [14, 17] d (18.10) [57, 100] 62.5 e Mean ovarian stromal vascularity (flow index) c (2.25) [36, 43] 4.00 c (2.00) [3, 6] d (11.20) [50, 80] 0.0 e c (10.75) [22, 43] 9.00 c (4.00) [3, 15] d (20.67) [50, 100] 47.2 e c (3.50) [28, 33] c (5.50) [10, 17] d (19.17) [60, 100] 50 e Note: Data are the median (interquartile range) [minimum, maximum]. a Kruskal Wallis test. b 2 test. c P.001. d Not significant. e Could not be interpreted owing to small frequencies. patients with greater antral follicle count, increased ovarian volume, and adequate ovarian stromal vascularity had better fertilization and pregnancy rates. When patients were classified on the basis of total antral follicle count ( 5, 5 to 10, or 10), these groups differed significantly in patient age. When we compared the number of oocytes retrieved with varying categories of total antral follicle count, total ovarian volume, and mean ovarian stromal vascularity, we found statistically significant differences, and improved oocyte retrieval correlated with increased values for ovarian measurements. No pregnancy occurred in patients with fewer than 5 antral follicles, whereas those with 5 to 10 antral follicles achieved a pregnancy rate of 37.9%. Patients with more than 10 antral follicles achieved a pregnancy rate of 68.8%, which was not statistically significant (Table 2). Patients were also analyzed by categories of ovarian volume ( 7 cm 3,7to10cm 3, and 10 cm 3 ). In general, larger ovarian volume was associated with more oocytes retrieved and higher clinical pregnancy rates. No pregnancy occurred in patients with low mean ovarian stromal vascularity (flow index 11), whereas pregnancy occurred in 17 of 36 patients (47.2%) with a flow index of 11 to 13 and in 5 of 10 patients (50.0%) with a flow index 13. All three-dimensional ultrasonographic variables negatively correlated with chronologic age; this correlation was significant at the.01 level (two-tailed) (Figs. 1, 2, and 3). When we analyzed the relationship between antral follicle count, total ovarian volume, and mean ovarian stromal flow index and age, three-dimensional ultrasonographic variables decreased linearly with advancing patient age. DISCUSSION Reproductive potential and ovarian reserve decrease with advancing female age (5, 6). However, age is not an independent factor that determines outcome in patients undergoing ART, nor is it a consistently reliable predictor for patients younger than 40 years of age. We evaluated trends of three-dimensional ultrasonographic ovarian measurements (total antral follicle count, total ovarian volume, and mean ovarian stromal flow index) in relation to patient age, number of oocytes recovered, and fertilization and pregnancy rates obtained in 56 consecutive women undergoing their first IVF treatment. All patients had normal basal serum FSH concentrations and received the same standard protocol of pituitary down-regulation and gonadotropin induction. Our results suggest that patient age, total antral follicle count, total ovarian volume, and mean ovarian stromal flow index (after pituitary suppression) are related to the subsequent number of mature oocytes retrieved and pregnancy rates. Patients with more than five antral follicles achieved higher number of oocytes retrieved and better pregnancy rates. Tomas et al. (17) and Chang et al. (18) first reported on the prediction of ovarian response by using the antral follicle FERTILITY & STERILITY 193

5 FIGURE 1 Relationship between number of antral follicles obtained in the early follicular phase and chronologic age. count. A study of 166 women undergoing their first IVF cycle after pituitary down-regulation showed that the number of antral follicles present before ovarian stimulation was a better predictor of the ovarian response than was ovarian volume or age alone (17). The number of antral follicles correlated more strongly with the number of recovered oocytes, whereas ovarian volume correlated with the number of antral follicles before the stimulation but not with the number of oocytes. Chang et al. (18) studied 130 women in 149 IVF cycles after 6 cycles of infertility treatment, including ovarian stimulation with or without intrauterine insemination, had failed. They found a significant correlation between the antral follicle count and the number of oocytes retrieved. A higher likelihood of cycle cancellation, lower E 2 concentration, and need for increased gonadotrophin dosage were detected in cycles with fewer than three antral follicles. The total antral follicles count offered the best predictive value of ovarian response in an IVF program, followed by basal FSH level, body mass index, and female age (19). Scheffer et al. (20) evaluated antral follicle counts on transvaginal ultrasonography in relation to age in women with proven natural fertility. Antral follicle count had the clearest correlation with age and decreased by a mean of 4.8% yearly before 37 years of age and 11.7% thereafter. The number of small antral follicles in both ovaries as measured by ultrasonography is thus clearly related to reproductive age and may reflect the size of the remaining primordial follicle pool. Our recent study (15) showed that total number of antral follicles offered the best predictive value for favorable IVF outcome, followed by ovarian stromal flow index, peak E 2 level on the day of hcg administration, total ovarian volume, total ovarian stromal area, and age. Using these six variables, we predicted favorable IVF outcome in 11 of 22 patients and poor IVF outcome in 29 of 34 patients (overall prediction rate, 71.43%). Zaidi et al. (21) first demonstrated the relationship between ovarian stromal blood flow velocity and ovarian follicular response. They found that poor responders had low peak systolic velocity of ovarian blood flow in the early follicular phase. Engmann et al. (14) found that in patients with normal basal FSH concentrations, the mean ovarian stromal peak systolic velocity on the day of pituitary suppression was a better predictor of ovarian responsiveness than was age. 194 Kupesic et al. 3-D ovarian measurements, IVF, and age Vol. 79, No. 1, January 2003

6 FIGURE 2 Relationship between total ovarian volume and chronologic age. The main limitation of previous studies on ovarian stromal blood flow is that they used peak systolic velocity as the main predictor. Accurate measurement of blood flow velocity requires knowledge of the angle of insonation to the blood vessels analyzed. Ovarian vessels inside the ovarian stroma are thin and tortuous, and it is therefore impossible to obtain accurately the angle between the ultrasonographic beam and the vessels. This potential limitation leads to subjective measurements and depends on experience of the sonographer, since he or she should search for the highest velocity of the ovarian stromal vessels. Our data indicate that ovarian stromal blood flow obtained by three-dimensional power Doppler imaging after pituitary suppression accurately reflects total ovarian stromal blood flow. We used a color histogram mode to assess the intensity of ovarian stromal blood flow, which seems to predict increased delivery of gonadotropins to target cells for stimulation of follicular growth. Our study adds to the data indicating the usefulness of three-dimensional and color Doppler ultrasonography in the evaluation of infertile patients (9, 13, 14, 16, 17). Mean ovarian stromal vascularity (flow index) between 11 and 13 is related to better pregnancy rate than are values less than 11 and higher than 13. Deficient intraovarian vascularity may serve as the initial marker of reduced ovarian reserve, which precedes an increased FSH level and reduction of ovarian volume (14). Syrop et al. (9) demonstrated that ovarian volume on transvaginal ultrasonography predicts ovarian response to ovulation induction and clinical pregnancy rates. Decreased total ovarian volume and volume of the smallest ovary were significant variables that predicted peak E 2 level and number of oocytes and embryos. Wu et al. (22) found that three-dimensional ultrasonography facilitates determination of ovarian volume in patients with and without polycystic ovary disease. Furthermore, the volume of the ovary on three-dimensional ultrasonography correlates better with direct measurement of the surgical specimen than does that obtained by two-dimensional ultrasonography (23, 24). Lass et al. (25) proposed that assessment of ovarian size should be an integral part of infertility evaluation. They found a strong association between ovarian volume and ovarian reserve. Small ovaries were associated with poor response to hmg and a very high cancellation rate during IVF. Previous studies have evaluated the effect of aging on ovarian volume measurements in infertile women (9, 25, 26). A significantly positive correlation between age and FSH level, but not ovarian volume, has been found (26). A significant correlation was noted between FSH level and the number of follicles, but not between FSH level and ovarian volume. Women with small ovarian volume ( 3 cm 3 )on FERTILITY & STERILITY 195

7 FIGURE 3 Relationship between mean ovarian stromal flow index and chronologic age. baseline ultrasonography achieved implantation and pregnancy rates similar to those in women with larger ovarian volumes with use of a higher dose of gonadotropin and microdose GnRH agonist stimulation (27). In our patients, we found that a decreased total ovarian volume on three-dimensional ultrasonography was associated with fewer oocytes retrieved and lower fertilization and pregnancy rates. Furthermore, patients with smaller total ovarian volume were significantly older than those with higher ovarian volume. All three-dimensional ultrasonographic ovarian variables tended to decrease with increasing age (Figs. 1, 2, and 3). Patients with greater reproductive age had diminished ovarian reserve (lower total antral follicle count), smaller ovaries (lower total ovarian volume), and poor ovarian stromal vascularity (flow index 11). These findings correlate well with those of other studies reporting that the number of antral follicles decreases with increasing female age (7). The decrease appears to depend on the size of the residual pool of inactive primordial follicles. We believe that three-dimensional ultrasonographic measurements can aid in individualizing the treatment of IVF patients: Those with few antral follicles, small ovarian volume, and poor intraovarian vascularity may benefit from a higher initial dose of gonadotropins and longer duration of treatment. Although the total number of follicles, ovarian volume, and mean ovarian stromal vascularity decrease with age, high pregnancy rates may be achieved in patients older than 40 years of age in whom three-dimensional ultrasonographic variables are considered normal. References 1. Serhal PF, Craft IA. Oocyte donation in 61 patients. Lancet 1989;1: Sauer MV, Paulson RJ, Lobo RA. A preliminary report on oocytes donation extending reproductive potential to women over 40. N Engl J Med 1990;323: Sauer MV, Paulson RJ, Lobo RA. Reversing the natural decline in human fertility: an extended clinical trial of oocytes donation to women of advanced reproductive age. JAMA 1992;268: Navot D, Bergh PA, Williams MA, Garrisa GJ, Guzman I, Sandler B, et al. Poor oocytes quality rather than implantation failure as a cause of age-related decline in female infertility. Lancet 1991;337: Scott RT, Hofmann GE. Prognostic assessment of ovarian reserve. Fertil Steril 1995;63: Wallach EE. Pitfalls in evaluating ovarian reserve. Fertil Steril 1995; 63: Reuss ML, Kline J, Santos R, Levin B, Timor-Tritsh I. Age and the ovarian follicle pool assessed with transvaginal ultrasonography. Am J Obstet Gynecol 1996;174: Baker TG. A quantitative and cytological study of germ cells in human ovaries. Proc R Soc Biol 1963;158: Syrop CH, Willhoite A, Van Voorhis BJ. Ovarian volume: a novel outcome predictor for assisted reproduction. Fertil Steril 1995;64: Kupesic et al. 3-D ovarian measurements, IVF, and age Vol. 79, No. 1, January 2003

8 10. Tan SL, Royston P, Campbell S, Jacobs HS, Betts J, Mason B, et al. Cumulative conception and live birth rats after in vitro fertilization. Lancet 1992;339: Scott R, Toner J, Muasher S, Oehinger S, Robinson S, Rosenwaks Z. Follicle-stimulating hormone levels on cycle day 3 are predictive of in vitro fertilization outcome. Fertil Steril 1989;51: Farhi J, Homburg R, Ferber A, Orvieto R, Ben Rafael Z. Non-response to ovarian stimulation in normogonadotrophic, normogonadal women: a clinical sign of impending onset of ovarian failure pre-empting the rise in basal follicle stimulating hormone levels. Hum Reprod 1997;12: Zaidi J, Barber J, Kyei-Mensah A, Bekir J, Campbell S, Tan SL. Relationship of ovarian stromal blood flow at baseline ultrasound to subsequent follicular response in an in vitro fertilization program. Obstet Gynecol 1996;88: Engmann L, Sladkevicius P, Agrawal R, Bekir JS, Campbell S, Tan SL. Value of ovarian stromal blood flow velocity measurement after pituitary suppression in the prediction of ovarian responsiveness and outcome of in vitro fertilization treatment. Fertil Steril 1999;71: Kupesic S, Kurjak A. Predictors of in vitro fertilization outcome by three-dimensional ultrasound. Hum Reprod 2002;17: Kupesic S. The present and future role of three-dimensional ultrasound in assisted conception. Ultrasound Obstet Gynecol 2001;18: Tomas C, Nuojua-Huttunen S, Matikainen H. Pre-treatment transvaginal ultrasound examination predicts ovarian responsiveness to gonadotrophins in in-vitro fertilization. Hum Reprod 1997;12: Chang MY, Chiang CH, Hsieh TT, Soong YK, Hsu KH. Use of antral follicle count to predict the outcome of assisted reproductive technologies. Fertil Steril 1998;69: Ng EH, Tang OS, Ho PC. The significance of the number of antral follicles prior to stimulation in predicting ovarian responses in an IVF programmed. Hum Reprod 2000;15: Scheffer GJ, Broekmans JM, Dorland M, Habbema JDF, Looman CWN, te Velde ER. Antral follicle counts by transvaginal ultrasonography are related to age in women with proven natural fertility. Fertil Steril 1999;72: Zaidi J, Campbell S, Pittrof R, Kyei-Mensah A, Shaker A, Jacobs HS, et al. Ovarian stromal blood flow in women with polycystic ovaries: a possible new marker for diagnosis? Hum Reprod 1995;6: Wu MH, Tang HH, Hsu CC, Wang ST, Huang E. The role of threedimensional ultrasonographic images in ovarian measurements. Fertil Steril 1998;69: Bonilla-Musoles F, Raga F, Osborne NG, Pellicer A, Blanes J. Use of three-dimensional ultrasound in reproductive medicine. Assist Reprod Rev 1995;5: Kyei-Mensah A, Zaidi J, Pittrof R, Shaker A, Campbell S, Tan SL. Transvaginal three dimensional ultrasound: accuracy of follicular volume measurements. Fertil Steril 1996;65: Lass A, Skull J, McVeigh E, Margara R, Winston RML. Measurement of ovarian volume by transvaginal sonography before ovulation induction with human menopausal gonadotrophin for in-vitro fertilization can predict poor response. Hum Reprod 1997;12: Sharara FI, McClarmrock HD. The effect of aging on ovarian volume measurements in infertile women. Obstet Gynecol 1999;94: Sharara FI, McClarmrock HD. Use of microdose GnRH agonist protocol in women with low ovarian volumes undergoing IVF. Hum Reprod 2001;16: FERTILITY & STERILITY 197