Value of ovarian stromal blood flow velocity measurement after pituitary suppression in the prediction of ovarian responsiveness and outcome of in vitro fertilization treatment

FERTILITY AND STERILITYt VOL. 71, NO. 1, JANUARY 1999 Copyright © 1998 American Society for Reproductive Medicine Published by Elsevier Science Inc. Printed on acid-free paper in U.S.A.

Value of ovarian stromal blood flow velocity measurement after pituitary suppression in the prediction of ovarian responsiveness and outcome of in vitro fertilization treatment Lawrence Engmann, M.D.,*† Povilas Sladkevicius, M.D.,*‡ Rina Agrawal, M.D.,*§ Jinan S. Bekir, M.D.,* Stuart Campbell, M.D.,*‡ and Seang Lin Tan, M.D.*† The London Women’s Clinic, London, United Kingdom and McGill University, Montreal, Canada

Received March 18, 1998; revised and accepted August 23, 1998. Presented at the 16th World Congress on Fertility and Sterility and the 54th Annual Meeting of the American Society for Reproductive Medicine, San Francisco, California, October 3–9, 1998. Reprint requests: Lawrence Engmann, M.D., Department of Obstetrics and Gynecology, McGill University, Royal Victoria Hospital, Women’s Pavilion, 687 Pine Avenue West, Montreal, H3A 1A1, Canada (FAX: 514-8431496, E-mail: lengmann @rvhob2.lan.mcgill.ca). * The London Women’s Clinic. † Department of Obstetrics and Gynecology, Royal Victoria Hospital, McGill University, Montreal, Quebec, Canada. ‡ St. George’s Hospital Medical School, London, United Kingdom. § The Middlesex Hospital, London, United Kingdom. 0015-0282/98/$19.00 PII S0015-0282(98)00406-3

22

Objective: To evaluate whether ovarian stromal blood flow velocity after pituitary suppression is predictive of ovarian response and the outcome of IVF treatment in patients with normal basal serum FSH levels and to compare the predictive value of this test with age, early follicular phase serum FSH level, E2 level, and FSH:LH ratio. Design: Prospective observational study of women undergoing IVF treatment. Setting: A tertiary referral center for assisted reproduction. Patient(s): Eighty-eight women who received the long buserelin acetate treatment protocol. Intervention(s): Transvaginal color and pulsed Doppler measurement of the ovarian stromal peak systolic velocity (PSV) after pituitary suppression and measurement of the basal serum FSH level, E2 level, and FSH:LH ratio. Main Outcome Measure(s): Number of mature oocytes retrieved and pregnancy rate. Result(s): Ovarian stromal PSV was the most important single independent predictor of ovarian response in patients with a normal basal serum FSH level, when compared with age, basal FSH level, E2 level, or FSH:LH ratio. Patients in group 2 (PSV $10 cm/s) had a significantly higher median number of mature oocytes retrieved (11 versus 5.5) and a higher clinical pregnancy rate (35.3% versus 11.3%) than patients in group 1 (PSV ,10 cm/s), even after controlling for age. Conclusion(s): Ovarian stromal blood flow velocity, after pituitary suppression is confirmed, is predictive of ovarian responsiveness and the outcome of IVF treatment. (Fertil Sterilt 1999;71:22–9. ©1998 by American Society for Reproductive Medicine.) Key Words: Color Doppler, ovarian stromal blood flow, predictive value, ovarian response, in vitro fertilization

Despite numerous improvements that have been made in ovarian stimulation protocols (1) and in the management of male factor infertility during IVF treatment, the embryo implantation rate has remained fairly stable at 10%– 15%. Endometrial receptivity and embryo quality are both essential factors for the success of implantation, and recent experience from oocyte donation programs supports the notion that oocyte quality is an important determinant of reproductive outcome (2). The success of IVF treatment is therefore dependent on the ability of the ovary to re-

spond to controlled stimulation by gonadotropins and to develop a reasonable number of mature follicles and oocytes simultaneously. Failure to respond is associated with cancellation of the cycle or poor outcome of treatment. Prior prediction of the likelihood of optimal ovarian response is therefore essential in identifying patients who are most likely to benefit from IVF treatment. Chronologic female age (3) and early follicular phase serum FSH level (4) are the most useful parameters for predicting ovarian reserve. Ovarian reserve, however, does not al-

ways correlate with chronologic age. Moreover, although an elevated basal FSH level has a good predictive value, a normal level does not necessarily predict an optimal response to controlled ovarian stimulation (5). There are conflicting reports regarding the predictive value of the early follicular phase E2 level (4, 6) and the FSH:LH ratio test (7). Other tests, such as the clomiphene citrate challenge test (8), the exogenous FSH ovarian reserve test (9), and the GnRH agonist stimulation test (10), also have been evaluated in an attempt to improve the prediction of ovarian response, but none has gained wide acceptance. Transvaginal color and pulsed Doppler ultrasound (US) has made feasible the noninvasive evaluation of vascular changes during the menstrual and stimulated cycles (11, 12). More important, the technique has been used to evaluate uterine receptivity in women undergoing IVF treatment (13– 15). Although Doppler assessment of ovarian stromal blood flow in the early follicular phase of the spontaneous menstrual cycle has been related to ovarian follicular response (16), it remains uncertain whether the measurement of ovarian stromal blood flow after pituitary suppression in women undergoing IVF treatment is useful. The primary purpose of this study was to evaluate the role of ovarian stromal blood flow velocity measurement, after pituitary suppression has been confirmed, in predicting ovarian responsiveness and the outcome of IVF treatment in patients with normal early follicular phase FSH levels. The secondary aim was to compare the ability of this test to predict ovarian response with other parameters of ovarian reserve, such as female age, basal serum FSH level, E2 level, and FSH:LH ratio.

MATERIALS AND METHODS Study Population The study included 88 patients who were attending The London Women’s Clinic for IVF treatment over a 10-month period. It was approved by the clinic’s institutional review board. Only patients who received the long protocol for pituitary suppression, who had an early follicular phase FSH level of ,10 mIU/mL, and who did not have any uterine fibroids, ovarian cysts, or ovarian endometriomas were recruited for the study. Basal serum FSH was considered abnormal if the day 3 concentration exceeded 10 mIU/mL (8). The median age of the patients was 34 years (range, 27– 47 years). The causes of infertility were male factor (56.8%), tubal disease (13.5%), unexplained (13.5%), and multiple factors (16%).

Treatment Protocol All patients underwent transvaginal US to assess uterine and ovarian morphology on day 2 or day 3 of the menstrual cycle, and SC administration of a 500-mg daily dose of the GnRH agonist buserelin acetate (Suprefact; Hoechst, Hounslow, United Kingdom) was begun. A US scan was perFERTILITY & STERILITYt

formed after 2 weeks to confirm that pituitary suppression had occurred, as shown by the absence of follicular activity and an endometrial thickness of ,5 mm. Once pituitary suppression was achieved, the administration of gonadotropin was commenced and the dose of buserelin acetate was reduced to 200 mg/d. The latter was continued until, and including, the day of hCG (Pregnyl, Organon, Cambridge, United Kingdom) administration. The standard starting dose of FSH or hMG was 2–5 ampules (150 –375 IU of FSH activity) depending on the patient’s age, previous response to ovarian stimulation, and early follicular phase serum FSH level, and on the presence or absence of polycystic ovaries (PCO) on US assessment. Follicular growth was monitored with serial US scans, and the dose of FSH or hMG was adjusted according to the follicular response. When the average diameters of the two or three leading follicles were at least 18 mm, as measured by US, hCG (10,000 IU) was administered as a single injection. Transvaginal US-directed oocyte retrieval was performed approximately 36 hours after hCG administration, and ET took place 48 hours after oocyte retrieval. The methods used for oocyte retrieval and ET have been described previously (17). Oocyte maturity was classified according to the criteria described by Veeck et al. (18). Progesterone pessaries (400 mg twice daily, Cyclogest; Hoechst) were given as luteal support, starting on the day of ET and continuing until 16 days thereafter. Fresh ET was not offered to patients who had serum E2 levels of .3,450 pg/mL on the day of hCG administration or E2 levels between 2,725 pg/mL and 3,540 pg/mL with .15 oocytes retrieved because of the increased risk of ovarian hyperstimulation syndrome (OHSS). All embryos for such patients were cryopreserved and transferred in a subsequent cycle.

Doppler Ultrasonography Transvaginal color and pulsed Doppler measurements of the ovarian stromal arteries were performed on the day that pituitary suppression was confirmed (approximately 2 weeks after buserelin acetate administration). All US scans were performed with the use of a 5-MHz endovaginal transducer for B-mode and color and pulsed Doppler examinations (Acuson Corporation, Mountain View, CA). The spatial peak temporal average intensity for B-mode and Doppler examinations was ,50 mW/cm2, which is well within the safety limits recommended by the Bioeffects Committee of the American Institute of Ultrasound in Medicine (19). Ultrasound assessments were performed with the patient in the lithotomy position and between 8 AM and 12 PM to minimize the influence of the circadian variation on US measurement (20). The coefficient of variation for ovarian stromal blood flow velocity was 11.9% (16). The arteries within the ovarian stroma were visualized 23

with the color Doppler technique, ensuring that no artery near the surface was measured. The blood flow velocity waveforms were obtained by placing the Doppler gate over the colored areas and activating the pulsed Doppler function. No correction was made for the angle of insonation in the vessels of the ovarian stroma; however, the highest achievable signals were sought. Blood flow velocity waveforms illustrating the frequency and intensity of the shifted Doppler frequencies were demonstrated. The arterial Doppler shift spectra were recorded on a video and analyzed, on a subsequent occasion, from an envelope of three uniform consecutive cardiac cycles, and the resulting values were averaged. The waveforms were characterized by the peak systolic velocity (PSV).

Endocrine Assessment Blood samples for serum FSH, LH, and E2 were obtained, for the purposes of the study, on day 2 or day 3 of the menstrual cycle and on the day of pituitary suppression. Serum FSH and LH levels were measured by microparticle enzyme immunoassay (Abbott AxSYM reagent pack; Abbott Laboratories, Chicago, IL) and serum E2 levels were measured by RIAs (Sorin Clinical Assays and Coated tubes; Diagnostic Products Corporation, Los Angeles, CA). The intra-assay and interassay coefficients of variation were 4% and 7.5% for FSH, 4% and 7.5% for LH, and 6% and 7.5% for E2, respectively.

Outcome Measures The primary outcome measure was the number of mature oocytes retrieved. The age of the patient, day 3 serum FSH level, E2 level, FSH:LH ratio, and ovarian stromal PSV were used to predict the number of mature oocytes retrieved. To investigate the effect of adequate or diminished ovarian blood flow velocity after pituitary suppression on ovarian responsiveness and cycle outcome, the patients were subdivided into two groups depending on the mean ovarian stromal PSV, with a cutoff level of 10 cm/s (derived from the mean ovarian PSV of the whole population). Group 1 consisted of patients who had a mean PSV of ,10 cm/s and group 2 consisted of patients who had a mean PSV of $10 cm/s. The two groups were compared in terms of duration and dose of gonadotropin administration; number of follicles, oocytes, and embryos produced; cumulative embryo score; implantation rate; and pregnancy rate (PR). The cumulative embryo score was calculated by assessing the number of blastomeres and the morphologic grade of each embryo. The morphologic grade of each embryo was multiplied by the number of blastomeres to produce a quality score of each embryo. The scores of all embryos transferred per patient were added to obtain the cumulative embryo score (21). Clinical pregnancy was defined as a positive urine b-hCG test result with US evidence of a gestational sac. The implantation rate was defined as the number of gestational sacs, 24

Engmann et al.

Predictive value of ovarian stromal PSV

as assessed by US at 6 weeks’ gestation, divided by the number of embryos transferred for each patient.

Statistical Analysis All data initially were entered into a patient’s data file and subsequently were transferred into a computer and analyzed with the use of the Stata statistical package (STATACORP, Stata Statistical software, release 5.0; Stata Corporation, College Station, TX). Double data entry was used to minimize errors. The paired t-test was used to assess any differences in ovarian Doppler indices between the left and right ovaries. Data relating to the prediction of ovarian response and other continuous variables were analyzed with the use of a linear regression approach. Data relating to groups 1 and 2 also were analyzed with the use of a multiple linear regression approach, controlling for potential confounding factors such as age. Differences in ovarian response between the two groups also were controlled for PCO status because women with PCO have an exaggerated response to gonadotropin stimulation (22). Data relating to other categorical outcome variables, such as pregnancy, were analyzed with the use of a logistic regression approach, controlling for age. A log transformation was used for continuous variables that were not normally distributed. Statistical significance was determined at the 5% level (P,.05).

RESULTS A total of 88 patients initially were recruited but 7 patients were excluded because of raised FSH levels (.10 mIU/mL). Therefore, 81 patients were used for the analysis. No cycles were abandoned because of poor response. There were 3 patients who did not undergo ET and were excluded from the analysis of clinical PRs. These included 2 patients whose embryos were cryopreserved electively after oocyte retrieval because of an increased risk of OHSS and 1 with a cycle that resulted in failed fertilization. There were no statistically significant differences in the Doppler measurements between the right and left ovaries, so the mean PSV was calculated for each patient and used for subsequent analysis.

Degree of Pituitary Suppression and Ovarian PSV To assess whether there was an association between the degree of pituitary suppression and ovarian blood flow velocity, we assessed the differences between the mean ovarian PSV and the duration of pituitary suppression and the endocrine parameters of pituitary suppression. There were no statistically significant differences between groups 1 and 2 in the mean duration of pituitary suppression (16.3 days versus 17.2 days; P 5 .30). There also were no statistically significant differences between groups 1 and 2 in the median serum FSH level (3.3 mIU/mL versus 3.8 mIU/mL; P 5 .13), LH level (5.3 mIU/mL versus 5 mIU/L; P 5 .46), and Vol. 71, No. 1, January 1999

TABLE 1 Prediction of the number of mature oocytes retrieved based on age, basal serum FSH level, E2 level, FSH:LH ratio, and ovarian stromal PSV. Characteristic Patient age Day 3 FSH level (mIU/mL) Day 3 E2 level (pg/mL) Day 3 FSH:LH ratio Ovarian stromal PSV (cm/s)

Model 1* P value

Model 2† P value

.22 .03 .69 .23 .0003

.64 .17 .91 .56 .001

Note: PSV 5 peak systolic velocity. The number of mature oocytes retrieved, patient age, and D3 E2 level were analyzed on the log scale. * Based on univariate analysis. † Based on analysis adjusted for all other variables in the model. Regression equation: log(no. of mature oocytes) 5 0.85 1 0.28 3 log(age) 2 0.05 3 FSH 1 0.02 3 log(E2) 2 0.08 3 FSH:LH 1 0.07 3 PSV (see text).

E2 level (20.5 pg/mL versus 27.9 pg/mL; P 5 .70) on the day of pituitary suppression.

Ovarian Response and Parameters for Predicting Response We attempted to predict the number of mature oocytes retrieved based on patient age, basal serum FSH level, E2 level, FSH:LH ratio, and ovarian stromal PSV with the use of multiple linear regression analysis (Table 1). Only the basal FSH level and ovarian stromal PSV significantly predicted the number of mature oocytes retrieved, with the use of univariate analysis (model 1). However, when we controlled for all the other factors in the model, ovarian stromal PSV was the only factor that was still highly associated with the number of mature oocytes retrieved (model 2). Ovarian stromal PSV therefore is likely to be an important independent predictor of ovarian response. When all the other predictive factors were held constant, the number of mature oocytes retrieved increased by 7% for

each additional 1-cm/s increase in ovarian stromal PSV: {No. of mature oocytes 5 Exp[0.85 1 0.28 3 log(age) 2 0.05 3 FSH 1 0.02 3 log(E2) 2 0.08 3 FSH:LH 1 0.07 3 PSV]}. Controlling for all the other predictive factors in the model, ovarian stromal PSV accounted for 13% of the variation in the number of mature oocytes retrieved (R 2 5 0.13, P 5 .001). The parameters for predicting ovarian response were compared between poor responders (n 5 5) and normal responders (n 5 76). Poor ovarian response was defined as the retrieval of ,4 mature oocytes and normal response was defined as the retrieval of $4 mature oocytes. Patients who responded poorly to gonadotropins had a significantly lower mean ovarian stromal PSV (7.4 cm/s versus 11.1 cm/s; P 5 .01) and a significantly higher median basal FSH concentration (8.7 mIU/mL versus 6 mIU/mL; P 5 .02) than normal responders. However, median age (39 years versus 34 years), basal E2 level (45.5 pg/mL versus 36.5 pg/mL), and FSH:LH ratio (1.5 versus 1.2) were not significantly different between the poor and normal responders.

Adequate and Diminished Ovarian PSV The baseline characteristics of groups 1 and 2 are given in Table 2. Patients in group 2 were significantly younger than those in group 1 but had similar basal serum FSH levels. There were no statistically significant differences between groups 1 and 2 in the causes of infertility (male factor, 57.1% versus 56.6%; tubal factor, 7.2% versus 17%; unexplained, 21.4% versus 9.4%, and multiple factors, 14.3% versus 17%). More mature oocytes were retrieved for patients in group 2 than for those in group 1 (Table 3). The number of follicles produced and the number of embryos created also were significantly higher for patients in group 2 than for those in group 1. After adjusting for patient age and PCO status, there still were statistically significant differences between the two groups in the number of follicles produced (P 5 .02), the number of mature oocytes retrieved (P,.01), and the number of embryos created (P 5 .02). Although patients in

TABLE 2 Baseline characteristics of patients with ovarian stromal PSV of ,10 cm/s and $10 cm/s. Characteristic Age* No. of previous attempts* Duration of infertility* (y) Body mass index* (kg/m2) Presence of polycystic ovaries (%) Day 3 FSH level (mIU/mL) Day 3 E2 level* (pg/mL) Day 3 FSH:LH ratio

Group 1 (PSV ,10 cm/s) (n 5 28)

Group 2 (PSV $10 cm/s) (n 5 53)

P value

37 (32–41) 1.5 (1–3) 5.0 (3–7) 23.1 (21.4–25.9) 21.4 6.1 (4.4–8.2) 31.3 (18.6–37.9) 1.4 (0.8–2.0)

34 (32–35) 1.0 (0–2) 6.0 (3–7) 23.4 (20.7–25.7) 49.1 6.0 (5.3–7.1) 32.5 (23.9–43.8) 1.2 (0.8–1.7)

.02 NS NS NS .01 NS NS NS

Note: Values are medians with 25th–75th percentiles (i.e., central 50% of distribution) in parentheses, unless otherwise stated. NS 5 not significant. * Analyzed on the log scale because of data that were not normally distributed.

FERTILITY & STERILITYt

25

TABLE 3 Characteristics of ovarian response in patients with ovarian stromal PSV of ,10 cm/s and $10 cm/s. Characteristic Duration of ovarian stimulation (d)† No. of ampules of hMG† No. of follicles† Endometrial thickness† (mm) No. of oocytes† No. of mature oocytes† Fertilization rate (%) Cleavage rate† (%) No. of embryos produced† No. of embryos transferred† Cumulative embryo score Mean implantation rate (%, 6SD) Percent pregnancy rate (no. of pregnancies/no. of ETs) Percent live birth rate (no. of live births/no. of ETs)

Group 1 (PSV ,10 cm/s) (n 5 28)

Group 2 (PSV $10 cm/s) (n 5 53)

P value

P value*

12 (11–13) 51 (28–75) 14 (9–22) 11.1 (9.7–12.6) 10 (7.5–14.5) 5.5 (2–8) 61 (48–76) 100 (91–100) 4 (2–7) 3 (2–3) 24 (15–39) 8.6 6 2.7 11.1 (3/27) 11.1 (3/27)

11 (10–13) 35 (26–46) 20 (14–30) 10.3 (9.8–11.8) 13 (10–17) 11 (7–15) 70 (50–83) 100 (76–100) 7 (4–10) 3 (2–3) 32 (24–36) 21.1 6 3.3 35.3 (18/51) 31.4 (16/51)

NS .03 .001 NS .007 .003 NS NS .001 NS NS NS .02 .04

NS NS .006 NS .02 .006 NS NS .005 NS NS NS .04 .09

Note: Values are medians with 25th–75th percentiles (i.e., central 50% of distribution) in parentheses, unless otherwise stated. NS 5 not significant. * Adjusted for age. † Analyzed on the log scale because of data that were not normally distributed.

group 2 required significantly fewer ampules of gonadotropins for ovarian stimulation, this difference was no longer statistically significant after adjusting for patient age. The clinical PR for patients in group 2 was significantly higher than for patients in group 1 (35.3% versus 11.1%; P 5 .02). The difference was still statistically significant after adjusting for patient age. The live birth rate for group 2 also was significantly higher than for group 1, but this difference was no longer statistically significant after adjusting for patient age. The predictive value of an abnormal result (i.e., the probability that no pregnancy would result if the ovarian PSV was ,10 cm/s) was 89% (24/27). When we restricted our analysis to patients with normal ovaries and excluded patients with PCO, there were still statistically significant differences between the two groups. Patients with normal ovaries in group 2 still produced a higher median number of follicles (15.5 versus 13; P 5 .04), mature oocytes (11 versus 7; P 5 .03), and embryos (6 versus

3; P,.01), and they had a higher PR (25.9% versus 4.5%; P 5 .04) than patients with normal ovaries in group 1. The patients in groups 1 and 2 were subdivided further into two age groups (,37 years old and $37 years old) (Table 4). Among those patients who were ,37 years old, there were 13 patients in group 1 and 43 patients in group 2, and among those who were $37 years old, there were 15 patients in group 1 and 10 patients in group 2. Patients in group 1 had fewer mature oocytes retrieved in both age groups compared with patients in group 2. However, the differences were not statistically significant. Further, although the PR for group 1 was almost half that for group 2 in each of the age groups, the differences were not statistically significant because of the small group sizes.

DISCUSSION The results of this study suggest that there is an association between ovarian stromal blood flow PSV after pituitary

TABLE 4 Ovarian response and outcome of treatment in groups 1 and 2 according to age group. No. of mature oocytes*

Age group ,37 y (n 5 56) $37 y (n 5 25)

Pregnancy rate per ET (%)

Group 1 (PSV ,10 cm/s)

Group 2 (PSV $10 cm/s)

Group 1 (PSV ,10 cm/s)

Group 2 (PSV $10 cm/s)

9.0 (6.5–11.5) 7.0 (4.0–12.0)

11.0 (7.0–16.0) 10.5 (7.0–15.0)

2/12 (16.7) 1/15 (6.7)

16/41 (39.0) 2/10 (20.0)

Note: Values are medians with 25th–75th percentiles (i.e., central 50% of distribution) in parentheses, unless otherwise stated. There were no statistically significant differences between any of the values for groups 1 and 2. * Analyzed on the log scale because of data that were not normally distributed.

26

Engmann et al.

Predictive value of ovarian stromal PSV

Vol. 71, No. 1, January 1999

suppression and the number of mature oocytes retrieved and that this association is independent of the age of the patient. In patients with normal basal serum FSH levels, it is the most important single independent predictor of ovarian response to gonadotropin stimulation, and diminished ovarian stromal blood flow velocity is associated with the retrieval of a lower number of mature oocytes and a lower PR. Zaidi et al. (16) were the first to show that there was a relation between ovarian stromal blood flow velocity and ovarian follicular response. They measured the ovarian stromal PSV in the early follicular phase and showed that poor responders had low ovarian blood flow PSV. The current study confirms the predictive value of ovarian PSV but differs from the earlier study in that the measurements were performed after pituitary suppression was confirmed. This is more useful in clinical practice because it generally is routine to perform a US scan after the administration of GnRH agonists in the long protocol to assess pituitary suppression. It may be argued that the relation between poor ovarian blood flow after pituitary suppression and poor ovarian response may be due to the effect of excessive GnRH agonist therapy in some women, leading to excessive pituitary suppression. Although a recent retrospective study showed that a longer duration of pituitary suppression resulted in the retrieval of a comparable number of oocytes but a lower PR than a shorter duration of pituitary suppression (23), several studies have confirmed that a longer duration of pituitary suppression is not detrimental and in fact might be beneficial to certain groups of patients (24, 25). Nevertheless, we found that there were no differences in the duration of pituitary suppression and in the serum FSH, LH, and E2 levels after pituitary suppression between patients with diminished and those with adequate ovarian blood flow velocity. We can only speculate that the ovarian stromal blood flow velocity after 2–3 weeks of pituitary suppression is a true representative of baseline ovarian blood flow because the ovaries are in a quiescent state. The primordial follicles in the ovary have no independent capillary network, lying simply among vessels of the stroma (26), and therefore depend on their proximity to the stromal vessels for the delivery of nutrients and hormones. The subsequent growth of primary follicles leads to the acquisition of a vascular sheath through the process of angiogenesis. The administration of a GnRH agonist suppresses follicular activity and consequently the ovaries become inactive; ovarian stromal blood flow at this time might be at its lowest and may truly reflect the baseline ovarian blood flow. Chronologic age is the most widely used and simplest parameter for evaluating ovarian responsiveness and fertility potential, and it is well documented that female fertility declines after 35 years of age (3). Our data suggest that, in patients with normal basal FSH concentrations, the mean ovarian stromal PSV on the day of pituitary suppression is a better predictor of ovarian responsiveness than is age. FERTILITY & STERILITYt

The onset of diminished ovarian reserve and the decline in reproductive potential is highly variable and can occur in younger patients (27). It is therefore worth noting that the PRs for patients who were ,37 years of age and who had diminished ovarian PSV were almost one-half of those for patients who had adequate ovarian PSV, although the difference was not statistically significant because of the small number of patients studied (Table 4). Therefore, there was a tendency toward a decrease in ovarian responsiveness and PRs associated with inadequate ovarian blood flow, which was generally poor irrespective of the patient’s age group. However, there was also a tendency toward a decline in ovarian responsiveness and PRs with age, which was not explained entirely by ovarian blood flow velocity alone. Early follicular phase serum FSH concentrations increase with declining ovarian function (28), and elevated levels are highly predictive of ovarian reserve (4). The main limitation of this test is the fact that routine testing does not always predict poor ovarian response, and the presence of wide intercycle variation in basal FSH levels, especially in patients with diminished ovarian reserve (29), might affect the predictive value of the test. In a study of 12 patients who initially had had a normal basal FSH concentration but failed to respond to ovarian stimulation, Farhi et al. (5) showed that within 19 months, these patients exhibited raised serum FSH concentrations and entered a state of menopause. A normal basal FSH test result, therefore, does not necessarily imply optimal ovarian response to gonadotropin stimulation. The group of patients we studied had normal basal FSH levels, and the data suggest that for this group of patients, ovarian PSV is an independent predictor of ovarian responsiveness. Our data also suggest that the predictive value of basal serum E2 levels is limited, and this is consistent with the findings of other investigators (4). The correlation between ovarian blood flow velocity and the outcome of IVF is not surprising because adequate ovarian blood flow is a prerequisite for normal ovarian function (26, 30). Animal studies have confirmed that increased follicular vascularity may be a primary determinant of follicular dominance and that dominant follicles have an increased uptake of serum gonadotropins (30). Patients in this study who had adequate ovarian blood flow velocity performed better in all aspects of ovarian stimulation and treatment outcome. Adequate ovarian blood flow velocity therefore may be associated with an increased delivery of gonadotropins to the target cells for stimulation of follicular growth. In addition to the fact that more oocytes were retrieved from these patients, the quality of the oocytes and of the embryos may be better, as evidenced by the higher PR. It is not entirely clear whether poor ovarian response is always related to follicular depletion (31), a decrease in gonadotropin responsiveness, or an autoimmune phenomenon (32). This study offers new insights into the physiologic 27

basis of poor response in patients with normal basal FSH levels. The decline in ovarian responsiveness may be related to changes in blood flow to the ovarian stroma. Patients with extensive pelvic adhesions (33), a history of pelvic surgery, or ovarian endometriosis (34) are known to respond poorly to gonadotropin stimulation. It is conceivable that disruption of the blood supply to the ovaries as a result of extensive adhesions or surgery impairs the delivery of gonadotropins, which has an adverse effect on the recruitment of follicles. Ovarian vascularization may be affected by increasing age or a disease process, and it is possible that diminished ovarian blood flow velocity is the initial marker of reduced ovarian reserve that precedes even an increased FSH level. One therefore can envisage that inadequate ovarian blood flow might indicate a depleted ovarian reserve that eventually will lead to nonresponse and possibly ovarian failure. Some investigators have suggested increasing the dose of gonadotropins in patients with poor ovarian reserve to improve the ovarian response (35). However, if the basic underlying pathophysiology in this group of patients is reduced blood flow to the ovaries, starting ovarian stimulation with an increased dose of gonadotropins may be of limited use in improving follicular response. It is therefore not surprising that some investigators have failed to show that increasing the dose of gonadotropins in poor responders has a significant beneficial effect (36, 37). Further, patients in this study who had diminished ovarian blood flow velocity required a higher total dose of gonadotropins for stimulation but produced fewer follicles and mature oocytes compared with patients who had adequate ovarian blood flow velocity. One might speculate that by improving the ovarian stromal blood flow velocity, the delivery of gonadotropins to the follicles will be improved and as a result, the number and quality of mature oocytes produced and the implantation rate will improve. Will the adjuvant use of angiogenic factors, such as LH, fulfill this role? In conclusion, ovarian stromal blood flow velocity after pituitary suppression is predictive of ovarian responsiveness and the outcome of IVF treatment. We have gained further insight into the possible underlying cause of diminished ovarian response, which may be amenable to treatment. It is therefore essential that patients with diminished ovarian blood flow velocity be identified early if appropriate intervention is to be applied. Further studies are required to characterize better the changes in ovarian stromal blood flow velocity that occur in the general population and in the general infertile population with increasing age. Prospective studies also are necessary to assess whether any role is played by angiogenic factors in 28

Engmann et al.

Predictive value of ovarian stromal PSV

improving ovarian responsiveness and IVF outcome in patients with diminished ovarian blood flow velocity.

Acknowledgments: The authors thank Noreen Maconochie, Ph.D., for her statistical advice and the nurses at the clinic for their help with the recruitment of patients. They also are grateful to the medical, nursing, and scientific staffs of the London Women’s Clinic who were involved in the care of the patients studied.

References 1. Tan SL, Maconochie N, Doyle P, Campbell S, Balen A, Bekir J, et al. Cumulative conception and live-birth rates after in vitro fertilization with and without the use of long, short and ultrashort regimens of the gonadotropin-releasing hormone agonist buserelin. Am J Obstet Gynecol 1994;171:513–20. 2. Navot D, Bergh PA, Williams M, Garrisi GJ, Guzman I, Sandler B, et al. An insight into early reproductive processes through the in vivo model of ovum donation. J Clin Endocrinol Metab 1991;72:408 –14. 3. Tan SL, Royton P, Campbell S, Jacobs HS, Betts J, Mason B, et al. Cumulative conception and live birth rates after in vitro fertilization. Lancet 1992;339:1390 – 4. 4. 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 fertilisation outcome. Fertil Steril 1989;51:651– 4. 5. Farhi J, Homburg R, Ferber A, Orvieto R, BenRafael 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: 241–3. 6. Smotrich DB, Widra EA, Gindoff PR, Levy MJ, Hall JL, Stillman RJ. Prognostic value of day 3 estradiol on in vitro fertilization outcome. Fertil Steril 1995;64:1136 – 40. 7. Murkjee T, Copperman A, Lapinski R, Sandler B, Bustillo M, Grunfeld L. An elevated day three follicle-stimulating hormone: luteinizing hormone ratio (FSH:LH) in the presence of normal day 3 FSH predicts a poor response to controlled ovarian stimulation. Fertil Steril 1996;65: 588 –93. 8. Scott RT, Leonardi MR, Hoffman GE, Illions EH, Neal GS, Navot DA. A prospective evaluation of clomiphene citrate challenge test screening in the general infertility population. Obstet Gynecol 1993;82:539 – 45. 9. Fanchin R, DeZiegler D, Olivennes F, Taieb J, Dzik A, Frydman R. Exogenous follicle stimulating hormone ovarian reserve test (EFORT): a simple and reliable screening test for detecting ‘poor responders’ in in-vitro fertilisation. Hum Reprod 1994;9:1607–11. 10. Winslow K, Tonor J, Brzyski R, Oehinger S, Acosta A, Muasher S. The gonadotropin-releasing hormone agonist stimulation test—a sensitive predictor of performance in the flare-up in vitro fertilisation. Fertil Steril 1991;56:711–7. 11. Sladkevicius P, Valentin L, Marsal K. Blood flow velocity in the uterine and ovarian arteries during the normal menstrual cycle. Ultrasound Obstet Gynecol 1993;3:199 –208. 12. Tan SL, Zaidi J, Campbell S, Doyle P, Collins W. Blood flow changes in the ovarian and uterine arteries during the normal menstrual cycle. Am J Obstet Gynecol 1996;175:625–31. 13. Steer CV, Campbell S, Tan SL, Crayford T, Mills C, Mason BA, et al. The use of transvaginal color flow imaging after in vitro fertilization to identify optimum uterine conditions before embryo transfer. Fertil Steril 1992;57:372– 6. 14. Zaidi J, Campbell S, Pittroff R, Tan SL. Endometrial thickness, morphology, vascular penetration and velocimetry in predicting implantation in an in vitro fertilization program. Ultrasound Obstet Gynecol 1995;6:191– 8. 15. Zaidi J, Pittrof R, Shaker A, Kyei-Mensah A, Campbell S, Tan SL. Assessment of uterine artery blood flow on the day of human chorionic gonadotropin administration by transvaginal color Doppler ultrasound in an in vitro fertilization program. Fertil Steril 1996;65:377– 81. 16. 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 fertilisation program. Obstet Gynecol 1996;88:779 – 84. 17. Tan SL, Bennet S, Parsons J. Surgical techniques for oocyte recovery and embryo transfer. Br Med Bull 1990;46:628 – 42. 18. Veeck LL, Wortham JWE, Witmeyer J, Sandow BA, Acosta AA,

Vol. 71, No. 1, January 1999

19.

20. 21.

22. 23.

24.

25. 26. 27. 28.

Garcia JE, et al. Maturation and fertilization of morphologically immature human oocytes in a program of in vitro fertilization. Fertil Steril 1983;39:594 – 602. American Institute of Ultrasound in Medicine/National Electrical Manufacturers Association. Standard for real-time display of thermal and mechanical acoustic output indices on diagnostic ultrasound equipment. Rockville (MD): American Institute for Ultrasound in Medicine, 1992. Zaidi J, Jurkovic D, Campbell S, Okokon E, Tan SL. Circadian variation in uterine artery blood flow during the follicular phase of the menstrual cycle. Ultrasound Obstet Gynecol 1995;5:406 –10. Steer CV, Mills CL, Tan SL, Campbell S, Edwards RG. The cumulative embryo score: a predictive embryo scoring technique to select the optimal number of embryos to transfer in an in-vitro fertilization and embryo transfer program. Hum Reprod 1992;7:117–9. MacDougall MJ, Tan SL, Balen A, Jacobs HS. A controlled study comparing patients with or without polycystic ovaries undergoing in vitro fertilisation. Hum Reprod 1993;8:233–7. Alvarez C, Cremades N, Blasco N, Bernabeu R. Influence of gonadotrophin-releasing hormone agonist total dose in the ovarian stimulation in the long down-regulation protocol for in-vitro fertilization. Hum Reprod 1997;12:2366 –9. Salat-Baroux J, Alvarez S, Antoine JM, Tibi C, Cornet D, Mandelbaum J, et al. Comparison between long and short protocols of LHRH agonist in the treatment of polycystic ovary disease by in-vitro fertilization. Hum Reprod 1988;3:535–9. Marcus S, Edwards RG. High rates of pregnancy after long-term down-regulation of women with severe endometriosis. Am J Obstet Gynecol 1994;171:812–7. Findlay JK. Angiogenesis in reproductive tissues. J Endocrinol 1986; 111:357– 66. Mosher WD, Pratt WF. Fecundity and infertility in the United States: incidence and trends. Fertil Steril 1991;56:192–3. Lenton EA, Sexton L, Lee S, Cooke ID. Progressive changes in LH and

FERTILITY & STERILITYt

29. 30.

31. 32.

33. 34.

35. 36. 37.

FSH and LH:FSH ratio in women throughout reproductive life. Maturitas 1988;10:35. Scott RT, Hoffman GE, Oeninger S, Muasher SJ. Intercycle variability of day 3 follicle-stimulating hormone levels and its effect on stimulation quality in in vitro fertilization. Fertil Steril 1990;54:297–302. Zeleznik AJ, Sciler HM, Reichert LE. Gonadotropin-binding sites in the rhesus monkey ovary: role of the vasculature in the selective distribution of human chorionic gonadotropin to the preovulatory follicle. Endocrinology 1981;109:356 – 61. Richardson SJ, Senikas V, Nelson JF. Follicular depletion during the menopausal transition: evidence for accelerated loss and ultimate exhaustion. J Clin Endocrinol Metab 1987;65:1231–7. Meyer WR, Lavy G, DeCherney AH, Visintin I, Economy K, Luborsky J. Evidence of gonadal and gonadotropin antibodies in women with a suboptimal ovarian response to exogenous gonadotropin. Fertil Steril 1990;75:795–9. Moloy D, Martin M, Speirs A, Lopata A, Clarke G, McBain J, et al. Performance of patient with ‘frozen pelvis’ in an in vitro fertilization program. Fertil Steril 1987;47:450 –5. Simon C, Gutierrez A, Vidal A, de los Santos MJ, Tann JJ, Remohi J, et al. Outcome of patients with endometriosis in assisted reproduction: results from in vitro fertilization and oocyte donation. Hum Reprod 1994;9:725–9. Ben-Rafael Z, Feldberg D. The poor responder patient in an in vitro fertilization-embryo transfer program. J Assist Reprod Genet 1993;10: 118 –20. Stadmauer L, Diktoff EC, Session D, Kelly A. High dosages of gonadotropins are associated with poor pregnancy outcomes after in vitro fertilization-embryo transfer. Fertil Steril 1994;61:1058 – 64. Land JA, Yarmolinskaya MI, Dumoulin JCM, Evers JL. High-dose human menopausal gonadotropin stimulation in poor responders does not improve in vitro fertilization outcome. Fertil Steril 1996;65:961–5.

29