Predictors of IVF outcome by three-dimensional ultrasound

S. Kupesic,1 and A. Kurjak

Department of Obstetrics and Gynaecology, Medical School University of Zagreb, Sveti Duh Hospital, Sveti Duh 64, HR-10000 Zagreb, Croatia


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
BACKGROUND: Our study was designed to evaluate whether ovarian antral follicle number, ovarian volume, stromal area and ovarian stromal blood flow are predictive of ovarian response and IVF outcome. A total of 56 women with normal basal serum FSH concentrations who had no history of ovarian surgery, no ovarian and/or uterine pathology, were non-smokers and undergoing their first IVF cycle using a standard long GnRH agonist protocol were examined. METHODS: Total ovarian antral follicle number, total ovarian volume, total stromal area and mean flow index (FI) of the ovarian stromal blood flow were determined by three-dimensional (3D) and power Doppler ultrasound after pituitary suppression. Pretreatment 3D ultrasound ovarian measurements were compared with subsequent ovulation induction parameters [peak estradiol (E2) on HCG administration day and number of oocytes] and cycle outcome (fertilization and pregnancy rates). RESULTS: The total antral follicle number achieved the best predictive value for favourable IVF outcome, followed by ovarian stromal FI, peak E2 on HCG administration day, total ovarian volume, total ovarian stromal area and age. Using these six parameters, we were able to predict a favourable IVF outcome in 50% (11/22) of patients and poor outcome in 85% (29/34) of patients. CONCLUSIONS: Three-dimensional ultrasound facilitates determination of the antral follicle number, ovarian volume calculation, evaluation of the ovarian stroma and analysis of the intensity of ovarian stromal blood flow in a short time without increasing the patient's discomfort.

Key words: antral follicle number/ovarian stromal blood flow/ovarian volume/stromal area/three-dimensional ultrasound


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Recruitment and development of multiple follicles in response to gonadotrophin stimulation are key factors leading to successful treatment by assisted reproductive methods. Exaggerated response on the one hand leads to increased risk of ovarian hyperstimulation syndrome and multiple pregnancy, while on the other hand, poor ovarian response is associated with cancellation of the cycle and poor pregnancy rates. Any modality which may predict ovarian responsiveness prior to ovulation induction will be helpful in counselling the patients and tailoring the dosage of gonadotrophin. Traditionally, chronological female age and early follicular phase serum FSH levels are the most useful parameters for prediction of the ovarian reserve (Padilla and Garcia, 1989Go; Toner et al., 1991Go; Tan et al., 1992Go; Check et al., 1994Go). Clinical studies suggest that a small ovarian volume, reduced stromal area and decreased number of antral follicles indicate a poor response to subsequent controlled ovarian stimulation (Syrop et al., 1995Go; Chang et al., 1998Go; Pellicer et al., 1998Go; Wu et al., 1998Go; Yu et al., 2000). The total number of antral follicles achieved the best predictive value as to the ovarian response in an IVF programme, followed by basal FSH, body mass index (BMI) and age of the woman (Ng et al., 2000Go). A study that evaluated the effect of ageing on ovarian volume measurements in infertile women found a significant positive correlation between age and FSH, but not between age and ovarian volume (Sharara and McClamrock, 1999Go). In the same study, a significant relationship was noted between FSH and the number of follicles, but not between FSH and volume. The same authors noted that women with small ovarian volumes at baseline ultrasound (<3 cm3) can have comparable implantation and pregnancy rates with those with larger ovarian volumes with the use of a higher dose gonadotrophin, microdose GnRH agonist stimulation (Sharara and McClamrock, 2001Go). Lass et al. proposed that assessment of ovarian size should be an integral part of infertility evaluation (Lass et al., 1997Go). Their results indicated 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.

To substantiate this observation, we undertook a three-dimensional (3D) ultrasound prospective study evaluating antral follicle number, ovarian volume measurements, stromal area and ovarian stromal blood flow after pituitary suppression. Pretreatment 3D ultrasound ovarian measurements were compared with subsequent ovulation induction parameters [peak estradiol (E2) on day of HCG administration and number of oocytes] and cycle outcome (fertilization and pregnancy rates).


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
This prospective study included 56 patients enrolled in an IVF/embryo transfer treatment cycle between May 1, 2000 and March 31, 2001. The mean age of patients was 33.5 years (range 22–43) and the mean duration of infertility was 6 years (range 2–14). All the patients met the following criteria: (i) an early follicular phase FSH level of <10 mIU/ml; (ii) both ovaries were present and contained no ovarian cysts >=10 mm in diameter; (iii) no history of partial or complete surgical resection of the ovary; (iv) no history or actual signs of ovarian endometriosis or uterine leiomyoma; (v) no hormonal or ultrasonographic features of polycystic ovaries; (vi) none had received steroid hormone treatment within 6 months before the ultrasonographic evaluation; (vii) all were undergoing their first IVF cycle; (viii) all were non-smokers; (ix) 3D ultrasound measurements were performed exclusively at our department by two experienced sonographers (S.K. and A.K.)

The intra- and interobserver coefficients of variation were low (7 and 8% respectively). The study protocol was approved by the local ethics committee and written informed consent was obtained from all subjects.

All patients underwent a standard regime of the GnRH agonist. Triptorelin (Decapeptyl; Ferring, Kiel, Germany) was administered s.c. at a daily dose of 100 µg starting in the mid-luteal phase. After confirmation of pituitary down-regulation (no ovarian cysts >1 cm, endometrial thickness <5 mm and E2 <50 pg/ml) a transvaginal ultrasound scan was performed using an electronic 7.5 MHz transducer with 3D facility (Combison 530D; Kretztechnique, Zipf, Austria).

Using this method, three rotatable perpendicular planes were displayed for each ovary to obtain the largest dimensions. The data set was then stored digitally on an internal disk for subsequent analysis. This ability to store 3D data on a hard disk allowed us to shorten the examination time (2–5 min).

Detailed analysis of antral follicle number, ovarian volume and stromal area was performed after the patient had gone and lasted 10–20 min. The stromal area was measured by meticulously outlining the region of interest. At the same time, the number of antral follicles was carefully obtained. Both measurements were summed up to obtain an individual's total stromal area and total antral follicle number. The volume of the ovary was measured by outlining the areas of multiple parallel sections at different distances from the ovary and was calculated using the trapezoid formula. At least 10 serial slices, 1–2 mm apart, were taken across the ovary for volume measurement. The actual volume was calculated by the built-in computer program. The volumetric data were then stored digitally on a hard disc (SyQuest; Technology Inc., Fremont, CA, USA) for subsequent analysis. Each ovarian volume was recorded and then summed to obtain an individual's total ovarian volume.

After B-mode analysis, power Doppler imaging was switched on together with the volume mode. To reduce the acquisition time, the volume of the colour box and sweep angle were reduced. The colour frame rate was adjusted as follows: both colour density and colour quality were as low as necessary to obtain a good image, whereas the pulsed repetition frequency was as high as possible to enable the display of targeted flow velocity. The spatial peak temporal average intensity was ~80 mW/cm2. Wall filters (50 Hz) were used to eliminate low frequency signals. The average patient examination time for 3D power Doppler sonography was 3 min. With the use of the lowest line density, the average acquisition time was 22 s (range 12–32). At the end of each examination, combined colour and grey scale rendering was performed and quantitative analysis of the blood flow in the outlined area was achieved by implementing the colour histogram mode. The flow index (FI) reflects the intensity of blood flow and was calculated automatically by the built-in computer: FI = weighted colour values/colour values. Since no significant differences were noted between the two sides for ovarian stromal blood flow in any patient, the data were calculated together and the mean measurement of both sides in an individual patient was used for statistical analysis. The subsequent analysis of the power Doppler reformatted sections lasted 5–10 min.

The standard starting dose of FSH was 2–4 ampoules (150–300 IU of FSH activity) depending on the patient age and basal serum FSH concentration. Follicular growth was monitored with serial ultrasound scans, and the dose of FSH was adjusted according to the follicular response. When the average diameters of the three leading follicles were at least 18 mm, as measured by ultrasound, 10 000 IU of HCG was administered as a single injection.

Transvaginal ultrasound-directed oocyte retrieval was performed ~36 h after HCG administration and embryo transfer took place 5 days after oocyte retrieval in the blastocyst stage. Progesterone pessaries (400 mg twice daily, Utrogestan; Laboratories Besins International, Paris, France) were given as luteal support, starting on the day of embryo transfer and continuing until 16 days thereafter.

Blood sampling for serum FSH and LH was performed on day 3 of the menstrual cycle, while E2 measurements were taken on the day of pituitary suppression and on HCG administration day. Serum FSH and LH levels were measured by microparticle enzyme immunoassay (Abbott AXSYM reagent pack; Abbott Diagnostics, Abbott Park, IL, USA) and serum E2 levels were assayed using a commercially available chemiluminescent immunoassay (Abbott). The intra- and inter-assay coefficients of variation were 4.2 and 7.2% for FSH, 4.5 and 7.5% for LH and 5.0 and 7.5% for E2 respectively.

All of the following outcome variables (total antral follicle number, total ovarian volume, total ovarian stromal area and mean FI of the ovarian stromal blood flow) were correlated with peak serum E2 concentration (measured in all patients on the day of HCG administration), number of oocytes retrieved, number of embryos resulting from the cycle and number of clinical pregnancies. Pregnancy was defined as the occurrence of a positive ß-HCG (>10 IU) value on day 12 after embryo transfer and a second higher value 2 days later. Only pregnancies reaching ß-HCG values >100 IU were considered for evaluation.

Statistical analysis
All data are expressed as median (range). Comparison between two outcomes (pregnant and non-pregnant) was carried out by the non-parametric Mann–Whitney test. A P-value (two-tailed) < 0.05 was considered statistically significant.

Correlation was assessed by the Spearman's rank method and correlation coefficients were determined for all measures and outcome variables. At the end of the study, logistic regression analysis was performed using the following predictors: age, E2 on day of HCG administration, total antral follicle number, total ovarian volume, total ovarian stromal area and mean ovarian stromal FI.


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
A total of 56 infertile women with normal basal serum FSH concentrations was recruited into this study: 28 tubal factor, 20 male factor and eight mixed causes. Comparisons of baseline characteristics and ovarian response data between pregnant and non-pregnant groups are summarized in Table IGo. All the initiated cycles reached oocyte retrieval and there were 22 clinical pregnancies (39.3%). The clinical pregnancy rate was similar in all three aetiological groups.


View this table:
[in this window]
[in a new window]
 
Table I. Comparison of baseline characteristics and ovarian responses data between the pregnant and non-pregnant group
 
Differences in age, duration of infertility, antral follicle number, ovarian volume, ovarian stromal area, ovarian stromal FI, E2 on day of HCG administration, number of aspirated oocytes and fertilization rates were statistically significant between pregnant and non-pregnant groups (Table IGo). In the pregnant group, a median of 2.36 (range 1–4) embryos were transferred versus a median of 2.2 (range 2–3) embryos in the non-pregnant group (not significant).

Spearman's correlation coefficients were determined for each measure and outcome variable. The total number of aspirated oocytes correlated well (P < 0.01) with total antral follicle number (r = 0.98), total ovarian volume (r = 0.89), total ovarian stromal area (r = 0.89) and mean ovarian stromal FI (r = 0.08) (Table IIGo). Antral follicle count and stromal FI were found to have a high degree of interaction with one another. When we took into consideration the ovaries with a comparable number of follicles and stromal vascularity, we found the following results: patients with less than five antral follicles had a mean FI of 10.6, patients with an antral follicle count of 5–10 had a mean FI of 12.1, and patients with more than 10 antral follicles had a mean FI of 12.9. Therefore, patients with higher antral follicle count presented higher stromal vascularity due to higher need for perfusion.


View this table:
[in this window]
[in a new window]
 
Table II. Correlation coefficients of total antral follicle number, total ovarian volume, total ovarian stromal area and mean ovarian stromal blood flow index (FI) with other parameters of ovarian response
 
Table IIIGo presents data on the logistic regression analysis of six variables (age, E2 on day of HCG administration, total antral follicle number, total ovarian volume, total ovarian stromal area and mean ovarian stromal FI) in predicting IVF outcome. The total antral follicle number was the best predictor of IVF outcome, followed by ovarian stromal FI, peak E2 on day of HCG administration, total ovarian volume, total ovarian stromal area and age. The group of patients with stromal FI <11 had 4.2 (mean) oocytes retrieved, 64.3% fertilization rate and 0% pregnancy rate; patients with an FI of 11–13 had 9.4 (mean) oocytes retrieved, 75.2% fertilization rate and 47.2% pregnancy rate; while patients with FI >13 had 14.1 (mean) oocytes retrieved, 89% fertilization rate and 50% pregnancy rate. These results provided evidence that the intensity of the stromal blood flow was predictive of increased delivery of gonadotrophins to target cells and resulted in a higher number of oocytes retrieved, and more favourable cycle outcome (higher fertilization and pregnancy rates).


View this table:
[in this window]
[in a new window]
 
Table III. Logistic regression analysis of six parameters: age, E2 on day of HCG administration, total ovarian volume, total antral follicle number, mean ovarian stromal flow index (FI) and total ovarian stromal area in predicting IVF outcome (significance)
 
Using six parameters from Table IIIGo we were able to predict a favourable IVF outcome in 11/22 patients and a poor IVF outcome in 29/34 patients. The logistic model achieved a statistically significant value (P = 0.0106) with –2 Log Likelihood of 58.385. Prediction was carried out and the overall prediction rate was 71.43%.


    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Our study compared predictive values of total antral follicle number, total ovarian volume, total ovarian stromal area and mean ovarian stromal FI in relation to the number of oocytes and fertilization and pregnancy rates obtained in 56 consecutive women undergoing IVF treatment. All of the patients had normal basal serum FSH concentrations and received the same standard protocol for pituitary down-regulation and gonadotrophin induction for the first time. The results of our study suggest that there is an association between total antral follicle number, mean ovarian stromal FI, total ovarian stromal area and total ovarian volume after pituitary suppression, and the number of mature oocytes retrieved and pregnancy rates. We found that in patients with normal basal serum FSH concentrations, total antral follicle number was the most important predictor of ovarian response and IVF outcome followed by ovarian stromal FI.

Tomas et al. and Chang et al. were the first to report on prediction of the ovarian response by counting the number of antral follicles (Tomas et al., 1997Go; Chang et al., 1998Go). In 166 women undergoing their first IVF cycle after pituitary down-regulation, it was concluded that the number of antral follicles present before ovarian stimulation was a better predictor of the ovarian response than the ovarian volume or age alone (Tomas et al., 1997Go). The authors revealed that the number of antral follicles was correlated with the number of recovered oocytes, while the ovarian volume was correlated with the number of antral follicles before the stimulation, but not with the number of oocytes. Chang et al. studied 130 women in 149 IVF cycles after they failed six cycles of infertility treatment including ovarian stimulation with or without intrauterine insemination (Chang et al., 1998Go). They found a highly significant correlation between the antral follicle number and the number of oocytes. A higher chance of cycle cancellation, a lower E2 concentration and a need for higher gonadotrophin dosage was detected in cycles with less than three antral follicles. Scheffer et al. evaluated antral follicle counts by transvaginal ultrasonography in relation to age in women with proven natural fertility (Scheffer et al., 1999Go). They found that the antral follicle count showed the clearest correlation with age, and a mean yearly decline of 4.8% before the age of 37 years, compared with 11.7% thereafter. The number of small antral follicles in both ovaries, as measured by ultrasound, is clearly related to reproductive age and could well reflect the size of the remaining primordial follicle pool.

The second most important predictor of IVF outcome in our study was ovarian stromal FI. Zaidi et al. were the first to show that there was a relationship between ovarian stromal blood flow velocity and ovarian follicular response (Zaidi et al., 1995Go). They measured the ovarian stromal peak systolic velocity (PSV) in the early follicular phase and showed that poor responders had low ovarian blood flow PSV. Engmann et al. found that in patients with normal basal FSH concentrations, the mean ovarian stromal PSV on the day of pituitary suppression was a better predictor of ovarian responsiveness than age (Engmann et al., 1999Go).

The main shortcoming of the previous studies on ovarian stromal blood flow is that they used PSV as a main predictor. The accurate measurement of blood flow velocity requires knowledge of the angle of insonation to the blood vessels analysed. Ovarian vessels inside the ovarian stroma are thin and tortuous and therefore it is impossible to obtain accurately the angle between the ultrasound beam and intra-ovarian vessels. This potential limitation leads to subjective measurements and depends on experience of the sonographer, since he/she should search for the highest velocity of the ovarian stromal vessels. Our data indicate that ovarian stromal blood flow obtained by 3D power Doppler imaging after pituitary suppression truly reflects total ovarian stromal blood flow. Using the colour histogram mode, we assessed the intensity of ovarian stromal blood flow, which seems to be predictive of increased delivery of gonadotrophins to target cells for stimulation of follicular growth. In addition to the fact that more oocytes are expected to be retrieved in aspiration procedures, better quality of the oocytes and embryos may lead to higher pregnancy rates.

Patients with a depleted ovarian reserve assessed by 3D ultrasound may require an increased starting dosage of gonadotrophins in order to improve follicular response (Ben-Rafael and Feldberg, 1993Go), although some studies failed to show any beneficial effect of increasing the dosage of gonadotrophins in poor responders (Stadmauer et al., 1994Go; Laud et al., 1996Go). We believe that some angiogenic factors could potentially be used for improving ovarian responsiveness and IVF outcome in patients with diminished ovarian blood flow. It is possible that deficient intra-ovarian vascularity may serve as the initial marker of reduced ovarian reserve which precedes an increased FSH level and reduction of the ovarian volume (Engmann et al., 1999Go). If this is so, exogeneous angiogenic factors may be used to improve ovarian blood flow and prevent the onset of ovarian failure.

Syrop et al. demonstrated that ovarian volume, as determined by transvaginal ultrasonography, is another predictor of ovarian response to ovulation induction and clinical pregnancy rates (Syrop et al., 1995Go). Decreased total ovarian volume and the volume of the smallest ovary were significant variables predicting peak E2 and number of the oocytes and embryos. Lass et al. recommend that ovarian volume should be measured by transvaginal scan before ovulation induction in all patients, regardless of age, and the stimulation protocol planned accordingly (Lass et al., 1997Go). They suggest that women who have a mean ovarian volume of <3 cm3 have a very high chance of failure to respond to exogenously applied stimuli. Wu et al. found that 3D ultrasound facilitates determination of the ovarian volume in patients with and without polycystic ovarian disease (Wu et al., 1998Go). Furthermore, volume of the ovary assessed by 3D ultrasound correlates better with direct measurement of the surgical specimen than that obtained by 2D ultrasound (Bonilla-Musoles et al., 1995Go; Kyei-Mensah et al., 1996Go).

In our selected group of patients, total ovarian volume and total stromal area assessed by 3D ultrasound were less predictive of IVF outcome than total antral follicle number and ovarian stromal vascularity. It seems that some patients may have diminished ovarian reserves without evident changes in ovarian volume. Although ovarian volume and stromal area measurements are useful to distinguish between multifollicular and polycystic ovaries (Wu et al., 1998Go), they are less significant predictors of IVF outcome in patients with normal basal serum FSH and LH concentrations.

Some investigators (Gougeon, 1996Go; Pellicer et al., 1998Go) suggest that the loss of resting follicles up to 30 years of age is due to atresia; thereafter, this loss is due principally to the entrance of resting follicles into the growth phase, with the decay rate of resting follicles accelerating from 38 years onward. If this hypothesis is correct, evaluation of the follicles in the growing phase may be an acceptable way to measure the decay rate of resting follicles. In this context, 3D ultrasound extends the boundaries of conventional ultrasonographic methods enabling more objective assessments of ovarian morphology and volume.

Using this method the examination time is short, and does not increase the patient's discomfort with serial ultrasound ovarian and endometrial monitoring. For the first time it becomes possible to assess pelvic structures of an infertile patient in three separate volume acquisitions (uterus, left and right adnexa). The total examination time is short and the patient can leave the examination room once the volumes have been stored and all further investigations can be performed without the presence of the patient. Retrospective evaluation of the ovarian measurements enables precise correlation with response to stimulating drugs and IVF outcome. In addition to ovarian measurements, endometrial volume measurements and assessment of endometrial perfusion by 3D power Doppler histogram on the day of embryo transfer could be used for determination of uterine receptivity (Kupesic et al., 2001Go). Therefore, it is believed that 3D ultrasound ovarian and endometrial measurements may become simple additional tests which could predict response to stimulating drugs, degree of endometrial receptivity and outcome of assisted reproductive techniques.


    Acknowledgements
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
The authors thank Drazena Bjelos, M.D. and Zdenko Sonicki, M.D. for their statistical advice, and Sanja Vujisic, Ph.D. for performing the hormone assays.


    Notes
 
1 To whom correspondence should be addressed. E-mail: sanja.kupesic{at}zg.tel.hr Back


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Ben-Rafael, Z. and Feldberg, D. (1993) The poor responder patient in an in-vitro fertilization–embryo transfer program. J. Assist. Reprod. Genet., 10, 118–120.[ISI][Medline]

Bonilla-Musoles, F., Raga, F. and Osborne, N.G. (1995) Three dimensional ultrasound evaluation of ovarian masses. Gynecol. Oncol., 59, 129–135.[ISI][Medline]

Chang, M.Y., Chiang, C.H., Hsieh, T.T., Soong, K.Y. and Hsu, K.H. (1998) Use of antral follicle count to predict the outcome of assisted reproduction technologies. Fertil. Steril., 69, 505–510.[ISI][Medline]

Check, J.H., Lurie, D., Callan, C., Baker, A. and Benfer, K. (1994) Comparison of the cumulative probability of pregnancy after in vitro fertilization–embryo transfer by infertility factor and age. Fertil. Steril., 61, 257–261.[ISI][Medline]

Engmann, L., Sladkevicius, P., Agrawal, R., Bekir, J.S., Campbell, S. and Tan, S.L. (1999) Value of stromal blood flow velocity measurement after pituitary suppression in the prediction of ovarian responsiveness and outcome of in vitro fertilization treatment. Fertil. Steril., 71, 22–29.[ISI][Medline]

Gougeon, A. (1996) Regulation of ovarian follicular development in primates: facts and hypothesis. Endocr. Rev., 17, 125–155.

Kupesic, S., Bekavac, I., Bjelos, D. and Kurjak, A. (2001) Assessment of endometrial receptivity by transvaginal colour Doppler and three-dimensional power Doppler ultrasonography in patients undergoing in vitro fertilization procedures. J. Ultrasound Med., 20, 125–134.[Abstract/Free Full Text]

Kyei-Mensah, A., Zaidi, J., Pittrof, R., Shaker, A., Campbell, S. and Tan, S.L. (1996) Transvaginal three dimensional ultrasound: accuracy of follicular volume measurements. Fertil. Steril., 65, 371–376.[ISI][Medline]

Lass, A., Skull, J., McVeigh, E., Margara, R. and Winston, R.M.L. (1997) Measurement of ovarian volume by transvaginal sonography before ovulation induction with human menopausal gonadotrphin for in-vitro fertilization can predict poor response. Hum. Reprod., 12, 294–297.[Abstract]

Laud, J.A., Yarmolinskaya, M.I., Daumolin, J.C.M. and Evers, J.L. (1996) High-risk human menopausal gonadotrophin stimulation in poor responders does not improve in-vitro fertilization outcome. Fertil. Steril., 65, 961–965.[ISI][Medline]

Ng, E.H., Tang, O.S. and Ho, P.C. (2000) The significance of the number of antral follicles prior to stimulation in predicting ovarian responses in an IVF programme. Hum. Reprod., 15, 1937–1942.[Abstract/Free Full Text]

Padilla, S.L. and Garcia, J.E. (1989) Effect of maternal age and number of in vitro fertilization procedures on pregnancy outcome. Fertil. Steril., 52, 270–273.[ISI][Medline]

Pellicer, A., Ardilles, G., Neuspiller, F., Remoni, J., Simon, C. and Bonilla-Musoles, F. (1998) Evaluation of the ovarian reserve in young low responders with normal basal levels of follicle stimulating hormone using three-dimensional ultrasonography. Fertil. Steril., 70, 671–675.[ISI][Medline]

Scheffer, G.J., Broekmans, J.M., Dorland, M., Habbema, J.D.F., Looman, C.W.N. and te Velde, E.R. (1999) Antral follicle counts by transvaginal ultrasonography are related to age in women with proven natural fertility. Fertil. Steril., 72, 845–851.[ISI][Medline]

Sharara, F.I. and McClarmrock, H.D. (1999) The effect of aging on ovarian volume measurements in infertile women. Obstet. Gynecol., 94, 57–60.[Abstract/Free Full Text]

Sharara, F.I. and McClarmrock, H.D. (2001) Use of microdose GnRH agonist protocol in women with low ovarian volumes undergoing IVF. Hum. Reprod., 16, 500–503.[Abstract/Free Full Text]

Stadmauer, L., Diktoff, E.C., Session, D. and Kelly, A. (1994) High dosages of gonadotrophins are associated with poor pregnancy outcomes after in-vitro fertilization–embryo transfer. Fertil. Steril., 61, 1058–1064.[ISI][Medline]

Syrop, C.H., Willhoite, A. and Voorhis, B.J. (1995) Ovarian volume: a novel outcome predictor for assisted reproduction. Fertil. Steril., 64, 1167–1171.[ISI][Medline]

Tan, S.L., Royston, P., Campbell, S., Jacobs, H.S., Betts, J., Mason, B., et al. (1992) Cumulative conception and livebirth rates after in vitro fertilisation. Lancet, 339, 1390–1394.[ISI][Medline]

Tomas, C., Nuojua-Huttunen, S. and Matikainen, H. (1997) Pre-treatment transvaginal ultrasound examination predicts ovarian responsiveness to gonadotrophins in in-vitro fertilisation. Hum. Reprod., 12, 220–223.[Abstract]

Toner, J.P., Philput, C.B., Jones, G.S. and Muasher, S.J. (1991) Basal follicle-stimulating hormone level is a better predictor of in vitro fertilization performance than age. Feril. Steril., 55, 784–791.

Wu, M.H., Tang, H.H., Hsu, C.C., Wang, S.T. and Huang, E. (1998) The role of three-dimensional ultrasonographic images in ovarian measurements. Fertil. Steril., 69, 1152–1155.[ISI][Medline]

Zaidi, J., Campbell, S., Pittrof, R., Kyei-Mensah, A., Shaker, A., Jacobs, H.S. and Tan, S.L. (1995) Ovarian stromal blood flow in women with polycystic ovaries—a possible new marker for diagnosis? Hum. Reprod., 6, 191–198.

Submitted on August 3, 2001; accepted on November 6, 2001.