Prediction of over-response to ovarian stimulation in an intrauterine insemination programme

H. Lashen1,3, M. Afnan1, L. McDougall1 and P. Clark2

1 Assisted Conception Unit, Birmingham Women's Hospital, Edgbaston, Birmingham, B15 2TG and 2 Regional Endocrine Laboratory, University Hospital, Birmingham B29 6JD, UK


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Prediction of poor-response is of equal importance to prediction of over-response in intrauterine insemination programmes. The gonadotrophin-releasing hormone agonist (GnRHa) stimulation test (GAST) was assessed as a predictor of over-response to ovarian stimulation in 81 patients. Blood samples were taken on cycle day 2 (before and 24 h after starting the GnRHa). Day 2 and 3 samples were assayed for oestradiol, follicle stimulating hormone (FSH) and luteinizing hormone (LH). Linear and logistic regression analyses were used to assess age, day 2 FSH, day 2 FSH/LH, oestradiol ratio (oestradiol on day 3/oestradiol on day 2) and FSH ratio (FSH on day 3/FSH on day 2) as predictors of the number of follicles (total and >=14 mm), oestradiol on HCG day, and clinical pregnancy rate as appropriate. Several parameters were also compared between the patients who produced <=3 (>=14 mm) follicles (group A) and those who produced >3 (>=14 mm) follicles (group B). The mean ± SEM age of the patients in the study was 32 ± 0.4 years. The mean total dose of recombinant FSH was 800 ± 20 IU and the mean duration of stimulation was 7.6 ± 0.2 days. Nine (11%) and 12 (15%) patients were cancelled for poor and over-response respectively. The oestradiol ratio was significantly positively correlated with oestradiol on HCG day (P < 0.001), and with the number of mature follicles (>=14 mm) (P = 0.01). Age, day 2 FSH and FSH ratio were not significantly correlated with oestradiol on HCG day, total follicles and follicles >=14 mm. None of the above-mentioned variables was correlated with clinical pregnancy rate. Group A had significantly lower oestradiol ratio (P = 0.007), longer duration of stimulation (P = 0.002), higher total FSH dose (P = 0.001), and lower oestradiol on HCG day (P = 0.001). GAST is therefore useful in predicting the high responders to gonadotrophin stimulation.

Key words: GnRH stimulation test/intrauterine insemination/ovarian response


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Ovulation induction and intrauterine insemination (IUI) has been used to treat couples with unexplained and mild male factor infertility (Dodson et al., 1987Go; Cruz et al., 1986Go; Serhal et al., 1988Go). The extent of ovarian stimulation in this modality of treatment is crucial, mainly to avoid the recruitment of an excessive number of follicles with the attendant risk of multiple pregnancy. The multiple pregnancy rate has increased dramatically with the introduction of assisted reproduction techniques. Cohen (1998) reviewed the data on the incidence of multiple pregnancy associated with in-vitro fertilization (IVF) treatment and recommended several approaches to reduce such risk with its associated perinatal mortality and morbidity. However, reducing the incidence of multiple pregnancy in IVF is relatively easy compared with that in ovarian stimulation/IUI due to the difficulties encountered in controlling the number of follicles produced given the wide variability in women's response to ovarian stimulation. High responders in this treatment category are usually cancelled unless IVF treatment is available. Therefore, prediction of ovarian response in this group of patients will help avoiding unnecessary cost and anguish to the patients. Age, basal follicle stimulating hormone (FSH) concentration, clomiphene citrate challenge test (CCCT), and gonadotrophin-releasing hormone (GnRH) agonist stimulation test (GAST) have been reported as good predictors of ovarian response and outcome in IVF patients (Fédération CECOS 1982; Navot et al., 1987Go; Padilla et al., 1990Go; Toner et al., 1991Go). As these tests were developed and tested mainly in IVF patients, the main aim was to predict poor response and the pregnancy outcome. However, predicting the patients' over-response to stimulation is equally important in the IUI programme. We prospectively assessed GAST as a predictor of ovarian over-response to stimulation.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Subjects
During the period from October 1, 1996 to September 30, 1997, patients with unexplained infertility undergoing IUI were invited to take part in the study. Those who agreed were asked to give a written consent and were then consecutively enrolled in the study. Unexplained infertility was diagnosed by confirming tubal patency with hysterosalpingography or laparoscopy and dye test, normal semen fluid analysis (semen volume >=2 ml, count >=20x106/ml, motility >=50% and normal forms >=20%), and confirming ovulation by a serum progesterone concentration >30 nmol/l in the mid-luteal phase.

Intrauterine insemination protocol
The ovarian stimulation protocol involved ultrasound scanning on cycle day 2 to rule out any ovarian pathology before starting intra-nasal nafarelin acetate 400 µg twice daily (Synarel; Searle, High Wycombe, Bucks, UK) from cycle day 2 until the administration of human chorionic gonadotrophin (HCG). On cycle days 3–5 the patients were given 150 IU of recombinant FSH (rFSH, Gonal-F; Serono UK Ltd, Garden City, Herts, UK) daily which was then reduced to 75 IU daily from day 6 onwards. The patients' response was monitored with ultrasound scanning starting on cycle day 9, and daily thereafter. When the leading follicle measured >=16 mm, 10 000 IU HCG was administered.

A maximum of five follicles measuring >=14 mm was the threshold above which the cycle was either cancelled or converted to IVF. A total of nine patients who failed to produce at least one follicle >=14 mm after 10 days of gonadotrophin stimulation were cancelled for poor response.

Design
Blood samples were taken between 0800 and 1100 h on cycle day 2 (before starting the GnRHa) and on cycle day 3. A third blood sample was taken on HCG day. Day 2 sample was assayed for oestradiol, FSH and luteinizing hormone (LH). Day 3 sample was assayed for oestradiol and FSH, and the HCG day sample was assayed for oestradiol.

Hormonal assays
The blood samples were centrifuged at 3000 g, the serum was separated and frozen at –20°C and assayed in one batch at the end of the study. Oestradiol was measured by direct radioimmunoassay (Sorin Biomedica Diagnostica, Wokingham, Berks, UK) with interassay coefficient of variation (CV) of 9.1% at 212 pmol/l and 8.5% at 765 pmol/l. FSH was measured by an automated chemiluminometric assay (Chiron Diagnostics, Halstead, Essex, UK) calibrated against the World Health Organization (WHO) second International Reference Preparation (IRP) 78/549. CV were <10% over the range 8.6–51 IU/l. LH was measured by chemiluminometric assay (Chiron Diagnostics) calibrated against WHO second IRP 80/552. CV were <9% over the range 4.6–53 IU/l.

Data analysis
Linear and logistic regression analyses were used where appropriate to establish the relationship between age, day 2 FSH, FSH/LH, oestradiol ratio (oestradiol day 3/oestradiol day 2), and FSH ratio (FSH day 3/FSH day 2) as predictors, and the number of follicles (total and >=14 mm), oestradiol on HCG day, and clinical pregnancy rate as the response. The t-value as well as the regression coefficient were provided since we believe that the t-value is more informative than the regression coefficients alone.

We also compared the age, oestradiol ratio, FSH ratio, day 2 FSH, and FSH/LH between the poor responders, over-responders and adequate responders (within the specified threshold of five follicles >=14 mm). The significance of GAST was assessed at an arbitrary threshold of three mature follicles (>=14 mm), which is the recommended threshold of the British Fertility Society (Balen, 1997Go). The patients were divided into two groups: A (1–3 follicles) and B (>3 follicles) and the two groups were compared with regard to age, oestradiol ratio, FSH ratio, day 2 FSH, day 2 oestradiol, and FSH/LH, duration of stimulation, total gonadotrophin dose, clinical pregnancy rate per insemination and multiple pregnancy rate.

ANOVA (analysis of variance) test was used to carry out the comparative analyses. P < 0.05 was considered significant. Statistical analysis was carried out using Minitab for Windows (Minitab Inc., State College, PA, USA).

Ethical committee approval and patients' written consent were obtained before starting the study.


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
A total of 81 patients with unexplained infertility was recruited for the study. The mean age ± SEM was 32 ± 0.4 years (range 22–42). The mean total rFSH dose was 800 ± 20 IU and the mean duration of stimulation by rFSH was 7.6 ± 0.2 days. Nine (11%) and 12 (15%) patients were cancelled for poor and over-response respectively giving a 26% cycle cancellation rate. The response of the remaining 60 patients was within the accepted protocol. For comparison purposes, the results of the 12 over-responders were included where appropriate. The clinical pregnancy rates (CPR) per started cycle and per insemination were 13/81 (16%) and 13/60 (22%) respectively. The multiple pregnancy rate (MPR) was 5/13 (38%) of whom three were twins and two were quadruplets.

Correlation analysis results
Oestradiol ratio was correlated significantly with oestradiol on HCG day (regression coefficient = 1017.8, t = 4, P < 0.001), and the number of follicles >=14 mm (regression coefficient = 0.5, t = 2.6, P = 0.01). Day 2 FSH/LH significantly inversely correlated with the total number of follicles (regression coefficient = –0.7, t = –2 and P = 0.02).

Age, day 2 FSH, day 2 oestradiol, and FSH ratio were not significantly correlated with oestradiol on HCG day, the total number of follicles, or the number of follicles >=14 mm. Neither age, oestradiol ratio, FSH ratio, day 2 FSH, day 2 oestradiol nor FSH/LH were significantly correlated with the clinical pregnancy rate.

Poor, over- and adequate responders
The oestradiol ratio was significantly different between the over-responders and the adequate responders (P = 0.04) while no significant difference was found between the poor responders and the other two groups (Table IGo). Considering the oestradiol ratio, the 95% CI for the poor responders (n = 9), adequate responders (n = 60) and over-responders (n = 12) respectively were 2.1–3.8, 2.9–3.7 and 3.5–5.4. No significant difference was found between the three groups with regard to age, day 2 oestradiol, day 2 FSH, and FSH/LH. The mean ± SD of the compared parameters is summarized in Table IGo.


View this table:
[in this window]
[in a new window]
 
Table I. Adequate, poor, and over-responders
 
Comparison of patients according to number of mature follicles (>=14 mm) produced
Forty-one patients produced 1–3 mature follicles (group A), and 31 patients produced >3 follicles (group B; including 12 over-responders whose cycles were cancelled). The two groups were similar with regard to age, FSH ratio, day 2 FSH, day 2 oestradiol, and FSH/LH ratio (Table IIGo). However, the oestradiol concentration on HCG day and the oestradiol ratio were significantly different between the two groups. In groups A and B, the 95% CI for the oestradiol ratio were 2.6–3.4 and 3.6–4.3 respectively.


View this table:
[in this window]
[in a new window]
 
Table II. Clinical characteristics of patients in group A (1–3 follicles) and in group B (>3 follicles, including 12 over-responders whose cycles were cancelled)
 
Sensitivity and predictive value of the test at different thresholds
The sensitivity and positive and negative predictive values of the GAST in predicting over-response (>3 mature follicles) were calculated at the following thresholds: 3.7, 4.0 and 4.3, which represent the mean ± SEM of oestradiol ratio in group B. The findings are summarized in Table IIIGo.


View this table:
[in this window]
[in a new window]
 
Table III. Sensitivity and predictive values of gonadotrophin-releasing hormone agonist stimulation test at different threshold values of oestradiol ratio
 

    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Considering the personal commitment and expenditure required by patients in assisted conception, sufficient prognostic parameters are not readily available. The stage of ovarian stimulation is crucial to the success of assisted reproduction techniques. Prediction of the individual patient's poor response to stimulation may be attempted using the patient's age, day 3 FSH, with or without oestradiol concentrations, FSH/LH ratio, CCCT, and GAST in IVF patients. However, there is a different challenge when ovarian stimulation is used for IUI where prediction of over-response and poor response are of equal importance; it is also important to stimulate only a certain number of follicles so that the risk of multiple pregnancy may be reduced. There has not been any test available to predict over-response to ovarian stimulation. Such a test is important in centres where conversion to IVF treatment is not available. In this study we assessed the possibility of predicting ovarian over-response using GAST in a group of patients undergoing ovarian stimulation for IUI. We modified the GAST which was reported to be a sensitive predictor of ovarian response in flare-up IVF cycles (Winslow et al., 1991Go). The reason for modifying the test in this study was 3-fold: (i) measuring the oestradiol concentration beyond the second day post GnRHa stimulation, in the study reported by Winslow et al. (1991), was not of added significance; (ii) a value (oestradiol ratio) rather than a pattern of oestradiol rise facilitates the statistical analysis and the identification of a useful threshold for future use; (iii) since the absolute value of oestradiol is assay dependent, a ratio of two values measured with the same assay could be more reproducible.

In this study, oestradiol ratio was the only factor that was significantly correlated with the number of mature follicles (>=14 mm), and oestradiol on HCG day. Furthermore oestradiol ratio was distinctly different between the adequate (1–5 follicles >=14 mm in diameter) and the over-responders (>5 follicles >=14 mm) despite the similarity between the two groups with regard to age, day 2 FSH, FSH ratio, day 2 oestradiol, and FSH/LH ratio. The ability of the GAST to detect a difference in oestradiol response to GnRHa despite the small number of patients was repeated when the threshold was arbitrarily changed to three mature follicles. The sensitivity and predictive values of the test at a threshold of oestradiol ratio of 4.0 (mean oestradiol ratio in group B) was the best compromise for an acceptable sensitivity and positive and negative predictive values as shown in Table IIIGo. However, it should be noted that the GAST is suggested only as a screening test and no major decisions should be made based on the result of the test alone. Though the predictive values are fairly low, they can be considered acceptable given the population size in the study and the incidence of the event screened for.

The use of GnRHa in conjunction with gonadotrophin prior to IUI has been a controversial issue (Dodson et al., 1991Go, Galgiardi et al., 1991, Sengoku et al., 1994Go, Manzi et al., 1995Go). Nevertheless, we believe it offers a degree of flexibility with regard to monitoring, cycle cancellation, and when conversion to IVF seems appropriate; hence it was used for ovarian stimulation in the IUI programme. Nevertheless, the use of the GAST does not require the continuation of the GnRHa provided the values of oestradiol ratio obtained in this study are validated accordingly.

The similarity of the oestradiol ratio, day 2 FSH, day 2 oestradiol, and day 2 FSH/LH ratio between the poor responders and the other groups could be due to the small number of the poor responders in the study and/or the heterogeneity of this group of patients. It should also be noted that the character of ovarian response is relative to the stimulation protocol and the gonadotrophin dose used. Therefore, in any prospective assessment of this study, validation of the test according to the stimulation protocol and the relevant population is essential. Although the number of pregnancies is too small to draw a conclusion, the data suggest that similar pregnancy and lower multiple pregnancy rates would be obtained when <=3 follicles were recruited, as when a threshold of five mature follicles is used (as in this study).

We can conclude that prediction of ovarian over-response to ovarian stimulation in an IUI programme could be achieved using the modified GAST as suggested in this study and the patients could be counselled regarding the risk of cycle cancellation or conversion to IVF and multiple pregnancy. Furthermore, gonadotrophin dose adjustment could be decided prior to starting the treatment. The use of GnRH antagonist with recombinant FSH in the late follicular phase to induce a modest increase in the follicle numbers over the natural cycle has been suggested (Edwards et al., 1996Go). Whichever approach is used, it will have to be tested prospectively in a comparative study.


    Notes
 
3 To whom correspondence may be addressed at: Department of Obstetrics and Gynaecology, Solihull Hospital, Solihull, West Midlands B91 2JL, UK Back


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Balen, A. (1997) Anovulatory infertility and ovulation induction. Hum. Reprod. (Natl. Suppl., JBFS), 2, 83–87.

Cohen, J. (1998) How to avoid multiple pregnancies in assisted reproduction. Hum. Reprod., 13 (Suppl. 3), 197–214.[Medline]

Cruz, R.I., Kenman, E., Brandeis, V.T. et al. (1986) A prospective study of intrauterine insemination of processed sperm from men with oligoasthenospermia in superovulated women. Fertil. Steril., 46, 673–677.[ISI][Medline]

Dodson, W.C., Whitesides, D.B., Hughes, C.L. Jr et al. (1987) Superovulation with intrauterine insemination in the treatment of infertility: a possible alternative to gamete intrafallopian transfer and in vitro fertilization. Fertil. Steril., 48, 441–445.[ISI][Medline]

Dodson, W.C., Walmer, D., Hughes, C.L. et al. (1991) Adjunctive leuprolide therapy does not improve cycle fecundity in controlled ovarian hyperstimulation and intrauterine insemination of subfertile women. Obstet. Gynecol., 78, 187–190.[Abstract]

Edwards, R.G., Lobo, R. and Bouchard, P. (1996) Time to revolutionize ovarian stimulation. Hum. Reprod., 11, 917–919.[ISI][Medline]

Fédération CECOS, Schwartz, D. and Mayaux, B.A. (1982) Female fecundity as a function of age: results of artificial insemination in 2193 nulliparous women with azoospermic husbands. N. Engl. J. Med., 306, 404–406.[ISI][Medline]

Gagliardi, C.L., Adelina, M.E., Weiss, G. et al. (1991) Gonadotrophin-releasing hormone agonist improves the efficiency of controlled ovarian hyperstimulation/intrauterine insemination. Fertil. Steril., 55, 939–944.[ISI][Medline]

Manzi, D.L., Dumez, S., Scott, L.B. et al. (1995) Selective use of leuprolide acetate in women undergoing superovulation and intrauterine insemination results in significant improvement in pregnancy outcome. Fertil. Steril., 63, 866–873.[ISI][Medline]

Navot, D., Rosenwaks, Z. and Margolioth, E.J. (1987) Prognostic assessment of female fecundity. Lancet, ii, 645–647.

Padilla, S.L., Bayati, J. and Garcia, J.E. (1990) Prognostic value of early serum estradiol response to leuprolide acetate in in vitro fertilisation. Fertil. Steril., 53, 288–294.[ISI][Medline]

Sengoku, K., Tamate, K., Takaoka, Y. et al. (1994) A randomised prospective study of gonadotrophin with or without gonadotrophin-releasing hormone agonist for treatment of unexplained infertility. Hum. Reprod., 9, 1043–1047.[Abstract]

Serhal, P.F., Katz, M., Little, V. and Woronowski, H. (1988) Unexplained infertility — the value of Pergonal superovulation combined with intrauterine insemination. Fertil. Steril., 49, 602–606.[ISI][Medline]

Toner, J.P., Philput, C.B., Jones, G.S. et al. (1991) Basal follicle stimulating hormone level is a better predictor of in vitro fertilisation performance than age. Fertil. Steril., 55, 784–791.[ISI][Medline]

Winslow, K.L., Toner, J.P., Brzyski, R.G. et al. (1991) The gonadotropin-releasing hormone agonist stimulation test — a sensitive predictor of performance in the flare-up in vitro fertilization cycle. Fertil. Steril., 56, 711–717.[ISI][Medline]

Submitted on February 12, 1999; accepted on August 2, 1999.





This Article
Abstract
FREE Full Text (PDF )
Alert me when this article is cited
Alert me if a correction is posted
Services
Email this article to a friend
Similar articles in this journal
Similar articles in ISI Web of Science
Similar articles in PubMed
Alert me to new issues of the journal
Add to My Personal Archive
Download to citation manager
Request Permissions
Google Scholar
Articles by Lashen, H.
Articles by Clark, P.
PubMed
PubMed Citation
Articles by Lashen, H.
Articles by Clark, P.