1 Center for Clinical Decision Sciences, Department of Public Health and 2 Division of Reproductive Medicine, Department of Obstetrics and Gynaecology, Erasmus MC- University Medical Center Rotterdam, The Netherlands
3 To whom correspondence should be addressed at: Division of Reproductive Medicine, Department of Obstetrics and Gynaecology, Erasmus MC-University Medical Center Rotterdam, Dr. Molewaterplein 40, 3015 GD Rotterdam, The Netherlands. e-mail: b.c.j.m.fauser{at}erasmusmc.nl
![]() |
Abstract |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Key words: anovulation/clomiphene citrate/hMG/pregnancy rate/prognosis
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Over the past 10 years, increased attention has focused towards alternative treatment options such as the insulin-sensitizing agents (Glueck et al., 2002; Heard et al., 2002
; Nestler et al., 2002
), aromatase inhibitors (Mitwally and Casper, 2001
), laparoscopic surgery of the ovaries (Cohen, 1996
; Farquhar et al., 2001
) or assisted reproduction such as intrauterine insemination (IUI) or IVF (Buyalos and Lee, 1996
; Child et al., 2001
). Most promising results of up to 80% pregnancies have been reported by individual studies, but the majority of these reports involved a limited number of selected patients and were retrospective and uncontrolled. It is often suggested that the multiple birth rate is reduced in comparison with ovulation induction, but reliable information on multiple births is mostly lacking.
There seems to be a clear need to firmly establish the success and complication rates of the conventional approach for ovulation induction (involving CC and FSH) in a large unselected cohort of patients. The present report contains an integrated analysis of chances for pregnancies resulting in singleton live childbirth, extending previously reported outcomes separately for CC (Imani et al., 2002a) and FSH (Imani et al., 2002b
; Mulders et al., 2003
). In addition, a prediction model is presented to help identify patients with poor chances of success. This information may serve as point of reference for future studies involving alternative approaches for ovulation induction.
![]() |
Materials and methods |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
The standardized initial clinical, sonographic and endocrine screening took place prior to the initiation of CC ovulation induction, as has been described previously (Imani et al., 2002a). The clinical screening included patients age, type of infertility (primary or secondary), cycle history (amenorrhoea or oligomenorrhoea), BMI, waist-to-hip ratio (WHR) and previous medication and/or surgery. Transvaginal sonography included assessment of ovarian volume (ml), and total number of follicles, as described previously (Pache et al., 1992
). Endocrine screening included serum assays for FSH, LH, E2, androstenedione, testosterone, sex-hormone binding globulin (SHBG), and fasting insulin and glucose. The hormone assays used, and the intra- and inter-assay coefficients of variation valid for this study, have all been described previously (Imani et al., 2000
).
The treatment protocol and assessment of ovarian response and pregnancy after CC administration have also been described previously (Imani et al., 2002a). In brief, the women received initial CC doses of 50 mg/day from cycle day 3 until day 7 after spontaneous or progestagen-induced withdrawal bleeding. In cases of absent ovarian response, the daily dosage was increased to 100 and 150 mg in subsequent cycles. Ovulation after CC or FSH medication was assessed by sonographic monitoring of disappearance of the pre-ovulatory follicle and/or the finding of midluteal progesterone levels >25 nmol/l. In case of CC-resistant anovulation (CRA)that is, absent ovulation after 3 months, despite stimulation with a maximum daily CC dose of 150 mgor if pregnancy failed to occur within six ovulatory cycles, ovulation induction with exogenous recombinant FSH (Puregon®; Organon NV, Oss, The Netherlands) was applied as second-line treatment, using a decremental dose regimen as published previously (van Santbrink and Fauser, 1997
). The FSH response dose was assessed during the first cycle applying a low-dose, step-up regimen (Imani et al., 2002b
; Mulders et al., 2003
).
Pregnancy was defined as a positive urinary pregnancy test (Clearview, hCG II; Unipath Ltd, Bedford, UK) at least 3 days after the expected menses. Ongoing pregnancy was defined as sonographic assessment at 12 weeks amenorrhoea of an intrauterine gestational sac with a positive heartbeat. Live birth was defined as the delivery of a baby after at least 28 weeks gestational age. Information regarding deliveries and the health condition of the babies born was collected using hospital records. In cases of home delivery, the information was collected directly from the patient and her general practitioner or midwife.
Statistical analysis
The KaplanMeier method was used to estimate the cumulative pregnancy rate leading to singleton live birth. The period from the start of CC treatment until the first pregnancy leading to live birth was the time variable in this analysis. Conceptions that ended in miscarriage were ignored by the analysis, and in these cases follow-up continued until pregnancy resulting in singleton live birth occurred. Patients who had a pregnancy ending in premature birth, stillbirth or multiple live birth, were considered censored at conception. Patients who did not become pregnant were censored at their time of last treatment. Spontaneous pregnancies resulting in singleton live birth that occurred during the first or second month after the end of treatment were also included in the analysis.
Cox regression was used for univariable and multivariable analysis relating initial screening characteristics to the cumulative pregnancy rate leading to singleton live birth. In univariable analysis the log-rank test was used, and a P-value < 0.05 was considered to be statistically significant. A multivariable prediction model was constructed using a backward stepwise elimination method with P < 0.15 for exclusion of predictors, corresponding to Akaikes Information Criterion (Akaike, 1973). Predictor variables with univariable P < 0.5 were candidate variables for the multivariable model. These liberal P-values were chosen because the resulting model may be expected to have a better predictive performance in new patients than when the strict P < 0.05 criterion is used (Harrell et al., 1996
). The prediction model was corrected by a shrinkage factor, determined by a bootstrapping technique with 200 replications, to provide better predictions in new patients (Van Houwelingen and Le Cessie, 1990
; Harrell et al., 1996
). The discriminative ability of the prediction model was assessed by the c-statistic, which is similar to the area under the receiver operating characteristic (ROC) curve for dichotomous outcomes, and can range from 0.5 to 1.0 (Harrell et al., 1985
). The bootstrap technique was used again, this time to correct the optimism in the apparent c-statistic (Harrell et al., 1996
). Missing values occurred for some screening characteristics. The technique of multiple imputation was used to avoid loss of information due to missing data in multivariable analyses (Little, 1992
). Missing data occurred in the WHR (45% missing), semen analysis (25% missing) and insulin:glucose ratio (26% missing). Statistical analyses were performed with SPSS for Windows (version 10) and S+ 2000.
![]() |
Results |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
|
|
|
![]() |
Discussion |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
A prognostic model based on age of the woman, the insulin:glucose ratio and the duration of infertility predicted probabilities of pregnancy resulting in singleton live birth at 12 months ranging from 15 to 82%. The apparent c-statistic of the model was 0.64, and the optimism-corrected value was 0.61, indicating only a moderate discriminative ability. However, the model was able to identify patients with low chances for success applying classical ovulation induction who may therefore qualify for alternative first-line treatment options such as insulin-sensitizing agents, assisted reproductive technologies or laparoscopic ovarian surgery. For example, a probability lower than 30% was predicted in 25 (10%) out of the 240 patients. However, this cut-off value for low chance of 30% was arbitrarily chosen and other values may be just as valid. In a general infertility setting, the individual choice between several treatment options should depend on the chances of success, the complication rates, the burden to the patient, and costs. Taking these considerations into account, clinicians can choose any desired cut-off value for the chance of success of ovulation induction, and derive from Figure 3 how many patients will be identified as having a low chance of success. In previous studies by the present authors, with regard to outcome following CC-treatment alone, a history of amenorrhoea (as opposed to oligomenorrhoea) was found to be an adverse factor for ovulation (Imani et al., 1998), but a positive factor for pregnancy once ovulation was achieved (Imani et al., 1999
). In the present study, amenorrhoea had no effect on outcome, indicating that probably both effects level out, as had been noted previously (Imani et al., 2002a
). Additional factors involved in the prediction of live birth following CC treatment included FAI, BMI and female age. Previously identified screening characteristics involved in the prediction of ongoing pregnancy following FSH ovulation induction included concentrations of both insulin-like growth factor-1 (IGF-I) and testosterone, as well as age (Mulders et al., 2003
).
Almost 50% of the CC-failure (CCF) patients did not proceed to FSH treatment and should be considered as drop-outs. In general, drop-outs may cause bias in the estimates of outcome when their prognosis differs from the patients who did not drop out (selective drop-out), as noted in a previous study (Imani et al., 2002a). However, the screening characteristics that were predictive of success were similar between these patients and the patients who continued with FSH treatment. Only endogenous FSH levels were different, but since that screening characteristic was not associated with outcome in the prognostic model, it can be concluded that there is no indication of selective drop-out. The 37 women who did not reach the end-point following FSH treatment had received a median number of five treatment cycles, which was close to the standard number of treatment cycles according to protocol. Therefore they were not to be considered as drop-outs.
It is generally perceived that the majority of patients who qualify for exogenous gonadotrophins for ovulation induction suffer from PCOS (White et al., 1996; Fauser and Van Heusden, 1997
). However, scientific evidence to support this believe is scant. It has recently been established that 70% of women suffering from normogonadotrophic anovulation present with polycystic ovaries as diagnosed by ultrasound, and 61% with endocrine signs associated with PCOS (i.e. hyperandrogenaemia) (Laven et al., 2002
). However, insulin resistanceas defined by an increased fasting insulin:glucose ratiowas observed in only 21% of women, whereas 44% were obese. All these clinical, endocrine and ultrasound abnormalities are involved in ovarian dysfunction in PCOS, and may therefore potentially predict response and outcome of ovarian stimulation by exogenous FSH (Balen et al., 1993
; Fulghesu et al., 1997
). It was surprising therefore to find that none of these featuresexcept for insulin resistancecould be identified in the present study as a predictor of pregnancy in multivariable analysis. Moreover, in case the overall heterogeneous WHO 2 group was classified according to definitive PCOS or definitive non-PCOS (see Results section for details), again no differences in pregnancy chances could be observed. Not even obesity (defined as an elevated BMI) or abnormal fat distribution (indicated by a high WHR) was capable of predicting outcome, despite much emphasis in recent literature on the impact of excess bodyweight (Hamilton-Fairley et al., 1992
) and weight reduction (Norman et al., 2002
) on the success of fertility therapies in anovulatory women. These discrepancies clearly underline the need for properly designed, prospective studies in a well-defined patient population.
In conclusion, the results of the present study showed that, overall, normogonadotrophic anovulatory women have good prospects of singleton live birth when treated according to the classical induction of ovulation algorithm, including CC as first-line treatment followed by exogenous FSH as second line. Using a multivariate prediction model, individual women with chances for success (far) above or below average can be identified. For women with a predicted low chance, the use of alternative treatment strategies may be appropriate.
![]() |
Acknowledgements |
---|
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Balen, A.H., Tan, S.L. and Jacobs, H.S. (1993) Hypersecretion of luteinising hormone: a significant cause of infertility and miscarriage. Br. J. Obstet. Gynecol., 100, 10821089.[ISI][Medline]
Buyalos, R.P. and Lee, C.T. (1996) Polycystic ovary syndrome: pathophysiology and outcome with in vitro fertilization. Fertil. Steril., 65, 110.[ISI][Medline]
Child, T.J., Abdul-Jalil, A.K., Gulekli, B. and Tan, S.L. (2001) In vitro maturation and fertilization of oocytes from unstimulated normal ovaries, polycystic ovaries, and women with polycystic ovary syndrome. Fertil. Steril., 76, 936942.[CrossRef][ISI][Medline]
Cohen, J. (1996) Laparoscopic procedures for treatment of infertility related to polycystic ovarian syndrome. Hum. Reprod. Update, 2, 337344.
Farquhar, C., Vandekerckhove, P. and Lilford, R. (2001) Laparoscopic drilling by diathermy or laser for ovulation induction in anovulatory polycystic ovary syndrome. Cochrane Database Syst. Rev., 4.
Fauser, B.C. and Van Heusden, A.M. (1997) Manipulation of human ovarian function: physiological concepts and clinical consequences. Endocr. Rev., 18, 71106.
Fulghesu, A.M., Villa, P., Pavone, V., Guido, M., Apa, R., Caruso, A. et al. (1997) The impact of insulin secretion on the ovarian response to exogenous gonadotropins in polycystic ovary syndrome. J. Clin. Endocrinol. Metab., 82, 644648.
Glueck, C.J., Wang, P., Goldenberg, N. and Sieve-Smith, L. (2002) Pregnancy outcomes among women with polycystic ovary syndrome treated with metformin. Hum. Reprod., 17, 28582864.
Hamilton-Fairley, D., Kiddy, D., Watson, H., Paterson, C. and Franks, S. (1992) Association of moderate obesity with a poor pregnancy outcome in women with polycystic ovary syndrome treated with low dose gonadotrophin. Br. J. Obstet. Gynecol., 99, 128131.[ISI][Medline]
Hammond, M.G., Halme, J.K. and Talbert, L.M. (1983) Factors affecting the pregnancy rate in clomiphene citrate induction of ovulation. Obstet. Gynecol., 62, 196202.[Abstract]
Harrell, F.E., Jr, Lee, K.L., Matchar, D.B. and Reichert, T.A. (1985) Regression models for prognostic prediction: advantages, problems, and suggested solutions. Cancer Treat. Rep., 69, 10711077.[ISI][Medline]
Harrell, F.E., Jr, Lee, K.L. and Mark, D.B. (1996) Multivariable prognostic models: issues in developing models, evaluating assumptions and adequacy, and measuring and reducing errors. Stat. Med., 15, 361387.[CrossRef][ISI][Medline]
Heard, M.J., Pierce, A., Carson, S.A. and Buster, J.E. (2002) Pregnancies following use of metformin for ovulation induction in patients with polycystic ovary syndrome. Fertil. Steril., 77, 669673.[CrossRef][ISI][Medline]
Imani, B., Eijkemans, M.J., te Velde, E.R., Habbema, J.D. and Fauser, B.C. (1998) Predictors of patients remaining anovulatory during clomiphene citrate induction of ovulation in normogonadotropic oligoamenorrheic infertility. J. Clin. Endocrinol. Metab., 83, 23612365.
Imani, B., Eijkemans, M.J., te Velde, E.R., Habbema, J.D. and Fauser, B.C. (1999) Predictors of chances to conceive in ovulatory patients during clomiphene citrate induction of ovulation in normogonadotropic oligoamenorrheic infertility. J. Clin. Endocrinol. Metab., 84, 16171622.
Imani, B., Eijkemans, M.J., de Jong, F.H., Payne, N.N., Bouchard, P., Giudice, L.C. et al. (2000) Free androgen index and leptin are the most prominent endocrine predictors of ovarian response during clomiphene citrate induction of ovulation in normogonadotropic oligoamenorrheic infertility. J. Clin. Endocrinol. Metab., 85, 676682.
Imani, B., Eijkemans, M.J., te Velde, E.R., Habbema, J.D. and Fauser, B.C. (2002a) A nomogram to predict the probability of live birth after clomiphene citrate induction of ovulation in normogonadotropic oligoamenorrheic infertility. Fertil. Steril., 77, 9197.[CrossRef][ISI][Medline]
Imani, B., Eijkemans, M.J., Faessen, G.H., Bouchard, P., Giudice, L.C. and Fauser, B.C. (2002b) Prediction of the individual follicle-stimulating hormone threshold for gonadotropin induction of ovulation in normogonadotropic anovulatory infertility: an approach to increase safety and efficiency. Fertil. Steril., 77, 8390.[CrossRef][ISI][Medline]
Kousta, E., White, D.M. and Franks, S. (1997) Modern use of clomiphene citrate in induction of ovulation. Hum. Reprod. Update, 3, 359365.
Laven, J.S., Imani, B., Eijkemans, M.J. and Fauser, B.C. (2002) New approach to polycystic ovary syndrome and other forms of anovulatory infertility. Obstet. Gynecol. Surv., 57, 755767.[CrossRef][ISI][Medline]
Legro, R.S. (1998) Polycystic ovary syndrome: current and future treatment paradigms. Am. J. Obstet. Gynecol., 179, S101S108.[ISI][Medline]
Little, R.J.A. (1992) Regression with missing Xs: a review. J. Am. Statist. Assoc., 87, 12271237.[ISI]
Messinis, I.E. and Milingos, S.D. (1997) Current and future status of ovulation induction in polycystic ovary syndrome. Hum. Reprod. Update, 3, 235253.
Mitwally, M.F. and Casper, R.F. (2001) Use of an aromatase inhibitor for induction of ovulation in patients with an inadequate response to clomiphene citrate. Fertil. Steril., 75, 305309.[CrossRef][ISI][Medline]
Mulders, A.G., Eijkemans, M.J., Imani, B. and Fauser, B.C. (2003) Prediction of chances for success or complications in gonadotropin ovulation induction in normogonadotropic anovulatory infertility. Reprod. Biomed. Online 7.4856.
Nestler, J.E., Stovall, D., Akhter, N., Iuorno, M.J. and Jakubowicz, D.J. (2002) Strategies for the use of insulin-sensitizing drugs to treat infertility in women with polycystic ovary syndrome. Fertil. Steril., 77, 209215.[CrossRef][ISI][Medline]
Norman, R.J., Davies, M.J., Lord, J. and Moran, L.J. (2002) The role of lifestyle modification in polycystic ovary syndrome. Trends Endocrinol. Metab., 13, 251257.[CrossRef][ISI][Medline]
Pache, T.D., Wladimiroff, J.W., Hop, W.C. and Fauser, B.C. (1992) How to discriminate between normal and polycystic ovaries: transvaginal US study. Radiology, 183, 421423.[Abstract]
The ESHRE Capri Workshop Group (1995) Anovulatory infertility. Hum. Reprod., 10, 15491553.[Abstract]
Van Houwelingen, J.C. and Le Cessie, S. (1990) Predictive value of statistical models. Stat. Med., 9, 13031325.[ISI][Medline]
van Santbrink, E.J. and Fauser, B.C. (1997) Urinary follicle-stimulating hormone for normogonadotropic clomiphene- resistant anovulatory infertility: prospective, randomized comparison between low dose step-up and step-down dose regimens. J. Clin. Endocrinol. Metab., 82, 35973602.
van Santbrink, E.J., Hop, W.C. and Fauser, B.C. (1997) Classification of normogonadotropic infertility: polycystic ovaries diagnosed by ultrasound versus endocrine characteristics of polycystic ovary syndrome. Fertil. Steril., 67, 452458.[ISI][Medline]
White, D.M., Polson, D.W., Kiddy, D., Sagle, P., Watson, H., Gilling-Smith, C. et al. (1996) Induction of ovulation with low-dose gonadotropins in polycystic ovary syndrome: an analysis of 109 pregnancies in 225 women. J. Clin. Endocrinol. Metab., 81, 38213824.[ISI][Medline]
World Health Organization (1993) WHO Manual for the Standardized Investigation and Diagnosis of the Infertile Couple. Cambridge University Press, Cambridge, UK.
Submitted on April 25, 2003; resubmitted on June 13, 2003; accepted on July 17, 2003.