The accuracy of multivariate models predicting ovarian reserve and pregnancy after in vitro fertilization: a meta-analysis.

BACKGROUND To review the accuracy of multivariate models for the prediction of ovarian reserve and pregnancy in women undergoing IVF compared with the antral follicle count (AFC) as single test. METHODS We performed a computerized MEDLINE and EMBASE search to identify articles published on multivariate models for ovarian reserve testing in patients undergoing IVF. In order to be selected, articles had to contain data on the outcome of IVF in terms of either pregnancy and/or poor response and on the prediction of these events based on a multivariate model. For the selected studies, sensitivity and specificity of the test in the prediction of poor ovarian response and non-pregnancy were calculated. Overall performance was assessed by estimating a summary receiver operating characteristic (ROC) curve, which was compared with the ROC curve for the AFC as the current best single test. RESULTS We identified 11 studies reporting on the predictive capacity of multivariate models in ovarian reserve testing. All studies reported on the prediction of poor ovarian response, whereas none reported on the occurrence of pregnancy. The sensitivity for prediction of poor ovarian response varied between 39% and 97% and the specificity between 50% and 96%. Logistic regression analysis indicated that cohort studies provided a significantly better discriminative performance than case-control studies. As cohort studies are superior to case-control studies, further analysis was limited to the cohort studies. For the cohort studies, a summary ROC curve could be estimated, which had a shape similar to that previously made for the AFC. CONCLUSIONS The accuracy of multivariate models for the prediction of ovarian response in women undergoing IVF is similar to the accuracy of AFC. No data are available on the capacity of these models to predict pregnancy, let alone live birth. On the basis of these findings, the use of more than one single test for the assessment of ovarian reserve cannot currently be supported.

Single and repeated GnRH agonist stimulation tests compared with basal markers of ovarian reserve in the prediction of outcome in IVF.

PURPOSE: To study the value of a single or repeated GnRH agonist stimulation test (GAST) in predicting outcome in IVF compared to basal ovarian reserve tests. METHODS: A total of 57 women was included. In a cycle prior to the IVF treatment, on day 3, an antral follicle count (AFC) was performed and blood taken for basal FSH, inhibin B and E2 measurements, followed by a subcutaneous injection of 100 microg triptorelin for the purpose of the GAST. Twenty-four hours later blood sampling was repeated. All the tests were repeated in a subsequent cycle. From the GAST E2 and inhibin B response were used as test parameters. The outcome measures were poor ovarian response and ongoing pregnancy. Group comparisons were done using the Mann-Whitney or chi-square test. Univariate and multivariate logistic regression was applied to assess which test revealed the highest predictive accuracy as expressed in the area under receiver-operating characteristic curve (ROC(AUC)). Clinical value was compared by calculating classical test characteristics for the best logistic models. RESULTS: All the basal and GAST variables were significantly different in the poor responders (n = 19) compared to normal responders (n = 38). In the univariate analysis on cycle 1 tests the AFC was the best predictor for poor ovarian response, while in cycle 2 the E2 response in the GAST performed best (ROC(AUC) of 0.91 for both). Multivariate analysis of the basal variables led to the selection of AFC and inhibin B in cycle 1, yielding a ROC(AUC) of 0.96. Mean E2 response was selected in a multivariate analysis of the repeated GAST variables (ROC(AUC) 0.91). At a specificity level of -0.90, several logistic models including GAST variables appeared to have a sensitivity (-0.80), positive predictive value (-0.82) and false positive rate (-0.18), comparable to a logistic model containing AFC and inhibin B. None of the test variables showed a significant relation with ongoing pregnancy. CONCLUSIONS: The GAST has a rather good ability to predict poor response in IVF. However, comparing the predictive accuracy and clinical value of the GAST with a day 3 AFC and inhibin B, it appeared that neither a single nor a repeated GAST performed better. In addition, the predictive ability towards ongoing pregnancy is poor. Therefore, the use of the GAST as a predictor of outcome in IVF should not be advocated.

Repeated clomiphene citrate challenge testing in the prediction of outcome in IVF: a comparison with basal markers for ovarian reserve.

BACKGROUND: The aim of this study was to investigate the predictive accuracy and clinical value of performing either a single or a repeated clomiphene citrate challenge test (CCCT) in predicting poor response in IVF, compared to that of currently used basal ovarian reserve markers. METHODS: Sixty-three patients undergoing their first IVF treatment were prospectively included. After measurement of basal markers on cycle day 3 (cd3) [FSH, inhibin B and antral follicle count (AFC)], a CCCT was performed. FSH and inhibin B levels were measured on day 10 (cd10). A second CCCT was performed after a washout period of one cycle. In all patients the tests were followed by an IVF treatment. Poor response (<4 oocytes or cancellation due to impaired (<3 follicles) or absent follicular growth) was used as primary outcome measure. RESULTS: Both the single as well as the repeated CCCT markers had a rather good discriminative potential for the prediction of poor response (area under the receiver operating characteristic curve (ROCAUC): FSH cd10=0.79, inhibin B cd10=0.79, mean FSH cd10=0.82 and mean inhibin B cd10=0.88). This compared well with the performance of the basal markers (FSH 0.82, inhibin B 0.72 and AFC 0.83). In a multivariate analysis on only the basal variables, FSH cd3 and AFC were selected (ROCAUC 0.89). Only stepwise forward analysis on the repeated CCCT variables revealed a better discriminating potential for the prediction of poor response (ROCAUC 0.92). At a specificity level of approximately 0.97, sensitivity and the positive predictive value were marginally improved in the CCCT models. CONCLUSIONS: Performing a CCCT (single or repeated) has a rather good ability to predict poor response in IVF. However, it appears that the predictive accuracy and clinical value of the CCCT is not clearly better than that of basal FSH in combination with an AFC. Therefore, the use of the CCCT as a predictor of outcome in IVF should not be advocated.

Use of stimulated serum estradiol measurements for the prediction of hyperresponse to ovarian stimulation in in vitro fertilization (IVF).

PURPOSE: In ovarian stimulation an exaggerated ovarian response is often seen and is related to medical complications, such as ovarian hyperstimulation syndrome (OHSS), and increased patient discomfort. If it were possible to identify hyperresponders at an early stage of the stimulation phase, adaptation of the stimulation protocol would become feasible to minimize potential complications. Therefore, we studied the usefulness of measuring stimulated serum estradiol (E2) levels in predicting ovarian hyperresponse. METHODS: A total of 109 patients undergoing their first IVF treatment cycle using a long protocol with GnRH agonist was prospectively included. The E2 level was evaluated on day 3 and 5 of the stimulation phase. Two outcome measures were defined. The first was ovarian hyperresponse (collection of > or = 15 oocytes at retrieval and/or peak E2 > 10000 pmol/L, or cancellation due to > or = 30 follicles growing and/or peak E2 > 15000 pmol/L, or OHSS developed). The second outcome measure comprised a subgroup representing the more severe hyperresponders. named extreme-response (cancellation or OHSS developed). RESULTS: The data of 108 patients were analyzed. The predictive accuracy of E2 measured on stimulation day 3 towards ovarian hyperresponse was clearly lower than that of E2 measured on stimulation day 5 (area under the receiver operating characteristic curve (ROCAUC) 0.75 and 0.81, respectively). For extreme-response the predictive accuracy of E2 measured on stimulation day 3 or 5 was comparable (ROCAUC 0.81 and 0.82, respectively). For both outcome measures the stimulated E2 tests yielded only acceptable specificity with moderate sensitivity at higher cutoff levels. Prediction of extreme-response seemed slightly more effective due to a lower error rate. CONCLUSIONS: There is a significant predictive association between E2 levels measured on stimulation day 3 and 5 and both ovarian hyperresponse and extreme-response in IVF. However, the clinical value of stimulated E2 levels for the prediction of hyperresponse is low because of the modest sensitivity and the high false positive rate. For the prediction of extreme-response the clinical value of stimulated E2 levels is moderate.

Serum anti-Müllerian hormone levels: a novel measure of ovarian reserve.

BACKGROUND: Anti-Müllerian hormone (AMH) is produced by the granulosa cells of preantral and small antral follicles and its levels can be assessed in serum. Since the number of ovarian follicles declines with increasing age, AMH levels might be used as a marker for ovarian ageing. Therefore, we studied the relationship between AMH levels and ovarian response during ovarian stimulation for IVF. METHODS: A total of 130 patients undergoing their first IVF treatment cycle using a long protocol with GnRH agonist was prospectively included. Blood withdrawal was performed and the number of antral follicles was assessed by ultrasound on day 3 of a spontaneous cycle. Poor response and the number of oocytes were used as primary outcome measures. In a random subset of 23 patients a GnRH agonist stimulation test was performed to investigate whether a rise in FSH and LH would affect AMH levels. RESULTS: The data of 119 patients were analysed. Serum AMH levels were highly correlated with the number of antral follicles (r = 0.77; P < 0.01) and the number of oocytes retrieved (r = 0.57, P < 0.01). A negative association was found between AMH levels and poor ovarian response (fewer than 4 oocytes or cycle cancellation; OR 0.82, 95% CI 0.75-0.90, P < 0.01). Inclusion of inhibin B and FSH concentrations to AMH in a multivariate model improved the prediction of ovarian response. The post GnRH agonist rise in FSH and LH levels did not influence AMH values. CONCLUSIONS: Poor response in IVF, indicative of a diminished ovarian reserve, is associated with reduced baseline serum AMH concentrations. In line with recent observations it appears that AMH can be used as a marker for ovarian ageing.