- Open Access
- Open Peer Review
Ovarian stimulation with human and recombinant gonadotropin – comparison of in vitro fertilization efficiency with use of time-lapse monitoring
© Wdowiak and Bojar. 2015
- Received: 24 July 2015
- Accepted: 4 December 2015
- Published: 15 December 2015
Achieving pregnancy by in vitro fertilization (IVF) treatment depends on many factors, including the ovaries’ capacity and the efficiency of ovarian stimulation. The aim of this study was to assess the influence of ovarian stimulation with human and recombinant gonadotropin, as well as specific hormonal parameters, on the effectiveness of IVF and the dynamics of embryonic development.
The study involved 221 women aged 25–35 years in whom intracytoplasmic sperm injection was performed. The ovarian stimulation was carried out according to the short protocol: injections of gonadotropin-releasing hormone analogue were followed by human (hFSH) and recombinant (rFSH) follicle-stimulating hormone administration. The growth of embryos was monitored with a time-lapse system. Levels of follicle-stimulating hormone (FSH), luteinizing hormone (LH), and anti-Müllerian hormone (AMH) were measured before ovarian stimulation, and levels of estradiol were assessed on the day of administration of recombinant chorionic gonadotropin.
Pregnancy was achieved in 77 women (group A) – 42 (54.55 %) of them were stimulated with hFSH and 35 (45.45 %) were stimulated with rFSH. Among the 144 women in whom pregnancy was not achieved (group B), hFSH was administered to 73 (50.69 %) women and rFSH to 71 (49.31 %) women. In both groups subsequent embryo development stages were usually noted earlier after hFSH stimulation than after rFSH stimulation. The average values of AMH, estradiol, and estradiol per >17 mm follicle were higher in group A; in turn, FSH and LH mean levels were higher in group B. ROC curve analysis showed no statistically significant differences between accuracy of using FSH and AMH levels to predict pregnancy after IVF.
The kind of gonadotropin applied to stimulate ovaries impacts the dynamics of embryo development - in women stimulated with hFSH, subsequent development stages were usually observed earlier than in women treated with rFSH; however, there was no statistically significant difference in pregnancy rates between women who were hFSH stimulated and those who were rFSH stimulated. The mean estradiol level was higher in women who achieved pregnancy than in women in whom pregnancy was not achieved AMH and FSH have the greater impact on achieving pregnancy than other hormones, and the value of AMH and FSH in predicting pregnancy is similar.
- In vitro fertilization (IVF)
- Intracytoplasmic sperm injection (ICSI)
- Ovarian stimulation
- Embryo development dynamics
- Follicle-stimulating hormone
- Anti-Müllerian hormone
- Time-lapse monitoring
Infertility is recognized as a social disease, and its prevalence is estimated to be around 9 % worldwide for women in reproductive age. The method the most often used to treat infertility is in vitro fertilization (IVF) with intracytoplasmic sperm injection (ICSI). This method gives the highest success rate per cycle in comparison with other treatment options . As a preparation for IVF/ICSI, ovarian stimulation is performed in order to induce the development of multiple follicles of the ovaries. Then, oocytes are aspirated, injected with sperm, and placed into special medium. After a few days of culture, the embryos regarded as the best-developed are transferred into the woman’s uterus . Achieving pregnancy by means of IVF/ICSI treatment depends on many factors, including the capacity of the ovaries and the performed ovarian stimulation. The ovarian reserve is assessed i.a. on the basis of gonadotropins (follicle-stimulating hormone – FSH and luteinizing hormone - LH) and anti-Müllerian hormone (AMH) levels . Ovarian stimulation may be performed with use human or recombinant FSH; however, it is still not clear which kind of gonadotropin gives better IVF outcomes [4–7]. On the other hand, although estrogens and progesterone have been identified as the key hormones involved in the course of oocyte maturation, which appears to be strongly linked to a successful result in assisted reproduction treatment, it has been not clarified whether or not they affect the probability of pregnancy [8, 9]. IVF/ICSI outcomes may also be influenced by proper selection of the embryo to transfer. For a long time embryos were evaluated based on morphological features. Recently, there has been a notable increase in monitoring embryos in real time using a camera placed inside the incubator. A time-lapse embryo monitoring system enables observation of all embryos without any disturbance in incubation conditions or alteration of daily routine [10–13]. The embryo growth is observed at fixed times, and those that show better dynamics of development are regarded as the most suitable for transfer .
The aim of this study was to assess the influence of ovarian stimulation with human and recombinant gonadotropin, as well as specified hormonal parameters, on the effectiveness of IVF and the dynamics of embryo development.
The presented study was conducted in the years 2013–2014 in the “Ovum Reproduction and Andrology” Non-Public Health Care Unit in Lublin (Poland). The study involved 223 women undergoing IVF treatment for the first time. The inclusion criteria were: age below 35 years, FSH level ≤10 mIU/mL, AMH level ≥1.5 ng/mL, and Body Mass Index (BMI) ≤30 kg/m2. Women with severe endometriosis, metabolic disease, or leiomyoma were excluded from the study. In all women ICSI was performed due to abnormal semen parameters. In two women embryo development stopped before achieving blastocyst stage, and therefore they were excluded from the study. Finally, analysis was made for the 221 women in whom embryos achieved the blastocyst stage and were transferred into uterus.
The growth of all the embryos was monitored continuously by obtaining images at 10-minute intervals. This imaging was achieved with use of a compact time-lapse microscope system (Primo Vision EVO Microscope, Cryo-Innovation, Hungary) placed inside incubator combined with a microwell embryo culture dish. During the time-lapse observation, the embryos were not moved. Between image acquisitions, the system was turned off completely to avoid exposure of the embryos to electromagnetic radiation.
The following terms for the timing of the stages of embryo growth were adopted: t0 - the time when ICSI was carried out; tF - the time of the first frame in which both pronuclei could be observed; tC - the time of the frame with the last observation of both pronuclei; t1 – the time when one cell stage was observed; t2, t3, t4, t5, t6, t7, t8, and t9 - the stages for the corresponding number of cells, e.g. t2 for 2 cells, t3 for 3 cells, etc. (stages were annotated at the first frame in which the cells were seen as separated by membranes); tM - the time of the first frame in which the embryos were compacting into the morula stage; tB - the time of the frame in which a crescent-shaped area began to emerge from the morula; and tEB – the time of the frame with expanded blastocyst with increased volume and expansion of the blastocoele cavity.
Based on the analysis of time-lapse records and in accordance with American Society for Reproductive Medicine and the European Society of Human Reproduction and Embryology consortium guidelines , a single blastocyst was selected to be transferred into the uterus. After 6 weeks, the presence of the embryo and its cardiac activity were assessed via ultrasound.
Determination of hormone levels and imaging studies
Levels of FSH, LH, and AMH were determined on the third day of the cycle preceding ovulation, before administration of drugs. Ultrasound examination was performed daily since 9th day of cycle and after observation of a follicle with dimension equal to or greater than 17 mm; daily measurement of estradiol levels was implemented. The estradiol values were then converted into the levels of estradiol per follicle equal to or greater than 17 mm (E2/f). All hormones levels were measured in serum obtained from morning blood samples (5 mL). Levels of FSH, LH, and estradiol were assessed with electrochemiluminescent method on a Cobas analyzer (Roche Diagnostics) – the reference values were: for FSH 3.5–12.5 mIU/mL; for LH 2.4–12.6 mIU/mL; and for estradiol 12.5–166 pg/mL. Levels of AMH were measured with AMH Gen II ELISA test (Beckman Coulter) on a Euroimmun analyzer (reference range > 1.5 ng/mL).
The obtained results were subjected to the Statistica 9.1 software system (StatSoft, Poland). The measurable parameters were reported as mean (M), standard deviation (SD), and minimum (Min) or maximum (Max) whereas the values of the immeasurable parameters were reported as proportion and quantity. For quality attributes, the Chi2 test was performed to show differences between the examined groups. The Shapiro-Wilk normality test was used to check the normality of distribution of variables in the examined groups. The Mann-Whitney U-test was carried out to examine differences between the groups. The r-Pearson correlation test was used to check the correlations between variables. The receiver operating characteristic (ROC) curve was used to assess diagnostic values of the tested parameters. The significance level was set at p <0.05, which indicates the existence of statistically significant differences or correlations.
The studies were approved by the Ethics Committee of the Institute of Rural Health in Lublin. All women were provided with oral and written information about the study and signed a written consent allowing the use of their data for research purposes.
Characteristics of the studied population and the effectiveness of IVF
Baseline characteristics of the entire studied population and groups
n = 77
n = 144
n = 221
M ± SD
30.70 ± 3.06
31.0 ± 2.68
30.90 ± 2.93
Min – Max
25 - 35
25 - 35
25 - 35
M ± SD
22.58 ± 3.12
23.04 ± 3.34
22.88 ± 3.27
Min – Max
17 - 30
17 - 30
17 - 30
Number of women stimulated with hFSH
Number of women stimulated with rFSH
In group B, faster embryo development after hFSH than after rFSH stimulation was observed for time stages: tF, tC, t1, t2, t3, t4, t9, and tB, and statistically significant differences were found for all these time stages except for t9: tF (Z = −3.572, p <0.001), tC (Z = −3.612, p <0.001), t1 (Z = −2.322, p = 0.020), t2 (Z = −3.823, p <0.001), t3 (Z = −3.512, p <0.001), t4 (Z = −3.929, p <0.001), and tB (Z = −4.133, p <0.001). Additionally, a statistically significant difference was noted for t6 stage that was recorded earlier for rFSH-stimulated embryos than for embryos stimulated with hFSH (Z = 2.026, p = 0.043).
Evaluation of hormone levels
Comparison of mean hormone levels in the group of women who achieved pregnancy and in the group of women in whom pregnancy was not achieved
Group A (n = 77)
Group B (n = 144)
Correlations between hormone levels and time stages of embryo development
Correlations between hormone levels and time stages of embryo development in women who achieved pregnancy and in women in whom pregnancy was not achieved
Group A (n = 77)
Group B (n = 144)
In group B, weak negative correlations were found between AMH levels and the following stages: t1 (r = −0.197, p = 0.018), t3 (r = −0.221, p = 0.008), t4 (r = −0.237, p = 0.004), t6 (r = −0.185, p = 0.026), and tB (r = −0.279, p = 0.001). Weak positive correlations were found for FSH level at tB stage (r = 0.281, p = 0.001) and for E2/f for t8 stage (r = 0.254, p = 0.002). For LH level, a weak negative correlation was noted only at tF stage (r = −0.190, p = 0.023). Weak negative correlations for estradiol levels were observed in the following time stages: tF (r = −0.214, p = 0.010), t3 (r = −0.166, p = 0.047), t4 (r = −0.225, p = 0.007), and tB (r = −0.234, p = 0.005).
Prognostic value of AMH and FSH levels on the outcomes of IVF
The dynamics of embryo development and IVF outcomes are conditioned by many factors. We evaluated the influence of gonadotropins on the timing of embryo development with use the time-lapse monitoring, and we compared hormones levels in women who achieved pregnancy and in women who did not achieve pregnancy with ICSI procedure. We are aware that other factors such as age, BMI, or quality of oocytes can affect embryo development and ICSI outcomes; however, in order to make the text more clear and understandable we decided to focus only on a few items.
Pregnancy, in our study, was achieved in almost 35 % of women (group A). This ratio is comparable with results obtained by other authors e.g. Kirkegaard et al. (31 %) and Polanski et al. (35 %) [16, 17]. Taking into consideration the kind of gonadotropin used, we noted that the pregnancy rate was slightly higher in women stimulated with hFSH than in women stimulated with rFSH; however, this difference was not statistically significant. The observations made by Flicori et al. were similar – they also noted higher frequency of pregnancies in women treated with hFSH than in women treated with rFSH, and their results also turn out to be statistically insignificant . In turn, Kilani et al. obtained 35 % frequency of pregnancies both for hFSH-stimulated women and rFSH-stimulated women. However, they observed that hFSH treatment was associated with a more efficient response than stimulation with rFSH .
In our study, in women who achieved pregnancy almost all subsequent embryo development stages were observed earlier than in women who did not achieve pregnancy, but only part of these differences were statistically significant (tC, t1,t2, t4, and tB stages). When evaluating the impact of gonadotropins on embryos growth we noted faster development after stimulation with hFSH than after rFSH administration, but the differences were significant only at stages tC, t2, and t3 in women who achieved pregnancy and at stages tF-t4 and tB in women who did not achieve pregnancy. Slightly different results were obtained by Muñoz et al. – they noted that after administration of rFSH embryos showed better timing of development than after stimulation with hFSH or both rFSh and hFSH; however, these differences turned out to be not significant. Contrary to our study, they observed no significant differences at t2 stage between women stimulated with rFSH or hFSH or both rFSH and hFSH. Muñoz et al. used a different protocol for oocyte stimulation, but it seems that these divergences are rather due to using different types of gonadotropins - in the case of hFSH they used gonadotropins containing LH activity, and we used a medicament free of LH .
Analyzing hormone levels in the studied groups, we noted that FSH mean concentration was higher in the group of women in whom pregnancy was not achieved, contrary to levels of AMH, estradiol, and E2/f that were higher in women who achieved pregnancy. Other studies also proved that a high estradiol level improves embryos development and IVF outcome, expressed as number of retrieved oocytes, number of high-grade embryos, number of transferred embryos, and implementation rate [18–22]. Munoz et al. also assessed that better IVF outcomes are achieved when the concentration of estradiol exceeds 2000 pg/mL. In turn, Kara et al. indicated a 4000 pg/mL value [18, 22]. Our results seems to confirm the observation of Munoz et al. because the mean estradiol level in women who achieved pregnancy was about 2600 pg/mL, in contrast to women in whom pregnancy was not achieved with mean estradiol level about 1700 pg/mL. In the present study the higher estradiol level obtained in women who achieved pregnancy may be due to the hFSH stimulation that was applied to over 50 % of these women. As shown by Kilani et al., treatment with hFSH is associated with higher estradiol concentration than stimulation with rFSH . Moreover, we observed negative correlations between estradiol level and timing of embryo development at stages t2, t4, and t5, which is in agreement with the results obtained by Muñoz et al. - they noted differences in the embryo dynamics at stage t5 and in cc2, dependent on estradiol level, which is the duration of the period as a 2-blastomere embryo (t3-t2) .
Endocrine markers thought to be useful in distinguishing good and poor responders for ovarian stimulation are, among others, FSH and AMH, but none of them is adequate for predicting pregnancy outcomes . In our study AMH levels were significantly higher in women who achieved pregnancy than in women in whom pregnancy was not achieved, and the sensitivity and specificity of AMH in predicting pregnancy were 46.75 and 86.11 %, respectively. It was shown by other authors that a low level of AMH may cause cycle cancelation, lower mean implantation rate, or lower chances of ongoing pregnancy [24, 25]. On the other hand, even with undetectable AMH, pregnancy after transfer is possible. Therefore, as the author suggested, extremely low levels of AMH should not be the only cause of exclusion of a patient from attempting IVF. Then, usefulness of AMH for forecasting pregnancy indicated in the current study is slightly different from the observation of Reichman et al. . We also tried to check which parameter – AMH or FSH – could be better in the prognosis of whether pregnancy will be achieved, but due to an assumed level for α at 0.05 and the obtained p value (0.597) we cannot prove a better predictive value of any one parameter. Hussain et al. also assessed the relationships between AMH and FSH in women undergoing IVF/ICSI, and they observed no difference in cycle cancellation, clinical pregnancy, and live birth/ongoing pregnancy between groups of women with different hormone levels .
Regarding LH level, we did not show significant differences in LH concentrations between the groups of women in whom pregnancy was or was not achieved. Our observation is supported by results of Ramachandran et al. They also showed no influence of LH level on IVF outcomes .
Recently it has been suggested that the genotype of receptor for FSH should be take into account for the pharmacological approach to infertility treatment with FSH because the response to FSH stimulation seems to be associated with genetic background . However, more clinical data are necessary to warrant routine use of the FSHR isoforms as a diagnostic test.
The kind of gonadotropin applied to stimulate ovaries impacts the dynamics of embryo development: in women stimulated with hFSH subsequent development stages were usually observed earlier than in women treated with rFSH.
There was no statistically significant difference in pregnancy rate between hFSH-stimulated women and rFSH-stimulated women.
The mean estradiol level was higher in women who achieved pregnancy than in women in whom pregnancy was not achieved, and its values negatively correlates with some time stages of embryo development.
AMH and FSH had greater impact on achieving pregnancy than other hormones, and the values of AMH and FSH in predicting pregnancy are similar.
We thank Mr Bartosz Lowczak for statistical analysis.
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
- European Society for Human Reproduction and Embryology: ART Factsheet. http://www.eshre.eu/ESHRE/English/Guidelines-Legal/ART-fact-sheet/page.aspx/1061 (2013). Accessed 12 Sep 2013.
- Wołczyński S, Radwan M (red). Algorytmy diagnostyczno-lecznicze w zastosowaniu do niepłodności. Polskie Towarzystwo Medycyny Rozrodu (2014). http://rozrodczosc.pl/ptmr/uploads/docs/Algorytmy_w_nieplodnosci_06_05_2011.pdf. Accessed 6 May 2011.
- Broekmans FJ, Kwee J, Hendriks DJ, Mol BW, Lambalk CB. A systematic review of tests predicting ovarian reserve and IVF outcome. Hum Reprod Update. 2006;12:685–718.View ArticlePubMedGoogle Scholar
- Zwart-van Rijkom JE, Broekmans FJ, Leufkens HG. From HMG through purified urinary FSH preparations to recombinant FSH: a substitution study. Hum Reprod. 2002;17:857–65.View ArticlePubMedGoogle Scholar
- Daya S. Updated meta-analysis of recombinant follicle-stimulating hormone (FSH) versus urinary FSH for ovarian stimulation in assisted reproduction. Fertil Steril. 2002;77:711–4.View ArticlePubMedGoogle Scholar
- Filicori M, Cognigni GE, Pocognoli P, Tabarelli C, Ferlini F, Perri T, et al. Comparison of controlled ovarian stimulation with human menopausal gonadotropin or recombinant follicle-stimulating hormone. Fertil Steril. 2003;80:390–7.View ArticlePubMedGoogle Scholar
- Kilani Z, Dakkak A, Ghunaim S, Cognigni GE, Tabarelli C, Parmegiani L, et al. A prospective, randomized, controlled trial comparing highly puri®ed hMG with recombinant FSH in women undergoing ICSI: ovarian response and clinical outcomes. Hum Reprod. 2003;18:1194–99.View ArticlePubMedGoogle Scholar
- Blazar AS. Serum estradiol positively predicts outcomes in patients undergoing in vitro fertilization. Fertil Steril. 2004;81:1707–9.View ArticlePubMedGoogle Scholar
- Ng EHY, Yeung WSB, Lau EYL, So WWK, Ho PC. High serum estradiol concentrations in fresh IVF cycles do not impair implantation and pregnancy rates in subsequent frozen- thawed embryo transfer cycles. Hum Reprod. 2000;15:250–5.View ArticleGoogle Scholar
- Wong C, Loewke KE, Bossert NL, Behr B, De Jonge CJ, Baer TM, et al. Non-invasive imaging of human embryos before embryonic genome activation predicts development to the blastocyst stage. Nat Biotechnol. 2010;28:1115–21.View ArticlePubMedGoogle Scholar
- Arav A, Aroyo A, Yavin S, Roth Z. Prediction of embryonic developmental competence by time-lapse observation and ‘shortest-half’ analysis. Reprod Biomed Online. 2008;17:669–75.View ArticlePubMedGoogle Scholar
- Azzarello A, Hoest T, Mikkelsen AL. The impact of pronuclei morphology and dynamicity on live birth outcome after time-lapse culture. Hum Reprod. 2012;27:2649–57.View ArticlePubMedGoogle Scholar
- Conaghan J, Chen AA, Willman SP, Ivani K, Chenette PE, Boostanfar R, et al. Improving embryo selection using a computer-automated time-lapse image analysis test plus day 3 morphology: results from a prospective multicenter trial. Fertil Steril. 2013;100:412–9.View ArticlePubMedGoogle Scholar
- Sullivan MW, Stewart-Akers A, Krasnow JS, Berga SL, Zeleznik AJ. Ovarian responses in women to recombinant follicle-stimulating hormone and luteinizing hormone (LH): a role for LH in the final stages of follicular maturation. J Clin Endocrinol Metab. 1999;84:228–32.PubMedGoogle Scholar
- Alpha Scientists in Reproductive Medicine and ESHRE Special Interest Group of Embryology. The Istanbul consensus workshop on embryo assessment: proceedings of an expert meeting. Hum Reprod. 2011;26:1270–83.View ArticleGoogle Scholar
- Kirkegaard K, Kesmodel US, Hindkjær JJ, Ingerslev HJ. Time-lapse parameters as predictors of blastocyst development and pregnancy outcome in embryos from good prognosis patients: a prospective cohort study. Hum Reprod. 2013;28:2643–51.View ArticlePubMedGoogle Scholar
- Polanski LT, Barbosa MA, Martins WP, Baumgarten MN, Campbell B, Brosens J, et al. Interventions to improve reproductive outcomes in women with elevated natural killer cells undergoing assisted reproduction techniques: a systematic review of literature. Hum Reprod. 2014;29:65–75.View ArticlePubMedGoogle Scholar
- Muñoz M, Cruz M, Humaidan P, Garrido N, Pérez-Cano I, Meseguer M. Dose of recombinant FSH and oestradiol concentration on day of HCG affect embryo development kinetics. Reprod Biomed Online. 2012;25:382–9.View ArticlePubMedGoogle Scholar
- Papageorgiou T, Guibert J, Goffinet F, Patrat C, Fulla Y, Janssens Y, et al. Percentile curves of serum oestradiol levels during controlled ovarian stimulation in 905 cycles stimulated with recombinant FSH show that high oestradiol is not detrimental to IVF outcome. Hum Reprod. 2002;17:2846–50.View ArticlePubMedGoogle Scholar
- Pena JE, Chang PL, Chan LK, Zeitoun K, Thornton MH, Sauer MV. Supraphysiological oestradiol levels do not affect oocyte and embryo quality in oocyte donation cycles. Hum Reprod. 2002;17:83–7.View ArticlePubMedGoogle Scholar
- Manno M, Cervi M, Zadro D, Fuggetta G, Adamo V, Tomei F. Different ART outcomes at increasing peak estradiol levels with long and antagonist protocols: retrospective insights from ten years’ experience. J Assist Reprod Genet. 2011;28:693–8.PubMed CentralView ArticlePubMedGoogle Scholar
- Kara M, Kutlu T, Sofuoglu K, Devranoglu B, Cetinkaya T. Association between serum estradiol level on the hCG administration day and IVF-ICSI outcome. Iran J Reprod Med. 2012;10:53–8.PubMed CentralPubMedGoogle Scholar
- Huang JY, Rosenwaks Z. In vitro fertilisation treatment and factors affecting success. Best Pract Res Clin Obstet Gynaecol. 2012;26:777–88.View ArticlePubMedGoogle Scholar
- Lehmann P, Vélez MP, Saumet J, Lapensée L, Jamal W, Bissonnette F, et al. Anti-Müllerian hormone (AMH): a reliable biomarker of oocyte quality in IVF. J Assist Reprod Genet. 2014;27:493–8.View ArticleGoogle Scholar
- Reichman DE, Goldschlag D, Rosenwaks Z. Value of antimüllerian hormone as a prognostic indicator of in vitro fertilization outcome. Fertil Steril. 2014;101:1012–18.View ArticlePubMedGoogle Scholar
- Hussain M, Cahill D, Akande V, Gordon U. Discrepancies between anti-Müllerian hormone and follicle-stimulating hormone in assisted reproduction. Obstet Gynecol Int. 2013;383278. doi: 10.1155/2013/383278.Google Scholar
- Ramachandran A, Jamdade K, Kumar P, Adiga SK, Bhat RG, Ferrao SR. Is there a need for luteinizing hormone (LH) estimation in patients undergoing ovarian stimulation with gonadotropin-releasing hormone (GnRH) antagonists and recombinantf-stimulating hormone (rFSH)? J Clin Diagn Res. 2014;8:90–2.PubMed CentralPubMedGoogle Scholar
- Perez Mayorga M, Gromoll J, Behre HM, Gassner C, Nieschlag E, Simoni M. Ovarian response to follicle-stimulating hormone (FSH) stimulation depends on the FSH receptor genotype. J Clin Endocrinol Metab. 2000;85:3365–9.PubMedGoogle Scholar