Comparison of anti-mullerian hormone and antral follicle count for assessment of ovarian reserve

Transcription

1 Original Article Comparison of anti-mullerian hormone and antral follicle count for assessment of ovarian reserve Sonal Panchal, Chaitanya Nagori Dr. Nagori s Institute for Infertility and IVF, Ellisbridge, Ahmedabad, Gujarat, India Address for correspondence: Dr. Sonal Panchal, Dr. Nagori s Institute for Infertility and IVF, 2 nd Floor Kedar, Opp. Petrol Pump, Nr. Parimal Garden, Ellisbridge, Ahmedabad , Gujarat, India. E mail: sonalyogesh@yahoo. com Received: Review completed: Accepted: ABSTRACT AIM: This study aims to compare the efficacy of AFC and AMH, as markers for ovarian reserve. MATERIALS AND METHODS: 75 patients with PCO (polycystic ) undergoing IVF were randomized with 75 non PCO patients. On day 3, volume of ovary was acquired, ovarian volume was defined by VOCAL, and Sono AVC was used to count the number of antral follicles. Sum total of antral follicles in both was taken as total antral follicle count (AFC). AMH was measured on the same day. Long agonist protocol with recombinant FSH (rfsh) was used for IVF stimulation till at least two follicles of 18 mm were seen. hcg 10,000 iu was given and ovum pick up was done after h. Primary end point was number of follicles >12 mm seen on day of hcg. Final end point was number of ova retrieved on ovum pick up. Correlation of AFC and AMH was checked for both end points and with each other. RESULTS: Correlation of AFC and follicles >12 mm on day of hcg in PCO group is 0.56 and non PCO group is 0.63, 1 and for AMH and follicles >12 mm on day of hcg in PCO group is 0.42 and non PCO group is Correlation of AFC with number of ova retrieved on OPU in PCO group is 0.44 and for non PCO group is The value for AMH is 0.39 in PCO and 0.43 for non PCO group. Comparing correlation of AFC and AMH for primary end point in PCO group has z value 1.11(onetailed , twotailed 0.267) and in non PCO group comparison shows a z value of 1.39 (one tailed , two tailed ). Therefore in both groups, AFC and AMH correlates with total number of follicles >12 mm on day of hcg, but both AFC and AMH have independent. Comparing correlation of AFC and AMH with number of ova retrieved on OPU, in non PCO group has z value of 0.54(one tailed , two tailed ). In PCO group, this comparison shows, z value of 0.36(one tailed , two tailed ). CONCLUSION: AFC and AMH correlate with each other but have independent for estimating follicles >12 mm on day of hcg. For ova retrieved on OPU, in PCO group, AMH and AFC have similar. In non PCO, AMH may increase accuracy of estimation of number of ova retrieved on OPU over AFC, but difference is not significant. AFC only may suffice for estimation of number of ova retrieved on OPU for all practical purposes. KEY WORDS: Anti-mullerian hormone, antral follicle count, ovarian reserve Access this article online Quick Response Code: Website: DOI: / INTRODUCTION Ovarian reserve relates to secondary, pre antral and antral ovarian follicle pool. Ovarian reserve was conventionally assessed by serum FSH (S. FSH). AMH (Anti mullerian hormone) has lately been found to be a very reliable and more accurate than S.FSH, for assessment of ovarian reserve. Anti mullerian hormone is produced by pre antral follicles. The highest level of AMH expression is present in granulosa cells of secondary, pre antral and small antral follicles up to 6 mm in diameter. [1] Therefore, level of AMH may represent the population of these follicles. It has been shown that the number of antral follicles in the is proportionally related to the size of primordial follicle stock from which they were recruited. [2] Therefore, the antral follicle count (AFC) is believed to represent the quantitative aspect of ovarian aging. [3] With advancement in ultrasound technology it is now possible to count antral follicles. This study was therefore designed to find out if there was any correlation between AFC and AMH and to analyze which of the two is more reliable. Comparison of AMH and AFC has been done in two groups: 274 Journal of Human Reproductive Sciences / Volume 5 / Issue 3 / Sep - Dec 2012

2 Polycystic PCOs and non PCOs (normal ). S. FSH did show high values for low reserve but its values are not found to be different for normal and hyper responding PCO groups. These two groups were selected for the study, therefore to assess whether values of these new markers were different in these groups. OBJECTIVE The study aims to find out whether AFC or AMH is more reliable parameter for assessment of ovarian reserve based on computation of correlation between AFC and AMH. MATERIALS AND METHODS Seventy five patients with PCOs, with BMI (Body mass index) between 22 and 26 (mean 24.5) undergoing IVF (In vitro fertilization) were randomized with 75 patients with same BMI (mean 23.2) but non PCO (normal responding) undergoing IVF (In vitro fertilization) for the study. In both groups, patients recruited were of years of age. Mean age in PCO group was 28.7 years and mean age in non PCO group was 29.2 years. Patients in PCO group were confirmed to have PCO by Rotterdam criteria, whereas patients in non PCO group had normal responding (ovarian diameter between 2 and 3 cm, ovarian volume between 3 and 6.6 cc, AFC 5-12), with regular menstrual cycles and no biochemical or clinical signs of hyperandrogenemia. Patients with other endocrinal derangements like thyroid or prolactin were excluded from the study. Baseline scan was done for all these women on day 3 of the menstrual cycle using 3D US (ultrasound). After acquiring a 3D volume of the ovary, VOCAL (Volume calculation software) was used to define the ovarian volume. Angle of rotation for VOCAL was 15. Then Sono AVC (Automated volume calculation) was used to count the number of antral follicles in each ovary. Antral follicles are the follicles between 2 and 9 mm on day 3 of menstrual cycle. Sono AVC is based on inversion mode and color codes each follicle along with its diameter in all three orthogonal planes (x, y, z), mean diameter and follicular volume. Post processing of Sono AVC was done in all patients to count the number of antral follicles more accurately. Post processing includes manually adding the follicles that were missed on automatic counting, removing the areas that were erroneously counted as antral follicles on automatic counting, cutting the area that was over measured on automatic calculation and merging the areas that were part of the same follicle but was counted as separate follicles. The sum total of antral follicles in both was taken as total AFC. All the scans were done by a single operator on Voluson E8 (GE Healthcare), with 5-9 MHz transvaginal volume probe. Anti mullerian hormone level in patient s blood was measured in all patients on the same day. Measurement of serum AMH levels was performed using AMH/MIS enzyme linked immunosorbent assay kit (Beckman Coulter Immunotech Com., Fullerton, CA). The lowest detection rate limit and intra assay and inter assay coefficients of variation were 1 pmol/l or 0.14 ng/ml, 12.3%, and 14.2%, respectively. The unit of measurement for AMH used was ng/ml. All patients were down regulated by long agonist protocol (Leuprolide acetate 1 mg daily from day 21 of previous cycle and 0.5 mg daily from the day of menstruation till the day of hcg). Recombinant follicle stimulating hormone (rfsh), (Inj. Gonal F, Merck Serono) 150 iu was used for ovulation induction in all patients starting from day 2 of the period. Transvaginal ultrasound scan (TVS) was done on day 6 of stimulation. If at least 2-3 follicles of 10 mm or more were seen, the same dose was continued. If not the dose of rfsh was increased. Stimulation was continued till there were at least 2-3 follicles of 18 mm. These follicles were assessed by color Doppler for their physiological maturity. Follicles when showed flow covering at least 3/4 th of its circumference with resistance index (RI) of <0.48 and peak systolic velocity (PSV) of >10 cm/s, hcg 10,000 was given for ovulation trigger. Ovum pick up was done h after hcg. For both groups the primary end point was number of follicles larger than 12 mm seen on the day of hcg. The final end point was the number of ova retrieved on ovum pick up. In both groups, correlation of AFC and AMH was checked with both these endpoints and correlation of AFC was also checked with AMH. RESULTS In PCO group, the AFC ranged from 25 to 59 with an average of 34 and in non PCO group, the AFC ranged from 11 to 23 follicles with an average of 14. AMH levels in PCO group ranged from 3.39 to 13.4 ng/ml with an average of 6.62 ng/ml and in non PCO group, AMH levels ranged from 0.9 to 2.88 ng/ml with an average of 1.84 ng/ml. Table 1 presents the correlation between AFC and AMH for both groups. Correlation between AFC and AMH in PCO group is 0.46, i.e., higher than the tabulated value (Pearson s correlation), (0.283) at 0.01 Table 1: Correlation of antral follicle count and anti-mullerian hormone Normal Polycystic AFC AFC AMH AMH AFC = Antral follicle count, AMH = Anti-mullerian hormone Journal of Human Reproductive Sciences / Volume 5 / Issue 3 / Sep - Dec

3 level. Similarly for non PCO group, correlation between AFC and AMH is 0.49, which is also higher than the tabulated value (Pearson s correlation), (0.283) at 0.01 level. Table 2 presents the correlation between AFC and follicles of >12 mm diameter on the day of hcg and AMH and follicles >12 mm on the day of hcg for PCO and non PCO groups. Correlation between AFC and follicles of >12 mm on the day of hcg in PCO group is 0.56, that is significant at level (99.9% as per Pearson s correlation coefficient). Similarly for non PCO group, correlation between AFC and follicles larger than 12 mm on the day of hcg is 0.63, which is also significant at 0.01 level. It means AFC is positively correlated with follicles >12 mm on the day of hcg for both groups. Table 2 also presents the correlation between AMH and follicles larger than 12 mm on the day of hcg for both the types of. Correlation between AMH and follicles larger than 12 mm on the day of hcg in PCO group is 0.42, which is significant at 0.01 level. Similarly for non PCO group, correlation between AMH and follicles larger than 12 mm on the day of hcg in non PCO group is 0.47, which is also significant at 0.01 level. This infers that AMH is positively correlated with number of follicles >12 mm on the day of hcg for both. Table 3 shows that comparing the correlation of AFC with number of follicles >12 mm diameter on the day of hcg and AMH and follicles larger than >12 mm on the day of hcg in non PCO group shows a z value of This value is higher than the tabulated one tailed value (0.0823), and also at two tailed (0.1645) value. This indicates that AFC and AMH both correlate with the number of follicles >12 mm on the day of hcg in non PCO group but have different results. It means both the tests need to be performed for estimation of number of follicles >12 mm in diameter available on the day of hcg for non PCO group. In PCO group, comparing the correlation of AFC with follicles >12 mm diameter on the day of hcg and AMH with follicles >12 mm diameter on the day of hcg, a z value of 1.11 is achieved. This value is higher than the tabulated one tailed value and two tailed value This indicates that for PCO group though AFC and AMH can be correlated with the total number of follicles larger than 12 mm diameter on the day of hcg, both have independent for estimation of number of follicles that would grow to a diameter larger than 12 mm on the day of hcg. This means both the tests need to be performed for the estimation of number of follicles >12 mm in diameter available on the day of hcg for PCO group. Results in Table 3 clearly confirm that both AFC and AMH despite having positive and high correlation are yet critical as the level of difference between values of correlation is significant as indicated by z value for both the groups. Table 4 presents the correlation between AFC and AMH and ova retrieved on OPU for both groups. Correlation between AFC and ova retrieved on OPU in PCO group is 0.44, which is significant at 0.01 level, whereas for AMH and ova retrieved on OPU, correlation value is Similarly for non PCO group, correlation between AFC and ova retrieved on OPU in non PCO ovary is 0.50, and correlation for AMH and ova retrieved on OPU the value is This is significant at 0.01 level for PCO also. One can therefore infer that the AFC can be used for estimation of the total number of ova likely to be retrieved on OPU in PCO and non PCO groups. It can also be inferred that AMH is positively correlated with number of ova retrieved on OPU in both groups. One can therefore infer Table 2: Correlation between AFC and follicles >12 mm on the day of hcg and AMH and follicles >12 mm on the day of hcg for PCO and non PCO groups, N=75 Normal AFC >12 mm Polycystic AFC >12 mm AFC AFC >12 mm >12 mm AMH >12 mm AMH >12 mm AMH AMH >12 mm >12 mm AFC = Antral follicle count, AMH = Anti-mullerian hormone Table 3: Z score for difference in correlation of AFC and AMH with number of follicles larger than >12 mm on the day of hcg for non PCO and PCO groups Non PCO >12 mm Test of Polycystic Test of Z value >12 mm Z value AMH * *a AFC Difference *Values for level of : (one tailed test) and (two tailed test) *a Values for level of : (one tailed test) and (two tailed test) Table 4: Correlation of AFC and AMH with ova retrieved on ovum pick up for PCO and non PCO groups, N=75 Normal AFC OPU Polycystic AFC OPU AFC AFC OPU OPU AMH OPU AMH OPU AMH AMH OPU OPU AFC = Antral follicle count, AMH = Anti-mullerian hormone, OPU = Ovum pick up 276 Journal of Human Reproductive Sciences / Volume 5 / Issue 3 / Sep - Dec 2012

4 that AMH also can be used for the estimation of number of ova retrieved on OPU. Comparing the correlation of AFC with number of ova retrieved on OPU and AMH and number of ova retrieved on OPU in non-pco group [Table 5] shows a z value of This value is higher than the tabulated one tailed value , but lower than two tailed value In PCO group, comparing the correlation of AFC to number of ova retrieved on OPU and AMH to number of ova retrieved on OPU, a z value of 0.36 is achieved. This value is almost same as the tabulated one tailed value and lower than two tailed value This indicates that for non PCO group, AFC and AMH do have minimally different but independent for estimating the ova retrieved on OPU, but for PCO group, AFC and AMH do not have independent for assessment of ovarian reserve and AFC only may suffice to estimate the number of ova retrieved on OPU. This infers that in PCO group, AFC only suffices for the estimation of ovarian reserve. For non PCO groups as the z score value is higher than one tailed value, but lower than two tailed value; this infers that though AMH may increase the accuracy of estimation of number of ova retrieved on OPU over AFC, this difference is not highly significant. DISCUSSION Success of any technique for artificial reproduction depends on selection of a correct protocol for ovarian stimulation. This can be decided only by proper assessment of ovarian reserve before commencing ovarian stimulation. Various tests have been tried at different times for assessment of ovarian reserve like S.FSH, AMH, AFC, Inhibin B, etc., Several studies have been done and published at different times assessing these different markers for assessment of ovarian reserve. AMH and AFC have been shown to be the most sensitive ones for assessment of ovarian reserve in these studies. AFC and ovarian volume provide direct measurements of ovarian reserve, while AMH, inhibin B and oestradiol are released from growing follicles and so their levels reflect the size of developing follicle cohort. [4] Table 5: Z score for difference in correlation of AFC and AMH with OPU for non-pco and PCO groups Non PCO OPU Test of Polycystic Test of Z value OPU Z value AMH * *a AFC Difference *Values for level of : (one tailed test) and (two tailed test) *a Values for level of : (one tailed test) and (two tailed test), OPU = Ovum pick up AFC can be reliably and accurately calculated by 3D US. 3D US with SonoAVC was used for the purpose of more accuracy and reproducibility. Inversion mode is most convenient method for counting antral follicles when they are multiple. [5] Number of follicles >12 mm on day of oocyte retrieval correlated significantly with AFC counted by 3D US rather than 2D US. Studies have been done to compare the accuracy and reliability of AFC and AMH for prediction of low response or for prediction of ovarian hyperstimulation. Serum AMH levels have been shown to strongly correlate with the number of antral follicles [6,7] and have appeared to be cycle independent. [8,9] The ROC curves do not suggest a clearly better predictive ability for AMH than for AFC, and the difference was not statistically significant. This implies that the best poor response predictor to date, AFC, [10] has obtained company from a test that may have some crucial advantages. This implies that AFC can predict the response of ovary to stimulation and can be used for selection of individual cycle for ovulation induction, but adding AMH also allows to assess the quality of the resultant ova and embryo. Broer et al. have compared AFC and AMH and have shown that sensitivity and specificity for AMH were 82 and 76%, respectively, and 82 and 80%, respectively, for AFC. Comparison of the summary estimates and ROC curves for AMH and AFC showed no statistical difference in this study. Both AMH and AFC are accurate predictors of excessive response to ovarian hyperstimulation. [11] Another study by Aflatoonian et al. has shown that small AFC and anti müllerian hormone have similar predictive accuracy for high ovarian response with area under curve of and 0.922, respectively. The sensitivity and specificity for prediction of high ovarian response were 89% and 92% for small AFC and 93% and 78% for anti müllerian hormone at the cutoff values of 16 and 34.5 pmol/l, respectively. [12] Still one more study compares AMH and AFC for assessment of ovarian response. Number of retrieved oocytes was statistically significant and correlated with D3 AMH, AFC, DhCG AMH, DhCG inhibin B, FSH, and age (r = 0.885, 0.874, 0.742, 0.732, 0.521, 0.385, respectively). [13] In a metaanalysis by Broer et al., it has been shown that AMH has at least the same level of accuracy and clinical value for the prediction of poor response and non pregnancy as AFC. [14] AMH may provide a more accurate assessment of the follicle pool in young hypergonadotropic patients, especially in the Journal of Human Reproductive Sciences / Volume 5 / Issue 3 / Sep - Dec

5 clinically challenging subgroups of patients with elevated FSH and regular menses (i.e., IOF: incipient ovarian failure) and in hypergonadotropic women with cycle disturbances not fulfilling the POF diagnostic criteria (i.e., TOF: transistional ovarian failure). [15] AFC and AMH are the most significant predictors of poor response to ovarian stimulation. Both alone or in combination have a similar predictive power. Combination of both tests do not significantly increase the predictive power. [16] Comparing AFC and AMH for assessment of ovarian reserve in poor responding has been done in several studies. But assessment of both as predictor for normal response or for hyperstimulation has been done only in a few. Therefore, the two groups PCO and non PCO (normal responding ) were selected for this study. The purpose was to decrease the number of tests that a patient undergoing ART, has to take, to decrease the stress as well as financial burden, without compromising with the information gained. Ultrasound is as such done before starting the ovarian stimulation for ART and AFC is a part of it. The study has shown that both AFC and AMH are equally accurate for assessment of ovarian reserve, in PCO as well as in patients with normal. AFC therefore may be thought of as a single test for assessment of ovarian reserve. CONCLUSION The results suggest that AFC and AMH both correlate with each other but have independent for estimating follicles larger than 12 mm on the day of hcg. But for estimation of ova expected on OPU, the difference of correlation coefficient between AFC and AMH is minimally significant in non PCO and is not significant in PCO group. As the ultimate end result is ova retrieved on OPU, it can be concluded that for both groups, AFC alone may suffice as a test for estimation of ovarian reserve. This is so because the z values for both non PCO as well as PCO group is equal to or below the tabulated one tailed value. ENDNOTE 1. Significance of correlation at 0.01 level is required to be higher than Significance of correlation at 0.01 level is required to be higher than REFERENCES 1. Weenen C. Laven JS, Von Bergh AR, Cranfield M, Groome NP, Visser JA, et al. Antimullerian hormone expression pattern in human oovary: Potential implications for initial and cyclic follicle recruitment. Mol Hum Reprod 2004;10: Gougeon A, Echochard R, Thalabard JC. Age related changes of the population of human ovarian follicles: Increase in the disappearance rate of nongrowing and early growing follicles in aging women. Biol Reprod 1994;50: BroekmansFJ, Kwee J, Hendriks DJ, Mol BW, Lambalk CB. A systematic review of tests of predicting ovarian reserve and IVF outcome. Hum Reprod Update 2006;12: Hendricks DJ, Kwee J, Mol BW, te Velde ER, Broekmans FJ. Ultrasonography as a tool for prediction of outcome in IVF patients: A comparative meta analysis of ovarian volume and antral follicle count. Fertil Steril 2007;87: Ng EH, Chan CC, Yeung WS, Ho PC. Effect of age on ovarian stromal flow measured by three dimensional ultrasound with power Doppler in Chinese women with proven fertility. Hum Reprod 2004;19: Gruijters MJ, Visser JA, Durlinger AL, Themmen AP. Antimullerian hormone and its role in ovarian function. Mol Cell Endocrinol 2003;211: Van Rooij IA, Broekmans FJ, te Velde ER, Fauser BC, Bancsi LF, Jong FH, et al. Serum antimullerian hormone levels: a novel measure of ovarian reserve. Hum Reprod 2002;17: Hehenkamp WJ, Looman CW, Themmen AP, de Jong FH, te Velde ER, Broekmans FJ. Antimullerian hormone levels in the spontaneous menstrual cycle do not show sunstantial fluctuation. J Clin Endocrinol Metab 2006;91: La MArca A, Stabile G, Artenisio AC, Volpe A. Serum antimullerian hormone throughout the human menstrual cycle. Hum Reprod 2006;21: Hendriks DJ, Mol BW, Bancsi LF, te Velde ER, Broekmans FJ. Antral follicle count in the prediction of poor ovarian response and pregnancy after in vitro fertilization: A meta analysis and comparision with basal follicle stimulating hormone level. Fertil Steril 2005;83: van Disseldorp J, Lambalk CB, Kwee J, Looman CW, Eijkemans MJ, Fauser BC, et al. Comparison of inter and intra cycle variability of anti Mullerian hormone and antral follicle counts. Hum Reprod 2010;25: Aflatoonian A, Oskouian H, Ahmadi S, Oskouian L. Prediction of high ovarian response to controlled ovarian hyperstimulation: Anti mullerian hormone versus small antral follicle count (2 6mm). J Assist Reprod Genet 2009;26: Wu CH, Chen YC, Wu HH, Yang JG, Chang YJ, Tsai HD. Serum anti Mullerian hormone predicts ovarian response and cycle outcome in IVF patients. J Assist Reprod Genet 2009;26: Broer SL, Mol BW, Hendriks D, Broekmans FJ. The role of antimullerain hormone in prediction of outcome after IVF: comparison with the antral follicle count. Fertil Steril 2009;91: Knauff EA, Eijkemans MJ, Lambalk CB, ten Kate Booij MJ, Hoek A, Beerendonk CC, et al. Antimullerian hormone, Inhibin B and antral follicle count in young women with ovarian failure. J Clin Endocrinol Metab 2009;94: Jayaprakasan K, Campbell B, Hopkisson J, Johnson I, Rainne Fenning N. A prospective, comparative analysis of anti mullerian hormone, inhibin B, and three dimensional ultrasound determinants of ovarian reserve in the prediction of poor response to controlled ovarian stimulation. Fertil Steril 2010;93:30 9. How to cite this article: Panchal S, Nagori C. Comparison of anti-mullerian hormone and antral follicle count for assessment of ovarian reserve. J Hum Reprod Sci 2012;5: Source of Support: Nil, Conflict of Interest: None declared. 278 Journal of Human Reproductive Sciences / Volume 5 / Issue 3 / Sep - Dec 2012