|Year : 2020 | Volume
| Issue : 4 | Page : 228-232
Potential application of anti-müllerian hormone in polycystic ovary syndrome according to chinese classification criteria: A retrospective analysis
Ling-Li Tang1, Ling-Shan Zhang2, Xiao-Yong Zhu3, Ying-Li Shi2
1 Department of Gynecology, Changsha Maternal and Child Health Hospital, Changsha 410007; Department of Gynecology, Obstetrics and Gynecology Hospital of Fudan University, Shanghai 200090, China
2 Department of Gynecology, Obstetrics and Gynecology Hospital of Fudan University, Shanghai 200090; Shanghai Key Laboratory of Female Reproductive Endocrine Related Diseases, Shanghai 200011, China
3 Department of Gynecology, Obstetrics and Gynecology Hospital of Fudan University, Shanghai 200090; Laboratory for Reproductive Immunology, Hospital and Institute of Obstetrics and Gynecology, Shanghai Medical School, Fudan University, Shanghai 200011, China
|Date of Submission||08-Feb-2020|
|Date of Decision||27-May-2020|
|Date of Acceptance||26-Jul-2020|
|Date of Web Publication||31-Dec-2020|
Department of Gynecology, Obstetrics and Gynecology Hospital of Fudan University, No. 218, Shenyang Road, Shanghai 200090
Source of Support: None, Conflict of Interest: None
Objective: Anti-Müllerian hormone (AMH) expression is elevated in patients with polycystic ovary syndrome (PCOS), however, its clinical significance is not clear. Owing to the strong correlation between AMH and polycystic ovarian morphology (PCOM), some studies believe that AMH alone can be used to diagnose PCOS. The aim of this study was to explore whether AMH can be used to diagnose PCOS and to differentiate the various PCOS subtypes.
Methods: This was a retrospective study of 503 patients with PCOS. Patients were divided into eight subtypes based on the presence/absence of hyperandrogenemia (HA), insulin resistance (IR), or obesity (OB). The expression characteristics of AMH in each subtype were analyzed. Due to the small number of patients with subtypes 7 and 8, only patients with subtypes 1–6 were included in the analysis.
Results: AMH showed a good positive correlation with PCOM (P = 0.000) and negative correlations with OB (P = 0.000) and IR (P = 0.003). The free testosterone index showed no correlation with AMH (P = 0.803). The percentages of patients with each subtype (excluding subtypes 7–8) and their respective AMH levels were as follows: Type 1 (HA + NIR + OB) 4.77% and 9.12 ng/mL; Type 2 (HA + IR + NOB) 20.68% and 10.34 ng/mL; Type 3 (HA + NIR + NOB) 23.66% and 9.47 ng/mL; Type 4 (HA + IR + OB) 30.82% and 8.32 ng/mL; Type 5 (NHA + NIR + NOB) 11.73% and 10.0 ng/mL; and Type 6 (NHA + IR + NOB) 6.16% and 9.76 ng/mL. The diagnostic rates of AMH (>8.09 ng/mL) and ultrasound for PCOM were 60.10% and 85.60%, respectively, suggesting that AMH did not completely predict PCOM.
Conclusions: High AMH levels can be used to evaluate the incidence trend of PCOS. However, due to clinical heterogeneity, accurately evaluating the severity of PCOS and identifying the subtype of PCOS in Chinese patients are difficult. Individualized treatment should be administered based on accurate clinical subtypes and other clinical characteristics.
Keywords: Anti-Müllerian Hormone; Application Value; Polycystic Ovary Syndrome; Subtype
|How to cite this article:|
Tang LL, Zhang LS, Zhu XY, Shi YL. Potential application of anti-müllerian hormone in polycystic ovary syndrome according to chinese classification criteria: A retrospective analysis. Reprod Dev Med 2020;4:228-32
|How to cite this URL:|
Tang LL, Zhang LS, Zhu XY, Shi YL. Potential application of anti-müllerian hormone in polycystic ovary syndrome according to chinese classification criteria: A retrospective analysis. Reprod Dev Med [serial online] 2020 [cited 2021 Apr 14];4:228-32. Available from: https://www.repdevmed.org/text.asp?2020/4/4/228/305927
| Introduction|| |
Polycystic ovary syndrome (PCOS) is a reproductive endocrine disease that begins before puberty. The clinical manifestations of PCOS are menstrual disorders caused by ovulation disorder, polycystic ovarian morphology (PCOM), hirsutism, acne, obesity (OB), and insulin resistance (IR). Owing to complex pathogenetic mechanisms, PCOS is characterized by considerable clinical heterogeneity and inter-individual variability with respect to the response to various treatments. Clinically, PCOS is classified mainly based on the presence or absence of hyperandrogenism (HA), PCOM, OB, and IR.
Patients with PCOS exhibit higher expression of anti-Müllerian hormone (AMH); however, its clinical significance is not well characterized., AMH is mainly expressed in the granulosa cells of primary follicles, secondary follicles, and small sinusoidal follicles. AMH shows a good correlation with PCOM, and its levels are not affected by the menstrual cycle; therefore, some scholars believe that AMH can replace PCOM as the criterion for diagnosis without considering HA and anovulation. The expression of AMH is also affected by other clinical features of PCOS. Studies have demonstrated a positive correlation between AMH and testosterone (T)., However, the correlation between AMH and the free testosterone index (FAI) is not clear, probably because sex hormone-binding globulin (SHBG) is affected by many factors., Studies have also demonstrated a negative correlation between AMH and OB,, however, the relationship between AMH and IR has not been clarified.,,
The clinical characteristics of PCOS in China and abroad are different owing to racial, regional, and other factors. In this study, we explored the clinical expression of AMH in Chinese patients with various PCOS subtypes to explore the value of AMH for the clinical diagnosis.
| Methods|| |
From April 2016 to February 2018, 503 patients with PCOS were enrolled from the reproductive endocrinology clinic of the Obstetrics and Gynecology Hospital, Fudan University. The inclusion criteria were: (1) a confirmed diagnosis of PCOS according to the diagnostic criteria of China (2011), which are as follows: chronic anovulation or oligo-ovulation and at least two of the following features: biochemical and clinical manifestations of HA or PCOM as defined by ultrasound and exclusion of other medical conditions that cause irregular menstrual cycles and androgen excess and (2) no use of hormones or medicines regulating glycol-metabolism in the preceding 6 months. According to the presence or absence of PCOM, HA, IR, and OB, the patients were categorized into eight clinical subtypes.
Assessment of anthropometric measures
Data pertaining to the following variables were collected: height, weight, waist circumference, blood pressure, menstrual cycle and duration, the last three menstrual periods, hirsutism, acne, acanthosis nigricans, and family history. The above parameters were measured and evaluated by the same doctor.
Fasting blood samples were collected on the 2nd–5th days of menstruation (any day if the patient was anovulatory). Endocrine indicators, including follicle-stimulating hormone, luteinizing hormone (LH), T, AMH, SHBG, prolactin, thyroid-stimulating hormone, 17-hydroxyprogesterone (17a-OHP), and dehydroepiandrosterone (DHEA-S); metabolic indicators, including total cholesterol, fasting triglycerides, high-density lipoprotein-cholesterol, low-density lipoprotein-cholesterol, apolipoprotein-A, apolipoprotein-B, and lipoprotein-a; and oral glucose tolerance test and insulin IR test were evaluated. All the parameters were tested in the hospital's biochemical laboratory. AMH was tested using an iFlash 3000-A chemiluminescence analyzer (Shenzhen Yahui Long Biotechnology Co., Ltd., China).
In this study, we defined an FAI =3.32 (ng × dL/nmol) as HA, an insulin IR index (HOMA-IR) =2.69 (mmol × mIU/L2) as IR, a body mass index (BMI) =25 kg/m2 as OB, and AMH =8.09 ng/mL as the threshold for the diagnosis of PCOM.
All patients underwent intracavity ultrasound (Philips I-IDI-4000) during the follicular phase; PCOM was confirmed according to the revised 2003 Rotterdam criteria.
Based on the 2018 Chinese PCOS classification standard, the 503 PCOS patients were classified into eight subtypes: subtype 1 (HA + NIR + OB), patients with a combination of HA and OB and the absence of IR; subtype 2 (HA + IR + NOB), patients with a combination of HA and IR and the absence of OB; subtype 3 (HA + NIR + NOB); subtype 4 (HA + IR + OB); subtype 5 (NHA + NIR + NOB), patients with a combination of NIR and NOB and the absence of HA; subtype 6 (NHA + IR + NOB); subtype 7 (NHA + IR + OB); and subtype 8 (NHA + NIR + OB).
Data were analyzed using SPSS Statistics 19 (IBM) software (Amonk, NY, USA). All data were subjected to normality and variance homogeneity testing. Normally distributed variables were analyzed using the t-test. Single-cause logistic regression analysis was used for statistical analysis because of the various factors that may affect the observation index. Statistical significance was tested by two-sided test. P < 0.05 was considered indicative of statistical significance.
| Results|| |
Correlation analysis between anti-Müllerian hormone and the main characteristics of polycystic ovary syndrome
The clinical characteristics of different subtypes were summarized [Table 1], and no significant difference in age was found, but BMI, IR, and HA differed among the corresponding subgroups.
The AMH level in the PCOM group was statistically significantly higher than that in the non-PCOM group (9.68 ± 4.65 ng/mL vs. 6.86 ± 4.06 ng/mL, P = 0.000). The between-group difference was statistically significant even after adjusting for BMI and age. AMH showed a good positive correlation with PCOM (P = 0.000).
The level of AMH in the HA group was higher than that in the NHA group; however, the between-group difference was not statistically significant (9.53 ± 3.93 ng/mL vs. 9.27 ± 4.21 ng/mL, P = 0.656). Even after adjusting for BMI and age, the FAI still showed no correlation with AMH (P = 0.803). Compared with those in the NIR subtypes, AMH levels in the IR subtypes were statistically significantly lower (9.95 ± 4.08 ng/mL vs. 8.43 ± 4.16 ng/mL, P = 0.001). Compared with those in the NOB subtypes, AMH levels in the OB subtypes were also statistically significantly lower (9.85 ± 4.13 ng/mL vs. 8.43 ± 4.09 ng/mL, P = 0.002) [Figure 1].
|Figure 1: Correlation analysis between AMH and the main characteristics of PCOS. HA: Presence of HA; NHA: Absence of HA; IR: Presence of IR; NIR: Absence of IR; OB: Presence of obesity; NOB: Absence of OB; AMH: Anti-Müllerian hormone; PCOS: Polycystic ovary syndrome *P< 0.05|
Click here to view
Clinical classification according to the Chinese standard for polycystic ovary syndrome
The distribution of the 503 PCOS patients grouped according to subtype is shown in [Figure 2]. Because the subtypes 7 and 8 only accounted for 2% of the total study population, the number of cases was too small to draw any credible conclusions; therefore, only 492 patients with subtypes 1–6 were included in the analysis.
|Figure 2: The distribution of PCOS subtypes according to the 2018 Chinese classification standard. HA: Presence of HA; NHA: Absence of HA; IR: Presence of IR; NIR: Absence of IR; OB: Presence of obesity; NOB: Absence of OB; PCOS: Polycystic ovary syndrome.|
Click here to view
The diagnostic rate of AMH (>8.09 ng/mL) for PCOM was 60.10% (300/492), and that of ultrasound for PCOM was 85.60% (421/492), suggesting that AMH may not completely predict PCOM.
Comparison of anti-Müllerian hormone levels in different clinical subtypes of polycystic ovary syndrome
The average serum AMH level in all subtypes was >8.09 ng/mL. The AMH level is known to be closely related to age. To eliminate the effect of age on the results, we analyzed the age distribution in each study group, and the results showed no significant age difference among the subtypes. According to the AMH level, the subgroups from high to low were as follows: subtype 2 (AMH 10.34 ng/mL), subtype 5 (10.00 ng/mL), subtype 6 (9.76 ng/mL), subtype 3 (9.47 ng/mL), subtype 1 (9.12 ng/mL), and subtype 4 (8.32 ng/mL). All patients with subtypes 5 and 6 had PCOM, while the proportion of patients with PCOM characteristics was lowest (71.36%) in subtype 4 (HA + IR + OB) [Figure 3]a, suggesting a positive correlation of AMH with PCOM. Interestingly, although not all patients with subtype 2 (HA + IR + NOB) had PCOM characteristics (90.40%), their AMH levels were the highest, suggesting that some other factors may also affect the expression of AMH. Furthermore, after subdividing the patients in subtypes 1–4, we found that the AMH levels in patients with PCOM were higher than those in patients without PCOM [Figure 3]b. These findings also show a positive correlation between AMH and PCOM.
|Figure 3: Comparison of AMH in different subtypes of PCOS. (a) AMH levels and PCOM proportions in different subtypes under the Chinese classification. (b) AMH levels in subtypes 1–6 with or without PCOM. PCOM%: Proportion of PCOM; Non-PCOM: Absence of PCOM; PCOM: Polycystic ovarian morphology; AMH: Anti-Müllerian hormone; PCOS: Polycystic ovary syndrome|
Click here to view
| Discussion|| |
In this study, we explored the expression profile of AMH in Chinese patients with PCOS and assessed its possible role in clinical diagnosis. A total of 503 patients with PCOS were divided into eight subtypes according to the classification criteria of HA, IR, and OB.
Our results demonstrated a positive correlation of AMH with PCOM based on the following points: (1) The AMH level in the PCOM group was significantly higher than that in the non-PCOM group; (2) all patients with subtypes 5 and 6 had PCOM characteristics, which accounted for the high level of AMH; (3) PCOM was least common in subtype 4 patients, resulting in the lowest level of AMH; and (4) each group was further divided into a PCOM subgroup and a non-PCOM subgroup, and the AMH level in the PCOM subgroup was higher than that in the non-PCOM subgroup.
AMH can increase gonadotropin-releasing hormone (GnRH)-dependent LH pulse and secretion, which is an important pathophysiological feature of PCOS. However, the exact pathophysiological mechanism is not yet clear and requires further studies. Notably, although AMH is positively correlated with PCOM, AMH cannot completely predict PCOM. In our study, the diagnostic rate of AMH was significantly lower than that of ultrasound for PCOM. Using AMH alone to diagnose PCOM may lead to a missed diagnosis. The inconsistency between the AMH threshold and ultrasound measurement of antral follicles for the diagnosis of PCOM suggests that AMH cannot completely replace ultrasound in predicting PCOM. Indeed, owing to regional and racial differences, the diagnostic value of AMH for PCOM is not yet clear., Some studies suggest that the combined use of AMH and PCOM for the diagnosis of PCOS may be more accurate.,
Although PCOM was identified in 83.3% of patients with subtype 2, this subtype had the highest AMH level among all subtypes, indicating that PCOM may not be independently related to AMH, and that other factors may also affect AMH expression., Our study found that IR and OB were negatively correlated with AMH levels, which may explain why AMH levels were highest in PCOS patients with subtype 2.
In this study, we found no significant difference in AMH levels between the NHA group and the HA group. However, previous studies showed a positive correlation between AMH and T., The discrepancy may be attributable to the fact that our research object was the free testosterone level, which is affected by the SHBG level.
PCOS is characterized by clinical heterogeneity owing to complex pathogenetic mechanisms. In our cohort, subtype 4 accounted for the largest proportion of PCOS patients (30.82%). Subtype 4 is the most typical and severe clinical type of PCOS, and these patients show HA, OB, and IR. Moreover, this subtype is associated with the highest risk of abnormal glycolipid metabolism. We speculated that the AMH levels of these patients should be high, but surprisingly, our results showed just the opposite.
Recent studies have assessed the prediction of clinical ovulation induction based on AMH levels and found that a higher AMH level in patients with PCOS corresponded to a longer duration or a greater amount of Gn use and a lower quality embryo rate., However, for patients with this subtype and fertility requirements, in the absence of any intervention (such as lifestyle modification) before ovulation induction treatment, compared with intervention groups, the dose of ovulation-promoting drugs is higher, the quality of follicles is poorer, and the pregnancy rate is lower., Therefore, from our perspective, the AMH level is not directly proportional to the severity of clinical symptoms, and the use of AMH to predict the dose and efficacy of ovulation-promoting drugs may yield inconsistent results.
Among patients with classical PCOS (subtypes 1–4), the AMH levels in non-PCOM patients with the same subtypes were significantly lower than those in PCOM patients (even as low as 5.01 ng/mL in subtype 1). Therefore, our findings suggest that AMH cannot be used as a single index for the diagnosis of PCOS.
In conclusion, in this study, AMH showed a good positive correlation with PCOM and negative correlations with OB and IR; however, AMH showed no correlation with HA. Thus, AMH should not be used as the sole index for the diagnosis of PCOS. AMH levels may not accurately reflect the severity of HA and metabolic disorders in Chinese patients with PCOS. In addition, AMH cannot distinguish between the various PCOS subtypes according to the Chinese classification standard. Owing to the clinical heterogeneity of PCOS, the treatment strategy should be individualized based on clinical characteristics rather than relying on AMH levels alone.
This study had some limitations. First, all patients were enrolled at our hospital, which may have introduced an element of regional (selection) bias. Second, the sample size was limited, and some subtypes (such as subtypes 7 and 8) were not analyzed due to the small sample size. A larger study is required to provide more definitive conclusions.
Financial support and sponsorship
This work was supported by the Natural Science Foundation of Shanghai (No. 19ZR1406700 to Ying-Li Shi).
Conflicts of interest
There are no conflicts of interest.
| References|| |
Saxena U, Ramani M, Singh P. Role of AMH as diagnostic tool for polycystic ovarian syndrome. J Obstet Gynaecol India 2018;68:117-22. doi: 10.1007/s13224-017-1066-4.
Cimino I, Casoni F, Liu X, Messina A, Parkash J, Jamin SP, et al
. Novel role for anti-Müllerian hormone in the regulation of GnRH neuron excitability and hormone secretion. Nat Comm 2016;7:10055. doi: 10.1038/ncomms10055.
Quinn MM, Kao CN, Ahmad AK, Haisenleder DJ, Santoro N, Eisenberg E, et al
. Age-stratified thresholds of anti-Mullerian hormone improve prediction of polycystic ovary syndrome over a population-based threshold. Clin Endocrinol (Oxf) 2017;87:733-40. doi: 10.1111/cen.13415.
Sathyapalan T, Al-Qaissi A, Kilpatrick ES, Dargham SR, Atkin SL. Anti-Mullerian hormone measurement for the diagnosis of polycystic ovary syndrome. Clin Endocrinol (Oxf) 2018;88:258-62. doi: 10.1111/cen.13517.
Feldman RA, O'Neill K, Butts SF, Dokras A. Anti-Müllerian hormone levels and cardiometabolic risk in young women with polycystic ovary syndrome. Fertil Steril 2017;107:276-81. doi: 10.1016/j.fertnstert.2016.10.009.
Cui Y, Shi Y, Cui L, Han T, Gao X, Chen ZJ. Age-specific serum anti-Müllerian hormone levels in women with and without polycystic ovary syndrome. Fertil Steril 2014;102:230-600. doi: 10.1016/j.fertnstert.2014.03.032.
Lin YH, Chiu WC, Wu CH, Tzeng CR, Hsu CS, Hsu MI. Anti-Müllerian hormone and polycystic ovary syndrome. Fertil Steril 2011;96:230-5. doi: 10.1016/j.fertnstert.2011.04.003.
Chinese Medical Doctor Association Branch of the Endocrinology Department of Endocrinology. An endocrinologist's consensus for the diagnosis and treatment of polycystic ovary syndrome. Chin J Endocrinol Metab 2018;34:1-7. doi: 10.3760/cma.j.issn.1000-6699.2018.01.001.
Cimino I, Casoni F, Liu X, Messina A, Parkash J, Jamin SP, et al
. Novel role for anti-Müllerian hormone in the regulation of GnRH neuron excitability and hormone secretion. Nat Commun 2016;7:10055. doi: 10.1038/ncomms10055.
Lie Fong S, Laven JSE, Duhamel A, Dewailly D. Polycystic ovarian morphology and the diagnosis of polycystic ovary syndrome: Redefining threshold levels for follicle count and serum anti-Müllerian hormone using cluster analysis. Hum Reprod 2017;32:1723-31. doi: 10.1093/humrep/dex226.
Fraissinet A, Robin G, Pigny P, Lefebvre T, Catteau-Jonard S, Dewailly D. Use of the serum anti-Müllerian hormone assay as a surrogate for polycystic ovarian morphology: Impact on diagnosis and phenotypic classification of polycystic ovary syndrome. Hum Reprod 2017;32:1716-22. doi: 10.1093/humrep/dex239.
Lauritsen MP, Bentzen JG, Pinborg A, Loft A, Forman JL, Thuesen LL, et al
. The prevalence of polycystic ovary syndrome in a normal population according to the Rotterdam criteria versus revised criteria including anti-Mullerian hormone. Hum Reprod 2014;29:791-801. doi: 10.1093/humrep/det469.
Song DK, Oh JY, Lee H, Sung YA. Differentiation between polycystic ovary syndrome and polycystic ovarian morphology by means of an anti-Müllerian hormone cutoff value. Korean J Intern Med 2017;32:690-8. doi: 10.3904/kjim.2016.038.
Saxena U, Ramani M, Singh P. Role of AMH as Diagnostic Tool for Polycystic Ovarian Syndrome. J Obstet Gynaecol India 2018;68:117-22. doi: 10.1007/s13224-017-1066-4.
Lefebvre T, Dumont A, Pigny P, Dewailly D. Effect of OB and its related metabolic factors on serum anti-Mullerian hormone concentrations in women with and without polycystic ovaries. Reprod Biomed Online 2017;35:325-30. doi: 10.1016/j.rbmo.2017.05.013.
Li J, Li R, Yu H, Zhao S, Yu Y, Qiao J. The relationship between serum anti-Müllerian hormone levels and the follicular arrest for women with polycystic ovary syndrome. Syst Biol Reprod Med 2015;61:103-9. doi: 10.3109/19396368.2014.973123.
Recabarren SE, Sir-Petermann T, Rios R, Maliqueo M, Echiburú B, Smith R, et al
. Pituitary and testicular function in sons of women with polycystic ovary syndrome from infancy to adulthood. J Clin Endocrinol Metab 2008;93:3318-24. doi: 10.1210/jc.2008-0255.
Gülsen MS, Ulu I, Yildirim Köpük S, Kiran G. The role of anti-Müllerian hormone in predicting clomiphene citrate resistance in women with polycystic ovarian syndrome. Gynecol Endocrinol 2019;35:86-9. doi: 10.1080/09513590.2018.1499085.
Chen Y, Ye B, Yang XJ, Zheng JJ, Lin JJ, Zhao JZ. Predicting the outcome of different protocols of in vitro
fertilization with anti-Müllerian hormone levels in patients with polycystic ovary syndrome. J Int Med Res 2017;45:1138-47. doi: 10.1177/0300060517704140.
Gaba A, Hörath S, Hager M, Marculescu R, Ott J. Basal Anti Mullerian hormone levels and endometrial thickness at midcycle can predict the outcome after clomiphene citrate stimulation in anovulatory women with PCOS, a retrospective study. Arch Gynecol Obstet 2019;300:1751-7. doi: 10.1007/s00404-019-05359-7.
[Figure 1], [Figure 2], [Figure 3]