European Journal of Obstetrics & Gynecology and Reproductive Biology 163 (2012) 180–184

Normal serum anti-Mullerian hormone levels in the general female populationand the relationship with reproductive history

Antonio La Marca a,*, Elena Spada b, Valentina Grisendi a, Cindy Argento a, Enrico Papaleo c,Silvano Milani b, Annibale Volpe a

a Mother-Infant Department, Institute of Obstetrics and Gynecology, University of Modena and Reggio Emilia, 41100 Modena, Italyb Dipartimento di Medicina del Lavoro ‘‘Clinica del Lavoro L Devoto’’, Sezione di Statistica Medica e Biometria ‘‘GA Maccacaro’’, Universita degli Studi di Milano, Milan, Italyc Centro Natalita, Gynecological-Obstetrics Department, San Raffaele Hospital, Vita-Salute San Raffaele, Milano, Italy

A R T I C L E I N F O

Article history:

Received 14 October 2011

Received in revised form 5 April 2012

Accepted 17 April 2012

Keywords:

AMH

Healthy women

Reproductive history

BMI

Smoking

A B S T R A C T

Objective: Anti-Mullerian hormone (AMH) has been evaluated by several groups as a potential novel

clinical marker of ovarian reserve. Considering the wide use of AMH measurement in daily clinical

practice and the large number of conditions in which it may be used, it is essential to establish reference

values in the healthy female population. In this study we aim to calculate the age-by-age normal values

of circulating AMH. In addition, we report on AMH levels in women according to BMI, smoking status and

reproductive history.

Study design: The study was performed at the Institute of Obstetrics and Gynecology, University of

Modena, between January 2008 and December 2010. A total of 416 healthy women (aged 18–51) were

recruited and serum AMH levels were measured for all of them. The centiles of AMH distribution were

estimated with the CG-LMS method. The relationship between AMH levels and the womens’

characteristics such as BMI, smoking status and reproductive history was analysed by using the uni-

and multi-variable regression analysis and the Chi-square test.

Results: Serum AMH concentrations show a progressive decline with female ageing. Age-related

nomograms for the 5th, 25th, 50th, 75th, and 95th percentiles of AMH were produced. Mean AMH

concentrations were not modified by smoking habit and BMI and were independent of parity of the women.

Conclusion: In the present study, we established age-specific reference values for circulating AMH levels

in the eumenorrheic female population. AMH measurement produces new information on ovarian

pathophysiology and ovarian reserve and the establishment of reference values for AMH is the first step

for a correct interpretation of the assay.

� 2012 Elsevier Ireland Ltd. All rights reserved.

Contents lists available at SciVerse ScienceDirect

European Journal of Obstetrics & Gynecology andReproductive Biology

jou r nal h o mep ag e: w ww .e lsev ier . co m / loc ate /e jo g rb

1. Introduction

Anti-Mullerian hormone (AMH) is a dimeric glycoproteinmember of the transforming growth factor-beta (TGF-b) super-family, expressed in the growing follicles in the ovary untilthey have reached the size and differentiation state at which theymay be selected for dominance. In the mouse, this occurs at theearly antral stage in small growing follicles, whereas in thehuman it is evident in antral follicles 4–6 mm in diameter [1].Thus, AMH is expressed in follicles that have undergonerecruitment from the primordial follicle pool but have not beenselected for dominance.

* Corresponding author at: Mother-Infant Department, Institute of Obstetrics and

Gynecology, University of Modena and Reggio Emilia, Policlinico di Modena, Largo

del Pozzo, 41100 Modena, Italy. Tel.: +39 338 4795303; fax: +39 059 4224394.

E-mail address: antlamarca@libero.it (A. La Marca).

0301-2115/$ – see front matter � 2012 Elsevier Ireland Ltd. All rights reserved.

http://dx.doi.org/10.1016/j.ejogrb.2012.04.013

AMH is secreted by the ovary into the circulation and ismeasurable in serum [2]. As serum AMH levels essentially reflectthe ovarian follicular pool, age-related reduction in the number ofsmall growing follicles may be followed by a reduction incirculating AMH [2].

In women, AMH levels show a progressive decline duringreproductive life as the follicular reserve decreases, becomingundetectable after the menopause [2–4]. The observed relation-ship between the follicular ovarian pool and serum AMH levelsindicates that serum levels could provide additional informationduring the diagnostic evaluation of ovarian dysfunction such ashypogonadotropic- and hypergonadotropic-hypogonadism [5–9]and polycystic ovary syndrome (PCOS) [10,11]. In other words,AMH may constitute a new endocrine parameter for theinvestigation of ovarian function.

Most importantly, AMH has been evaluated by several groupsas a potential novel clinical marker of ovarian reserve andresponse to gonadotropins, permitting the identification of both

Fig. 1. Age-specific centiles (5th, 25th, 50th, 75th and 95th) of circulating AMH.

Grey dots represent the observations.

A. La Marca et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 163 (2012) 180–184 181

the poor and the excessive response to gonadotropin administra-tion [12–17].

Therefore, considering the wide use of AMH measurement indaily clinical practice and the large number of conditions in whichit may be used, it is essential to establish the reference values in thehealthy female population. Indeed, we recently published thenormal values for AMH in women of different age groups [18]. Inthe present study, on the basis of a large group of healthy womenwe aim to calculate the age-by-age normal values of circulatingAMH. In addition, we report on AMH levels in women according toBMI, smoking status and reproductive history.

2. Materials and methods

The study was performed at the Institute of Obstetrics andGynecology, University of Modena, between January 2008 andDecember 2010. Volunteers (n = 416) were recruited amongpatients attending the Institute of Obstetrics and Gynecology forthe following reasons: participation in the programme of cervicalcancer screening (n = 65), preconception counselling (n = 224),contraception counselling (n = 89), and request for tubal steriliza-tion (n = 38). Women were recruited when the following criteriawere met: regular menstrual cycle (length 25–35 days) with <5days difference between cycles, age 18–51 years, and presence ofboth ovaries. Exclusion criteria were: history of hormoneadministration in the previous 6 months; history of pelvicinflammatory disease or ovarian surgery; history of infertility;known chronic, systemic, metabolic, and endocrine diseaseincluding hyperandrogenism. Some of these data were also usedin a study previously published by our group [18].

At the time of blood sampling women were asked to fill in aquestionnaire regarding the following information: height andweight measurements, smoking status, previous ovarian surgery,reproductive history and regularity of their menstrual cycle. At thetime of sampling 187 (44.9%) and 229 (55.1%) women were in thefollicular and luteal phase of the menstrual cycle, respectively. Allwomen gave their written informed consent before the bloodsampling.

2.1. AMH assay

The blood sample for AMH determination was performed onthe day in which patients were recruited, independently of thelast menstrual cycle. The blood was centrifuged at 2000 � g for10 min and the serum was stored in polypropylene tubes at�80 8C.

Serum AMH was measured by enzyme-linked immunosorbentassay (ELISA) using the Beckman Coulter, Inc. (Chaska, MN, USA)AMH ELISA kit (Immunotech version, Marseilles, France). AMHvalues are presented in concentration of ng/ml (conversion factorto pmol/L = 7.143 ng/ml). The detection limit of the assay was0.14 ng/ml; intra- and inter-assay coefficients of variation were12.3% and 14.2%, respectively. The immunoassay is specific forAMH. No cross-reaction was observed with transforming growthfactor-beta.

2.2. Statistical analysis

The centiles of AMH distribution were estimated with the CG-LMS method [19]. The centiles are modelled according to threeage-specific smoothing splines called L(t), M(t) and S(t). The M(t)and S(t) curves correspond to the median and coefficient ofvariation of AMH at age t, whereas the L(t) curve allows for theskewness of the distribution. The value of a given centile (ya) iscomputed as ya = M(t) � (1 + SDSa � L(t) � S(t))(1/L(t)) where SDSa

is the normal deviate of the centile, and the standard deviation

score corresponding to the value (y) of AMH is obtained asSDS = [(y/M)L(t) � 1]/(L(t) � S(t)).

The relationship between AMH levels and patients’ character-istics such as BMI, smoking status, and reproductive history wasanalysed by using uni- and multi-variable regression analysis andthe Chi-square test. p < 0.05 was assumed to be statisticallysignificant. Statistical analysis was performed using the softwareStata 10 (StataCorp, TX, USA) and LMS Program version 1.29.

3. Results

3.1. AMH levels in healthy women

The median age (25th–75th range) of regularly menstruatingwomen (n = 416) included for the nomogram construction was 34(27–40 years, range 18–51) and the median (25th–75th range) ofserum AMH levels was 3.45 ng/ml (1.64–5.64 ng/ml, range 0–19.34). Fig. 1 shows AMH values as a function of age as well as theestimates of selected centiles (5th, 25th, 50th, 75th, 95th).The model used to trace L(t), M(t) and S(t) smoothing splinesand the 5th, 25th, 75th, 95th centiles (Table 1) was L2M2S2 (i.e. amodel with 2 edf for each spline). The values of L(t), which arelower than 1 in the whole range of age under study, denote that thedistribution of AMH is positively skewed.

3.2. AMH, BMI and smoking status

AMH levels and BMI were significantly and negativelycorrelated (r: �0.1; p = 0.03). As BMI significantly increased withage (r: 0.2; p < 0.01), however, a stepwise multiple regression wasperformed, considering AMH as a dependent variable and BMI andage as covariates. Analysis revealed that changes in AMH may beexplained only by changes in age and BMI was not included in themodel (data not shown).

Eighty-three out of 416 women were smokers. The median (andinterquartiles) age and AMH levels were similar in smoking andnon-smoking groups (35 years, 27–43 versus 34 years, 27–39and 3.1 ng/ml, 1.1–5 versus 3.5 ng/ml, 1.9–5.7 ng/ml, respectively).The distribution of smoking patients in the age-specific AMH quartileswas not different when compared to that of non-smoking women(<258: 30 versus 28%, 25–508: 21.6 versus 20%, 50–758: 19.4 versus23%, >758: 29 versus 29%). Regression analysis confirmed a non-significant effect of smoking on circulating AMH (data not shown).

3.3. AMH and reproductive history

Reproductive history was known for 348 of the 416 women:138 women had never been pregnant, whereas 210 had been

Table 1Values of L(t) and S(t), and of 5th, 25th, 50th, 75th and 95th centiles for AMH (ng/ml) as function of age. The values of a centile, at age t, can be computed as

y = M(t) � (1 + SDS � L(t) � S(t))�1/L(t) where SDS is the value of normal deviate.

Age (years) L S 5th 25th 50th (M) 75th 95th

18 0.4471 0.5709 1.74 3.86 5.891 8.40 12.90

19 0.4341 0.5811 1.67 3.73 5.729 8.23 12.74

20 0.4210 0.5912 1.59 3.60 5.566 8.05 12.58

21 0.4080 0.6013 1.52 3.47 5.404 7.86 12.40

22 0.3949 0.6115 1.45 3.34 5.241 7.68 12.23

23 0.3818 0.6216 1.39 3.22 5.078 7.49 12.04

24 0.3688 0.6318 1.33 3.09 4.916 7.31 11.85

25 0.3557 0.6419 1.26 2.97 4.753 7.11 11.65

26 0.3427 0.6521 1.21 2.85 4.590 6.91 11.44

27 0.3296 0.6622 1.15 2.73 4.428 6.72 11.23

28 0.3166 0.6723 1.09 2.62 4.265 6.52 11.01

29 0.3035 0.6825 1.04 2.50 4.102 6.31 10.78

30 0.2905 0.6926 0.99 2.38 3.940 6.11 10.54

31 0.2774 0.7028 0.94 2.27 3.777 5.90 10.29

32 0.2643 0.7129 0.89 2.16 3.615 5.68 10.04

33 0.2513 0.7230 0.84 2.05 3.452 5.47 9.77

34 0.2382 0.7332 0.79 1.94 3.289 5.25 9.49

35 0.2252 0.7433 0.75 1.84 3.127 5.03 9.21

36 0.2121 0.7535 0.70 1.73 2.964 4.80 8.91

37 0.1991 0.7636 0.66 1.63 2.801 4.57 8.59

38 0.1860 0.7738 0.62 1.52 2.639 4.34 8.27

39 0.1730 0.7839 0.58 1.42 2.476 4.11 7.93

40 0.1599 0.7940 0.53 1.32 2.313 3.86 7.57

41 0.1469 0.8042 0.49 1.22 2.151 3.62 7.20

42 0.1338 0.8143 0.45 1.12 1.988 3.38 6.82

43 0.1207 0.8245 0.42 1.03 1.826 3.13 6.41

44 0.1077 0.8346 0.38 0.93 1.663 2.87 5.98

45 0.0946 0.8447 0.34 0.84 1.500 2.61 5.53

46 0.0816 0.8549 0.30 0.74 1.338 2.35 5.07

47 0.0685 0.8650 0.26 0.65 1.175 2.09 4.57

48 0.0555 0.8752 0.23 0.55 1.012 1.81 4.04

49 0.0424 0.8853 0.19 0.46 0.850 1.53 3.49

50 0.0294 0.8955 0.15 0.38 0.687 1.26 2.91

51 0.0163 0.9056 0.12 0.28 0.525 0.95 2.29

A. La Marca et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 163 (2012) 180–184182

pregnant at least once. The relationship between pregnancy andAMH was studied using two different approaches. Firstly, thedistribution of the number of pregnancies, abortions, anddeliveries was analysed according to the age-specific AMHquartiles (Table 2) and secondly regression analysis (data notshown) was performed. Both statistical approaches showed thatthe number of pregnancies, abortions and deliveries were similarin women with low or high age-specific AMH levels, indicating anAMH-unrelated pattern in reproductive outcome for the generalpopulation.

4. Comment

To our knowledge this is the largest study on AMH performed inthe female adult general population. In the present study themedian reduction in AMH levels across the reproductive period isof about 0.16 ng/ml per year. In previous studies circulating AMHconcentrations declined with increasing reproductive age in a

Table 2Reproductive outcome in eumenorrheic women according to their age-specific AMH le

AMH

quartiles

Na Women with

at least one

pregnancy (n)

Mean number of

pregnancies per

woman (with

gravidity � 1)

Number of

offsprings

(n)

Mean

of off

wom

gravid

<258 105 52 1.92 76 1.46

25–508 106 44 1.75 58 1.31

50–758 86 45 2.04 69 1.53

>758 119 69 1.97 103 1.49

a Women have been divided into four groups according to their age-specific AMH l

outcome has also been confirmed by multivariate regression analysis (data not shown

manner optimally described by a quadratic equation [18,20]. Thisdiscrepancy stems from the fact that the data in the present studywere treated without logarithmic conversion. Until today severalnomograms for AMH have been published [20–22], but all thesenomograms are based on measurements performed in infertilewomen, which logically may not be suitable for the generalpopulation, since the prevalence of conditions associated withabnormal AMH levels, such as menstrual irregularities, PCOS,ovarian dysfunction, previous ovarian surgery and endometriosis,is expected to be higher in an infertile than in a general population.

More importantly, all the above-mentioned studies are basedon the DSL assay, whereas the present study is based on the IOTassay. The DSL and IOT assays utilize two different primaryantibodies against AMH and different standards, and consequentlythe crude values reported can differ substantially between papers,with the IOT assay giving values for AMH that are higher than thoseobtained with the DSL assay [16]. With the recent consolidation ofthese two companies by Beckman Coulter, however, finally there is

vels.

number

spring per

an (with

ity � 1)

Women (with

gravidity � 1)

with at least

one offspring, n (%)

Women (with

gravidity � 1) with

at least one

miscarriage, n (%)

Mean number of

miscarriages per

woman (with

gravidity � 1)

44/52 (84.6) 13/52 (25) 0.32

41/44 (93.1) 7/44 (15.9) 0.22

40/45 (88.8) 12/45 (26.6) 0.31

64/69 (92.7) 13/69 (18.8) 0.37

evels. The non significant correlation between AMH and the reproductive female

).

A. La Marca et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 163 (2012) 180–184 183

a single commercially available assay—the AMH Gen II assay. Thisnew assay will fully replace the DSL kit and the IOT kit, but thestandards used for the older IOT assay have been maintained in theGeneration II assay, and consequently, as noted, similar values aregenerated [23]. At present a nomogram for the new Gen II AMHassay has yet to be developed but the nomogram based on the IOTkit presented in this study should also be valid for the newgeneration assay.

The only paper reporting normal values for the IOT assay in thegeneral population was developed by our group on a cohort of 272women, but with analysis limited to 5-year age groups [18]. Hencethe present study, introducing the year by year reference values,expands and improves our previous published observation.

In the present study AMH measurement was performedindependently of the menstrual cycle, and indeed AMH serumlevels do not significantly change throughout the menstrual cycle[5,24–26]. Others, however, have reported significant fluctuationsin AMH levels [27–29]. Nevertheless these variations are similarto reported inter-cycle fluctuations for AMH [26], indicating thatin the clinical setting the inter- and intra-cycle variability inserum AMH levels may be considered to be low enough to permitrandom timing of AMH measurements during the menstrualcycle.

It has been reported that obese women show reduced levels ofinhibin B and AMH [30], suggesting that obesity may be associatedwith impaired ovarian reserve. A recent study [31] examined thecorrelation of obesity with hormonal and ultrasound-derivedmarkers of ovarian reserve and found that serum AMH levels arelower in obese women compared with age-matched women ofnormal weight, despite similar antral follicular count. Thissuggests that AMH levels in obese women may be lower forphysiological reasons related to obesity itself and may not benecessarily indicative of impaired ovarian reserve [31]. In thepresent study, AMH levels in the general, normally menstruatingwomen decreased with increasing BMI, but a careful analysishighlighted an indirect relationship between AMH and BMI. Asmean BMI increased with age whereas AMH levels decreased, therelationship between BMI and AMH seems to be secondary to thestronger relationship with age.

There is continuing debate concerning the possibility thatsmoking may directly accelerate ovarian follicular depletion,reducing the median age of menopause. Contradictory resultshave been reported on the relationship between smoking andAMH, with some authors [32] reporting reduced AMH levels insmokers versus non-smokers, and others [33] reporting similarvalues in both groups of women. In the present study we reportsimilar values of AMH in smokers and non-smokers, indicating thatif smoking may really be able to reduce the ovarian follicular pool,this effect may be so small that not all studies are able to find itthrough statistical analysis. Of course several methodologicallimitations may have flawed our results, for example, the absenceof information on duration and quantity of smoking.

The relationship between serum AMH levels and the women’sreproductive history has been analysed. Our study clearly showedthat pregnancies, miscarriages and the number of offspring aredistributed in an AMH-unrelated pattern, at least in the generalnormally menstruating female population. In particular theproportion of women with at least one miscarriage, the meannumber of miscarriages per woman, and the mean number ofpregnancies and offspring were similar in the groups of womenwith AMH levels in the lowest and highest quartiles for AMHvalues (Table 2). This allows us to conclude that at least in thenormal range of 5th–95th centile of AMH values for the specificage, modifications in ovarian reserve as assessed by AMH do notseem to give reproductive advantage or disadvantage. Of courseonly well-designed longitudinal studies including women from

healthy couples seeking pregnancies may answer the questionwhether AMH measurement can predict future female fertility.

The identification of women with AMH at the extremes of thenormal range may have several possible clinical implications.Women with an AMH low for their age may be very likely toexperience menopause at a younger age than expected from agealone. The clinical relevance of this diagnostic improvement mayrely on the known correlation existing between the quantitativeaspect of ovarian reserve and duration of reproductive period.Hence women with abnormally low AMH values may have areduced reproductive lifespan. Although this hypothesis needs tobe confirmed in specifically designed longitudinal studies in thegeneral population, we think that patients should be aware of thispossibility. Hence such a nomogram could be an important tool forforecasting the reproductive lifespan.

In conclusion, in the present study we established age-specificreference values for circulating AMH levels in the eumenorrheicfemale population. AMH measurement produces new informationon ovarian pathophysiology and ovarian reserve. Whether thisinformation is clinically more relevant than the measurement ofother known markers of ovarian function is still under investiga-tion [16], but the establishment of reference values for AMH is thefirst step towards answering these questions in the near future.

References

[1] Weenen C, Laven JS, Von Bergh AR, et al. Anti-Mullerian hormone expressionpattern in the human ovary: potential implications for initial and cyclic folliclerecruitment. Molecular Human Reproduction 2004;10:77–83.

[2] La Marca A, Volpe A. Anti-Mullerian hormone (AMH) in female reproduction:is measurement of circulating AMH a useful tool? Clinical Endocrinology(Oxford) 2006;64:603–10.

[3] La Marca A, De Leo V, Giulini S, et al. Anti-Mullerian hormone in premeno-pausal women and after spontaneous or surgically induced menopause.Journal of the Society for Gynecologic Investigation 2005;12:545–8.

[4] van Rooij IA, Tonkelaar I, Broekmans FJ, et al. Anti-Mullerian hormone is apromising predictor for the occurrence of the menopausal transition. Meno-pause 2004;11:601–6.

[5] La Marca A, Pati M, Orvieto R, Stabile G, Carducci Artenisio A, Volpe A. Serumanti-mullerian hormone levels in women with secondary amenorrhea. Fertili-ty and Sterility 2006;85:1547–9.

[6] van Elburg AA, Eijkemans MJ, Kas MJ, et al. Predictors of recovery of ovarianfunction during weight gain in anorexia nervosa. Fertility and Sterility2007;87:902–8.

[7] Meduri G, Massin N, Guibourdenche J, et al. Serum anti-Mullerian hormoneexpression in women with premature ovarian failure. Human Reproduction2007;22:117–23.

[8] Knauff EA, Eijkemans MJ, Lambalk CB, et al. Anti-Mullerian hormone, inhibin B,and antral follicle count in young women with ovarian failure. Journal ofClinical Endocrinology and Metabolism 2009;94:786–92.

[9] La Marca A, Marzotti S, Brozzetti A, et al. Primary ovarian insufficiency due tosteroidogenic cell autoimmunity is associated with a preserved pool of func-tioning follicles. Journal of Clinical Endocrinology and Metabolism2009;94:3816–23.

[10] Pigny P, Merlen E, Robert Y, et al. Elevated serum level of anti-mullerianhormone in patients with polycystic ovary syndrome: relationship to theovarian follicle excess and to the follicular arrest. Journal of Clinical Endocri-nology and Metabolism 2003;88:5957–62.

[11] Laven JS, Mulders AG, Visser JA, Themmen AP, De Jong FH, Fauser BC. Anti-Mullerian hormone serum concentrations in normoovulatory and anovulatorywomen of reproductive age. Journal of Clinical Endocrinology and Metabolism2004;89:318–23.

[12] Hazout A, Bouchard P, Seifer DB, Aussage P, Junca AM, Cohen-Bacrie P. Serumantimullerian hormone/mullerian-inhibiting substance appears to be a morediscriminatory marker of assisted reproductive technology outcome thanfollicle-stimulating hormone, inhibin B, or estradiol. Fertility and Sterility2004;82:1323–9.

[13] Fanchin R, Schonauer LM, Righini C, Frydman N, Frydman R, Taieb J. Serumanti-Mullerian hormone dynamics during controlled ovarian hyperstimula-tion. Human Reproduction 2003;18:328–32.

[14] Muttukrishna S, Suharjono H, McGarrigle H, Sathanandan M. Inhibin B andanti-Mullerian hormone: markers of ovarian response in IVF/ICSI patients?BJOG 2004;111:1248–53.

[15] La Marca A, Giulini S, Tirelli A, et al. Anti-Mullerian hormone measurement onany day of the menstrual cycle strongly predicts ovarian response in assistedreproductive technology. Human Reproduction 2007;22:766–71.

[16] La Marca A, Broekmans FJ, Volpe A, Fauser BC, Macklon NS, ESHRESpecial Interest Group for Reproductive Endocrinology—AMH Round Table.

A. La Marca et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 163 (2012) 180–184184

Anti-Mullerian hormone (AMH): what do we still need to know? HumanReproduction 2009;24:2264–75.

[17] La Marca A, Sighinolfi G, Radi D, et al. Anti-Mullerian hormone (AMH) as apredictive marker in assisted reproductive technology (ART). Human Repro-duction Update 2010;16:113–30.

[18] La Marca A, Sighinolfi G, Giulini S, et al. Normal serum concentrations of anti-Mullerian hormone in women with regular menstrual cycles. ReproductiveBiomedicine Online 2010;21:463–9.

[19] Cole TJ, Green PJ. Smoothing reference centile curves: the LMS method andpenalized likelihood. Statistics in Medicine 1992;11:1305–19.

[20] Nelson SM, Messow MC, Wallace AM, Fleming R, McConnachie A. Nomogramfor the decline in serum antimullerian hormone: a population study of 9,601infertility patients. Fertility and Sterility 2011;95:736–41.

[21] Seifer DB, Baker VL, Leader B. Age-specific serum anti-Mullerian hormonevalues for 17,120 women presenting to fertility centers within the UnitedStates. Fertility and Sterility 2011;95:747–50.

[22] Shebl O, Ebner T, Sir A, et al. Age-related distribution of basal serum AMH levelin women of reproductive age and a presumably healthy cohort. Fertility andSterility 2011;95:832–4.

[23] Kumar A, Kalra B, Patel A, McDavid L, Roudebush WE. Development of a secondgeneration anti-Mullerian hormone (AMH) ELISA. Journal of ImmunologicalMethods 2010;362:51–9.

[24] Hehenkamp WJ, Looman CW, Themmen AP, de Jong FH, Te Velde ER, Broek-mans FJ. Anti-Mullerian hormone levels in the spontaneous menstrual cycledo not show substantial fluctuation. Journal of Clinical Endocrinology andMetabolism 2006;91:4057–63.

[25] Tsepelidis S, Devreker F, Demeestere I, Flahaut A, Gervy Ch, Englert Y.Stable serum levels of anti-Mullerian hormone during the menstrual cycle: a

prospective study in normo-ovulatory women. Human Reproduction 2007;22:1837–40.

[26] Streuli I, Fraisse T, Pillet C, Ibecheole V, Bischof P, de Ziegler D. Serumantimullerian hormone levels remain stable throughout the menstrual cycleand after oral or vaginal administration of synthetic sex steroids. Fertility andSterility 2008;90:395–400.

[27] Wunder DM, Bersinger NA, Yared M, Kretschmer R, Birkhauser MH. Statisti-cally significant changes of antimullerian hormone and inhibin levels duringthe physiologic menstrual cycle in reproductive age women. Fertility andSterility 2008;89:927–33.

[28] Streuli I, Fraisse T, Chapron C, Bijaoui G, Bischof P, de Ziegler D. Clinical uses ofanti-Mullerian hormone assays: pitfalls and promises. Fertility and Sterility2009;91:226–30.

[29] Sowers M, McConnell D, Gast K, et al. Anti-Mullerian hormone and inhibin Bvariability during normal menstrual cycles. Fertility and Sterility 2010;94:1482–6.

[30] Gracia CR, Freeman EW, Sammel MD, Lin H, Nelson DB. The relationshipbetween obesity and race on inhibin B during the menopause transition.Menopause 2005;12:559–66.

[31] Su HI, Sammel MD, Freeman EW, Lin H, DeBlasis T, Gracia CR. Body size affectsmeasures of ovarian reserve in late reproductive age women. Menopause2008;15:857–61.

[32] Plante BJ, Cooper GS, Baird DD, Steiner AZ. The impact of smoking on anti-mullerian hormone levels in women aged 38 to 50 years. Menopause2010;17:571–6.

[33] Dafopoulos A, Dafopoulos K, Georgoulias P, et al. Smoking and AMH levels inwomen with normal reproductive history. Archives of Gynecology and Ob-stetrics 2010;282:215–9.