|Home | About | Journals | Submit | Contact Us | Français|
To compare serum mullerian inhibiting substance (MIS) levels between white, black and Hispanic women to determine if ovarian aging occurs at a different time course for women of different racial groups.
Longitudinal study of serum MIS levels in women of different race/ethnicity over two different time points.
Women’s Interagency HIV Study, a multicenter prospective cohort study.
Serum samples obtained from 809 participants (122 white, 462 black and 225 Hispanic women).
Comparison of serum MIS between women of different race/ethnicity at two time points (median age 37.5 years and 43.3 years).
Variation in MIS by race/ethnicity over time, controlling for age, BMI, HIV status and smoking.
Compared to white women, average MIS values were lower among black (25.2% lower, p=0.037) and Hispanic (24.6% lower, p=0.063) women, adjusting for age, BMI, smoking and HIV status.
There is an independent effect of race/ethnicity on the age-related decline in MIS over time.
There is a growing body of medical literature which indicates that female reproductive function may differ by race. Within the context of the treatment of infertility, race differences have been addressed by several studies examining in vitro fertilization (IVF) outcomes. Initial small studies (1–4) on racial differences in outcomes from assisted reproductive technology (ART) produced conflicting results. More recently two large studies demonstrated differences in pregnancy rates and miscarriage rates by race (5,6). Seifer and colleagues (5) examined The Society for Assisted Reproductive Technology (SART) database of over 80,000 cycles the years for 1999 and 2000, and found lower pregnancy and higher miscarriage rates among black as compared to white women when controlled for age. Also using the SART database, Purcell et al (6) demonstrated lower pregnancy rates in Asian-American women as compared to white women when examining the SART database of more than 25,000 cycles.
Furthermore, race may also influence the prevalence of premature menopause. Luborsky et al (7) noted differences between white, black and Hispanic women and concluded that the prevalence of premature ovarian failure appears to vary by race. Thus, the longevity of ovarian function may be influenced by race/ethnicity within the context of IVF, a common treatment of infertility and the epidemiology of premature ovarian failure.
These racial/ethnic differences in reproductive outcomes may result from racial differences in ovarian reserve. Mullerian Inhibiting Substance (MIS) or Antimullerian Hormone (AMH) has been recently identified as the earliest, most sensitive serum marker of ovarian reserve (8–10). It is also a measure that is relatively stable throughout the menstrual cycle so MIS/AMH is an optimal analyte for exploring the question of whether or not ovarian aging may differ by race (10–15). Here we report the findings of a comparison of serum MIS/AMH levels between white, black and Hispanic women enrolled in the Women’s Interagency HIV Study (WIHS) to explore if ovarian aging occurs at a different time course for different racial groups.
This study was nested in the Women’s Interagency HIV Study (WIHS), the largest ongoing multicenter prospective cohort study of HIV infection and related health conditions among HIV-infected women and high-risk seronegative women in the United States. The WIHS protocols and procedures have been previously described (16). WIHS participants were eligible to contribute data to this analysis if they had completed a study visit at age 36–40 years, and another between the ages of 41 and 46, and had stored serum available from both visits. Samples were withdrawn from the repository accordingly, in order to investigate the effect of race/ethnicity on age-related decline in MIS/AMS. Of the 891 potentially eligible participants, we excluded from analysis 42 with incomplete BMI data, 14 with incomplete smoking data, 3 who seroconverted between their two visits, and 23 whose self-identified race was not black, white or Hispanic.
HIV infection was previously detected using EIA/Western blot methods. Race was categorized as either white (women who self-identified as white non-Hispanic), black (women who self-identified as black non-Hispanic), or Hispanic (which included self-identified Hispanic, white Hispanic, and black Hispanic women). Written informed consent was obtained from all WIHS participants and study procedures were approved by the human subjects protection committees at all participating institutions (Bronx, NY; Brooklyn, NY; Chicago, IL; Los Angeles, CA; San Francisco, CA; and Washington, DC). The laboratory protocol for this study was approved by the Institutional Review Board at Women and Infants Hospital, Providence, RI.
Serum levels of MIS/AMH were measured using an enzyme-linked immunosorbent assay (ELISA) kit (DSL-10-14400, Diagnostic Systems Laboratories, Webster, TX) according to manufacturer recommendations. Samples were run in duplicate by a single operator without knowledge of group assignment. The lower limit of sensitivity was 0.10 ng/mL and inter-assay coefficients of variation were 6%, 7% and 10% at doses of 1.5, 2.2, and 8.1 ng/mL, respectively.
Univariate comparisons were made by race/ethnicity and MIS/AMH for the following potential confounders: age, body mass index (BMI), HIV infection status, and smoking (current/former/never). Medians and interquartile ranges (IQR) were reported for continuous variables. Statistical comparisons conducted include chi-square tests for categorical variables and the Kruskal-Wallis test for continuous variables. Multivariate linear regression analysis was conducted in order to assess the effect of race/ethnicity on serum MIS/AMH, adjusting for potential confounders as listed above. In order to assess the effect of race/ethnicity on serum MIS/AMH, we specified a repeated measures model with a random intercept for log-transformed AMH which incorporated left-censoring for undetectable values and adjusted for potential confounders as listed above (17). Age, BMI, and smoking status were time-dependent. Race/ethnicity and HIV status were fixed. Because a log-transformed outcome was used, the exponentiated betas represent the ratio of MIS levels in the raw scale for one group compared to the reference group for categorical factors or a one-unit difference in continuous factors. Subtracting one from the ratio and multiplying the result by 100 yields the percent difference. All analyses were conducted in SAS version 9 (SAS Institute Inc., Cary, NC).
We investigated longitudinal serum MIS/AMH levels among 809 WIHS participants, of whom 122 (15.1%) were white, 462 (57.1%) were black, and 225 (27.8%) were Hispanic. By design, the age range at visit 1 was 36–40 (median 37.5 years), and at visit 2 was 41–46 years (median 43.3 years). The median time between visits was 5.3 years (range: 1.5–8.7 years). Of the 809 participants, 628 (77.6%) were HIV-infected and 181 (22.4%) were HIV-uninfected; all participants’ HIV status was the same at both study visits.
White women were more likely to be HIV-infected, were significantly older at the second visit (p=0.04) and had significantly lower BMI at both visits (p<0.001), as compared to black and Hispanic women (Table 1). A greater proportion of white and Hispanic women had never smoked compared to black women (p<0.001). Median MIS/AMH was lower at the second visit than the first for all three racial categories, as all subjects were older (Table 1; Figure 1). MIS/AMH did not differ significantly by race/ethnicity at either time point (p>0.05 for both visits) and there were no significant differences in magnitude of change in MIS/AMH values between time points by race. The median decrease between visits in MIS/AMH was 0.7 ng/mL for white women, and 0.6 ng/mL for black and Hispanic women. However, while all three racial groups had similar proportions of undetectables at the first visit, black and Hispanic women had a higher proportion at the second visit.
We first conducted univariate linear regression analysis to assess the effect of age on MIS/AMH, stratified by race. For a 1-year difference in age, MIS/AMH for the older women was 20.9% lower for white women, 22.8% lower for black women, and 22.1% lower for Hispanic women (p<0.001 for all 3 models; data not shown). Multivariate linear regression analysis (Table 2) suggested a difference in the level of MIS/AMH by race. Controlling for age, BMI, HIV status, and smoking black women had an average MIS/AMH value that was 25.2% lower than the average MIS/AMH value for white women (p=0.037). The average MIS/AMH value for Hispanic women was 24.6% lower than for white women; however, this difference was only marginally significant (p=0.063). In addition to black race, another significant predictor of lower MIS/AMH was age (22.3% lower per 1-year increase in age, p<0.001). BMI, smoking, and HIV infection were not significantly associated with MIS/AMH in this analysis.
Almost all clinical studies to date have shown an inverse correlation between MIS/AMH and age but none have controlled for race or ethnicity. One reason for this is that many of the original longitudinal clinical studies of MIS/AMH were conducted in The Netherlands which has a homogenous population. Van Rooij et al (8) reported on 81 female volunteers with a mean age of 39.5 who were followed for a mean duration of 4 years and were noted to have a 58% decline in MIS/AMH serum levels over that time interval. However, that cohort of women was not stratified by race or ethnicity. Most recently, van Disseldorp et al. (18) demonstrated good conformity between observed distribution of age at menopause and that predicted from declining MIS/AMH levels. However, they also did not adjust for race or ethnicity.
The data from the current study demonstrate, for the first time, a significant difference in the mean level over time of MIS/AMH, a surrogate marker of ovarian aging, as a function of race or ethnicity. Our study presents biochemical evidence of a difference in ovarian aging as assessed by serum MIS/AMH between black and white women, and suggests a possible difference between white and Hispanic women, when controlling for age, BMI, smoking and HIV status. Ovarian function of white women appears to be sustained longer over time than in black or Hispanic women. Specifically, on average minority women have 25% lower serum MIS/AMH values than white women when adjusting for age, BMI, smoking history and HIV status. Implications of these findings may have potential broad applications for the life planning of minority women. As such, this information may influence minority women to seek medical assistance earlier if pregnancy is not easily accomplished. This information could influence physicians in their initial choice of fertility treatment options for minority women who may show earlier evidence of diminished ovarian reserve than their white counterparts. Finally, these data, if confirmed in followup studies, could have important implications for minority women who may experience an earlier onset of menopause compared to white women (19–21).
Our data further suggests that MIS/AMH levels are also unaffected by BMI, smoking and HIV status. Our findings of BMI and HIV status having no influence upon MIS/AMH serum levels are consistent with others (22,10). However, our findings of smoking not affecting MIS/AMH are not consistent with a recent report by Freour et al (23) demonstrating that MIS/AMH is diminished in current cigarette smokers. Further studies are needed to clarify this issue.
In conclusion, there is an independent effect of race/ethnicity on the age-related decline in MIS/AMH over time. It may be that the racial differences in pregnancy and miscarriage rates noted in large sample size IVF studies (5,6) and in the onset of premature ovarian failure (7) may be attributable in part to inherent differences in ovarian reserve. Furthermore, the influence race/ethnicity may have upon the measurement of serum MIS/AMH should be noted by both clinicians and investigators working with women of reproductive age.
Data in this manuscript were collected by the Women’s Interagency HIV Study (WIHS) Collaborative Study Group with centers (Principal Investigators) at New York City/Bronx Consortium (Kathryn Anastos); Brooklyn, NY (Howard Minkoff); Washington DC Metropolitan Consortium (Mary Young); The Connie Wofsy Study Consortium of Northern California (Ruth Greenblatt); Los Angeles County/Southern California Consortium (Alexandra Levine); Chicago Consortium (Mardge Cohen); Data Coordinating Center (Stephen Gange). The WIHS is funded by the National Institute of Allergy and Infectious Diseases (UO1-AI-35004, UO1-AI-31834, UO1-AI-34994, UO1-AI-34989, UO1-AI-34993, and UO1-AI-42590) and by the National Institute of Child Health and Human Development (UO1-HD-32632). The study is co- funded by the National Cancer Institute, the National Institute on Drug Abuse, and the National Institute on Deafness and Other Communication Disorders. Funding is also provided by the National Center for Research Resources (MO1-RR-00071, MO1-RR-00079, MO1-RR-00083).
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.