Flexible parametric survival models built on age-specific antimüllerian hormone percentiles are better predictors of menopause

Anti-Mullerian hormone and cardiometabolic status: a systematic review

BACKGROUND

To summarise the relationship between Anti-mullerian hormone (AMH) levels and cardiometabolic status in different populations.

METHODS

PubMed, Scopus, and Embase were searched for retrieving observational studies published up to February 2022 investigating the relationship between AMH level and cardiometabolic status.

RESULTS

Of 3,643 studies retrieved from databases, a total of 37 observational studies were included in this review. The majority of the included studies revealed an inverse association between AMH and lipid profiles, including triglycerides (TG), total cholesterol (TC), low-density lipoprotein (LDL), and a positive correlation with high-density lipoprotein (HDL). While some studies have revealed a significant inverse association between AMH and glycemic parameters, including fasting plasma glucose (FPG), fasting insulin, and HOMA-IR, others found no such relationships. There is also an inconsistency among studies regarding the association of AMH with adiposity indices and blood pressure. Evidence indicates a significant association between AMH and some vascular markers, such as intima-media thickness and coronary artery calcification. Of 3 studies evaluating the relationship between AMH and cardiovascular events, two studies showed an inverse relationship between AMH levels and cardiovascular (CVD), whereas another study showed no significant association.

CONCLUSIONS

The results of this systematic review suggest that serum AMH levels can be associated with CVD risk. This may provide new insight into the use of AMH concentrations as a predictive marker for assessing the risk of cardiovascular disease, although more well-design longitudinal studies are still necessary for this area. Future studies on this topic will hopefully provide an opportunity to run a meta-analysis; it will increase the persuasiveness of this interpretation.

Anti-Müllerian hormone for the diagnosis and prediction of menopause: a systematic review

BACKGROUND

The early onset of menopause is associated with increased risks of cardiovascular disease and osteoporosis. As a woman's circulating anti-Müllerian hormone (AMH) concentration reflects the number of follicles remaining in the ovary and declines towards the menopause, serum AMH may be of value in the early diagnosis and prediction of age at menopause.

OBJECTIVE AND RATIONALE

This systematic review was undertaken to determine whether there is evidence to support the use of AMH alone, or in conjunction with other markers, to diagnose menopause, to predict menopause, or to predict and/or diagnose premature ovarian insufficiency (POI).

SEARCH METHODS

A systematic literature search for publications reporting on AMH in relation to menopause or POI was conducted in PubMed®, Embase®, and the Cochrane Central Register of Controlled Trials up to 31 May 2022. Data were extracted and synthesized using the Synthesis Without Meta-analysis for diagnosis of menopause, prediction of menopause, prediction of menopause with a single/repeat measurement of AMH, validation of prediction models, short-term prediction in perimenopausal women, and diagnosis and prediction of POI. Risk-of-bias was evaluated using the Tool to Assess Risk of Bias in Cohort Studies protocol and studies at high risk of bias were excluded.

OUTCOMES

A total of 3207 studies were identified, and 41, including 28 858 women, were deemed relevant and included. Of the three studies that assessed AMH for the diagnosis of menopause, one showed that undetectable AMH had equivalent diagnostic accuracy to elevated FSH (>22.3 mIU/ml). No study assessed whether AMH could be used to shorten the 12 months of amenorrhoea required for a formal diagnosis of menopause. Studies assessing AMH with the onset of menopause (27 publications [n = 23 835 women]) generally indicated that lower age-specific AMH concentrations are associated with an earlier age at menopause. However, AMH alone could not be used to predict age at menopause with precision (with estimates and CIs ranging from 2 to 12 years for women aged <40 years). The predictive value of AMH increased with age, as the interval of prediction (time to menopause) shortened. There was evidence that undetectable, or extremely low AMH, may aid early diagnosis of POI in young women with a family history of POI, and women presenting with primary or secondary amenorrhoea (11 studies [n = 4537]).

WIDER IMPLICATIONS

The findings of this systematic review support the use of serum AMH to study the age of menopause in population studies. The increased sensitivity of current AMH assays provides improved accuracy for the prediction of imminent menopause, but diagnostic use for individual patients has not been rigorously examined. Prediction of age at menopause remains imprecise when it is not imminent, although the finding of very low AMH values in young women is both of clinical value in indicating an increased risk of developing POI and may facilitate timely diagnosis.

Age at menopause among rural and urban women in the state of Campeche, Mexico

OBJECTIVE

The purpose of this study was to examine age at natural menopause among women of Maya and non-Maya ancestry living in urban and rural communities in the state of Campeche, Mexico.

METHODS

Women ages 40 to 60 (n = 543) participated in semi-structured interviews and anthropometric measures. The last names, languages spoken, and the birthplace of the woman, her parents, and her grandparents were used to determine Maya or non-Maya ethnicity. Recalled age at natural menopause was compared across four communities; analysis of variance was used to compare means and Kaplan-Meier analyses were used to compare medians. Probit analysis was also used to estimate median ages at menopause. Cox regression analyses were applied to identify variables associated with age at menopause.

RESULTS

Mean recalled age at natural menopause across all sites was 46.7 years, ranging from 47.8 years in the city of Campeche to 43.9 years in the rural Maya communities in the municipality of Hopelchén. Median ages at menopause across all sites were 50.55 years by probit analysis and 50.5 years by Kaplan-Meier. Variables associated with a later age at menopause included higher socioeconomic status, higher parity, and a later age at menarche.

CONCLUSIONS

The early mean recalled age at menopause in southern Hopelchén was consistent with previous studies in the Yucatán peninsula. As expected, probit and Kaplan-Meier analyses demonstrated later ages at menopause. Contrary to our expectations, Maya/non-Maya ethnicity was not associated with age at menopause. Demographic and reproductive factors were more important than ethnicity in explaining variation in age at menopause within the state of Campeche, Mexico.

Improving Prediction of Age at Menopause Using Multiple Anti-Müllerian Hormone Measurements: the Tehran Lipid-Glucose Study

CONTEXT

Several statistical models were introduced for the prediction of age at menopause using a single measurement of anti-müllerian hormone (AMH); however, individual prediction is challenging and needs to be improved.

OBJECTIVE

The objective of this study was to determine whether multiple AMH measurements can improve the prediction of age at menopause.

DESIGN

All eligible reproductive-age women (n = 959) were selected from the Tehran Lipid and Glucose Study. The serum concentration of AMH was measured at the time of recruitment and twice after that at an average of 6-year intervals. An accelerated failure-time model with Weibull distribution was used to predict age at menopause, using a single AMH value vs a model that included the annual AMH decline rate. The adequacy of these models was assessed using C statistics.

RESULTS

The median follow-up period was 14 years, and 529 women reached menopause. Adding the annual decline rate to the model that included single AMH improved the model's discrimination adequacy from 70% (95% CI: 67% to 71%) to 78% (95% CI: 75% to 80%) in terms of C statistics. The median of differences between actual and predicted age at menopause for the first model was -0.48 years and decreased to -0.21 in the model that included the decline rate. The predicted age at menopause for women with the same amount of age-specific AMH but an annual AMH decline rate of 95 percentiles was about one decade lower than in those with a decline rate of 5 percentiles.

CONCLUSION

Prediction of age at menopause could be improved by multiple AMH measurements; it will be useful in identifying women at risk of early menopause.

Do dietary intakes influence the rate of decline in anti-Mullerian hormone among eumenorrheic women? A population-based prospective investigation

BACKGROUND

Dietary intakes are suggested to affect age at menopause but associations between dietary factors and ovarian reserve reduction have not yet been investigated. We aimed to examine dietary intakes in relation to the rate of decline in anti-Mullerian hormone (AMH), an indicator of ovarian reserve, in a generally healthy cohort of women.

METHODS

This prospective investigation was conducted among 227 eumenorrheic women, aged 20-50 years, from the Tehran Lipid and Glucose study, who were followed over a mean of 16 years. AMH was measured twice, at baseline and the 5th follow-up examination cycle, and yearly rate of decline in AMH was calculated. Rapid decline in AMH was defined as the annual percent change AMH > 5.9%/year based on tertile 3 of the variable. Average usual dietary intakes were estimated using the food frequency questionnaires administered at the second, third, and the fourth follow-up examinations. After adjusting for potential covariates, the association between dietary factors and both risk of rapid decline in AMH and also annual percent decline of AMH (as a continuous variable) were examined using logistic regression and the Spearman correlation, respectively.

RESULTS

The baseline age of the participants and the median rate of decline in AMH were 37.2 years and was 5.7% yearly, respectively. The odds of rapid decline in AMH was reduced by 47% for dairy products (95% CIs = 0.36, 0.79; p = 0.002), 38% for milk (95% CIs = 0.41, 0.93; p = 0.020), and 36% for fermented dairy (95% CIs = 0.45, 0.93, p = 0.018) per one standard deviation (SD) increase in their dietary intakes. The odds of rapid decline in AMH was significantly reduced with higher intakes of fat, carbohydrate, protein, and calcium intakes from dairy sources, lactose and galactose. Annual rate of AMH decline was inversely correlated with dairy products, milk, fermented dairy, fruits, dairy carbohydrate, dairy fat, dairy protein, total calcium and dairy calcium, lactose and galactose, and positively correlated with organ meats.

CONCLUSION

Dairy foods consumption may reduce the rate of AMH decline in regularly menstruating women. Life style modification in terms of dietary advice may be considered as a preventive strategy for reduction in the rate of ovarian reserve loss.

Association between ovarian reserve and preeclampsia: a cohort study

Background

The risk of cardiovascular disease in women increases after menopause. It has been shown that women with lower pre-menopausal ovarian reserve may experience increased cardiovascular risk. We sought to determine whether there is any association between ovarian reserve, as assessed by Anti-Mullerian hormone (AMH), and preeclampsia (PE).

Methods

Subjects of this study were selected from among participants of the Tehran Lipid and Glucose Study (TLGS), a population-based cohort with a 15-year follow-up (1998–2014). Out of 2412 women aged 20–50 years, there were 781 women who met eligibility criteria, including having comprehensive data on their reproductive assessment and ovarian reserve status, identified based on age-specific AMH levels according to the exponential–normal three-parameter model that was measured before pregnancy.

There were 80 and 701 participants in the preeclampsia and non-PE groups, respectively. The association between dichotomous outcome variable PE and age-specific AMH quartiles was evaluated using pooled logistic regression.

Results

PE was observed in 23 (11.1%), 12 (6.4%), 26 (13.3%) and 19 (10%) women in the 1st, 2nd, 3rd and 4th quartiles of pre-pregnancy age-specific AMH, respectively (P = 0.16). Median and inter-quartile range of serum AMH levels was 1.05 (0.36–2.2) mg/L in women who experienced PE compared with 0.85 (0.28–2.1) mg/L in women with normotensive pregnancies (P = 0.53). Based on the pooled logistic regression analysis, the effect of age-specific AMH quartiles on PE progression (adjusted for age, BMI, smoking status, and family history of hypertension) were not significant (OR_{1st vs 4th}: 1.5, P-value: 0.1, CI: (0.9, 2.4)).

Conclusions

Age-specific AMH may not be a suitable marker for prediction of PE. Further longitudinal studies, considering pre-conception measurement of AMH, are recommended for better interpretation of the association between ovarian reserve status and PE.

Overtime trend of thyroid hormones and thyroid autoimmunity and ovarian reserve: a longitudinal population study with a 12-year follow up

BACKGROUND

Ovarian reserve, vital for reproductive function, can be adversely affected by thyroid diseases. Despite alternations of thyroid hormones with ageing, data on interactions between the overtime trend of thyroid functions and ovarian reserve status has rarely been reported. We aimed to examine the overtime trend of thyroid hormones, thyroid peroxidase antibody (TPO Ab) and their associations with ovarian reserve status, identified by levels of age specific anti-mullerian hormone (AMH) in reproductive aged women, who participated in 12-year cohort of Tehran Thyroid Study (TTS).

METHODS

Reproductive age women(n = 775) without any thyroid disease or ovarian dysfunction were selected from the Tehran Thyroid Study cohort. Participants were divided into four age specific AMH quartiles (Q1-Q4), Q1, the lowest and Q4, the highest. AMH was measured at the initiation of study and thyroid stimulating hormone (TSH), free T4 (FT4), and TPO Ab were measured at baseline and at three follow up visits.

RESULTS

At baseline, there was no statistically significant difference in thyroid hormones between women of the four quartiles, although TPO Ab levels were higher in women of Q1. During the follow ups, FT4 was decreased in all quartiles (p < 0.05), whereas TPO Ab increased in Q1 (p = 0.02). Odds ratio of overall TPO Ab positivity in women of Q1 was 2.08 fold higher than those in Q4. (OR: 2.08, 95%CI: 1.16, 3.72; p = 0.01).

CONCLUSION

Women with the lowest ovarian reserves had higher levels of TPO Ab, with a positive trend of this antibody overtime in comparison to other quartiles, indicating that this group may be at a higher risk of hypothyroidism over time.

Variation in levels of AMH among Maya and non-Maya women in Campeche, Mexico

OBJECTIVES

The purpose of this study was to examine anti-Müllerian hormone (AMH) levels among women of Maya and non-Maya ancestry in the city of Campeche, Mexico. Levels of AMH can potentially predict age at menopause. Previous studies have indicated an early mean age at menopause among the Maya.

MATERIALS AND METHODS

Women aged 40-60 (n = 97) participated in semistructured interviews, anthropometric measures, and blood samples. Maya/non-Maya ethnicity was determined by the last names, languages spoken, and birthplace of the woman, her parents, and her grandparents. AMH values were categorized as detectable (0.05-4.19 ng/mL) and undetectable (<0.05 ng/mL). Logistic regressions calculated odds ratios (OR) for undetectable AMH.

RESULTS

Women were categorized as Maya (n = 44), not Maya (n = 39), or not able to be clearly defined (n = 14). In bivariate comparisons, women with detectable levels of AMH were younger, more likely to be pre-menopausal, and not Maya. Age, menopausal status, and ethnicity remained significant in a logistic regression models after controlling for age at menarche. Maya women were more than five times as likely to have nondetectable AMH levels as non-Maya women.

DISCUSSION

Increasing age and progression through the menopausal transition were both associated with declining levels of AMH. The association between Maya ethnicity and a lower likelihood of detecting AMH is consistent with the early ages at menopause reported in previous studies. We considered a rapid life history model as an explanatory framework, and suggest, from an ecological perspective, that future research should consider measures of developmental stress that may compromise ovarian reserves.

The current application of the Royston-Parmar model for prognostic modeling in health research: a scoping review

BACKGROUND

Prognostic models incorporating survival analysis predict the risk (i.e., probability) of experiencing a future event over a specific time period. In 2002, Royston and Parmar described a type of flexible parametric survival model called the Royston-Parmar model in Statistics in Medicine, a model which fits a restricted cubic spline to flexibly model the baseline log cumulative hazard on the proportional hazards scale. This feature permits absolute measures of effect (e.g., hazard rates) to be estimated at all time points, an important feature when using the model. The Royston-Parmar model can also incorporate time-dependent effects and be used on different scales (e.g., proportional odds, probit). These features make the Royston-Parmar model attractive for prediction, yet their current uptake for prognostic modeling is unknown. Thus, the objectives were to conduct a scoping review of how the Royston-Parmar model has been applied to prognostic models in health research, to raise awareness of the model, to identify gaps in current reporting, and to offer model building considerations and reporting suggestions for other researchers.

METHODS

Five electronic databases and gray literature indexed in web sources from 2001 to 2016 were searched to identify articles for inclusion in the scoping review. Two reviewers independently screened 1429 articles, and after applying exclusion criteria through a two-step screening process, data from 12 studies were abstracted.

RESULTS

Since 2001, only 12 studies were identified that used the Royston-Parmar model in some capacity for prognostic modeling, 10 of which used the model as the basis for their prognostic model. The restricted cubic spline varied across studies in the number of interior knots (range 1 to 6), and only three studies reported knot placement. Three studies provided details about the baseline function, with two studies using a figure and the third providing coefficients. However, no studies provided adequate information on their restricted cubic spline to permit others to validate or completely use the model.

CONCLUSIONS

Despite the advantages of the Royston-Parmar model for prognostic models, they are not widely used in health research. Better reporting of details about the restricted cubic spline is needed, so the prognostic model can be used and validated by others.

REGISTRATION

The protocol was registered with Open Science Framework (https://osf.io/r3232/).

Back to the basics of ovarian aging: a population-based study on longitudinal anti-Müllerian hormone decline

BACKGROUND

Anti-Müllerian hormone (AMH) is currently used as an ovarian reserve marker for individualized fertility counseling, but very little is known of individual AMH decline in women. This study assessed whether the decline trajectory of AMH is uniform for all women, and whether baseline age-specific AMH levels remain consistently high or low during this trajectory.

METHODS

A total of 3326 female participants from the population-based Doetinchem Cohort Study were followed with five visits over a 20-year period. Baseline age was 40 ± 10 years with a range of 20-59 years. AMH was measured in 12,929 stored plasma samples using the picoAMH assay (AnshLabs). Decline trajectories of AMH were studied with both chronological age and reproductive age, i.e., time to menopause. Multivariable linear mixed effects models characterized the individual AMH decline trajectories.

RESULTS

The overall rate of AMH decline accelerated after 40 years of age. Mixed models with varying age-specific AMH levels and decline rates provided the significantly best fit to the data, indicating that the fall in AMH levels over time does not follow a fixed pattern for individual women. AMH levels remained consistent along individual trajectories of age, with an intraclass correlation coefficient (ICC) of 0.87. The ICC of 0.32 for AMH trajectories with time to menopause expressed the large variation in AMH levels at a given time before the menopause. The differences between low and high age-specific AMH levels remained distinguishable, but became increasingly smaller with increasing chronological and reproductive age.

CONCLUSIONS

This is the first study to characterize individual AMH decline over a long time period and broad age range. The varying AMH decline rates do not support the premise of a uniform AMH decline trajectory. Although age-specific AMH levels remain consistently high or low with increasing age, the converging trajectories and variance of AMH levels at a given time before menopause shed doubt on the added value of AMH to represent individualized reproductive age.