Diagnostic Considers Antimüllerian Hormone, BMI to Predict Oligoanovulation Disorders

Antimüllerian hormone (AMH) has an important role in regulating folliculogenesis, and it's possible that its overproduction is associated with anovulation. Some reports have linked AMH levels to ovulatory dysfunction and the severity of polycystic ovary syndrome (PCOS), the most common cause of oligoanovulation.

Subscribe to the latest updates from OB/GYN Advances in Motion

Thank you for subscribing!

Error: Please enter a valid email address.

Email Address Submit

Studies have also supported the use of AMH in the diagnosis of oligoanovulation, either alone or as an adjunct to existing criteria. However, AMH levels have yet to be accepted as a diagnostic marker, perhaps because of the variety of commercially available assays.

Irene Souter, MD, fertility specialist at the Fertility Center and Reproductive Endocrinology Unit at Massachusetts General Hospital, and colleagues have developed a model that predicts the probability of an oligoanovulation diagnosis based on an assessment of AMH in combination with body mass index (BMI). In Fertility and Sterility, they provide materials for diagnosing individual patients.

Study Methods

The researchers reviewed data on 1,010 women who underwent intrauterine insemination with or without ovulation induction between May 2011 and March 2019 and did not have diminished ovarian reserve. The patients were stratified into three groups based on their Society for Assisted Reproductive Technology diagnosis:

• PCOS: 228 women (23%)
• Other ovulatory dysfunction disorders (OVDYS): 93 (9%)
• "Other" (other cause of infertility, single women or same-sex couple): 689 (68%)

Correlations Between AMH and BMI

The correlations between serum AMH level and BMI were:

• Overall: r, −0.06; P=0.04
• PCOS: r, −0.19; P=0.004
• OVDYS: r, −0.12; P=0.24
• Other: r, −0.14; P<0.001

After controlling for age, the researchers constructed curves (available in the article) that combine AMH and BMI to predict the oligoanovulation diagnosis. For women with higher BMI, lower AMH was able to predict PCOS diagnosis compared with the AMH values needed for PCOS prediction in normal-weight or underweight women.

Receiver Operating Curve (ROC) Analysis: AMH

Even without taking BMI into consideration, AMH was very good at predicting a PCOS diagnosis:

• In a derivation dataset of 503 patients, the area under the curve (AUC) was 0.86 (P<0.001); a cutoff of AMH ≥6.25 ng/mL (>70th percentile) produced the highest combined sensitivity and specificity
• In the validation dataset, that cutoff was 73% sensitive and 85% specific for a PCOS diagnosis
• In the total population at that cutoff:
  • PCOS vs. OVDYS: OR, 4.71 (P<0.001)
  • PCOS vs. other: OR, 17.20 (P<0.001)

ROC Analysis: AMH + BMI

When the ROC analysis was expanded for each of three different BMI groups, the model retained its very good ability to predict a PCOS diagnosis:

• Overall: AUC, 0.87
• 18.5–24.9 kg/m2: 0.86
• 25.0–29.9 kg/m2: 0.91
• ≥30.0 kg/m2: 0.88

A Tool for Diagnosis and Patient Counseling

This tool is unique because the provided curves and supplemental tables can be used to quantify the probability of anovulatory infertility for individual patients. Expressing AMH as both absolute serum values and percentiles should make the findings applicable to any assay used or population studied.

view original journal article Subscription may be required

Learn more about the Mass General Fertility Center

Refer a patient to the Mass General Fertility Center