Comparison of clinical and hormonal characteristics among four phenotypes of polycystic ovary syndrome based on the Rotterdam criteria.
Level 4 - case-series / case-control
Observational case-control study
PubMed 26408006 · doi:10.1007/s00404-015-3889-5
What was done
An observational case-control study compared clinical parameters (BMI, waist circumference) and hormonal profiles (AMH, LH/FSH ratio, total testosterone, FSH, SHBG, and free androgen index) between 263 women diagnosed with PCOS according to Rotterdam criteria and 263 healthy controls. Women with PCOS were categorized into four phenotypes: Phenotype A (oligo-anovulation + hyperandrogenism + polycystic ovary morphology), Phenotype B (oligo-anovulation + hyperandrogenism), Phenotype C (hyperandrogenism + polycystic ovary morphology), and Phenotype D (oligo-anovulation + polycystic ovary morphology).
What was found
Phenotype A presented the most severe profile, with significantly higher waist circumference than phenotype D, higher BMI than phenotype C, and higher total testosterone than phenotypes B, C, D, and controls. Phenotype A also had a higher LH/FSH ratio than phenotype D and controls. AMH levels were highest in phenotype A; phenotypes A, C, and D had significantly higher AMH compared to phenotype B and controls (phenotype B AMH levels were low and similar to controls). Compared to phenotype A, phenotype D and controls had significantly lower LH/FSH ratios, total testosterone, and free androgen index, and higher FSH and SHBG levels (P < 0.001). In logistic regression, AMH and LH were identified as independent predictors of PCOS. Exact numerical values, standard deviations, and odds ratios were not reported in the abstract.
Why it matters
The findings highlight the biological heterogeneity of PCOS under the Rotterdam criteria, indicating that patients with phenotype A carry the most pronounced endocrine and metabolic disturbances.
Limits
The observational case-control design cannot establish causality or track metabolic trajectory over time. The abstract omits exact point estimates, numerical spread, confidence intervals, and demographic details of the study population. Potential confounding factors such as age matching and metabolic comorbidities were not detailed in the abstract.
Cited by
- supports PCOS has four distinct phenotypes.