Associations of cyst size, bilaterality, and age with serum anti-Müllerian hormone levels in women with ovarian endometriomas: a retrospective cohort study of 321 patients

Anti-Müllerian hormone (AMH) is a widely used marker of ovarian reserve. Although ovarian endometrioma (OMA) has been associated with reduced ovarian reserve in prior studies, the relationship between lesion characteristics and AMH among women with OMA remains uncertain. Cyst size is a readily available imaging parameter that often contributes to clinical assessment, particularly when fertility preservation and the potential ovarian impact of intervention are considered in young women. This study evaluated the associations of cyst laterality, total cyst diameter, total cyst volume, and age with AMH levels. We performed a single-center retrospective observational study of 321 women aged 18–35 years with ovarian endometriomas, including 203 with unilateral and 118 with bilateral disease. Associations with AMH were evaluated using Spearman rank correlation with bootstrap 95% confidence intervals and Mann–Whitney U tests. Because AMH was right-skewed, multivariable linear regression models used natural-log-transformed AMH as the dependent variable; age, laterality, and either total cyst diameter or log-transformed total cyst volume were included as covariates. Among 321 patients, the median age was 30.0 years (IQR 27.0–33.0), median total cyst diameter was 7.90 cm (IQR 5.40–10.70), and median AMH was 2.436 ng/mL (IQR 1.365–3.650). AMH was inversely correlated with age (Spearman ρ=-0.284, 95% CI -0.384 to -0.177, P < 0.001), whereas neither total cyst diameter (ρ=-0.021, 95% CI -0.129 to 0.086, P = 0.711) nor total cyst volume (ρ = 0.025, 95% CI -0.090 to 0.132, P = 0.660) was significantly correlated with AMH in the overall cohort. Consistent with this overall pattern, patients with bilateral OMA had a substantially greater cumulative cyst diameter than those with unilateral OMA (10.70 vs. 6.50 cm, P < 0.001), yet AMH levels did not differ significantly between the groups (2.37 vs. 2.46 ng/mL, P = 0.455). In multivariable models, age remained independently associated with lower log-transformed AMH ( P < 0.001), whereas laterality, total diameter, and total volume were not statistically significant. In this cohort of women with OMA, age showed a consistent inverse association with AMH, whereas no statistically significant independent associations were observed for cyst diameter, volume, or bilaterality. These findings suggest that cyst size alone may be an incomplete surrogate for ovarian reserve and support interpreting lesion dimensions together with AMH, age, fertility intentions, and other clinical characteristics when individualized management is considered. These findings should not be interpreted as evidence that OMA has no effect on ovarian reserve, because the study did not include a non-OMA control group and very small associations cannot be excluded.