Association of serum uric acid with women's ovarian reserve: observational study and Mendelian randomization analyses.

OBJECTIVE:To investigate the association between serum uric acid and women's ovarian reserve. DESIGN:Retrospective observational study and Mendelian randomization study. SETTING:University-affiliated in vitro fertilization center. PATIENTS:Observational analyses were undertaken using data from 8,257 women with infertility who finished their first in vitro fertilization treatments between May 2017 and December 2021. Mendelian randomization analyses were based on genome-wide association summary statistics from several biobanks of predominantly European ancestries. INTERVENTIONS:Observational study involved testing log2 transformed serum uric acid levels (for linear, negative regression, and logistic regression analyses); original uric acid levels (for nonlinear association analyses). Mendelian randomization study involved testing genetically predicted uric acid levels. MAIN OUTCOME MEASURES:Biomarkers including antimüllerian hormone, basal antral follicle count, follicle-stimulating hormone, luteinizing hormone, ratio of follicle-stimulating hormone to luteinizing hormone, estradiol; indices of ovarian response to stimulation including poor ovarian response according to different criteria and oocyte yield. RESULTS:In retrospective observational study, all ovarian reserve-related outcomes demonstrated significant differences across serum uric acid quartiles. A two-fold uric acid increase was associated with increased antimüllerian hormone (adjusted β = 0.69; 95% confidence interval [CI], 0.43-0.95), antral follicle count (adjusted incidence rate ratio = 1.10, 95% CI, 1.05-1.14), luteinizing hormone (adjusted β = 0.53, 95% CI, 0.28-0.78), decreased risks of Bologna poor ovarian response (adjusted odds ratio = 0.97; 95% CI, 0.95-0.99) and groups 2-4 Poseidon poor ovarian response (group 2: 0.63, 0.56-0.71; group 3: 0.71, 0.65-0.78; group 4: 0.50, 0.46-0.55), whereas an increased risk of group 1 (1.26, 1.13-1.41). Nonlinear analyses showed a common inflection point at 320-340 μmol/L of uric acid. Interactions between uric acid and antimüllerian hormone and antral follicle count were presented in association with oocyte yield. Mendelian randomization results suggested a significant association between genetically predicted uric acid levels and antimüllerian hormone levels (β = 0.08; 95% CI, 0.04-0.12) but none for uric acid in relation to polycystic ovarian syndrome or other related hormones. CONCLUSION:Higher uric acid levels were associated with better ovarian reserve and increased levels of antimüllerian hormone albeit an increased risk of unexpected poor ovarian response.