CHARACTERIZING PRIMARY AMENORRHEA IN A LARGE DIVERSE COHORT

Abstract

To our knowledge, no recent large studies in the US have reexamined the etiologies of primary amenorrhea; particularly, in the context of increasing obesity rates and diverse ethnic backgrounds. The purpose of this study is to identify and characterize underlying causes of primary amenorrhea amongst a large, diverse cohort. We performed a retrospective chart review of the Kaiser Permanente Southern California database for patients with primary amenorrhea using ICD9 and ICD10 diagnosis codes 626.0 and N91.0, selecting for females ages 0-21 years old at first diagnosis from 2008-2019. Charts were then reviewed to determine which patients met diagnostic criteria for primary amenorrhea, as defined by the American Academy of Pediatrics (AAP): lack of menarche by age 15 years old or by >3 years after onset of secondary sexual characteristics. BMI-for-age percentile (BMIPCT) was used per CDC and AAP guidelines for patients ages 2-19 to account for differences in body composition with sex and age. Medical records were reviewed for demographics and clinical characteristics. Data were analyzed using ANOVA and Tukey analysis; p<0.05. Out of 936 patients from initial search, 748 were diagnosed with primary amenorrhea by chart review. Of those diagnosed, 45% identified as Hispanic (n=340), 32% as White (n=238), 10% as Black (n=73), and 7% as Asian (n=51). The remaining 6% (n=46) identified as Native American/Alaskan, Pacific Islander, Multiple, Other, or Unknown were excluded from final statistical analyses due to small sample size. Within ethnicity groups, patients were normally distributed by age. There was no significant difference in age of presentation between ethnicity groups. BMIPCT ranged from <1 to >99 with a median percentile of 69. The mean BMIPCT was 71 for Hispanic, 46 for White, 63 for Black, and 45 for Asian patients. There was a significant difference for both BMI and BMIPCT when comparing Hispanic vs. White (P<.001) and Hispanic vs. Asian patients (P<.001). A similar trend was seen comparing Black vs. Asian (P<.001) and Black vs. White patients (P<.001). There was no significant difference comparing Hispanic vs. Black, nor White vs. Asian patients. Hispanic and Black populations have significantly higher BMI and BMI-for-age percentiles at time of primary amenorrhea diagnosis when compared to White and Asian populations. However, there is no difference in age of presentation between ethnicity groups. Traditionally, the most common diagnosis underlying primary amenorrhea is hypogonadotropic hypogonadism or gonadal dysgenesis, followed by Mullerian anomaly, physiologic delay, and polycystic ovarian syndrome. Over the past 10 years, limited research in Europe and Asia evaluating differing ethnicities suggest alternative etiologies for primary amenorrhea.