ORF36-03 - Mortality in Women with Primary Ovarian Insufficiency, Early Menopause and Turner Syndrome

Background: Previous epidemiology and large database studies suggest that primary ovarian insufficiency (POI) and early menopause are associated with increased mortality. However, these studies have not examined mortality risk on the basis of POI etiology. We therefore examined mortality in women with POI and in women with Turner syndrome to determine whether etiology plays a role in early mortality.

Methods: Women with POI (≤40 years), early menopause (>40 and < 45 years) and Turner syndrome were identified using ICD9 and 10 codes in electronic medical records (1995-2021) from two major healthcare systems in Utah and reviewed for accuracy. Controls were chosen 10:1 to match age, sex and birth inside or outside of Utah. Identified women were linked to their relatives using genealogy information and all subjects were linked to mortality information from death certificates through the Utah Population Database (UPDB). Mortality was compared using hazard ratios.

Results: We identified 518 women with POI and 166 women with early menopause from the University of Utah and Intermountain Healthcare. These women with POI and early menopause were diagnosed at age 35.3±7.8 years and were currently 50.1±13.0 years. We identified 390 women with confirmed Turner syndrome. Their ages were 8.7±8.3 years at diagnosis and were currently 32.0±16.6 years. The karyotype was not available for the majority of Turner syndrome subjects (n=175). The rest were documented 45,X, mosaic 45,X/46,XX, 45,X/47,XXX, 45,X/46,XY, and other.
Mortality was increased in women with Turner syndrome compared to controls (hazard ratio [95% confidence interval]; 2.59 [1.89, 3.54]; p< 3.0X10-9). There was no difference in mortality in women with POI compared to controls (0.98 [0.72, 1.34]; p=0.9). There was no interaction with smoking. Parity decreased mortality risk in women with Turner syndrome (0.58 [0.37,0.92]; 0.02), likely because these women had mosaic karyotypes with potentially fewer affected tissues. There was also no difference in the mortality of the relatives compared to controls. Limitations included more readily identified Turner syndrome, failure to seek healthcare for POI and possible death before presenting with POI.

Conclusion: Our data suggest that earlier mortality in women with POI may derive from a Turner syndrome etiology. These studies suggest that previous epidemiology results may be influenced by the Turner syndrome subset of POI.