Hippocampus. 2025 May;35(3):e70012. doi: 10.1002/hipo.70012.
ABSTRACT
Early midlife bilateral salpingo-oophorectomy (BSO) is associated with greater Alzheimer’s disease risk compared to spontaneous/natural menopause. Previously, we found that participants with BSO had lower volume in the hippocampal dentate gyrus and cornu ammonis 2/3 composite subfield (DG-CA2/3). We sought to extend those hippocampal subfield findings by assessing whether BSO affected volumes along the anteroposterior hippocampal axis, anterolateral entorhinal cortex, and perirhinal cortex subregions (Brodmann area (BA) 35 and 36). We also correlated volumes with key demographic and wellbeing-related factors (age, depressive mood, education), hormone therapy characteristics, and recognition memory performance. Early midlife participants with BSO (with and without 17β-estradiol therapy (ET)) and age-matched control participants with intact ovaries (AMC) completed high-resolution T2-weighted structural magnetic resonance imaging (MRI). Medial temporal lobe volumes and Remember-Know task recognition memory performance were compared between groups-BSO (n = 23), BSO + ET (n = 28), AMC (n = 34) using univariate analyses. Multivariate Partial Least Squares (PLS) analyses were used to examine how volumes related to demographic and wellbeing-related factors, as well as hormone therapy characteristics. Relative to BSO + ET, BSO had lower posterior hippocampal and DG-CA2/3 volumes but greater perirhinal BA 36 volumes. Compared to age, depressive mood, and education, ET was the strongest positive predictor of hippocampal volumes and negative predictor of perirhinal BA 36 volumes. For BSO + ET, hippocampal volumes were negatively related to ET duration and positively related to concurrent progestogen therapy. Relative to AMC, BSO had greater anterolateral entorhinal cortex and perirhinal BA 35 and BA 36 volumes. BSO groups (with and without ET) relied more on familiarity than recollection for successful recognition memory. BSO and ET may have distinct effects on medial temporal lobe volumes, with potential implications for memory processes affected by Alzheimer’s disease risk.
PMID:40156318 | DOI:10.1002/hipo.70012
