This study showed that MHT menopausal women have greater cortical GM volumes than non-MHT women within the same age range (mean, 57 years), which was positively correlated with MHT treatment. With DARTEL-based VBM, the volumes of subcortical structures in MHT women differed from those in noMHT women. In particular, localized GM volumes, including the SFG/MFG/IFG, Hy, ITG, PHG, Hi, Cb, PoG, PCu, AnG, MOs, SOG and PrG were larger in MHT women than those in noMHT women. Furthermore, MHT women showed positive correlations between treatment period and volumes in the AnG and Hy. The GM volume values of the Hy were negatively correlated with the levels of FSH and LH (menstrual cycle-related hormones associated with gonadal function). The GM volume values of the IFG and AnG were negatively correlated with PG levels in MHT women. However, using conventional volume measurement, there were no significant differences between the two groups in total intracranial volume.
Regarding the effect of hormone therapy, previous studies25,2627 reported that estrogen increased the neutral effect or affected structural integrity in brain tissue, especially in the prefrontal, temporal, and parietal lobes. In the study, estradiol levels in pre-menopausal women were significantly higher than in post-menopausal women, while FSH was significantly lower. Estrogen production decreases with age during the transition to menopause; a higher level of estradiol positively affects cognitive function, neuroprotection, and associated brain functions33. ET may influence clinical outcomes through vascular changes or effects on regional brain volume, including neuronal architecture and synaptic density26. A morphological study6 evaluating the effect of menopause-related hormone in post-menopausal women reported that the reduced volumes observed in the superior temporal gyrus, IFG, olfactory cortex, and MOs are closely associated with menopause-related structural changes. In our study, the mean period of MHT is approximately 6 years. The serum estradiol level in MHT women, such as pre-menopausal women, was higher than that of noMHT women, while the FSH level was lower. These hormonal changes in MTH women were similar to previous studies25,2627, demonstrating level values before and after hormone therapy. Compared to noMHT women, the increments in local GM volume observed in MHT women could be associated with protection or suppression of cell loss or/and cell shrinkage related to MHT treatment.
Menopausal women with MHT showed larger GM volumes in key brain regions including of the SFG/MFG/IFG, Hy, ITG, PHG, Hi, Cb, PoG, PCu, AnG, MOs, SOG and PrG compared to noMHT women. ET as a MHT can modulate cell proliferation, differentiation, and survival in brain tissues34; moreover, estrogen treatment significantly increased the density of neurotransmitter serotonin-binding sites in the anterior frontal, cingulate, and olfactory cortex35. Estrogen receptors were ubiquitously localized in the frontal cortex; therefore, the SFG/MFG/IFG, and MOs may be capable of neuroprotection by MHT treatment. Additionally, MHT menopausal women showed larger GM volumes, including the PoG, PCu, and AnG in somatosensory cortices and the PrG and MOs in closely adjacent primary and secondary motor areas. There are differences in the somatosensory cortex by hormone therapy36,37. Khan et al.37 reported that estradiol enhanced rapid increases in dendritic spines in the somatosensory cortex, an area that contributed to working memory, as well as the prefrontal cortex. A previous study38 suggested that hormone therapy and estrogen exposure positively affect the motor system, thus counteracting aging-related reorganization. Thus, these volume changes might help improve the execution of the voluntary movement-related function and the receiving and processing of sensory information from throughout the body by MHT treatment. Moreover, hormone replacement therapy (HRT) was positively associated with better performance on tasks of verbal memory39. A positron emission tomography (PET) study showed evidence that ET (either with or without progestagen addition) modulates longitudinal changes in blood flow of the brain areas involved in cognitive function40. HRT group performing on tasks of verbal and visual memory demonstrated differences in the activation of regions subserving memory (especially the frontal lobe) compared to non-HRT control group41. Also, greater activation of the temporal gyrus, Hi and insula regions, and the PHG and IFG, was observed during verbal and visual memory tasks, respectively, in long-term HRT users (> 2 years) compared to non-HRT users40. Therefore, cerebral volume changes by MHT are closely associated with the alteration of brain blood flow and brain function.
Interestingly, MHT women showed greater hypothalamic volumes. The Hy, which is located in the center of the brain between the pituitary gland and the thalamus42, plays an important role in the production of hormones such as FSH, LH, dopamine, vasopressin, and gonadotropin-releasing hormone (GnRH); and helps to stimulate many important processes while maintaining the stable state of the body (homeostasis), such as body temperature, emotions, sleep cycles, sexual drive, blood pressure, and heart rate43. Especially in the feedback mechanism for ovarian hormone production44, the secretion of female sex hormones starts with the secretion of a GnRH in the Hy of the brain. GnRH stimulates the pituitary to release two gonadotropins, FSH and LH. Released FSH and LH promote ovulation and stimulate secretion of the sex steroids estradiol (the most common estrogen, E2) and PG. Positive and negative feedback by E2 and PG acts on FSH and LH production as follows: high levels of E2 and PG stimulate FSH and LH secretion, and the low levels of E2 and PG cause negative feedback to suppress FSH and LH secretion. In the present study, MHT-treated women had lower levels of FSH and LH than noMHT women (as Table 1), and their hypothalamic volume is negatively correlated with FSH and LH levels, as shown in Fig. 3. In old reproductive-age women as noMHT menopausal women, the patterns of hormone secretion with decreased ovarian function show significant alterations of hypothalamic-pituitary feedback mechanisms by menopause43. The study showed relative hypothalamic-pituitary insensitivity to estrogen in aging menopausal women manifested by positive and negative feedback mechanisms (from FSH and LH levels). In this study, both groups enrolled are age-matched and BMI-matched. Hypothalamic hormonal changes (lower FSH/LH and higher estradiol levels like premenopausal women) reflected the improvement in hypothalamic pituitary sensitivity by MHT treatment; therefore, these findings may support the idea that MHT treatment improves menopausal symptoms that disrupt quality of life in menopausal women.
This study included several limitations. First, this study used the DARTEL-based VBM method to analyze the GM difference between MHT and noMHT groups. The DARTEL method can be replaced by ‘Geodesic Shooting,’ which is supposedly more mathematically correct45. Further study will be performed to clarify the difference between DARTEL and Geodesic Shooting methods for assessing GM difference associated with MHT. Second, sex hormones, aging, and metabolic status are important factors in brain volume changes in post-menopausal women that may interact. To overcome aging and metabolic status-related morphometric changes, we directly compared age-matched and BMI-matched post-menopausal women as the MHT and noMHT groups. Our study focused on GM volume in response to MHT; more studies are needed to evaluate changes in WM volume to understand the effect of MHT, particularly with the addition of diffusion tensor imaging, which might help assess MHT-related morphometric alterations. The information of other tissues such as WM, and CSF could be available for understanding MHT-related differences. Further study will be performed to clarify WM and CSF difference by MHT. Third, MHT treatment plays a protective role in cognitive decline; however, we did not measure cognitive function using assessment tools, such as the Mini-Mental State Examination or other intelligence tests. Despite such limitations, we showed MHT-related morphometric differences in specific areas of the brain, suggesting a neuroprotective effect of hormone therapy in the cerebral cortex.
In conclusion, MHT-treated women might have larger GM volumes compared to menopausal women without MHT. The anatomical structures that showed greater volume in association with MHT included the deep brain areas (Hy, PHG, Hi), cerebellar cortex, frontal (SFG/MFG/IFG, MOs, PrG), inferior temporal, parietal (PoG, PCu, AnG ), and superior occipital gyri. This study found a correlation between the levels of sex hormones and localized brain volumes following MHT. These findings would help us understand the interaction between brain volumes and levels of sex hormones following menopause and MHT.