Cognition in perimenopause: the effect of transition stage

Abstract

Objective: The aims of this cross-sectional study were to determine if cognitive function differs across stages of reproductive aging and to evaluate whether hormones or menopausal symptoms predict cognition in perimenopause. We hypothesized that women in late menopausal transition and early postmenopause would perform more poorly than those in the late reproductive stage on attention and verbal memory tasks, and that estradiol, depressive symptoms, anxiety symptoms, hot flashes, and sleep disturbance would predict cognitive performance on those tasks.

Methods: One hundred seventeen middle-aged women enrolled in the Rochester Investigation of Cognition Across Menopause were categorized into late reproductive stage (n = 34), early menopausal transition stage (n = 28), late menopausal transition stage (n = 41), or early postmenopause stage (n = 14) according to criteria from the Stages of Reproductive Aging Workshop +10. We administered a neuropsychological battery assessing six domains of cognition, assessed menopausal symptoms, and measured serum levels of estradiol and follicle-stimulating hormone. Multivariate regressions were conducted to determine the impact of menopausal stage and symptoms on cognition.

Results: Women in the first year of postmenopause performed significantly worse than women in the late reproductive and late menopausal transition stages on measures of verbal learning, verbal memory, and motor function. They also performed significantly worse than women in the late menopausal transition stage on attention/working memory tasks.

Conclusions: Cognitive function does not change linearly across perimenopause. Decreases in attention/working memory, verbal learning, verbal memory, and fine motor speed may be most evident in the first year after the final menstrual period.

Figure 1

Adjusted z-scores and standard errors as a function of menopausal transition stage for verbal learning, verbal memory, fine motor skills, and attention/working memory cognitive composites. * p<0.05.Analyses adjusted for age, education, vasomotor symptoms, sleep symptoms, depression, anxiety, estradiol, and FSH. Consequently, adjusted z-scores will not add to 0.