Serum estrogen level, attention, memory and other cognitive functions in middle-aged women

To evaluate the relationship between serum estradiol level and cognitive processing efficiency and memory.

Sixty-three healthy women aged 45-65 years were recruited through a newspaper announcement. The subjects were divided into two subgroups (low-estrogen group, n = 37 and high-estrogen group, n = 26) according to their serum follicle stimulating hormone (FSH) and estradiol levels. In the high-estrogen group, estrogen was either endogenous or supplied by estrogen replacement therapy. Automatic and controlled cognitive processing and attentional resources were measured using CogniSpeed software, together with conventional tests of cognitive performance: similarities, digit span, digit symbol, block design, object naming and recall, paired word associates (PWA) recall, Benton visual retention and paced auditory serial addition test (PASAT). The Beck depression inventory was also assessed.

Cognitive reaction speeds were similar in both groups. Women with low estrogen levels made more errors in the vigilance test (sustained attention, p = 0.040). There were no differences in short-term or long-term memory, or verbal, visual or working memory between the study groups. Older women were slower in the ten-choice reaction time (10-CRT) test (r = 0.25, p = 0.047) and made more errors in the test of suppressing attention (Stroop incongruence test; r = 0.34, p = 0.007) and in the sustaining attention test (vigilance test; r = 0.47, p < 0.001). Depression scores did not correlate with cognitive variables.

Cognitive performance was well preserved in healthy middle-aged women. Cognitive speed, accuracy, attentional resources and memory did not show impairment with decline of serum estrogen level in this age group.