Menopausal hormone therapy does not improve some domains of memory: A systematic review and meta-analysis

Abstract

Background: Aged women appear to be at a higher risk of developing memory impairment than men. Whether menopausal hormone therapy (MHT) could improve memory in postmenopausal women remains unclear. We thus conducted a meta-analysis to investigate the potential effect of MHT on memory, especially verbal memory, in postmenopausal women.

Methods: PubMed, Cochrane, Embase, Chinese Biomedical Literature Database, and web of ClinicalTrials.gov were systematically searched for randomized controlled trials comparing MHT versus placebo in postmenopausal women. Our primary outcome of interest is memory function.

Results: We included 10 studies with 2,818 participants in the final analysis. There was no significant differences in immediate recall (weighted mean difference [WMD] 0.34, 95% confidence interval [CI]: -0.73, 1.40), delayed recall (WMD 0.99, 95% CI: -0.51, 2.48), short-delay (WMD -0.00, 95% CI: -0.37, 0.37), and long-delay (WMD -0.19, 95% CI: -0.69, 0.31) recall between WMT and placebo. WMT was associated with a lower digit span forward (mean reduction -0.20, 95% CI: -0.36, -0.03). In women within 5 years of menopause, MHT did not differ in immediate (0.45, 95% CI: -0.75, 1.65) or delayed recall (1.03, 95% CI: -0.93, 3.00), and digit span forward (-0.11, 95% CI: -0.72, 0.50), when compared with placebo.

Conclusion: This meta-analysis suggested that MHT had no effect on verbal memory in postmenopausal women, and may impair some domains of short-term memory. Current available evidence does not support MHT for improving memory in women less than 60 years, even in recently menopausal women.

Systematic review registration: https://www.crd.york.ac.uk/PROSPERO, identifier CRD42021233255.

Keywords: cognition; memory; menopausal hormone therapy; menopause; meta-analysis; systematic review; women.

Figure 1

Flow diagram of study selection process.

Figure 2

Meta-analyses of MHT versus placebo on comparing memory. Outcomes assessed are as follows: immediate and delayed recall of logical memory (A), short- and long-delayed recall from CVLT (B), and digit span including forward, backward, and total scores (C). Log Mem Immed, immediate recall of logical memory; Log Mem delay, delayed recall of logical memory.

Figure 3

Sensitivity analyses of MHT versus placebo on comparing memory. Outcomes assessed are as follows: immediate and delayed recall of logical memory (A), short- and long-delayed recall from CVLT (B), and digit span including forward and backward (C).

Figure 4

Subgroup analyses of MHT versus placebo on comparing memory according to the length of menopause. Outcomes assessed are as follows: immediate recall of logical memory (A), delayed recall of logical memory (B), and digit span forward (C).

Figure 5

Subgroup analyses of MHT versus placebo on comparing memory stratified by the formulation of HT. Outcomes assessed are as follows: immediate recall of logical memory (A), delayed recall of logical memory (B), short delayed recall from CVLT (C), long-delayed recall from CVLT (D), digit span forward (E), and digit span backward (F).