TABLE 1.
Primary RCTs that assess the role of estrogen and cognitive function in postmenopausal women1
| Reference | Study design | Population | Intervention | Control | Duration | Outcome measure(s) | Outcome assessment(s) | Results |
|---|---|---|---|---|---|---|---|---|
| Grady et al. (2002) (89) | RCT (parallel), HERS | Postmenopausal women (age 71 ± 6 y) with coronary disease | 0.625 mg/d CEE + 2.5 mg/d MPA (n = 517) | Placebo (n = 546) | 4.2 ± 0.4 y | General cognitive function | MMSE, verbal fluency, Boston Naming, Word List Memory, Word List Recall and Trails B | No observed differences in age-adjusted cognitive test scores except estrogen + progestin scored worse on verbal fluency test (15.9 ± 4.8 vs. 16.6 ± 4.8, P = 0.02) |
| Shumaker et al. (2003) (38) | RCT (parallel), WHIMS | Postmenopausal women free of probable dementia, age 65+ y at baseline | 0.625 mg/d CEE + 2.5 mg/d MPA (n = 2229) | Placebo (n = 2303) | 4.05 ± 1.19 y | Incidence of probable dementia and MCI | MMSE | 61 women diagnosed with probable dementia, 40 in estrogen + progestin group vs. 21 in placebo group. HR for probable dementia was 2.05 (95% CI, 1.21, 3.48). Treatment effects on MCI did not differ between groups |
| Shumaker et al. (2004) (37) | RCT (parallel), WHIMS | Postmenopausal women free of probable dementia (age 65–79 y) at baseline | 0.625 mg/d CEE-alone or 0.625 mg/d CEE + 2.5 mg/d MPA (n = 5157) | Placebo (n = 5269) | 5.21 ± 1.73 y (CEE-alone) and 4.05 ± 1.19 y (CEE + MPA) | Incidence of probable dementia and MCI | MMSE | Women diagnosed with probable dementia: 28 in estrogen vs. 19 in placebo group. HR for probable dementia: 1.49 (95% CI: 0.83, 2.66); 76 women diagnosed with mild cognitive impairment in estrogen vs. 58 in placebo group. HR for MCI: 1.34 (95% CI: 0.95, 1.89); 178 women diagnosed with probable dementia or mild cognitive impairment in combined estrogen-alone or estrogen + progestin group vs. 132 in placebo group. HR for probable dementia or MCI: 1.41 (95% CI: 1.12, 1.76) |
| Dumas et al. (2006) (52) | RCT (crossover) | Healthy cognitively normal, older women (age 48–84 y) at baseline | 1 mg/d 17-β estradiol (n = 15) | Placebo (n = 15) | 3 mo | Subjects performed cognitive task during 5 anticholinergic challenge sessions and were administered 1 of 2 acute doses scopolamine, mecamylamine, or placebo | Cognitive battery included tests of attention (CFFT, CRT, DSST, DAT, CCPT), verbal learning and memory (BSRT, VPAT, and IPR), and nonverbal learning and memory (RAT and BVRT) | Estrogen pretreatment attenuated anticholinergic drug–induced impairments on tests of attention and tasks with speed components |
| Resnick et al. (2006) (36) | RCT (parallel), WHIMS | Postmenopausal women free of probable dementia, age 65–79 y at baseline | 0.625 mg/d CEE + 2.5 mg/d MPA (n = 690) | Placebo (n = 726) | Mean 3 y treatment and followed for a mean of 1.35 ± 0.61 y | Annual rates of change in specific cognitive functions (mainly verbal and figural memory), adjusted for time since randomization | WHISCA cognitive battery | Estrogen + progestin had a negative impact on verbal memory (P ≤ 0.01) and positive impact on figural memory (P = 0.01), compared to placebo. Both effects on memory were evident only after long-term therapy. Other cognitive domains were not significantly affected |
| Dumas et al. (2008) (53) | RCT (crossover) | Healthy younger (age 50–62 y) and older (age 70–84 y) postmenopausal women at baseline | 1 mg/d 17-β estradiol for 1 mo then 2 mg/d for 2 mo (n = 22) | 1 mg of 17-β estradiol for 1 mo then placebo for 2 mo (n = 22) | 3 mo | Verbal memory performance and attention after 3 anticholinergic challenge sessions, where subjects were administered an acute dose of scopolamine, mecamylamine, or placebo | Cognitive battery composed of: BSRT, IPR, CFFT, CRT, and CCPT | Estradiol pretreatment significantly attenuated anticholinergic drug-induced impairments on a test of episodic memory (i.e., BSRT) for younger group only, while estradiol treatment impaired performance of older group; study suggests younger subjects may experience greater benefit from estradiol treatment vs. older subjects, supporting concept of a critical period for postmenopausal estrogen use |
| Gleason et al. (2015) (34) | RCT (parallel), KEEPS–Cog | Recently postmenopausal women (age 52.6 ± 2.6 y) | 0.45 mg/d oral CEE + 200 mg/d micronized progesterone (n = 220) or 50 μg/d transdermal estradiol + 200 mg/d micronized progesterone (n = 211) | Placebo (n = 262) | 4 y with 2.85 ± 0.49 y follow-up | Verbal learning and memory, auditory attention and working memory, visual attention and executive function, and speeded language and mental flexibility | MMSE | No treatment related benefits found for cognitive outcomes |
| Henderson et al. (2016) (35) | RCT (parallel) | Healthy women (age 41–84 y) were recruited into early and late postmenopause groups | 1 mg/d of 17-β estradiol (n = 284) | Placebo (n = 283) | 57 ± 5.8 mo | Verbal memory | Comprehensive neuropsychological battery (table e-1 on Neurology® website) that emphasized standardized tests sensitive to age-related change | Estradiol did not affect measures of verbal memory, executive functions, or global cognition vs. placebo. Estradiol initiated within 6 y of menopause vs. 10+ y after does not differently affect verbal memory, executive functions, or global cognition |
| Espeland et al. (2017) (90) | Postintervention follow-up of 2 RCT (parallel) cohorts: WHIMSY and WHIMS-ECO | Younger (age 50–54 y) and older (age 65–79 y) postmenopausal women | 0.625 mg/d CEE in women with hysterectomy or 0.625 mg/d CEE + 2.5 mg/d MPA in women without hysterectomy (n = 2,103) | Placebo (n = 2113) | Follow-up mean 7.2 ± 1.0 y for WHIMSY and 6.4 ± 1.0 y for WHIMS-ECO | Global cognitive functioning, verbal memory, attention, and semantic verbal fluency | Cognitive battery composed of: TICS–m, EBMT, OTMT, and VF–A | Hormone therapy when prescribed to younger women had no significant effect on long-term cognition, but was associated with reduced global cognitive function, working memory, and executive function in older women |
| Manson et al. (2017) (39) | Postintervention follow-up of 2 RCT (parallel) cohorts: WHI and WHI–OS | Postmenopausal women (age 63.4 ± 7.2 y) | CEE 0.625 mg/d alone or + 2.5 mg MPA (n = 13,816) | Placebo (n = 13,531) | Cumulative 18-y follow-up | AD or dementia mortality | Regular surveillance of cohort through National Death Index and by reports of next of kin or postal service | Estrogen-alone reduced morality risk from AD or dementia vs. placebo, HR: 0.74 (95% CI: 0.59, 0.94). Estrogen + progestin no effect vs. placebo, HR: 0.93 (95% CI: 0.77, 1.11). Pooled results from 2 cohorts |
| suggest hormone therapy may reduce mortality risk from AD or dementia, HR: 0.85 (95% CI: 0.74, 0.98) | ||||||||
| Yoon et al. (2018) (91) | RCT (parallel) | Postmenopausal women (age 52–82 y) with multiple-domain, amnestic subtype MCI | 0.1%, 2 mg/d percutaneous estradiol gel (n = 19) | Placebo (n = 18) | 24 mo | General cognitive function | ADAS and Korean versions of MMSE and MCA | Dementia progression rates 44.4% (8/18) in estradiol group vs. 52.9% (9/17) in placebo group. Estradiol treatment reduced global cognition deterioration in MCI when adjusted for apoE genotype (ε4 allele) (P = 0.0261) and resulted in better cognitive battery scores after 24 mo vs. placebo on both Korean versions of MCA (MD: 3.85; 95% CI: –0.46, 8.16; P = 0.043) and MMSE (MD: 3.26; 95% CI: 0.04, 6.48; P = 0.0319) |
| Moradi et al. (2019) (92) | RCT (parallel) | Postmenopausal women (age 55.97 ± 4.92 y) free of chronic disease | 0.625 mg/d CEE + 2.5 mg/d MPA + 500 mg/d calcium and 200 IU/d vitamin D tablet (n = 70) | 500 mg/d calcium and 200 IU/d vitamin D tablet (n = 70) | 4 mo | General cognitive function and climacteric symptoms | MCA and GCS | All MCA domains except orientation (i.e., visuospatial/executive, memory, attention, language, abstraction, total MCA score GCS improved in estrogen and progestin group vs. control group (P < 0.001) MCA and GCS negatively correlated after intervention (r = –0.235; P = 0.006) |
1Values presented as means ± SDs unless otherwise indicated. AD, Alzheimer's disease; ADAS, Alzheimer's Disease Assessment Scale; BSRT, Buschke Selective Reminding Task; BVRT, Benton Visual Retention Test; CEE, Conjugated Equine Estrogens; CFFT, Critical Flicker Fusion Task; CCPT, Connors Continuous Performance Test; CRT, Choice Reaction Time task from the Milford Test Battery; DDST, Digit Symbol Substitution Test; DAT, Divided Attention Test; EBMT, East Boston Memory Test; GCS, Green Climacteric Scale; HERS, Heart and Estrogen/Progestin Replacement Study; IPR, New York University (NYU) Immediate Paragraph Recall; KEEPS–Cog, Kronos Early Estrogen Prevention Study–Cognitive Affective Study; MCA, Montreal Cognitive Assessment; MCI, mild cognitive impairment; MMSE, Mini-Mental State Examination; MPA, medroxyprogesterone acetate; OTMT, Oral Trail Making Test; RAT, Repeated Acquisition Task; TICS–m, Telephone Interview for Cognitive Status–modified; VF–A, Verbal Fluency–Animals; VPAT, Verbal Paired Associates Test; WHI, Women's Health Initiative; WHIMS, Women's Health Initiative Memory Study; WHIMS–ECHO, Women's Health Initiative Memory Study–Epidemiology of Cognitive Health Outcomes; WHIMSY, Women's Health Initiative Memory Study of Younger Women; WHI–OS, Women's Health Initiative–Observational Study.