Climacteric

ISSN: (Print) (Online) Journal homepage: https://www.tandfonline.com/loi/icmt20

Menopause and cognitive complaints: are ovarian

hormones linked with subjective cognitive decline?

R. Reuben, L. Karkaby, C. McNamee, N. A. Phillips & G. Einstein

To cite this article: R. Reuben, L. Karkaby, C. McNamee, N. A. Phillips & G. Einstein (2021):
Menopause and cognitive complaints: are ovarian hormones linked with subjective cognitive
decline?, Climacteric, DOI: 10.1080/13697137.2021.1892627

To link to this article: https://doi.org/10.1080/13697137.2021.1892627

Published online: 15 Mar 2021.

Submit your article to this journal

Article views: 94

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=icmt20
CLIMACTERIC
https://doi.org/10.1080/13697137.2021.1892627

REVIEW

Menopause and cognitive complaints: are ovarian hormones linked with

subjective cognitive decline?
R. Reubena , L. Karkabya,b, C. McNameea, N. A. Phillipsc and G. Einsteina,b,d,e
a
Department of Psychology, University of Toronto, Toronto, ON, Canada; bTema Genus, Linko€ping University, Link€oping, Sweden;
c
Department of Psychology, Concordia University, Montreal, QC, Canada; dDalla Lana School of Public Health, University of Toronto,
Toronto, ON, Canada; eRotman Research Institute, Baycrest, Toronto, Ontario, Canada

ABSTRACT ARTICLE HISTORY

Subjective cognitive decline (SCD) and the loss of ovarian hormones after menopause have been inde- Received 25 September 2020
pendently linked to later-life Alzheimer’s disease (AD). The objective of this review was to determine Revised 8 February 2021
whether menopause and the loss of ovarian hormones contribute to cognitive complaints and SCD in Accepted 12 February 2021
women. This would suggest that SCD at the menopausal transition might be an important marker of Published online 15 March
2021
eventual cognitive decline and AD. We conducted a literature search using PubMed, PsycINFO and
Web of Science in July 2020. All English-language studies assessing SCD and cognitive complaints with KEYWORDS
respect to menopause and ovarian hormones were included. A total of 19 studies were included. Menopause; ovarian
Studies found that cognitive complaints increased across the menopause transition and were associ- hormones; estrogens;
ated with reductions in attention, verbal and working memory, and medial temporal lobe volume. cognitive complaints;
Women taking estrogen-decreasing treatments also had increased cognitive complaints and reduced subjective cognitive decline;
working memory and executive function. The current literature provides impetus for further research Alzheimer’s disease
on whether menopause and the loss of ovarian hormones are associated with cognitive complaints
and SCD. Clinicians may take particular note of cognitive complaints after menopause or ovarian hor-
mone loss, as they might presage future cognitive decline.

Introduction may be especially relevant when predicting eventual cogni-

tive decline in women.
Recent research has identified subjective cognitive decline
Women frequently report cognitive complaints in postme-
(SCD) as a potential warning sign for progressing toward
nopause [13]. Although concerns about cognition can be a
future cognitive decline and Alzheimer’s disease (AD) [1].
benign part of the menopausal transition and aging, cogni-
SCD is defined as self-experienced cognitive decline in the
tive complaints may also be the harbingers of later demen-
presence of normal performance on standardized cognitive
tia. A key transition in menopause is the lowered production
tests that cannot be explained by any neurologic or psychi-
of hormones produced principally by the ovaries, including
atric disease, medical disorder, medication or substance use
[1]. SCD is associated with an increased risk of AD [2] and estrogens [14]. Many studies demonstrate that estrogens
objective measures of cognitive decline, such as longitudinal modulate neurogenesis and synaptic plasticity, interact with
cognitive decrements [3], differences in neuronal glucose neurotransmitter systems and maintain cognitive function
metabolism [4] and cortical thinning in the medial temporal and brain health [15,16]. In oophorectomized rats, estradiol
lobe (MTL) and frontal cortex [5–7]. administration leads to rapid and reversible increases in the
Cognitive complaints, the hallmark of SCD, may be par- potentiation of CA1 hippocampal neurons [17] and increases
ticularly important to study in women. Two-thirds of individ- in choline acetyltransferase (acetylcholine enzyme; neuro-
uals with AD are women, and by age 45 years the lifetime transmitter involved in memory) in the frontal cortex and
risk for AD is twice as high in women as in men [8]. This hippocampus [18]. Spontaneously (i.e. naturally) menopausal
increased prevalence goes beyond demographic–survival dif- women have decrements in verbal memory, working mem-
ferences and indicates female sex as a risk factor for AD [9]. ory and executive function compared to perimenopausal and
While SCD increases the risk of progressing to AD in both premenopausal women [13], as well as declines in frontal
sexes [2], studies that differentiate by sex show that SCD pre- cortical volumes that may be ameliorated with estradiol ther-
dicts worse verbal memory performance [10] and increased apy [19].
AD risk in women only [11]. Further, the presence of cogni- Other types of menopause, such as surgical menopause
tive complaints in women aged 65 years and older predicts due to oophorectomy (ovarian removal) or induced meno-
an increased risk of cognitive impairment 18 years later [12]. pause due to cancer treatments, may provide human models
This evidence suggests that cognitive complaints and SCD of ovarian hormone loss or dysregulation. Women with

CONTACT R. Reuben [email protected] Department of Psychology, University of Toronto, Toronto M5S 1A1, Ontario, Canada
ß 2021 International Menopause Society
2 R. REUBEN ET AL.

surgical or induced menopause have abruptly reduced ovar- Results

ian hormones at an early age, and are associated with a
General findings
greater risk of later-life AD and cognitive decline compared
to women with spontaneous menopause [20,21]. For Fifty articles were identified. After the removal of duplicates,
example, women with oophorectomy at midlife without 31 articles remained for title and abstract review. Based on
estradiol therapy have poorer performance on episodic and inclusion criteria, 11 studies were removed, leaving 20
working memory compared to age-matched controls [22,23]. articles for full-text review. One article was removed upon
Thus, it is important to examine the role of ovarian hormone full-text review as it did not meet the inclusion criteria, leav-
loss and dysregulation in cognitive decline in women with ing 19 total articles for inclusion (see Preferred Reporting
any type of menopause. Items for Systematic reviews and Meta-Analyses [PRISMA]
Independently, both SCD and the loss of ovarian hor- diagram shown in Figure 1).
mones are associated with cognitive decline and AD in later Publication dates ranged from 1996 to 2020, with 11
life. However, the link between SCD and ovarian hormones articles (58%) published within the past 10 years and five
has not been thoroughly explored. Further examination may articles (26%) published since 2014 (the date of the SCD
offer a better understanding of why more women than men framework consensus [1]). Cohorts included from 16 to 3044
have AD. Therefore, we reviewed the current literature on persons, and included ages from 20 to 86 years. Eighteen
subjective cognitive complaints to determine whether meno- studies comprised women exclusively; we included one study
pause or the loss of ovarian hormones contributes to cogni- with a mixed-sex cohort that disaggregated their data by
sex, yielding a total of 19 studies.
tive complaints and SCD in women.
The terms used to designate cognitive complaints varied
between studies, with subjective memory complaints (26%)
the most frequent. Methods used to measure cognitive com-
Methods plaints also varied considerably. Eleven studies (58%)
reported using at least one validated questionnaire to meas-
We conducted a review of the literature on SCD, menopause
ure subjective cognition; the remaining studies used other
and ovarian hormones in PubMed, PsycINFO and Web of
methods including single yes/no response questions, inter-
Science in July 2020. We used Boolean combinations of the
views and diaries.
following terms: subjective cognitive impairment, subjective
Of the 19 total studies, 12 (63%) assessed cognition. Of
cognitive decline, cognitive complaints or subjective memory
these 12 studies, two did not determine the normal cogni-
complaints and menopause, postmenopause, postmeno-
tive status, while one reported that 13% of their participants
pausal, oophorectomy, ovarian removal, surgical menopause,
performed outside the normal range. Another two studies
ovarian hormones, estrogen, estradiol, hormone therapy, labeled participants as having SCD, and seven specified that
estrogen therapy or hormone replacement therapy. The all participants were cognitively normal. Therefore, we con-
search was unbounded by dates. Bibliographies from rele- sidered nine of 12 studies (75%) to have assessed SCD
vant articles were also reviewed and some references (Table 1).
included. After removal of duplicates, the titles and abstracts Studies used three main approaches to investigate cogni-
of all articles were reviewed by at least two authors tive complaints with menopause and/or ovarian hormones:
for inclusion. frequency of cognitive complaints; neuropsychological test-
ing; and neuroimaging. Studies of cognitive complaints and/
or cognition in women taking estrogen-decreasing treat-
Inclusion criteria ments were categorized separately. We organized the pres-
entation of studies according to these approaches.
Inclusion criteria were: written in English, human participants
and SCD or cognitive complaints assessed with respect to
menopause, ovarian hormones or hormonal loss/replace- Frequency of cognitive complaints
ment. Reviews, case studies and conference abstracts Seven studies investigated the frequency of cognitive com-
were excluded. plaints with menopause status and ovarian hormone levels.
Three studies assessed cognitive complaints in combined
menopause status groups. One reported on 230 American
Data extraction women (mean age ¼ 46.7 years) across the early (n ¼ 53),
middle (n ¼ 54) and late (n ¼ 27) stages of the menopausal
The following information was extracted from each article transition [24]. The remaining 96 women included took hor-
after reading the full text: study purpose, participants, study mone therapy (HT) or oral contraceptives of any type, had a
design, term for cognitive complaints, measure of cognitive hysterectomy or were not classifiable into a menopause
complaints, whether normal cognition was determined, stage due to inadequate data. Frequency analyses were con-
measure of menopause or hormones, outcome measures, ducted on the group as a whole and showed that 62%
statistical analysis and results. At least two authors discussed (n ¼ 142) reported cognitive complaints. Another study, con-
and agreed on the data extracted for each study included. sisting of 103 American spontaneously perimenopausal or
 CLIMACTERIC 3

Figure 1. PRISMA flow diagram of the study selection process. PRISMA, Preferred Reporting Items for Systematic reviews and Meta-Analyses; SCD, subjective cogni-
tive decline.

postmenopausal women (mean age ¼ 51.2 years) and 48 perimenopause (n ¼ 25; mean age ¼ 45.44 years), postmeno-
premenopausal women (mean age ¼ 39.6 years) found that pause (n ¼ 38; mean age ¼ 55.63 years), HT (any type; n ¼ 36;
those in perimenopause and postmenopause were three mean age ¼ 55.03 years) and oral contraceptives (n ¼ 31;
times more likely to have complaints than those in premeno- mean age ¼ 34.74 years) [28]. However, there were no group
pause, even after adjustment for age [25]. The third study, differences in complaints on the Subjective Memory
comprising Italian women who were postmenopausal for any Questionnaire.
reason (88 recruited from a menopause clinic; 78 spontan- Two studies focused on the relationship between cogni-
eously menopausal and 10 surgical; 43 taking HT of any tive complaints and circulating estrogen levels. In one, 3044
type), found that 70% (n ¼ 62) related to captioned cartoon spontaneously menopausal American women (mean age at
depictions of forgetfulness. Those not taking HT reported hormone assessment ¼ 60.1 years; not taking HT) were fol-
more forgetfulness than those taking HT [26]. lowed for up to 23 years [29]. Higher levels of plasma estrone
Two studies compared the prevalence of cognitive com- sulfate (estradiol precursor) at midlife predicted decreased
plaints across different menopause groups. One used per- odds of cognitive complaints (assessed by a mail-in question-
formance on the Prospective and Retrospective Memory naire) in later life. However, a study of 292 American women
Questionnaire to assess memory complaints in Brazilian (mean age at baseline ¼ 41.4 years) comparing groups at dif-
women in early postmenopause (within the first 5 years; ferent stages of menopausal transition – late premenopause
n ¼ 27), late postmenopause (after 5 years; n ¼ 62) and pre- (cycle regularity), early (cycle irregularity) or late perimeno-
menopause (n ¼ 233; none taking HT) [27]. Accounting for pause (persistent skipped periods), or early postmenopause
age, women in early postmenopause had significantly more (within 5 years of last period) – found no relationship
memory complaints than those in late postmenopause and between difficulty concentrating or forgetfulness (assessed
premenopause. The other study compared five distinct via yearly diary entries) and menopause stage or estrone glu-
groups: premenopause (n ¼ 42; mean age ¼ 33.68 years), curonide (urinary estradiol metabolite) [30]. Instead, difficulty
Table 1. Characteristics of the included studies.
Menopause or
Reference Purpose Participants Study design Term for CC CC measure Normal cognition hormone measure Outcome measures Statistical analysis Results
Armeni Evaluate association 44 from menopause Cross-sectional Subjective Yes to: ’Do you have Not assessed 12 consecutive months HVLT; BVMT; Verbal Digits Multivariate linear Women with
et al. [35] between memory clinic with subjective memory any memory amenorrhea Backwards Test regression E2 > 10.0 pg/ml
performance and sex memory complaints complaints problems E2 levels < 50 pg/ml and Grouping by E2 levels had higher HVLT
hormones in Age ¼ 42–77 years compared to FSH > 25 mIU/ml total scores
postmenopausal (58.7 ± 6.5; not the past?’ Free estrogens and compared to
women taking HT) androgens calculated women with
using total E2 and E2  10.0 pg/ml;
total testosterone E2 predicted >
in serum 10.0 pg/ml
predicted HVLT
total scores
Betti Identify daily life areas 88 from menopause Cross-sectional Memory Cartoons Not assessed Patients followed since Frequency of menopause Frequency and count 70% of
et al. [26] that need support clinic loss complaints accompanied by a beginning of experiences; State- statistics; postmenopausal
postmenopause Age ¼ 52.56 ± 4.48 years yes/no question perimenopause Trait Anxiety Mann–Whitney women
78 Amenorrhea at least Inventory; Self-Rating U test complained of
spontaneously 1–3 years Depression Scale memory loss
menopausal
Conley Identify whether SCD 44 postmenopausal Cross-sectional SCD % of all items Defined as SCD Menopause Symptom SRT; N-back sequential Linear regression Participants with
et al. [38] associated with gray Age ¼ 50–60 years endorsed on Checklist (MSC) for letter task; regional model; more cognitive
matter volume and (56.1 ± 2.7 the CCI severity of gray matter mediation analyses complaints had
cognitive performance Hysterectomy or postmenopausal volume (GMV) lower GMV at the
change in early unilateral symptoms right MTL; N-back
postmenopause oophorectomy (n ¼ 8) or SRT was not
At least one functioning significantly
ovary and associated with
spontaneous either the CCI or
menopause GMV;
11 prior HT (1 MSC did not mediate
year ago) the relationship
between cognitive
complaints
and GMV
Devi Determine whether 151 healthy Cross-sectional Memory Yes to questions Not assessed Menopause stage Presence of Logistic regression Perimenopausal and
et al. [25] memory complaints Age ¼ 30–60 years loss complaints about the determined by age, menopausal postmenopausal
are reported in the 103 menopausal or presence of changes in menstrual symptoms women were 3
menopausal symptom perimenopausal memory loss cycle and reports more likely to
complex in healthy (51.2 ± 5.0) of changes complain of
community-dwelling 48 premenopausal memory loss than
postmenopausal controls (39.6 ± 7.2) premenopausal
women women
Devi Determine whether 26 postmenopausal Double-blind placebo- Subjective Subjective measures Yes Cessation of periods  BRCS; WMS-III; WAIS-III; Pre/post ratio Trending
et al. [36] donepezil Age ¼ 46–60 years (no controlled trial cognitive function of cognitive 1 year Buschke SRT; Boston calculated; t-tests; improvement in
(acetylcholinesterase HT) function on Brief Diagnostic Aphasia general subjective
inhibitor; AD 12 placebo (53.71 ± 3.56), Cognitive Rating Examination; linear model cognition in both
management drug) is 14 donepezil Scale (BCRS) Controlled Oral Word groups of
an effective treatment 5 mg (55.21 ± 3.47) Association Test postmenopausal
for cognitive and women treated
memory complaints with donepezil
postmenopause and placebo
Drogos Examine relationship 68 perimenopausal (21%) Cross-sectional Subjective Memory Functioning Yes Last period 6 months–10 MFQ; GCS; California Multivariable linear Better CVLT
et al. [33] between CC and and postmenopausal memory Questionnaire years prior to Verbal Learning Test regression analysis performance
objective memory (79%) complaints (MFQ) recruitment; (CVLT); Logical predicted fewer
performance Reported 35þ hot Overall current Greene Climacteric Scale Memory; Benton complaints about
flushes/week; no HT; memory rating (GCS) for Visual Retention Task; current memory
age ¼ 44–62 defined as MFQ psychological, Modified Card function; greater
years (53.00 ± 4.3) item 1: ’How vasomotor, sexual Rotations Test; Letter frequency of
would you rate dysfunction and Fluency; Digit Span; forgetting
your memory in somatic symptoms Brief Test of Attention; predicted by poor
terms of the kinds of menopause Finding A’s Test Digit Span forward
of problems that
you have?’
(continued)
Table 1. Continued.
Menopause or
Reference Purpose Participants Study design Term for CC CC measure Normal cognition hormone measure Outcome measures Statistical analysis Results
Ford et al. [28] Establish whether 172 from general practice Cross-sectional Perceived memory Subjective Memory Not assessed Self-report SMQ; Hospital Anxiety ANOVAs, ANCOVAs No significant
variations in patient list difficulties Questionnaire and Depression Scale; MANOVA; difference on total
menopausal status Age ¼ 26–64 years (SMQ) Women’s Health multiple regression score between
and age predict (45.70 ± 11.01) Questionnaire groups
prevalence of CC 42 premenopausal No difference in
(33.68 ± 6.04), 25 memory concern
perimenopausal level
(45.44 ± 4.94), No group differences
38 postmenopausal in overall memory
(55.63 ± 6.27), 36 HT rating
(55.03 ± 6.64), 31 Neither menopausal
OC (34.74 ± 7.88) status nor
hormonal
treatments
predicted
perceived
memory problems
Koyama Determine whether 3044 postmenopausal Longitudinal Subjective Mailed questionnaire yes Self-report at baseline Telephone Interview of Multivariate logistic Higher levels of
et al. [29] plasma sex hormones from Nurses’ Health prospective cognitive concerns with items Plasma levels for: estrone, Cognitive Status regressions plasma estrone
affect subjective and Study assessing memory estrone sulfate, E2, composite score associated with
objective Age ¼ 43–69 years (age and cognitive testosterone across all time points decreased odds of
cognitive function at blood draw ¼ function on a androstenedione, reporting SCC
60.1); not taking HT variety of tasks DHEA and DHEA-S
Mitchell and Determine memory 230 healthy Longitudinal cohort Memory changes Interview questions: Not assessed Menstrual calendars and Self-reported cognitive Thematic content 62% reported a
Woods [24] changes at the MT Age ¼ 40–60 years ’Have you noticed interviews; MTS-E, changes in interviews analysis; frequency negative memory
(46.7 ± 4.4) any changes in changes in flow and analysis; chi- change in
Subgroups of MT stages: your memory over cycle length but no squared tests recent years
early (MTS-E; n ¼ 53), the past few change in regularity;
middle (MTS-M; years?’; ’What kind MTS-M, persistent
n ¼ 54) and late of changes?’; ’How irregularity without
(MTS-L; n ¼ 27); long ago did you skipped periods; MTS-
n ¼ 93 HT, OC, first notice a L, skipped periods
hysterectomy or not change in your
classified into memory?’;
MT stage ’Describe the
changes you have
noticed about
your memory.’;
’What do you
think are the
reasons for your
memory changes?’
Mitchell and Determine whether 292 in late Longitudinal Cognitive symptoms Diary rating Not assessed Menopausal stage STRAW Diary cognitive symptoms Mixed- Best predictors in final
Woods [30] cognitive symptoms premenopause, early concentration criteria, participant effects modeling difficulty
correlate with CC or late difficulties and menstrual calendar; concentrating
perimenopause, or forgetfulness urinary E1G, model: age,
early postmenopause testosterone and FSH; depressed mood,
Age at baseline) ¼ 35–55 menopausal anxiety, awakening
years (41.4 ± 4.3) symptoms assessed during the night,
At least uterus and in diary perceived stress,
one ovary perceived positive
health and current
employment
Best predictors in final
forgetfulness
model: age, hot
flashes, anxiety,
depressed mood,
perceived stress,
perceived health
and history of
sexual abuse
(continued)
Table 1. Continued.
Menopause or
Reference Purpose Participants Study design Term for CC CC measure Normal cognition hormone measure Outcome measures Statistical analysis Results
Newton Determine effect of GnRH 16 with endometriosis Randomized Memory complaints Memory Observation Not assessed Serum E2 concentrations MOQ; Profile of Mood Analyses of variance Treatment led to
et al. [40] agonist on memory Age ¼ 32.4 ± 5.5 years prospective Questionnaire obtained pre and post States; Health with planned greater memory
and assess role of (MOQ); adverse treatment at first Concerns Scale; orthogonal complaints than in
emotional, somatic effects diary rating menstrual bleed adverse effects diary contrasts controls
and of memory After discontinuation of
personality factors problems GnRH-agonist
treatment, perceived
memory functioning
improved
to baseline
Piauilino Evaluate factorial 664 (n ¼ 341 women) Cross-sectional Meta-memory PRMQ Not assessed Menopause  1 year of PRMQ scores Confirmatory factor More memory
et al. [27] structure of the from Sao Paulo amenorrhea and a analysis of complaints in
Prospective and Epidemiologic Sleep plasma FSH level PRMQ items women than men;
Retrospective Memory Study above 30 mUI/ml women in the first
Questionnaire (PRMQ) Age ¼ 20–80 years (measured by 5 years after
premenopause (n ¼ 233), competitive menopause had
early postmenopause immunoassay) more memory
(first 5 years since complaints than
menopause; n ¼ 27), premenopausal
late postmenopause women and
(more than 5 years women after the
after menopause) first 5 years
with no HT (n ¼ 62) of menopause
Schaafsma Assess CC and subjective 120 (22 Cross-sectional Subjective Response to the Not assessed Menopausal status Semi-structured ANCOVAs and Perimenopause and HT
et al. [37] attention and premenopausal,48 cognitive questions: ‘Do you determined by questionnaire; MANCOVAs; groups more likely
cognitive perimenopausal, 38 complaints have problems bleeding patterns and Menopausal ordinal logistic than the
performance; postmenopausal, 12 with your STRAW criteria Symptom Scale regression analyses premenopause
determine hormonal, HT) from the general memory?’ and ‘Do group to report
psychosocial and population you have attention problems;
cognitive predictors Age ¼ 45–60 years problems with perimenopause
of CC attention/ and
concentration?’ postmenopause
(assessed for groups more likely
duration, severity than
and interference premenopause to
with daily make spontaneous
function); ‘Describe cognitive
your experience of complaints;
the menopause’ perimenopause
and
postmenopause
groups associated
with longer
duration of
attention problems;
HT group
associated with
longer duration of
memory problems
Schilder Examine effects of 128 postmenopausal Cross-sectional Cognitive complaints Interview about daily Yes Menopausal symptoms RAVLT; WMS-R visual Chi-squared tests/ Patients more likely to
et al. [41] exemestane and Age ¼ 48–71 years cognitive assessed by Endocrine memory subtest; univariate ANOVAs report daily
tamoxifen on 30 tamoxifen after AC complaints; Subscale of the Visual Association MANOVAs memory issues
cognition chemotherapy Cognitive Failures Functional Assessment Test; WAIS-III ANOVAS, than controls
postmenopause (57.9 ± 3.9), 50 Questionnaire of Cancer Therapy – Letter–Number logistic regression Participants receiving
exemestane after AC (CFQ); EORTC Breast questionnaire Sequencing; Stroop treatment
chemotherapy Quality of Life (FACT-B ES) Card 1, 2 and 3; Trail performed more
(58.5 ± 5.4)), Questionnaire – Making Task A þ B; poorly on category
48 controls (60.2 ± 5.1) Cognitive FePsy Reaction Times; fluency and
cancer-free at time Functioning Scale FePsy finger tapping information
of testing and visual reaction processing tasks
times; letter fluency, than controls
categorical fluency; CFQ scores correlated
Dutch Adult with menopausal
Reading Test symptoms but not
overall cognitive
performance
(continued)
Table 1. Continued.
Menopause or
Reference Purpose Participants Study design Term for CC CC measure Normal cognition hormone measure Outcome measures Statistical analysis Results
Schilder Evaluate self-reported 299 postmenopausal with Longitudinal, Cognitive complaints Yes to: ‘do you have Yes No definition/criteria of RAVLT; Visual Association Logistic Adjuvant therapy with
et al. [42] cognitive functioning breast cancer and participants from complaints with menopause given, Test; WMS visual regression; tamoxifen and
before and during healthy international regard to menopausal memory; Trails A þ B; ANCOVA exemestane did
anticancer endocrine postmenopausal randomized trial memory?’ and ‘do complaints assessed Stroop Card 1–3, not influence the
treatment; determine controls you have by FACT-B-ES WAIS-III self-reported
associations between 80 tamoxifen (68.7 ± 7.6), complaints with Letter–Number frequency of
CC, cognitive 99 exemestane regard to Sequencing; Hopkins cognitive failures;
performance and (68.3 ± 6.8) attention/ Symptom Checklist tamoxifen but not
anxiety/depression, 120 controls (66.2 ± 7.9) concentration?’; (HSCL-25); European exemestane
fatigue and CFQ Organization for increased
postmenopausal Research and attention/
complaints Treatment of Cancer concentration
Quality of Life complaints;
questionnaire; memory
Endocrine Subscale of complaints and
FACT-B ES task performance
before and 1 year
after treatment
unassociated
Triantafyllou Investigate association 39 postmenopausal from Cross-sectional Subjective Yes to: ‘Do you have Yes 12 consecutive months MMSE; Clock Drawing ANOVA; multivariate Free estrogens
et al. [34] between independent a menopause clinic memory any memory amenorrhea; levels of Test; HVLT; BVMT regression analysis predicted worse
perimenopausal with subjective complaints problems E2 < 50 pg/ml and BVMT performance
symptoms on memory complaints compared to FSH > 25mIU/ml (total and delayed
cognition Age ¼ 42–77 the past?’ Free estrogens measured recall) and HVLT
and memory years (58.7 ± 6.8) using blood samples discrimination
index
Vega Investigate relationship of 31 spontaneously Cross-sectional Subjective Endorsed >20% of Defined as SCD Amenorrhea for 1 year; SRT immediate and Pearson correlation; CCI score positively
et al. [39] resting state activity postmenopausal cognitive items on the CCI MSC determined delayed; resting second-level correlated with
and SCD Age ¼ 50–60 years complaints severity of state fMRI random MSC; CCI not
(56.0 ± 3.3) menopausal effects analysis related to
No severe menopause symptoms performance on
symptoms measures of the
No HT SRT
Higher CCI scores
correlated with
stronger positive
functional
connectivity in the
executive control
network and
worse negative
functional
connectivity in the
left middle
frontal gyrus
Weber and Examine relationship 24 spontaneously Cross-sectional Subjective memory MFQ Yes Self-report Digit Span subtest of Pearson correlation; 79% of the women
Mapstone between CC and perimenopausal complaints; Early perimenopause WMS-III; D2 Test of stepwise reported some
[31] performance on memory ¼ presence of Attention; linear regression degree of memory
neuropsychological Age ¼ 40–57 years complaints irregular periods; late Letter–Number loss of which 46%
tests; determine (49.7 ± 4.1) perimenopause  two Sequencing subtest of reported at least
whether deficits 12 early perimenopause, skipped periods WMS-III; Controlled moderate severity
correlate with 12 late during the past 12 Oral Word Association Women with few
hormone levels, mood perimenopause months and one Test; Grooved memory
or sleep quality No HT period of amenorrhea, Pegboard Test; Hooper complaints
with  one menstrual Visual Organization outperformed
cycle during the past Test; RAVLT those with
12 months complaints in
Serum E2 and FSH attention,
collected testing day encoding and
and measured fluency
Women with
significant memory
complaints
performed worse
on tests
of encoding
(continued)
 plaints; CCI, Cognitive Complaints Index; DHEA, dehydroepiandrosterone; DHEA-S, dehydroepiandrosterone sulfate; EORTC, European Organization for Research and Treatment of Cancer; E1G, estrone glucuronide; E2, estradiol; fMRI,
functional magnetic resonance imaging; FSH, follicle stimulating hormone; GnRH, gonadotropin-releasing hormone; HVLT, Hopkins Verbal Learning Test; HT, hormone therapy; MANOVA, multivariate analysis of variance; MMSE, Mini-
AC, adjuvant doxorubicin and cyclophosphamide; AD, Alzheimer’s disease; ANCOVA, analysis of covariance; ANOVA, analysis of variance; BRCS, Brief Cognitive Rating Scale; BVMT, Brief Visuospatial Learning Test; CC, cognitive com-

Mental State Examination; MT, menopausal transition; MTL, medial temporal lobe; OC, oral contraceptives; RAVLT, Rey Auditory Verbal Learning Task; SCC, subjective cognitive complaints; SCD, subjective cognitive decline; SRT,
complex attention/
forgetting to be at

memory task and

memory loss and
(n ¼ 50) reported

least moderately

Memory complaints

performance on
some degree of

associated with

complaints and

hormone levels
vigilance task;
41% (n ¼ 31)

performance
unrelated to
the working
67% of women
Results
concentrating and forgetfulness correlated with night-time

reported

memory

memory
serious

poorer
awakening and hot flash severity, respectively.

Neuropsychological testing
Pearson correlations;
Statistical analysis

linear regression

Seven studies investigated cognitive complaints with meno-

stepwise

pause and ovarian hormones and performance on neuro-

psychological tasks.
Two studies examined the relationship between cognitive
complaints and neuropsychological tasks in perimenopausal
Outcome measures
Cognitive battery and

women. Twenty-four perimenopausal American women

questionnaires

(aged 40–57 years; not taking HT; cognitively normal) were

self-report

evaluated for memory complaints using the Memory

Functioning Questionnaire (MFQ) [31]. Forty-six percent
reported memory problems of moderate severity or greater.
at least one menstrual
period in the prior 12
menstrual cycles; and
Self-report of changes in

Higher MFQ scores were associated with worse encoding on

collected testing day
uterus; women with

or HT in the prior 3
surgical menopause
months; with intact
hormone measure

months excluded
Menopause or

the Rey Auditory Verbal Learning Task (RAVLT) and working

Serum E2 and FSH

and measured

memory performance on the Letter–Number Sequencing

task. Serum estradiol levels collected at testing predicted
Selective Reminding Task; STRAW, Stages of Reproductive Aging Workshop; WAIS, Wechsler Adult Intelligence Scale; WMS, Wechsler Memory Scale. encoding performance, but were not correlated with memory
complaints. In a study of 75 American perimenopausal
(n ¼ 10) performed

women (aged 40–60 years; not taking HT), 67% (n ¼ 50)

impaired range on
within borderline
Normal cognition

reported some degree of memory loss and 41% (n ¼ 31)

impaired to
Reported 13%

the RAVLT

reported moderate severity or greater [32]. While memory

complaints were not related to serum estradiol, MFQ scores
were negatively associated with working memory perform-
ance on the Letter–Number Sequencing task and complex
attention/vigilance on the D2 Test of Attention. However,
CC measure

13% of participants (n ¼ 10) performed within the borderline

impaired to impaired range on the RAVLT.
MFQ

Another study including 68 American perimenopausal and

postmenopausal women not taking HT (79% postmeno-
pausal, mean age ¼ 53 years; cognitively normal) found that
Term for CC

better verbal memory performance on the delay trial of the

complaints
memory

California Verbal Learning Task predicted better current

Subjective

memory function on the MFQ. Further, poorer performance

on the Digit Span Forward predicted higher frequency of for-
getting on the MFQ [33].
Study design

Three studies related cognitive complaints with neuro-

Cross-sectional

psychological task performance in postmenopausal women.

Two came from the same research group, studying spontan-
eously menopausal Greek women with cognitive complaints
from a menopause clinic. The first (n ¼ 39, mean age ¼
58.7 years; cognitively normal) found that menopausal symp-
Age ¼ 40–60 years
perimenopausal
Participants

tomatology and serum estradiol levels predicted total and

75 spontaneously

(49.3 ± 4.3)
Not taking HT

delayed visuospatial memory on the Brief Visuospatial

Memory Test [34]. The second study (n ¼ 44, mean age ¼
58.7 years; normal cognition unspecified) found that women
with serum estradiol levels greater than 10.0 pg/ml had bet-
cognitive performance

ter verbal recall on the Hopkins Verbal Learning Test com-

Determine relationship
between CC and

pared to those with less than 10.0 pg/ml [35]. A third study
Purpose

of 28 American community-dwelling women in spontaneous

menopause (age 40–60 years; HT unspecified; cognitively nor-
Table 1. Continued.

mal) with cognitive complaints were given either a placebo

or donepezil (acetylcholinesterase inhibitor for AD symp-
et al. [32]

toms); neither affected the frequency of cognitive complaints

Reference

or performance on neuropsychological tasks [36].

Weber
 CLIMACTERIC 9

In a study of American women, 22 premenopausal, 48 participants taking tamoxifen and those taking exemestane.
perimenopausal, 38 postmenopausal and 12 taking unspeci- However, both treatment groups had higher scores on the
fied HT (aged 45–60 years; normal cognition unspecified), CFQ and poorer performance on the verbal fluency and
subjective memory and attention problems were assessed FePsy visual reaction time tasks compared to controls [41].
via single-item yes/no questions. Those in perimenopause The same group used the CFQ to determine the relationship
and taking HT were more likely to have memory and atten- between cognitive complaints before and after 1 year of
tion complaints than those in premenopause. Hot flash fre- estrogen-decreasing treatment in cognitively normal women
quency on the Menopause Symptom Scale was associated taking tamoxifen (n ¼ 80; mean age ¼ 68.7 years) or exemes-
with unprompted cognitive complaints, while hot flash inten- tane (n ¼ 99; mean age ¼ 68.3 years) and controls (n ¼ 120;
sity was associated with greater interference of memory mean age ¼ 66.2 years). They found no differences between
problems on daily functioning. For all groups, memory and groups on change in CFQ scores pre and post treatment,
attention problems were associated with poorer verbal mem- although women taking tamoxifen had increased attention/
ory and slower attentional reaction time on the Weschler concentration complaints (measured by a single-item yes/no
Memory Scale and CalCAP computerized reaction time question) after 1 year of treatment [42].
tasks [37].

Discussion
Neuroimaging
We conducted a review to determine whether menopause
Two studies investigated brain changes in postmenopausal and the loss of ovarian hormones contribute to cognitive
women with cognitive complaints. complaints and SCD in women. Overall, the current research
Cognitive complaints were assessed using the Cognitive provides some early support for this association, and war-
Complaints Index (CCI) in 44 spontaneously menopausal rants further targeted investigations with improved study
American women (mean age ¼ 56.1 years; cognitively nor- design and quality.
mal; eight with hysterectomy or unilateral oophorectomy, 11 Four studies found that the majority of women across the
taking past unspecified HT). Complaints were associated with menopausal transition reported cognitive complaints.
reduced gray matter volumes in the right MTL and increased Further, three studies determined that perimenopausal and
Menopause Symptom Checklist (MSC) scores. However, the postmenopausal women had greater complaints than preme-
relationship between cognitive complaints and MTL volume nopausal women, while two found no group differences in
was not mediated by MSC scores, nor was there a relation- complaints by menopause status. These findings suggest
ship between complaints and performance on the N-back or that cognitive complaints may arise more frequently during
the Selective Reminding Task [38]. postmenopause and the menopausal transition.
The relationship between resting state functional connect- Supporting this, studies also found relationships between
ivity and CCI scores was assessed in 31 spontaneously meno- cognitive complaints and menopausal symptoms. Thought to
pausal American women (mean age ¼ 56 years; not taking be caused by hormonal fluctuations, menopausal symptoms
HT; cognitively normal). CCI scores were positively correlated themselves may also contribute to memory decline. Hot
with functional connectivity in the executive control network flashes are known to disrupt sleep and daily functioning [43],
and right middle temporal gyrus, but negatively correlated which, in turn, can affect memory. Ovarian hormones are
with functional connectivity in the left middle frontal gyrus. implicated in a variety of bodily systems, including regulation
The CCI score was also positively correlated with the MSC. of blood flow, temperature, sleep, cellular energy balance
There was no correlation between the CCI score and delayed and inflammation – all of which may affect memory and
verbal recall on the Selective Reminding Task [39]. brain health [15,44]. Thus, postmenopausal cognitive com-
plaints may be part of a larger network of dysregulation after
the loss of ovarian hormones. Future studies should aim to
Estrogen-decreasing treatments
understand how the dysregulation of each of these systems
Three studies investigated cognitive complaints in women contributes to both menopausal symptoms and later-life cog-
taking estrogen-decreasing treatments. nitive decline.
Measured over a period of 24 weeks, 16 Canadian women Studies that included neuropsychological and imaging
with endometriosis (mean age ¼ 34.4 years) taking gonado- measures found that cognitive complaints were predictive of
tropin-releasing hormone agonists (inducing rapid estrogen objective cognitive outcomes, including differences in verbal
depletion) had more cognitive complaints than an age- memory, attention, working memory and AD-related brain
matched community sample. After treatment cessation, cog- regions, such as the MTL. While numerous studies support
nitive complaints were reduced to baseline levels [40]. the role of estrogens in promoting cognition and frontal cor-
A comparison of spontaneously menopausal Dutch tical maintenance [45,46] and have shown similar neuro-
women taking tamoxifen (n ¼ 30; mean age ¼ 57.9 years), psychological and brain changes after estrogen loss
exemestane (n ¼ 50; mean age ¼ 58.5 years; cognitively nor- [19,22,23], fewer have correlated the subjective experiences
mal) and controls (n ¼ 48; mean age ¼ 60.2 years; cognitively of their participants with objective outcomes. Within postme-
normal) revealed no difference on the Cognitive Failures nopausal populations, cognitive complaints after ovarian hor-
Questionnaire (CFQ) or cognitive performance between mone loss might serve as an important indicator of potential
10 R. REUBEN ET AL.

future decline and contribute to the identification of those at effects of SCD in menopausal women to better understand
risk for dementia. Further, hormonal fluctuations in peri- this relationship and its implications for AD risk.
menopause have been related to cognitive decrements and
increased cognitive complaints – making perimenopause an
Limitations of the literature
important window for potential intervention and future
investigation of long-term effects [47]. Although not all The total of 19 studies published between 1996 and 2020
women with cognitive complaints at the menopausal transi- denotes a paucity of studies on SCD in menopause. Perhaps
tion will show future impairment, they will likely benefit reflecting the relative newness of SCD research, there was
from clinicians taking their complaints seriously. Thus, those wide variation in terminology, study design and measure-
with cognitive complaints showing measurable cognitive ment, creating challenges in determining the strength of this
and/or brain changes form an important group to follow association. Although the majority of studies (58%) utilized
over time as they may presage continuing changes. validated questionnaires, several relied on single-item ques-
Studies also provided direct support for a relationship tions, interviews and diary entries to measure cognitive com-
between cognitive complaints and ovarian hormone levels. plaints. Similarly, while the majority of studies (47%)
One study determined that higher estrone sulfate, an estra- examined group differences across different menopause
diol precursor, at midlife led to decreased cognitive com- stages, studies also used regressions and frequency statistics
plaints in later years [29], and others found that higher within a single menopause group. As previously noted, the
serum estradiol predicted better verbal and visuospatial various terms, research designs and lack of widely used
memory performance in postmenopausal women with cogni- measurement criteria make it difficult to compare studies
tive complaints [34,35]. However, three studies did not find a [49]. Further, while there was a range of cognitive domains
direct relationship between ovarian hormones and cognitive assessed across the studies, only two studies utilized neuroi-
complaints. As all studies measured hormones peripherally maging and four studies evaluated participants longitudin-
(blood or urine), the lack of correspondence may be due to ally. Considering that SCD is an indicator of longitudinal
not knowing levels in the brain. Measuring the unbound decline [1], and brain changes are often observable earlier
ovarian hormones in saliva may provide a stronger correlate than cognitive changes [50], future work should aim to
of cognition, as the unbound portion more easily crosses the address these areas in particular.
blood–brain barrier [48]. While nine of the 11 studies assessing cognition directly
Studies of participants taking treatments that block estro- determined that participants had cognitive scores within nor-
genic action in the brain or eliminate the production of mal limits, they were not described by the authors as meet-
estradiol provide more direct support for a relationship ing the criteria for SCD. One study [32] determined that 13%
between estradiol depletion and cognitive complaints. of participants scored outside the normal range, putting an
Women taking tamoxifen, aromatase inhibitors and estradiol interpretation of SCD in question. Since only five studies
blockers had increased cognitive complaints and poorer per- were published post SCD criteria, only two studies specific-
formance on verbal working memory and executive function. ally labeled their participants as having SCD [38,39].
Interestingly, one study of premenopausal women taking Furthermore, none of the studies investigating the frequency
gonadotropin-releasing hormone agonists found that, when of cognitive complaints defined SCD or assessed cognitive
estrogen production returned after cessation of treatment, status within their participants. Future studies in all areas of
cognitive complaints receded to pretreatment levels [40]. investigation will benefit from utilizing standardized SCD cri-
Even at a younger age than is typical for menopause and teria and explicitly defining participants as having SCD [1].
SCD, these findings demonstrate memory’s sensitivity to Multiple studies did not separate groups by menopause
estradiol loss which may underlie increased cogni- status or type, nor created separate groups of women taking
tive complaints. HT and oral contraceptives without specifying which meno-
Several studies focused solely on cognitive complaints, pause types and statuses comprised these groupings.
providing important insight into the frequency of cognitive Previous work has shown that full differentiation between
complaints across the menopausal transition and suggesting menopause groups strengthens the evidence for clinical eval-
that cognitive complaints alone are associated with cognitive uations and treatment of cognitive concerns [51].
decline. However, to qualify as SCD, complaints must be in Menopause staging was similarly inconsistent across studies,
the presence of performance within normal limits on stand- with only three studies (16%) referencing the standardized
ardized cognitive tests [1]. Overall, studies were aware of the Stages of Reproductive Aging Workshop (STRAW) staging
importance in establishing normal cognition in their partici- system [52]. The use of reliable standardized staging will be
pants, as 75% of studies assessing cognition met this criter- critical for the accuracy and interpretation of future investi-
ion. This evidence suggests that the menopausal transition gations of the menopausal transition [53]. The menopausal
and ovarian hormone loss may be an entry point for SCD in transition is an important window of cognitive and brain
women. Considering that SCD increases the risk of later AD changes in women and an opportunity for therapeutic inter-
[2], it is important to understand how cognitive complaints ventions for AD [54]. Thus, a more nuanced consideration of
during normal cognition may later progress to objective menopause groups and staging is necessary to ensure preci-
decline. Future work should aim to examine the long-term sion medicine for women.
 CLIMACTERIC 11

In this regard, cognitive complaints in relation to surgical References

menopause and HT were under-researched. Multiple studies
[1] Jessen F, Amariglio RE, Van Boxtel M, et al. A conceptual frame-
excluded women with surgical menopause and/or those tak- work for research on subjective cognitive decline in preclinical
ing HT. In some, they were included in small numbers with- Alzheimer’s disease. Alzheimers Dement. 2014;10:844–852.
out consideration of the differential effects of oophorectomy [2] Jessen F, Wiese B, Bachmann C, et al. Prediction of dementia by
on cognition. Women with oophorectomy prior to spontan- subjective memory impairment: effects of severity and temporal
association with cognitive impairment. Arch Gen Psychiatry. 2010;
eous menopause are at particular risk of AD in later life [21], 67:414–422.
and show early neuropsychological markers for cognitive [3] Dik MG, Jonker C, Comijs HC, et al. Memory complaints and
decline [23], suggesting that identifying SCD in women with APOE-epsilon4 accelerate cognitive decline in cognitively normal
surgical menopause would be particularly important for elderly. Neurology. 2001;57:2217–2222.
[4] Scheef L, Spottke A, Daerr M, et al. Glucose metabolism, gray
later-life risk. matter structure, and memory decline in subjective memory
Similarly, further consideration of HT on cognitive com- impairment. Neurology. 2012;79:1332–1339.
plaints in postmenopause is warranted. Although some stud- [5] Sanchez-Benavides G, Grau-Rivera O, Suarez-Calvet M, et al. Brain
ies included an HT group, none provided any further and cognitive correlates of subjective cognitive decline-plus fea-
tures in a population-based cohort. Alzheimer Res Ther. 2018;10:
information outside current or past use. There are many
113.
types, regimens and durations of use of HT, each of which [6] Schultz SA, Oh JM, Koscik RL, et al. Subjective memory com-
may engender unique cognitive outcomes [55]. Therefore, plaints, cortical thinning, and cognitive dysfunction in middle-
whenever possible, future research would benefit from speci- aged adults at risk for AD. Alzheimers Dement (Amst). 2015;1:
33–40.
fying and stratifying by HT types.
[7] Rabin LA, Smart CM, Amariglio RE. Subjective cognitive decline in
preclinical Alzheimer’s disease. Annu Rev Clin Psychol. 2017;13:
369–396.
[8] Alzheimer’s Association. 2019 Alzheimer’s disease facts and fig-
Conclusion ures. Alzheimers Dement. 2019;15:321–387.
[9] Mielke MM. Sex and gender differences in Alzheimer’s disease
These studies prompt further investigation of women’s cog- dementia. Psychiatric Times. 2018;35:14.
nitive complaints after menopause and ovarian hormone [10] Sundermann EE, Edmonds EC, Delano-Wood L, et al. Sex influen-
ces the accuracy of subjective memory complaint reporting in
loss. While more targeted research designs and improved
older adults. J Alzheimers Dis. 2018;61:1163–1178.
methodologies are needed, the current findings advise that [11] Heser K, Kleineidam L, Wiese B, et al. Subjective cognitive decline
cognitive complaints and SCD at the menopausal transition may be a stronger predictor of incident dementia in women than
are worthy of note, as they may presage further decline. We in men. J Alzheimer’s Dis. 2019;68:1469–1478.
[12] Kaup AR, Nettiksimmons J, LeBlanc ES, et al. Memory complaints
recommend precise measurement of ovarian hormones,
and risk of cognitive impairment after nearly 2 decades among
studies of early hormone loss for any reason and longitudinal older women. Neurology. 2015;85:1852–1858.
assessments of cognition and brain changes, including meas- [13] Weber MT, Maki PM, McDermott MP. Cognition and mood in
urement of premenopausal and postmenopausal hormone perimenopause: a systematic review and meta-analysis. J Steroid
Biochem Mol Biol. 2014;142:90–98.
levels. The increased prevalence of cognitive complaints dur-
[14] Li R, Cui J, Shen Y. Brain sex matters: estrogen in cognition and
ing the menopausal transition suggests that, for some, this Alzheimer’s disease. Mol Cell Endocrinol. 2014;389:13–21.
may be a branching point into SCD, mild cognitive impair- [15] Maki PM, Dumas J. Mechanisms of action of estrogen in the
ment and the AD continuum [1,56]. Future studies should brain: insights from human neuroimaging and psychopharmaco-
logic studies. Semin Reprod Med. 2009;27:250–259.
consider how increased cognitive complaints during any
[16] Sahab-Negah S, Hajali V, Moradi HR, et al. The impact of estradiol
type of menopause or ovarian hormone loss may signal pro- on neurogenesis and cognitive functions in Alzheimer’s Disease.
gression toward cognitive decline and AD. Cell Mol Neurobiol. 2020;40:283–299.
[17] Wong M, Moss RL. Long-term and short-term electrophysiological
effects of estrogen on the synaptic properties of hippocampal
CA1 neurons. J Neurosci. 1992;12:3217–3225.
Potential conflict of interest The authors report no conflict [18] Luine VN. Estradiol increases choline acetyltransferase activity in
of interest. specific basal forebrain nuclei and projection areas of female rats.
Exp Neurol. 1985;89:484–490.
[19] Kantarci K, Tosakulwong N, Lesnick TG, et al. Brain structure and
cognition 3 years after the end of an early menopausal hormone
Source of funding Canadian Consortium on Neurodegeneration in therapy trial. Neurology. 2018;90:e1404–e1412.
Aging Phase II [CCNA 049-04]; Canadian Institute of Health Research [20] Liao KF, Lin CL, Lai SW. Nationwide case-control study examining
[WJP-150643]; Ontario Brain Institute, Women’s Brain Health Initiative the association between tamoxifen use and Alzheimer’s disease
Chair; Wilfred and Joyce Posluns Chair in Women’s Brain Health and in aged women with breast cancer in Taiwan. Front Pharmacol.
Aging: Women’s Brain Health Initiative [WJP-150643]. 2017;8:612.
[21] Rocca WA, Bower JH, Maraganore DM, et al. Increased risk of cog-
nitive impairment or dementia in women who underwent
ORCID oophorectomy before menopause. Neurology. 2007;69:
1074–1083.
R. Reuben http://orcid.org/0000-0003-2275-9504 [22] Phillips SM, Sherwin BB. Effects of estrogen on memory function
N. A. Phillips http://orcid.org/0000-0003-0855-2961 in surgically menopausal women. Psychoneuroendocrinology.
G. Einstein http://orcid.org/0000-0002-0770-5471 1992;17:485–495.
12 R. REUBEN ET AL.

[23] Gervais NJ, Au A, Almey A, et al. Cognitive markers of dementia [39] Vega JN, Zurkovsky L, Albert K, et al. Altered brain connectivity in
risk in middle-aged women with bilateral salpingo-oophorectomy early postmenopausal women with subjective cognitive impair-
prior to menopause. Neurobiol Aging. 2020;94:1–6. ment. Front Neurosci. 2016;10:433.
[24] Mitchell ES, Woods NF. Midlife women’s attributions about per- [40] Newton C, Slota D, Yuzpe AA, et al. Memory complaints associ-
ceived memory changes: observations from the Seattle Midlife ated with the use of gonadotropin-releasing hormone agonists: a
Women’s Health Study. J Women’s Health Gender-Based Med. preliminary study. Fertil Steril. 1996;65:1253–1255.
2001;10:351–362. [41] Schilder CM, Eggens PC, Seynaeve C, et al. Neuropsychological
[25] Devi G, Hahn K, Massimi S, et al. Prevalence of memory loss com- functioning in postmenopausal breast cancer patients treated
plaints and other symptoms associated with the menopause tran- with tamoxifen or exemestane after AC-chemotherapy: cross-sec-
sition: a community survey. Gender Med. 2005;2:255–264. tional findings from the neuropsychological TEAM-side study.
[26] Betti S, Orsini MR, Sciaky R, et al. Attitudes towards menopause Acta Oncol. 2009;48:76–85.
[42] Schilder CM, Seynaeve C, Linn SC, et al. Self-reported cognitive
in a group of women followed in a public service for menopause
functioning in postmenopausal breast cancer patients before and
counseling. Aging (Milano). 2001;13:331–338.
during endocrine treatment: findings from the neuropsycho-
[27] Piauilino DC, Bueno OF, Tufik S, et al. The Prospective and
logical TEAM side-study. Psychooncology. 2012;21:479–487.
Retrospective Memory Questionnaire: a population-based random
[43] Pinkerton JV, Abraham L, Bushmakin AG, et al. Relationship
sampling study. Memory. 2010;18:413–426. between changes in vasomotor symptoms and changes in meno-
[28] Ford N, Slade P, Butler G. An absence of evidence linking per- pause-specific quality of life and sleep parameters. Menopause.
ceived memory problems to the menopause. Br J Gen Pract. 2016;23:1060–1066.
2004;54:434–438. [44] Brinton RD, Yao J, Yin F, et al. Perimenopause as a neurological
[29] Koyama AK, Tworoger SS, Eliassen AH, et al. Endogenous sex hor- transition state. Nat Rev Endocrinol. 2015;11:393–405.
mones and cognitive function in older women. Alzheimers [45] Galea LA, Frick KM, Hampson E, et al. Why estrogens matter for
Dement. 2016;12:758–765. behavior and brain health. Neurosci Biobehav Rev. 2017;76:
[30] Mitchell ES, Woods NF. Cognitive symptoms during the meno- 363–379.
pausal transition and early postmenopause. Climacteric. 2011;14: [46] Hampson E. Estrogens, aging, and working memory. Curr
252–261. Psychiatry Rep. 2018;20:109.
[31] Weber M, Mapstone M. Memory complaints and memory per- [47] Greendale GA, Derby CA, Maki PM. Perimenopause and cognition.
formance in the menopausal transition. Menopause. 2009;16: Obstet Gynecol Clin North Am. 2011;38:519–535.
694–700. [48] Yaffe K, Lui LY, Grady D, et al. Cognitive decline in women in
[32] Weber M, Mapstone M, Staskiewicz J, et al. Reconciling subjective relation to non-protein-bound oestradiol concentrations. Lancet.
memory complaints with objective memory performance in the 2000;356:708–712.
menopausal transition. Menopause. 2012;19:735 [49] Studart A, Nitrini R. Subjective cognitive decline: the first clinical
[33] Drogos LL, Rubin LH, Geller SE, et al. Objective cognitive perform- manifestation of Alzheimer’s disease? Dement Neuropsychol.
ance is related to subjective memory complaints in midlife 2016;10:170–177.
women with moderate to severe vasomotor symptoms. [50] Beason-Held LL, Goh JO, An Y, et al. Changes in brain function
Menopause. 2013;20:1236–1242. occur years before the onset of cognitive impairment. J Neurosci.
[34] Triantafyllou N, Armeni E, Christidi F, et al. The intensity of meno- 2013;33:18008–18014.
pausal symptoms is associated with episodic memory in postme- [51] Edwards H, Duchesne A, Au AS, et al. The many menopauses:
searching the cognitive research literature for menopause types.
nopausal women. Climacteric. 2016;19:393–399.
Menopause. 2019;26:45–65.
[35] Armeni E, Apostolakis M, Christidi F, et al. Endogenous sex hor-
[52] Soules MR, Sherman S, Parrott E, et al. Executive summary: stages
mones and memory performance in middle-aged Greek women
of reproductive aging workshop (STRAW). Climacteric. 2001;4:
with subjective memory complaints. Neurol Sci. 2018;39:259–266.
267–272.
[36] Devi G, Massimi S, Schultz S, et al. A double-blind, placebo-con-
[53] Sherman S. Defining the menopausal transition. Am J Med. 2005;
trolled trial of donepezil for the treatment of menopause-related 118:3–7.
cognitive loss. Gender Med. 2007;4:352–358. [54] Scheyer O, Rahman A, Hristov H, et al. Female sex and
[37] Schaafsma M, Homewood J, Taylor A. Subjective cognitive com- Alzheimer’s risk: the menopause connection. J Prev Alzheimers
plaints at menopause associated with declines in performance of Dis. 2018;5:225–230.
verbal memory and attentional processes. Climacteric. 2010;13: [55] Sherwin BB, Henry JF. Brain aging modulates the neuroprotective
84–98. effects of estrogen on selective aspects of cognition in women: a
[38] Conley AC, Albert KM, Boyd BD, et al. Cognitive complaints are critical review. Front Neuroendocrinol. 2008;29:88–113.
associated with smaller right medial temporal gray-matter vol- [56] Aisen PS, Cummings J, Jack CR, et al. On the path to 2025: under-
ume in younger postmenopausal women. Menopause. 2020;27: standing the Alzheimer’s disease continuum. Alzheimers Res
1220–1227. Ther. 2017;9:60.