ORIGINAL RESEARCH

Effect of Sex Hormones on Brain Connectivity Related to Sexual

Function in Perimenopausal Women: A Resting-State fMRI Functional
Connectivity Study
Weizhao Lu, MS,1,2 Wei Guo, MS,3 Dong Cui, PhD,2 Kejiang Dong, BS,2 and Jianfeng Qiu, PhD1,2

ABSTRACT

Background: Perimenopause is associated with increased risk of depression, vasomotor symptoms, and sexual
dysfunction.
Aims: To explore the effect of sex hormones on the functional connectivity (FC) of different brain regions
related to sexual function in perimenopausal women.
Methods: 32 premenopausal women (mean age, 47.75 ± 1.55 years) and 25 perimenopausal women (mean age,
51.60 ± 1.63 years) underwent sex hormone level measurements and resting-state fMRI.
Main Outcome Measures: Serum levels of sex hormones, including prolactin (PRL), follicle-stimulating
hormone (FSH), luteotropic hormone (LH), estradiol (E2), free testosterone (free-T), and progesterone (P),
were measured. 10 brain regions related to sexual function were selected according to a meta-analysis, and FCs of
the selected regions of interest were calculated as Pearson’s correlation coefficient.
Results: Compared with premenopausal women, perimenopausal women showed increased FC between the
right area 13 (A13_r) and the right medial superior frontal gyrus (mSFG), between the left dorsal granular insula
(dIg_L) and the right superior frontal gyrus (SFG) (Gaussian random field-corrected at the voxel level, P < .001,
and cluster level, P < .025). Furthermore, the PRL level was negatively correlated with the FC of A13_R with the
right mSFG and the FC of dIg_L with the right SFG.
Clinical Translation: These findings may be applicable to assessing brain dysfunction with FC changes in
women approaching menopause.
Strengths & Limitations: This study is the first to evaluate a direct relationship between sex hormone levels
and brain FC changes in women approaching menopause. Sexual function was not assessed, which may weaken
the conclusions related to sexual function.
Conclusions: The results show that women approaching menopause suffered from aberrant intrinsic FC in
regions related to sexual function, and reveal a direct relationship between serum sex hormone levels and FC
changes related to sexual function. Lu W, Guo W, Cui D, et al. Effect of Sex Hormones on Brain Con-
nectivity Related to Sexual Function in Perimenopausal Women: A Resting-State fMRI Functional Con-
nectivity Study. J Sex Med 2019;XX:XXXeXXX.
Copyright
2019, International Society for Sexual Medicine. Published by Elsevier Inc. All rights reserved.
Key Words: fMRI; Functional Connectivity; Premenopause; Perimenopause; Menopause Transition; Sexual
Function; Sex Hormones

INTRODUCTION
Received January 3, 2019. Accepted March 2, 2019.
1
A rapidly growing number of studies have documented sexual
Medical Engineering and Technical Center, Taishan Medcial University, Tai’an,
influences on brain structure and function.1 Stoléru et al2
China;
2 identified cortical activation in response to visually evoked sex-
Department of Radiology, Taishan Medical University, Tai’an, China;
3 ual arousal using positron emission tomography (PET).2 To our
Affiliated Hospital of Taishan Medical University, Tai’an, China
knowledge, this was the first study of sexual-related brain regions
Copyright ª 2019, International Society for Sexual Medicine. Published by
Elsevier Inc. All rights reserved.
in human using neuroimaging techniques. fMRI, a popular tool
https://doi.org/10.1016/j.jsxm.2019.03.004 in neuroscience, also has been used to study brain regions related

J Sex Med 2019;-:1e10 1

2 Lu et al
to sexual function.3e6 A study by Maravilla et al7 was the first to
evaluate female cerebral response to sexual stimuli using blood
oxygen level- dependent (BOLD)-fMRI. Park et al8 observed
activations of certain cerebral regions, including the occipital
cortex, inferior frontal lobe, and insula gyrus, during sexual
arousal in women using fMRI.8 Neuroscience studies have
shown that human sexuality relies on several brain regions,
including the insula and cingulate, and involves several functions,
including arousal, reward, memory, and inhibition.1,5,6
Perimenopause is defined as the transition between normal
ovarian function and menopause, which is accompanied by in-
ternal biological changes, such as endocrine and metabolic
changes.9 Symptoms of depression, vasomotor dysfunction, sleep
disturbance, vaginal dryness, and migraine are widely observed at
different stages of the menopausal transition.10,11 Sexual
dysfunction is a common problem in women during the meno-
pausal transition; various studies have reported declining sexual
activity, diminished sexual desire, and vaginal dryness in peri-
menopausal women.12,13 Avis et al14 reported that menopause and
other factors associated with menopause and aging leads to sexual
degradation. Several state-of-the-art studies have confirmed an
association between reproductive hormonal dynamics and sexual
dysfunction in women approaching menopause.15,16
Despite the brain’s central role in sexual function, little is known
about the relationship between brain functional changes and sex
hormone levels in women during menopausal transition. Recently,
emerging neuroimage techniques have provided convenient ap-
proaches to assess brain regions related to sexual function for women
approaching menopause.4,5 Among these neuroimaging tech-
niques, fMRI offers a noninvasive method for in vivo investigation
of brain function.4,17 Several fMRI studies have revealed differences
in brain activation related to sexual arousal between premenopausal
and menopausal women.4,17 Jeong et al4 compared brain activations
in response to visual sexual stimuli between premenopausal and
menopausal women using BOLD fMRI,4 and the same group
localized brain functional changes related to visual erotic stimuli in
postmenopausal women using functional magnetic resonance
spectroscopy.18 Kim et al. analyzed fMRI time course data in
response to sexual stimuli in premenopausal and menopausal
women using BOLD fMRI.17 Besides sexual stimuli, fMRI is also
used for neuropsychological assessment after hormone therapy in
women undergoing the menopausal transition.19
Most of the neuroscience research studies related to sexual
function in women undergoing the menopausal transition con-
ducted to date have been done with sexual stimuli. The brain ac-
tivations associated with sexual function were found to be spatially
remote.5 Few studies have focused on the effects of serum sex
hormone levels. In the present study, resting-state fMRI FC was
used to assess brain connectivities in the cortical areas related to
sexual function. We hypothesized that sex hormone levels may be
correlated with brain connectivities related to sexual function in
women approaching menopause. To test this hypothesis, we per-
formed correlation analyses between FC values and sex hormone
levels. Although there is ample evidence for subcortical involvement
in sexual function, subcortical structures were not selected, because
subcortical structures such as the hippocampus, thalamus, and
amygdala have close relationships with sex hormones.6 The corre-
lations between sex hormone levels and the FC of subcortical
structures may be affected by the double-dipping phenomenon.
MATERIALS AND METHODS
Subjects
Between June 2017 and January 2018, premenopausal and
perimenopausal subjects were recruited by the university-
affiliated hospital. Criteria for enrollment were as follows: (i)
female, age 45e55 years, (ii) high school degree or above, (iii)
right-handed, (iv) heterosexual, and (v) a diagnosis of perimen-
opause based on the Stages of Reproductive Aging Workshop
(STRAW) þ10: menstrual cycle change larger than 7 days or a
12-month period of amenorrhea.20 Premenopausal women were
recruited based on not meeting the diagnosis criterion of
STRAW þ10 and had a regular ovulation day based on the
rhythm method. Exclusion criteria were (i) history of psychiatric
or neurological illnesses according to clinical records, (ii) history
of hormone or steroid treatment or oral contraceptive use in the
month before the study, (iii) premenstrual syndrome and pre-
menstrual dysphoric disorder according to clinical queries, and
(iv) contradictions for MRI. The study protocol was approved by
the university’s Medical Ethics Committee in accordance with
the Declaration of Helsinki. All subjects provided written,
informed consent before participating in this study.
32 premenopausal women ranging in age from 45 to 51 years
(mean age, 47.75 ± 1.55 years) and 25 perimenopausal women
ranging in age from 49 to 55 years (mean age, 51.60 ± 1.63
years) were finally recruited. All participants were engaged in
ordinary office work, with no obvious statistical differences in
educational level and physical ability.
Sex Hormone Level Measurements
All participants underwent sex hormone level testing, which
included prolactin (PRL), follicle-stimulating hormone (FSH),
luteotropic hormone (LH), estradiol (E2), free testosterone (free-
T), and progesterone (P). Measurements were performed at 3e5
days after the onset of menstruation (in the early follicular
phase). For women with amenorrhea, there was no such re-
striction on the testing time. Participants were instructed to
remain calm and get a good night’s sleep before the test. Testing
was carried out between 8:30 and 9:30 AM, which included
venous blood collection, serum separation, and measurement of
the 6 sex hormones by chemiluminescent immunoassay (E170
Immunology Analyzer; Roche Diagnostics, Basel, Switzerland).
fMRI Acquisition and Processing
fMRI was done using a 3.0-T magnetic resonance scanner
with an 8-channel head array coil (Discovery MR 750; GE
J Sex Med 2019;-:1e10
Sex Hormones and Brain Connectivity in Sexual Function of Perimenopausal Women 3

Table 1. MNI coordinates of the 10 selected ROIs

MNI coordinates Interoception,
Sexuality Z
Label Location Abbreviation Hemisphere X Y Z score

19 Middle frontal gyrus A46_L Left 28 56 12 1.66

42 Orbital gyrus A14m_R Right 7 54 7 5.81
49 Orbital gyrus A13_L Left 10 18 19 6.14
50 Orbital gyrus A13_R Right 9 20 19 3.29
97 Inferior temporal A37vl_L Left 55 60 6 1.75
gyrus
108 Fusiform gyrus A37lv_R Right 43 49 19 2.75
134 Superior parietal A7ip_R Right 31 54 53 4.97
lobule
171 Insula gyrus dIg_L Left 38 8 8 4.21
187 Cingulate gyrus A32sg_L Left 4 39 2 1.92
202 Lateral occipital cortex V5_MTþ_R Right 48 70 1 1.96
A46 ¼ area 46; A14m ¼ medial area 14; A13 ¼ area 13; A37vl ¼ ventrolateral area 37; A37lv ¼ lateroventral area37; A7ip ¼ intraparietal area 7; A32sg ¼
subgenual area 32; dIg ¼ dorsal granular insula; L ¼ left; MNI ¼ Montreal Neurological Institute; R ¼ right; V5/MTþ ¼ visual area 5/middle temporal
complex.

Healthcare, Chicago, IL, USA) . The participants were 3e7 days
postmenstrual period (in the luteal phase) for the fMRI scan. For
participants with amenorrhea, there was no such time restriction.
Subjects were scanned in the supine, head-first position with
cushions on both sides and at top of the head to control motion
confounding. A 3-dimensional brain volume (T1 3D-BRAVO)
scan was first performed to exclude substantial lesions in brain
with the following parameters: repetition time (TR)/echo time
(TE)/inversion time (TI) ¼ 6.656/2.928 /450 ms, field of view
(FOV) ¼ 240 mm  240 mm, slice thickness ¼ 1 mm, slice
gap ¼ 1 mm, matrix ¼ 256  256, number of signal averages
(NEX) ¼ 1, flip angle (FA) ¼ 12 , and bandwidth ¼ 122.07
Hz. Before resting-state BOLD examination, all subjects were
instructed to open their eyes, breathe calmly, keep a clear con-
sciousness and not engage in any specific thinking activity. To
acquire BOLD-fMRI data, a gradient recalled echo-echo planar
imaging sequence was used with the following parameters: TR/
TE ¼ 2,000/30 ms, scan duration ¼ 480 seconds (240 vol-
umes), FOV ¼ 240 mm  240 mm, matrix ¼ 64  64, slice
thickness ¼ 3.5 mm, slice gap ¼ 1.2 mm, FA ¼ 90 , and 33
axial slices covering the whole brain.
Data Processing and Analysis for Brain Imaging (DPABI,
version 3.0; http://restfmri.net/forum/DPARSF) was used for
fMRI data preprocessing, FC calculation, and statistical analysis.
DPABI is based on Statistical Parametric Mapping 12 software
running on the Matrix Laboratory platform (MATLAB R2014b;
MathWorks, Cambridge, MA, USA).
Data preprocessing steps began with data conversion, con-
verting the original DICOM format to NIFTI format, discarding
of the first 10 time points, and slice timing and realignment.
Head motion excluding thresholds were set. fMRI data with >1
mm mean translation, 1 mean rotation, or 0.5 mm mean
framewise displacement (FD) were discarded. 9 participants were
excluded and were contacted to undergo fMRI rescans. Finally,
all the participants passed the head motion excluding thresholds.
The tested data were normalized to the Montreal Neurological
Institute (MNI) space using the Diffeomorphic Anatomical
Registration Through Exponentiated Lie Algebra algorithm.
Smoothing with a Gaussian kernel of 6 mm full-width half-
maximum and the linear trend of time courses was assessed.
Head motion and other nuisance covariates (eg, white matter,
cerebrospinal fluid signal) were removed, and bandpass temporal
filtering (0.01e0.08 Hz) was performed. Scrubbing was
conducted.
The preprocessed data were used for FC calculation. In this
study, voxel-wise FC values were obtained by calculating Pear-
son’s correlation coefficient between regions of interest (ROIs)
and rest of the brain. For ROI selection, brain regions related to
sexual function were selected from the brainnetome atlas
(BNA)21 based on the BrainMap meta-analysis.22 BrainMap’s
meta-data category label “Interoception, Sexuality” was selected
to evaluate the association of brain regions with sexual func-
tion.22 The “Interoception, Sexuality” label describes the need
for sexual activity as tested by fMRI contrasts in various exper-
imental designs, including 3 major sexual arousal elements:
watching erotic videos, listening to scripts of autobiographical
events characterized by maximally pleasant sexual arousal, and
viewing sexual images.23e25 In other words, the “Interoception,
Sexuality” label describes the possibility of brain regions being
activated in response to sexual stimuli. The dataset for this label
contains 50 relevant publications with hundreds of subjects,
including Chinese men and women.23,24 Then forward and
reverse inferences were used for functional characterization of
each subregions in the BNA, with significance assessed by sta-
tistical tests.21 The resulting Z scores indicated the closeness to
sexual function. A higher Z score indicated that the relevant

J Sex Med 2019;-:1e10

4 Lu et al
subregion was more likely associated with sexual function.21
Finally, 10 subregions in the cortical areas with highest Z
scores were selected as the ROIs. Table 1 provides locations and
Z scores for “Interoception, Sexuality” of the selected 10 ROIs.
The 10 ROIs were the left area 46, right medial area 14, left area
13, right area 13 (A13_R), left ventrolateral area 37, right lat-
eroventral area 37, right intraparietal area 7, left dorsal granular
insula (dIg_L), left subgenual area 32, and right visual area 5/
middle temporal complex. The locations of the 10 ROIs are
shown in Figure 1. FCs between each ROI and the rest of the
brain were calculated with Pearson’s correlation coefficient.
Whole-brain FC maps of the 10 ROIs were obtained.
Quality Control and Statistical Analysis
Head motion has a significant impact on the interpretation of
fMRI results.26 For quality control (QC) of the study, head
translation, rotation, and FD excluding thresholds were set, but
the mean absolute value of 9 head motionerelated parameters
(x-, y-, and z-axis translations; pitch; roll; yaw; FD by Jenkinson;
FD by Power; and FD by Van Dijk)26 between premenopausal
and perimenopausal women were also compared using an inde-
pendent t test in SPSS (IBM, Armonk, NY, USA). The mean
absolute value of x- and z-axis translation and 3 FD metrics
showed significant differences between the premenopausal and
perimenopausal groups. These 9 parameters were also considered
as nuisance covariates in statistical analysis, and the results were
almost the same. The statistical analysis of head motion param-
eters is summarized in Table 2.
Figure 1. Ten regions of interest selected from the brainnetome atlas: sagittal view (A), coronal view (B), axial view (C). Figure 1 is available
in color online at www.jsm.jsexmed.org.
The QC module of DPABI was used for a quality check of
fMRI data. Visual inspections of raw T1-weighted and raw
functional images and normalization were performed by an
experienced doctor from the affiliated hospital. A QC score 1e5
was assigned based on the quality of raw T1-weighted and raw
functional images and normalization of each subject. fMRI data
with three points scored at >4 were qualified for statistical
analysis; otherwise relevant steps were redone to guarantee high-
quality raw data and accurate registration. Finally, all data
reached the QC thresholds.
SPSS 20.0 was used to run independent t tests for de-
mographic information between premenopausal and perimeno-
pausal women. The ManneWhitney U test was used to compare
the levels 6 sex hormones between the 2 groups. Statistical sig-
nificance was set at P < .05.
The statistical analysis module of DPABI was used to analyze
FC maps. Specifically, a general linear model was used to detect
any significant statistical differences in FC maps between pre-
menopausal and perimenopausal women. Age was treated as a
covariate and entered into the design matrix. For multiple
comparison correction, Gaussian random field (GRF) correction
with voxel level P < .001 (Z > 3.29) and cluster level P < .025
was used. The threshold of cluster size was set at 50 voxels.
Correlation Analyses
Correlation analyses between FC values and sex hormone
levels were conducted in the perimenopausal group. After group
J Sex Med 2019;-:1e10
Sex Hormones and Brain Connectivity in Sexual Function of Perimenopausal Women 5

Table 2. Mean absolute value of head motion characteristics in the premenopausal women demonstrated a decreased FC between
premenopausal and perimenopausal women A13_R and the right medial superior frontal gyrus (mSFG), a
Premenopausal Perimenopausal decreased FC between dIg_L and the right superior frontal gyrus
Head motion women women P value (SFG) (GRF corrected at voxel level P < .001 and cluster level P <
.025). Figure 2 shows the location of these clusters in the standard
x , mm 0.16 ± 0.35 0.12 ± 0.15 .005 brain. The brain regions that have significant FC differences with
y , mm 0.14 ± 0.21 0.14 ± 0.25 .320 A13_R and dIg_L are identified in Table 4 by the BNA region, the
z , mm 0.36 ± 0.46 0.22 ± 0.21 .011
volume of brain region with FC differences, MNI coordinates, the
Pitch, degree 0.005 ± 0.015 0.004 ± 0.01 .424
difference between 2 groups, and the type of FC changes seen in
Roll, degree 0.005 ± 0.001 0.005 ± 0.002 .760
premenopausal group. The difference is designated the “T value,”
Yaw, degree 0.005 ± 0.007 0.005 ± 0.001 .072
FD_Jenkinson, 0.09 ± 0.19 0.04 ± 0.05 <.001
representing the calculated t statistic. A higher T value represents a
mm greater difference between the 2 groups.
FD_Power, mm 0.16 ± 0.34 0.08 ± 0.10 <.001
FD_VanDijk, 0.04 ± 0.09 0.02 ± 0.03 <.001 Correlation Analyses
mm The results of correlation analyses are shown in Figure 3. FC
FD ¼ framewise displacement. between the A13_R and the right mSFG showed a negative
correlation with PRL level (r ¼ -0.578; P ¼ .003). Of note,
analysis, the regions showing significant FC changes between the when excluding 3 perimenopausal outliers, the correlation was
2 groups were identified, and the mean FC of each region was attenuated (r ¼ -0.161). FC between the dIg_L and the right
extracted in the perimenopausal group. Spearman partial corre- SFG also had a negative correlation with PRL level (r ¼ -0.402;
lation analyses were conducted to evaluate the relationship be- P ¼ .05). The negative correlation between FC with the dIg_L
tween the mean FC values of these regions and sex hormone and PRL was also attenuated (r ¼ -0.400) when 3 perimeno-
levels, with age considered a nuisance covariate. pausal outliers were excluded.

RESULTS
Demographic and Clinical Data
Demographic information and sex hormone levels in the pre-
menopausal and perimenopausal women are presented in Table 3.
There was a significant difference in age but no significant dif-
ference in educational level between the 2 groups. There were
significant between-group differences in the levels of PRL, FSH,
LH, E2, and P, but no significant difference in free-T level.
FC Changes
There were no significant differences in FC maps of 8 selected
ROIs: the left area 46, right medial area 14, left area 13, left
ventrolateral area 37, right lateroventral area 37, right intraparietal
area 7, left subgenual area 32, and right visual area 5/middle
temporal complex. Compared with perimenopausal women,
DISCUSSION
Functional Connectivity
Sex hormones are well known for their involvement in pro-
creation and sexual maturity; however, recent neuroendocri-
nology studies have demonstrated their roles in brain plasticity as
well.27e29 In this study, resting-state FC was used to assess
functional changes in brain connectivity related to sexual func-
tion. Brain regions related to sexual function were selected based
on the Z score of BrainMap’s meta-data category label “Inter-
oception, Sexuality,” which refers to the need for sexual activity.
Of note, sexual function status was not assessed for the enrolled
subjects. In other words, ROI selection was based totally on a
meta-analysis of previous studies without the clinical support of
subjects’ sexual function. This is because broadly used measures,
such as the Female Sexual Function Index (FSFI), have not been

Table 3. Demographic information and sex hormone levels between premenopausal and perimenopausal women
Parameter Premenopausal women* Perimenopausal women* P value†

Age, y 47.75 ± 1.55 51.60 ± 1.63 <.001

Education, y 16.18 ± 1.31 15.84 ± 0.80 .248
PRL, ng/mL 21.67 ± 17.71 12.93± 10.68 .011
FSH, mIU/mL 5.74 ± 4.21 52.63 ± 29.75 <.0001
LH, mIU/mL 7.31 ± 13.13 22.12 ± 11.32 .0013
E2, pg/mL 109.70 ± 64.55 35.70 ± 22.70 <.0001
Free-T, ng/dL 28.47 ± 16.35 19.13 ± 1.27 .403
P, ng/mL 2.37 ± 4.23 0.22 ± 0.06 .001
E2 ¼ estradiol; FSH ¼ follicle-stimulating hormone; LH ¼ luteotropic hormone; P ¼ progesterone; PRL ¼ prolactin; T ¼ testosterone.
*Normal hormonal ranges: PRL, 1.9e25 ng/mL; FSH, 1.2e153 mIU/mL; LH, 1.1e77 mIU/mL; E2, 20e400 pg/mL; free-T, 0e73 ng/dL; P, 0.1e24 ng/mL.

J Sex Med 2019;-:1e10

6 Lu et al

Figure 2. Two-sample functional connectivity (FC) T-map between right area 13 (BNA50), left dorsal granular insula (BNA171), and the
rest of the brain between premenopausal and perimenopausal women with sagittal, coronal, and axial views. (A) FC t-map between A13_R
and the voxels of the remaining brain. (B) FC t-map between dIg_L and the voxels of the remaining brain. A t test was performed using
Gaussian random field (GRF) correction (voxel level, P < .001; cluster level, P < .025) for multiple comparisons; the color bar indicates
T-score. Figure 2 is available in color online at www.jsm.jsexmed.org.

validated in Chinese owing to conservative attitudes toward
sexual issues of Chinese women.30 Moreover, the Chinese
version of the FSFI has remained a preliminary version without
wide adoption.30 Therefore, clinical assessment of differences in
sexual function between Chinese premenopausal and perimen-
opausal women remains difficult.
Nevertheless, the integrity of the hormonal milieu is assumed
to be closely related to sexual function.31 Therefore, correlation
analyses were used to assess the relationship between serum sex
hormone levels and aberrant FCs in women approaching
menopause as a supplement. The results revealed that certain
regions related to sexual function experienced FC changes in
women approaching menopause, and that aberrant sex hormone
levels might account for the FC abnormalities.
Increased FC between A13_R and the right mSFG was found
in perimenopausal women compared with premenopausal
women. A13_R is part of the orbitofrontal cortex (OFC), also
known as the gyrus rectus, which involves in reward, punish-
ment, and decision.32 In terms of sexual function, A13_R is
related to pleasure, sexual reward, orgasm-related activity, mate
selection, and facial attractiveness in women.5,6,32,33 Specifically,
Rupp et al33 found increased neural activation in the right OFC

Table 4. FC of A13_R (BNA50) and dIg_L (BNA171) in premenopausal women compared with perimenopausal women (Gaussian random
fieldecorrected at voxel level P < .001, cluster level P < .025)
MNI coordinates
Type of FC
ROI Brain region BNA Cluster size X Y Z T value change

A13_R mSFG_R 14 80 9 63 9 5.2159 Decreased FC

dIg_L SFG_R 16 204 27 48 39 5.9077 Decreased FC
A13 ¼ area 13; BNA ¼ brainnetome atlas; dIg ¼ dorsal granular insula; L ¼ left; MNI ¼ Montreal Neurological Institute; FC ¼ functional connectivity; mSFG ¼
medial superior frontal gyrus; R ¼ right; ROI ¼ region of interest; SFG ¼ superior frontal gyrus.

J Sex Med 2019;-:1e10

Sex Hormones and Brain Connectivity in Sexual Function of Perimenopausal Women 7

Figure 3. Correlation analyses results between FC values and sex hormone levels. (A) Scatter plots between FC of the right area 13
(A13_R) with the right medial superior frontal gyrus (mSFG) and the level of prolactin (PRL). (B) Scatter plots between FC of the left dorsal
granular insula (dIg_L) with the right superior frontal gyrus (SFG) and the level of PRL. Figure 3 is available in color online at www.jsm.
jsexmed.org.

for women in response to male faces across the menstrual cycle. A
study has reported increased sex desires or even hypersexuality in
patients with lesions of the medial OFC,34 demonstrating a
relationship between OFC activation and sexual function. Pre-
vious studies have linked stronger activation of the gyrus rectus
(mOFC) to hypoactive sexual disorder (sexual dysfunction
involving lack of desire for sexual activity).35,36 In line with the
previous studies,5,6,32,33 the increased FC between A13_R and
the right mSFG may reflect a high prevalence of sexual
dysfunction in women during the transition to menopause.
The SFG, including the supplementary motor area (SMA) and
pre-SMA, has been implicated in the motor control of sexual
behavior, including erectile responses of heterosexual males and
premature ejaculation.6,37 Kuhtz-Buschbeck et al38 reported
significant activation of the SMA during pelvic floor muscle
contractions and linked the SMA to pelvic floor muscle control.
The pelvic floor muscle plays a crucial role in sexual arousal and
orgasm in both males and females.37,38 In accordance with
previous studies,37,38 the increased FC between A13_R and the
right mSFG may affect the central motor control of sexual
motivation and may be a potential cause of pelvic floor muscle
hypotonus and sexual dysfunction in perimenopausal women.
This finding was also consistent with a former task-fMRI study
reporting a greater enhancement of the SFG signal in response to
visual erotic stimuli in menopausal women compared with pre-
menopausal women.4
In addition, abnormal FC between the gyrus rectus and SFG is
often seen in patients with depression and schizophrenia with
emotional and memory impairment.39,40 During menopausal
transition, aberrant hormonal levels and other risk factors, such
as poor sleep, age, and stress, can lead to depressive symptoms.10
Freeman10 has proposed that the perimenopausal period opens a
“window of vulnerability” for women. Therefore, the increased
FC between A13_R and the right mSFG that we observed may
reflect an increased risk of depression during the transition from
premenopause to postmenopause, which closely corresponds to
the findings reported by Freeman.10
In the present study, increased FC between dIg_L and the
right SFG in was observed in the perimenopausal group. The
dIg_L is part of the insula cortex, which is involved in the
integration of interoception, affection, and cognition.41 Human
insula is believed to be responsible for emotional feelings, such as
sexual arousal, maternal love, and romantic love.5,42,43 Several
researchers have reported functional abnormalities in the insula
in females in the process of sexual arousal.4,5,8,44 The increased
FC in perimenopausal women may reflect the fact that the
transition to menopause increases the risk of sexual dysfunc-
tion,12,13 which was also supported by sex hormone level testing
in the 2 groups. Furthermore, this finding is also consistent with
the study by Jeong et al,4 in which neurologic activation of the
SFG in response to erotic visual stimuli was greater in meno-
pausal women compared with premenopausal women. Georgia-
dis et al45 reported that insula-frontal connectivity is related to
sexual arousal and dearousal during the sexual cycle. Therefore,
the abnormal FC in perimenopausal women in this study may be
the potential cause of difficulties facing sexual arousal in

J Sex Med 2019;-:1e10

8 Lu et al
perimenopausal women. However, to confirm this prediction,
further investigation is needed using clinical information on
sexual function.
Correlation Analyses
Sex hormones act throughout the entire brain with the
participation of neuroendocrine phenomena and establishment
of behavioral patterns, including mood and cognition.29 Hor-
mone receptor expression has been identified in several regions of
the brain, including the cerebral cortex, hypothalamus, and
hippocampus.46 These receptors include LH, FSH,28 and PRL46
receptors. Studies have also identified increased activation of
several brain regions after hormone therapy.47
In this study, we found significant negative correlations between
FC of the A13_R with the right mSFG and PRL level and between
FC values of the dIg_L with the right SFG and PRL level. PRL is a
versatile sex hormone related to such neuroactivities as stress
adaptation, neurogenesis, cognition, and emotional responses
including anxiety and depression.48 PRL may profoundly impact
brain structures and functions via PRL receptor expression.27,29,46
Indeed, previous studies have found PRL receptor expression in
several brain regions, including the OFC, frontal cortex, and
cingulate cortex.27,46 Seo et al36 have reported significant associ-
ations between PRL concentration and the activity in several brain
regions involved in sexual behavior, including the SFG and cere-
bellum. In line with these previous studies, the negative correla-
tions between FC values and PRL levels suggest that certain sexual
hormones might affect specific brain connectivities related to
sexual function and the brain’s sexual network. Of note, outlier
removal attenuated the correlation results; therefore, the negative
correlations may be due to the effect of outliers.
Study Limitations
This study has several limitations. First, the study lacked
clinical information related to participants’ sexual function,
which will be taken into consideration in future study designs.
Second, this study’s cross-sectional design is not optimal; a
longitudinal study is preferred in the future to understand how
FC may change during the process of menopausal transition.
Finally, our conclusions may be limited by our relatively small
sample size; however, using strict statistical multiple comparison
correction methods like GRF guaranteed good reliability for the
fMRI study.
CONCLUSION
In this study, resting-state fMRI was applied to assess func-
tional changes in brain connectivities related to sexual function.
Correlation analyses were used to evaluate the relationship be-
tween serum sex hormone levels and brain function in women
approaching menopause. The results demonstrate connections
between intrinsic brain FC and serum sex hormone levels in
perimenopausal women. In addition, our findings suggest that
brain networks related to sexual function might exhibit abnor-
malities in women approaching menopause, and may provide an
explanation of the sexual dysfunction seen in many in peri-
menopausal women. Future studies with relevant clinical infor-
mation, larger sample sizes, and a longitudinal design are needed
to investigate the relationship between sex hormone levels and
brain connectivities related to sexual function in the process of
menopause.
Corresponding Author: Jianfeng Qiu, PhD, No. 619 Chang-
cheng Road, Tai’an, Shandong Province, China 271016. Tel:
0538-6222138; Fax: 0535-6222138; E-mail: jfqiu100@gmail.
com
Conflicts of interest: The authors report no conflicts of interest.
Funding: This work was supported by the National Key Research
and Development Program of China (Grant 2016YFC
0103400), Key Research and Development Program of Shan-
dong Province (Grant 2017GGX201010), and Taishan Scholars
Program of Shandong Province (Grant TS201712065).
STATEMENT OF AUTHORSHIP
Category 1
(a) Conception and Design
Weizhao Lu, Wei Guo; Jianfeng Qiu
(b) Acquisition of DataWei Guo; Kejiang Dong
(c) Analysis and Interpretation of DataWeizhao Lu; Dong Cui
Category 2
(a) Drafting the ArticleWeizhao Lu; Wei Guo; Kejiang Dong
(b) Revising It for Intellectual ContentWeizhao Lu; Dong Cui;
Jianfeng Qiu
Category 3
(a) Final Approval of the Completed ArticleWeizhao Lu; Wei Guo;
Dong Cui; Kejiang Dong; Jianfeng Qiu
REFERENCES
1. Cahill L. Why sex matters for neuroscience. Nat Rev Neurosci
2006;7:477-484.
2. Stoléru S, Grégoire MC, Gérard D, et al. Neuroanatomical
correlates of visually evoked sexual arousal in human males.
Arch Sex Behav 1999;28:1-21.
3. Gizewski ER, Krause E, Karama S, et al. There are differences
in cerebral activation between females in distinct menstrual
phases during viewing of erotic stimuli: A fMRI study. Exp
Brain Res 2006;174:101-108.
4. Jeong GW, Park K, Youn G, et al. Assessment of cere-
brocortical regions associated with sexual arousal in premen-
opausal and menopausal women by using BOLD-based
functional MRI. J Sex Med 2010;2:645-651.
5. Ruesink GB, Georgiadis JR. Brain imaging of human sexual
response: Recent developments and future directions. Curr
Sex Health Rep 2017;9:183-191.
J Sex Med 2019;-:1e10
Sex Hormones and Brain Connectivity in Sexual Function of Perimenopausal Women
6. Georgiadis JR, Kringelbach ML. The human sexual response
cycle: Brain imaging evidence linking sex to other pleasures.
Prog Neurobiol 2012;98:49-81.
7. Maravilla KR, Deliganis AV, Heiman J, et al. BOLD fMRI
evaluation of normal female sexual arousal response: Sites of
cerebral activation correlated with subjective and objective
measures of arousal. Radiology 2000;214:614.
8. Park K, Kang HK, Seo JJ, et al. Blood-oxygenation-level-
dependent functional magnetic resonance imaging for evalu-
ating cerebral regions of female sexual arousal response.
Urology 2001;57:1189-1194.
9. Mckinlay SM, Brambilla DJ, Posner JG. The normal meno-
pause transition. Maturitas 1992;14:103-115.
10. Freeman EW. Associations of depression with the transition to
menopause. Menopause 2010;17:823-827.
11. Woods NF, Mitchell ES. Symptoms during the perimenopause:
Prevalence, severity, trajectory, and significance in women’s
lives. Am J Med 2005;118:14-24.
12. Novi JM, Book NM. Sexual dysfunction in perimenopause a
review. Obstet Gynecol Surv 2009;64:624-631.
13. Sierra B, Marta E, German R. The prevalence of female sexual
dysfunction and metabolic syndrome in perimenopausal and
menopausal women in Colombia. Presented at the 14th World
Congress on Menopause; May 1e4, 2014; Cancun, Mexico.
14. Avis NE, Stellato R, Crawford S, et al. Is there an association
between menopause status and sexual functioning. Meno-
pause 2000;7:297-309.
15. Gracia CR, Freeman EW, Sammel MD, et al. Hormones and
sexuality during transition to menopause. Obstet Gynecol
2007;109:831-840.
16. Graziottin A, Leiblum SR. Biological and psychosocial patho-
physiology of female sexual dysfunction during the meno-
pausal transition. J Sex Med 2010;2:133-145.
17. Kim GW, Jeong GW. Menopause-related brain activation pat-
terns during visual sexual arousal in menopausal women: An
fMRI pilot study using time-course analysis. Neuroscience
2017;343:449-458.
18. Kim TH, Kang HK, Park K, et al. Localized brain metabolite
changes during visual sexual stimulation in postmenopausal
women: A pilot study using functional magnetic resonance
spectroscopy. Menopause 2013;21:59-66.
19. Berent-Spillson A, Briceno E, Pinsky A, et al. Distinct cognitive
effects of estrogen and progesterone in menopausal women.
Psychoneuroendocrinology 2015;59:25-36.
20. Harlow SD, Gass M, Hall JE, et al. Executive summary of the
Stages of Reproductive Aging Workshop þ10: Addressing the
unfinished agenda of staging reproductive aging. Fertil Steril
2012;97:843-851.
21. Fan L, Li H, Zhuo J, et al. The human brainnetome atlas: A new
brain atlas based on connectional, architecture. Cereb Cortex
2016;26:3508-3526.
22. Laird AR, Lancaster JJ, Fox PT. BrainMap. Neuroinformatics
2005;3:65-77.
J Sex Med 2019;-:1e10
9
23. Zhu X, Wang X, Parkinson C, et al. Brain activation evoked by
erotic films varies with different menstrual phases: An fMRI
study. Behav Brain Res 2010;206:279-285.
24. Hu SH, Wang QD, Xu Y, et al. Hemodynamic brain response to
visual sexual stimuli is different between homosexual and
heterosexual men. J Int Med Res 2011;39:199-211.
25. Stark R, Schienle A, Girod C, et al. Erotic and disgust-inducing
pictures—differences in the hemodynamic responses of the
brain. Biol Psychol 2005;70:19-29.
26. Yan CG, Craddock RC, He Y, et al. Addressing head motion
dependencies for small-world topologies in functional con-
nectomics. Front Hum Neurosci 2013;7:910.
27. Shun Y, Jian S, Junfeng G, et al. Cognitive function and serum
hormone levels are associated with gray matter volume
decline in female patients with prolactinomas. Front Neurol
2018;8:742.
28. Lu W, Guo W, Hou K, et al. Gray matter differences associated
with age and sex hormone levels between premenopausal and
perimenopausal women, a voxel-based morphometry study.
J Neuroendocrinol 2018;30:e12655.
29. Lentini E, Kasahara M, Arver S, et al. Sex differences in the
human brain and the impact of sex chromosomes and sex
hormones. Cereb Cortex 2013;23:2322-2336.
30. Sun X, Li C, Jin L, et al. Development and validation of Chinese
version of female sexual function index in a Chinese pop-
ulation—a pilot study. J Sex Med 2011;8:1101-1111.
31. Goldstein I, Traish A, Kim N, et al. The role of sex steroid
hormones in female sexual function and dysfunction. Clin
Obstet Gynecol 2004;47:471-484.
32. Spinella M. The role of prefrontal systems in sexual behavior.
Int J Neurosci 2007;117:369-385.
33. Rupp HA, James TW, Ketterson ED, et al. Neural activation in
the orbitofrontal cortex in response to male faces increases
during the follicular phase. Horm Behav 2009;56:66-72.
34. Politis M, Loane C, Wu K, et al. Neural response to visual
sexual cues in dopamine treatment-linked hypersexuality in
Parkinson’s disease. Brain 2013;136:400-411.
35. Stoléru S, Redouté J, Costes N, et al. Brain processing of visual
sexual stimuli in men with hypoactive sexual desire disorder.
Psychiat Res 2003;124:67-86.
36. Holstege G. How the emotional motor system controls the
pelvic organs. Auton Neurosci 2016;4:303-328.
37. Rosenbaum TY. Pelvic floor involvement in male and female
sexual dysfunction and the role of pelvic floor rehabilitation in
treatment: A literature review. J Sex Med 2007;4:4-13.
38. Kuhtz-Buschbeck JP, Van der Horst C, Wolff S, et al. Activation
of the supplementary motor area (SMA) during voluntary
pelvic floor muscle contractions: An fMRI study. Neuroimage
2007;35:449-457.
39. Ballmaier M, Toga AW, Blanton RE, et al. Anterior cingulate,
gyrus rectus, and orbitofrontal abnormalities in elderly
depressed patients: An MRI-based parcellation of the pre-
frontal cortex. Am J Psychiat 2004;161:99-108.
40. Zhao C, Zhu J, Liu X, et al. Structural and functional brain
abnormalities in schizophrenia: A cross-sectional study at
10
different stages of the disease. Prog Neuro-Psychoph 2017;
83:27-32.
41. Hong JY, Kilpatrick LA, Labus JS, et al. Sex and disease-
related alterations of anterior insula functional connectivity in
chronic abdominal pain. J Neurosci 2014;34:14252-14259.
42. Clark L, Bechara A, Damasio H, et al. Differential effects of
insular and ventromedial prefrontal cortex lesions on risky
decision-making. Brain 2008;131:1311-1322.
43. Vilares I, Howard JD, Fernandes HL, et al. Differential repre-
sentations of prior and likelihood uncertainty in the human
brain. Curr Biol 2012;22:1641-1648.
44. Kim JH, Choi SH, Jang JH, et al. Impaired insula functional
connectivity associated with persistent pain perception in
patients with complex regional pain syndrome. PloS One
2017;12:e0180479.
Lu et al
45. Georgiadis JR, Mjboessen F. Dynamic subcortical blood flow
during male sexual activity with ecological validity: A perfusion
fMRI study. Neuroimage 2010;50:208-216.
46. Cabrera-Reyes EA, Limón-Morales O, Rivero-Segura NA, et al.
Prolactin function and putative expression in the brain.
Endocrine 2017;52:199-213.
47. Persad CC, Zubieta JK, Love T, et al. Enhanced neuroactivation
during verbal memory processing in postmenopausal women
receiving short-term hormone therapy. Fertil Steril 2009;
92:197-204.
48. Luz T. Actions of prolactin in the brain: From physiological
adaptations to stress and neurogenesis to psychopathology.
Front Endocrinol 2016;7:25.
J Sex Med 2019;-:1e10