Lan et al.

Reproductive Biology and Endocrinology (2020) 18:12

https://doi.org/10.1186/s12958-020-0569-5

RESEARCH Open Access

Levels of sex steroid hormones and their

receptors in women with preeclampsia
Kuo-Chung Lan1,2, Yun-Ju Lai1, Hsin-Hsin Cheng1, Ni-Chin Tsai1,3, Yu-Ting Su1, Ching-Chang Tsai1* and
Te-Yao Hsu1*

Abstract
Background: Pregnant women have high serum concentrations of sex steroid hormones, which are major
regulators of paracrine and autocrine responses for many maternal and placental functions. The main purpose of
this study was to compare patients with preeclampsia and patients with uncomplicated pregnancies in terms of
serum steroid hormones (estradiol [E2], progesterone [P4], dehydroepiandrosterone sulfate [DHEAS], and
testosterone [T]) throughout pregnancy and the levels of cord blood and placental steroid receptors during the
third trimester.
Methods: Quantitative real-time reverse transcription PCR, western blotting, and immunohistochemistry were used
to determine the levels of steroid hormones in the serum and cord blood and the placental levels of estrogen
receptor-α (ERα), ERβ, androgen receptor (AR), and progesterone receptor (PR).
Results: There were 45 women in the uncomplicated pregnancy group and 30 women in the preeclampsia group.
Serum levels of T were greater and serum levels of E2 were reduced in the preeclampsia group, but the two
groups had similar levels of P4 and DHEAS during the third trimester. Cord blood had a decreased level of DHEAS
in the preeclampsia group, but the two groups had similar levels of P4, E2, and T. The two groups had similar
placental mRNA levels of ERα, ERβ, AR, and PR, but the preeclampsia group had a higher level of ERβ protein and a
lower level of ERα protein. Immunohistochemistry indicated that the preeclampsia group had a greater level of ERβ
in the nucleus and cytoplasm of syncytiotrophoblasts and stromal cells.
Conclusions: Women with preeclampsia had lower levels of steroid hormones, estrogen, and ERα but higher levels
of T and ERβ. These molecules may have roles in the pathogenesis of preeclampsia.
Keywords: Preeclampsia, Placenta, Sex steroids, Steroid receptor

Background to placental ischemia and the release of antiangiogenic

Preeclampsia is a pregnancy-specific hypertensive dis- factors from the ischemic placenta into the maternal cir-
order with multisystem involvement that is associated culation, thus contributing to endothelial damage [2].
with an increased risk of future cardiovascular disease During pregnancy, the placenta is the primary endocrine
[1]. The pathogenesis of preeclampsia has not been fully organ for maintaining pregnancy and fetal growth. The
elucidated, but much progress has been made in recent placenta releases hormones, including androgens, estro-
decades [2]. In particular, it is widely accepted that pre- gens, and progestogens, and these hormones occur at
eclampsia is associated with abnormal placentation, extremely high concentrations in the maternal circulation.
reduced placental perfusion, and systemic vasospasm. A These hormones are important paracrine and autocrine
two-stage model of preeclampsia proposed that incom- regulators that affect the growth and differentiation of the
plete spiral artery remodeling in the uterus contributes placental trophoblast, growth and maturation of the
placental vascular tree, and uterine endovascular invasion
* Correspondence: [email protected]; [email protected] by the extravillous cytotrophoblast [3, 4].
1
Department of Obstetrics and Gynecology, Kaohsiung Chang Gung
Memorial Hospital and Chang Gung University College of Medicine, 123
Progesterone and estrogens also function as modula-
Ta-Pei Road, Niao-Sung District, Kaohsiung City, Kaohsiung, Taiwan tors of uterine vessels in that they decrease the resistance
Full list of author information is available at the end of the article

© The Author(s). 2020 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0
International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and
reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to
the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver
(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
Lan et al. Reproductive Biology and Endocrinology (2020) 18:12 Page 2 of 7

of the spiral uterine arteries and modulate the synthesis samples during the first (gestational age: 8–14 weeks), sec-
and release of angiogenic factors by placental cells. Andro- ond (gestational age: 20–24 weeks), and third (gestational
gens have the opposite effect. Previous studies have com- age: 32–38 weeks) trimesters of pregnancy and to provide
pared the serum levels of sex steroid hormones and cord blood and placenta samples during delivery (Fig. 1).
signaling in women who had preeclampsia with women Among the 50 participants who donated 3 serum samples,
who had uncomplicated pregnancies [4–7]. Unsurpris- 45 women had uncomplicated pregnancies and 5 women
ingly, women with preeclampsia exhibited altered serum (10%) developed preeclampsia. During the study period,
concentrations of sex steroid hormones. However, no 25 women with preeclampsia during the second or third
studies have yet examined the role of placental steroid trimesters were referred from other hospitals to the high-
hormone receptors and cord blood steroid hormone con- risk pregnancy prenatal care center of our institute. Thus,
centrations in the pathogenesis of preeclampsia. there were 45 women in the uncomplicated pregnancy
The main purpose of the present study is to compare group and 30 women in the preeclampsia group.
pregnant women with preeclampsia and women with Women in the uncomplicated pregnancy group had no
uncomplicated pregnancies in terms of the serum and diagnoses of preeclampsia or hypertension during preg-
cord blood levels of sex steroid hormones (estradiol nancy and no hospitalization due to premature delivery or
[E2], progesterone [P4], dehydroepiandrosterone sulfate bleeding. Women in the preeclampsia group were diag-
[DHEAS], and testosterone [T]) and placental steroid nosed based on the presence of new-onset hypertension
receptors during the third trimester. (systolic blood pressure of 140 mmHg or more, diastolic
blood pressure of 90 mmHg or more, or both) on two oc-
Materials and methods casions 6 h or more apart after 20 weeks of gestation and
Study design the presence of significant proteinuria (≥300 mg/24 h).
Following approval of the Ethics Committee of the Institu- None of the patients had a previous history of any known
tional Review Board of Chang Gung Memorial Hospital endocrine disorder. Women were excluded if they were
(CGMH 201601484A3) and obtainment of informed writ- smokers, alcoholics, or had chronic maternal disease
ten consent from all subjects, this prospective study was (essential hypertension, connective tissue diseases, hyper-
conducted from June 2017 to January 2019. Singleton thyroidism, hypothyroidism, chronic glomerulonephritis,
pregnant women (n = 78) were recruited to donate blood renal failure, and diabetes mellitus) or gestational diabetes.

Fig. 1 Flow chart of the study population

Lan et al. Reproductive Biology and Endocrinology (2020) 18:12 Page 3 of 7

Tissue and blood collection RNA extraction and quantitative real-time reverse
Human placental tissues and blood samples of the transcription PCR
healthy group (n = 45) and the preeclampsia group Total RNA was isolated from placental tissues using the
(n = 30) were collected and immediately stored at − RNA Clean & Concentrator-5 kit (R1014, Zymo Re-
80 °C. Blood was collected in plastic tubes under search, CA, USA) and reverse transcribed. Quantitative
aseptic conditions, with EDTA as an anticoagulant, real-time reverse transcription-PCR (qRT-PCR) was per-
and centrifuged at 18,472×g for 10 min at 4 °C to formed using Fast SYBR® Green Master Mix (Applied
separate the serum. Serum concentrations of E2, P4, Biosystems, CA, USA) and the ABI 7500 Fast Real-Time
testosterone and dehydroepiandrosterone sulfate PCR System (Applied Biosystems), with the primers
(DHEA-S) were analyzed using commercially available listed in Table 1.
immunoassay systems (ADVIA Centaur XP; Siemens
USA). The intraassay and interassay coefficients of
Statistical analysis
variation were 5.0 and 4.1%, respectively, for E2; 5.2
SPSS version 10.0 (SPSS, Inc., Chicago, IL, USA) was
and 3.5% for P4; and 2.6 and 4.3% for testosterone.
used for data analysis. Continuous data are summarized
The intraassay coefficient of variation was 3.9% for
as the mean ± standard deviation. Visual inspection and
DHEA-S.
the Shapiro-Wilk normality test were used to check for
normality of distributions. The Mann-Whitney rank-
Immunohistochemistry sum test was used for the comparison of means.
Tissues were paraffin-embedded and subjected to immu- Categorical variables, reported as proportions, were
nohistochemical staining. For this procedure, 4-μm compared using the chi-square test or Fisher’s exact test,
sections were deparaffinized and rehydrated, rinsed in as appropriate, and a P value below 0.05 was considered
purified water, and treated with 3% H2O2 for 15 min at statistically significant.
room temperature. After rinsing three times with puri-
fied water, they were heated in an autoclave for 1 h with
Results
10 mM citrate buffer and then incubated with primary
The preeclampsia group (n = 30) and the uncomplicated
antibodies against estrogen receptor-α (ERα; EP1 diluted
pregnancy group (n = 45) had similar maternal age, par-
1:50, Bio SB, CA, USA), ERβ (14C8 diluted 1:100,
ity, gravidity, and hemoglobin levels (Table 2). However,
Abcam, Cambridge, UK), progesterone receptor (PR;
the preeclampsia group had higher blood pressure, ma-
NCL-L-PGR-312 diluted 1:50, Leica Biosystems, Benton
ternal body mass index, nulliparity proportion and
Lane, UK), and androgen receptor (AR; Clone SP107
markedly decreased infant birth weight and gestational
diluted 1:50, ZECA, CA, USA). After the addition of the
weeks.
appropriate secondary IgG antibody, sections were incu-
Comparison of serum data during the third trimester
bated with DAB (K5007, Dako, Denmark) and counter-
indicated that the uncomplicated pregnancy group had a
stained with hematoxylin and eosin (H&E; 1.05174,
greater level of serum T and a decreased level of E2, but
Merck, MA, USA). Finally, sections were dehydrated in a
the two groups had similar levels of P4 and DHEAS
graded series of ethanol, cleared with xylene, mounted
(Fig. 2a). Comparison of cord blood indicated that the
using Histomount (008030, Life Technologies, MD,
preeclampsia group had a decreased level of DHEAS,
USA), and coverslips were applied for evaluation by light
but the two groups had similar levels of P4, E2, and T
microscopy.
(Fig. 2b).
We determined the expression of placental hormone
Western blot analysis receptors in the two groups using western blotting and
Placental tissues were washed in PBS and lysed in RIPA qRT-PCR. The two groups had similar mRNA levels of
lysis buffer (20–188, Merck, MA, USA). Proteins were ERα, ERβ, AR, and PR (Fig. 3). However, the preeclamp-
separated by electrophoresis using 8% SDS/PAGE and sia group had a greater protein level of ERβ and
then transferred to PVDF Blotting Membranes (10,600,
022, GE Healthcare, Germany). Blots were probed with a Table 1 Primer sequences
primary antibody and then developed using Immobilon™ Forward Reverse
Western (WBKLS0500, Millipore, MA, USA). The pri- 18S GTAACCCGTTGAACCCCATT CCATCCAATCGGTAGTAGTG
mary antibodies were against ERα (MA1–39540, Thermo ER-α CAGGAACCAGGGAAAATGTG AACCGAGATGATGTAGCCAGC
Fisher, IL, USA), ERβ (PA1-310B, Thermo Fisher, IL, ER-β ACTTGCTGAACGCCGTGACC CAGATGTTCCATGCCCTTGTT
USA), PR (MA1–411, Thermo Fisher, IL, USA), AR PR TGAATCCGGCCTCAGGTAGTT CGCGCTCTACCCTGCACTC
(06–680, Millipore, CA, USA), and GAPDH (MAB374,
AR TCACCGCACCTGATGTGTG ACATGGTCCCTGGCAGTCTC
Millipore, CA, USA).
Lan et al. Reproductive Biology and Endocrinology (2020) 18:12 Page 4 of 7

Table 2 Patient characteristics

Normal pregnant women (N = 45) Preeclampsia women (N = 30) p
Age (yr) 33.6 ± 4.7 (24–45) 35.3 ± 5.1 (22–44) NS
Hemoglobin (g/dL) 11.6 ± 1.53 12.17 ± 1.48 NS
Gestational week (wk.) 38.6 ± 2.8 35.8 ± 2.2 < 0.001
Gravidity 2.1 ± 1.1 (1–5) 2.5 ± 2.2 (1–9) NS
Parity 1.7 ± 0.7 (1–3) 1.6 ± 0.9 (1–4) NS
Nulliparity 18 (40%) 19(63.3%) 0.048
Multiparity 27(60%) 11 (36.7%) 0.048
Body Mass Index (BMI) 27.8 ± 4.9 32.3 ± 6.3 0.002
Birth weight (g) 3046.0 ± 717.33 2273.6 ± 886.3 < 0.001
Systolic blood pressure (mmHg) 118.9 ± 23.7 163.0 ± 18.7 < 0.001
Diastolic blood pressure (mmHg) 74.2 ± 12.0 101.8 ± 12.4 < 0.001
NS not significant
Data are expressed as mean ± standard deviation

decreased protein level of ERα. The two groups had Discussion

similar protein levels of AR and PR (Fig. 4b). Female sex steroid hormones, including E2 and P4, have
We further characterized the expression patterns of significant physiological roles during menstruation and
these hormone receptors in placental tissues by H&E pregnancy. The ovaries, adrenal glands, and placenta
staining and immunohistochemical analysis of formalin- produce these hormones, and the placenta is the major
fixed tissues. The H&E staining results indicated that the endocrine organ during pregnancy. During pregnancy,
cytotrophoblast was surrounded by a layer of syncytio- the complex process of steroidogenesis occurs in mul-
trophoblasts, a general morphological feature of the pla- tiple organs, including the maternal uterus, placenta,
centa. Immunohistochemical staining indicated that ERβ fetal membrane, and the maternal and fetal
was present in cyto- and syncytiotrophoblast cells of hypothalamic-pituitary-adrenal axis, although most ste-
women in the uncomplicated pregnancy group but was roidogenesis occurs in the placenta.
mostly present in the nuclei of syncytiotrophoblast cells During pregnancy, the placenta primarily produces es-
of women in the preeclampsia group (Fig. 4a). The two trogen by conversion of androgen precursors that origin-
groups had no evident differences in the placental distri- ate in the maternal and fetal adrenal glands. Many
butions of other hormone receptors. studies have reported that serum estrogen levels increase
Measurements of the serum levels of hormones progressively from 22 weeks of gestation and are import-
throughout pregnancy indicated that the preeclampsia ant regulators of placental and embryo growth [8]. This
group had increased T levels but decreased E2 levels is consistent with our findings (Fig. 5).
(Fig. 5). However, the two groups had similar P4 levels Furthermore, we also found that women in the pre-
throughout pregnancy. eclampsia group had lower serum estrogen levels in the

Fig. 2 Levels of estradiol, testosterone, DHEAS, and progesterone in women with preeclampsia compared to women with uncomplicated
pregnancies. a serum in third trimester b cord blood. * P < 0.05
Lan et al. Reproductive Biology and Endocrinology (2020) 18:12 Page 5 of 7

Fig. 3 Quantitative evaluation of ERα, ERβ, AR and PR mRNA in the placenta by real-time reverse transcriptase polymerase chain reaction. The
two groups had similar mRNA levels of ERα, ERβ, AR, and PR

Fig. 4 a The localization and expression of ERα, ERβ, AR and PR proteins in uncomplicated pregnancies (upper panel) and preeclampsia placenta
(bottom panel) were analyzed by immunohistochemical analysis. Specific and robust ERβ immunostaining was detected exclusively in the
nucleus and cytoplasm of syncytiotrophoblasts (arrowhead) and stromal cells (arrow). Magnification × 400. b Comparison of ERα, ERβ, AR and PR
protein expression in placenta from preeclampsia between the two groups by western blot. ERα and ERβ protein expression was significantly
lower and higher, respectively, in preeclampsia
Lan et al. Reproductive Biology and Endocrinology (2020) 18:12 Page 6 of 7

Fig. 5 Trajectories of hormone concentrations in women with preeclampsia compared to women with uncomplicated pregnancies. a Estradiol b
Progesterone c Testosterone d DHEA- S

third trimester and lower ERα and higher ERβ protein the concentration of estrogen during pregnancy, but
expression in their placentas. A recent study reported only Yin et al. examined ERα expression in the placenta
that altered serum concentrations of steroid hormones, of patients with preeclampsia [10]. Their findings are
including E2, may be associated with preeclampsia [9]. consistent with our findings, although they did not
Estrogen has specific metabolic effects in that it stimu- measure the level of ERβ. P4 reduces vascular resistance
lates the expression of vascular endothelial growth factor by decreasing the sensitivity to angiotensin and increas-
(VEGF) and angiogenesis. However, aberrant production ing the production of endothelial vasodilators, which dir-
of estrogens could promote preeclampsia because they ectly affect muscles [10]. A previous study reported that
are exclusively produced by the placenta, and they pro- the serum P4 level was markedly increased in women
mote angiogenesis and vasodilation. More recent studies with preeclampsia compared with healthy pregnant
suggested a marked decrease in E2 levels in women with women [6]. These results are inconsistent with our data.
preeclampsia [7]. This is consistent with our findings The relationship between androgens and maternal
(Figs. 2a and 5). cardiovascular and placental function deserves par-
Estrogen acts by binding to ERα or ERβ, which are ticular consideration because the serum T level in
both located in the nucleus but have distinct tissue- women with preeclampsia is elevated and correlates
specific expression patterns, or by binding to other re- with vascular dysfunction [5]. Our finding of elevated
ceptors located in the plasma membrane and cytoplas- serum T in women with preeclampsia supports previ-
mic organelles. ERα and ERβ are similar in terms of ous studies, thus suggesting that increased androgens
ligand-binding and nuclear DNA-binding regions, and and androgen signaling may contribute to preeclamp-
there are numerous mRNA splice variants for both re- sia [5, 11, 12].
ceptors in diseased and normal tissues. In a ligand- The adrenal cortex produces large amounts of DHEAS
dependent fashion, each ER pool contributes to the over- and androstenedione during the fetal period, but this
all effects of estrogens on biological outcomes. Kim et al. production decreases rapidly after birth. However, the
suggested that placental ERα and ERβ had higher ex- physiological role of these changes is still unclear. DHEA
pression at term period compared with early preterm is a precursor to testosterone. Interestingly, our results
and that they were located in cyto- and syncytiotropho- demonstrated that women in the uncomplicated preg-
blast cells [8]. The placentas of our preeclampsia group nancy and preeclampsia groups had similar serum levels
had a decreased protein level of ERα and an increased of DHEAS, but the preeclampsia group had a lower cord
protein level of ERβ but no alterations in the levels of blood level of DHEAS. The physiological role of DHEAS
the corresponding mRNAs. Many studies have measured requires further study.
Lan et al. Reproductive Biology and Endocrinology (2020) 18:12 Page 7 of 7

Our research provided novel insights into the relation- Competing interests
ship between sex hormone status and preeclampsia. The authors declare that they have no competing interests.

However, our study was limited by the small number of Author details
patients. Thus, confirmation requires examination of a 1
Department of Obstetrics and Gynecology, Kaohsiung Chang Gung
larger cohort, with longitudinal data, using highly reli- Memorial Hospital and Chang Gung University College of Medicine, 123
Ta-Pei Road, Niao-Sung District, Kaohsiung City, Kaohsiung, Taiwan. 2Center
able technology such as gas chromatography/mass spec- for Menopause and Reproductive Medicine Research, Kaohsiung Chang
trometry [7]. In addition, a high BMI is strongly Gung Memorial Hospital and Chang Gung University College of Medicine,
associated with preeclampsia [13], but there is evidence Kaohsiung, Taiwan. 3Graduate Institute of Clinical Medicine, College of
Medicine, Kaohsiung Medical University, Kaohsiung, Taiwan.
to suggest that obesity may vary by hormone receptor
status and by moderating the synthesis and metabolism Received: 28 November 2019 Accepted: 27 January 2020
of circulating sex steroid hormones and related binding
proteins [14]. Furthermore, the serum levels of sex hor- References
mones change as pregnancy increases, as shown in Fig. 5 1. Shah S, Gupta A. Hypertensive disorders of pregnancy. Cardiol Clin. 2019;37:
and other reports [8]. According to Table 2, the two 345–54.
2. Phipps E, Prasanna D, Brima W, Jim B. Preeclampsia: updates in pathogenesis,
groups included in this study showed significant differ- definitions, and guidelines. Clin J Am Soc Nephrol. 2016;11:1102–13.
ences in gestational age and BMI that might interfere 3. Armaly Z, Jadaon JE, Jabbour A, Abassi ZA. Preeclampsia: novel mechanisms
with the levels of hormones and/or the expression of and potential therapeutic approaches. Front Physiol. 2018;9:973.
4. Maliqueo M, Echiburu B, Crisosto N. Sex steroids modulate uterine-placental
their receptors in the placenta. This important fact is an- vasculature: implications for obstetrics and neonatal outcomes. Front
other limitation of the study. Physiol. 2016;7:152.
5. Kumar S, Gordon GH, Abbott DH, Mishra JS. Androgens in maternal vascular
and placental function: implications for preeclampsia pathogenesis.
Conclusions Reproduction. 2018;156:R155–67.
Our findings suggest that preeclampsia is associated with 6. Hertig A, Liere P, Chabbert-Buffet N, Fort J, Pianos A, Eychenne B, Cambourg
A, Schumacher M, Berkane N, Lefevre G, et al. Steroid profiling in
a distinct hormonal milieu. In particular, we found that preeclamptic women: evidence for aromatase deficiency. Am J Obstet
E2 and ERα are downregulated and that T and ERβ are Gynecol. 2010;203:477 e471–479.
upregulated in preeclamptic pregnancies. This suggests 7. Berkane N, Liere P, Oudinet JP, Hertig A, Lefevre G, Pluchino N, Schumacher
M, Chabbert-Buffet N. From pregnancy to preeclampsia: a Key role for
the involvement of these molecules in the pathogenesis estrogens. Endocr Rev. 2017;38:123–44.
of preeclampsia. 8. Kim SC, Park MN, Lee YJ, Joo JK, An BS. Interaction of steroid receptor
coactivators and estrogen receptors in the human placenta. J Mol
Abbreviations Endocrinol. 2016;56:239–47.
DHEA: Dehydroepiandrosterone; DHEAS: Dehydroepiandrosterone sulfate; 9. Zheng JJ, Wang HO, Huang M, Zheng FY. Assessment of ADMA, estradiol, and
E2: Estradiol; ERα: Estrogen receptor-α; ERβ: Estrogen receptor-β; progesterone in severe preeclampsia. Clin Exp Hypertens. 2016;38:347–51.
P4: Progesterone; qRT-PCR: Quantitative real-time reverse transcription-PCR; 10. Albrecht ED, Pepe GJ. Estrogen regulation of placental angiogenesis and fetal
T: Testosterone; VEGF: Vascular endothelial growth factor; VEGF: Vascular ovarian development during primate pregnancy. Int J Dev Biol. 2010;54:397–408.
endothelial growth factor 11. Hsu TY, Lan KC, Tsai CC, Ou CY, Cheng BH, Tsai MY, Kang HY, Tung YH, Wong YH,
Huang KE. Expression of androgen receptor in human placentas from normal
Acknowledgements and preeclamptic pregnancies. Taiwan J Obstet Gynecol. 2009;48:262–7.
Not applicable. 12. Shin YY, Jeong JS, Park MN, Lee JE, An SM, Cho WS, Kim SC, An BS, Lee KS.
Regulation of steroid hormones in the placenta and serum of women with
preeclampsia. Mol Med Rep. 2018;17:2681–8.
Authors’ contributions
13. Dayan N, Pilote L, Opatrny L, Basso O, Messerlian C, El-Messidi A,
Conception and design of study: CCT, TYH, YJL; enrolled the subjects: CCT,
Daskalopoulou SS. Combined impact of high body mass index and in vitro
TYH, YJL; analysis of data and writing of the manuscript: KCL, NCT; critical
fertilization on preeclampsia risk: a hospital-based cohort study. Obesity
revision of the article for intellectual content: KCL NCT, YJL, HHC, YTS. All
(Silver Spring). 2015;23:200–6.
authors read and approved the final manuscript.
14. Key TJ, Allen NE, Verkasalo PK, Banks E. Energy balance and cancer: the role
of sex hormones. Proc Nutr Soc. 2001;60:81–9.
Authors’ information
Not applicable.
Publisher’s Note
Funding Springer Nature remains neutral with regard to jurisdictional claims in
This study was supported by CMRPG8G0071–73, CMRPG8G0101, published maps and institutional affiliations.
CMRPG8F0831 from Chang Gung Memorial Hospital.

Availability of data and materials

The datasets used and/or analyzed during the current study are available
from the corresponding author on reasonable request.

Ethics approval
This study was approved by the Institutional Review Board of Chang Gung
Memorial Hospital (CGMH 201601484A3).

Consent for publication

Not applicable.