Available via license: CC BY-NC 4.0
Content may be subject to copyright.
1
ChenY, etal. BMJ Open 2019;9:e023003. doi:10.1136/bmjopen-2018-023003
Open access
Association between homocysteine,
vitamin B12, folic acid and erectile
dysfunction: a cross-sectional study
in China
Yang Chen,1,2,3 Jie Li,2,4 Tianyu Li,1,2,3 Jianxiong Long,2,5 Jinling Liao,2
Gong-Hong Wei,6 Zengnan Mo,1,2,3 Jiwen Cheng1,2,3
To cite: ChenY, LiJ, LiT,
etal. Association between
homocysteine, vitamin
B12, folic acid and erectile
dysfunction: a cross-sectional
study in China. BMJ Open
2019;9:e023003. doi:10.1136/
bmjopen-2018-023003
►Prepublication history and
additional material for this paper
are available online. To view
please visit the journal online
(http:// dx. doi. org/ 10. 1136/
bmjopen- 2018- 023003).
YC and JL contributed equally.
Received 29 March 2018
Revised 29 November 2018
Accepted 30 November 2018
For numbered afliations see
end of article.
Correspondence to
DrJiwen Cheng;
chengjiwen1977@ foxmail. com
Research
© Author(s) (or their
employer(s)) 2019. Re-use
permitted under CC BY-NC. No
commercial re-use. See rights
and permissions. Published by
BMJ.
ABSTRACT
Objectives Erectile dysfunction (ED) affects up to 53.4%
of men aged 30–80 years. In this study, we aimed to
examine the association between homocysteine (HCY),
vitamin B12 (B12), folic acid (FA) and ED.
Design Cross-sectional study.
Setting Guangxi, China.
Participants A total of 1381 participants who completed
questionnaires were included, between September 2009
and December 2009.
Measures ED was evaluated by the International Index of
Erectile Function scores. Also, the values of HCY, B12 and
FA were acquired. Then, regression and between-group
analyses were performed.
Results No association between FA and ED was found.
Signicant correlations between HCY and ED were
found—the relationships between these two parameters
were most notable in men aged over 60 years and in
men living alone (bachelors or bachelorhood). B12 levels
were higher in men with ED (718.53±234.37 pg/mL vs
688.74±229.68, p=0.015). Using multinomial logistic
regression analyses, B12 levels were related to mild ED
(multivariate adjusted analysis: OR 1.620, 95% CI 1.141 to
2.300, p=0.007), especially among men aged 40–49 years
(OR 2.907, 95% CI 1.402 to 6.026, p=0.004).
Conclusions We report, for the rst time, a relationship
between B12 levels and ED. We also found specic
cohorts of men for whom the relationship between
HCY levels and ED is most prominent. Further studies
are required to elucidate the mechanisms underlying
these relationships—these may ultimately result in new
therapies for ED.
INTRODUCTION
Erectile dysfunction (ED) is defined as the
inability to acquire and maintain satisfying
sexual intercourse with a sufficient erec-
tion, affecting up to 53.4% of men aged 30
to 80 years.1 The morbidity increases sharply
among men over 40 years of age.2 3 It has been
estimated that the prevalence of ED will reach
322 million worldwide by the year 2025.4
Various factors including smoking, hyper-
tension and hyperlipidaemia have been
identified to influence the development
of ED. Among these factors, the vascular
component is dominant.5 6 Moreover, ED
may be one of the indicators of cardiovas-
cular disease (CVD).7 Homocysteine (HCY),
a CVD-associated factor, was recently defined
as an independent risk factor for ED.8 9 HCY
is a thiol-containing amino acid, mainly from
methionine, with two steps of transformation.
First, methionine is catalysed to form S-ade-
nosylmethionine (SAM) by the enzyme SAM
synthase. As a major methyl group donor for
various methylation reactions, SAM is mainly
transformed into S-adenosylhomocysteine
(SAH) after loss of the methyl group. In the
second step, SAH is hydrolysed by SAH hydro-
lase to form HCY and adenosine. Biologically,
HCY is involved in two pathways, including
remethylation (RM) and transsulfuration
(TS). In the RM pathway, HCY regenerates
methionine by methylenetetrahydrofolate
reductase with cofactors of folic acid (FA)
and vitamin B12 (B12). In the TS pathway,
Strengths and limitations of this study
►Our study is a cross-sectional study, mainly based
on the large Fangchenggang Area Male Health and
Examination Survey project, including a total of 4303
men.
►This study includes comprehensive analyses of
baseline, linear and logistic regression, and multino-
mial logistic regression.
►According to the changes in the homocysteine(HCY),
vitaminB12 (B12) and folic acid (FA) levels, and the
order of erectile dysfunction (ED) severity, we inves-
tigated the associations between HCY, B12, FA and
ED.
►The study also took into consideration the effects of
age, marital statusand educational status.
►Nevertheless, as a cross-sectional analysis, the ex-
act mechanisms of the relationship between HCY,
B12, FA and ED cannot be clearly dened.
on 23 May 2019 by guest. Protected by copyright.http://bmjopen.bmj.com/BMJ Open: first published as 10.1136/bmjopen-2018-023003 on 22 May 2019. Downloaded from
2ChenY, etal. BMJ Open 2019;9:e023003. doi:10.1136/bmjopen-2018-023003
Open access
HCY is catalysed by the cystathione-β-synthase and γ-cysta-
thionase.10 11
FA and B12, as the cofactors of HCY, have also been
identified to be associated with ED.12 However, limited
studies have been focused on the relevance of their levels
to ED. On the basis of previous studies, we hypothesised
that there are likely associations between HCY, B12, FA
and ED. In order to comprehensively investigate the
exact association between HCY, B12, FA and ED, our
study is conducted based on the Fangchenggang Area
Male Health and Examination Survey (FAMHES) project.
Our study may pave the way to the treatment of ED on the
basis of the balance among HCY, B12 and FA.
METHODS AND MATERIALS
Population and data collection
FAMHES is a population-based project, which was mainly
performed to investigate environmental and genetic
factors, as well as their inter-relations. From September
2009 to December 2009, 4303 men coming for routine
physical examination at the Medical Center in Fangc-
henggang First People’s Hospital were enrolled. Then,
3593 participants responded for further interviews
(response rate=83.5%).13 No distinct differences were
detected between the men who participated in the inter-
views and those who did not.
All participants were asked to provide blood samples
between 8:00 and 11:00, after fasting for at least 8 hours
(overnight). Then, these blood samples were transported
to the Department of Clinical Laboratory at the First Affil-
iated Hospital of Guangxi Medical University in Nanning
within 2–3 hours, where they were centrifuged within
15–25 min and stored at −80°C. Serum B12 and FA were
detected with electrochemiluminescence immunoassays,
while serum HCY was measured with enzymatic cycling
methods.
Then, all the participants were invited to complete
a comprehensive questionnaire. This process was
performed by the trained investigators using a stan-
dardised protocol with a face-to-face interview. Essen-
tial information (eg, age, sex, smoking and drinking)
was collected, and complete physical examinations (eg,
height, weight, waistline and hipline) were performed.
Smoking status and alcohol consumption were defined as
Yes or No. Marital status was classified into living together
(married or cohabitation without marriage) and alone
(bachelors or bachelorhood). Meanwhile, according to
the years of education, three groups could be defined
(0–6 years, primary education; 7–12 years, intermediate
education; and ≥13 years, superior education). In the
physical examination, body weight with thin clothing
and height without shoes were measured. Then, body
mass index (BMI) was calculated with the formula of
weight/(height)2. Waist circumference was measured at
the midpoint between the inferior costal margin and the
superior iliac crest in the midaxillary line. The hipline
was defined as the maximum circumference over the
buttocks. The waist:hip ratio (WHR) was calculated as
waist circumference/hipline. These processes above
including initial examination (including height, weight,
waistline and hipline), further interviews (essential infor-
mation, such as age, sex, smoking and drinking, etc)
and blood collection were performed on the same days
coherently. The flow of participants’ collection is shown
in online supplementary figure S1.
Patient and public involvement
Patients and the public were not involved in the develop-
ment of the research question and design or recruitment
of this study.
ED denition and grouping
In this study, the International Index of Erectile Function
(IIEF-5) was applied to define ED.14 The IIEF-5 system
has five questions, which mainly cover the conditions
of erection confidence, erection firmness, maintenance
ability, maintenance frequency and satisfaction, with
the scores ranging from 5 to 25. Each question has six
selections. According to the orders of answers, the scores
are defined as 0–5. Then, participants can be divided
into ED (IIEF-5 ≤21) and non-ED (IIEF-5 >21) groups.
According to the symptoms, ED can also be classified into
five groups: none (IIEF-5 score 22–25), mild (17–21),
moderate (12–16) and severe symptoms (5–11).13 15
In addition, HCY level can also be divided into normal
(HCY 5–15 µmol/L) and hyperhomocysteinaemia
(HCY >15 µmol/L).16
Participant screening
In order to acquire the eligible participants for this study,
we developed rigorous exclusion criteria: (1) incomplete
data for the individual information and IIEF-5 score;
(2) incomplete data for HCY, B12 and FA or refused
to provide the blood samples; (3) with diseases such as
CVDs, inflammatory/immune diseases and kinds of
cancers, which might influence the levels of HCY, B12 and
FA (online supplementary table S1); and (4) currently
taking drugs that might affect HCY, B12 and FA levels,
such as vitamins, antidiabetic medicines, non-steroidal
anti-inflammatory drugs, antibiotics, cimetidine or gluco-
corticoids (online supplementary table S1). Then, 1381
participants were included for further analyses. The flow
for screening the eligible participants is shown in online
supplementary figure S2.
Statistical analysis
Before analysis, HCY, B12 and FA levels were tested for
Gaussian distribution with the Shapiro-Wilks test. If data
were not Gaussian in distribution, they were logarithmically
transformed in order to ensure the approximate Gaussian
distribution. Based on the 22 IIEF-5 scores, two groups
were defined (ED and non-ED), and Student’s t-test and
χ2 test were applied in the baseline analysis. Then, linear
and logistic regression analyses were used, with the IIEF-5
scores and binary variable (ED or non-ED) as the depen-
dent factors, respectively. Three adjusted models were
on 23 May 2019 by guest. Protected by copyright.http://bmjopen.bmj.com/BMJ Open: first published as 10.1136/bmjopen-2018-023003 on 22 May 2019. Downloaded from
3
ChenY, etal. BMJ Open 2019;9:e023003. doi:10.1136/bmjopen-2018-023003
Open access
used: unadjusted, age-adjusted and multivariate adjusted.
In the multivariate-adjusted model, the covariates were as
follows: age, smoking status, alcohol consumption, BMI
and WHR. Among them, BMI and WHR are the indexes
applied to estimate obesity. However, BMI tends to evaluate
body fatness but has a weak ability to differentiate fatness as
central or visceral.17 Alternatively, WHR is said to be more
effective in reflecting the visceral fat and central adiposity
but is not suitable for an estimation of body fat.17 18 Addi-
tionally, the predictive effects of BMI and WHR in diseases
are different.19 20 So, in our study, these two obesity indexes
were treated as the covariates.
Then, multinomial logistic regression analysis was also
performed to discover the potential association between
HCY, B12, FA and ED, along with the order of severity of
ED or the changes in the HCY, B12 and FA levels quartile
(Q1<25%, 25%≤Q2≤50%, 50%<Q3≤75%, Q4>75%). Addi-
tionally, considering the non-negligible influences of age on
the risk of ED, we also grouped the participants on the basis
of age (<40, 40–49, 50–59 and ≥60 years). The Bernoulli
correction was applied, with the significant threshold of
p value <0.0125 (=0.05/4 tests) for multinomial logistic
regression analysis. Additionally, according to the groups
of marital status and educational status, logistic regression
analyses were also conducted. In these analyses, the missing
data were deleted. All statistical tests were two-tailed, which
were performed with SPSS V.16.0 software. The threshold
for significance was p value <0.05.
RESULTS
In the baseline analysis, based on IIEF-5, the ED and non-ED
groups were defined. In line with previous studies, the age
of the ED group (37.99±10.75 years) was older than the
non-ED group (34.18±8.47 years, p<0.001). Meanwhile, B12
levels were significantly higher in the ED group (p=0.015).
However, no significant difference was shown for HCY
levels, and the proportion of hyperhomocysteinaemia was
higher in the ED group (43.02%) than that in the non-ED
group (37.52%, p=0.037). In addition, the proportion of
alcohol consumption (p=0.032) and educational status
Table 1 Characteristics of the eligible participants in the analysis
ED Non-ED Pvalue
N 688 693
Age, years 37.99±10.75 34.18±8.47 <0.001*
BMI, kg/m223.27±3.26 23.37±3.48 0.591*
WHR 0.88±0.06 0.88±0.06 0.253*
HCY, µmol/L 14.97±4.11 15.34±11.09 0.524*
Normal HCY 392 (56.98%) 433 (62.48%)
Hyperhomocysteinaemia 296 (43.02%) 260 (37.52%) 0.037*
B12, pg/mL 718.53±234.37 688.74±229.68 0.015*
FA, ng/mL 9.56±2.72 9.89±11.28 0.594*
Smoke
Yes 392 (56.98%) 385 (55.56%)
No 296 (43.02%) 308 (44.44%) 0.594†
Drink
Yes 586 (85.17%) 617 (89.03%)
No 102 (14.83%) 76 (10.97%) 0.032†
Marital status‡
Live together 595 (86.48%) 578 (83.41%)
Alone 93 (13.52%) 115 (16.59%) 0.110†
Educational status§
Primary 19 (2.76%) 5 (0.72%)
Intermediate 488 (70.93%) 411 (59.39%)
Superior 181 (26.31%) 276 (39.89%) <0.001†
Normal HCY: 5–15µmol/L; hyperhomocysteinaemia:>15 µmol/L.
*Student's t-test.
†χ2 test.
‡Marital status was classied into living together (married or cohabitation without marriage) and alone (bachelors or bachelorhood).
§One participant without the information of educational status in the non-ED group.
BMI, body mass index; B12, vitamin B12; ED, erectile dysfunction; FA, folic acid;HCY, homocysteine; WHR, waist:hip ratio.
on 23 May 2019 by guest. Protected by copyright.http://bmjopen.bmj.com/BMJ Open: first published as 10.1136/bmjopen-2018-023003 on 22 May 2019. Downloaded from
4ChenY, etal. BMJ Open 2019;9:e023003. doi:10.1136/bmjopen-2018-023003
Open access
(p<0.001) were also identified to have a statistically signif-
icant difference in the two groups (table 1).
Signal for the association between HCY and ED
While we discovered no significant association between
HCY levels and ED in the comprehensive analyses
(tables 2–5), a slight association of HCY with ED was
observed in the participants grouped by age, especially in
the old men (age ≥60) (online supplementary table S2).
Similar relevance was confirmed in marital status (alone,
unadjusted severe ED: OR 4.385, 95% CI 1.070 to 17.974,
p=0.040; age-adjusted severe ED: OR 5.085, 95% CI 1.195
to 21.636, p=0.028) (online supplementary table S3).
In the latter analysis, HCY was divided into normal
(HCY 5–15 µmol/L) and hyperhomocysteinaemia
(HCY >15 µmol/L). The significant association between
HCY and ED seemed to be more prominent in men living
alone (age-adjusted severe ED: OR 2.448, 95% CI 1.046 to
5.733, p=0.039) (online supplementary table S3).
B12 level is signicantly associated with ED
To investigate the association between ED and B12, we
applied linear and logistic regression analyses, resulting
in no significant association for B12 in the linear regres-
sion analysis (in which IIEF-5 scores were treated as the
dependent factor). For the binary logistic regression (the
status of ED evaluated by IIEF-5 was treated as the depen-
dent factor), B12 was identified to be associated with ED
in the unadjusted model (OR 1.438, 95% CI 1.070 to
1.933, p=0.016). However, the association signal dimin-
ished in other adjusted models (table 3). We next investi-
gated the relationship between B12 and ED based on the
severity grades of ED. Interestingly, the positive correla-
tion between B12 and ED was further confirmed, espe-
cially among men with mild ED (unadjusted: OR 1.694,
95% CI 1.207 to 2.376, p=0.002; age-adjusted: OR 1.596,
95% CI 1.135 to 2.244, p=0.007; multivariate adjusted:
OR 1.620, 95% CI 1.141 to 2.300, p=0.007) (table 4).
Subsequently, the levels of B12 were divided into quar-
tiles. The result showed that B12 might be significantly
associated with ED, especially at the higher levels (unad-
justed: Q2: OR 0.917, p=0.569; Q3: OR 0.988, p=0.939;
Q4: OR 1.452, p=0.015; and p for trend <0.001) (table 5).
After adjusting age for the above analyses, the signif-
icant association between B12 and ED diminished
(tables 2 and 3 and table 5), suggesting that age cannot
be excluded while investigating the relationship between
B12 and ED. We thus grouped the participants into four
age groups (ages <40, 40–49, 50–59 and ≥60 years). Our
results showed that the significant correlations between
B12 and mild ED (IIEF-5=17–21) mainly presented in the
age group 40–49 years (OR 2.907, 95% CI 1.402 to 6.026,
p=0.004) (online supplementary table S2).
Our baseline analysis discovered different proportions
of educational status in the ED and non-ED groups. In
Table 2 Linear regression analyses for the ED and HCY, B12 and FA
Unadjusted Ageadjusted Multivariate adjusted
β95% CI Pvalue β 95% CI Pvalue β 95% CI Pvalue
IIEF-5
HCY −0.202 −1.080to0.676 0.651 0.139 −0.732to 1.009 0.755 0.084 −0.787to 0.956 0.850
Binary HCY −0.338 −0.817to 0.142 0.167 −0.146 −0.622to 0.330 0.548 −0.186 −0.663to 0.291 0.444
B12 0.048 −0.600to 0.696 0.885 0.212 −0.428to 0.852 0.515 0.404 −0.259to 1.068 0.232
FA 0.112 −0.668to 0.891 0.779 0.388 −0.384to 1.160 0.986 0.324 −0.496to 1.145 0.438
IIEF-5 scores were the dependent factor for the linear regression analysis.Multivariate adjusted: age, BMI, WHR, smoke and
drink.BinaryHCY: normal HCY (5–15µmol/L); hyperhomocysteinaemia (>15 µmol/L).
BMI, body mass index; B12, vitamin B12; ED, erectile dysfunction; FA, folic acid; HCY, homocysteine; IIEF-5, International Index of Erectile
Function; WHR, waist:hip ratio.
Table 3 Binary regression analyses for ED and HCY, B12 and FA
Binary
Unadjusted Ageadjusted Multivariate adjusted
OR 95% CI Pvalue OR 95% CI Pvalue OR 95% CI Pvalue
HCY 1.137 0.766 to 1.687 0.524 0.959 0.641 to 1.435 0.839 0.986 0.658 to 1.479 0.986
Binary HCY 1.258 1.014 to 1.560 0.037 1.151 0.923 to 1.435 0.212 1.174 0.940 to 1.466 0.157
B12 1.438 1.070 to 1.933 0.016 1.338 0.992 to 1.805 0.057 1.311 0.961 to 1.788 0.087
FA 1.100 0.775 to 1.561 0.594 0.956 0.668 to 1.367 0.804 1.041 0.710 to 1.527 0.835
In the binary regression analysis, the ED status (ED: IIEF-5≤21; non-ED: IIEF-5>21) was treated as the dependent factor.Multivariate
adjusted: age, BMI, WHR, smoke and drink.BinaryHCY: normal HCY (5–15µmol/L); hyperhomocysteinaemia (>15 µmol/L).
BMI, body mass index; B12, vitamin B12; ED, erectile dysfunction; FA, folic acid; HCY, homocysteine; IIEF-5, International Index of Erectile
Function; WHR, waist:hip ratio.
on 23 May 2019 by guest. Protected by copyright.http://bmjopen.bmj.com/BMJ Open: first published as 10.1136/bmjopen-2018-023003 on 22 May 2019. Downloaded from
5
ChenY, etal. BMJ Open 2019;9:e023003. doi:10.1136/bmjopen-2018-023003
Open access
Table 4 Multinomial logistic regression for the association between ED and HCY, B12 and FA
HCY BinaryHCY B12 FA
OR 95% CI Pvalue OR 95% CI Pvalue OR 95% CI Pvalue OR 95% CI Pvalue
ED—unadjusted
None (IIEF-5=22–25) 1 1 1 1 1 1 1 1 1 1 1 1
Mild (17–21) 1.081 0.695 to 1.681 0.730 1.246 0.980 to 1.583 0.072 1.694 1.207 to 2.376 0.002 1.180 0.801 to 1.740 0.402
Moderate (12–16) 1.258 0.649 to 2.439 0.497 1.298 0.896 to 1.880 0.168 1.187 0.715 to 1.972 0.508 0.960 0.519 to 1.776 0.896
Severe (5–11) 1.259 0.562 to 2.820 0.575 1.258 0.799 to 1.980 0.322 0.891 0.496 to 1.599 0.698 0.923 0.434 to 1.963 0.834
ED—ageadjusted
None (IIEF-5=22–25) 1 1 1 1 1 1 1 1 1 1 1 1
Mild (17–21) 0.948 0.607 to 1.480 0.814 1.166 0.914 to 1.486 0.217 1.596 1.135 to 2.244 0.007 1.052 0.710 to 1.558 0.800
Moderate (12–16) 0.929 0.455 to 1.898 0.840 1.095 0.747 to 1.605 0.643 1.409 0.635 to 1.733 0.851 0.762 0.404 to 1.437 0.401
Severe (5–11) 1.068 0.467 to 2.443 0.876 1.150 0.726 to 1.820 0.552 0.839 0.471 to 1.495 0.551 0.794 0.370 to 1.705 0.554
ED—multivariate adjusted
None (IIEF-5=22–25) 1 1 1 1 1 1 1 1 1 1 1 1
Mild (17–21) 0.973 0.621 to 1.524 0.904 1.188 0.929 to 1.517 0.169 1.620 1.141 to 2.300 0.007 1.184 0.775 to 1.808 0.435
Moderate (12–16) 0.965 0.472 to 1.971 0.922 1.119 0.762 to 1.643 0.567 0.972 0.576 to 1.640 0.915 0.777 0.401 to 1.507 0.456
Severe (5–11) 1.132 0.491 to 2.613 0.771 1.187 0.748 to 1.885 0.467 0.717 0.388 to 1.324 0.288 0.821 0.368 to 1.831 0.631
BinaryHCY: normal HCY (5–15µmol/L); hyperhomocysteinaemia (>15 µmol/L).The categorical dependent variables were the various ED groups, based on the IIEF-5. The
symptoms of ED were divided into None (IIEF-5=22–25), Mild (17–21), Moderate (12–16) and Severe (5–11). The None group (22–25) was treated as the reference.Multivariate
adjusted: age, BMI, WHR, smoke and drink.
BMI, body mass index; B12, vitamin B12; ED, erectile dysfunction; FA, folic acid;HCY, homocysteine;IIEF-5, International Index of Erectile Function; WHR, waist:hip ratio.
on 23 May 2019 by guest. Protected by copyright.http://bmjopen.bmj.com/BMJ Open: first published as 10.1136/bmjopen-2018-023003 on 22 May 2019. Downloaded from
6ChenY, etal. BMJ Open 2019;9:e023003. doi:10.1136/bmjopen-2018-023003
Open access
order to discuss the influences of marital and educa-
tional status in the relevance to ED and B12, we further
performed between-group analyses. Similar to previous
results, B12 was also identified to be associated with mild
ED, even after multivariate adjustment (marital status,
living together: OR 1.501, 95% CI 1.035 to 2.175, p=0.032;
alone: OR 3.449, 95% CI 1.113 to 10.692, p=0.032; and
educational status, intermediate: OR 1.858, 95% CI 1.214
to 2.845, p=0.004) (online supplementary table S3).
DISCUSSION
ED is a common disorder affecting a large number of
men.1–4 Recent studies suggest HCY may be an indepen-
dent risk factor for ED.8 9 In order to test this associa-
tion, we conducted the current study based on the larger
population-based FAMHES project. We confirmed
that HCY is significantly associated with ED, especially
severe ED. Moreover, B12 may also be relevant to mild
ED. In contrast, we observed no significant association
between FA and ED in our study.
HCY was reported to be associated with many
diseases and health conditions, such as psychological
disorders,21 22 lipid profiles,23 renal impairment24 and
inflammatory/immune factors.25 Moreover, HCY is also
identified to be a useful marker for CVD.26 27 Mean-
while, ED could be a potentially predictive factor for
cardiovascular and other chronic diseases.28 Based on
the relevance, it was assumed that HCY might be a risk
factor for ED.8 9 Consistent with this, we revealed that
HCY was significantly associated with ED, especially
severe ED. The main mechanism might be that HCY
could influence endothelial dysfunction and nitric
oxide (NO) diffusion. Previously, in vitro and in vivo
studies discovered that HCY could be a toxin for the
vasculature by inducing endothelial dysfunction.29
Additionally, NO is mainly involved in vascular dilation,
smooth muscle relaxation (including genitourinary
smooth muscle) and permitting penile erection.30 31
Studies showed that increased HCY could inhibit NO
synthase, probably influencing the production of NO,
and the development of ED.32 So, on the basis of these
relevance, we could understand the risk effect of HCY
on ED. Additionally, the status of living alone for men
would also influence this association, hinting the patho-
genesis of psychological factors for ED.
B12 is also known as cobalamin. Similar to FA, it is
an important cofactor in methionine synthesis and HCY
metabolism.33 Although previous studies identified that
FA might be a potential protective factor for ED,34 no
significant association has been detected. In contrast to
B12, HCY has been found to protect against ED.35 Our
study also identifies the potential association between
B12 and ED, though ED tends to have high levels of B12
(ED: 718.53±234.37, non-ED: 688.74±229.68, p=0.015).
Meanwhile, the significant association between B12
and ED was more prominent for mild ED at higher B12
levels. There are two possible explanations. First, our
results suggest that the function of B12 in ED might be
Table 5 Association between HCY, B12, FA and ED along with the increased levels of these indexes
Unadjusted Ageadjusted Multivariate adjusted
OR 95% CI Pvalue OR 95% CI Pvalue OR 95% CI Pvalue
HCY
Q1 1 1 1 1 1 1 1 1 1
Q2 1.130 0.838 to 1.525 0.422 1.022 0.753 to 1.387 0.887 1.012 0.744 to 1.377 0.938
Q3 1.292 0.958 to 1.743 0.094 1.187 0.875 to 1.610 0.271 1.210 0.890 to 1.646 0.224
Q4 1.276 0.946 to 1.720 0.110 1.091 0.803 to 1.483 0.578 1.103 0.810 to 1.502 0.534
B12
Q1 1 1 1 1 1 1 1 1 1
Q2 0.917 0.680 to 1.236 0.569 0.893 0.659 to 1.209 0.464 0.899 0.662 to 1.221 0.496
Q3 0.988 0.733 to 1.333 0.939 0.972 0.717 to 1.316 0.853 0.986 0.726 to 1.338 0.927
Q4 1.452 1.076 to 1.961 0.015 1.299 0.955 to 1.765 0.095 1.286 0.945 to 1.752 0.110
FA
Q1 1 1 1 1 1 1 1 1 1
Q2 1.313 0.974 to 1.770 0.074 1.325 0.978 to 1.795 0.069 1.332 0.981 to 1.811 0.067
Q3 1.300 0.963 to 1.755 0.086 1.243 0.916 to 1.687 0.163 1.286 0.944 to 1.751 0.111
Q4 1.198 0.888 to 1.616 0.236 1.094 0.806 to 1.487 0.564 1.116 0.819 to 1.522 0.487
In the multinomial logistic regression, the levels of HCY, B12 and FA were divided into quartiles (Q1<25%, 25%≤Q2≤50%, 50%<Q3≤75%,
Q4>75%), which were treated as the categorical dependent variables. Q1 was the reference. As a binary categorical variable, the ED was put
as the ‘Factors’.Multivariate adjusted: age, BMI, WHR, smoke and drink.
BMI, body mass index; B12, vitamin B12; ED, erectile dysfunction; FA, folic acid;HCY, homocysteine; WHR, waist:hip ratio.
on 23 May 2019 by guest. Protected by copyright.http://bmjopen.bmj.com/BMJ Open: first published as 10.1136/bmjopen-2018-023003 on 22 May 2019. Downloaded from
7
ChenY, etal. BMJ Open 2019;9:e023003. doi:10.1136/bmjopen-2018-023003
Open access
dose dependent. Excessive B12 levels would increase the
risk of mild ED with some unclear mechanisms. Second,
increased B12 might provide negative feedback for this
disease. At the beginning of the disease, defence mecha-
nisms are triggered. As a potential protective factor, the
absorption of B12 is enhanced. Combining the limited
reports, our study can also propose that B12 is signifi-
cantly associated with ED. As for the exact effects of B12
on ED, further studies are needed, which might pave
the way for the treatment of ED with B12 in the future.
Limitations
Our study verified the previous conclusions that HCY
could increase the risk of ED. However, some limitations
still need to be noted: (1) this study is a cross-sectional
analysis, which just reflects the status of specific time
points and populations; (2) there are limited numbers
of participants with primary educational status. So, the
results need to be examined further; (3) although we
have identified a significant association between B12
and ED, the exact mechanisms and effects were unclear
until now; and (4) after multiple testing, no positive
association can be detected, suggesting that our results
might be unstable. So, further studies will be needed.
CONCLUSIONS
ED is one of the most common male diseases. This study
was conducted in order to discover the functions of
HCY, B12 and FA in ED. Our results confirmed the posi-
tive correlations of HCY and ED. Meanwhile, B12 was
also likely to be significantly associated with ED. Further
studies with larger cohorts of participants should be
focused on the potential mechanisms and therapeutic
effects of B12 on ED.
Author afliations
1Department of Urology, The First Afliated Hospital of Guangxi Medical University,
Nanning, China
2Center for Genomicand Personalized Medicine, Guangxi Medical University,
Nanning, China
3Institute of Urology and Nephrology, The First Afliated Hospital of Guangxi Medical
University, Nanning, China
4Department of Reproduction, The Guangxi Zhuang Autonomous Region Family
Planning Research Center, Nanning, China
5School of Public Health, Guangxi Medical University, Nanning, China
6Faculty of Biochemistry and Molecular Medicine, University of Oulu, Oulu, Finland
Contributors YC, JieL, ZM and JC participated in participants’ collection, eld
investigation, design, writing and modication of all the paper. YC and JieL took part
in the statistical analysis. ZM and JC provided important advice for this paper. TL,
JiaL, JinL and G-HW provide efforts in the processes of modication.
Funding This study was funded by National Natural Science Foundation of China
(81770759, 81370857), and Innovation Project of Guangxi Graduate Education
(YCBZ2017037) .
Competing interests None declared.
Ethics approval This study was approved by the Medical Ethics Committee of
Guangxi Medical University.
Provenance and peer review Not commissioned; externally peer reviewed.
Data sharing statement The data for this study are available in the
supplementary materials. Further questions could be sent to ZM ( zengnanmo@
hotmail. com) and JC ( chengjiwen1977@ foxmail. com).
Open access This is an open access article distributed in accordance with the
Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which
permits others to distribute, remix, adapt, build upon this work non-commercially,
and license their derivative works on different terms, provided the original work is
properly cited, appropriate credit is given, any changes made indicated, and the use
is non-commercial. See: http:// creativecommons. org/ licenses/ by- nc/ 4. 0/.
REFERENCES
1. Braun M, Wassmer G, Klotz T, et al. Epidemiology of erectile
dysfunction: results of the 'Cologne Male Survey'. Int J Impot Res
2000;12:305–11.
2. Pinnock CB, Stapleton AM, Marshall VR. Erectile dysfunction in the
community: a prevalence study. Med J Aust 1999;171:353–7.
3. Nicolosi A, Moreira ED, Shirai M, et al. Epidemiology of erectile
dysfunction in four countries: cross-national study of the prevalence
and correlates of erectile dysfunction. Urology 2003;61:201–6.
4. Ayta IA, McKinlay JB, Krane RJ. The likely worldwide increase in
erectile dysfunction between 1995 and 2025 and some possible
policy consequences. BJU Int 1999;84:50–6.
5. Kloner RA, Speakman M. Erectile dysfunction and atherosclerosis.
Curr Atheroscler Rep 2002;4:397–401.
6. Sullivan ME, Keoghane SR, Miller MA. Vascular risk factors and
erectile dysfunction. BJU Int 2001;87:838–45.
7. Nehra A, Jackson G, Miner M, et al. Diagnosis and treatment of
erectile dysfunction for reduction of cardiovascular risk. J Urol
2013;189:2031–8.
8. Khan MA, Thompson CS, Emsley AM, et al. The interaction of
homocysteine and copper markedly inhibits the relaxation of rabbit
corpus cavernosum: new risk factors for angiopathic erectile
dysfunction? BJU Int 1999;84:720–4.
9. Zhang Z, Xu Z, Dai Y, et al. Elevated serum homocysteine level as an
independent risk factor for erectile dysfunction: a prospective pilot
case–control study. Andrologia 2017;49.
10. Long Y, Nie J. Homocysteine in renal injury. Kidney Dis 2016;2:80–7.
11. Lai WK, Kan MY. Homocysteine-induced endothelial dysfunction.
Ann Nutr Metab 2015;67:1–12.
12. Sansone M, Sansone A, Romano M, et al. Folate: a possible role in
erectile dysfunction? Aging Male 2018;21:1–5.
13. Chen Y, Xin X, Zhang H, et al. Immunization associated with erectile
dysfunction based on cross-sectional and genetic analyses. PLoS
One 2014;9:e111269.
14. Rosen RC, Cappelleri JC, Smith MD, et al. Development and
evaluation of an abridged, 5-item version of the International Index of
Erectile Function (IIEF-5) as a diagnostic tool for erectile dysfunction.
Int J Impot Res 1999;11:319–26.
15. Kupelian V, Araujo AB, Chiu GR, et al. Relative contributions
of modiable risk factors to erectile dysfunction: results from
the Boston Area Community Health (BACH) Survey. Prev Med
2010;50:19–25.
16. Guo H, Chi J, Xing Y, et al. Inuence of folic acid on plasma
homocysteine levels & arterial endothelial function in patients with
unstable angina. Indian J Med Res 2009;129:279–84.
17. McDonnold M, Mele LM, Myatt L, et al. Waist-to-hip ratio versus
body mass index as predictor of obesity-related pregnancy
outcomes. Am J Perinatol 2016;33:618–24.
18. Suchanek P, Kralova Lesna I, Mengerova O, et al. Which index best
correlates with body fat mass: BAI, BMI, waist or WHR? Neuro
Endocrinol Lett 2012;33(Suppl 2):78–82.
19. Tang B, Han CT, Zhang GM, et al. Waist-hip ratio (WHR), a better
predictor for prostate cancer than Body Mass Index (BMI): results
from a Chinese hospital-based biopsy cohort. Sci Rep 2017;7:43551.
20. Welborn TA, Dhaliwal SS. Preferred clinical measures of central
obesity for predicting mortality. Eur J Clin Nutr 2007;61:1373–9.
21. Salagre E, Vizuete AF, Leite M, et al. Homocysteine as a peripheral
biomarker in bipolar disorder: a meta-analysis. Eur Psychiatry
2017;43:81–91.
22. Elstgeest LE, Brouwer IA, Penninx BW, et al. Vitamin B12,
homocysteine and depressive symptoms: a longitudinal study
among older adults. Eur J Clin Nutr 2017;71:468–75.
23. Momin M, Jia J, Fan F, et al. Relationship between plasma
homocysteine level and lipid proles in a community-based Chinese
population. Lipids Health Dis 2017;16:54.
24. Chen J, Li G, Xu Z, et al. Elevated plasma homocysteine level
increased the risk of early renal impairment in acute ischemic
stroke patients. Cell Mol Neurobiol 2017;37:1399–405.
25. Li T, Chen Y, Li J, et al. Serum homocysteine concentration is
signicantly associated with inammatory/immune factors. PLoS
One 2015;10:e0138099.
on 23 May 2019 by guest. Protected by copyright.http://bmjopen.bmj.com/BMJ Open: first published as 10.1136/bmjopen-2018-023003 on 22 May 2019. Downloaded from
8ChenY, etal. BMJ Open 2019;9:e023003. doi:10.1136/bmjopen-2018-023003
Open access
26. Sahu A, Gupta T, Kavishwar A, et al. Cardiovascular diseases
risk prediction by homocysteine in comparison to other markers:
a study from Madhya Pradesh. J Assoc Physicians India
2015;63:37–40.
27. Yeh JK, Chen CC, Hsieh MJ, et al. Impact of homocysteine level
on long-term cardiovascular outcomes in patients after coronary
artery stenting. J Atheroscler Thromb 2017;24:696–705.
28. Baumann F, Hehli D, Makaloski V, et al. Erectile dysfunction—
overview from a cardiovascular perspective. Vasa 2017;46:347–53.
29. McDowell IF, Lang D. Homocysteine and endothelial dysfunction: a
link with cardiovascular disease. J Nutr 2000;130:369S–72.
30. Rajfer J, Aronson WJ, Bush PA, et al. Nitric oxide as a mediator of
relaxation of the corpus cavernosum in response to nonadrenergic,
noncholinergic neurotransmission. N Engl J Med 1992;326:90–4.
31. Deaneld J, Donald A, Ferri C, et al. Endothelial function and
dysfunction. Part I: methodological issues for assessment in the
different vascular beds: a statement by the working group on
endothelin and endothelial factors of the european society of
hypertension. J Hypertens 2005;23:7–17.
32. Eikelboom JW, Lonn E, Genest J, et al. Homocyst(e)ine and
cardiovascular disease: a critical review of the epidemiologic
evidence. Ann Intern Med 1999;131:363–75.
33. O'Leary F, Samman S. Vitamin B12 in health and disease. Nutrients
2010;2:299–316.
34. Yan WJ, Yu N, Yin TL, et al. A new potential risk factor in patients
with erectile dysfunction and premature ejaculation: folate
deciency. Asian J Androl 2014;16:902–6.
35. Giovannone R, Busetto GM, Antonini G, et al.
Hyperhomocysteinemia as an early predictor of erectile
dysfunction. Medicine 2015;94:e1556.
on 23 May 2019 by guest. Protected by copyright.http://bmjopen.bmj.com/BMJ Open: first published as 10.1136/bmjopen-2018-023003 on 22 May 2019. Downloaded from