Original Article
Dynamic penile peak systolic velocity predicts major
adverse cardiovascular events in hypertensive patients
with erectile dysfunction
Nikolaos Ioakeimidis a, Charalambos Vlachopoulos a, Konstantinos Rokkas a, Zisis Kratiras b,
Athanasios Angelis a, Alexis Samentzas a, Charalambos Fassoulakis b, and Dimitris Tousoulis a
Objective: Hypertension is associated with an abnormal
penile blood flow. Reduced dynamic penile peak systolic
velocity (D-PSV) correlates with adverse cardiovascular
outcomes. The aim of this study is to investigate whether
abnormal penile blood flow predicts major adverse
cardiovascular events (MACE) in hypertensive men.
Methods: In total, 298 hypertensive men (55
9 y/o)
without known cardiovascular disease or diabetes were
evaluated for cavernous vascular disease severity by
dynamic penile Doppler ultrasound. The whole population
was divided into tertiles according to D-PSV reduction (low
tertile <25 cm/s; middle tertile 25–35 cm/s; and high tertile
>35 cm/s). Predictive performance was evaluated with
calibration, discrimination, and reclassification.
Results: During the mean follow-up period of 4.9 years, a
total of 22 (7%) MACE occurred. D-PSV level was
associated with MACE and the differences between the
tertiles were significant (Mantel log-rank test: 6.54;
P < 0.01). A Cox proportional hazard model showed that
study participants in the lowest D-PSV tertile (<25 cm/s)
had an approximately 3.5-fold higher MACE risk compared
with those in the highest D-PSV tertile (>35 cm/s) after
adjustment for age, systolic pressure, metabolic
parameters, smoking, C-reactive protein, and testosterone
levels. Low D-PSV did not significantly improve the C-
statistic model (0.774 vs. 0.767; P ¼ 0.44), whereas the
calibration was satisfactory (Hosmer–Lemeshow X2 ¼ 8.73,
P ¼ 0.30). When only intermediate-risk patients were
evaluated, the risk reclassification beyond traditional risk
factors resulted in a clinical net reclassification index of
9.2% that was marginally significant (P ¼ 0.07). The
integrated discrimination improvement index showed
better performance of the model that included D-PSV
compared with the reference model in identifying MACE
(improvement index: 0.047, P ¼ 0.038).
Conclusion: Low-penile blood flow predicts MACE in
hypertensive patients free of clinical atherosclerosis. This
predictive ability is independent of the severity of
hypertension and decreased testosterone that is often
present in such patients.
Keywords: erectile dysfunction, major adverse
cardiovascular events, penile peak systolic velocity
860 www.jhypertension.com
Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.
Abbreviations: ANCOVA, analysis of covariance; ARB,
angiotensin II receptor blocker; CAD, coronary artery
disease; CVD, cardiovascular disease; D-PSV, dynamic
penile peak systolic velocity; FRS, Framingham Risk Score;
ICI, intracavernous injection; IDI, integrated discrimination
improvement; MACE, major adverse cardiovascular events;
NRI, net reclassification index; PCDU, penile color Doppler
ultrasound
INTRODUCTION
E
rectile dysfunction is defined as the persistent
inability to attain and maintain an erection suffi-
cient to permit satisfactory sexual performance [1].
Solid evidence indicates that erectile dysfunction is an
early and powerful predictive marker of cardiovascular
disease (CVD). Indeed, erectile dysfunction and CVD
share the same risk factors and probably the same under-
lying pathophysiologic mechanisms [2,3]. Presence of
erectile dysfunction appears to be indicative of asympto-
matic CVD [4] and predictive of forthcoming major
adverse cardiovascular events (MACE) [5] and precede a
cardiac event by 2–5 years [6,7].
Hypertension represents a great health issue and popu-
lation burden given that almost 25% of the adult population
worldwide is hypertensive [8]. Furthermore, it is a major risk
factor of both CVD and erectile dysfunction [9,10]. The
relative risk of erectile dysfunction is almost two times
greater in hypertensive than in normotensive adults [11],
whereas hypertension is associated with an approximately
two-fold increase in the likelihood of impaired cavernous
vasculature [12].
Journal of Hypertension 2016, 34:860–868
a
1st Department of Cardiology, Hypertension and Cardiovascular Diseases and Sexual
Health Units, Athens Medical School and bDepartment of Urology, Hippokration
Hospital, Athens, Greece
Correspondence to Charalambos Vlachopoulos, 1st Cardiology Department, Athens
Medical School, Hippokration Hospital, Profiti Elia 24, Athens 14575, Greece. Tel: +30
697 227 2727; fax: +30 210 747 3374; e-mail: cvlachop@otenet.gr
Received 27 July 2015 Revised 16 December 2015 Accepted 8 January 2016
J Hypertens 34:860–868 Copyright ß 2016 Wolters Kluwer Health, Inc. All rights
reserved.
DOI:10.1097/HJH.0000000000000877
Volume 34  Number 5  May 2016
 Peak systolic velocity and major adverse cardiovascular events
Penile color Doppler ultrasound (PCDU) is considered
as the mainstay in the evaluation of the penile vasculature
and in the assessment of vasculogenic erectile dysfunc-
tion [13]. Impaired cavernous vasculature, particularly
decreased peak systolic velocity (PSV) has been related
to ‘silent’ CVD [14] and increased risk of future MACE
[15,16]. However, the predictive value of low PSV in hyper-
tensive men has not been investigated yet. Thus, the aim
of our study was to evaluate whether the impaired penile
blood flow, evaluated by PCDU, could be associated with
forthcoming MACE in a pure cohort of arterial hypertension
patients, without diabetes or any clinical manifestation
of CVD.
METHODS
Study population
From January 2004 to June 2010, we enrolled consecutive
male hypertensive patients that were referred to the Hy-
pertension Unit and the Cardiovascular Diseases and
Sexual Health Unit of the 1st Department of Cardiology
of Athens Medical School for evaluation of erectile dys-
function. These patients had erectile dysfunction symp-
toms of recent onset but they were neither symptomatic of
CVD nor had they documented CVD disease. These
patients were evaluated comprehensively with noninva-
sive tests (exercise stress test and stress echocardiography)
and coronary angiography, when indicated, to reveal
occult coronary artery disease (CAD) [5]. Initially, 366
men were assessed for inclusion in the study. In total, 51
(14%) had positive one or both of the two noninvasive
procedures. Of them, 44 proved to have significant cor-
onary stenosis (44/366, 12%). All patients with angiograph-
ically documented CAD at onset, as well as 24 (6.5%) men
who were lost to follow-up, were excluded from the study.
Finally, 298 men were analyzed.
Patients who had diabetes, evidence of antihypertensive
drug-related hypotension, secondary hypertension, renal
insufficiency (creatinine >2 mg/dl), or an acute or chronic
inflammatory/infectious disease were excluded. Malig-
nancy, overt endocrine disease and use of steroid hor-
mones were also excluded, because these conditions
may have a significant influence on both plasma sex hor-
mones and clinical outcome. The study complies with the
Declaration of Helsinki; it was approved by our Institutional
Research Ethics Committee and all study participants gave
informed consent.
Study design–study protocol
Evaluation of hypertension and other risk factors
Arterial blood pressure (BP) of each patient was measured
three times on the right arm with a mercury sphygmoman-
ometer. Mean BP was calculated as DBPþ[(SBPDBP)/3].
Brachial pulse pressure was calculated as SBPDBP. Accord-
ing to the guidelines of the European Society of Hyperten-
sion/European Society of Cardiology (ESH/ESC) [17],
hypertension was defined as SBP of 140 mmHg or higher,
DBP of 90 mmHg or higher, or taking antihypertensive drug
treatment. Stage 1 hypertension was defined as SBP of 140–
159 mmHg and/or DBP of 90–99 mmHg, stage 2 as SBP of
Journal of Hypertension
Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.
160–179 mmHg and/or DBP of 100–109 mmHg and stage 3
as SBP of at least 180 mmHg and/or DBP of at least
110 mmHg. BP control was defined as poor when SBP
and/or DBP (measured at recruitment phase) were above
normal values in patients under antihypertensive medication.
All hypertensive men in our unit were evaluated compre-
hensively at initial presentation with echocardiography and
vascular tests for detection of target organ damage (TOD).
Organ damage was defined when patients had either echo-
cardiographically documented left ventricular hypertrophy
(index above 115 g/m2) and/or high-carotid intima–media
thickness (above 0.9 mm)/plaque, and/or or high-carotid–
femoral pulse wave velocity (above 10 m/s) [17].
A 10-year risk for cardiovascular events according to the
Framingham was calculated as described by D’ Agostino [18]
and patients were divided into three risk categories (low
<10%), intermediate (10–20%), and high (>20%) as recom-
mended by Wilson [19]. Furthermore, the whole population
was stratified in different 10-year risk of mortality categories
(low, moderate, moderate-high, and high) on the basis of BP
category, number of other cardiovascular risk factors, and the
presence of asymptomatic target-organ damage as defined
by the 2013 ESH/ESC guidelines [17].
The other risk factors were defined according to the
European Society of Cardiology guidelines [20] as follows:
hypercholesterolemia as total and/or low-density lipopro-
tein (LDL) cholesterol level higher than 5 mmol/l (190 mg/
dl) and 3 mmol/l (115 mg/dl), respectively, obesity as BMI
higher than 30 kg/m2.
Morning blood samples were taken from each patient to
determine total testosterone, glucose, LDL cholesterol,
high-density lipoprotein (HDL) cholesterol, triglycerides
and high-sensitivity C-reactive protein.
Penile color duplex ultrasound
Patients were initially evaluated for erectile dysfunction by
completing the validated International Index Erectile Func-
tion-5 questionnaire [21]. All patients underwent a PCDU in
accordance with ‘Standard Practice in Sexual Medicine’
developed by the International Society for Sexual Medicine
Standards Committee [13,22]. Penile morphology and
hemodynamics were evaluated in flaccid state and after
pharmacologically induced erection, using intracavernous
injection (ICI) of a vasoactive agent. All PCDUs were
performed by one experienced sonographer, in an isolated
quite room with the presence of only the operator. A high-
resolution linear array ultrasound probe (7–12 MHz),
specific for small parts, was used. The penis was initially
evaluated in flaccid state with transverse and sagital scan-
ning in B-mode imaging, highlighting presence of plaques,
cavernosal heterogeneity, and calcification.
Pharmacologically induced erection was achieved by ICI
of 10 mg of Prostaglandin E1 (Alprostadol) using a 30 G
needle. Prostaglandin E1 as a single agent was preferred
rather than bimix or trimix to prevent episodes of pro-
longed erection and priapism [22]. Manual genital self-
stimulation was allowed in order to achieve better and
more rigid erections, while audiovisual equipment for
sexual stimulation was not available. Using the Doppler
color mode both cavernosal arteries were visualized and
dynamic D-PSV were calculated for each cavernosal artery
www.jhypertension.com 861
Ioakeimidis et al.
at various time points up to 20 min (every 5 min). Finally, for
patients where sufficient erections and stable state was not
achieved, a last measurement of D-PSV 30 min after the ICI
was performed. Intra and interobserver variability for PSV
measurements were all lower than 5%.
Main end point follow-up
The incidence of MACE was the outcome end point of our
study. We investigated the association between penile
hemodynamics and MACE and tried to identify whether
a decreased D-PSV could predict increased incidence of
MACE in our study population. Information on MACE up to
February 2012 was obtained by visits to our unit, phone
interview, and/or mail. Each study participant or a family
member completed the questionnaire on MACE, current
medication, and health status. MACE analyzed as the end
points of this study included cardiovascular death, CAD,
cerebrovascular disease, and peripheral arterial disease. All
MACE included in analysis were validated by research
doctors who were unaware of the patient’s PCDU using
source data, including death certificates, hospital record
forms, and interview. Also, nonfatal cases of CVD requiring
hospitalization were identified through the regional hospi-
tal discharge system.
Statistical analysis
Sample size estimation and power analysis
Sample size calculation was based on the hypothesis that
patients with low D-PSV would have a relative risk of at
least 2.5 for MACE compared with patients with high D-PSV
over a median follow-up of 5 years. Therefore, we esti-
mated that 286 study participants in total would provide
80% power at the 5% level of significance to detect a MACE
difference corresponding to a relative risk of at least 2.5
between groups (low D-PSV and high D-PSV).
Population analysis
We assessed differences in normal variables by the Student
t-test and differences in nonnormal variables by the Mann–
Whitney U test. The Pearson x2 test was used for frequency
comparison. The analysis of covariance (ANCOVA) was
used to determine whether there are any significant differ-
ences between the D-PSV means of two or more independ-
ent groups (specifically, the adjusted means).
Outcome and prognostic impact of dynamic penile
peak systolic velocity
Free of MACE survival among D-PSV tertiles was estimated
by Kaplan–Meier curves and compared by the Mantel log-
rank test. In addition, we calculated the expected number
of events at 10-years in each risk category using Kaplan–
Meier estimates. The primary analyses used the unadjusted
Cox proportional hazards estimates to evaluate the risk
ratios of cardiovascular risk factors, such as age, hyper-
cholesterolemia, obesity, smoking, hypertension
parameters such as BP level (systolic, diastolic, and pulse
pressure), degree of hypertension and antihypertensive
drugs, and biological variables, such as fasting glucose,
total cholesterol, triglycerides, HDL total testosterone, and
862 www.jhypertension.com
Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.
high-sensitivity C-reactive protein. As 10 events per cova-
riate is recommended, a series of multivariate Cox
regression models with no more than four variables at a
time were applied to determine the independent effect of
D-PSV on MACE. The Cox models were adjusted for D-PSV,
the two most important covariates according to univariate
Cox analysis (age and SBP), and each of the other variables
at a time in rotation. The robustness of the estimated hazard
ratios was evaluated using bootstrap resampling method of
1000 datasets. A bias of the estimate less than 0.01 was
considered adequate for the performance of the calculated
hazard ratio.
Prognostic performance
To identify D-PSV that gave optimal sensitivities and spe-
cificities receiver operating characteristic curve were drawn
by plotting sensitivity versus 1-specificity. Cutoff values
were determined using individual data collected prospec-
tively during the first visit after January 2004. Sensitivity,
specificity, positive and negative predictive values, and
likelihood ratios were calculated to determine the predic-
tive power of D-PSV in distinguishing between MACE and
no MACE. Their corresponding 95% confidence intervals
(CI) correspond to a sensitivity of 90%, thus allowing only
a 10% of false negative results.
Incremental predictive usefulness of dynamic penile
peak systolic velocity
First, we assessed model calibration (i.e., concordance of
observed risk and that predicted by the model with PSV by
calculating the Hosmer–Lemeshow x2 statistic for Cox
models [23]. Second, we evaluated the discriminatory capa-
bility of D-PSV and traditional risk factors using C-statistics,
which is an extension of the traditional receiver–operator
characteristic curve analysis to survival analysis. Finally, we
assessed the ability of D-PSV to reclassify risk in our
population. As we had only 4.9 years of follow-up in our
study, for the reclassification analysis we did not use stand-
ard risk categories based on Framingham Risk Score (FRS).
Instead, we computed the predicted risk for all patients
using a Cox regression model that included only the stand-
ard risk factors used for calculation of FRS (global cardio-
vascular risk) (age, SBP, cholesterol, HDL-cholesterol, and
smoking). To better identify a group that the dynamic PSV
reclassifies patients in a considerable percentage we
defined cutoff points for risk groups according to the
quartiles of the predicted risk in patients who experienced
an event during the 4.9-year follow-up (25th percentile of
risk¼3.5%, 50th percentile of risk¼5.5%, 75th percentile of
risk¼8.5%). Then, we constructed three groups of risk (low:
<3.5%, intermediate: 3.5–8.5%, high: >8.5%) with the two
middle quartiles combined into a single ‘intermediate’
category, since the reliable assessment of reclassification
ability is of utmost importance in patients belonging to such
risk category (intermediate). We cross-classified groups of
risk (<3.5%, 3.5–8.5%, >8.5%) based on a model that
included standard risk factors against groups of risk based
on a model where D-PSV was added. Cross-classification
was evaluated separately in patients who did or did not
experience an event. We then calculated the net
Volume 34  Number 5  May 2016
 Peak systolic velocity and major adverse cardiovascular events
reclassification index (NRI), which quantifies the ability of
D-PSV to predict risk when added to the classical risk
factors. The NRI was calculated as the sum of the difference
in proportions of patients reclassified into a higher risk
group minus the proportion reclassified into a lower risk
group among men who developed events, plus the differ-
ence in the proportion of patients reclassified into a lower
risk category minus the proportion reclassified into a higher
risk group among men who did not develop events. The
significance of the net reclassification improvement was
assessed with an asymptotic test as described by Pencina
et al. [24].
To investigate whether reclassification with D-PSV is
affected by the presence of target-organ damage, an
additional reclassification chart was performed by using
Cox regression models that included: the standard risk
factors used for calculation of FRS (age, SBP, cholesterol,
HDL-cholesterol, and smoking) and the presence of TOD
according to the paradigm of ESC/ESH guidelines where
hypertensive patients with TOD were assigned to a higher
risk category.
Given that the NRI is highly dependent upon the pre-
specified categories of risk, model performance with the
addition of the D-PSV was further evaluated using the
integrated discrimination improvement (IDI) index, which
is a continuous assessment of reclassification improvement
independent on the risk categories used. The IDI compares
two models according to the average difference in pre-
dicted risk between those who have the event and those
who do not [24]. If the new model assigns a higher risk to
those who will have a MACE and a lower risk to those who
will not, as compared with the reference model, the IDI will
be above zero. Therefore, the IDI can be interpreted as the
average net improvement in the predicted risk of the out-
come in the new model compared to the reference model.
The 95% CIs were calculated from the binomial distri-
bution. Data analysis was performed with SPSS software,
version 18.0 (Chicago, Illinois, USA) and MedCalc statistical
software, version 11.5.1 (Mariakerke, Belgium). Analyses
were completed with STATA version 13 (StataCorp, College
Station, Texas, USA).
RESULTS
Dynamic penile peak systolic velocity and
hypertension
The mean age of participants at entry was 56 years (range
31–72 years, Table 1). Figure 1a illustrates the inverse
association between severity of hypertension and D-PSV
(P < 0.001 by ANCOVA, post hoc P < 0.05 for comparing
grade 3 with grade 2 hypertension patients and grade 2 with
grade 1 hypertension patients). Duration of hypertension
also affected severity of penile vascular disease as shown in
Figure 1b (P < 0.01 by ANCOVA, post hoc P < 0.01 for
comparing men with long-standing hypertension with
men with duration less than 3 years). Treated hypertensive
patients had significantly lower Doppler velocities as com-
pared to untreated men (31.5 vs. 34.6 cm/s, P < 0.05),
however the number of antihypertensive drugs did not
affect erectile dysfunction severity (P ¼ NS, Fig. 1c). Patients
receiving either a b-blocker or a calcium channel blocker as
Journal of Hypertension
Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.
monotherapy had lower mean PSV compared to men
receiving an angiotensin-converting enzyme inhibitor
(ACEi) or angiotensin II receptor blocker (ARB) (31.2
and 30.5 cm/s vs. 34.5 cm/s, respectively), however the
differences after adjustment for age and BP level were
not significant (P ¼ 0.21 and P ¼ 0.14). Among patients with
a 2-drug combination therapy, the group of men who are
taking ACEi/ARB plus diuretic had significantly lower age
and BP level-adjusted PSV compared with study paticipants
taking either a ACEi/ARB plus calcium channel blocker or
other combinations without a diuretic (27.2 vs. 31.9 cm/s or
30.6 cm/s, respectively. all P < 0.05).
Occurrence of major adverse cardiovascular
events
The mean follow-up time was 4.9 years, during which 22
(7.4%) study participants developed MACE (18 nonfatal, 4
fatal). Of those with CAD events (n ¼ 14), two were CAD
deaths, two were myocardial infarction with survival
beyond 1 day and 10 were documented angina pectoris
or coronary revascularization. Other MACE events were
cerebrovascular events (n ¼ 5), and new onset of peripheral
arterial disease (n ¼ 3, 1 fatal).
The baseline clinical characteristics of the 298 partici-
pants are given in Table 1. Compared to patients who did
not experience MACE, hypertensive study participants
who developed MACE were older (P < 0.01) and had
higher pulse pressure (P < 0.05) and lower total testos-
terone concentration (P < 0.05). SBP, DBP, and mean BP
were not significantly different between groups. The
prevalence of grade II/III hypertension was not different
between patients with MACE and study participants
without MACE. Also, the prevalence of patients treated
with antihypertensive drugs was greater in patients with
MACE compared to men who did not experience a MACE
(74 vs. 55%, respectively). Finally, the group of hyper-
tensive patients suffering a MACE had higher FRS com-
pared with study participants without MACE (P < 0.001), a
greater prevalence of asymptomatic organ damage
(P ¼ 0.003) and a marginally greater likelihood of high
10-year risk of cardiovascular mortality according to ESH/
ESC guidelines (P ¼ 0.06).
Regarding erectile dysfunction characteristics, men who
developed MACE had reduced D-PSV (28
10 vs.
32
8 cm/s, P < 0.01) and a greater prevalence of severe
arterial insufficiency compared with men without MACE (50
vs. 15%, P < 0.001). Mean International Index Erectile Func-
tion-5 was not different between the two groups.
Outcome and prognostic impact of D-PSV
The whole population was divided into tertiles according to
the D-PSV level reduction (low tertile <25 cm/s; middle
tertile 25–35 cm/s; and high tertile >35 cm/s). Kaplan–
Meier survival curve for MACE by tertiles of D-PSV at
baseline is illustrated in Figure 2. Penile D-PSV was signifi-
cantly associated with MACE and the differences between
the D-PSV tertiles were significant (Mantel log-rank test:
8.37; P < 0.01). The estimated cumulative proportion sur-
viving at 10 years for the low, middle, and high D-PSV tertile
was 0.58 (58%), 0.77 (77%), and 0.85 (85%), respectively.
www.jhypertension.com 863
Ioakeimidis et al.

TABLE 1. Baseline characteristics of patients with a major adverse cardiovascular events and study participants without major adverse
cardiovascular events
All patients (n ¼ 298) MACE (n ¼ 22) No MACE (n ¼ 276) P valuea
Age (years) 56
10 61
6 56
10 <0.01
BMI (kg/m2) 28.4
4.2 27.6
2.5 28.6
3.0 0.24
SBP (mmHg) 134
17 140
17 133
17 0.09
DBP (mmHg) 83
10 84
7 83
8 0.69
Pulse pressure (mmHg) 51
13 56
13 50
13 <0.05
Mean pressure (mmHg) 100
11 103
10 100
10 0.27
Total cholesterol (mg/dl) 209
35 208
29 209
32 0.89
HDL cholesterol (mg/dl) 45
9 45
9 45
9 0.92
Triglycerides (mg/dl) 109 (81–133) 114 (78–139) 109 (80–131) 0.97
Glucose (mg/dl) 105 (88–117) 108 (89–120) 105 (87–116) 0.55
hsCRP (mg/l) 1.7 (1.3–2.1) 1.9 (1.5–2.4) 1.7 (1.3–2.0) 0.62
Total testosterone (ng/ml) 4.5
1.2 4.0
0.8 4.6
1.5 <0.05
Grade II/III hypertension n (%) 74 (22) 10 (45) 64 (23) 0.26
Hypercholesterolemia n (%) 170 (67) 11 (50) 159 (57) 0.26
Smoking n (%) 154 (56) 10 (45) 144 (52) 0.35
FRS 16
6 20
6 15
5 <0.001
High FRS n (%) 83 (27) 11 (50) 72 (24) 0.02
10-year risk of CV mortality stratificationb
High risk category n (%) 69 (23) 9 (41) 60 (22) 0.06
Target organ damage n (%) 52 (17) 9 (41) 43 (15) 0.003
PWV (m/s) 8.8
1.3 9.7
1.5 8.6
1.2 <0.001
cIMT (mm) 0.7
0.3 1.2
0.4 0.7
0.2 <0.001
LVM index (g/m2) 83
35 92
32 81
36 0.005
Drug therapy, n (%)
Antihypertenive therapy 156 (57) 13 (59) 143 (52) 0.35
Statins 73 (20) 7 (32) 66 (24) 0.47
Poor BP controlc 50 (17) 7 (32) 43 (16) 0.08
ED parameters
IIEF-5 13.1
4 12.5
3 13
4 0.48
ED duration (years) 2.2
2.0 2.5
2.0 2.2
2.0 0.53
D-PSV (cm/s) 33
9 29
10 34
8 0.01
SAI, n (%)d 52 (17) 11 (50) 41 (15) <0.001

BP, blood pressure; cIMT, carotid intima media thickness; CV, cardiovascular; D-PSV, dynamic peak systolic velocity; ED, erectile dysfunction; FRS, Framingham Risk Score; HDL, high-
density lipoprotein; hsCRP, high-sensitivity C-reactive protein; IIEF-5, International Index Erectile Function-5; LVM, left ventricular mass, MACE, major adverse cardiovascular events; PWV,
pulse wave velocity; SAI, severe arterial insufficiency. Categorical variables are presented as absolute (relative) frequencies; continuous variables, as mean
SD or median (interquartile
range).
a
P value: refers to differences between patients with a MACE and study participants without MACE.
b
As defined by the 2013 European Society of Hypertension/European Society of Cardiology guidelines.
c
BP control was defined as poor when SBP and /or DBP (measured at recruitment phase) were above normal values in patients under antihypertensive medication.
d
SAI: D-PSV < 25 cm/s.

The unadjusted Cox models show a significant associ-
ation of age (P ¼ 0.008), total testosterone (P ¼ 0.015), and
D-PSV (P ¼ 0.02) with MACE. Table 2 shows MACE by PSV
tertiles and all multivariate Cox regression models con-
structed to investigate the association between D-PSV
and MACE. In all multivariate Cox models, study partici-
pants at the lowest D-PSV tertile, but not those in the middle
tertile, had a significantly increased risk for MACE com-
pared to those with the highest tertile after adjustment for
age, SBP, and each of the other variables examined in

(a) (b) (c)

60
P < 0.001 P < 0.01 P = NS

50
D-PSV (cm/s)

40

30

20

10

Stage 1 Stage 2 Stage 3 <2 2–5 >5 1 2 >2

Hypertension severity Duration of hypertension (years) Antihypertensive drug (N)
FIGURE 1 Dynamic peak systolic velocity (D-PSV) values in subgroups of erectile dysfunction patients categorised by (a) hypertension severity, (b) duration of hypertension
and (c) number of antihypertensive drugs. Bars represent mean; vertical lines represent standard deviation. The P value was obtained by one-way analysis of variance across
the three subgroups.

864 www.jhypertension.com Volume 34  Number 5  May 2016

Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.

 Peak systolic velocity and major adverse cardiovascular events

cholesterol, HDL, and current smoking [yes/no]), and hy-

1.0
pertension categorized according to BP readings by ESC
definition (stage I, and stage II–III), was 0.767. Addition of
0.8 D-PSV to this model offered only a small improvement in
the resulting C statistic to 0.774 (P ¼ 0.44, for comparison
Cumulative survival

between the area under the curves).

0.6
Calibration
The X2 ¼ 8.73 of the Hosmer–Lemeshow test was not
0.4 statistically significant (P ¼ 0.30), indicating satisfactory
calibration of the model (that is, the proportion of MACE
>35 cm/s
predicted by D-PSV does not significantly differ from
0.2 observed MACE).
25–35 cm/s
< 25 cm/s
0.0
Reclassification
In the whole study population, addition of D-PSV to the
0 2 4 6 8 standard model including the risk factors resulted in
Follow-up (years) upward reclassification (correct reclassification into a
FIGURE 2 Kaplan–Meier curves for major adverse cardiovascular events (MACE) by higher risk group) of four (18.1%) of patients who experi-
tertile group of dynamic peak systolic velocity (D-PSV). Cutoffs of the tertiles for enced a MACE, as well as upward reclassification (wrong
D-PSV were 25 and 35 cm/s.
reclassification into a higher risk group) of 29 (10.5%) of
patients who did not experience an event (Table 3).
univariate analysis in rotation. The association between D- Furthermore, addition of D-PSV resulted in downward
PSV and MACE remained statistically significant even after reclassification (wrong reclassification into a lower risk
adjusting for total testosterone levels. The bootstrap pro- group) of 0 (0%) of patients who experienced a MACE,
cedure showed that the bias of the hazard ratio (HR) as as well as downward reclassification (correct reclassifica-
regard the D-PSV was very low, that is, 0.006, indicating tion into a lower risk group) of two (0.7%) of patients who
robustness of the estimated HR. did not experience an event. The overall NRI was 8.1%
(¼[18.1–10.5%] þ [0.7–0%]), which was not statistically
Prognostic performance significant (P ¼ 0.27). When only intermediate-risk patients
A cutoff of 38.6 cm/s gave a sensitivity of 87%, a specificity (3.5–8.5% risk group) were evaluated, the respective
of 74%, a positive likelihood ratio of 3.4 and a negative reclassifications were 13.6, 0, 0.7%, and 5.1%, resulting in
likelihood ratio of 0.16. The cutoff value was associated an NRI (clinical NRI) of 9.2% that was marginally significant
with a negative predictive value (ability to ‘rule out’ MACE) (P ¼ 0.07).
of 94% (95% CI 90.3–98.6%), indicating that a patient with When patients had been already reclassified to a higher
D-PSV higher than 38.6 cm/s has a 0.94 probability of being risk category with inclusion of target-organ damage, the
free of MACE during follow-up. addition of D-PSV in the groups of predicted risk resulted in
a relatively small overall NRI 6.9 (¼ [18.1–11.9%] þ [0.7–
Discrimination 0%]) (P ¼ 0.48).
The C-statistic for the full multivariate model that incorp- The IDI index demonstrated a significantly better dis-
orated variables that are part of the FRS (age, total crimination by the model including D-PSV as compared

TABLE 2. Major adverse cardiovascular events by tertiles of dynamic peak systolic velocity
1st tertile (<25 cm/s) 2nd tertile (25–35 cm/s) 3rd tertile (>35 cm/s)
HR (95% CI) for MACE
Unadjusted 4.58 (1.3–16.9) 1.97 (0.49–7.87) 1 (Ref)
Adjusted for age, SP 3.74 (1.06–13.05)y 1.89 (0.74–7.15) 1 (Ref)
Adjusted for age, SP along with each of the following covariates:
BMI 3.53 (0.98–12.72)y 1.82 (0.45–7.35) 1 (Ref)
Glucose 3.47 (0.97–12.70)y 1.90 (0.47–7.62) 1 (Ref)
Cholesterol 3.48 (0.97–12.48)y 1.91 (0.31–7.66) 1 (Ref)
CRP 3.39 (0.94–12.09)y 1.88 (0.47–7.55) 1 (Ref)
TT 3.56 (1.00–12.64)y 1.79 (0.44–7.24) 1 (Ref)
Hypertension stage>1 (1, yes; 0, no) 3.71 (1.05–13.12)y 1.98 (0.49–7.91) 1 (Ref)
Antihypertensive drugs (1, yes; 0, no) 3.46 (0.98–12.68)y 1.87 (0.46–7.38) 1 (Ref)
Poor BP control (1, yes; 0, no) 3.37 (0.92–12.17)y 2.01 (0.48–7.57) 1 (Ref)
Hypercholesterolemia (1, yes; 0, no) 3.59 (1.00–12.93)y 1.85 (0.46–7.41) 1 (Ref)
Current smoking (1, yes; 0, no) 3.44 (0.96–12.30)y 1.79 (0.44–7.23) 1 (Ref)

BP, blood pressure control; CI, confidence interval; CRP, C reactive protein; HR, hazard ratio; MACE, major adverse cardiovascular events; SP, systolic pressure; TT, total testosterone.

P < 0.01.
y
P < 0.05 for 3rd tertile vs. 1st tertile.

Journal of Hypertension www.jhypertension.com 865

Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.
Ioakeimidis et al.

TABLE 3. Predicted risk for a major cardiovascular event before and after reclassification with dynamic peak systolic velocity in patients
who did (a) or did not (b) experience a major cardiovascular event
MACE

Model with D-PSV

<3.5% 3.5–8.5% >8.5% Total
Model without D-PSV
<3.5% 9 (40.9%)a 1 (4.5%)b 0 (0%)b 10 (45.5%)
3.5–8.5% 0 (0%)c 6 (27.3%)a 3 (13.6%)b 9 (40.9%)
>8.5% 0 (0%)c 0 (0%)c 3 (13.6%)a 3 (13.6%)
Total 9 (40.9%) 7 (31.8%) 6 (27.3%) 22 (100%)

No MACE

Model with D-PSV

<3.5% 3.5–8.5% >8.5% Total
Model without D-PSV
<3.5% 49 (17.7%)a 15 (5.4%)c 0 (0%)c 64 (23.2%)
3.5–8.5% 2 (0.7%)b 124 (44.9%)a 14 (5.1%)c 140 (50.7%)
>8.5% 0 (0%)b 0 (0%)b 72 (26.1%)a 72 (26.1%)
Total 51 (18.4%) 139 (50.3%) 86 (31.2%) 276 (100%)

Values on each row to the right of the value with an asterisk were upwardly classified, and those to the left were downwardly classified by the model that included D-PSV. D-PSV,
major adverse cardiovascular events; MACE, major adverse cardiovascular events.
a
No reclassification.
b
Correct reclassification.
c
Wrong reclassification.

with the reference model. The IDI index was positive and
significantly different from zero (IDI: 0.047; z ¼ 2.09,
P ¼ 0.038), suggesting improvement in reclassification
when the model incorporating D-PSV was compared with
the reference model.
DISCUSSION
Our study is the first to investigate the prognostic role of
ultrasonographically evaluated severity of penile vascular
disease in middle-aged hypertensive individuals with erec-
tile dysfunction. The principal finding is that penile arterial
insufficiency defined when D-PSV is lower than 25 cm/s
predicts MACE in long-term follow-up independent of age,
severity of hypertension, and testosterone levels. Also, a D-
PSV value of 38.6 cm/s was associated with a negative
predictive value (ability to rule out MACE) of 94%. In
intermediate-risk patients, the clinical NRI was marginally
significant.
Clinical implications
Identification of those erectile dysfunction patients who are
truly at high risk for future cardiovascular events is of
paramount importance. Indeed, erectile dysfunction is fre-
quent in patients with established CAD [6], it coexists with
silent CAD [5], and it is an independent predictor of car-
diovascular outcomes [4]. We have previously shown that
aortic stiffness (assessed by pulse wave velocity), which has
been integrated in recommendations for the assessment
and management of arterial hypertension [17], predicts
MACE in men with erectile dysfunction and no evidence
of clinical CVD [25,26]. However, it is currently unknown
whether evaluation of penile vascular function is related to
any additional risk in patients with arterial hypertension
within the context of erectile dysfunction. Our results
suggest that identification of significantly impaired penile
vascular inflow in hypertensive males is accompanied by an
unfavorable cardiovascular outcome independent of age,
BP level, and testosterone.
Our study focused on hypertensive patients. The signifi-
cant association between low PSV and cardiovascular
events has been reported in previous studies in a cohort
of erectile dysfunction patients with similar mean age and
risk factors including hypertension, diabetes, and estab-
lished CVD [15,16]. Interestingly, these studies reported
higher year event rates in hypertensive patients compared
to normotensive individuals. To the best of our knowledge,
our study is the first that enrolled a cohort of purely hyper-
tensive patients allowing us to assess the level and pre-
dictive role of PSV on future MACE, in the context of
hypertension itself and not confounded by other conditions
or risk factors.
Identification of hypertensive men with low D-PSV
would be of important prevention strategy. First, hyper-
tensive men with cavernous arterial insufficiency should
receive ‘close follow-up’ for prevention of MACE in the long
term [9]. Second, these patients are candidates of aggressive
management of hypertension and concomitant risk factors
[27–29]. Management of risk factors includes also modifi-
cation of lifestyle that, apart from the expected reduction of
cardiovascular risk through known mechanisms, has been
shown to improve penile arterial function [30].
We should acknowledge that despite the statistically
significant association between D-PSV and risk of MACE,
tests that aimed at demonstrating whether penile arterial
insufficiency improves significantly the prediction of MACE
incidence in our population beyond that offered by the FRS-
yielded inconclusive results. Despite satisfactory calibration
of the model, the addition of low D-PSV did not significantly
improve the C-statistic model and risk reclassification
beyond traditional risk factors. The nonsignificant NRI,
and low discrimination ability may be expected given
the strength of the association between traditional risk
factors and CVD and that numerous studies have shown

866 www.jhypertension.com Volume 34  Number 5  May 2016

Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.

 Peak systolic velocity and major adverse cardiovascular events
that the factors that comprise the FRS such as age, high BP
levels are associated with low D-PSV itself. However, we
applied further testing since the NRI is highly sensitive to
prespecified categories and in our population it is not
possible to translate risk estimates into the clinically
relevant standard 10-year risk estimate [31]. Accordingly,
the IDI, a newer metric that assesses whether novel markers
correctly assign individuals to higher or lower MACE prob-
abilities independent of prespecified categories improved
significantly when D-PSV was added to the reference
model. Overall, it should be noted that disparate results
are often observed when assessing model performance of
novel biomarkers and in such cases the clinical relevance
warrants further investigation [31].
Therefore, from the practical standpoint, a broad per-
formance of PCDU as a prognostic tool cannot be recom-
mended; however, for those who undergo this test for
diagnostic purposes for erectile dysfunction, the risk infor-
mation imparted are apparent and could aid further strat-
ification and management of patients, especially those
belonging in the intermediate-risk category.
Another indirect finding is worth mentioning. NRI was
numerically decreased when TOD was incorporated in
classification of risk. This implies that when TOD is used
the risk of a hypertensive patient is better assessed leaving
less room for improvement when an additional biomarker,
such as PSV, is included. Accordingly, the concept of
TOD in the assessment of risk of hypertensive patients
is reinforced.
Pathophysiological mechanisms
The finding that cavernosal arterial insufficiency is associ-
ated with increased risk of MACE, even after adjustment for
age and BP level, is not surprising. Peripheral cavernosal
arteries are end-arteries, with smaller diameter than the
coronary, having almost no ability to form collaterals [32].
As a result the consequences of arterial insufficiency and
loss of vasodilation are earlier recognizable in the penile
corpora than in heart, creating a 2–5 year window between
erectile dysfunction and cardiovascular events [33]. Further-
more, hypertension induces several functional and struc-
tural abnormalities in the cavernosal arteries [34,35].
Indeed, arterial hypertension is associated with an approxi-
mately two-fold increase in the likelihood of having an
abnormal penile blood flow [11]. The pathogenesis of
impaired penile blood inflow involves a complex interplay
between stenotic lesions [32,36], smooth muscle hypertro-
phy [34], endothelial dysfunction [37,38], subclinical inflam-
mation [39], and low-testosterone levels [40], which all from
a prominent pathogenetic milieu in hypertensive men with
erectile dysfunction. Interestingly, the significant corre-
lation between PSV and MACE risk remained significant
after adjustment for total testosterone, an important pre-
dictor of adverse cardiovascular outcomes in hypertensive
individuals [41].
Specific comments and limitations
The number of MACE was relatively small, primarily
because of the small sample size of our study and secon-
darily because of the relatively short follow-up. Although
Journal of Hypertension
Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.
our study has adequate power, we acknowledge that a
larger number of MACE would be desirable, especially for
dealing in a robust way with potential confounders.
Furthermore, although all individuals were examined by
the same urologist/radiologist, dynamic PCDU remains an
invasive examination, providing operator dependent
results, with specific rare adverse effects (i.e. priapism)
[22]. However, dynamic PCDU allows for more precise
(compared to investigation in the flaccid state) evaluation
and characterization of the cavernosal arteries because
frequent anatomical and pathological variations in the
penile arteries are more visible during dynamic investi-
gation [13].
In conclusion, low D-PSV is associated with increased
risk for a MACE in hypertensive nondiabetic men without
CVD. This predictive ability of D-PSV is independent of age,
hypertension severity, and related testosterone levels.
When only intermediate-risk patients were evaluated, the
risk reclassification beyond traditional risk factors resulted
in a marginally significant clinical NRI.
Although further research is warranted, the principal
findings of this study have important clinical implications
for those patients who undergo this test in terms of risk
stratification and further management, given the increasing
prevalence of hypertension in the erectile dysfunction
population and the higher risk for cardiovascular events
that erectile dysfunction confers by itself.
ACKNOWLEDGEMENTS
Conflicts of interest
There are no conflicts of interest.
REFERENCES
1. Hatzimouratidis K, Amar E, Eardley I, Giuliano F, Hatzichristou D,
Montorsi F, et al., European Association of Urology. Guidelines on
male sexual dysfunction: erectile dysfunction and premature ejacula-
tion. Eur Urol 2010; 57:804–814.
2. Vlachopoulos C, Jackson G, Stefanadis C, Montorsi P. Erectile dys-
function in the cardiovascular patient. Eur Heart J 2013; 34:2034–2046.
3. Gandaglia G, Briganti A, Jackson G, Kloner RA, Montorsi F, Montorsi P,
Vlachopoulos C. A systematic review of the association between
erectile dysfunction and cardiovascular disease. Eur Urol 2014;
65:968–978.
4. Vlachopoulos CV, Terentes-Printzios DG, Ioakeimidis NK, Aznaouridis
KA, Stefanadis CI. Prediction of cardiovascular events and all-cause
mortality with erectile dysfunction: a systematic review and meta-analysis
of cohort studies. Circ Cardiovasc Qual Outcomes 2013; 6:99–109.
5. Vlachopoulos C, Rokkas K, Ioakeimidis N, Aggeli C, Michaelides A,
Roussakis G, et al. Prevalence of asymptomatic coronary artery disease
in men with vasculogenic erectile dysfunction: a prospective angio-
graphic study. Eur Urol 2005; 48:996–1002.
6. Montorsi P, Ravagnani PM, Galli S, Rotatori F, Veglia F, Briganti A, et al.,
Association between erectile dysfunction and coronary artery disease.
Role of coronary clinical presentation and extent of coronary vessels
involvement: the COBRA trial. Eur Heart J 2006; 22:2632–2639.
7. Montorsi P, Montorsi F, Schulman CC. Is erectile dysfunction the ‘tip of
the iceberg’ of a systemic vascular disorder? Eur Urol 2003; 44:352–354.
8. Kearney PM, Whelton M, Reynolds K, Whelton PK, He J. Global burden
of hypertension: analysis of worldwide data. Lancet 2005; 365:217–
223.
9. Viigimaa M, Vlachopoulos C, Doumas M. Erectile dysfunction in
hypertension and cardiovascular disease. A guide for clinicians.
Springer 2014.
10. Manolis A, Doumas M. Sexual dysfunction: the ’prima ballerina’ of
hypertension-related quality-of-life complications. J Hypertens 2008;
26:2074–2084.
www.jhypertension.com 867
Ioakeimidis et al.
11. Aranda P, Ruilope LM, Calvo C, Luque M, Coca A, Gil de Miguel A.
Erectile dysfunction in essential arterial hypertension and effects of
sildenafil: results of a Spanish national study. Am J Hypertens 2004;
17:139–145.
12. Kendirci M, Trost L, Sikka SC, Hellstrom WJ. The effect of vascular risk
factors on penile vascular status in men with erectile dysfunction. J Urol
2007; 178:2516–2520.
13. Aversa A, Sarteschi LM. The role of penile color-duplex ultrasound for
the evaluation of erectile dysfunction. J Sex Med 2007; 4:1437–1447.
14. Kawanishi Y, Lee KS, Kimura K, Koizumi T, Nakatsuji H, Kojima K, et al.
Screening of ischemic heart disease with cavernous artery blood flow
in erectile dysfunctional patients. Int J Impot Res 2001; 13:100–103.
15. Corona G, Monami M, Boddi V, Cameron-Smith M, Lotti F, de Vita G, et
al., Male sexuality and cardiovascular risk. A cohort study in patients
with erectile dysfunction. J Sex Med 2010; 7:1918–1927.
16. Rastrelli G, Corona G, Monami M, Melani C, Balzi D, Sforza A, et al.
Poor response to alprostadil ICI test is associated with arteriogenic
erectile dysfunction and higher risk of major adverse cardiovascular
events. J Sex Med 2011; 8:3433–3445.
17. Mancia G, Fagard R, Narkiewicz K, Redón J, Zanchetti A, Böhm M,
Christiaens T. 2013ESH/ESC guidelines for the management of arterial
hypertension: the Task Force for the Management of Arterial Hyper-
tension of the European Society of Hypertension (ESH) and of the
European Society of Cardiology (ESC). J Hypertens 2013; 31:1925–
1938.
18. D’Agostino RB, Vasan RS, Pencina MJ, Wolf PA, Cobain M, Massaro JM,
Kannel WB. General cardiovascular risk profile for use in primary care:
the Framingham Heart Study. Circulation 200812; 117:743–753.
19. Wilson PW, D’Agostino RB, Levy D, Belanger AM, Silbershatz H,
Kannel WB. Prediction of coronary heart disease using risk factor
categories. Circulation 1998; 97:1837–1847.
20. Perk J, De Backer G, Gohlke H, Graham I, Reiner Z, Verschuren M, et al.,
European Guidelines on cardiovascular disease prevention in clinical
practice (version 2012). The Fifth Joint Task Force of the European
Society of Cardiology and Other Societies on Cardiovascular Disease
Prevention in Clinical Practice (constituted by representatives of nine
societies and by invited experts). Eur Heart J 2012; 33:1635–1701.
21. Cappelleri JC, Rosen RC. The Sexual Health Inventory for Men (SHIM):
a 5-year review of research and clinical experience. Int J Impot Res
2005; 17:307–319.
22. Porst H and Buvat J and the Standards Committee of the International
Society for Sexual Medicine. Standard Practice in Sexual Medicine 2006,
Blackwell Publishing
23. Hosmer DW, Hosmer T, Le Cessie S, Lemeshow S. A comparison of
goodness-of-fit tests for the logistic regression model. Stat Med 1997;
16:965–980.
24. Pencina MJ, D’Agostino RB Sr, D’Agostino RB Jr, Vasan RS. Evaluating
the added predictive ability of a new marker: from area under the ROC
curve to reclassification and beyond. Stat Med 2008; 27:157–172.
25. Vlachopoulos C, Ioakeimidis N, Aznaouridis K, Terentes-Printzios D,
Rokkas K, Aggelis A, et al. Prediction of cardiovascular events with
aortic stiffness in patients with erectile dysfunction. Hypertension 2014;
64:672–678.
Reviewers’ Summary Evaluations
Reviewer 1
The work by Ioakeimidis et al. is another example of the
cardiovascular system as an integrated system giving us
different signals to characterize its dysfunction. We are
learning that the clinically-guided combination of different
tests (genetics, blood pressure, blood biomarkers, pharma-
cological- and nonpharmacological test of endothelial func-
tion, imaging-tests,. . .) is giving us useful information to
understand the cardiovascular dysfunction and to identify
the subjects at higher risk to develop an event. The better
identification of subjects at high risk will make them to take
that risk more seriously and physicians to better adjust the
preventive strategy and the therapeutic objectives.
868 www.jhypertension.com
Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.
26. Vlachopoulos C, Ioakeimidis N, Stefanadis C. Biomarkers erectile
dysfunction and cardiovascular risk prediction: the latest of an evolving
concept. Asian J Androl 2015; 17:17–20.
27. Viigimaa M, Doumas M, Vlachopoulos C, Anyfanti P, Wolf J, Narkie-
wicz K, Mancia G. European Society of Hypertension Working Group
on Sexual Dysfunction. J Hypertens 2011; 29:403–407.
28. Nehra A, Jackson G, Miner M, Billups KL, Burnett AL, Buvat J, et al. The
Princeton III Consensus recommendations for the management of
erectile dysfunction and cardiovascular disease. Mayo Clin Proc
2012; 87:766–778.
29. Doumas M, Anyfanti P, Triantafyllou A. Management of erectile dysfunc-
tion: do not forget hypertension. Arch Intern Med 2012; 172:597–598.
30. La Vignera S, Condorelli R, Vicari E, D’Agata R, Calogero A. Aerobic
physical activity improves endothelial function in the middle-aged
patients with erectile dysfunction. Aging Male 2011; 14:265–272.
31. Kerr KF, McClelland RL, Brown ER, Lumley T. Evaluating the incre-
mental value of new biomarkers with integrated discrimination
improvement. Am J Epidemiol 2011; 174:364–374.
32. Ioakeimidis N, Vlachopoulos C. Microvascular function and incident
erectile dysfunction: a new perspective for their relationship. Coron
Artery Dis 2014; 25:545–547.
33. Jackson G, Montorsi P, Adams MA, Anis T, El-Sakka A, Miner M, et al.
Cardiovascular aspects of sexual medicine. J Sex Med 2010; 7 (4 Pt
2):1608–1626.
34. Toblli JE, Stella I, Inserra F, Ferder L, Zeller F, Mazza ON. Morpho-
logical changes in cavernous tissue in spontaneously hypertensive rats.
Am J Hypertens 2000; 13 (6 Pt 1):686–692.
35. Vlachopoulos C, Aznaouridis K, Ioakeimidis N, Rokkas K, Tsekoura D,
Vasiliadou C, Stefanadi E. Arterial function and intima-media thickness
in hypertensive patients with erectile dysfunction. J Hypertens 2008;
26:1829–1836.
36. Bocchio M, Scarpelli P, Necozione S, Pelliccione F, Spartera C, Fran-
cavilla F, Francavilla S. Penile duplex pharmaco-ultrasonography of
cavernous arteries in men with erectile dysfunction and generalized
atherosclerosis. Int J Androl 2006; 29:496–501.
37. Vlachopoulos C, Ioakeimidis N, Terentes-Printzios D, Stefanadis C. The
triad: erectile dysfunction: endothelial dysfunction–cardiovascular dis-
ease. Curr Pharm Des 2008; 14:3700–3711.
38. Ioakeimidis N, Vlachopoulos C, Rokkas K, Aggelis A, Terentes-Print-
zios D, Samentzas A, et al. Relationship of asymmetric dimethylarginine
with penile Doppler ultrasound parameters in men with vasculogenic
erectile dysfunction. Eur Urol 2011; 59:948–955.
39. Vlachopoulos C, Aznaouridis K, Ioakeimidis N, Rokkas K, Vasiliadou C,
Alexopoulos N, et al. Unfavourable endothelial and inflammatory state
in erectile dysfunction patients with or without coronary artery disease.
Eur Heart J 2006; 27:2640–2648.
40. Fogari R, Zoppi A, Preti P, Rinaldi A, Marasi G, Vanasia A, Mugellini A.
Sexual activity and plasma testosterone levels in hypertensive males.
Am J Hypertens 2002; 15:217–221.
41. Vlachopoulos C, Ioakeimidis N, Terentes-Printzios D, Aznaouridis K,
Rokkas K, Aggelis A, et al. Plasma total testosterone and incident
cardiovascular events in hypertensive patients. Am J Hypertens 2013;
26:373–378.
Reviewer 2
In this study Ioakeimidis et al. demonstrated that an
abnormal penile blood flow predicted major adverse
cardiovascular events (MACE) in hypertensive males
without known cardiovascular disease or diabetes, thus
adding new insight on the predictive value of erectile
dysfunction as a marker of future events. However, the
proposed penile Doppler parameter did not improve the
risk classification in this relatively small cohort of hyper-
tensive patients. Thus, the clinical utility of evaluating
erectile dysfunction by penile Doppler for assessing
cardiovascular risk beyond classical risk factors has to
be confirmed in a larger population of patients with and
without hypertension.
Volume 34  Number 5  May 2016