Review Articleiju_2426 38..47

Endothelial dysfunction and erectile dysfunction in theaging manAntonio Aversa,1 Roberto Bruzziches,1 Davide Francomano,1 Marco Natali,1 Pietro Gareri2 andGiovanni Spera1

1Department of Medical Pathophysiology, ‘Sapienza’ University of Rome, Rome and 2Operative Unit ‘Elderly Health Care’, ASPCatanzaro, Italy

Abstract: Penile erection is a vascular event that requires an intact endothelium to occur. A dysfunctional endotheliumis an early marker for the development of atherosclerotic changes and can also contribute to the occurrence of acutecardiovascular events. The pathogenesis of both endothelial and erectile dysfunction (ED) is intimately linked throughdecreased expression and activation of endothelial nitric oxide (NO) synthase, and the subsequent blunted physiologicalactions of NO naturally occurring with aging. It is now well-understood that ED is a symptom of underlying disease ratherthan a disease itself; for this reason in the near future both general practitioners, internal medicine practitioners and manyspecialists will have to interplay with sexual medicine. Aging in the man is also associated with several changes in arterialstructure and function, part of them related to the decline of circulating levels of steroids, that is, testosterone andestradiol. These changes may be responsible, in part, for the lack of efficacy of ED treatments. The recent discovery thatchronic administration of phosphodiesterase type 5 inhibitors may improve erectile and endothelial responsiveness ofmen previously non-responsive to on-demand regimes, and the knowledge that testosterone is one of the main modula-tors of the expression of penile phosphodiesterase type 5 isoenzyme, opens a new scenario in the treatment of men withED and co-morbidities. The aim of this review is to discuss the pathophysiology of endothelial dysfunction and itsrelationship with ED in the aging male, and to suggest possible strategies to improve arterial function with regard to sexualdysfunctions.

Key words: cardiovascular risk factors, endothelial dysfunction, erectile dysfunction, nitric oxide, phosphodiesterasetype 5 inhibitors.

Introduction

Erectile dysfunction (ED) is not a life-threatening disorder,but it influences the daily routine, social interactions, well-being and quality of life of the patient. It may often consti-tute the first manifestation of important systemic (inadvancing age) or relational (in younger age) pathologiesand it is considered a possible marker of clinically undiag-nosed disease, thus representing the ‘tip of the iceberg’ of asystemic vascular disorder.1 It is more common in men withdiabetes mellitus (DM) and is often associated with otherdisseminated vascular diseases, such as coronary andperipheral atherosclerosis, and an increased risk of futurecardiovascular events similar to that of men with coronaryartery disease (CAD).2,3 Aging per se represents a risk factorfor ED and pathological conditions that are commonlyencountered in the aging male, as well as chronic diseasesthat represent other causes of organic ED, and are often

treated with medications that can interfere with sexual func-tion at a central and peripheral level. Obesity, cigarettesmoking and the presence of diabetic complications (periph-eral neuropathy, vascular disease, retinopathy and nephropa-thy) show a significant relationship with the occurrence ofED in both type 1 and type 2 DM.4 The association betweenED and clinical atherosclerosis has been documented.5

There is a high incidence of cardiovascular disease (CVD) inmen with ED, and data suggest that ED may be an earlymanifestation of endothelial dysfunction (EDys) in the pres-ence or absence of cardiovascular risk factors (CRF).6 EDand CAD overlap in CRF, the pathological basis of diseaseand disease progression.7 The common underlying factor isEDys; men with penile vascular damage may have EDys inother vascular beds, as well.8 Therefore, men with ED maybe at increased risk for cardiovascular adverse events andED may be considered as a sentinel symptom in patientswith occult CVD.9 In fact, EDys measured as impairedvasodilation of the brachial artery is prognostic for cardio-vascular events, such as myocardial ischemia and stroke.10,11

The aim of this article will be to give fundamentalsregarding endothelial pathophysiology and its relevance toED in the aging male. Moreover, the role of age-related

Correspondence: Antonio Aversa MD PhD, Dipto FisiopatologiaMedica, Room 37, Viale Policlinico 155, 00161 Rome, Italy.Email: antonio.aversa@uniroma1.it

Received 1 October 2009; accepted 21 October 2009.Online publication 25 November 2009

International Journal of Urology (2010) 17, 38–47 doi: 10.1111/j.1442-2042.2009.02426.x

38 © 2009 The Japanese Urological Association

decline of testosterone (T) levels with respect to erectile andendothelial function and the beneficial effects of T supple-mentation along with phosphodiesterase type 5 inhibitor(PDE5-i) administration will be discussed.

Endothelial dysfunction

EDys has gained increasing notoriety as a key player in thepathogenesis of atherosclerosis.12 As atherosclerosis is themost common cause of vasculogenic erectile dysfunction inolder men, it is frequently considered another manifestationof vascular disease.13 The mutual risk factors shared by EDand CAD each contribute to EDys. Just as the presence ofthese risk factors overlap within patient populations, so dotheir effects on the endothelium. Several traditional CRF,such as aging, smoking, hypertension, dyslipidemia andDM, and some less-traditional risk factors, includinginflammation, hypoxia, oxidative stress and hyperhomocys-teinemia, are related to EDys.14 EDys is also a feature ofacute coronary syndromes, heart failure, reperfusion injury,renal failure, systemic inflammatory disorders and ED.15 Anumber of novel plasma markers have been associated withatherosclerosis and EDys. Erectile function is dependentupon nitric oxide (NO) production by penile endotheliumand thus ED is associated with reduced plasma NO levels.

EDys is characterized by significant modifications in thephysiological and biochemical parameters. These include:vascular stiffness, increased vascular tone, production ofinflammatory cytokines, increased permeability, suscepti-bility to invasion of immunocytes, decrease in endothelialcell growth, and dysregulation of fibrinolytic factors. Clini-cal and biochemical markers of EDys include: (i) reducedexpression and activity of endothelial nitric-oxide synthase(eNOS), reduced synthesis of NO, and increased productionof the asymmetric dimethylarginine (ADMA), a competi-tive, endogenous inhibitor of eNOS; (ii) increased produc-tion of reactive oxygen species (ROS); (iii) increasedsynthesis and release of the vasoconstrictor peptideendothelin-1 (ET-1); (iv) increased production of inflamma-tory cytokines such as interleukin-6 (IL-6), C-reactiveprotein (CRP) and tumor necrosis factor-a (TNF-a); (v)increased expression of markers of cell adhesion such asE-selectin,16 intracellular adhesion molecules (ICAM),17 andvascular cell adhesion molecules (VCAM); (vi) dysregula-tion of fibrinolytic factors such as Von Willebrand Factor(vWF), tissue plasminogen activator (tPA), and plasminogenactivator inhibitor (PAI-1); (vii) inability to regenerate fromendothelial progenitor cells (EPC); (viii) increased endothe-lial apoptosis; (ix) increased cellular permeability; and (x)increased vascular tone.18 Overall, a reduction in NO resultsin EDys.19 These mechanisms, along with a reduction inbone-marrow-derived EPC, could underpin a commonpathogenesis for both ED and EDys.20,21 In individuals withestablished ED, elevated ADMA levels have been shown to

correlate with the severity of various CRF as in patients withrenal disease,22 insulin resistance and DM23, and CAD,24,25

indicating the impact they could have on endothelialfunction.26

In addition to the biochemical markers of EDys, diagnos-tic tools of EDys are characterized by the evaluation offlow-mediated dilation (FMD) of the brachial artery.27 Thisclinical measurement of EDys is strongly linked to coronaryEDys and predicts cardiovascular events.28–30 EDys can betested in vivo using several techniques that rely principallyon measuring change in arterial diameter or flow in responseto stimuli.31 Longitudinal observations confirmed that dys-function of the endothelium of the coronary and peripheralcirculation is predictive of cardiovascular events, the sensi-tivity and specificity being greater for coronary artery EDysthan for peripheral dysfunction.32

Recently, a new operator-independent technique, that is,digital pulse amplitude by fingertip peripheral arterialtonometry (PAT) device, became very popular for diagnos-ing endothelial dysfunction in CAD patients.33 A recentstudy was carried out by investigating men with ED usingPAT and indicated that the average FMD was not different inmen with or without ED, independently from the presence orabsence of VRFs. By contrast, the augmentation index,which is an indirect measurement of arterial stiffness, washigher in men with ED compared with controls even whencontrolled for age and VRF. In that study, it was concludedthat an increased arterial stiffness but not EDys, is present inmen with vascular ED and may represent an early detectionof vascular impairment that may precede endothelial dys-function in populations at low risk for developing vascularED34 (Fig. 1).

Insulin-resistance and endothelialdysfunction

The past decade has witnessed a dramatic increase in theprevalence of obesity. Co-morbidities of obesity includeaging, type 2 DM, hypertension, and lipid abnormalities, allcontribute to CVD and are associated with EDys. Theseabnormalities frequently cluster in individuals, and the termmetabolic syndrome (MeS) is now widely used to define thiscluster. The syndrome is frequently (although not invariably)associated with insulin resistance and CVD.35 In recentyears, it has become clear that the correlation betweeninsulin resistance and EDys plays a central role in the patho-genesis of atherosclerosis. EDys is an important componentof the metabolic or insulin resistance syndrome, as demon-strated by inadequate vasodilation and/or paradoxical vaso-constriction in coronary and peripheral arteries in responseto stimuli that release NO.36 Metabolic actions of insulin topromote glucose disposal are augmented by vascular actionsof insulin in endothelium to stimulate production of thevasodilator NO. Indeed, NO-dependent increases in blood

ED3 in the elderly man

© 2009 The Japanese Urological Association 39

flow to skeletal muscle account for 25% to 40% of theincrease in glucose uptake in response to insulin stimulation.Phosphatidylinositol 3-kinase-dependent (PI3K) insulin-signaling pathways related to production of NO in endothe-lium share striking similarities with metabolic pathways thatpromote glucose uptake in skeletal muscle. Other distinctnon-metabolic branches of insulin-signaling pathways regu-late secretion of the vasoconstrictor ET-1 in endothelium.Metabolic insulin resistance is characterized by pathway-specific impairment in PI3K-dependent signaling, which inendothelium may cause imbalance between production ofNO and secretion of ET-1, leading to decreased bloodflow, which worsens insulin resistance.24 Deficiency ofendothelial-derived NO is believed to be the primary defectthat links insulin resistance and EDys. NO deficiency resultsfrom decreased synthesis and/or release, in combinationwith exaggerated consumption in tissues by high levels ofreactive oxygen (ROS) and nitrogen (RNS) species, whichare produced by cellular disturbances in glucose and lipidmetabolism. EDys contributes to impaired insulin action, byaltering the transcapillary passage of insulin to targettissues. Reduced expansion of the capillary network, withattenuation of microcirculatory blood flow to metabolicallyactive tissues, contributes to the impairment of insulin-stimulated glucose and lipid metabolism. This establishes areverberating negative feedback cycle in which progressiveEDys and disturbances in glucose and lipid metabolismdevelop secondarily to the insulin resistance.37 Hyperglyce-

mia may impair endothelial function by promoting releaseof free radicals. Vascular damage, which results from lipiddeposition and oxidative stress to the vessel wall, triggers aninflammatory reaction, and the release of chemoattractantsand cytokines that worsen the insulin resistance and EDys. Arecent study provided consistent evidence for a role ofinsulin-resistance in determining endothelial dysfunction inED patients; it was found that chronic cannabis use deter-mined early EDys, and that the hallmark of that condition inyoung cannabis users was a lower insulin sensitivity (asdetected by the homeostatic model assessment index) com-pared with the control group. It was hypothesized that pro-longed activation of the endocannabinoid system is able,somehow, to alter endothelial function and to determine anearly damage of the erectile process.38

Role for testosterone in erectile andendothelial function/dysfunction

The observation that androgen decline in aging men is asso-ciated with a reduction in the number and quality of erectionsdates long ago.39 T has a central role on endothelial health(Fig. 2); in fact its decline over the aging process may affecteither arterial reactivity or sexual function.40 Hypogonadismand ED are common disorders found in aging men present-ing to andrology clinics. Increasing evidence indicates thatboth disorders have important associations with the MeS,DM and CVD, all conditions with an increased morbidity

Age-relateddiseases

Tx

Tx

PDE5-i ±

Testosterone

Modify

risk factors

Evaluation of

arterial stiffness

Specific daily Tx

Periodic antio

xidants

CVD - Diabetes

ED3

Adequate Lifestyle + Diet + Physical activity

Erectile Dysfunction

Endothelial Dysfunction

Early Detection of vascular disease

Oxidative Stress

AGING MAN

Obesity Dyslipidemia Hyperglycemia HypertensionMetabolicSyndrome

Fig. 1 The aging male ‘pyramid’: ED3 represents the mainstay for the prevention of common diseases associated with increasedmortality in the aging male. CVD, cardiovascular disease; PDE5-i, phosphodiesterase type 5 inhibitors; Tx, pharmacologicaltreatment.

A AVERSA ET AL.

40 © 2009 The Japanese Urological Association

and mortality.41 A low T-level is positively associated withthe presence and severity of atherosclerosis and a reductionin plasma T might contribute to increased arterial stiffness,which in turn has been associated with increased cardiovas-cular risk.42 Numerous epidemiological and interventionalstudies reported a controversial relationship between T andCVD. T inversely correlates with the severity of atheroscle-rosis and has beneficial effects upon vascular reactivity,inflammatory cytokine, adhesion molecules, insulin resis-tance, serum lipids, and hemostatic factors.43 Interestingly,men with established CAD often have reduced circulating Tlevels44 that are often associated with a certain degree ofEDys independently of other VRF, suggesting a protectiverole of endogenous T on the endothelium.37 The Rotterdamstudy, a population-based cohort study, showed that lowlevels of endogenous androgens are associated withincreased likelihood of atherosclerosis in elderly men.45

Svaartberg demonstrated an inverse relationship betweentotal T levels and carotid intima-media thickness (IMT).46

This finding was not independent of body mass index (BMI).Accumulating evidence has shown an association of low T tocardiovascular mortality, morbidity in men of varying age,and CRF.47 Men have a higher rate of CVD than women. Thelikely culprits appear to be T, dihydrotestosterone (DHT),dehydroepiandrosterone (DHEAS) and their metabolites.Capaldo et al.48 showed that men with sex steroid deficiencyhad a greater IMT and that low plasma T has also beenassociated with cardiovascular risk in healthy men.49

Akishita et al.50 also found that DHEAS levels correlate with

flow-mediated dilation analysis and it was irrespective ofother confounding factors in women. A conclusive view onthe effects of androgen deficiency on vascular function canbe seen with the adverse effects of androgen deprivationtherapy (ADT) in prostate cancer patients. Whether viaorchiectomy or use of gonadotropin-releasing hormone ago-nists or antagonists and AR agonists, the net result of ADT islow circulating T/DHT with concomitant changes in bodycomposition, insulin resistance and vascular disease.51 Thereis a decrease in lean muscle mass and an increase in fat mass.A long-term study (1–8 months) comparing men undergoingADT to eugonadal men found an increase in fat mass com-pared to controls of eugonadal men.52 ADT has been impli-cated in inducing MeS.53 Overall, ADT in men with prostatecancer has been shown to increase the risk of cardiovascularevents.54 Keating reported that men undergoing ADT hada 25% increase in risk of CAD compared with non-ADTpatients.55 A large study consisting of 23 000 men undergo-ing ADT for at least 12 months had an increase of cardio-vascular morbidity by 20% compared with non-ADT menafter controlling for confounding factors.56 A recent reportfound that men receiving ADT were approximately 2.6 timesat greater risk of cardiovascular mortality than non-ADTcontrols after adjusting for confounding factors.57 Interest-ingly, a new study by D’amico et al.58 suggested that elderlymen with T-1 to T-2 localized prostate cancer should not begiven primary ADT due to reduced cancer as well as overallsurvival in these patients. Montalcini et al. showed that post-menopausal women in the lowest T tertile had the least

Fig. 2 Schematic representation ofthe molecular mechanisms underly-ing the control of endothelial functionby testosterone and phosphodi-esterase type 5 (PDE5) inhibitors.NOS, nitric-oxide synthase.

TESTOSTERONE

PDE5 inhibitors

APOPTOSIS

microparticles

REGENERA TION

Endothelialprogenitor cells

GENOMIC EFFECT

L-Arginine

O2

Sexual Stimulation

Endothelial Cell

NOS 1/2/3

Smooth muscle cell

Smooth MuscleCell Relaxation

ReductionCa2+

Ca2+

Cavernousnerve

NOTerminalNon-adrenergicNon-colinergic

Guanylatecyclasi

GTP

GMPc

PDE 5K+

5’ GMP

GMPcProtein Kinases

NON GENOMIC EFFECT

ED3 in the elderly man

© 2009 The Japanese Urological Association 41

flow-mediated dilation, which implies that not only doesestrogen deficiency play a role in CVD, but T deficiency doesas well.59

The role of androgens in determining vasodilatation hasbeen investigated recently. T may activate the endotheliumand stimulate the NO-cyclic guanosine monophosphateand/or the hyperpolarization-mediated vascular relaxationpathway and may thus add potential beneficial effectsagainst coronary artery atherosclerosis. Additionalendothelium-independent effects of T may involve inhibi-tion of the signaling mechanism of vascular smooth musclecontraction, such as intracellular concentration [Ca2+] andprotein kinase C, whereas a significant portion of the vasore-laxing effect of T appears to be endothelium independentbecause no significant difference is observed between therelaxing effect of the hormone in isolated vessels with orwithout endothelium.60 Also, inhibition of NO-synthase,prostaglandin synthase, and guanylate cyclase do not appearto affect the vasorelaxing effect of T, suggesting that theT-induced vascular relaxation may involve inhibition of themechanism of vascular smooth muscle contraction.61,62

Several studies have shown that acute administration of Tinduces a rapid relaxation in vascular tissues of differentspecies including humans,63,64 suggesting a non-genomiceffect of this hormone on vasomotion.65 Different mecha-nisms have been proposed to explain T-induced vasodilata-tion,66 but as to which are the effective mechanisms andwhich are the mediators involved with the T-induced vasore-laxation remain a matter of debate. T might induce relax-ation in human isolated corpora cavernosa strips byactivation of smooth muscle adenosine triphosphate-sensitive K(+) channels (Fig. 2). This finding suggests that T,in addition to its known endothelial action, might regulateerectile function locally by its action on the smooth muscleof the human corpus cavernosum.67 It is now accepted thatdifferent thresholds for sexual desire and erectile functionare present in humans, the former being quite higher thanthe latter.68 In humans, T deficiency determines a sequenceof molecular penile events leading to reduced capacity ofsmooth muscle and endothelial cells to relaxation as well asto increased sensitivity to contractile factors, that is,a-adrenergic agonists and deficiency of NO-induced relax-ation during sexual stimulus. Recent evidences in humanssuggest that T may directly control the expression and activ-ity of PDE5 in human corpus cavernosum so that in someselected patients, that is, total-T < 10 nmol/L and/or free-Tbelow 200 nmol/L, androgen supplementation may improvetherapeutic efficacy to PDE5-i.

Erectile dysfunction and endothelialdysfunction in the aging male

ED is predominately a disease of vascular origin. It is widelyknown that it dramatically increases in men with diabetes

mellitus, hypercholesterolemia and cardiovascular disease.It has a high prevalence and mainly affects men between 40and 70 years of age.69,70 Loss of the functional integrity ofthe endothelium and subsequent endothelial dysfunctionplays a pivotal role in the occurrence of ED. There is a closerelationship between ED, aging and EDys. ED problems dueto organic causes comprise up to 80% of cases, while vas-cular disease is the most common pathophysiology of ED.71

ED is the first clinical presentation of subclinical endothelialdysfunction disease prior to the appearance of cardiovascu-lar disease or CRF.72 In particular, it was shown that 64% ofmen who presented with myocardial infarction had ED priorto their heart problems, and 57% of men undergoing coro-nary artery bypass graft surgery had ED before the opera-tion.73 The more coronary vessels are involved, the moresevere ED is. EDys is an early abnormality of the artery;it may occur many years before vascular disease, such asatherosclerosis. As it is widely known, advanced age is arisk factor for atherosclerosis and ischemic heart disease.74

EDys occurring in the healthy elderly might have apathogenic role in the increased risk of atherosclerosisand its complications in the elderly.75 EDys and atheroscle-rosis may affect all major vascular beds with differentsizes (i.e. 1–2 mm of the penile artery, 3–4 mm of theproximal left anterior descending artery, 5–7 mm of theinternal carotid artery and 6–8 mm of the femoralartery).70,72 Symptoms of vascular disease are more frequentin elderly people and the threshold for the development ofsymptoms is reached when the lumen of the artery isoccluded by 50%.72 As there is a close relationship amongEDys, ED and aging and as EDys appears to precede notonly ED but cardiovascular complications too, the assess-ment of endothelial function in the elderly is a crucial point.In recent years venous occlusion plethysmography, arteriog-raphy and intracoronary Doppler catheter techniques duringan intra-arterial infusion were performed for this purpose.74

Recently, brachial artery flow-mediated dilation (FMD) wasused as a measure of endothelial function in humans.76 Theaim of FMD is to increase brachial artery blood flow inresponse to transient hyperemia, which is provoked byinducing post-ischemic dilation of distal vascular beds.72,77

Reactive hyperemia is reproduced by inflating a pediatricblood pressure cuff up to 200 mm Hg above systolic bloodpressure for five minutes to occlude arterial flow. Sixtyseconds after cuff deflation arterial diameter and flowvelocity measurements are performed.76 Patients withimpaired endothelial function present a lower increase inarterial diameter. FMD is a commonly used, accurate,non-invasive and reproducible method for evaluating endot-helial function. Limb vessels are suitable for assessingendothelial function in the elderly, as EDys is a diffuseprocess and peripheral vessel examination is non-invasivecompared to coronary artery assessment.75 Furthermore,FMD was shown to be nearly as sensitive as exercise

A AVERSA ET AL.

42 © 2009 The Japanese Urological Association

electrocardiography in the detection of coronary arterydisease and more specific for diagnosis than exercise elec-trocardiography. In a recent paper, endothelial function wasevaluated by FMD of the brachial artery by ultrasound in 30elderly vs 30 younger subjects.75 FMD of the elderly groupwas significantly lower than the younger group (7.9 � 3.1 vs10.8 � 1.9, respectively, P < 0.001).75 FMD and age werefound to be inversely correlated. This result is remarkable,as it was found in healthy elderly subjects. Celermajeret al.78 examined patients aged 15–72 years and foundthat aging is associated with impaired flow-mediatedendothelium-dependent vasodilation in the brachial artery;in this study elderly healthy men were reported to beaffected earlier than women.

Therefore, the final message is that advanced age is astrong predictor of EDys in the elderly; EDys may not onlyprecede ED onset, but also cardiovascular events. As it iswidely shown, it usually occurs in several age-related dis-eases, such as chronic renal failure, heart failure, hyperten-sion, diabetes mellitus, hypercholesterolemia, rheumaticdiseases and Alzheimer’s disease.79,80 For such diseases, thecontinuous use of a PDE5-i for treating ED is indicated, butto avoid the high dropout rate we recommend an adequateinitial instruction and long-term follow up, even for patientswith successful treatment.81

PDE5-i in the aging male: daily vson-demand therapy?

There are numerous conditions in the aging male that areassociated with ED, including DM, obesity, MS and CVD,that require specific treatments that may in turn determinesexual disturbances.82 The debate as to whether in such casesa daily or on-demand therapy for ED with PDE5-i should beinstituted is still open and many concerns have been raisedregarding rehabilitative effects of PDE5-i given daily in menwith ED. There have been several publications in the pasttwo years regarding daily dosing of PDE5-i for two majorindications, penile rehabilitation after prostate cancer andmanagement of lower urinary tract symptoms (LUTS). Evi-dence from basic science investigations has indicated that adaily dose of PDE5-i may improve erectile function andexert a number of beneficial tissue effects on the penis aswell as in other districts.83 Encouraging animal studies indi-cate that EDys and oxidative stress associated with insulinresistance can be reversed by daily sildenafil84 and insulinsensitivity is significantly improved in an animal modelof diet-induced obesity;85 chronic PDE5 inhibition alsoresulted in increased energy expenditure, suggesting thatimproved energy balance and weight reduction might bepartially responsible for the enhanced insulin action withoutany adverse effects on cardiac morphology or blood pres-sure measured in vivo, supporting human studies thatshowed no association between long-term use of sildenafil

and risk of ischemic events. Unfortunately, data from humanseries of routine dose PDE5-i for penile rehabilitation afterradical prostatectomy are conflicting, with the two largeststudies showing no benefit to daily dose therapy in thepost-radical prostatectomy and the general ED populations.Clinical studies revealed good safety and efficacy of sildena-fil given at bedtime in recovering spontaneous erections bothin aging men with mild-to-moderate arteriogenic etiology86

or in prostate cancer patients.87 A recent large controlledtrial comparing on-demand and nightly dosing of vardenafilon recovery of erectile function in bilateral nerve-sparingradical prostatectomy patients with ED demonstrated thatnightly dosing did not have any additive effect over that ofon-demand use.88 Similar results were reported in a con-trolled trial by Zumbè et al.,89 who compared the efficacy ofvardenafil given on-demand vs daily in men with mild-to-moderate ED. It was concluded that daily vardenafil did notproduce greater sustained effects on erectile function thanthe on-demand regime with no sustained clinical benefitsbeyond cessation of treatment above those observed withon-demand administration. By contrast, numerous studieshave shown the beneficial acute and chronic effects ofsildenafil and tadalafil on arterial function in men with ED,thus presenting them as potential candidates for therapiesaimed at reducing the burden of EDys in selected categoriesof aging men.90

Vascular damage, insulin resistance and EDys have beenrecently implicated in the pathogenesis of benign prostatehyperplasia (BPH), thus suggesting the hypothesis of anincreased sympathetic nerve activity in men with BPH.91

Evidences show that diabetic men are at higher risk for bothED and BPH,92 and vascular damage being considered as themainstay of both;92 hyperplasia of the stromal and glandularprostate compartments might be induced by stromal growthsecondary to hypoxia, which in turn results from reducedblood flow patterns at the Duplex ultrasound.93 It appearsclear that EDys and the consequent adrenergic overtonecan be considered as the common denominator linkingatherosclerosis, cardiovascular disease and ED-BPH. To thispurpose, daily PDE5-i use in clinical trials has been reportedto be effective and safe in improving LUTS with no changein urinary flow rates in men with BPH/LUTS, suggesting anewer utilization to improve the quality of life for aging menwith symptoms secondary to BPH.94

Conclusions

ED is associated with the same underlying risk factors asvascular disease and includes hypertension, DM, hyperlipi-demia, smoking, obesity, and aging. Some evidence showsthat ED can be greatly improved not only with PDE5-i butalso by treating the risk factors directly. These include ces-sation of smoking, correction of hyperlipidemia, and ame-lioration of obesity through weight loss, all of which result

ED3 in the elderly man

© 2009 The Japanese Urological Association 43

in amelioration of endothelial health (Fig. 1). Generally, inthese patients the co-administration of T when a deficiencysyndrome is present, does not produce detrimental orobstructive effects on the prostate, but must be used withcaution in the long term. ED and LUTS frequently coexist inmen of advancing age but further studies are required toelicit the exact mechanism of action in LUTS. There appearsto be an emerging role for PDE5-i as a treatment for bothconditions. Finally, it is important to remember that prostatecancer is the most common cause of death in men withadvancing age; radical prostatectomy is often associatedwith ED and no rehabilitative oral treatment has demon-strated proven efficacy. When androgen deprivation therapyis instituted, it is important to remember that it is able toinduce rapid metabolic changes, such as increased insulinresistance and visceral adipose tissue accumulation, whichare associated with increased cardiovascular mortality.

Acknowledgment

Dr Aversa and Dr Gareri wish to thank Professor MariannaMorrone for her ongoing assistance with the Englishlanguage.

References

1 Aytac IA, McKinlay JB, Krane RJ. The likely worldwideincrease in erectile dysfunction between 1995 and 2025 andsome possible policy consequences. BJU Int. 1999; 84:450–6.

2 Kannel WB, McGee DL. Diabetes and cardiovasculardisease: the Framingham study. JAMA 1979; 241: 2035–8.

3 Haffner SM, Lehto S, Rönnemaa T, Pyörälä K, Laakso M.Mortality from coronary heart disease in subjects with type2 diabetes and in non diabetic subjects with and withoutprior myocardial infarction. N. Engl. J. Med. 1998; 339:229–34.

4 Vitale C, Mercuro G, Cornoldi A, Fini M, Volterrani M,Rosano GM. Metformin improves endothelial function inpatients with metabolic syndrome. J. Intern. Med. 2005;258: 250–6.

5 Guay AT. ED2: erectile dysfunction = Endothelialdysfunction. Endocrinol. Metab. Clin. North Am. 2007; 36:453–63.

6 Gazzaruso C, Solerte SB, Pujia A et al. Erectiledysfunction as a predictor of cardiovascular events anddeath in diabetic patients with angiographically provenasymptomatic coronary artery disease. A potentialprotective role for statins and 5-phosphodiesteraseinhibitors. J. Am. Coll. Cardiol. 2008; 51: 2040–4.

7 Akilli H, Gok H, Soylu A, Kayrak M. Severity of coronaryartery disease and symptoms of erectile dysfunction inmales with a positive exercise treadmill test. Int. J. Urol.2007; 14: 733–7.

8 Kaiser DR, Billups K, Mason C, Wetterling R, LundbergJL, Bank AJ. Impaired brachial artery

endothelium-dependent and – independent vasodilation inmen with erectile dysfunction and no other clinical CVD.J. Am. Coll. Cardiol. 2004; 43: 179–84.

9 Thompsom IM, Tangen CM, Goodman PJ, Probstfield JL,Moinpour CM, Coltman CA. Erectile dysfunction andsubsequent CVD. JAMA 2005; 294: 2996–3002.

10 Schachinger V, Britten MB, Zeiher AM. Prognostic impactof coronary vasodilator dysfunction on adverse long-termoutcome of coronary heart disease. Circulation 2000; 101:1899–906.

11 Suwaidi JA, Hamasaki S, Higano ST, Nishimura RA,Holmes DR Jr, Lerman A. A long-term follow-up ofpatients with mild coronary artery disease and endothelialdysfunction. Circulation 2000; 101: 948–54.

12 Ross R. The pathogenesis of atherosclerosis: a perspectivefor the 1990s. Nature 1993; 362: 801–9.

13 Russell S, Nehra A. The physiology of erectile dysfunction.Herz 2003; 28: 277–83.

14 Brunner H, Cockcroft JR, Deanfield J et al. Endothelialfunction and dysfunction. Part II: association withcardiovascular risk factors and diseases. A statement by theWorking Group on Endothelins and Endothelial Factors ofthe European Society of Hypertension. J. Hypertens. 2005;23: 233–46.

15 Bonetti PO, Lerman LO, Lerman A. Endothelialdysfunction: a marker of atherosclerotic risk. Arterioscler.Thromb. Vasc. Biol. 2003; 23: 168–75.

16 Blankenberg S, Rupprecht HJ, Bickel C et al. Circulatingcell adhesion molecules and death in patients with coronaryartery disease. Circulation 2001; 104: 1336–42.

17 Witte DR, Broekmans WM, Kardinaal AF,Klopping-Ketelaars IA, van Poppel G, Bots ML. Solubleintercellular adhesion molecule 1 and flow-mediateddilatation are related to the estimated risk of coronary heartdisease independently from each other. Atherosclerosis2003; 170: 147–53.

18 Traish AM, Abu-Zahra H, Guay AT. The brain, the penisand steroid hormones: clinical correlates with endothelialdysfunction. Curr. Pharm. Des. 2008; 14: 3723–36.

19 Browne D, Meeking D, Shaw K. Endothelial dysfunctionand pre-symptomatic atherosclerosis in type 1 diabetes –pathogenesis and identification. Br. J. Diabetes 2003; 3:27–34.

20 Elesber AA, Solomon H, Lennon RJ et al. Coronaryendothelial dysfunction is associated with erectiledysfunction and elevated asymmetric dimethylarginine inpatients with early atherosclerosis. Eur. Heart. 2006; 27:824–31.

21 Baumhakel M, Werner N, Bohm M, Nickenig G.Circulating endothelial progenitor cells correlate witherectile function in patients with coronary heart disease.Eur. Heart. 2006; 27: 2184–8.

22 Kielstein JT, Frolich JC, Haller H, Fliser D. ADMA(asymmetric dimethylarginine): an atherosclerotic diseasemediating agent in patients with renal disease? Nephrol.Dial. Transplant. 2001; 16: 1742–5.

23 Schiel R, Franke S, Busch M et al. Effect of smoking onrisk factors for cardiovascular disease in patients with

A AVERSA ET AL.

44 © 2009 The Japanese Urological Association

diabetes mellitus and renal insufficiency. Eur. J. Med. Res.2003; 8: 283–91.

24 Kim JA, Montagnani M, Koh KK, Quon MJ. Reciprocalrelationships between insulin resistance and endothelialdysfunction: molecular and pathophysiologicalmechanisms. Circulation 2006; 113: 1888–904.

25 Lu TM, Ding YA, Lin SJ, Lee WS, Tai HC. Plasma levelsof asymmetrical dimethylarginine and adversecardiovascular events after percutaneous coronaryintervention. Eur. Heart J. 2003; 24: 1912–19.

26 Wierzbicki AS, Solomon H, Lumb PJ, Lyttle K,Lambert-Hammill M, Jackson G. Asymmetricaldimethylarginine levels correlate with cardiovascular riskfactors in patients with erectile dysfunction. Atherosclerosis2006; 185: 421–5.

27 Corretti MC, Anderson TJ, Benjamin EJ, Celermajer D,Charbonneau F, Creager MA. Guidelines for the ultrasoundassessment of endothelial-dependent flow-mediatedvasodilation of the brachial artery: a report of theInternational Brachial Artery Reactivity Task Force. J. Am.Coll. Cardiol. 2002; 39: 257–65.

28 Anderson TJ, Uehata A, Gerhard MD, Meredith IT, KnabS, Delagrange D. Close relation of endothelial function inthe human coronary and peripheral circulations. J. Am.Coll. Cardiol. 1995; 26: 1235–41.

29 Wu WC, Sharma SC, Choudhary G, Coulter L, Coccio E,Eaton CB. Flow-mediated vasodilation predicts thepresence and extent of coronary artery disease assessed bystress thallium imaging. J. Nucl. Cardiol. 2005; 12: 538–44.

30 Yavuz BB, Yavuz B, Sener DD et al. Advanced age isassociated with endothelial dysfunction in healthy elderlysubjects. Gerontology 2008; 54: 153–6.

31 Farouque HMO, Meredith IT. The assessment of endothelialfunction in humans. Coron. Artery Dis. 2001; 12: 445–54.

32 Vita JA, Keaney JF Jr. Endothelial function: a barometerfor cardiovascular risk? Circulation 2002; 106: 640–2.

33 Hamburg NM, Keyes MJ, Larson MG et al. Cross-sectionalrelations of digital vascular function to cardiovascular riskfactors in the Framingham Heart Study. Circulation 2008;117: 2467–74.

34 Aversa A, Francomano D, Bruzziches R, Pili M, Natali M,Spera G. The application of digital pulse amplitude in thediagnostic investigation of patients with erectiledysfunction. Andrologia (in press).

35 Fonseca V, Jawa A. Endothelial and erectile dysfunction,diabetes mellitus, and the metabolic syndrome: commonpathways and treatments? Am. J. Cardiol. 2005; 96:13M–18M.

36 Cersosimo E, DeFronzo RA. Insulin resistance andendothelial dysfunction: the road map to cardiovasculardiseases. Diabetes Metab. Res. Rev. 2006; 22: 423–36.

37 Akishita M, Hashimoto M, Ohike Y. Low testosterone levelis an independent determinant of EDys in men. Hypertens.Res. 2007; 30: 1029–34.

38 Aversa A, Rossi F, Francomano D et al. Early endothelialdysfunction as a marker of vasculogenic erectiledysfunction in young habitual cannabis users. Int. J. Impot.Res. 2008; 20: 566–73.

39 Gooren LJG. The age-related decline of androgen levels inmen: clinically significant? Br. J. Urol. 1996; 78: 763.

40 Allan CA, McLachlan RI. Age-related changes intestosterone and the role of replacement therapy in oldermen. Clin. Endocrinol. (Oxf.) 2004; 60: 653–70.

41 Jones TH. Testosterone associations with erectiledysfunction, diabetes, and the metabolic syndrome. Eur.Urol. 2007; 6: 847–57.

42 Hougaku H, Fleg JL, Najjar SS, Lakatta EG, Harman SM,Blackman MR, Metter EJ. Relationship between androgenichormones and arterial stiffness, based on longitudinalhormone measurements. Am. J. Physiol. Endocrinol. Metab.2006; 290: E234–42.

43 Fukui M, Kitagawa Y, Ose H, Hasegawa G, Yoshikawa T,Nakamura N. Role of endogenous androgen against insulinresistance and atherosclerosis in men with type 2 diabetes.Curr. Diabetes Rev. 2007; 3: 25–31.

44 Rosano GM, Sheiban I, Massaro R et al. Low testosteronelevels are associated with coronary artery disease inmale patients with angina. Int. J. Impot. Res. 2007; 19:176–82.

45 Hak AE, Witteman JC, de Jong FH. Low levels ofendogenous androgens increase the risk of atherosclerosisin elderly men: the Rotterdam study. J. Clin. Endocrinol.Metab. 2002; 87: 3632–9.

46 Svartberg J, von Mühlen D, Mathiesen E, Joakimsen O,Bønaa KH, Stensland-Bugge E. Low testosterone levels areassociated with carotid atherosclerosis in men. J. Intern.Med. 2006; 259: 576–82.

47 Laughlin GA, Barrett-Connor E, Bergstrom J. Low serumtestosterone and mortality in older men. J. Clin.Endocrinol. Metab. 2008; 93: 68–75.

48 Capaldo B, Patti L, Oliviero U et al. Increased arterialintima-media thickness in childhood-onset growthhormone deficiency. J. Clin. Endocrinol. Metab. 1997; 82:1378–81.

49 Simon D, Charles MA, Nahoul K et al. Associationbetween plasma total testosterone and cardiovascular riskfactors in healthy adult men: the Telecom Study. J. Clin.Endocrinol. Metab. 1997; 82: 682–5.

50 Akishita M, Hashimoto M, Ohike Y et al. Association ofplasma dehydroepiandrosterone-sulfate levels withendothelial function in postmenopausal women withcoronary risk factors. Hypertens. Res. 2008; 31: 69–74.

51 Chodak GW, Keane T, Klotz L. Critical evaluation ofhormonal therapy for carcinoma of the prostate. Urology2002; 60: 201–8.

52 Basaria S, Lieb J II, Tang A et al. Long-term effects ofandrogen deprivation therapy in prostate cancer patients.Clin. Endocrinol. 2002; 56: 779–86.

53 Shahani S, Braga-Basaria M, Basaria S. AndrogenDeprivation therapy in prostate cancer & metabolic risk foratherosclerosis. J. Clin. Endocrinol. Metab. 2008; 93:2042–9.

54 D’Amico AV, Denham JW, Crook J et al. Influence ofandrogen suppression therapy for prostate cancer on thefrequency and timing of fatal myocardial infarctions.J. Clin. Oncol. 2007; 25: 2420–5.

ED3 in the elderly man

© 2009 The Japanese Urological Association 45

55 Keating NL, O’Malley AJ, Smith MR. Diabetes andcardiovascular disease during androgen deprivation therapyfor prostate cancer. J. Clin. Oncol. 2006; 24: 4448–56.

56 Saigal CS, Gore JL, Krupski TL, Hanley J, Schonlau M,Litwin MS. Androgen deprivation therapy increasescardiovascular morbidity in men with prostate cancer.Cancer 2007; 110: 1493–500.

57 Tsai HK, D’Amico AV, Sadetsky N, Chen MH, Carroll PR.Androgen deprivation therapy for localized prostate cancerand the risk of cardiovascular mortality. J. Natl. CancerInst. 2007; 99: 1516–24.

58 D’Amico AV, Chen MH, Renshaw AA, Loffredo M,Kantoff PW. Androgen suppression and radiation vsradiation alone for prostate cancer: a randomized trial.JAMA 2008; 299: 289–95.

59 Montalcini T, Gorgone G, Gazzaruso C, Sesti G, PerticoneF, Pujia A. Endogenous testosterone and endothelialfunction in postmenopausal women. Coron. Artery Dis.2007; 18: 9–13.

60 Crews JK, Khalil RA. Antagonistic effects of 17beta-estradiol, progesterone and testosterone on Ca2+ entrymechanism of coronary vasoconstriction. Arterioscler.Thromb. Vasc. Biol. 1999; 19: 1034–40.

61 Murphy JG, Khalil RA. Decreased [Ca2]i during inhibitionof coronary smooth muscle contraction by 17b-estradiol,progesterone, and testosterone. J. Pharmacol. Exp. Ther.1999; 291: 44–52.

62 Yue P, Chatterjee K, Beale C, Poole-Wilson PA, Collins P.Testosterone relaxes rabbit coronary arteries and aorta.Circulation 1995; 91: 1154–60.

63 Costarella CE, Stallone JN, Rutecki GW, Whittier FC.Testosterone causes direct relaxation of rat thoracic aorta.J. Pharm. Exp. Ther. 1996; 277: 349–54.

64 Perusquia M, Hernandez R, Morales MA, Campos MG,Villalon MG. Role of endothelium in the vasodilating effectof progestins and androgens on the rat thoracic aorta. Gen.Pharmacol. 1996; 27: 181–5.

65 English KM, Jones RD, Jones TH, Morice AH, ChannerKS. Testosterone acts as a coronary vasodilator bycalcium antagonist action. J. Endocrinol. Invest. 2002; 25:455–8.

66 Tep-areenan P, Kendall DA, Randall MD.Testosterone-induced vasorelaxation in the rat mesentericarterial bed is mediated predominantly via potassiumchannels. Br. J. Pharmacol. 2002; 135: 735–40.

67 Yildiz O, Seyrek M, Irkilata HC, Yildirim I, Tahmaz L,Dayanc M. Testosterone might cause relaxation of humancorpus cavernosum by potassium channel opening action.Urology 2009; 74: 229–32.

68 Isidori AM, Giannetta E, Gianfrilli D et al. Effects oftestosterone on sexual function in men: results of ameta-analysis. Clin. Endocrinol. (Oxf.). 2005; 63:381–94.

69 Lue TF. Erectile dysfunction. N. Engl. J. Med. 2000; 342:1802–13.

70 Bortolotti A, Parazzini F, Colli E. The epidemiology oferectile dysfunction and its risk factors. Int. J. Androl.1997; 20: 323–34.

71 Chiurlia E, D’Amico R, Ratti C, Granata AR, RomagnoliR, Modena MG. Subclinical coronary artery atherosclerosisin patients with erectile dysfunction. J. Am. Coll. Cardiol.2005; 46: 1503–6.

72 Lojanapiwat B, Weerusawin T, Kuanprasert S. Erectiledysfunction as a sentinel marker of endothelial dysfunctiondisease. Singapore Med. J. 2009; 50: 698–701.

73 Montorsi P, Montorsi F, Schulman CC. Is erectiledysfunction the ‘Tip of the iceberg’ of a systemic vasculardisorder? Eur. Urol. 2003; 44: 352–4.

74 Lakatta EG, Yin FCP. Myocardial aging: functionalalterations and related cellular mechanisms. Am. J. Physiol.1982; 242: H927–41.

75 Singh N, Prasad S, Singer DRJ, Macallister RJ. Ageing isassociated with impairment of nitric oxide and prostanoiddilator pathways in the human forearm. Clin. Sci. 2002;102: 595–600.

76 Yavuz BB, Yavuz B, Sener DD, Cankurtaran M, Halil M,Ulger Z. Advanced age is associated with endothelialdysfunction in healthy elderly subjects. Gerontology 2008;54: 153–6.

77 Faulx MD, Wright AT, Hoit BD. Detection of endothelialdysfunction with brachial artery ultrasound scanning. Am.Heart J. 2003; 145: 943–51.

78 Celermajer DS, Sorensen KE, Spiegelhalter DJ,Georgakopoulos D, Robinson J, Deanfield JE. Aging isassociated with endothelial dysfunction in elderly healthymen years before the age-related decline in women. J. Am.Coll. Cardiol. 1994; 24: 471–16.

79 Fernandez-Cruz A, Martin P, Fernandez L et al. Plasmaendothelin is increased in young essential hypertensives butnot in elderly essential or diabetic hypertensives.J. Hypertens. 1993; 11: S146–7.

80 Dede DS, Yavuz B, Yavuz BB et al. Assessment ofendothelial function in Alzheimer’s disease: is Alzheimer’sdisease a vascular disease? J. Am. Geriatr. Soc. 2007; 55:1613–17.

81 Sato Y, Tanda H, Kato S, Onishi S, Nitta T, Koroku M.How long do patients with erectile dysfunction continue touse sildenafil citrate? Dropout rate from treatment course asoutcome in real life. Int. J. Urol. 2007; 14: 339–42.

82 Kimoto Y, Nagao K, Sasaki H et al. JSSM Guidelines forerectile dysfunction. Int. J. Urol. 2008; 15: 564–76.

83 Masuda H. Significance of nitric oxide and its modulationmechanisms by endogenous nitric oxide synthase inhibitorsand arginase in the micturition disorders and erectiledysfunction. Int. J. Urol. 2008; 15: 128–34.

84 Behr-Roussel D, Oudot A, Caisey S et al. Daily treatmentwith sildenafil reverses endothelial dysfunction andoxidative stress in an animal model of insulin resistance.Eur. Urol. 2008; 53: 1272–80.

85 Ayala JE, Bracy DP, Julien BM, Rottman JN, Fueger PT,Wasserman DH. Chronic treatment with sildenafil improvesenergy balance and insulin action in high fat-fed consciousmice. Diabetes 2007; 56: 1025–33.

86 Sommer F, Klotz T, Engelmann U. Improved spontaneouserectile function in men with mild-to-moderate arteriogenicerectile dysfunction treated with a nightly dose of sildenafil

A AVERSA ET AL.

46 © 2009 The Japanese Urological Association

for one year: a randomized trial. Asian J. Androl. 2007; 9:134–41.

87 Padma-Nathan H, McCullough AR, Levine LA et al.Randomized , double-blind , placebo-controlled study ofpostoperative nightly sildenafil citrate for the preventionof erectile dysfunction after bilateral nerve-sparing radicalprostatectomy. Int. J. Impot. Res. 2008; 20: 479–86.

88 Montorsi F, Brock G, Lee J et al. Effect of nightly versuson-demand vardenafil on recovery of erectile function inmen following bilateral nerve-sparing radicalprostatectomy. Eur. Urol. 2008; 54: 924–31.

89 Zumbé J, Porst H, Sommer F, Grohmann W, Beneke M,Ulbrich E. Comparable efficacy of once-daily versuson-demand vardenafil in men with mild-to-moderateerectile dysfunction: findings of the RESTORE Study.Eur. Urol. 2008; 54: 204–12.

90 Bruzziches R, Greco EA, Pili M, Francomano D, Spera G,Aversa A. Redefining the role of long-acting

phosphodiesterase inhibitor tadalafil in the treatment ofdiabetic erectile dysfunction. Curr. Diabetes Rev. 2008; 4:24–30.

91 Hammarsten J, Högstedt B. Hyperinsulinaemia as a riskfactor for developing benign prostatic hyperplasia. Eur.Urol. 2001; 39: 151–8.

92 Berger AP, Deibl M, Leonhartsberger N et al. Vasculardamage as a risk factor for benign prostatic hyperplasia anderectile dysfunction. BJU Int. 2005; 96: 1073–8.

93 McVary K. Lower urinary tract symptoms and sexualdysfunction: epidemiology and pathophysiology. BJU Int.2006; 97: 23–8.

94 Shindel AW. 2009 update on phosphodiesterase type 5inhibitor therapy part 1: recent studies on routine dosing forpenile rehabilitation, lower urinary tract symptoms, andother indications. J. Sex. Med. 2009; 6: 1794–808.

ED3 in the elderly man

© 2009 The Japanese Urological Association 47