Do cigarette smokers with erectile dysfunction benefit from stopping?: a prospective study


Gholamreza Pourmand, Department of Urology, Sina Hospital, Imam Khomeini Street, Tehran, Iran.



To assess whether stopping smoking can improve erectile dysfunction (ED) in smokers, as cigarette smoking is a known risk factor for ED.


Smokers who requested nicotine replacement therapy (NRT) and complained of ED were first evaluated for hypertension, dyslipidaemia, diabetes, psychiatric disorders and drug history. The grade of ED in smokers with none of these risk factors was then determined using the five-item version of the International Index of Erectile Function (IIEF-5) before NRT, and the grading repeated after 1 year of follow-up. The correlation between the exposure to smoking (pack-years) and severity of ED was assessed before the follow-up. The ED status between patients who stopped smoking after NRT and those who continued during the follow-up was then compared before and after the follow-up.


The severity of ED correlated significantly with the level of exposure to smoking. Age and ED status before the follow-up were not significantly different between 118 patients who stopped (ex-smokers) and 163 who continued smoking (current smokers). After 1 year the ED status improved in ≥ 25% of ex-smokers but in none of the current smokers; 2.5% of ex-smokers and 6.8% of current smokers had a deterioration in ED. Ex-smokers had a significantly better ED status after the follow-up (P = 0.009). Among ex-smokers, patients with advanced ED and those who were older had less improvement.


There is a strong association between the intensity of cigarette smoking and degree of ED. Stopping cigarette smoking can improve ED in a considerable proportion of smokers. Age and the severity of ED before stopping are inversely related to the chance of improvement.


erectile dysfunction


nitric oxide


nicotine replacement therapy


fasting blood sugar.


Erectile dysfunction (ED) is defined as the recurrent or persistent inability to achieve and/or maintain an erection adequate for satisfactory intercourse [1]. Penile erection depends largely on an adequate inflow of blood to the erectile tissue, and requires coordinated arterial endothelium-dependent vasodilatation and sinusoidal endothelium-dependent corporal smooth muscle relaxation [2]. Nitric oxide (NO) has been identified as the main vasodilator mediating penile erection [3–5]. Therefore, any factors interfering with the arterial inflow of blood to the corpora cavernosa or the synthesis/release of NO are prime factors involved in the pathophysiology of ED.

Numerous studies have shown dysfunction in the endothelium and of vasodilatation in smokers [6–8]. Compounds in cigarette smoke, e.g. free radicals, aromatic compounds and superoxide anions, can impair dilatation of the penile arteries and arterioles by impeding NO synthesis and degradation [9]. Cigarette smoking is also cited as an independent risk factor for atherosclerosis in the internal, pudendal and common penile arteries of young men with ED [10]. Therefore, it is not surprising that smoking, independently or related to other risk factors, is reportedly associated with ED in many studies [10–14]. Even the number of cigarettes smoked was related to the degree of abnormalities of nocturnal tumescence in one study [14]. Although the risk of some other smoking-induced vascular disorders, e.g. coronary artery disease, has been shown to decrease substantially in the initial 2–3 years after stopping smoking [15,16], no such period has been established for ED, probably because of the unique adverse effects of smoking on the physiology of erection [13].

Stopping smoking by middle-aged men did not significantly decrease the risk of ED in one study [17] but did so in another [18]. In the present study we prospectively evaluated the ED status in smokers with ED before stopping smoking and 1 year afterward, comparing the results with a matched group of smokers with ED who continued to smoke during the same period. We focused on the influence of age and severity of ED before, and the reversibility after, stopping smoking.


Patients were recruited from centres offering nicotine replacement therapy (NRT) and psychological counselling to smokers; 2837 smokers (aged 30–60 years) were visited between December 2000 and November 2002 in 10 centres, in which smokers received 2 or 4 mg of nicotine (gum) in different regimens for 1–2 months. Smokers were asked about any ‘recurrent inability to achieve and/or maintain a penile erection adequate for satisfactory sexual performance’[1]. Patients who reported having such a problem, that the problem had begun ≥ 5 years after starting to smoke cigarettes, and who agreed to participate in the study, were evaluated for their medical and drug history. Patients who had history of diabetes, hypertension, dyslipidaemia, cardiovascular disease, peripheral vascular disease, renal failure, psychiatric disorders, trauma or surgery in the pelvic region were excluded.

Special attention was given to the psychiatric history and psychiatric aspects of the sexual history of patients, to avoid missing psychogenic ED. Those who had taken medications which were presumed to involve ED were also excluded. The remaining patients provided a blood sample for evaluating fasting blood sugar (FBS), total cholesterol, low-density lipoprotein, creatinine, blood urea nitrogen and liver function tests. Blood pressure was measured in three consecutive recordings with the patient at rest. Hypertension was defined as a blood pressure of >140/90 mmHg, dyslipidaemia as a total cholesterol level of >6.21 mmol/L or low-density lipoprotein level of >4.14 mmol/L, and diabetes as a FBS of >7.8 mmol/L (checked twice). Patients with any of these risk factors were also excluded. At this point there were 346 smokers with ED who had no risk factor except smoking. The patients then received NRT for 1–2 months. The patients were followed for 1 year after stopping NRT; there were then two groups, those who had successfully stopped smoking after NRT (ex-smokers) and those who had not and had continued smoking (current smokers). In ex-smokers we excluded those who received NRT during the follow-up, and in both groups patients who received medication for ED during the follow-up were also excluded.

Patients were completely informed about the methods and aims of the study, and informed consent was obtained before and after the follow-up. Patients who fulfilled all the inclusion criteria had their erectile function assessed using the five-item version of International Index of Erectile Function questionnaire [19], before and after the follow-up. According to the score there are five grades: normal function (>21), mild dysfunction (17–21), mild to moderate dysfunction (12–16), moderate dysfunction (8–11) and severe dysfunction (≤7). The level of smoking exposure was calculated in patients and three levels defined, i.e. <10, 10–20 and >20 pack-years (one pack-year is defined as smoking 20 cigarettes per day for 1 year). We assessed whether there was a significant correlation between pack-years and the severity of ED (dose-dependent correlation). Age and grade of ED were compared between the groups (ex- and current smokers) before NRT; this comparison of ED status of the two groups was repeated after 1 year to determine the effects of stopping smoking. The rate of change in each ED grade was recorded after the follow-up to assess the reversibility of erectile function in advanced cases of ED. Ex-smokers in each ED grade were categorized in three age subgroups, i.e. 30–39, 40–49 and 50–60 years, and the beneficial effect of stopping smoking reported in older and younger patients.


The possible correlation between the exposure to smoking and grade of ED was assessed using Spearman's correlation coefficient; the Mann–Whitney U-test was used to compare ED grade between ex- and current smokers before and after the follow-up, with the age of the two groups compared using and independent t-test; in all tests, P < 0.05 was taken to indicate significance.


Of 2837 smokers (aged 30–60 years), 637 (22.5%) reported having ED. After evaluating them for the exclusion criteria, 346 patients had no risk factor except smoking; 40 current smokers and 25 ex-smokers were excluded after the follow-up because they were unwilling to continue participation or were taking medication for ED. Finally there were 118 ex-smokers (mean age 43.6, sd 7.58 years) and 163 current smokers (45.7, sd 8.4 years); the age was not significantly different between the groups (P = 0.12). There was a significant correlation between exposure (pack-years of smoking) and ED status in all patients (Spearman's correlation coefficient, 0.533). The initial grade of ED in both groups in shown in Table 1; there were no significant difference in ED status between the groups (Mann-Whitney, P = 0.349). The ED status after the follow-up is also shown in Table 1; the ED status improved by at least 1 grade in 30 (25%) ex-smokers (49% mild, 35% mild-to-moderate, 17% moderate, none severe) but in none of the current smokers (Table 1); three (2.5%) ex-smokers and 11 (7%) current smokers had a deterioration. Ex-smokers had a significantly better ED status after the follow-up (Mann-Whitney, P = 0.09). The proportion ‘improved’ in each age category of patients with a similar ED status is shown in Table 2.

Table 1.  ED status of patients in both groups before and after the follow-up, the change in ED in each grade after the follow-up
Variable n/N (%)Ex-smokers Current smokers 
  • *

    The denominators indicate the number of patients in each grade at baseline (before follow-up); ND, not defined.

ED gradeBeforeAfterBeforeAfter
Normal 0/11824/118 (20.4 0/163 0/163
Mild35/118 (29.6)21/118 (17.8)41/163 (25.2)39/163 (23.9)
Mild-to-moderate17/118 (14.4)13/118 (11)27/163 (16.5)25/163 (15.4)
Moderate42/118 (35.6)34/118 (28.8)63/163 (38.3)62/163 (38)
Severe24/118 (20.4)26/118 (22)32/163 (19.6)37/163 (22.7)
Mild17/35 (49) 0/35 0/41 2/41 (5)
Mild to moderate 6/17 1/17 0/27 4/27 (15)
Moderate 7/42 (17) 2/42 (5) 0/63 5/63 (8)
Severe 0/24ND 0/32ND
Total30/118 (25.4) 3/118 (2.5) 0/163 11/163 (6.8)
Table 2.  The proportion improved in each age category of ex-smokers
ED gradeAge group, years
  1. The denominators indicate the number of patients in each grade at baseline.

Mild-to-moderate 4/82/60/3
Moderate 5/192/160/7
Severe 0/60/80/10
Total19/50 (38)9/34 (27)2/26 (8)


The exact pathophysiology of ED in smokers is yet to be clarified [13], but impairment of the biosynthesis and degradation of NO [9], endothelial dysfunction and decreased endothelium-dependent vasodilatation [6–8], thickening of the intima-media of the arterial vasculature [7], increased blood coagulability [20], reduced serum antioxidants [21], and alteration of glucose and lipid metabolism [22,23] provide many mechanisms for smoking as a major risk factor for a ‘complex vascular event’ like penile erection, and its proven role in various vascular diseases, e.g. coronary artery [15,16], cerebrovascular [24] and peripheral artery disease [25]. Smoking is also an independent risk factor equivalent to hypertension and hypercholesterolaemia.

Although in several reports the relationship of ED with smoking is not confirmed, a close association is likely [13]. Feldman et al.[11], in a follow-up of the Massachusetts Male Aging Study, reported that after adjusting for age and other covariates, smokers at baseline were more likely than nonsmokers to have moderate or complete ED (24% vs 14%). Mannino et al.[12], in a cross-sectional survey of 4462 USA army veterans, found that after controlling for multiple confounders, the prevalence of ED was significantly higher in current smokers than veterans who had never smoked. Other case series and retrospective studies showed additional similar associations [10,13,14].

A biological gradient or dose–response is a key component for establishing a causal explanation when addressing any epidemiological question [26]. In the present study the grade of ED in all patients before NRT was strongly associated with the exposure to cigarette smoking. This result suggests that greater exposure to the detrimental effects of cigarette smoke and the resulting greater vascular damage is associated with advanced ED; this agrees with the results of Hirshkowitz et al.[14], who reported that the number of cigarettes smoked was related to the degree of abnormality in nocturnal penile tumescence, and with Rosen et al.[10], who identified a significant dose–response relationship of smoking and cavernosal arterial occlusive disease. Such a dose-dependent relation is also reported in cigarette smoking and dysfunctional endothelium-dependent vasodilatation [6], intima-media thickening of the coronary artery [7], myocardial infarction [27], ischaemic stroke [24], plasma fibrinogen concentration [20] and intermittent claudication [25].

The reversibility of the relative risk by removing the exposure is a key component in any attempt to establish causality. For this to be applicable the disease must not be so advanced that reversibility is unlikely [13]. Guay et al.[17], in a study of 10 heavy smokers, reported that 24 h of not smoking produced a significant improvement in nocturnal penile tumescence and rigidity. The prevalence of ED was significantly lower in former than in active smokers (2% vs 3.7%) in one study [12], implying that stopping smoking might decrease the risk of ED. In the present study, smokers who had stopped for 1 year had a significantly better ED status than the matched group of current smokers. Indeed, 25% of ex-smokers had an improvement of at least one grade, while none of the current smokers improved. Considering that age and ED status were not significantly different between the groups before the follow-up, and that patients in both groups had no risk factor except smoking, the improvement in ED status in ex-smokers would seem to be a result of eliminating the detrimental effects of smoking. The reversibility of other smoking-induced vascular diseases after stopping smoking is well reported. The excess risk of coronary artery disease and ischaemic stroke decrease substantially during the initial 2–3 years after stopping smoking [15,16]. That there may be a similar period of reversibility for smoking-induced ED seems likely.

Among ex-smokers the patients with more advanced ED before NRT had less chance of improving (49% of those with mild ED and none of those with severe ED; Table 1). The pathophysiology of smoking-induced ED is poorly understood, but for atherogenic vascular changes as a probable cause, early lesions up to pre-atheroma, as classified by Stary et al.[28], are considered reversible, but advanced lesions (type IV, V and VI) can just be stabilized by anti-atherosclerotic measures [29]. Thus fully established smoking-induced penile vascular damage manifested as advanced ED may need a longer period after stopping smoking to improve, if improvement is at all possible.

Many age-related changes in the vascular wall are known [29], e.g. increase in thickness of the media and intima, regressive lesions and defects of the lamina elastica interna, deviation in the morphological orientation of smooth muscle cells, the occurrence of endothelial cell-free areas, decreased DNA content, progressive fibrosis of the arterial wall, and decreased elastin-collagen ratio (although there is considerable heterogeneity of these changes). Among the present ex-smokers with a similar degree of ED, after stopping for 1 year the proportion of those improved was much higher in younger than older patients. Perhaps the overall age-related degenerative changes in older patients reduce the reparability of vascular structures after stopping, so it cannot be clinically manifested after only 1 year. Obviously, a long-term follow-up of these patients is required to reach a reliable conclusion.

Depending on the main pathophysiology of each smoking-induced vascular disease, the rate of reduced risk differs, e.g. the risk of myocardial infarction reduces sooner than that of coronary artery disease [30] because of their respective pathophysiology, and hypercoagulability vs atherosclerosis [13]. It is logical that there is such a period for smoking-induced ED, but a longer follow-up is needed for it to be accurately established; however, an improvement in a quarter of ex-smokers after only 1 year is good evidence for such a period.

In conclusion, this is the first prospective study of the beneficial effects of stopping smoking in smokers with ED who have no risk factor except smoking. Cigarette smoking had a strong dose-dependent association with ED, and a significant proportion of patients could benefit by stopping smoking; they should be urged to do so when younger, before the disease becomes advanced.


None declared. Source of funding: grant from Iranian Ministry of Health and personal funding of authors.