were again found to have a risk ratio of 1.97 (95% CI

1.07

–

3.63) to develop ED adjusted for age and hyperten-

sive medication.

In Finland, a cohort of 1130 men aged 50

–

70 were

followed 10 years in a fashion similar to that done in

the Massachusetts Male Health Study and Olmstead

County survey. This study produced an odds ratio of

1.4 that, while similar to that of the two aforementioned

cohorts, did not reach statistical signi±cance (95% CI

0.9

–

2.2) (Shiri

et al.

, 2005). Further analysis found that

smokers who developed vascular disease had three times

the risk of developing ED compared to nonsmokers

without vascular disease (RR 3.2, 1.3

–

7.5). In contrast,

smokers without vascular disease had no increased risk

of ED (RR 1.0 CI 0.5

–

1.8) (Shiri

et al.

, 2006). These

data have also been the subject of a recent systematic

review, which compiled 8 case

–

control and cohort stu-

dies composed of 28 586 men (the majority of which

are from the Male Health Professions Study) creating a

pooled OR of 1.81 (95% CI 1.34

–

2.44) for smokers hav-

ing increased risk of developing erectile dysfunction

(Cao

et al.

, 2013).

Smoking effects on ED are dose dependent

Cigarette smoking has been suggested to act with dose

dependency as a risk factor for heart disease as well as

ED. In the subgroup analysis of other larger studies, odds

ratios of patients who developed ED showed a signi±cant

difference when men smoked

10 cigarettes per day

(Austoni

et al.

, 2005). Among smokers, a positive but

nonsigni±cant trend towards increased ED occurred in

relation to daily cigarette intake (Chew

et al.

, 2009).

In a younger, less comorbid population, heavy smokers

(

20 cigarettes per day) had doubled the likelihood of

severe ED compared to those who smoked less (Natali

et al.

, 2005). Others have also noted that cumulative

smoking history was also a risk factor for ED. For exam-

ple, cumulative indices such as pack-years were related to

higher risk of ED. In this instance, Gades

et al.

(2005)

found that a 29 pack-year history was responsible for a

signi±cantly increased risk for ED compared to a

pack-year history which carried with it the same risk as a

nonsmoker. In similar ±ndings, the Boston Area Commu-

nity Health Survey found that it was only at a threshold

of 20 pack-years where the OR for developing ED became

signi±cant (Kupelian

et al.

, 2007). However, an earlier

Vietnam Experience Study did not show such a dose rela-

tionship (Mannino

et al.

, 1994).

Overall, it appears that the cumulative dose of cigarette

exposure does predict for the odds of developing ED.

Examining the severity of the ED, current evidence

appears to suggest that heavy smoking causes more severe

ED that appears to be not reversible following smoking

cessation.

Smoking and effects on other comorbidities in

men with ED

Cigarette smoking is also known to affect numerous other

comorbidities associated with ED. For example, athero-

sclerosis and cardiovascular disease are known to affect

erectile function by decreasing penile perfusion pressures,

resulting in increased time to maximal erection and

decreased rigidity during erection (Shabsigh

et al.

, 1991;

Sullivan

et al.

, 1999). Cigarette smoking is associated with

arteriogenic ED and is a component of the general pro-

cess of atherosclerosis (Shabsigh

et al.

, 1991; Sullivan

et al.

, 1999). Arterio-insuf±ciency also hinders erectile

function by decreasing penile perfusion pressures, result-

ing in increased times to maximal erection and decreased

rigidity during erection (Dean & Lue, 2005).

Diabetes also contributes to ED through both micro-

vascular and macrovascular damage (Maiorino

et al.

2014). Studies have shown that

50% of diabetics have

some degree of ED (Giuliano

et al.

, 2004; Thorve

et al.

2011). Furthermore, men with diabetes have a 3-fold

increase in risk for developing ED (Feldman

et al.

, 1994).

Among a group of 51 464 middle-aged and elderly Chi-

nese men, smokers were found to have a hazard ratio of

1.25 with regard to developing type 2 diabetes compared

to nonsmokers (Shi

et al.

, 2013). As such, not only does

cigarette smoking directly impact the physiologic mecha-

nisms of erectile function, but it also contributes to the

development of other medical conditions independently

associated with ED.

Effects of smoking cessation on erectile function

The current literature has yet to reach a consensus as to

the magnitude of the bene±ts for smoking cessation spe-

ci±cally with regard to ED. Indeed, in multiple cross-sec-

tional studies, former smokers (de±ned as having quit

smoking

1 year prior to the study) have an increased

risk of suffering from any form of ED compared to men

who have never smoked (Austoni

et al.

, 2005; Gades

et al.

, 2005; Bacon

et al.

, 2006). Former smokers have

also been shown to have increased risk compared to cur-

rent smokers, even when adjusted for age (Ghalayini

et al.

, 2010). However, the study was compromised due

to a small sample size and no data to describe total

smoking history or the presence of speci±c confounders

such as vascular disease (Ghalayini

et al.

, 2010). In a sep-

arate study that excluded patients with cardiovascular dis-

ease, former smokers had same signi±cantly increased risk

for ED as current smokers (He

et al.

, 2007). Thus, it

2014 Blackwell Verlag GmbH

1089

Andrologia

2015,

, 1087–1092

J. R. Kovac

et al.

Smoking and male erectile function