Karin Coyne, Senior Research Leader, United BioSource Corporation, 7101 Wisconsin Ave, Suite 600, Bethesda, MD 20814, USA. e-mail: Karin.Coyne@unitedbiosource.com
To evaluate the association between International Continence Society categories of lower urinary tract symptoms (LUTS; storage, voiding, and postmicturition) and individual LUTS (associated with decreased sexual activity and sexual satisfaction in men) with erectile dysfunction (ED), ejaculatory dysfunction (EjD) and premature ejaculation (PE).
SUBJECTS AND METHODS
The impact of LUTS on men’s sexual health was captured as part of a cross-sectional epidemiological study to assess the prevalence LUTS among men and women aged ≥40 years in the USA, the UK and Sweden.
The analysis included 11 834 men with a mean age of 56.1 years, 71% of whom reported being currently sexually active. The primary reason for not being sexually active was no partner (35%), followed by personal health (23%) and no desire (23%). Of the men, 26% had mild to severe ED, 7% had EjD, and 16% PE. Men with multiple LUTS had more severe ED and more frequent EjD and PE. Logistic regression analysis showed that greater age, hypertension, diabetes, depression, urgency with fear of leaking, and leaking during sexual activity were significantly associated with ED. The results were similar in the logistic regression analysis for EjD, whereas being younger and the absence of prostatitis were significantly associated with PE, as were the presence of terminal dribble, incomplete emptying, and split stream.
LUTS are associated with common sexual dysfunctions in men. The results of this study highlight the importance of assessing the sexual health of men presenting with LUTS.
LUTS are classified generally as storage (increased daytime frequency, nocturia, urgency, and/or urinary incontinence), voiding (terminal dribble, slow/weak stream, and/or hesitancy), and postmicturition (incomplete emptying and postmicturition dribble) symptoms, and can also include symptoms associated with sexual intercourse and genital and lower urinary tract pain . Urgency, with or without urgency incontinence, can be suggestive of overactive bladder (OAB) , whereas voiding symptoms in men might be suggestive of BOO and/or BPH . Both OAB and BPH are believed to be common in men; a recent population-based survey in Canada, Germany, Italy and the UK found that 12% of men aged ≥18 years had OAB , while findings from autopsy studies suggest that BPH affects ≈40% of men in their 50s and 70% in their 60s .
Increasing evidence supports that LUTS suggestive of OAB and BOO/BPH are related to sexual dysfunction in men. Studies show that LUTS are associated with decreased sexual activity and sexual satisfaction  and with erectile dysfunction (ED) [6–9]. LUTS, particularly voiding symptoms and nocturia, have been identified as independent risk factors for ED [7,8,10], whereas OAB in combination with other storage symptoms has been linked to ED and decreased sexual quality of life in men . Although the nature of this relationship is not known, it is possible that LUTS and ED have a common pathophysiology .
LUTS are also associated with other sexual difficulties in men, including painful ejaculation and ejaculatory dysfunction (EjD, i.e. reduced ejaculate or complete absence of ejaculate). Ejaculatory problems are reported to increase with age and increasing severity of LUTS in men able to achieve an erection . Painful ejaculation was reported by ≈20% of sexually active men with LUTS suggestive of BPH . These men were characterized by being younger, having reduced ejaculate, greater bother severity, more severe LUTS, higher prevalence of ED, UTIs, and a history of macroscopic haematuria.
The relationship of LUTS with the most prevalent sexual dysfunction, premature ejaculation (PE), which is believed to affect 25–40% of men , has not been evaluated in a population-based study. Although historically considered a psychological phenomenon, there is evidence that there are several subtypes of PE with different aetiologies. Recently, steps were taken to better define PE and an evidence-based definition of primary PE was developed . Primary PE is defined as ejaculation that always or nearly always occurs before or within ≈1 min of vaginal penetration, the inability to delay ejaculation on all or nearly all vaginal penetrations, and negative personal consequences such as distress, bother, frustration, and/or the avoidance of sexual intimacy .
Although previous research has identified links between LUTS and specific aspects of sexual dysfunction, most notably ED, the relationship of individual LUTS and ICS categories of urinary symptoms with men’s sexual health has not been evaluated. Furthermore, the possible link between LUTS and sexual dysfunction such as PE and EjD has been largely unexplored. As such, data from the Epidemiology of LUTS (EpiLUTS) study were used to assess the impact of LUTS, as defined by current ICS terminology , on sexual enjoyment and frequency and to describe the relationship between LUTS and ED, EjD and PE.
SUBJECTS AND METHODS
In all, 30 000 men and women aged ≥40 years from the USA (20 000), the UK (7500) and Sweden (2500) participated in the EpiLUTS study. Details of the design and results of the main study are reported elsewhere [16,17].
A survey of LUTS was developed to assess the prevalence of LUTS using current ICS definitions, and in terminology understood by patients. The frequency of affirmative responses was defined by dividing the five Likert response options in two ways; ‘never’ and ‘rarely’, which were combined into ‘no/minimal LUTS’, vs ‘sometimes’ or greater; and ‘never’, ‘rarely’, and ‘sometimes’, which were combined into ‘no/minimal LUTS’ vs ‘often’ or greater.
Participants were also asked to complete the IPSS, which includes seven questions covering frequency, nocturia, weak urinary stream, hesitancy, intermittency, incomplete emptying, and urgency . Symptom severity is categorized as no symptoms (IPSS 0), mild symptoms (IPSS 1–7), moderate symptoms (IPSS 8–19) and severe symptoms (IPSS 20–35) .
After the main survey was completed, men and women were invited to participate in the Sexual Health Survey or were given the opportunity to exit from the survey. This report focuses on the results from the male participants.
Men accepting the invitation to complete the Sexual Health questions were asked questions framed ‘within the past 4 weeks’ about current sexual activity level and were asked about the effect of urinary symptoms on sexual enjoyment and sexual health.
Sexual health assessments included the six-item Erectile Function (EF) domain of the International Index of Erectile Function (IIEF) [19,20], one question modified from the Male Sexual Health Questionnaire to assess EjD (‘Over the past 4 weeks, how often have you been able to ejaculate when having sexual activity?’) , and one slightly modified question from the Index of PE (‘Over the past 4 weeks, when you had sexual intercourse, how often did you ejaculate before you wanted to ejaculate?’) .
For the statistical analysis, sample matching was used to construct a population-representative sample of respondents within each country’s Internet-based panel, and poststratification weights were calculated to correct small amounts of imbalance based on differences in response rates .
Demographic variables, sexual health outcomes, and the relationship between the IIEF-EF domain and IPSS were evaluated by descriptive analyses. Logistic regressions were used to evaluate the predictors of sexual impact and dysfunction in men. Individual regressions used the following dependent variables: the presence of ED (defined by IIEF-EF domain score of <21); note that ED severity is categorized by the IIEF-EF domain score as follows: severe, 1–10; moderate 11–16; mild to moderate 17–21; mild 22–25; and no ED 26–30) ; the presence of EjD (ability to ejaculate half the time or less); the presence of PE (more than half the time); decrease in sexual enjoyment (somewhat of the time or more); and decrease in sexual activity (somewhat of the time or more). Each regression model included age, race, body mass index (BMI), country and the comorbid conditions of heart disease, hypertension, diabetes, bladder cancer, prostate cancer, prostatitis, neurological conditions, depression, and pain during sex as well as the frequency of each individual LUTS. Given the large sample size and multiple analyses conducted, P < 0.01 was used as the threshold for statistical significance.
The demographic characteristics for participants in EpiLUTS who completed the Sexual Health Survey and those who opted not to do so are presented in Table 1 . There were a few small statistically significant differences between those who agreed to participate in the Sexual Health Survey and those who did not; those who opted in were slightly younger and more likely to be white, college-educated, and married/living with partner or divorced as opposed to single. Of these men, 13% reported having an enlarged prostate, and 2.5% reported having prostate cancer. Overall, 84% of men completed the Sexual Health Survey, with 71% of the sexual health participants reporting being sexually active (Table 2). The most frequent reasons for not being currently sexually active were no partner (34.9%), participant’s health (23.3%) and no sexual desire (23.2%) (Table 2).
Table 1. The demographics and comorbid conditions of those who opted in and those who opted out of the sexual health survey*
All numbers presented are weighted; subgroups might or might not equal total N because of rounding or weighted values; IC, interstitial cystitis; PBIS, painful bladder syndrome; IBS, irritable bawel syndrome.
No. of men
Mean (sd) age, years
Race, n (%)
Marital status, n (%)
Married/living with partner
Education, n (%)
High school or less
Comorbid conditions, n (%)
High blood pressure
IC or PBlS
Sleep apnoea/sleep disorder
Table 2. Overall sexual activity of the 11 841 men*
All numbers presented were weighted; subgroups might or might not equal total N because of rounding or weighted values.
†Calculated as a percentage of men who answered this question.
Decreased enjoyment of sex due to LUTS (11094 men)†
Not at all
Quite a bit
A great deal
Decreased/stopped sexual activity due to LUTS (10671 men)†
Not at all
Quite a bit
A great deal
In all, 8508 (72%) Sexual Health respondents reported at least one ICS category of LUTS at least ‘sometimes’. Of these, 17.2% reported voiding symptoms only, 11.6% reported storage symptoms only, and 4.4% reported postmicturition symptoms only. Combinations of ICS categories of LUTS were reported in many men; 14.3% reported both voiding and storage symptoms, 14.9% reported both voiding and postmicturition symptoms, 2.8% reported both storage and postmicturition symptoms, and 34.7% reported a combination of storage, voiding and postmicturition symptoms (data not shown).
In general, sexual enjoyment declined and sexual activity decreased with overlapping LUTS when LUTS were reported at least sometimes (Fig. 1A,B). Of the 2954 respondents reporting a combination of voiding, storage, and postmicturition symptoms at least sometimes, 852 (28.8%) responded that their sexual enjoyment was ‘somewhat’, ‘quite a bit’, or ‘a great deal’ decreased because of LUTS, and 733 (24.8%) reported that they had decreased or stopped sexual activity because of LUTS (Fig. 1A). By contrast, of the 3326 respondents with no/minimal LUTS, only 47 (1.4%) responded that their sexual enjoyment was ‘somewhat’, ‘quite a bit’, or a ‘great deal’ decreased because of LUTS, and 43 (1.3%) reported that they had decreased or stopped sexual activity (Fig. 1B).
Logistic regression analysis showed that several individual LUTS and other health conditions were associated with decreased sexual enjoyment and decreased sexual activity (Table 3). Significant associations with decreased sexual enjoyment included weak stream, split stream, perceived urinary frequency, urgency with fear of leaking, leaking during sex, incomplete emptying, bladder area pain, dysuria, bladder and prostate cancer, prostatitis, and reports of pain during sex. There was an inverse relationship with stress urinary incontinence (laughing, coughing, and/or sneezing). Significant associations with decreased sexual activity included weak stream, perceived urinary frequency, leaking during sex, leaking for no reason, incomplete emptying, dysuria, hypertension, bladder and prostate cancer, prostatitis, and reports of pain during sex.
Table 3. Logistic regression of decreased sexual enjoyment and activity, predictors of ED, EjD and PE
ED (IIEF-EF < 21)
− indicates negative values and + indicates positive values for point estimates; (S)UI, (stress) urinary incontinence. *P < 0.01, **P < 0.001. †Not having a history of prostatitis is associated with PE.
In the total sample, ED of any severity (IIEF-EF domain score <26) was reported in 2864 (26.0%) men (Table 4). As expected, severe ED (IIEF-EF domain ≤10) increased with age. The percentage of men who reported severe ED increased from 3.2% of those aged 40–45 years, to 4.1%, 5.8%, 6.6%, 10.1%, 10.3%, 12.7% and 12.5% of men aged 46–50, 51–55, 56–60, 61–65, 66–70, 71–75 and ≥76 years, respectively.
Table 4. ED (determined with the IIEF-EF domain), ability to ejaculate (Male Sexual Health Questionnaire–Ejaculatory Dysfunction), and control of ejaculation during sex (Index of Premature Ejaculation)
‘How often have you been able to ejaculate when having sexual activity?’;
‘When you had sexual intercourse, how often did you ejaculate before you wanted to ejaculate?’ All numbers presented were weighted; subgroups might or might not equal total n due to rounding or weighted values.
Most men with no/minimal LUTS did not have ED; using both the ‘sometimes’ and ‘often’ classifications of presence of any LUTS, more than half of men with no/minimal LUTS reported no ED (Fig. 2A). Specifically, in all 3326 men reported no/minimal LUTS according to the ‘sometimes’ or more definition of prevalence; of these, 63.7% reported no ED (Fig. 2A). The proportion of men in each of the eight ICS categories of LUTS (no/minimal LUTS, voiding only, storage only, postmicturition only, voiding + storage, voiding + postmicturition, storage + postmicturition, and voiding + storage + postmicturition) who reported mild, mild-to-moderate, moderate and severe ED is illustrated in Fig. 2B. In general, ED severity increased in relation to more overlapping categories of LUTS. This relationship was also found using the IPSS to group LUTS severity (Fig. 3), in which IIEF-EF domain scores were inversely correlated with the IPSS (r = −0.28; P < 0.001), indicating that increasing ED severity was associated with increasing IPSS symptom severity.
The logistic regression analysis showed that greater age, hypertension, diabetes, depression, and the some LUTS (urgency with fear of leaking, weak stream, split stream, leaking during sexual activity and dysuria) were significantly associated with ED (Table 3). Swedish men were less likely to report ED than men in the USA.
Most men reported being able to ejaculate all of the time (71%) or most of the time (18%), with only 6.7% reporting EjD (defined as ability to ejaculate less than half the time or none of the time; Table 4). Responses to the EjD question among sexually active men according to the eight ICS categories of LUTS are shown in Fig. 4. Those with voiding, storage and postmicturition LUTS had the highest proportion of ejaculation problems, with 47% reporting that they ejaculate ‘most of the time’ or less. The proportion of men who reported that they could not ejaculate at all increased with age and was reported in 1.2%, 1.4%, 2.0%, 3.2%, 3.7%, 6.9%, 12.1% and 14.1% of those aged 40–45, 46–50, 51–55, 56–60, 61–65, 66–70, 71–75 and ≥76 years, respectively.
In the logistic regression analysis, significant associations with EjD were increasing age, prostate cancer, depression, and the individual LUTS of leaking during sex and urgency with fear of leaking (Table 3).
Overall, 4670 (52.8%) men reported that they almost never or never ejaculated before they wanted to ejaculate when they had sexual intercourse, and the prevalence of PE (defined as ejaculating prematurely ‘more than half the time’ or ‘almost always or always’) was 15.7%. Responses to the PE question among sexually active men according to the eight ICS categories of LUTS are shown in Fig. 5. Those with a combination of storage, voiding and postmicturition symptoms reported the greatest proportion of PE ‘more than half the time’ or more (21.3%).
Logistic regression analysis showed that being younger and the absence of prostatitis were significantly associated with PE, as were the presence of split stream, terminal dribble, incomplete emptying and bladder area pain (Table 3). However, the PE model was less robust (with a concordance index of 0.62) than the other logistic regression analyses noted in Table 3, and these results should be interpreted with caution.
Sexual problems are common in men of all ages and affect health-related quality of life, well-being, and interpersonal relationships [23,24]. Several aetiological factors are associated with sexual dysfunction, including age, chronic medical illnesses and comorbid conditions (e.g. diabetes, hypertension), and mood or psychiatric disorders (e.g. depression). The link between LUTS and diminished sexual activity and impaired sexual function has been reported in several studies [8,10,11,25–28]. The mounting evidence of this connection (in particular the link between LUTS and ED) has led to the suggestion that one or several pathophysiological mechanisms might underlie this relationship, although causal links have not yet been empirically established .
In the present study, almost 30% of men with LUTS reported that their urinary symptoms reduced their enjoyment of sex, and 25% reported that they had decreased or stopped sexual activity because of their urinary symptoms. This finding corroborates previous results from studies conducted in clinical samples. Among male urology clinic outpatients aged ≥45 years, Frankel et al. found that 46% reported that their sexual life was spoiled by LUTS. Findings from logistic regression analysis showed that symptoms commonly associated with BOO/benign prostatic obstruction and OAB, including increased daytime frequency, urgency, weak stream, and incomplete emptying, as well as leaking during sex, dysuria and bladder area pain, were linked to decreased sexual enjoyment and decreased sexual activity.
Overall, 26% of men reported some ED (mild to severe), and those with several ICS symptom categories of LUTS were more likely to have moderate and severe levels of ED. Although previous estimates of the prevalence of ED have been somewhat higher, ranging from 35% to 49%, this is probably because of sampling and questionnaire selection. For example, the Massachusetts Male Aging Study found that 34.8% men aged 40–70 years had moderate to complete ED based on a single question , while the Multinational Survey of the Aging Male (MSAM), which found that 49% of men had difficulty achieving an erection and 10% had a complete absence of erections, was conducted in an older cohort of men, aged 50–80 years .
The current study results also support the connection between LUTS and ED. Significant predictors of ED in the logistic regression model included many of the same voiding and storage symptoms associated with decreased sexual enjoyment and activity, such as weak stream, urgency, and leaking during sexual activity, as well as dysuria. To some extent, these findings are supported by previous reports showing an association between the severity of LUTS and ED. Paralleling our findings, several studies have reported that erection problems are more common in men with LUTS [8,10,28]. In the Cologne Male Survey, in which LUTS were determined with the IPSS and ED with the Kölner Erfassungsbogen der Erektile Dysfunktion questionnaire, the prevalence of LUTS in men with ED was 72%, compared with 38% in men with normal erections . Similarly, data from the MSAM study showed that the frequency of erectile difficulties was strongly related to LUTS severity within three 10-year age categories (40–59, 60–69 and 70–80 years) . Although previous research has not disentangled the individual contribution of various LUTS, Rosen et al. found that LUTS severity was independently associated with erection problems as assessed by both the Danish Prostate Symptom Score-Sex items and IIEF domain scores using logistic regression analysis. In addition to the significant associations noted for individual LUTS in relation to ED, the present results also show a significant relationship between IPSS and ED, with increasing ED severity observed in relation to increasing LUTS severity. The significant links found between greater age and comorbid conditions, including hypertension, diabetes and depression, in relation to ED in this study are also consistent with previous research. ED and LUTS are both associated with increasing age [10,30]. However, there is speculation that they might share a common pathophysiological aetiology, such as endothelial dysfunction . Hypertension, hyperlipidaemia, diabetes mellitus, and depression were found to be more prevalent in men with ED . ED might share common risk factors with these conditions, and it has been suggested that some men with ED might have underlying depression .
In this population-representative sample, most men reported being able to ejaculate all of the time (71%) or most of the time (18%), and the prevalence of EjD was 7%. Although epidemiological estimates of EjD are sparse, data from a large, community-based longitudinal study in the Netherlands indicated that a complete absence of ejaculations occurred in 4% of men aged <65 years, and another 4% in this age group reported significantly reduced quantity of ejaculate . In men aged ≥65 years the percentages were much higher (16% and 10%, respectively), with quantity of ejaculatory volume positively related to EF. Rosen et al. found that 5% of men had a complete absence of ejaculation, while a much larger proportion (46%) reported a reduced quantity. In a clinical sample, more than half of men with prostatitis were found to have EjD . Logistic regression results in the present study showed that EjD was associated with urgency, leaking during sexual activity and dysuria, as well as increased age, prostate cancer and depression. Previous work also has shown LUTS to be associated with an inability to ejaculate or a decreased ejaculation volume [10,25], as well as with pain and discomfort upon ejaculation . The finding that EjD is related to advancing age is consistent with previous reports [10,25]. Other research examining correlates of EjD showed that advancing age, decreased EF and previous prostate surgery were significantly linked with this outcome . An important caveat to our findings on EjD is that we do not know if participants are currently on α1-adrenoceptor antagonists for their LUTS, which could increase the frequency of EjD.
In all, 16% of men reported ejaculating before they wanted to more than half the time or almost always, which was defined as presence of PE in this study. Previous estimates of the prevalence of PE in men across all age groups suggest that it is 25–40%. That EpiLUTS was conducted among men aged ≥40 years, coupled with differences across studies in how PE was assessed and defined, might explain the lower prevalence estimate reported here. Also, PE is a multifactorial disorder with many subtypes, but the EpiLUTS study did not differentiate between men with primary PE (inability to control ejaculation from the beginning of their sexual lives) and those with secondary or acquired PE (developed after years of normal sexual functioning) [15,37,38]. Correlates of PE in the current study included younger age and the absence of prostatitis, as well as the LUTS of terminal dribble, split stream, incomplete emptying, and bladder area pain. However, this model was less robust than those for sexual enjoyment, ED and EjD, and the results should be interpreted with caution. Previous studies have also shown that PE is associated with being younger [39,40], and with hyperprolactinaemia [39,40] and hyperthyroidism . Contrary to our findings, prostatitis was associated with both primary and secondary PE in a previous study .
Interestingly and contrary to previous findings, BMI was not related to any of the sexual health outcomes in the present study. By contrast, El Sakka  reported a significant association between increased BMI and increased ED severity and presence of PE. Given the association of increased BMI with many of the comorbid conditions included in the models (e.g. diabetes and heart disease), perhaps the significance of BMI was reduced. Another interesting finding was that increasing age was not associated with either decreased sexual enjoyment or decreased sexual activity.
The cross-sectional design of this study limits the results to observational and descriptive outcomes for use in hypothesis-building. Importantly, some pharmacological treatments for LUTS, including 5α-reductase inhibitors, and α1-adrenoceptor antagonists, can result in abnormal ejaculation, such as retrograde ejaculation, reduced ejaculate volume, or absence of ejaculate . Thus, although some LUTS were significantly associated with EjD in this study, the direction underlying these associations is indeterminate, and it is possible that the observed relationship might be partly explained by pharmacological intervention. An additional limitation is that no physical examination was conducted; all data collected were based on self-reporting, which involves subjective interpretation of the questions on the part of the participant. Although the IIEF-EF domain has been shown to correlate well with physician diagnosis , the presence of PE and EjD was based on single-item questions. Furthermore, none of the outcomes in this study were confirmed by a clinical diagnosis. Also, because this was an Internet-based survey, the study population might have been biased toward those persons with access to and the skills to use the Internet, notwithstanding the efforts made to ensure that the sample was representative of the general population.
In conclusion, individual LUTS are significantly associated with decreased sexual health and health-related quality of life in men aged ≥40 years. The results of the study suggest that men presenting to their physician with common urological symptoms associated with BPH/BOO, OAB and/or incontinence should also receive a sexual health assessment. Similarly, when treating men for sexual problems, urological conditions, which men often fail to mention because of shame or embarrassment, should be assessed, because management of these conditions could also alleviate the sexual dysfunction.
Future studies should focus on understanding the underlying pathophysiology of LUTS and domains of sexual health, to facilitate better diagnosis, management and the development of new therapeutic approaches for physicians and the patients they treat.
This study was funded by Pfizer Inc. Editorial support was provided by Nancy Price, PhD, at Complete Healthcare Communications, Inc., and was funded by Pfizer Inc.
CONFLICTS OF INTEREST
Alan Wein is a scientific consultant with Allergan, Astellas, Novartis, and Pfizer. Karin Coyne and Chris Sexton are employees of United BioSource Corporation and scientific consultants to Pfizer in connection with the development of the study and manuscript. Andrea Tubaro is a scientific consultant for Astellas, GlaxoSmithKline, Novartis, and Ferring. Zoe Kopp and Lalitha Aiyer are employees of Pfizer Inc.