PATIENTS AND METHODS
- Top of page
- PATIENTS AND METHODS
- CONFLICT OF INTEREST
The Diabetic Center of the Capital Governorate in Kuwait, where the study took place, serves all primary-care clinics in the governorate, and covers a catchment with a population of ≈ 290 000. All patients found or suspected to have DM in these primary-care clinics are referred to this Center. In the Diabetic Center, from June 2004 to June 2005, those men found to have type 2 DM as defined below were included in the study. We also included men known to have type 2 DM diagnosed within 1 year from the time of the interview. In this way, 323 men with type 2 DM, aged 21–65 years, were recruited. In accordance with WHO recommendations , type 2 DM was diagnosed when the following conditions were met: (i) for patients with no symptoms of hyperglycaemia, a fasting capillary whole-blood glucose measurement of ≥ 6.1 mmol/L or a fasting plasma glucose level (FPG) of ≥ 7.0 mmol/L on two separate occasions; (ii) for patients with symptoms of hyperglycaemia, one fasting capillary whole-blood glucose measurement of ≥ 6.1 mmol/L or one FPG of ≥ 7.0 mmol/L.
All of the men were interviewed by an experienced diabetologist, following a protocol approved by the local ethical committee. After obtaining informed consent, the men were asked about their sex life; during the interview, the diabetologist explained the meaning of ED before asking whether they had experienced ED in the past 12 months. An abbreviated version of the International Index of Erectile Function (IIEF)-5  was administered to all men. Men with a score of ≤ 21 (range of score of 5–25) were considered to have ED.
A personal and medical history was taken for each man. Risk factors such as age, education level, smoking habits, and nationality were assessed by direct questioning. Body mass index (BMI) was based on the weight within 3 months of diagnosis. Blood pressure was calculated as the mean of two measurements taken while sitting after 5 min of rest. Men were considered hypertensive if they had a diastolic blood pressure of ≥ 90 mmHg, a systolic blood pressure of ≥ 140 mmHg, and/or were taking antihypertensive medications . Serum total cholesterol was measured using enzymatic techniques. Hypercholesterolaemia was defined as a serum total cholesterol level of >6.2 mmol/L, while borderline hypercholesterolaemia was defined as a level of 5.2–6.2 mmol/L. Income was recorded in the local currency and converted to US dollars. The glycosylated haemoglobin (HbA1c) percentage was determined by ion-exchange HPLC.
Statistical analysis began with descriptive methods followed by tests of associations between ED and potential risk factors. As the predictor variables were categorical or grouped, chi-square tests, including the test for trend (when the predictor variables were ordinal), were used to examine the association between the risk of ED and individual risk factors. Odds ratios of ED associated with the predictor variables were computed using univariate and multivariate logistic regression. Multivariate logistic regression allowed the identification of the independent predictors that were significantly associated with ED; P < 0.05 was considered to indicate statistical significance.
- Top of page
- PATIENTS AND METHODS
- CONFLICT OF INTEREST
Table 1 shows the medical and social characteristics of the men. Of 651 men newly diagnosed with type 2 DM, 305 refused to participate and 23 could not be screened because of hearing or vision impairment, thus 323 men participated in the study. They had a mean age of 41.7 years, and about a third were Kuwaiti nationals. Current smokers constituted 13.9% of the sample, and about a quarter had hypertension.
Table 1. The medical and social characteristics of the study patients (323 men) according to presence of ED
|All||With no ED||With ED|
|Number of men||323||223||100|| |
|Mean (sem) age, years|| 41.7 (0.6)|| 39.1 (0.6)|| 47.5 (1.0)||<0.001*|
|Age group, years; n, (%)|| || || ||<0.001‡|
| <40||152 (47.1)||133 (59.6)|| 19 (19.0)|| |
| 40–50||109 (33.7)|| 67 (30.0)|| 42 (42.0)|| |
| >50|| 62 (19.2)|| 23 (37.1)|| 39 (39.0)|| |
|Nationality|| || || ||0.02†|
| Kuwaiti||107 (33.1)|| 83 (37.2)|| 24 (24)|| |
| Not Kuwaiti||216 (66.9)||140 (62.8)|| 76 (76)|| |
|Monthly income, US $|| || || ||0.70†|
| <1500||275 (85.1)|| 191 (85.7)|| 84 (84)|| |
| >1500|| 48 (14.9)|| 32 (14.3)|| 16 (16)|| |
|Education level|| || || ||0.003†|
| <High school||128 (39.6)||100 (44.8)|| 28 (28)||<0.001‡|
| High school||159 (49.2)||105 (47.1)|| 54 (54)|| |
| >High school|| 36 (11.1)|| 18 (8.1)|| 18 (18)|| |
|Current smoking|| || || ||<0.001†|
| No||278 (86.1)||204 (91.5)|| 74 (74)|| |
| Yes|| 45 (13.9)|| 19 (8.5)|| 26 (26)|| |
|Mean (sem) duration of smoking, pack years|| 12.8 (0.9)|| 15.23 (1.1)|| 9.47 (1.2)||<0.001*|
|Mean (sem) HbA1c, %|| 6.59 (0.05)|| 6.44 (0.07)|| 6.93 (0.08)||<0.001*|
|Mean (sem) serum cholesterol, mmol/L|| 5.1 (0.04)|| 5.06 (0.05)|| 5.2 (0.07)||0.127*|
|BMI, kg/m2|| || || ||<0.001†|
| 18.5–25||186 (57.6)||146 (65.5)|| 40 (40.0)||<0.001‡|
| 25–30|| 110 (34.1)|| 66 (29.6)|| 44 (44.0)|| |
| >30|| 27 (8.4)|| 11 (4.9)|| 16 (16)|| |
|Hypertension|| || || ||0.001†|
| Yes|| 87 (26.9)|| 48 (21.5)|| 39 (39)|| |
| No||236 (73.1)||175 (78.5)|| 61 (61)|| |
The overall mean (sem) prevalence of ED in the study population was 31 (0.26)%. The prevalence was significantly associated with age (P < 0.05); as shown in the chi-square test for trend. The prevalence increased with increasing age; men aged 60–69 years had a prevalence of 63% compared to 11% among men aged 20–29 years.
The association of other risk factors with ED is shown in Tables 1 and2. Compared to potent men, men with ED had significant differences for serum HbA1c level, current smoking habits, duration of smoking, BMI, nationality, and hypertension. The other risk factor that was significantly different between the groups was the education level, which was inversely related to the prevalence of ED (P = 0.003). There were no significant associations between monthly income, serum cholesterol level and the risk of ED.
Table 2. Risk factors associated with ED quantified by univariate and multivariate logistic regression among 323 men with newly diagnosed type 2 DM
|Risk Factor||Univariate logistic regression||Multivariate logistic regression|
|Frequency||Crude OR (95% CI)||P||Adjusted OR (95% CI)||P|
| <40||152||Reference|| ||Reference|| |
| 40–50||109||4.39 (2.37–8.13)||<0.001||4.79 (2.48–9.25)||<0.001|
| >50|| 62||11.87 (5.87–24.02)||<0.001||9.39 (4.42–19.97)||<0.001|
| Kuwaiti||216||Reference|| || || |
| Not Kuwaiti||107||1.88 (1.10–3.20)||0.021|| || |
| <High school||128||Reference|| || || |
| High school||159||1.84 (1.08–3.13)||0.25|| || |
| >High school|| 36||3.57 (1.64–7.76)||0.01|| || |
|Monthly income, US $|
| <1500||275||Reference|| || || |
| >1500|| 48||1.13 (0.59–2.18)||0.74|| || |
| No||278||Reference|| ||Reference|| |
| Yes|| 45||3.78 (1.97–7.22)||<0.001||3.15 (1.49–6.65)||0.003|
| No||236||Reference|| ||Reference|| |
| Yes|| 87||2.33 (1.40–3.90)||<0.001||2.00 (1.10–3.64)||0.023|
| 18.5–25||186||Reference|| ||Reference|| |
| 25–30|| 110||2.43 (1.45–4.08)||<0.001||1.90 (1.06–3.40)||0.032|
| >30|| 27||5.31 (2.28-12.34)||<0.001||5.00 (1.84–13.57)||0.002|
| <8||305||Reference|| || || |
| >8.1|| 18||2.99 (1.14–7.81)||0.03|| || |
|Serum cholesterol, mmol/L|
| <5.2||216||Reference|| || || |
| 5.2–6.2|| 81||1.04 (0.60–1.81)||0.90|| || |
| >6.2|| 26||1.46 (0.63–3.37)||0.39|| || |
When multivariate logistic models were used to identify the independent predictors of the ED, four important risk factors were significantly and independently associated with ED (Table 2), the most important of which was age. Compared to men aged <40 years, men aged 40–50 years had almost five times the risk of having ED, whilst men aged >50 years had nine times the risk. Current smoking, hypertension and BMI were all independently associated with ED. Smokers were almost three times more likely to have ED than non-smokers, and men with hypertension had twice the risk of having ED after adjusting for other important factors. Obese and overweight men had a greater risk of ED, (five times and twice, respectively) than men with a normal BMI.
- Top of page
- PATIENTS AND METHODS
- CONFLICT OF INTEREST
While the prevalence of undiagnosed type 2 DM in ED has been reported , we are not aware of previous data in men with newly diagnosed type 2 DM. In the present study, the prevalence of ED was 31% and it increased with age. The main determinants of risk were age, hypertension, smoking and BMI. Many studies report that the prevalence of ED in established diabetics is 20–75%[12–14,19–23]. These studies were conducted during different periods, in different settings and in populations with different mean ages, and might be difficult to compare with the present study. In addition, many of those studies showed a significant positive correlation between the prevalence and the duration of diabetes. Therefore, it is not surprising that the prevalence of ED in known diabetics tends to be higher than in newly diagnosed cases.
The influence of age on the prevalence of ED is well established in normal men  and in men with known DM [9,22]. Advancing age was significantly correlated with a higher ED prevalence, as shown in Tables 1 and 2. Compared with similar age groups, the prevalence in the present study, as expected, is one of the lowest reported. The longer the duration of diabetes, the greater the chance of ED, supporting the aetiological significance of DM in the development of ED.
Expatriates in Kuwait are mainly from the Indian subcontinent, Middle East or North Africa, and the vast majority are unskilled workers with salaries much lower than the average Kuwaiti. Whereas most Kuwaitis have partners, most expatriates are living without partners, either by choice or because they cannot afford to bring their families with them to Kuwait, given their low salaries and high living expenses, hence they are not in a stable sexual relationship. These circumstances might explain why, after adjusting for age, expatriates were 1.8 times more likely to report ED than citizens (Table 2). Moreover, it is possible that most expatriates were not embarrassed by the questions, whereas locals are sensitive to sex-related queries, as it is a taboo in Islamic culture. Therefore, they might have reported their ED more faithfully.
In the present study, unlike Western reports [22,25], men with a higher education level had higher risks of ED. We attribute this to probably greater stresses associated with the careers of the educated men. An association between smoking and risk of ED was reported . This association might be due to the positive association of cigarette smoking with atherosclerosis, which itself is thought to be an important cause of early ED in men with diabetes . In the present study there was also a greater risk of ED in smokers; among those who smoked, the mean duration of smoking was significantly higher in men with ED (Table 1). When adjusted for age, hypertension and BMI, smokers had more than three times the risk of ED (Table 2). Smoking was also shown to increase the risk of ED in nondiabetic men .
Medical conditions such as hypertension, depression, cardiac disease and pelvic trauma or surgery can be associated with a greater risk of ED . In hypertension, the relationship with ED is complex and might involve both the direct effects of elevated blood pressure and the effects of antihypertensive medication on the microcirculation . In the present study, while adjusting for age, BMI and smoking, men with hypertension were twice as likely to report ED as those who were not hypertensive. This is consistent with the finding in Italian diabetic men where, while controlling for age and other risk factors, there was a 67% increase in the relative risk of ED in those who were hypertensive compared to those who were normotensive .
Although there was no significant association between monthly income and the risk of ED, this does not necessarily mean that income level is not related to ED. This is because we measured income level in two categories (< or >$1500). Most men earned <$1500, so the variability in income levels might not have been captured well. It is also possible that other factors related to income, such as smoking or hypertension, could confound the income-ED relationship.
Elevated HbA1c end-products associated with hyperglycaemia in men with DM have been postulated to decrease nitric oxide activity and reduce endothelium-dependent relaxation factors associated with an increased risk of ED . McCulloch et al. found a positive association of poor glycaemic control and the 5-year incidence of ED in men with DM. Those findings support the present results where, compared to men with HbA1c levels of <8%, men with HbA1c levels of >8.1% were almost three times more likely to report ED. In the present study there was also a significant association between ED and BMI even after adjusting for other significant risk factors, indicating that BMI is independently related to ED. In addition, there was a significant linear trend that enabled us to refine this association to a suggestion that the proportion of men with ED increased with an increase in BMI.
The study population consisted of newly diagnosed men referred to the diabetes centre in the Capital governorate. They might not represent all newly diagnosed diabetics, as other (albeit smaller) centres in the country treat DM. Most patients seen in this clinic are expatriates, and this might lead to under-representation of newly diagnosed men. Also, some newly diagnosed diabetics with more severe forms of ED might go directly to the governorate hospital for treatment, especially those with multiple medical problems, who will need ambulatory hospital care. Therefore, the present study might include some men with less severe comorbid medical conditions. Population-based studies would therefore be needed to estimate the true prevalence of ED in newly diagnosed diabetics in our community. The strengths of the present study were that this was an opportunity to analyse the prevalence of ED in men newly diagnosed with type 2 DM for the first time, and was conducted by direct interviews with the patients by an experienced full-time diabetologist. This should minimize patient misunderstanding of the questions, and obviate interobserver differences.
In conclusion, ED appears to be common in men with newly diagnosed type 2 DM and its prevalence is affected by many factors, including age, hypertension, smoking and BMI.