SEARCH

SEARCH BY CITATION

Keywords:

  • LUTS;
  • erectile dysfunction;
  • incontinence;
  • epidemiology;
  • Asian population

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. SUBJECTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONFLICT OF INTEREST
  8. REFERENCES

OBJECTIVE

To determine, in a population-based survey, the prevalence of lower urinary tract symptoms (LUTS), erectile dysfunction (ED) and incontinence in community-dwelling men in multiethnic Malaysia, as currently available Western demographic data might not be applicable in the Asian population.

SUBJECTS AND METHODS

A cross-sectional population-based survey was carried out in the State of Penang, Malaysia, with a target population of men aged ≥ 40 years. Using a multistage study design, random systematic sampling was used to represent the target population, who were weighted based on ethnicity and rural-urban ratios so as to represent the general population distribution. Trained field-workers conducted direct interviews and administered the American Urological Association Symptom Index (AUA-SI), the International Index of Erectile Function (IIEF-5) questionnaire and questions on incontinence based on the International Continence Society 2002 definition.

RESULTS

In all, 418 men aged ≥ 40 years were interviewed, of whom 353 completed the AUA-SI questionnaire (84.5% response rate). The prevalence of mild, moderate and severe LUTS was 80.6%, 6% and 0.3%, respectively. The prevalent symptoms were frequency and nocturia. There was moderate and severe ED in 45.9% of men, whereas incontinence was reported by 8.2%. The AUA-SI correlated strongly with age (R = 0.291, P < 0.001), IIEF-5 (R = − 0.265, P < 0.001) and diabetes mellitus.

CONCLUSION

The prevalence and severity of LUTS, ED and incontinence increased with age in this multiethnic Asian population, in which ED correlated strongly with LUTS. Compared to the Western population, the prevalence of LUTS was significantly lower, while the prevalence of ED and incontinence were comparable.


Abbreviations
SI

Symptom Index

IIEF

International Index of Erectile Function

ED

erectile dysfunction

BMI

body mass index.

INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. SUBJECTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONFLICT OF INTEREST
  8. REFERENCES

LUTS and erectile dysfunction (ED) appear to be highly prevalent in ageing men and recent publications confirm the correlation between these conditions, suggesting a possible common pathophysiology [1–3]. The general opinion among urologists in Malaysia is that the prevalence of LUTS, ED and incontinence must be different from that in the West. Malaysia is a multiethnic nation comprising mainly Malay (the ethnic majority), Chinese and Indian people. This putative difference stems from the common assumption that Malaysians are more stoical and seek care for what might be perceived as ‘universal ageing problems’ only when symptoms become unbearable. Data from hospital/clinic records and ‘prostate awareness campaigns’ reveal a high prevalence of the above conditions in men presenting to doctors [4]. High-quality population-based prevalence data on these common urological conditions are scarce. Although regional data from uni-ethnic nations are available in a wider context [5], the unique position of Malaysia in being multiethnic deserves attention. Furthermore, there is a need for sound, local epidemiological data to formulate local healthcare policies and form the much needed groundwork for future research.

The primary aim of the present study was to determine the prevalence of LUTS, ED and incontinence in community-dwelling men aged ≥ 40 years in this multiethnic country. The secondary aims were to: (i) correlate LUTS, ED and incontinence; (ii) identify associations between these conditions with diabetes mellitus, hypertension, smoking and body mass index (BMI) in Malaysia; and (iii) compare our findings with previously published studies, with the hypothesis that LUTS and ED, although interrelated, are not as prevalent or severe as in the West.

SUBJECTS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. SUBJECTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONFLICT OF INTEREST
  8. REFERENCES

We conducted a cross-sectional population-based survey in Penang, one of the states in Malaysia, with a population of 1.2 million and an ethnic/socio-economic distribution that represents the whole country (population ≈ 23 million); it also has a mixture of urban and rural settlements. A multistage sampling design was used: In stage 1 of the study, the sampling units were the two Eastern districts in Penang and subsequent stages involved the four other districts. From sample-size calculation, with a power of 80% and 95% CI, assuming the prevalence of incontinence (which was expected to be the least prevalent condition) in the population was 10%, we required 300 sample units (in this case, men aged ≥ 40 years). A systematic random selection of households (the enumeration units) based on telephone subscribers was used, as the availability of this utility was ubiquitous. The more valid electricity or water utility subscriber lists were not available to us, for national security reasons. The sample units were weighted based on ethnicity and rural-urban ratios so as to represent the general population distribution. This ultimately provided a large prospective cohort of randomly selected households, where the elementary (sample) units were non-institutionalized men and women aged ≥ 40 years. Data on incontinence in women was used for a separate analysis.

Independent field-workers who had no interest in the study outcome were trained to conduct the direct interviews, obtain a pertinent medical history, register height and weight of participants and administer the questionnaires. We used the AUA Symptom Index (AUA-SI) for LUTS, the abridged International Index of Erectile Function (IIEF-5) [6] questionnaire for ED and the question ‘Have you involuntarily leaked urine in the last 2 months during any one or a combination of (a) coughing, sneezing or with effort (b) after a sense of urgency [3], continuous urinary leakage or [4] bed-wetting’. All three questionnaires were administered in English, Malay and Chinese. The AUA-SI and IIEF-5 questionnaires were previously validated in the Malaysian community [7]. Eligible respondents were men and women aged ≥ 40 years; the criteria for non-respondents were those who were unavailable for interview on three separate occasions or who were unwilling to participate.

The men were stratified into groups with 10-year increments for a better description of prevalence data. Men were defined as having LUTS if the AUA-SI was > 0, while mild, moderate and severe LUTS were defined by AUA-SI scores of 1–7, 8–19 and 20–35. ED was present if the IIEF-5 was ≤ 21, and mild, moderate and severe ED were defined as scores of 17–21, 8–16 and 0–7, respectively. Mantel-Haenszel and Fisher’s exact two-sided tests were used to compare categorical variables and Spearman’s ρ used for nonparametric variables. Pearson’s correlation coefficient was used for continuous parametric variables. The findings were compared with those from other published series.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. SUBJECTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONFLICT OF INTEREST
  8. REFERENCES

In all, 799 adults (418 men and 381 women aged ≥ 40 years) from 396 randomly selected households were approached during the various stages of the study. From the households approached, 51 men were unavailable for interview despite three separate attempts, whereas 14 men declined to participate. The number (and proportion) of men completing the individual questionnaires (responders) were 353 (84.5%), 344 (82.3%) and 330 (78.9%) for the AUA-SI, IIEF-5 and incontinence questionnaires, respectively.

The demographics of the men completing the AUA-SI are detailed in Table 1. The mean AUA-SI in this sample was 2.9. Overall, 80% of men reported some degree of LUTS (AUA-SI ≥ 1), but the prevalence of moderate to severe LUTS was only 6.2%. In the different ethnic groups, the prevalence of moderate to severe LUTS was 7.8%, 5.8% and 2% in Malays, Chinese and Indians, respectively.

Table 1.  The demographics of the respondents completing the AUA-SI, the prevalence of various degrees and types of LUTS, and the prevalence of ED
VariableAge group, years
40–4950–5960–69≥70
Sample, n (%)133 (86.7)110 (31.4)68 (19.4)39 (11.1)
Mean age, years 44.0 53.963.375.2
Mean BMI, kg/m2 24.5 24.322.421.4
N (%):
Ethnicity
 Malay 75 (56.4) 63 (57.3)33 (48.5)24 (61.5)
 Chinese 27 (20.3) 33 (30)28 (41.2)15 (38.5)
 Indian 31 (23.3) 14 (12.8) 7 (10.3) 0
Occupation
 Farmers 15 (11.3) 19 (17.3) 9 (13.2) 4 (10.3)
 Industrial worker 29 (21.8) 10 (9.1) 5 (7.4) 0
 Businessman 28 (21.1) 25 (22.7)15 (22.1) 5 (12.8)
 Public sector 32 (24.1) 18 (16.4) 8 (11.8) 0
 Retired  3 (2.3) 10 (9.1)11 (16.2)15 (38.5)
 No fixed job  9 (6.8)  8 (7.3) 9 (13.2) 6 (15.4)
 Others 36 (27.1) 12 (10.9) 6 (8.8) 4 (10.3)
Smokers 61 (45.9) 52 (47.3)32 (47.1)14 (35.9)
Diabetes mellitus  9 (6.8) 16 (14.6)12 (17.7) 8 (20.5)
Hypertension  2 (1.5) 14 (12.7)13 (19.1) 7 (17.9)
LUTS
Mean AUA-SI  1.5  2.6 3.4 4.2
AUA-SI
 0 (no LUTS) 22 (16.5) 18 (16.4) 4 (5.9) 2 (5.1)
 1–7 (mild)110 (82.7) 86 (78.2)56 (50.9)30 (76.9)
 8–19 (moderate)  1 (0.75)  6 (5.4) 7 (10.3) 7 (17.9)
 20–35 (severe)  0  0 1 (1.5) 0
Incomplete emptying 10 (7.5) 27 (24.6)13 (19.1)15 (38.5)
Frequency 72 (54.1) 68 (61.8)54 (79.4)24 (61.5)
Intermittency 11 (8.3) 15 (13.6)11 (16.2)10 (25.6)
Urgency  7 (5.3) 15 (13.6)11 (16.2) 9 (23.1)
Poor stream  3 (2.6) 17 (15.5)12 (17.7)12 (30.8)
Straining  4 (3.0) 11 (10) 5 (7.4) 7 (17.9)
Nocturia 72 (54.1) 63 (57.3)37 (54.4)32 (82.1)
Incontinence  2 (1.5) 17 (15.5)11 (16.2)11 (28.2)
ED
Number1271086739
 No ED 72 (56.7) 18 (16.7) 4 (5.9) 2 (5.1)
 Mild ED 37 (29) 36 (33.3)13 (19.4) 1 (2.6)
 Moderate ED 14 (1.6) 45 (13.9)42 (28.4)19 (35.9)
 Severe ED  4 (3.2)  9 (8.3) 8 (11.9)17 (43.6)

When stratified into one-decade intervals, most men had mild to moderate LUTS in each age stratum (Table 1). The prevalence and severity of LUTS increased with age (three participants did not provide ‘age’ and were excluded from age stratification). Men aged ≥ 50 years had a significantly higher AUA-SI than younger men (P = 0.001; Table 1). When compared with the USA and Japan, the prevalence of moderate and severe LUTS in Malaysia was significantly lower (P < 0.001; Fig. 1[8–10]).

image

Figure 1. Age-related prevalence of moderate and severe LUTS in Malaysia, compared with Singapore [8], Japan [9] and Olmsted County, USA [10].

Download figure to PowerPoint

The symptoms most often reported in this sample were frequency and nocturia (Fig. 2[8,10,11]). When compared with the cohort from USA [10], the Malaysians had a significantly lower prevalence of voiding symptoms (i.e. poor stream, intermittency, straining and incomplete emptying), but the prevalence of nocturia was similar to that in the Olmsted county series [10].

image

Figure 2. Prevalence of different LUTS in Malaysia, Singapore [8], Olmsted County [10] and Scotland [11] (all differences between Malaysia and Olmsted county study were statistically significant at P < 0.001).

Download figure to PowerPoint

Of the 344 men completing the IIEF-5 questionnaire, 245 (71.2%) had some degree of ED with 87 (25.3%), 120 (34.9%), 38 (11.1%) reporting mild, moderate and severe ED, respectively. Table 1 also details the age-stratified prevalence of the different degrees of ED; there appeared to be a significant age-related linear increase in the prevalence of severe ED (P < 0.001).

The overall prevalence of incontinence in Malaysian men aged ≥ 40 years was 8.2%; it was 6.6%, 7.9%, 10.6% and 10.3% in those aged 40–49, 50–59, 60–69 and ≥ 70 years, respectively.

Table 2 describes the correlation and association between the prevalence and severity of LUTS and ED and other variables. Men with moderate to severe LUTS had 1.4 times the risk of also having ED. When controlled for age, this relationship was marginally insignificant (R = −0.1008, P = 0.064). Although BMI did not correlate with the AUA-SI, men who were overweight appeared to be at slightly greater risk of having moderate to severe LUTS. Smokers, hypertensives and diabetics were more likely to have ED. There was no association between occupation and LUTS or ED, nor between the severity of LUTS and the prevalence of incontinence.

Table 2.  The correlations between LUTS, ED and other variables, with odds ratios from logistic regression analysis, correlation coefficients (R): Spearman’s ρ for nonparametric and Pearson’s R for parametric variables
VariableIPSSIIEF-5
  • P < 0.001;

  • *

    < 0.05.

Age, R  0.233−0.536
BMI, R −0.141 0.229*
Risk of moderate to severe LUTS (AUA-SI ≥ 8) or ED (IIEF-5 < 17), odds ratio (95% CI)
Overweight (BMI ≥ 25 kg/m2)  1.7 (0.6–4.7) 1.6 (0.9–2.7)
Smoking  0.6 (0.2–1.5) 2.95 (1.6–5.5)*
Diabetes mellitus 10.1 (2–56) 4.63 (1.5–15.7)
Hypertension  2.7 (0.9–9.4) 3.57 (1.2–12.3)*
AUA-SI ≥ 8 1.4 (1.3–1.6)
Incomplete emptying141.7 (19.5–289.6) 1.5
Frequency  6.7 (1.5–44) 0.53 (0.3–0.9)*
Intermittency 12.8 (8.8–18.7)
Urgency 22.4 (7.7–67.1) 2.9 (1.1–7.8)*
Poor stream 27.8 (9.4–84.7)20.1 (2.9–39.8)
Straining 55.1 (17.1–186.7) 3.3 (0.91–14.1)*
Nocturia  2.9 (2.4–3.4) 2.5 (1.5–4.2)
Incontinence  1.3 (2.4–6.4) 1.07 (0.42–2.77)
Incontinent partner 0.93 (0.49–1.78)

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. SUBJECTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONFLICT OF INTEREST
  8. REFERENCES

This is the first large-scale, cross-sectional population-based survey addressing LUTS, ED and incontinence in multiethnic Malaysia. With a high response rate we showed that although the overall prevalence of LUTS in this multiethnic sample was ≈ 87%, the prevalence of moderate to severe LUTS (AUA-SI ≥ 8) was only 6.2%. However, moderate to severe ED and incontinence was reported by 46% and 8.2% of the sample, respectively. The prevalence of all three conditions increased with age. The study also confirmed the correlation between LUTS and ED.

When compared with the West and more industrialized nations, the prevalence of moderate and severe LUTS was significantly lower in Malaysia. This difference could be attributed to recent findings from the Baltimore Longitudinal study of ageing men, which found a strong correlation between BPH and obesity and diabetes mellitus [12], which are conditions less prevalent in Asians.

The obvious differences between the symptoms in the Malaysian and the Olmsted County samples were also apparent, i.e. a smaller proportion of Malaysians had voiding (obstructive) symptoms, with more of the storage (irritative) symptoms. From a prospective study on frequency-volume charts of Asian and Caucasian men presenting with LUTS, Mariappan et al. found that Asians had smaller maximum voided volumes and were more likely to have detrusor overactivity than their Western counterparts. This finding (Mariappan P et al., unpublished data) was also corroborated by a large Japanese study [13].

The worldwide prevalence of ED is 7–52%[14]; comparison with other studies showed marked ethnic variation in the prevalence of ED internationally (Fig. 3[15–18]), with the prevalence in Malaysia comparable with that in the USA. This is in keeping with a United Nations report [19], which suggested that the prevalence of ED seems to be higher in Asia and South America, and is projected to increase further by 2025. The trend of an age-related increasing prevalence of ED in the present survey is similar to that seen in other series, both from this region [16] and the West [17].

image

Figure 3. Prevalence of moderate and severe ED in Malaysia, Thailand [15], Japan [16], Finland [17] and USA [18].

Download figure to PowerPoint

To our knowledge there have been no published regional data on incontinence in men; the value from a Scandinavian population-based survey was 9.2% with a linear age-related increase [20].

The correlation between LUTS and ED identified in large population-based studies [1–3] was attributed to: (i) a common alteration in nitric oxide synthase/nitric oxide in the prostate and penile smooth muscle; (ii) autonomic hyperactivity affecting prostate growth and ED; (iii) increased rho-kinase activity; and (iv) prostatic and penile ischaemia [21]. The present findings of an association between the type of LUTS, i.e. voiding symptoms (obstructive), nocturia and ED can also be corroborated by similar findings from North American studies [22].

We used a separate Emotional Quotient questionnaire to measure the quality of life in these men, instead of using the quality-of-life question of the IPSS; this is part of a separate analysis. Population data from other sources (electricity and water supply registers) was not available to us and therefore we were limited to the telephone subscriber list. Although this subscription was almost universal, we realise that this could contribute to some bias. Nevertheless, we think that our data provides the most random representation of true symptoms in the Malaysian community. A recent publication by Li et al.[5] from the multicentre Asian Survey of Aging Men reported a much higher prevalence of LUTS from the Malaysian men in their cohort. These authors did not specify the method of selection of their sample (whether they were men recruited from hospital presenters or prostate health campaign attendees) and used a combination of mail and direct interview surveys. The regions in which the men were recruited also appeared to have an urban bias.

In conclusion, the results of this population-based survey showed that the prevalence, severity and types of LUTS in this multiethnic Asian population of Malaysia differs from that of the Western population. The prevalence of ED and incontinence was comparable with that in other published series. LUTS appeared to correlate with ED, diabetes and obesity.

CONFLICT OF INTEREST

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. SUBJECTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONFLICT OF INTEREST
  8. REFERENCES

Source of funding: Pfizer, Malaysia, and Sanofi, Malaysia.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. SUBJECTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONFLICT OF INTEREST
  8. REFERENCES