The prevalence of lower urinary tract symptoms (LUTS) in Brazil: Results from the epidemiology of LUTS (Brazil LUTS) study

1 | INTRODUCTION

Lower urinary tract symptoms (LUTS) is an umbrella term for a wide variety of storage, voiding and post-micturition symptoms.1 Large, international studies conducted in Europe and North America have reported that the prevalence of LUTS is over 60% in men and women aged >40 years.2, 3

No population-based studies have been published on the prevalence of LUTS across the whole of Brazil, one of the largest and most demographically diverse countries in the world. However, several LUTS studies have been performed in single cities of Brazil. One such study was performed in adults aged ≥30 years in Salvador, a city in the Northeastern region, and found that 67% of both men and women had LUTS according to International Continence Society (ICS) definitions, with nocturia defined as at least two episodes per night.4 The prevalence of urinary incontinence (UI) in 519 adults aged ≥18 years was reported to be 20% in a study performed in Pouso Alegre, Minas Gerais.5 In another study, 657 residents of São Paulo were interviewed using a survey with questions regarding UI and nocturia.6 Current UI, sporadic UI, and UI upon moderate efforts were present in 11%, 51%, and 74% of the participants.

The impact of LUTS is wide-ranging and significant, with impairment of quality of life, social functioning, sexual quality of life, and workplace productivity.7 These factors, particularly workplace productivity, can result in economic effects. There is considerable need to understand the extent of the problem, and to ensure that affected individuals are treated effectively.

Overactive bladder (OAB), which is defined by the ICS as urgency, with or without urgency incontinence, usually with frequency and nocturia, is also likely to cause bothersome symptoms that affect sleep, mental health, and work productivity.7 A survey in Porto Alegre, Brazil reported that OAB affects 19% of individuals aged 15-55 years.8

We performed a cross-sectional study to determine the prevalence and bother of LUTS and OAB in adults aged ≥40 years in five major cities, each from a different geographic region of Brazil.

2 | MATERIALS AND METHODS

We conducted a computer-assisted telephone survey between 1 September and 31 December 2015. Study participants were adults aged ≥40 years with residential phone lines (landlines) residing in major cities from each of the five geographical regions of Brazil: São Paulo (Southeast), Porto Alegre (South), Recife (Northeast), Belém (North) and Goiânia (Central-west). Pregnant women and individuals who reported a current or past (within 1 month) urinary tract infection were excluded. Selection procedures are outlined below and the telephone interviews are detailed in supplementary information.

The study was approved by a local ethics committee (Ethics Comittee of Faculdade de Medicina da Universidade de São Paulo) and was performed in compliance with Good Clinical Practice and in accordance with the Declaration of Helsinki. Informed consent was provided by all participants.

LUTS were assessed using a standardized protocol based on ICS definitions.1 The International Prostate Symptom Score (IPSS)9 and the Overactive Bladder—Validated 8-question Screener (OAB-V8)10 questionnaires were also included. All terms and questionnaires were validated in Portuguese. Participants were asked to rate how often they experienced individual LUTS during the past month (at least one voiding, storage [nocturia, two or more episodes] or post-micturition symptom). A Likert scale was used with the following options: none (score 0), <1 in 5 times (score 1), <half the time (score 2), about half the time (score 3), >half the time (score 4), or almost always (score 5). The degree of associated bother was also assessed (Likert scale: not at all [score 0], a little bit [score 1], somewhat [score 2], quite a bit [score 3], a great deal [score 4], or a very great deal [score 5]).

The primary study objective was to estimate the prevalence and associated symptom bother of LUTS in men and women aged ≥40 years in five major cities in Brazil. Two definitions were used for the presence of LUTS: definition 1 (Likert score 2-5; symptoms occurring less than half the time or more), and definition 2 (Likert score 3-5; symptoms occurring half the time or more). Secondary endpoints included the bother of LUTS, prevalence of specific symptoms, prevalence of OAB, and quality of life according to IPSS (question 8).

The sample size was calculated based on the population age distribution and expected LUTS prevalence rates (17.0-30.0% in men and 17.8-22.5% in women).11 Assuming a coefficient of variation of 0.15, a sample size of 1000 interviews in each of the five cities was calculated to exceed the required sample size for estimating LUTS prevalence. Telephone numbers of potential study participants were selected randomly, with stratification by zip code to ensure equal representation by city.

Data were weighted to compensate for differences in the probabilities of respondent selection. Depending on the random selection of a resident within each household, data were weighted initially according to number of residents then by number of telephones. To reduce potential bias due to landline coverage and non-response, data were further adjusted post-stratification, matching respondents' distribution by age, sex and education to that observed for each city's population in the 2010 census.

Descriptive statistics were used for initial data analysis. Chi-squared tests evaluated differences in LUTS prevalence between the sexes and between age groups. In addition, linear association between frequency and bother of symptoms was evaluated using Spearman's rank correlation coefficient. All statistical analyses were performed using SPSS (v20 for Microsoft Windows).

3 | RESULTS

A total of 17 600 telephone numbers were called, and 11 483 individuals met the eligibility criteria. Refusals were received from 2959 individuals (26%), and the interview was not completed by 3340 individuals (29%) for other reasons (rescheduled interviews, interrupted calls, calls not answered). Thus, 5184 individuals (45%) completed the interview: 1004 from São Paulo, 1002 from Porto Alegre, 1013 from Recife, 1004 from Belém, and 1161 from Goiânia. Demographic characteristics are shown in Table 1. Women comprised 53% of study participants, and the age group with most participants (34%) was 50-59 years.

3.1 Prevalence of LUTS

Using definition 1, the prevalence of LUTS was 75% (69% in men and 82% in women; P < 0.001). The corresponding values using definition 2 were 49% (40% and 59%; P < 0.001). The prevalence of LUTS increased with age (Figure 1; definition 1: P < 0.001 in men and in women). In women aged ≥70 years, 95.6% had LUTS occurring about half the time or more, compared with 71.3% of men.

Symptoms with the highest overall prevalence were perceived frequency and nocturia (Table 2). Terminal dribble and slow stream were the most common voiding symptoms. In general, voiding symptoms were more common in men than women, while storage symptoms were more common in women. The most common voiding symptom, terminal dribble, affected 25.2% of men and 17.7% of women, while perceived frequency (the most common storage symptom) was present in 27.7% of men and 32.3% of women. The most prevalent ICS symptom category in both men and women was storage (55.1% and 76.9%, respectively for definition 1 and 29.4% and 53.5%, respectively for definition 2).

Table 2. Prevalence of specific symptoms according to definition 1 (symptoms occurring less than half the time or more) and definition 2 (symptoms occurring about half the time or more) and associated bother in men and women

	Men (n = 2433)						Women (n = 2751)
	Symptom prevalence [definition 1]a		Symptom prevalence [definition 2]a		Prevalence of bother [at least quite a bit]a, b		Symptom prevalence [definition 1]a		Symptom prevalence [definition 2]a		Prevalence of bother [at least quite a bit]a, b
	n	%	n	%	n	%	n	%	n	%	n	%
Voiding symptoms
Slow stream	460	18.9	228	9.4	57	25.1**	437	15.9	206	7.5	96	46.6
Splitting	422	17.4*	194	8.0	57	29.6**	376	13.7	192	7.0	104	54.4
Intermittency	322	13.3	159	6.5	71	44.8	412	15.0	200	7.3	99	49.7
Hesitancy	329	13.5**	138	5.7	58	42.1	253	9.2	114	4.2	49	42.8
Straining	157	6.4*	65	2.7	40	62.0	113	4.1	59	2.2	33	56.3
Terminal dribble	614	25.2***	312	12.8**	94	30.1	487	17.7	237	8.6	102	43.0
Storage symptoms
Perceived frequency	673	27.7*	397	16.3*	123	31.0	888	32.3	556	20.2	217	39.0
Nocturia ≥2 episodes	659	27.1*	287	11.8*	112	39.0	891	32.4	413	15.0	169	41.0
Urgency	455	18.7*	170	7.0***	84	49.4**	647	23.5	324	11.8	238	73.6
Urgency with fear of leaking	356	14.6***	129	5.3***	85	66.0	603	21.9	309	11.2	218	70.6
Urgency urinary incontinence	228	9.4***	100	4.1***	67	66.7	409	14.9	214	7.8	160	74.5
Stress urinary incontinence	63	2.6***	24	1.0***	16	66.8	562	20.4	268	9.7	204	76.2
Leak for no reason	95	3.9***	51	2.1**	25	49.7*	284	10.3%	134	4.9	106	79.2
Nocturnal enuresis	38	1.6**	18	0.7*	8	47.5	95	3.5	52	1.9	40	78.0
Leak during sexual activityc	22	1.1*	10	0.5	4	36.9	32	2.5	14	1.1	10	73.3
Post-micturition symptoms
Incomplete emptying	306	12.6	150	6.2	88	59.0	381	13.8	190	6.9	128	67.4
Post-micturition dribble	418	17.2***	158	6.5	87	55.3*	330	12.0	165	6.0	128	77.4

• ^a Weighted values (subgroups may not equal total n because of rounding, weighted values or missing data).
• ^b Based on definition 2.
• ^c 1871 cases with no sexual activity in the past 6 months and 33 cases with missing information.
• * P ≤ 0.05, men vs women.
• ** P ≤ 0.01, men vs women.
• ^*** P ≤ 0.001, men vs women.

When considering symptom subtypes and combinations (Figure 2), using definition 1, the most common LUTS subtype was storage symptoms alone in both men (prevalence: 20.6%) and women (36.4%). The second and third most common subtypes/combinations were voiding plus storage symptoms, and the combination of all three subtypes. With definition 2, the most common subtype remained storage symptoms only, with prevalence rates of 13.3% and 30.0% in men and women, respectively. Approximately 40-50% of individuals with LUTS according to either definition 1 or definition 2 had more than one symptom subtype.

The most bothersome symptoms among men were urgency with fear of leaking, urgency urinary incontinence (UUI), and stress urinary incontinence (SUI; Table 2). In women, the most bothersome symptoms were leaking for no reason, nocturnal enuresis and post-micturition dribble. The prevalence of bother at least “quite a bit” was numerically higher in women than men for all symptoms except hesitancy.

There was a statistically significant association between the frequency and bother intensity of each symptom (P < 0.001; Table 3). The correlation coefficients for these relationships ranged from 0.504 (nocturia in men) to 0.984 (leaking during sexual activity in women). The highest correlation coefficients (≥0.9) were all observed with storage symptoms: nocturnal enuresis in men and women, leaking for no reason in women and leaking during sexual activity in women.

Table 3. Correlation between frequency of symptoms and bother intensity

	Sex
	Men (n = 2433)a	Women (n = 2751)a
	rS*	rS*
Voiding symptoms
Slow stream	0.643	0.748
Splitting	0.615	0.751
Intermittency	0.648	0.717
Hesitancy	0.724	0.764
Straining	0.789	0.917
Terminal dribble	0.623	0.711
Storage symptoms
Perceived frequency	0.576	0.652
Nocturia ≥1	0.504	0.543
Urgency	0.750	0.820
Urgency with fear of leaking	0.752	0.806
Urgency urinary incontinence	0.830	0.881
Stress urinary incontinence	0.820	0.871
Leak for no reason	0.839	0.910
Nocturnal enuresis	0.901	0.938
Leak during sexual activitya	0.734	0.984
Post-micturition symptoms
Incomplete emptying	0.801	0.862
Post-micturition dribble	0.771	0.898

• r^S Spearman correlation (from −1, perfect negative correlation to +1, perfect positive correlation).
• ^a n = 2033 (men) and 1280 (women) for leak during sexual activity.
• * P < 0.001 for all correlations.

4 DISCUSSION

This cross-sectional study, performed across major cities representing all five geographical regions of Brazil, is the first robust population-based epidemiological study of LUTS in Brazil. Significantly more women than men were affected by LUTS, with symptoms occurring less than half the time or more in 69% of men and 82% of women and symptoms occurring half the time or more affecting 40% of men and 59% of women. IPSS-defined LUTS with at least moderate symptom severity was reported by 22.5% of participants. Symptoms were commonly bothersome: for most symptoms, over half of affected individuals rated bother as at least quite a bit. Storage symptoms represented the most common ICS category in both men and women, although these symptoms tended to be more common in women than in men. Co-existence of multiple symptoms was common: with definition 1, larger numbers of men and women had multiple symptom subtypes than voiding, storage or post-micturition symptoms only.

International studies have estimated the prevalence of LUTS, according to ICS criteria, to be around 45% (adults aged ≥20 years),12 65% (adults aged ≥40 years),3 and 74% (adults aged >40 years).2 In a population-based survey, conducted by Chapple et al. in adults aged ≥40 years in China, South Korea and Taiwan, LUTS prevalence according to ICS criteria was 61%.13 Our observation of LUTS in 75% of adults aged ≥40, with more women than men affected (82% vs 69%) suggests similar LUTS prevalence in Brazil compared with other countries. However, the difference between the sexes in our study was larger than in the international studies (eg, prevalence rates of 76% and 72% in women and men, respectively, in the EpiLUTS study2).

Of the three ICS-defined categories of LUTS (voiding, storage, and post-micturition), storage symptoms represented the most prevalent category in both men and women in our study. Male LUTS, particularly voiding symptoms, are often attributed to the prostate and a supposed bladder outlet obstruction.14, 15 Our findings indicate that voiding symptoms are not the only LUTS in men, suggesting the existence of additional causes. In women, storage symptoms are generally considered as the most common category,4 but we observed at least one voiding symptom in more than one-third of women. Therefore, it appears that there is no single, dominant cause of LUTS in either men or women.

Our results are in accordance with other epidemiological studies, corroborating the current concept of LUTS not being dependent on a single disease or sex.14 A similar pattern of ICS symptom category frequencies was observed in the EPIC study, which used the same method as the current study for assessing LUTS prevalence.3 ICS symptom categories were also assessed in the EpiLUTS study, in which 66% of men and 61% of women with LUTS had symptoms from multiple categories.16 In both sexes, storage symptoms were reported in 46% and 67%, respectively, voiding symptoms were present in 57% and 48%, respectively while post-micturition symptoms (which occurred mainly in association with storage or voiding symptoms) were reported in 40% and 32.5%, respectively. In patients with multiple, co-existent LUTS categories, a single cause cannot be pinpointed and a broad, symptom-driven approach is likely to be needed for effective treatment.17

We found similar OAB prevalence between the sexes (25% in men and 24% in women). However, among those with OAB, approximately one-third of the men versus two-thirds of the women had wet OAB. Correspondingly, urgency-related symptoms were more common in women than in men. A previous Brazilian study of OAB was performed in Porto Alegre among individuals aged 15-55 years. Using the King's Health Questionnaire (validated for OAB) the prevalence of OAB was reported as 19%.8 In a study of adults aged ≥30 years in Salvador, Brazil, OAB prevalence was also lower than in our study (5.1% in men and 10% in women).4 Frequency and severity of urgency were used to define OAB, whereas our study used the OAB-V8 questionnaire, which has demonstrated high sensitivity and specificity in identifying OAB.10 In international studies, OAB has been estimated to affect 11% of men and 13% of women aged ≥18 years,3 and to occur at least sometimes in 22-24% of men and 41-42% of women aged ≥40 years.7 The Asian population-based study showed that 21% of adults aged ≥40 years have OAB,18 and the US National Overactive Bladder Evaluation (NOBLE) study reported that OAB affects 16.5% of adults aged >18 years.19 Considering differences in study populations and methods of measuring OAB, overall OAB prevalence rate in the current study appears consistent with both international studies and other Brazilian studies.

In the present study, there was no significant association between OAB prevalence and age, similar to findings reported in a study performed in Porto Alegre.8 However, international studies indicate that OAB prevalence increases with increasing age in both men and women.20

The IPSS is used globally and can be used for comparing LUTS prevalence across studies. In the Asian study, 39.8% and 43.2% of individuals aged 55-60 and >60 years, respectively, had an IPSS score ≥8.13 In adults aged >50 years in Denmark, LUTS prevalence has been estimated by this measure as 28% in men and 20% in women,21 while prevalence rates ranging between 28% and 72% were reported in men aged 40-79 years in Boxmeer (the Netherlands), Auxerre (France), Birmingham (UK), and Seoul (Republic of Korea).22 The prevalence in UK men aged >50 years has been reported as 41%,23 and in Singapore the rate among men aged ≥40 was estimated to be 16.5%.24 Despite the variability in these data, the prevalence rates we observed in adults aged ≥40 years (men, 21%; women, 24%) suggest broadly consistent rates of IPSS-defined LUTS in Brazil as in other countries. Interestingly, differences between the sexes in the prevalence of IPSS-defined LUTS were much smaller than with ICS-defined LUTS. Different questionnaires (eg, IPSS, ICS definitions) ask about LUTS in different ways and provide different response option formats. Respondent interpretations of the questions and response options might result in inconsistencies between studies, and translations into different languages can increase variability. The IPSS is the most used questionnaire for LUTS assessment, but it is limited by the inclusion of only seven questions, and the assessment of storage LUTS is particularly restricted. Moreover, the IPSS was developed prior to the development of the OAB concept.

Storage symptoms have previously been acknowledged as the most bothersome LUTS. In the population-based study in China, Taiwan, and South Korea, two out of the three symptoms causing most patients to have “quite a bit or greater” bother were related to storage (nocturia and urgency).13 Symptoms most commonly reported as bothersome in the EpiLUTS study included leaking urine during sexual activity, urgency with fear of leaking, SUI associated with physical activities, leaking for no reason, and nocturnal enuresis.2 In accordance with these data, storage-related symptoms showed greatest bother in this study.

We found statistically strong associations between the frequency and bother of each symptom, and the highest correlation coefficients were observed with symptoms of incontinence. Thus, the more frequent the occurrence of UI (including urgency incontinence, SUI, post-micturition dribble, leakage during sexual activity, leakage for no reason, and nocturnal enuresis) the more severe the bother. Similar correlations between symptoms and bother have been reported previously.2

Strengths of this study include the large sample size and the inclusion of all five geographical regions of Brazil, which have diverse climatic conditions and cultures. In addition, our study included balanced numbers of men and women. A wide variety of questions were included in the survey, and well-established, validated diagnostic tools were employed. As with all studies of this nature, limitations included self-reporting of LUTS without medical evaluation, reliance upon telephone interviews during which individuals may not always provide accurate answers, and possible selection bias because of the requirement for telephone contact. Many areas of Brazil—including all rural areas—were not included. Rural populations would have been more difficult to reach, and their exclusion should have had little impact on the results as LUTS prevalence has been reported to be similar in rural and urban populations.25

5 CONCLUSIONS

This was the first nationwide, population-based epidemiological study of LUTS to be performed in Brazil. LUTS are highly prevalent and often bothersome among men and women aged ≥40 years. Women are more likely to be affected than men, a finding which appears magnified in Brazil compared with other countries. Specific symptoms and categories of symptoms were not sex-specific, and there was a significant level of symptom overlap. Comprehensive assessment of LUTS and their effects may help ensure appropriate diagnosis and treatment in the general population.