Male Lower Urinary Tract Symptoms and Cardiovascular Events: A Systematic Review and Meta-analysis

2.1. Eligibility criteria

All studies that specifically evaluated the presence of MACE in patients with or without moderate or severe LUTS were included in the analysis. The following inclusion criteria were adopted: (1) The study reported original data and was published in a peer-reviewed journal (ie, not a meeting abstract or review article), (2) the study was a cohort study (prospective or historical cohort) consisting of male human adults or the male cohort data could be extrapolated from the paper, and (3) the authors reported the incidence or the prevalence of cardiovascular events in patients with and without LUTS.

For the definition of MACE, we used the International Classification of Primary Care (ICPC) definitions for CVDs, including K74, angina pectoris; K75, acute myocardial infarction; K76, other chronic ischemic heart disease; K77, congestive heart failure; K89, transient ischemic attack; and K90, cerebrovascular accident [12].

2.2. Information source and search strategy

An extensive search of Medline, Embase, and Cochrane was performed and included the following words: benign prostatic hyperplasia (BPH) and/or lower urinary tract symptoms (LUTS) and cardiovascular, cardio, major adverse cardiac events, MACE, heart disease, heart, myocardial infarction, myocardial, infarction, stroke, ischemic events, ischemic, cardiac death, coronary syndrome. The search through September 30, 2015, was restricted to English-language articles and studies of human participants. Bibliographies of retrieved papers were hand searched for additional references. Details of the literature search process are outlined in the flow chart. Identification of relevant abstracts, selection of studies based on the criteria described, and subsequent data extraction were performed independently by two authors (A.S., M.G.), and conflicts were resolved by a third investigator (G.C.). Full-text articles and meeting abstracts were included.

2.3. Study selection

We included all cross-sectional and longitudinal trials enrolling men and comparing the prevalence or incidence of MACE in patients with moderate to severe LUTS compared with those with mild or without LUTS [9], [10], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], and [24] (Fig. 1 and Table 1). Studies not specifically stating the occurrence or absence of cardiovascular-related events or the severity of LUTS were excluded from the analysis. In addition, articles in which only nocturia was evaluated were excluded from the analysis [25]. One study was excluded because only the OR was reported, not allowing proper data analysis.

2.4. Outcome and quality assessment

The principal outcome of this analysis was to evaluate the incidence of MACE in patients with moderate to severe LUTS in comparison to those with mild or no LUTS. A secondary outcome included the prevalence of MACE in participants with the same characteristics. The quality of the studies was assessed using the Cochrane criteria [26].

2.5. Statistical analysis

Heterogeneity of prevalence of MACE in patients with moderate to severe LUTS in comparison to those with mild or no LUTS was assessed using I² statistics. Even when low heterogeneity was detected, a random-effects model was applied because the validity of tests of heterogeneity can be limited with a small number of component studies. To estimate possible publication or disclosure bias, we used funnel plots, the Begg adjusted rank correlation test, and Egger's test. A metaregression analysis was performed to test the effect of age and associated morbidities (including diabetes and hypertension) on MACE prevalence. In addition, a linear regression analysis model weighting each study for the number of participants enrolled was carried out to verify the independent effect of diabetes on LUTS-related MACE after adjustment for age. All data were calculated using Comprehensive Meta-Analysis software version 2 (Biostat, Englewood, NJ, USA). Logistic multivariate analysis was done in SPSS for Windows 20.1 (IBM Corp, Armonk, NY, USA).

3.1. Results

Of 477 retrieved articles, 463 articles were excluded for different reasons. The flow of the meta-analysis is summarized in Fig. 1, and the characteristics of the trials included in the meta-analysis are summarized in Table 1[9], [10], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], and [24]. Thirty-one full-text articles were considered as potentially eligible for the meta-analysis; however, 17 of them were excluded because they did not include clear information about MACE or LUTS.

Among the 14 selected studies, 10 studies [9], [13], [14], [15], [16], [17], [18], [19], [20], and [21] reported the prevalence of history of MACE at enrollment in patients with moderate to severe LUTS compared with those with mild or no LUTS (Table 1). In addition, five trials longitudinally collected the incidence of MACE in participants with the same characteristics (Table 1) [10], [21], [22], [23], and [24].One of the trials [21] reported both the prevalence of MACE in the populations of interest and the incidence of MACE after the exclusion of men with a history of CVD.

Among the cross-sectional studies, 38 218 patients and 2527 MACE were included in the meta-analysis. The mean age of enrolled patients was 62.2 ± 8.0 yr. In addition, the mean prevalence of diabetes and hypertension was 18.6% and 33.8%, respectively. I² in cross-sectional trials was 79.9 (p < 0.0001). The funnel plot and Begg adjusted rank correlation test (Kendall τ: 0.0; p = 1.0) suggested no major publication bias. The presence of moderate to severe LUTS is significantly associated with an increased prevalence of reported history of MACE (OR: 2.38; 95% confidence interval [CI], 2.56–3.07; p < 0.001) (Fig. 2).

Metaregression analysis showed that the risk of MACE was lower in older patients and higher in participants with reported higher prevalence of diabetes (Fig. 3). Accordingly, a higher risk of LUTS-associated MACE was detected in those studies enrolling participants with a mean age < 60 yr at enrollment (OR: 3.40 [95% CI, 2.78–4.17] vs 1.96 [95% CI, 1.62–3.37]; Q = 15.12; p < 0.0001). A linear regression analysis model weighting each study for the number of participants enrolled was carried out to verify the independent effect of diabetes on moderate/severe LUTS–associated MACE. When age and diabetes were introduced as possible confounders and LUTS-related MACE was considered as a dependent variable, diabetes was significantly associated with MACE, even after adjustment for age (adjusted r = 0.498; p < 0.0001). Similar results were observed when trial follow-up was introduced in the model as a possible further confounder (adjusted r = 0.478; p < 0.0001). Conversely, no relationship between MACE and hypertension was observed (data not shown).

Longitudinal trials included 25 494 patients and 2291 MACE. The mean age of enrolled patients was 52.5 ± 5.5 yr, and the mean follow-up was 86.8 ± 22.1 mo. The presence of moderate to severe LUTS was associated with a significantly increased incidence of MACE compared with the rest of the sample (OR: 1.68; 95% CI, 1.13–2.50; p = 0.01) (Fig. 4).

3.2. Discussion

MACE and LUTS are health problems that are becoming more prevalent in the worldwide aging population. The association between CVD and micturition disorders has been investigated previously in white, black, and Chinese populations [19]. In the present meta-analytic systematic review, we showed that patients with moderate to severe LUTS have an increased risk of MACE, most probably because the two conditions share similar determinants. In particular, men with metabolic conditions, including diabetes, may present with alterations in vascular supply and innervation of several tissues (which may include bladder and prostate), finally leading to increased cardiovascular risk and MACE [27]. Accordingly, the present data indicated that the risk of LUTS-associated MACE increased as a function of diabetes prevalence.

In the EpiLUTS study, heart disease was independently associated with voiding symptoms combined with storage and postmicturition symptoms, suggesting that LUTS and MACE could share common pathophysiologic pathways [17]. MetS represents a constellation of pathologic conditions (hypertension, obesity, hyperglycemia, and dyslipidemia), reflecting insulin resistance and association with increased cardiovascular and metabolic risk [28]. Accordingly, in a meta-analysis of available evidence, Mottillo et al [8] reported a twofold increase in the risk of CVD in men with MetS. Interestingly, we previously reported that MetS was associated with a larger total and transitional prostatic volume, confirming a pathologic link between MetS and BPH [6]. In this review, we showed that BPH is significantly associated with higher cardiovascular risk tightly related to the prevalence of diabetes mellitus. The latter association was confirmed even when adjusted for age, suggesting that BPH is an early marker of cardiovascular risk rather than playing a direct pathogenic role in cardiovascular risk stratification. Accordingly, the risk was lower in older participants (particularly those aged >60 yr), for whom other major conventional cardiovascular risk might smooth the significance of LUTS. Our data are in line with the findings of the Boston Area Community Health (BACH) population-based epidemiological survey of urologic symptoms [18], [24], and [29], showing that, after a median follow-up of 4.8 yr, LUTS was associated with earlier diabetes onset in younger men (aged <50 yr). The association of moderate to severe LUTS with diabetes was evident only in men, not in women, and the increased risk of diabetes and heart disease occurred in obese LUTS patients, suggesting that LUTS could be considered as a sentinel marker for diabetes and heart disease.

In a recent study of MetS and prostate abnormalities in young men (mean age: 36 yr) of infertile couples, Lotti et al [30] reported that MetS was positively associated not only with biochemical (seminal interleukin 8) and ultrasound-derived signs of prostatic inflammation but also with prostate enlargement, suggesting that MetS could be a determinant of an early onset form of BPE, as a possible response of prostatic tissue to a chronic inflammation. The same authors reported that insulin levels were associated with prostate volume even after adjustment for confounders [30], suggesting that prostate overgrowth in young men may be due to a hyperinsulinemic state leading to chronic prostatic inflammation.

Although historically considered a “normal” consequence of the aging process, LUTS/BPH should be faced proactively as a preventable disorder in elderly patients. Type of diet and level of physical activity are emerging as other important factors affecting prostate health in aging men, most probably reducing risk factors such as MetS, hypogonadism, and inflammation [31] and [32]. Whether physical exercise, weight loss, and modifications of dietary habit can really alter the natural history of LUTS/BPH remains to be determined.

In a meta-analysis limited to five longitudinal trials, including 6027 men with LUTS and 18 993 men without LUTS, Bouwman et al [33] failed to find an association between LUTS severity and CVD. Some important differences in trial selection compared with the present study must be emphasized. Bouwman et al [33] excluded one trial [24] because the authors applied logistic regression analyses instead of survival analyses, whereas they included a trial targeted exclusively at nocturia [25] that was excluded from our analysis. We excluded four trials on nocturia from our review (Fig. 1) because nocturia can be caused by a variety of other factors, including aberrant behavior (fluid intake), systemic disorders (respiratory, neurogenic, endocrine, nephrologic), sleep disorders, and impaired storage function. Moreover, because lower urinary tract dysfunction is the cause of nocturia in only a minority of cases, it may be more appropriate to define nocturia as a systemic symptom rather than as LUTS [34]. Study selection, adjustment for different items, statistical methods, and different LUTS and MACE definitions could cause different outcomes between the reviews.

The findings of our review may have clinical implications (1) for prevention, by redefining the role of lifestyle (diet and physical activity) for men at risk of both urinary symptoms and heart disease; (2) for diagnosis, by suggesting screening for CVD in men with LUTS and vice versa; and (3) for treatment, by promoting a holistic approach (general practitioner, urologist, endocrinologist, cardiologist) to treat men with comorbid MetS, LUTS, and CVD. The main limitation of our review is surveillance bias, as men with LUTS or MACE might be more likely to seek health care and receive medical diagnoses. Moreover, none of the trials included in the review estimated the impact of behavioral education and medical treatment for MACE on LUTS and vice versa. Furthermore, our data were not adjusted for some relevant issues such as family history (of both LUTS and MACE), diet, smoking, physical activity, and pertinent comorbidities including depression. Finally, we were not able to include some other relevant urologic parameters, such as prostate volume, uroflowmetry, or postvoid residual volume, in the analysis.

2.1. Eligibility criteria

All studies that specifically evaluated the presence of MACE in patients with or without moderate or severe LUTS were included in the analysis. The following inclusion criteria were adopted: (1) The study reported original data and was published in a peer-reviewed journal (ie, not a meeting abstract or review article), (2) the study was a cohort study (prospective or historical cohort) consisting of male human adults or the male cohort data could be extrapolated from the paper, and (3) the authors reported the incidence or the prevalence of cardiovascular events in patients with and without LUTS.

For the definition of MACE, we used the International Classification of Primary Care (ICPC) definitions for CVDs, including K74, angina pectoris; K75, acute myocardial infarction; K76, other chronic ischemic heart disease; K77, congestive heart failure; K89, transient ischemic attack; and K90, cerebrovascular accident [12].

2.2. Information source and search strategy

An extensive search of Medline, Embase, and Cochrane was performed and included the following words: benign prostatic hyperplasia (BPH) and/or lower urinary tract symptoms (LUTS) and cardiovascular, cardio, major adverse cardiac events, MACE, heart disease, heart, myocardial infarction, myocardial, infarction, stroke, ischemic events, ischemic, cardiac death, coronary syndrome. The search through September 30, 2015, was restricted to English-language articles and studies of human participants. Bibliographies of retrieved papers were hand searched for additional references. Details of the literature search process are outlined in the flow chart. Identification of relevant abstracts, selection of studies based on the criteria described, and subsequent data extraction were performed independently by two authors (A.S., M.G.), and conflicts were resolved by a third investigator (G.C.). Full-text articles and meeting abstracts were included.

2.3. Study selection

We included all cross-sectional and longitudinal trials enrolling men and comparing the prevalence or incidence of MACE in patients with moderate to severe LUTS compared with those with mild or without LUTS [9], [10], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], and [24] (Fig. 1 and Table 1). Studies not specifically stating the occurrence or absence of cardiovascular-related events or the severity of LUTS were excluded from the analysis. In addition, articles in which only nocturia was evaluated were excluded from the analysis [25]. One study was excluded because only the OR was reported, not allowing proper data analysis.

2.4. Outcome and quality assessment

The principal outcome of this analysis was to evaluate the incidence of MACE in patients with moderate to severe LUTS in comparison to those with mild or no LUTS. A secondary outcome included the prevalence of MACE in participants with the same characteristics. The quality of the studies was assessed using the Cochrane criteria [26].

2.5. Statistical analysis

Heterogeneity of prevalence of MACE in patients with moderate to severe LUTS in comparison to those with mild or no LUTS was assessed using I² statistics. Even when low heterogeneity was detected, a random-effects model was applied because the validity of tests of heterogeneity can be limited with a small number of component studies. To estimate possible publication or disclosure bias, we used funnel plots, the Begg adjusted rank correlation test, and Egger's test. A metaregression analysis was performed to test the effect of age and associated morbidities (including diabetes and hypertension) on MACE prevalence. In addition, a linear regression analysis model weighting each study for the number of participants enrolled was carried out to verify the independent effect of diabetes on LUTS-related MACE after adjustment for age. All data were calculated using Comprehensive Meta-Analysis software version 2 (Biostat, Englewood, NJ, USA). Logistic multivariate analysis was done in SPSS for Windows 20.1 (IBM Corp, Armonk, NY, USA).

3.1. Results

Of 477 retrieved articles, 463 articles were excluded for different reasons. The flow of the meta-analysis is summarized in Fig. 1, and the characteristics of the trials included in the meta-analysis are summarized in Table 1[9], [10], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], and [24]. Thirty-one full-text articles were considered as potentially eligible for the meta-analysis; however, 17 of them were excluded because they did not include clear information about MACE or LUTS.

Among the 14 selected studies, 10 studies [9], [13], [14], [15], [16], [17], [18], [19], [20], and [21] reported the prevalence of history of MACE at enrollment in patients with moderate to severe LUTS compared with those with mild or no LUTS (Table 1). In addition, five trials longitudinally collected the incidence of MACE in participants with the same characteristics (Table 1) [10], [21], [22], [23], and [24].One of the trials [21] reported both the prevalence of MACE in the populations of interest and the incidence of MACE after the exclusion of men with a history of CVD.

Among the cross-sectional studies, 38 218 patients and 2527 MACE were included in the meta-analysis. The mean age of enrolled patients was 62.2 ± 8.0 yr. In addition, the mean prevalence of diabetes and hypertension was 18.6% and 33.8%, respectively. I² in cross-sectional trials was 79.9 (p < 0.0001). The funnel plot and Begg adjusted rank correlation test (Kendall τ: 0.0; p = 1.0) suggested no major publication bias. The presence of moderate to severe LUTS is significantly associated with an increased prevalence of reported history of MACE (OR: 2.38; 95% confidence interval [CI], 2.56–3.07; p < 0.001) (Fig. 2).

Metaregression analysis showed that the risk of MACE was lower in older patients and higher in participants with reported higher prevalence of diabetes (Fig. 3). Accordingly, a higher risk of LUTS-associated MACE was detected in those studies enrolling participants with a mean age < 60 yr at enrollment (OR: 3.40 [95% CI, 2.78–4.17] vs 1.96 [95% CI, 1.62–3.37]; Q = 15.12; p < 0.0001). A linear regression analysis model weighting each study for the number of participants enrolled was carried out to verify the independent effect of diabetes on moderate/severe LUTS–associated MACE. When age and diabetes were introduced as possible confounders and LUTS-related MACE was considered as a dependent variable, diabetes was significantly associated with MACE, even after adjustment for age (adjusted r = 0.498; p < 0.0001). Similar results were observed when trial follow-up was introduced in the model as a possible further confounder (adjusted r = 0.478; p < 0.0001). Conversely, no relationship between MACE and hypertension was observed (data not shown).

Longitudinal trials included 25 494 patients and 2291 MACE. The mean age of enrolled patients was 52.5 ± 5.5 yr, and the mean follow-up was 86.8 ± 22.1 mo. The presence of moderate to severe LUTS was associated with a significantly increased incidence of MACE compared with the rest of the sample (OR: 1.68; 95% CI, 1.13–2.50; p = 0.01) (Fig. 4).

3.2. Discussion

MACE and LUTS are health problems that are becoming more prevalent in the worldwide aging population. The association between CVD and micturition disorders has been investigated previously in white, black, and Chinese populations [19]. In the present meta-analytic systematic review, we showed that patients with moderate to severe LUTS have an increased risk of MACE, most probably because the two conditions share similar determinants. In particular, men with metabolic conditions, including diabetes, may present with alterations in vascular supply and innervation of several tissues (which may include bladder and prostate), finally leading to increased cardiovascular risk and MACE [27]. Accordingly, the present data indicated that the risk of LUTS-associated MACE increased as a function of diabetes prevalence.

In the EpiLUTS study, heart disease was independently associated with voiding symptoms combined with storage and postmicturition symptoms, suggesting that LUTS and MACE could share common pathophysiologic pathways [17]. MetS represents a constellation of pathologic conditions (hypertension, obesity, hyperglycemia, and dyslipidemia), reflecting insulin resistance and association with increased cardiovascular and metabolic risk [28]. Accordingly, in a meta-analysis of available evidence, Mottillo et al [8] reported a twofold increase in the risk of CVD in men with MetS. Interestingly, we previously reported that MetS was associated with a larger total and transitional prostatic volume, confirming a pathologic link between MetS and BPH [6]. In this review, we showed that BPH is significantly associated with higher cardiovascular risk tightly related to the prevalence of diabetes mellitus. The latter association was confirmed even when adjusted for age, suggesting that BPH is an early marker of cardiovascular risk rather than playing a direct pathogenic role in cardiovascular risk stratification. Accordingly, the risk was lower in older participants (particularly those aged >60 yr), for whom other major conventional cardiovascular risk might smooth the significance of LUTS. Our data are in line with the findings of the Boston Area Community Health (BACH) population-based epidemiological survey of urologic symptoms [18], [24], and [29], showing that, after a median follow-up of 4.8 yr, LUTS was associated with earlier diabetes onset in younger men (aged <50 yr). The association of moderate to severe LUTS with diabetes was evident only in men, not in women, and the increased risk of diabetes and heart disease occurred in obese LUTS patients, suggesting that LUTS could be considered as a sentinel marker for diabetes and heart disease.

In a recent study of MetS and prostate abnormalities in young men (mean age: 36 yr) of infertile couples, Lotti et al [30] reported that MetS was positively associated not only with biochemical (seminal interleukin 8) and ultrasound-derived signs of prostatic inflammation but also with prostate enlargement, suggesting that MetS could be a determinant of an early onset form of BPE, as a possible response of prostatic tissue to a chronic inflammation. The same authors reported that insulin levels were associated with prostate volume even after adjustment for confounders [30], suggesting that prostate overgrowth in young men may be due to a hyperinsulinemic state leading to chronic prostatic inflammation.

Although historically considered a “normal” consequence of the aging process, LUTS/BPH should be faced proactively as a preventable disorder in elderly patients. Type of diet and level of physical activity are emerging as other important factors affecting prostate health in aging men, most probably reducing risk factors such as MetS, hypogonadism, and inflammation [31] and [32]. Whether physical exercise, weight loss, and modifications of dietary habit can really alter the natural history of LUTS/BPH remains to be determined.

In a meta-analysis limited to five longitudinal trials, including 6027 men with LUTS and 18 993 men without LUTS, Bouwman et al [33] failed to find an association between LUTS severity and CVD. Some important differences in trial selection compared with the present study must be emphasized. Bouwman et al [33] excluded one trial [24] because the authors applied logistic regression analyses instead of survival analyses, whereas they included a trial targeted exclusively at nocturia [25] that was excluded from our analysis. We excluded four trials on nocturia from our review (Fig. 1) because nocturia can be caused by a variety of other factors, including aberrant behavior (fluid intake), systemic disorders (respiratory, neurogenic, endocrine, nephrologic), sleep disorders, and impaired storage function. Moreover, because lower urinary tract dysfunction is the cause of nocturia in only a minority of cases, it may be more appropriate to define nocturia as a systemic symptom rather than as LUTS [34]. Study selection, adjustment for different items, statistical methods, and different LUTS and MACE definitions could cause different outcomes between the reviews.

The findings of our review may have clinical implications (1) for prevention, by redefining the role of lifestyle (diet and physical activity) for men at risk of both urinary symptoms and heart disease; (2) for diagnosis, by suggesting screening for CVD in men with LUTS and vice versa; and (3) for treatment, by promoting a holistic approach (general practitioner, urologist, endocrinologist, cardiologist) to treat men with comorbid MetS, LUTS, and CVD. The main limitation of our review is surveillance bias, as men with LUTS or MACE might be more likely to seek health care and receive medical diagnoses. Moreover, none of the trials included in the review estimated the impact of behavioral education and medical treatment for MACE on LUTS and vice versa. Furthermore, our data were not adjusted for some relevant issues such as family history (of both LUTS and MACE), diet, smoking, physical activity, and pertinent comorbidities including depression. Finally, we were not able to include some other relevant urologic parameters, such as prostate volume, uroflowmetry, or postvoid residual volume, in the analysis.

2.1. Eligibility criteria

All studies that specifically evaluated the presence of MACE in patients with or without moderate or severe LUTS were included in the analysis. The following inclusion criteria were adopted: (1) The study reported original data and was published in a peer-reviewed journal (ie, not a meeting abstract or review article), (2) the study was a cohort study (prospective or historical cohort) consisting of male human adults or the male cohort data could be extrapolated from the paper, and (3) the authors reported the incidence or the prevalence of cardiovascular events in patients with and without LUTS.

For the definition of MACE, we used the International Classification of Primary Care (ICPC) definitions for CVDs, including K74, angina pectoris; K75, acute myocardial infarction; K76, other chronic ischemic heart disease; K77, congestive heart failure; K89, transient ischemic attack; and K90, cerebrovascular accident [12].

2.2. Information source and search strategy

An extensive search of Medline, Embase, and Cochrane was performed and included the following words: benign prostatic hyperplasia (BPH) and/or lower urinary tract symptoms (LUTS) and cardiovascular, cardio, major adverse cardiac events, MACE, heart disease, heart, myocardial infarction, myocardial, infarction, stroke, ischemic events, ischemic, cardiac death, coronary syndrome. The search through September 30, 2015, was restricted to English-language articles and studies of human participants. Bibliographies of retrieved papers were hand searched for additional references. Details of the literature search process are outlined in the flow chart. Identification of relevant abstracts, selection of studies based on the criteria described, and subsequent data extraction were performed independently by two authors (A.S., M.G.), and conflicts were resolved by a third investigator (G.C.). Full-text articles and meeting abstracts were included.

2.3. Study selection

We included all cross-sectional and longitudinal trials enrolling men and comparing the prevalence or incidence of MACE in patients with moderate to severe LUTS compared with those with mild or without LUTS [9], [10], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], and [24] (Fig. 1 and Table 1). Studies not specifically stating the occurrence or absence of cardiovascular-related events or the severity of LUTS were excluded from the analysis. In addition, articles in which only nocturia was evaluated were excluded from the analysis [25]. One study was excluded because only the OR was reported, not allowing proper data analysis.

2.4. Outcome and quality assessment

The principal outcome of this analysis was to evaluate the incidence of MACE in patients with moderate to severe LUTS in comparison to those with mild or no LUTS. A secondary outcome included the prevalence of MACE in participants with the same characteristics. The quality of the studies was assessed using the Cochrane criteria [26].

2.5. Statistical analysis

Heterogeneity of prevalence of MACE in patients with moderate to severe LUTS in comparison to those with mild or no LUTS was assessed using I² statistics. Even when low heterogeneity was detected, a random-effects model was applied because the validity of tests of heterogeneity can be limited with a small number of component studies. To estimate possible publication or disclosure bias, we used funnel plots, the Begg adjusted rank correlation test, and Egger's test. A metaregression analysis was performed to test the effect of age and associated morbidities (including diabetes and hypertension) on MACE prevalence. In addition, a linear regression analysis model weighting each study for the number of participants enrolled was carried out to verify the independent effect of diabetes on LUTS-related MACE after adjustment for age. All data were calculated using Comprehensive Meta-Analysis software version 2 (Biostat, Englewood, NJ, USA). Logistic multivariate analysis was done in SPSS for Windows 20.1 (IBM Corp, Armonk, NY, USA).

3.1. Results

Of 477 retrieved articles, 463 articles were excluded for different reasons. The flow of the meta-analysis is summarized in Fig. 1, and the characteristics of the trials included in the meta-analysis are summarized in Table 1[9], [10], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], and [24]. Thirty-one full-text articles were considered as potentially eligible for the meta-analysis; however, 17 of them were excluded because they did not include clear information about MACE or LUTS.

Among the 14 selected studies, 10 studies [9], [13], [14], [15], [16], [17], [18], [19], [20], and [21] reported the prevalence of history of MACE at enrollment in patients with moderate to severe LUTS compared with those with mild or no LUTS (Table 1). In addition, five trials longitudinally collected the incidence of MACE in participants with the same characteristics (Table 1) [10], [21], [22], [23], and [24].One of the trials [21] reported both the prevalence of MACE in the populations of interest and the incidence of MACE after the exclusion of men with a history of CVD.

Among the cross-sectional studies, 38 218 patients and 2527 MACE were included in the meta-analysis. The mean age of enrolled patients was 62.2 ± 8.0 yr. In addition, the mean prevalence of diabetes and hypertension was 18.6% and 33.8%, respectively. I² in cross-sectional trials was 79.9 (p < 0.0001). The funnel plot and Begg adjusted rank correlation test (Kendall τ: 0.0; p = 1.0) suggested no major publication bias. The presence of moderate to severe LUTS is significantly associated with an increased prevalence of reported history of MACE (OR: 2.38; 95% confidence interval [CI], 2.56–3.07; p < 0.001) (Fig. 2).

Metaregression analysis showed that the risk of MACE was lower in older patients and higher in participants with reported higher prevalence of diabetes (Fig. 3). Accordingly, a higher risk of LUTS-associated MACE was detected in those studies enrolling participants with a mean age < 60 yr at enrollment (OR: 3.40 [95% CI, 2.78–4.17] vs 1.96 [95% CI, 1.62–3.37]; Q = 15.12; p < 0.0001). A linear regression analysis model weighting each study for the number of participants enrolled was carried out to verify the independent effect of diabetes on moderate/severe LUTS–associated MACE. When age and diabetes were introduced as possible confounders and LUTS-related MACE was considered as a dependent variable, diabetes was significantly associated with MACE, even after adjustment for age (adjusted r = 0.498; p < 0.0001). Similar results were observed when trial follow-up was introduced in the model as a possible further confounder (adjusted r = 0.478; p < 0.0001). Conversely, no relationship between MACE and hypertension was observed (data not shown).

Longitudinal trials included 25 494 patients and 2291 MACE. The mean age of enrolled patients was 52.5 ± 5.5 yr, and the mean follow-up was 86.8 ± 22.1 mo. The presence of moderate to severe LUTS was associated with a significantly increased incidence of MACE compared with the rest of the sample (OR: 1.68; 95% CI, 1.13–2.50; p = 0.01) (Fig. 4).

3.2. Discussion

MACE and LUTS are health problems that are becoming more prevalent in the worldwide aging population. The association between CVD and micturition disorders has been investigated previously in white, black, and Chinese populations [19]. In the present meta-analytic systematic review, we showed that patients with moderate to severe LUTS have an increased risk of MACE, most probably because the two conditions share similar determinants. In particular, men with metabolic conditions, including diabetes, may present with alterations in vascular supply and innervation of several tissues (which may include bladder and prostate), finally leading to increased cardiovascular risk and MACE [27]. Accordingly, the present data indicated that the risk of LUTS-associated MACE increased as a function of diabetes prevalence.

In the EpiLUTS study, heart disease was independently associated with voiding symptoms combined with storage and postmicturition symptoms, suggesting that LUTS and MACE could share common pathophysiologic pathways [17]. MetS represents a constellation of pathologic conditions (hypertension, obesity, hyperglycemia, and dyslipidemia), reflecting insulin resistance and association with increased cardiovascular and metabolic risk [28]. Accordingly, in a meta-analysis of available evidence, Mottillo et al [8] reported a twofold increase in the risk of CVD in men with MetS. Interestingly, we previously reported that MetS was associated with a larger total and transitional prostatic volume, confirming a pathologic link between MetS and BPH [6]. In this review, we showed that BPH is significantly associated with higher cardiovascular risk tightly related to the prevalence of diabetes mellitus. The latter association was confirmed even when adjusted for age, suggesting that BPH is an early marker of cardiovascular risk rather than playing a direct pathogenic role in cardiovascular risk stratification. Accordingly, the risk was lower in older participants (particularly those aged >60 yr), for whom other major conventional cardiovascular risk might smooth the significance of LUTS. Our data are in line with the findings of the Boston Area Community Health (BACH) population-based epidemiological survey of urologic symptoms [18], [24], and [29], showing that, after a median follow-up of 4.8 yr, LUTS was associated with earlier diabetes onset in younger men (aged <50 yr). The association of moderate to severe LUTS with diabetes was evident only in men, not in women, and the increased risk of diabetes and heart disease occurred in obese LUTS patients, suggesting that LUTS could be considered as a sentinel marker for diabetes and heart disease.

In a recent study of MetS and prostate abnormalities in young men (mean age: 36 yr) of infertile couples, Lotti et al [30] reported that MetS was positively associated not only with biochemical (seminal interleukin 8) and ultrasound-derived signs of prostatic inflammation but also with prostate enlargement, suggesting that MetS could be a determinant of an early onset form of BPE, as a possible response of prostatic tissue to a chronic inflammation. The same authors reported that insulin levels were associated with prostate volume even after adjustment for confounders [30], suggesting that prostate overgrowth in young men may be due to a hyperinsulinemic state leading to chronic prostatic inflammation.

Although historically considered a “normal” consequence of the aging process, LUTS/BPH should be faced proactively as a preventable disorder in elderly patients. Type of diet and level of physical activity are emerging as other important factors affecting prostate health in aging men, most probably reducing risk factors such as MetS, hypogonadism, and inflammation [31] and [32]. Whether physical exercise, weight loss, and modifications of dietary habit can really alter the natural history of LUTS/BPH remains to be determined.

In a meta-analysis limited to five longitudinal trials, including 6027 men with LUTS and 18 993 men without LUTS, Bouwman et al [33] failed to find an association between LUTS severity and CVD. Some important differences in trial selection compared with the present study must be emphasized. Bouwman et al [33] excluded one trial [24] because the authors applied logistic regression analyses instead of survival analyses, whereas they included a trial targeted exclusively at nocturia [25] that was excluded from our analysis. We excluded four trials on nocturia from our review (Fig. 1) because nocturia can be caused by a variety of other factors, including aberrant behavior (fluid intake), systemic disorders (respiratory, neurogenic, endocrine, nephrologic), sleep disorders, and impaired storage function. Moreover, because lower urinary tract dysfunction is the cause of nocturia in only a minority of cases, it may be more appropriate to define nocturia as a systemic symptom rather than as LUTS [34]. Study selection, adjustment for different items, statistical methods, and different LUTS and MACE definitions could cause different outcomes between the reviews.

The findings of our review may have clinical implications (1) for prevention, by redefining the role of lifestyle (diet and physical activity) for men at risk of both urinary symptoms and heart disease; (2) for diagnosis, by suggesting screening for CVD in men with LUTS and vice versa; and (3) for treatment, by promoting a holistic approach (general practitioner, urologist, endocrinologist, cardiologist) to treat men with comorbid MetS, LUTS, and CVD. The main limitation of our review is surveillance bias, as men with LUTS or MACE might be more likely to seek health care and receive medical diagnoses. Moreover, none of the trials included in the review estimated the impact of behavioral education and medical treatment for MACE on LUTS and vice versa. Furthermore, our data were not adjusted for some relevant issues such as family history (of both LUTS and MACE), diet, smoking, physical activity, and pertinent comorbidities including depression. Finally, we were not able to include some other relevant urologic parameters, such as prostate volume, uroflowmetry, or postvoid residual volume, in the analysis.

2.1. Eligibility criteria

All studies that specifically evaluated the presence of MACE in patients with or without moderate or severe LUTS were included in the analysis. The following inclusion criteria were adopted: (1) The study reported original data and was published in a peer-reviewed journal (ie, not a meeting abstract or review article), (2) the study was a cohort study (prospective or historical cohort) consisting of male human adults or the male cohort data could be extrapolated from the paper, and (3) the authors reported the incidence or the prevalence of cardiovascular events in patients with and without LUTS.

For the definition of MACE, we used the International Classification of Primary Care (ICPC) definitions for CVDs, including K74, angina pectoris; K75, acute myocardial infarction; K76, other chronic ischemic heart disease; K77, congestive heart failure; K89, transient ischemic attack; and K90, cerebrovascular accident [12].

2.2. Information source and search strategy

An extensive search of Medline, Embase, and Cochrane was performed and included the following words: benign prostatic hyperplasia (BPH) and/or lower urinary tract symptoms (LUTS) and cardiovascular, cardio, major adverse cardiac events, MACE, heart disease, heart, myocardial infarction, myocardial, infarction, stroke, ischemic events, ischemic, cardiac death, coronary syndrome. The search through September 30, 2015, was restricted to English-language articles and studies of human participants. Bibliographies of retrieved papers were hand searched for additional references. Details of the literature search process are outlined in the flow chart. Identification of relevant abstracts, selection of studies based on the criteria described, and subsequent data extraction were performed independently by two authors (A.S., M.G.), and conflicts were resolved by a third investigator (G.C.). Full-text articles and meeting abstracts were included.

2.3. Study selection

We included all cross-sectional and longitudinal trials enrolling men and comparing the prevalence or incidence of MACE in patients with moderate to severe LUTS compared with those with mild or without LUTS [9], [10], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], and [24] (Fig. 1 and Table 1). Studies not specifically stating the occurrence or absence of cardiovascular-related events or the severity of LUTS were excluded from the analysis. In addition, articles in which only nocturia was evaluated were excluded from the analysis [25]. One study was excluded because only the OR was reported, not allowing proper data analysis.

2.4. Outcome and quality assessment

The principal outcome of this analysis was to evaluate the incidence of MACE in patients with moderate to severe LUTS in comparison to those with mild or no LUTS. A secondary outcome included the prevalence of MACE in participants with the same characteristics. The quality of the studies was assessed using the Cochrane criteria [26].

2.5. Statistical analysis

Heterogeneity of prevalence of MACE in patients with moderate to severe LUTS in comparison to those with mild or no LUTS was assessed using I² statistics. Even when low heterogeneity was detected, a random-effects model was applied because the validity of tests of heterogeneity can be limited with a small number of component studies. To estimate possible publication or disclosure bias, we used funnel plots, the Begg adjusted rank correlation test, and Egger's test. A metaregression analysis was performed to test the effect of age and associated morbidities (including diabetes and hypertension) on MACE prevalence. In addition, a linear regression analysis model weighting each study for the number of participants enrolled was carried out to verify the independent effect of diabetes on LUTS-related MACE after adjustment for age. All data were calculated using Comprehensive Meta-Analysis software version 2 (Biostat, Englewood, NJ, USA). Logistic multivariate analysis was done in SPSS for Windows 20.1 (IBM Corp, Armonk, NY, USA).

3.1. Results

Of 477 retrieved articles, 463 articles were excluded for different reasons. The flow of the meta-analysis is summarized in Fig. 1, and the characteristics of the trials included in the meta-analysis are summarized in Table 1[9], [10], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], and [24]. Thirty-one full-text articles were considered as potentially eligible for the meta-analysis; however, 17 of them were excluded because they did not include clear information about MACE or LUTS.

Among the 14 selected studies, 10 studies [9], [13], [14], [15], [16], [17], [18], [19], [20], and [21] reported the prevalence of history of MACE at enrollment in patients with moderate to severe LUTS compared with those with mild or no LUTS (Table 1). In addition, five trials longitudinally collected the incidence of MACE in participants with the same characteristics (Table 1) [10], [21], [22], [23], and [24].One of the trials [21] reported both the prevalence of MACE in the populations of interest and the incidence of MACE after the exclusion of men with a history of CVD.

Among the cross-sectional studies, 38 218 patients and 2527 MACE were included in the meta-analysis. The mean age of enrolled patients was 62.2 ± 8.0 yr. In addition, the mean prevalence of diabetes and hypertension was 18.6% and 33.8%, respectively. I² in cross-sectional trials was 79.9 (p < 0.0001). The funnel plot and Begg adjusted rank correlation test (Kendall τ: 0.0; p = 1.0) suggested no major publication bias. The presence of moderate to severe LUTS is significantly associated with an increased prevalence of reported history of MACE (OR: 2.38; 95% confidence interval [CI], 2.56–3.07; p < 0.001) (Fig. 2).

Metaregression analysis showed that the risk of MACE was lower in older patients and higher in participants with reported higher prevalence of diabetes (Fig. 3). Accordingly, a higher risk of LUTS-associated MACE was detected in those studies enrolling participants with a mean age < 60 yr at enrollment (OR: 3.40 [95% CI, 2.78–4.17] vs 1.96 [95% CI, 1.62–3.37]; Q = 15.12; p < 0.0001). A linear regression analysis model weighting each study for the number of participants enrolled was carried out to verify the independent effect of diabetes on moderate/severe LUTS–associated MACE. When age and diabetes were introduced as possible confounders and LUTS-related MACE was considered as a dependent variable, diabetes was significantly associated with MACE, even after adjustment for age (adjusted r = 0.498; p < 0.0001). Similar results were observed when trial follow-up was introduced in the model as a possible further confounder (adjusted r = 0.478; p < 0.0001). Conversely, no relationship between MACE and hypertension was observed (data not shown).

Longitudinal trials included 25 494 patients and 2291 MACE. The mean age of enrolled patients was 52.5 ± 5.5 yr, and the mean follow-up was 86.8 ± 22.1 mo. The presence of moderate to severe LUTS was associated with a significantly increased incidence of MACE compared with the rest of the sample (OR: 1.68; 95% CI, 1.13–2.50; p = 0.01) (Fig. 4).

3.2. Discussion

MACE and LUTS are health problems that are becoming more prevalent in the worldwide aging population. The association between CVD and micturition disorders has been investigated previously in white, black, and Chinese populations [19]. In the present meta-analytic systematic review, we showed that patients with moderate to severe LUTS have an increased risk of MACE, most probably because the two conditions share similar determinants. In particular, men with metabolic conditions, including diabetes, may present with alterations in vascular supply and innervation of several tissues (which may include bladder and prostate), finally leading to increased cardiovascular risk and MACE [27]. Accordingly, the present data indicated that the risk of LUTS-associated MACE increased as a function of diabetes prevalence.

In the EpiLUTS study, heart disease was independently associated with voiding symptoms combined with storage and postmicturition symptoms, suggesting that LUTS and MACE could share common pathophysiologic pathways [17]. MetS represents a constellation of pathologic conditions (hypertension, obesity, hyperglycemia, and dyslipidemia), reflecting insulin resistance and association with increased cardiovascular and metabolic risk [28]. Accordingly, in a meta-analysis of available evidence, Mottillo et al [8] reported a twofold increase in the risk of CVD in men with MetS. Interestingly, we previously reported that MetS was associated with a larger total and transitional prostatic volume, confirming a pathologic link between MetS and BPH [6]. In this review, we showed that BPH is significantly associated with higher cardiovascular risk tightly related to the prevalence of diabetes mellitus. The latter association was confirmed even when adjusted for age, suggesting that BPH is an early marker of cardiovascular risk rather than playing a direct pathogenic role in cardiovascular risk stratification. Accordingly, the risk was lower in older participants (particularly those aged >60 yr), for whom other major conventional cardiovascular risk might smooth the significance of LUTS. Our data are in line with the findings of the Boston Area Community Health (BACH) population-based epidemiological survey of urologic symptoms [18], [24], and [29], showing that, after a median follow-up of 4.8 yr, LUTS was associated with earlier diabetes onset in younger men (aged <50 yr). The association of moderate to severe LUTS with diabetes was evident only in men, not in women, and the increased risk of diabetes and heart disease occurred in obese LUTS patients, suggesting that LUTS could be considered as a sentinel marker for diabetes and heart disease.

In a recent study of MetS and prostate abnormalities in young men (mean age: 36 yr) of infertile couples, Lotti et al [30] reported that MetS was positively associated not only with biochemical (seminal interleukin 8) and ultrasound-derived signs of prostatic inflammation but also with prostate enlargement, suggesting that MetS could be a determinant of an early onset form of BPE, as a possible response of prostatic tissue to a chronic inflammation. The same authors reported that insulin levels were associated with prostate volume even after adjustment for confounders [30], suggesting that prostate overgrowth in young men may be due to a hyperinsulinemic state leading to chronic prostatic inflammation.

Although historically considered a “normal” consequence of the aging process, LUTS/BPH should be faced proactively as a preventable disorder in elderly patients. Type of diet and level of physical activity are emerging as other important factors affecting prostate health in aging men, most probably reducing risk factors such as MetS, hypogonadism, and inflammation [31] and [32]. Whether physical exercise, weight loss, and modifications of dietary habit can really alter the natural history of LUTS/BPH remains to be determined.

In a meta-analysis limited to five longitudinal trials, including 6027 men with LUTS and 18 993 men without LUTS, Bouwman et al [33] failed to find an association between LUTS severity and CVD. Some important differences in trial selection compared with the present study must be emphasized. Bouwman et al [33] excluded one trial [24] because the authors applied logistic regression analyses instead of survival analyses, whereas they included a trial targeted exclusively at nocturia [25] that was excluded from our analysis. We excluded four trials on nocturia from our review (Fig. 1) because nocturia can be caused by a variety of other factors, including aberrant behavior (fluid intake), systemic disorders (respiratory, neurogenic, endocrine, nephrologic), sleep disorders, and impaired storage function. Moreover, because lower urinary tract dysfunction is the cause of nocturia in only a minority of cases, it may be more appropriate to define nocturia as a systemic symptom rather than as LUTS [34]. Study selection, adjustment for different items, statistical methods, and different LUTS and MACE definitions could cause different outcomes between the reviews.

The findings of our review may have clinical implications (1) for prevention, by redefining the role of lifestyle (diet and physical activity) for men at risk of both urinary symptoms and heart disease; (2) for diagnosis, by suggesting screening for CVD in men with LUTS and vice versa; and (3) for treatment, by promoting a holistic approach (general practitioner, urologist, endocrinologist, cardiologist) to treat men with comorbid MetS, LUTS, and CVD. The main limitation of our review is surveillance bias, as men with LUTS or MACE might be more likely to seek health care and receive medical diagnoses. Moreover, none of the trials included in the review estimated the impact of behavioral education and medical treatment for MACE on LUTS and vice versa. Furthermore, our data were not adjusted for some relevant issues such as family history (of both LUTS and MACE), diet, smoking, physical activity, and pertinent comorbidities including depression. Finally, we were not able to include some other relevant urologic parameters, such as prostate volume, uroflowmetry, or postvoid residual volume, in the analysis.

2.1. Eligibility criteria

All studies that specifically evaluated the presence of MACE in patients with or without moderate or severe LUTS were included in the analysis. The following inclusion criteria were adopted: (1) The study reported original data and was published in a peer-reviewed journal (ie, not a meeting abstract or review article), (2) the study was a cohort study (prospective or historical cohort) consisting of male human adults or the male cohort data could be extrapolated from the paper, and (3) the authors reported the incidence or the prevalence of cardiovascular events in patients with and without LUTS.

For the definition of MACE, we used the International Classification of Primary Care (ICPC) definitions for CVDs, including K74, angina pectoris; K75, acute myocardial infarction; K76, other chronic ischemic heart disease; K77, congestive heart failure; K89, transient ischemic attack; and K90, cerebrovascular accident [12].

2.2. Information source and search strategy

An extensive search of Medline, Embase, and Cochrane was performed and included the following words: benign prostatic hyperplasia (BPH) and/or lower urinary tract symptoms (LUTS) and cardiovascular, cardio, major adverse cardiac events, MACE, heart disease, heart, myocardial infarction, myocardial, infarction, stroke, ischemic events, ischemic, cardiac death, coronary syndrome. The search through September 30, 2015, was restricted to English-language articles and studies of human participants. Bibliographies of retrieved papers were hand searched for additional references. Details of the literature search process are outlined in the flow chart. Identification of relevant abstracts, selection of studies based on the criteria described, and subsequent data extraction were performed independently by two authors (A.S., M.G.), and conflicts were resolved by a third investigator (G.C.). Full-text articles and meeting abstracts were included.

2.3. Study selection

We included all cross-sectional and longitudinal trials enrolling men and comparing the prevalence or incidence of MACE in patients with moderate to severe LUTS compared with those with mild or without LUTS [9], [10], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], and [24] (Fig. 1 and Table 1). Studies not specifically stating the occurrence or absence of cardiovascular-related events or the severity of LUTS were excluded from the analysis. In addition, articles in which only nocturia was evaluated were excluded from the analysis [25]. One study was excluded because only the OR was reported, not allowing proper data analysis.

2.4. Outcome and quality assessment

The principal outcome of this analysis was to evaluate the incidence of MACE in patients with moderate to severe LUTS in comparison to those with mild or no LUTS. A secondary outcome included the prevalence of MACE in participants with the same characteristics. The quality of the studies was assessed using the Cochrane criteria [26].

2.5. Statistical analysis

Heterogeneity of prevalence of MACE in patients with moderate to severe LUTS in comparison to those with mild or no LUTS was assessed using I² statistics. Even when low heterogeneity was detected, a random-effects model was applied because the validity of tests of heterogeneity can be limited with a small number of component studies. To estimate possible publication or disclosure bias, we used funnel plots, the Begg adjusted rank correlation test, and Egger's test. A metaregression analysis was performed to test the effect of age and associated morbidities (including diabetes and hypertension) on MACE prevalence. In addition, a linear regression analysis model weighting each study for the number of participants enrolled was carried out to verify the independent effect of diabetes on LUTS-related MACE after adjustment for age. All data were calculated using Comprehensive Meta-Analysis software version 2 (Biostat, Englewood, NJ, USA). Logistic multivariate analysis was done in SPSS for Windows 20.1 (IBM Corp, Armonk, NY, USA).

3.1. Results

Of 477 retrieved articles, 463 articles were excluded for different reasons. The flow of the meta-analysis is summarized in Fig. 1, and the characteristics of the trials included in the meta-analysis are summarized in Table 1[9], [10], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], and [24]. Thirty-one full-text articles were considered as potentially eligible for the meta-analysis; however, 17 of them were excluded because they did not include clear information about MACE or LUTS.

Among the 14 selected studies, 10 studies [9], [13], [14], [15], [16], [17], [18], [19], [20], and [21] reported the prevalence of history of MACE at enrollment in patients with moderate to severe LUTS compared with those with mild or no LUTS (Table 1). In addition, five trials longitudinally collected the incidence of MACE in participants with the same characteristics (Table 1) [10], [21], [22], [23], and [24].One of the trials [21] reported both the prevalence of MACE in the populations of interest and the incidence of MACE after the exclusion of men with a history of CVD.

Among the cross-sectional studies, 38 218 patients and 2527 MACE were included in the meta-analysis. The mean age of enrolled patients was 62.2 ± 8.0 yr. In addition, the mean prevalence of diabetes and hypertension was 18.6% and 33.8%, respectively. I² in cross-sectional trials was 79.9 (p < 0.0001). The funnel plot and Begg adjusted rank correlation test (Kendall τ: 0.0; p = 1.0) suggested no major publication bias. The presence of moderate to severe LUTS is significantly associated with an increased prevalence of reported history of MACE (OR: 2.38; 95% confidence interval [CI], 2.56–3.07; p < 0.001) (Fig. 2).

Metaregression analysis showed that the risk of MACE was lower in older patients and higher in participants with reported higher prevalence of diabetes (Fig. 3). Accordingly, a higher risk of LUTS-associated MACE was detected in those studies enrolling participants with a mean age < 60 yr at enrollment (OR: 3.40 [95% CI, 2.78–4.17] vs 1.96 [95% CI, 1.62–3.37]; Q = 15.12; p < 0.0001). A linear regression analysis model weighting each study for the number of participants enrolled was carried out to verify the independent effect of diabetes on moderate/severe LUTS–associated MACE. When age and diabetes were introduced as possible confounders and LUTS-related MACE was considered as a dependent variable, diabetes was significantly associated with MACE, even after adjustment for age (adjusted r = 0.498; p < 0.0001). Similar results were observed when trial follow-up was introduced in the model as a possible further confounder (adjusted r = 0.478; p < 0.0001). Conversely, no relationship between MACE and hypertension was observed (data not shown).

Longitudinal trials included 25 494 patients and 2291 MACE. The mean age of enrolled patients was 52.5 ± 5.5 yr, and the mean follow-up was 86.8 ± 22.1 mo. The presence of moderate to severe LUTS was associated with a significantly increased incidence of MACE compared with the rest of the sample (OR: 1.68; 95% CI, 1.13–2.50; p = 0.01) (Fig. 4).

3.2. Discussion

MACE and LUTS are health problems that are becoming more prevalent in the worldwide aging population. The association between CVD and micturition disorders has been investigated previously in white, black, and Chinese populations [19]. In the present meta-analytic systematic review, we showed that patients with moderate to severe LUTS have an increased risk of MACE, most probably because the two conditions share similar determinants. In particular, men with metabolic conditions, including diabetes, may present with alterations in vascular supply and innervation of several tissues (which may include bladder and prostate), finally leading to increased cardiovascular risk and MACE [27]. Accordingly, the present data indicated that the risk of LUTS-associated MACE increased as a function of diabetes prevalence.

In the EpiLUTS study, heart disease was independently associated with voiding symptoms combined with storage and postmicturition symptoms, suggesting that LUTS and MACE could share common pathophysiologic pathways [17]. MetS represents a constellation of pathologic conditions (hypertension, obesity, hyperglycemia, and dyslipidemia), reflecting insulin resistance and association with increased cardiovascular and metabolic risk [28]. Accordingly, in a meta-analysis of available evidence, Mottillo et al [8] reported a twofold increase in the risk of CVD in men with MetS. Interestingly, we previously reported that MetS was associated with a larger total and transitional prostatic volume, confirming a pathologic link between MetS and BPH [6]. In this review, we showed that BPH is significantly associated with higher cardiovascular risk tightly related to the prevalence of diabetes mellitus. The latter association was confirmed even when adjusted for age, suggesting that BPH is an early marker of cardiovascular risk rather than playing a direct pathogenic role in cardiovascular risk stratification. Accordingly, the risk was lower in older participants (particularly those aged >60 yr), for whom other major conventional cardiovascular risk might smooth the significance of LUTS. Our data are in line with the findings of the Boston Area Community Health (BACH) population-based epidemiological survey of urologic symptoms [18], [24], and [29], showing that, after a median follow-up of 4.8 yr, LUTS was associated with earlier diabetes onset in younger men (aged <50 yr). The association of moderate to severe LUTS with diabetes was evident only in men, not in women, and the increased risk of diabetes and heart disease occurred in obese LUTS patients, suggesting that LUTS could be considered as a sentinel marker for diabetes and heart disease.

In a recent study of MetS and prostate abnormalities in young men (mean age: 36 yr) of infertile couples, Lotti et al [30] reported that MetS was positively associated not only with biochemical (seminal interleukin 8) and ultrasound-derived signs of prostatic inflammation but also with prostate enlargement, suggesting that MetS could be a determinant of an early onset form of BPE, as a possible response of prostatic tissue to a chronic inflammation. The same authors reported that insulin levels were associated with prostate volume even after adjustment for confounders [30], suggesting that prostate overgrowth in young men may be due to a hyperinsulinemic state leading to chronic prostatic inflammation.

Although historically considered a “normal” consequence of the aging process, LUTS/BPH should be faced proactively as a preventable disorder in elderly patients. Type of diet and level of physical activity are emerging as other important factors affecting prostate health in aging men, most probably reducing risk factors such as MetS, hypogonadism, and inflammation [31] and [32]. Whether physical exercise, weight loss, and modifications of dietary habit can really alter the natural history of LUTS/BPH remains to be determined.

In a meta-analysis limited to five longitudinal trials, including 6027 men with LUTS and 18 993 men without LUTS, Bouwman et al [33] failed to find an association between LUTS severity and CVD. Some important differences in trial selection compared with the present study must be emphasized. Bouwman et al [33] excluded one trial [24] because the authors applied logistic regression analyses instead of survival analyses, whereas they included a trial targeted exclusively at nocturia [25] that was excluded from our analysis. We excluded four trials on nocturia from our review (Fig. 1) because nocturia can be caused by a variety of other factors, including aberrant behavior (fluid intake), systemic disorders (respiratory, neurogenic, endocrine, nephrologic), sleep disorders, and impaired storage function. Moreover, because lower urinary tract dysfunction is the cause of nocturia in only a minority of cases, it may be more appropriate to define nocturia as a systemic symptom rather than as LUTS [34]. Study selection, adjustment for different items, statistical methods, and different LUTS and MACE definitions could cause different outcomes between the reviews.

The findings of our review may have clinical implications (1) for prevention, by redefining the role of lifestyle (diet and physical activity) for men at risk of both urinary symptoms and heart disease; (2) for diagnosis, by suggesting screening for CVD in men with LUTS and vice versa; and (3) for treatment, by promoting a holistic approach (general practitioner, urologist, endocrinologist, cardiologist) to treat men with comorbid MetS, LUTS, and CVD. The main limitation of our review is surveillance bias, as men with LUTS or MACE might be more likely to seek health care and receive medical diagnoses. Moreover, none of the trials included in the review estimated the impact of behavioral education and medical treatment for MACE on LUTS and vice versa. Furthermore, our data were not adjusted for some relevant issues such as family history (of both LUTS and MACE), diet, smoking, physical activity, and pertinent comorbidities including depression. Finally, we were not able to include some other relevant urologic parameters, such as prostate volume, uroflowmetry, or postvoid residual volume, in the analysis.

2.1. Eligibility criteria

All studies that specifically evaluated the presence of MACE in patients with or without moderate or severe LUTS were included in the analysis. The following inclusion criteria were adopted: (1) The study reported original data and was published in a peer-reviewed journal (ie, not a meeting abstract or review article), (2) the study was a cohort study (prospective or historical cohort) consisting of male human adults or the male cohort data could be extrapolated from the paper, and (3) the authors reported the incidence or the prevalence of cardiovascular events in patients with and without LUTS.

For the definition of MACE, we used the International Classification of Primary Care (ICPC) definitions for CVDs, including K74, angina pectoris; K75, acute myocardial infarction; K76, other chronic ischemic heart disease; K77, congestive heart failure; K89, transient ischemic attack; and K90, cerebrovascular accident [12].

2.2. Information source and search strategy

An extensive search of Medline, Embase, and Cochrane was performed and included the following words: benign prostatic hyperplasia (BPH) and/or lower urinary tract symptoms (LUTS) and cardiovascular, cardio, major adverse cardiac events, MACE, heart disease, heart, myocardial infarction, myocardial, infarction, stroke, ischemic events, ischemic, cardiac death, coronary syndrome. The search through September 30, 2015, was restricted to English-language articles and studies of human participants. Bibliographies of retrieved papers were hand searched for additional references. Details of the literature search process are outlined in the flow chart. Identification of relevant abstracts, selection of studies based on the criteria described, and subsequent data extraction were performed independently by two authors (A.S., M.G.), and conflicts were resolved by a third investigator (G.C.). Full-text articles and meeting abstracts were included.

2.3. Study selection

We included all cross-sectional and longitudinal trials enrolling men and comparing the prevalence or incidence of MACE in patients with moderate to severe LUTS compared with those with mild or without LUTS [9], [10], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], and [24] (Fig. 1 and Table 1). Studies not specifically stating the occurrence or absence of cardiovascular-related events or the severity of LUTS were excluded from the analysis. In addition, articles in which only nocturia was evaluated were excluded from the analysis [25]. One study was excluded because only the OR was reported, not allowing proper data analysis.

2.4. Outcome and quality assessment

The principal outcome of this analysis was to evaluate the incidence of MACE in patients with moderate to severe LUTS in comparison to those with mild or no LUTS. A secondary outcome included the prevalence of MACE in participants with the same characteristics. The quality of the studies was assessed using the Cochrane criteria [26].

2.5. Statistical analysis

Heterogeneity of prevalence of MACE in patients with moderate to severe LUTS in comparison to those with mild or no LUTS was assessed using I² statistics. Even when low heterogeneity was detected, a random-effects model was applied because the validity of tests of heterogeneity can be limited with a small number of component studies. To estimate possible publication or disclosure bias, we used funnel plots, the Begg adjusted rank correlation test, and Egger's test. A metaregression analysis was performed to test the effect of age and associated morbidities (including diabetes and hypertension) on MACE prevalence. In addition, a linear regression analysis model weighting each study for the number of participants enrolled was carried out to verify the independent effect of diabetes on LUTS-related MACE after adjustment for age. All data were calculated using Comprehensive Meta-Analysis software version 2 (Biostat, Englewood, NJ, USA). Logistic multivariate analysis was done in SPSS for Windows 20.1 (IBM Corp, Armonk, NY, USA).

3.1. Results

Of 477 retrieved articles, 463 articles were excluded for different reasons. The flow of the meta-analysis is summarized in Fig. 1, and the characteristics of the trials included in the meta-analysis are summarized in Table 1[9], [10], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], and [24]. Thirty-one full-text articles were considered as potentially eligible for the meta-analysis; however, 17 of them were excluded because they did not include clear information about MACE or LUTS.

Among the 14 selected studies, 10 studies [9], [13], [14], [15], [16], [17], [18], [19], [20], and [21] reported the prevalence of history of MACE at enrollment in patients with moderate to severe LUTS compared with those with mild or no LUTS (Table 1). In addition, five trials longitudinally collected the incidence of MACE in participants with the same characteristics (Table 1) [10], [21], [22], [23], and [24].One of the trials [21] reported both the prevalence of MACE in the populations of interest and the incidence of MACE after the exclusion of men with a history of CVD.

Among the cross-sectional studies, 38 218 patients and 2527 MACE were included in the meta-analysis. The mean age of enrolled patients was 62.2 ± 8.0 yr. In addition, the mean prevalence of diabetes and hypertension was 18.6% and 33.8%, respectively. I² in cross-sectional trials was 79.9 (p < 0.0001). The funnel plot and Begg adjusted rank correlation test (Kendall τ: 0.0; p = 1.0) suggested no major publication bias. The presence of moderate to severe LUTS is significantly associated with an increased prevalence of reported history of MACE (OR: 2.38; 95% confidence interval [CI], 2.56–3.07; p < 0.001) (Fig. 2).

Metaregression analysis showed that the risk of MACE was lower in older patients and higher in participants with reported higher prevalence of diabetes (Fig. 3). Accordingly, a higher risk of LUTS-associated MACE was detected in those studies enrolling participants with a mean age < 60 yr at enrollment (OR: 3.40 [95% CI, 2.78–4.17] vs 1.96 [95% CI, 1.62–3.37]; Q = 15.12; p < 0.0001). A linear regression analysis model weighting each study for the number of participants enrolled was carried out to verify the independent effect of diabetes on moderate/severe LUTS–associated MACE. When age and diabetes were introduced as possible confounders and LUTS-related MACE was considered as a dependent variable, diabetes was significantly associated with MACE, even after adjustment for age (adjusted r = 0.498; p < 0.0001). Similar results were observed when trial follow-up was introduced in the model as a possible further confounder (adjusted r = 0.478; p < 0.0001). Conversely, no relationship between MACE and hypertension was observed (data not shown).

Longitudinal trials included 25 494 patients and 2291 MACE. The mean age of enrolled patients was 52.5 ± 5.5 yr, and the mean follow-up was 86.8 ± 22.1 mo. The presence of moderate to severe LUTS was associated with a significantly increased incidence of MACE compared with the rest of the sample (OR: 1.68; 95% CI, 1.13–2.50; p = 0.01) (Fig. 4).

3.2. Discussion

MACE and LUTS are health problems that are becoming more prevalent in the worldwide aging population. The association between CVD and micturition disorders has been investigated previously in white, black, and Chinese populations [19]. In the present meta-analytic systematic review, we showed that patients with moderate to severe LUTS have an increased risk of MACE, most probably because the two conditions share similar determinants. In particular, men with metabolic conditions, including diabetes, may present with alterations in vascular supply and innervation of several tissues (which may include bladder and prostate), finally leading to increased cardiovascular risk and MACE [27]. Accordingly, the present data indicated that the risk of LUTS-associated MACE increased as a function of diabetes prevalence.

In the EpiLUTS study, heart disease was independently associated with voiding symptoms combined with storage and postmicturition symptoms, suggesting that LUTS and MACE could share common pathophysiologic pathways [17]. MetS represents a constellation of pathologic conditions (hypertension, obesity, hyperglycemia, and dyslipidemia), reflecting insulin resistance and association with increased cardiovascular and metabolic risk [28]. Accordingly, in a meta-analysis of available evidence, Mottillo et al [8] reported a twofold increase in the risk of CVD in men with MetS. Interestingly, we previously reported that MetS was associated with a larger total and transitional prostatic volume, confirming a pathologic link between MetS and BPH [6]. In this review, we showed that BPH is significantly associated with higher cardiovascular risk tightly related to the prevalence of diabetes mellitus. The latter association was confirmed even when adjusted for age, suggesting that BPH is an early marker of cardiovascular risk rather than playing a direct pathogenic role in cardiovascular risk stratification. Accordingly, the risk was lower in older participants (particularly those aged >60 yr), for whom other major conventional cardiovascular risk might smooth the significance of LUTS. Our data are in line with the findings of the Boston Area Community Health (BACH) population-based epidemiological survey of urologic symptoms [18], [24], and [29], showing that, after a median follow-up of 4.8 yr, LUTS was associated with earlier diabetes onset in younger men (aged <50 yr). The association of moderate to severe LUTS with diabetes was evident only in men, not in women, and the increased risk of diabetes and heart disease occurred in obese LUTS patients, suggesting that LUTS could be considered as a sentinel marker for diabetes and heart disease.

In a recent study of MetS and prostate abnormalities in young men (mean age: 36 yr) of infertile couples, Lotti et al [30] reported that MetS was positively associated not only with biochemical (seminal interleukin 8) and ultrasound-derived signs of prostatic inflammation but also with prostate enlargement, suggesting that MetS could be a determinant of an early onset form of BPE, as a possible response of prostatic tissue to a chronic inflammation. The same authors reported that insulin levels were associated with prostate volume even after adjustment for confounders [30], suggesting that prostate overgrowth in young men may be due to a hyperinsulinemic state leading to chronic prostatic inflammation.

Although historically considered a “normal” consequence of the aging process, LUTS/BPH should be faced proactively as a preventable disorder in elderly patients. Type of diet and level of physical activity are emerging as other important factors affecting prostate health in aging men, most probably reducing risk factors such as MetS, hypogonadism, and inflammation [31] and [32]. Whether physical exercise, weight loss, and modifications of dietary habit can really alter the natural history of LUTS/BPH remains to be determined.

In a meta-analysis limited to five longitudinal trials, including 6027 men with LUTS and 18 993 men without LUTS, Bouwman et al [33] failed to find an association between LUTS severity and CVD. Some important differences in trial selection compared with the present study must be emphasized. Bouwman et al [33] excluded one trial [24] because the authors applied logistic regression analyses instead of survival analyses, whereas they included a trial targeted exclusively at nocturia [25] that was excluded from our analysis. We excluded four trials on nocturia from our review (Fig. 1) because nocturia can be caused by a variety of other factors, including aberrant behavior (fluid intake), systemic disorders (respiratory, neurogenic, endocrine, nephrologic), sleep disorders, and impaired storage function. Moreover, because lower urinary tract dysfunction is the cause of nocturia in only a minority of cases, it may be more appropriate to define nocturia as a systemic symptom rather than as LUTS [34]. Study selection, adjustment for different items, statistical methods, and different LUTS and MACE definitions could cause different outcomes between the reviews.

The findings of our review may have clinical implications (1) for prevention, by redefining the role of lifestyle (diet and physical activity) for men at risk of both urinary symptoms and heart disease; (2) for diagnosis, by suggesting screening for CVD in men with LUTS and vice versa; and (3) for treatment, by promoting a holistic approach (general practitioner, urologist, endocrinologist, cardiologist) to treat men with comorbid MetS, LUTS, and CVD. The main limitation of our review is surveillance bias, as men with LUTS or MACE might be more likely to seek health care and receive medical diagnoses. Moreover, none of the trials included in the review estimated the impact of behavioral education and medical treatment for MACE on LUTS and vice versa. Furthermore, our data were not adjusted for some relevant issues such as family history (of both LUTS and MACE), diet, smoking, physical activity, and pertinent comorbidities including depression. Finally, we were not able to include some other relevant urologic parameters, such as prostate volume, uroflowmetry, or postvoid residual volume, in the analysis.

2.1. Eligibility criteria

All studies that specifically evaluated the presence of MACE in patients with or without moderate or severe LUTS were included in the analysis. The following inclusion criteria were adopted: (1) The study reported original data and was published in a peer-reviewed journal (ie, not a meeting abstract or review article), (2) the study was a cohort study (prospective or historical cohort) consisting of male human adults or the male cohort data could be extrapolated from the paper, and (3) the authors reported the incidence or the prevalence of cardiovascular events in patients with and without LUTS.

For the definition of MACE, we used the International Classification of Primary Care (ICPC) definitions for CVDs, including K74, angina pectoris; K75, acute myocardial infarction; K76, other chronic ischemic heart disease; K77, congestive heart failure; K89, transient ischemic attack; and K90, cerebrovascular accident [12].

2.2. Information source and search strategy

An extensive search of Medline, Embase, and Cochrane was performed and included the following words: benign prostatic hyperplasia (BPH) and/or lower urinary tract symptoms (LUTS) and cardiovascular, cardio, major adverse cardiac events, MACE, heart disease, heart, myocardial infarction, myocardial, infarction, stroke, ischemic events, ischemic, cardiac death, coronary syndrome. The search through September 30, 2015, was restricted to English-language articles and studies of human participants. Bibliographies of retrieved papers were hand searched for additional references. Details of the literature search process are outlined in the flow chart. Identification of relevant abstracts, selection of studies based on the criteria described, and subsequent data extraction were performed independently by two authors (A.S., M.G.), and conflicts were resolved by a third investigator (G.C.). Full-text articles and meeting abstracts were included.

2.3. Study selection

We included all cross-sectional and longitudinal trials enrolling men and comparing the prevalence or incidence of MACE in patients with moderate to severe LUTS compared with those with mild or without LUTS [9], [10], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], and [24] (Fig. 1 and Table 1). Studies not specifically stating the occurrence or absence of cardiovascular-related events or the severity of LUTS were excluded from the analysis. In addition, articles in which only nocturia was evaluated were excluded from the analysis [25]. One study was excluded because only the OR was reported, not allowing proper data analysis.

2.4. Outcome and quality assessment

The principal outcome of this analysis was to evaluate the incidence of MACE in patients with moderate to severe LUTS in comparison to those with mild or no LUTS. A secondary outcome included the prevalence of MACE in participants with the same characteristics. The quality of the studies was assessed using the Cochrane criteria [26].

2.5. Statistical analysis

Heterogeneity of prevalence of MACE in patients with moderate to severe LUTS in comparison to those with mild or no LUTS was assessed using I² statistics. Even when low heterogeneity was detected, a random-effects model was applied because the validity of tests of heterogeneity can be limited with a small number of component studies. To estimate possible publication or disclosure bias, we used funnel plots, the Begg adjusted rank correlation test, and Egger's test. A metaregression analysis was performed to test the effect of age and associated morbidities (including diabetes and hypertension) on MACE prevalence. In addition, a linear regression analysis model weighting each study for the number of participants enrolled was carried out to verify the independent effect of diabetes on LUTS-related MACE after adjustment for age. All data were calculated using Comprehensive Meta-Analysis software version 2 (Biostat, Englewood, NJ, USA). Logistic multivariate analysis was done in SPSS for Windows 20.1 (IBM Corp, Armonk, NY, USA).

3.1. Results

Of 477 retrieved articles, 463 articles were excluded for different reasons. The flow of the meta-analysis is summarized in Fig. 1, and the characteristics of the trials included in the meta-analysis are summarized in Table 1[9], [10], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], and [24]. Thirty-one full-text articles were considered as potentially eligible for the meta-analysis; however, 17 of them were excluded because they did not include clear information about MACE or LUTS.

Among the 14 selected studies, 10 studies [9], [13], [14], [15], [16], [17], [18], [19], [20], and [21] reported the prevalence of history of MACE at enrollment in patients with moderate to severe LUTS compared with those with mild or no LUTS (Table 1). In addition, five trials longitudinally collected the incidence of MACE in participants with the same characteristics (Table 1) [10], [21], [22], [23], and [24].One of the trials [21] reported both the prevalence of MACE in the populations of interest and the incidence of MACE after the exclusion of men with a history of CVD.

Among the cross-sectional studies, 38 218 patients and 2527 MACE were included in the meta-analysis. The mean age of enrolled patients was 62.2 ± 8.0 yr. In addition, the mean prevalence of diabetes and hypertension was 18.6% and 33.8%, respectively. I² in cross-sectional trials was 79.9 (p < 0.0001). The funnel plot and Begg adjusted rank correlation test (Kendall τ: 0.0; p = 1.0) suggested no major publication bias. The presence of moderate to severe LUTS is significantly associated with an increased prevalence of reported history of MACE (OR: 2.38; 95% confidence interval [CI], 2.56–3.07; p < 0.001) (Fig. 2).

Metaregression analysis showed that the risk of MACE was lower in older patients and higher in participants with reported higher prevalence of diabetes (Fig. 3). Accordingly, a higher risk of LUTS-associated MACE was detected in those studies enrolling participants with a mean age < 60 yr at enrollment (OR: 3.40 [95% CI, 2.78–4.17] vs 1.96 [95% CI, 1.62–3.37]; Q = 15.12; p < 0.0001). A linear regression analysis model weighting each study for the number of participants enrolled was carried out to verify the independent effect of diabetes on moderate/severe LUTS–associated MACE. When age and diabetes were introduced as possible confounders and LUTS-related MACE was considered as a dependent variable, diabetes was significantly associated with MACE, even after adjustment for age (adjusted r = 0.498; p < 0.0001). Similar results were observed when trial follow-up was introduced in the model as a possible further confounder (adjusted r = 0.478; p < 0.0001). Conversely, no relationship between MACE and hypertension was observed (data not shown).

Longitudinal trials included 25 494 patients and 2291 MACE. The mean age of enrolled patients was 52.5 ± 5.5 yr, and the mean follow-up was 86.8 ± 22.1 mo. The presence of moderate to severe LUTS was associated with a significantly increased incidence of MACE compared with the rest of the sample (OR: 1.68; 95% CI, 1.13–2.50; p = 0.01) (Fig. 4).

3.2. Discussion

MACE and LUTS are health problems that are becoming more prevalent in the worldwide aging population. The association between CVD and micturition disorders has been investigated previously in white, black, and Chinese populations [19]. In the present meta-analytic systematic review, we showed that patients with moderate to severe LUTS have an increased risk of MACE, most probably because the two conditions share similar determinants. In particular, men with metabolic conditions, including diabetes, may present with alterations in vascular supply and innervation of several tissues (which may include bladder and prostate), finally leading to increased cardiovascular risk and MACE [27]. Accordingly, the present data indicated that the risk of LUTS-associated MACE increased as a function of diabetes prevalence.

In the EpiLUTS study, heart disease was independently associated with voiding symptoms combined with storage and postmicturition symptoms, suggesting that LUTS and MACE could share common pathophysiologic pathways [17]. MetS represents a constellation of pathologic conditions (hypertension, obesity, hyperglycemia, and dyslipidemia), reflecting insulin resistance and association with increased cardiovascular and metabolic risk [28]. Accordingly, in a meta-analysis of available evidence, Mottillo et al [8] reported a twofold increase in the risk of CVD in men with MetS. Interestingly, we previously reported that MetS was associated with a larger total and transitional prostatic volume, confirming a pathologic link between MetS and BPH [6]. In this review, we showed that BPH is significantly associated with higher cardiovascular risk tightly related to the prevalence of diabetes mellitus. The latter association was confirmed even when adjusted for age, suggesting that BPH is an early marker of cardiovascular risk rather than playing a direct pathogenic role in cardiovascular risk stratification. Accordingly, the risk was lower in older participants (particularly those aged >60 yr), for whom other major conventional cardiovascular risk might smooth the significance of LUTS. Our data are in line with the findings of the Boston Area Community Health (BACH) population-based epidemiological survey of urologic symptoms [18], [24], and [29], showing that, after a median follow-up of 4.8 yr, LUTS was associated with earlier diabetes onset in younger men (aged <50 yr). The association of moderate to severe LUTS with diabetes was evident only in men, not in women, and the increased risk of diabetes and heart disease occurred in obese LUTS patients, suggesting that LUTS could be considered as a sentinel marker for diabetes and heart disease.

In a recent study of MetS and prostate abnormalities in young men (mean age: 36 yr) of infertile couples, Lotti et al [30] reported that MetS was positively associated not only with biochemical (seminal interleukin 8) and ultrasound-derived signs of prostatic inflammation but also with prostate enlargement, suggesting that MetS could be a determinant of an early onset form of BPE, as a possible response of prostatic tissue to a chronic inflammation. The same authors reported that insulin levels were associated with prostate volume even after adjustment for confounders [30], suggesting that prostate overgrowth in young men may be due to a hyperinsulinemic state leading to chronic prostatic inflammation.

Although historically considered a “normal” consequence of the aging process, LUTS/BPH should be faced proactively as a preventable disorder in elderly patients. Type of diet and level of physical activity are emerging as other important factors affecting prostate health in aging men, most probably reducing risk factors such as MetS, hypogonadism, and inflammation [31] and [32]. Whether physical exercise, weight loss, and modifications of dietary habit can really alter the natural history of LUTS/BPH remains to be determined.

In a meta-analysis limited to five longitudinal trials, including 6027 men with LUTS and 18 993 men without LUTS, Bouwman et al [33] failed to find an association between LUTS severity and CVD. Some important differences in trial selection compared with the present study must be emphasized. Bouwman et al [33] excluded one trial [24] because the authors applied logistic regression analyses instead of survival analyses, whereas they included a trial targeted exclusively at nocturia [25] that was excluded from our analysis. We excluded four trials on nocturia from our review (Fig. 1) because nocturia can be caused by a variety of other factors, including aberrant behavior (fluid intake), systemic disorders (respiratory, neurogenic, endocrine, nephrologic), sleep disorders, and impaired storage function. Moreover, because lower urinary tract dysfunction is the cause of nocturia in only a minority of cases, it may be more appropriate to define nocturia as a systemic symptom rather than as LUTS [34]. Study selection, adjustment for different items, statistical methods, and different LUTS and MACE definitions could cause different outcomes between the reviews.

The findings of our review may have clinical implications (1) for prevention, by redefining the role of lifestyle (diet and physical activity) for men at risk of both urinary symptoms and heart disease; (2) for diagnosis, by suggesting screening for CVD in men with LUTS and vice versa; and (3) for treatment, by promoting a holistic approach (general practitioner, urologist, endocrinologist, cardiologist) to treat men with comorbid MetS, LUTS, and CVD. The main limitation of our review is surveillance bias, as men with LUTS or MACE might be more likely to seek health care and receive medical diagnoses. Moreover, none of the trials included in the review estimated the impact of behavioral education and medical treatment for MACE on LUTS and vice versa. Furthermore, our data were not adjusted for some relevant issues such as family history (of both LUTS and MACE), diet, smoking, physical activity, and pertinent comorbidities including depression. Finally, we were not able to include some other relevant urologic parameters, such as prostate volume, uroflowmetry, or postvoid residual volume, in the analysis.

2.1. Eligibility criteria

All studies that specifically evaluated the presence of MACE in patients with or without moderate or severe LUTS were included in the analysis. The following inclusion criteria were adopted: (1) The study reported original data and was published in a peer-reviewed journal (ie, not a meeting abstract or review article), (2) the study was a cohort study (prospective or historical cohort) consisting of male human adults or the male cohort data could be extrapolated from the paper, and (3) the authors reported the incidence or the prevalence of cardiovascular events in patients with and without LUTS.

For the definition of MACE, we used the International Classification of Primary Care (ICPC) definitions for CVDs, including K74, angina pectoris; K75, acute myocardial infarction; K76, other chronic ischemic heart disease; K77, congestive heart failure; K89, transient ischemic attack; and K90, cerebrovascular accident [12].

2.2. Information source and search strategy

An extensive search of Medline, Embase, and Cochrane was performed and included the following words: benign prostatic hyperplasia (BPH) and/or lower urinary tract symptoms (LUTS) and cardiovascular, cardio, major adverse cardiac events, MACE, heart disease, heart, myocardial infarction, myocardial, infarction, stroke, ischemic events, ischemic, cardiac death, coronary syndrome. The search through September 30, 2015, was restricted to English-language articles and studies of human participants. Bibliographies of retrieved papers were hand searched for additional references. Details of the literature search process are outlined in the flow chart. Identification of relevant abstracts, selection of studies based on the criteria described, and subsequent data extraction were performed independently by two authors (A.S., M.G.), and conflicts were resolved by a third investigator (G.C.). Full-text articles and meeting abstracts were included.

2.3. Study selection

We included all cross-sectional and longitudinal trials enrolling men and comparing the prevalence or incidence of MACE in patients with moderate to severe LUTS compared with those with mild or without LUTS [9], [10], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], and [24] (Fig. 1 and Table 1). Studies not specifically stating the occurrence or absence of cardiovascular-related events or the severity of LUTS were excluded from the analysis. In addition, articles in which only nocturia was evaluated were excluded from the analysis [25]. One study was excluded because only the OR was reported, not allowing proper data analysis.

2.4. Outcome and quality assessment

The principal outcome of this analysis was to evaluate the incidence of MACE in patients with moderate to severe LUTS in comparison to those with mild or no LUTS. A secondary outcome included the prevalence of MACE in participants with the same characteristics. The quality of the studies was assessed using the Cochrane criteria [26].

2.5. Statistical analysis

Heterogeneity of prevalence of MACE in patients with moderate to severe LUTS in comparison to those with mild or no LUTS was assessed using I² statistics. Even when low heterogeneity was detected, a random-effects model was applied because the validity of tests of heterogeneity can be limited with a small number of component studies. To estimate possible publication or disclosure bias, we used funnel plots, the Begg adjusted rank correlation test, and Egger's test. A metaregression analysis was performed to test the effect of age and associated morbidities (including diabetes and hypertension) on MACE prevalence. In addition, a linear regression analysis model weighting each study for the number of participants enrolled was carried out to verify the independent effect of diabetes on LUTS-related MACE after adjustment for age. All data were calculated using Comprehensive Meta-Analysis software version 2 (Biostat, Englewood, NJ, USA). Logistic multivariate analysis was done in SPSS for Windows 20.1 (IBM Corp, Armonk, NY, USA).

3.1. Results

Of 477 retrieved articles, 463 articles were excluded for different reasons. The flow of the meta-analysis is summarized in Fig. 1, and the characteristics of the trials included in the meta-analysis are summarized in Table 1[9], [10], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], and [24]. Thirty-one full-text articles were considered as potentially eligible for the meta-analysis; however, 17 of them were excluded because they did not include clear information about MACE or LUTS.

Among the 14 selected studies, 10 studies [9], [13], [14], [15], [16], [17], [18], [19], [20], and [21] reported the prevalence of history of MACE at enrollment in patients with moderate to severe LUTS compared with those with mild or no LUTS (Table 1). In addition, five trials longitudinally collected the incidence of MACE in participants with the same characteristics (Table 1) [10], [21], [22], [23], and [24].One of the trials [21] reported both the prevalence of MACE in the populations of interest and the incidence of MACE after the exclusion of men with a history of CVD.

Among the cross-sectional studies, 38 218 patients and 2527 MACE were included in the meta-analysis. The mean age of enrolled patients was 62.2 ± 8.0 yr. In addition, the mean prevalence of diabetes and hypertension was 18.6% and 33.8%, respectively. I² in cross-sectional trials was 79.9 (p < 0.0001). The funnel plot and Begg adjusted rank correlation test (Kendall τ: 0.0; p = 1.0) suggested no major publication bias. The presence of moderate to severe LUTS is significantly associated with an increased prevalence of reported history of MACE (OR: 2.38; 95% confidence interval [CI], 2.56–3.07; p < 0.001) (Fig. 2).

Metaregression analysis showed that the risk of MACE was lower in older patients and higher in participants with reported higher prevalence of diabetes (Fig. 3). Accordingly, a higher risk of LUTS-associated MACE was detected in those studies enrolling participants with a mean age < 60 yr at enrollment (OR: 3.40 [95% CI, 2.78–4.17] vs 1.96 [95% CI, 1.62–3.37]; Q = 15.12; p < 0.0001). A linear regression analysis model weighting each study for the number of participants enrolled was carried out to verify the independent effect of diabetes on moderate/severe LUTS–associated MACE. When age and diabetes were introduced as possible confounders and LUTS-related MACE was considered as a dependent variable, diabetes was significantly associated with MACE, even after adjustment for age (adjusted r = 0.498; p < 0.0001). Similar results were observed when trial follow-up was introduced in the model as a possible further confounder (adjusted r = 0.478; p < 0.0001). Conversely, no relationship between MACE and hypertension was observed (data not shown).

Longitudinal trials included 25 494 patients and 2291 MACE. The mean age of enrolled patients was 52.5 ± 5.5 yr, and the mean follow-up was 86.8 ± 22.1 mo. The presence of moderate to severe LUTS was associated with a significantly increased incidence of MACE compared with the rest of the sample (OR: 1.68; 95% CI, 1.13–2.50; p = 0.01) (Fig. 4).

3.2. Discussion

MACE and LUTS are health problems that are becoming more prevalent in the worldwide aging population. The association between CVD and micturition disorders has been investigated previously in white, black, and Chinese populations [19]. In the present meta-analytic systematic review, we showed that patients with moderate to severe LUTS have an increased risk of MACE, most probably because the two conditions share similar determinants. In particular, men with metabolic conditions, including diabetes, may present with alterations in vascular supply and innervation of several tissues (which may include bladder and prostate), finally leading to increased cardiovascular risk and MACE [27]. Accordingly, the present data indicated that the risk of LUTS-associated MACE increased as a function of diabetes prevalence.

In the EpiLUTS study, heart disease was independently associated with voiding symptoms combined with storage and postmicturition symptoms, suggesting that LUTS and MACE could share common pathophysiologic pathways [17]. MetS represents a constellation of pathologic conditions (hypertension, obesity, hyperglycemia, and dyslipidemia), reflecting insulin resistance and association with increased cardiovascular and metabolic risk [28]. Accordingly, in a meta-analysis of available evidence, Mottillo et al [8] reported a twofold increase in the risk of CVD in men with MetS. Interestingly, we previously reported that MetS was associated with a larger total and transitional prostatic volume, confirming a pathologic link between MetS and BPH [6]. In this review, we showed that BPH is significantly associated with higher cardiovascular risk tightly related to the prevalence of diabetes mellitus. The latter association was confirmed even when adjusted for age, suggesting that BPH is an early marker of cardiovascular risk rather than playing a direct pathogenic role in cardiovascular risk stratification. Accordingly, the risk was lower in older participants (particularly those aged >60 yr), for whom other major conventional cardiovascular risk might smooth the significance of LUTS. Our data are in line with the findings of the Boston Area Community Health (BACH) population-based epidemiological survey of urologic symptoms [18], [24], and [29], showing that, after a median follow-up of 4.8 yr, LUTS was associated with earlier diabetes onset in younger men (aged <50 yr). The association of moderate to severe LUTS with diabetes was evident only in men, not in women, and the increased risk of diabetes and heart disease occurred in obese LUTS patients, suggesting that LUTS could be considered as a sentinel marker for diabetes and heart disease.

In a recent study of MetS and prostate abnormalities in young men (mean age: 36 yr) of infertile couples, Lotti et al [30] reported that MetS was positively associated not only with biochemical (seminal interleukin 8) and ultrasound-derived signs of prostatic inflammation but also with prostate enlargement, suggesting that MetS could be a determinant of an early onset form of BPE, as a possible response of prostatic tissue to a chronic inflammation. The same authors reported that insulin levels were associated with prostate volume even after adjustment for confounders [30], suggesting that prostate overgrowth in young men may be due to a hyperinsulinemic state leading to chronic prostatic inflammation.

Although historically considered a “normal” consequence of the aging process, LUTS/BPH should be faced proactively as a preventable disorder in elderly patients. Type of diet and level of physical activity are emerging as other important factors affecting prostate health in aging men, most probably reducing risk factors such as MetS, hypogonadism, and inflammation [31] and [32]. Whether physical exercise, weight loss, and modifications of dietary habit can really alter the natural history of LUTS/BPH remains to be determined.

In a meta-analysis limited to five longitudinal trials, including 6027 men with LUTS and 18 993 men without LUTS, Bouwman et al [33] failed to find an association between LUTS severity and CVD. Some important differences in trial selection compared with the present study must be emphasized. Bouwman et al [33] excluded one trial [24] because the authors applied logistic regression analyses instead of survival analyses, whereas they included a trial targeted exclusively at nocturia [25] that was excluded from our analysis. We excluded four trials on nocturia from our review (Fig. 1) because nocturia can be caused by a variety of other factors, including aberrant behavior (fluid intake), systemic disorders (respiratory, neurogenic, endocrine, nephrologic), sleep disorders, and impaired storage function. Moreover, because lower urinary tract dysfunction is the cause of nocturia in only a minority of cases, it may be more appropriate to define nocturia as a systemic symptom rather than as LUTS [34]. Study selection, adjustment for different items, statistical methods, and different LUTS and MACE definitions could cause different outcomes between the reviews.

The findings of our review may have clinical implications (1) for prevention, by redefining the role of lifestyle (diet and physical activity) for men at risk of both urinary symptoms and heart disease; (2) for diagnosis, by suggesting screening for CVD in men with LUTS and vice versa; and (3) for treatment, by promoting a holistic approach (general practitioner, urologist, endocrinologist, cardiologist) to treat men with comorbid MetS, LUTS, and CVD. The main limitation of our review is surveillance bias, as men with LUTS or MACE might be more likely to seek health care and receive medical diagnoses. Moreover, none of the trials included in the review estimated the impact of behavioral education and medical treatment for MACE on LUTS and vice versa. Furthermore, our data were not adjusted for some relevant issues such as family history (of both LUTS and MACE), diet, smoking, physical activity, and pertinent comorbidities including depression. Finally, we were not able to include some other relevant urologic parameters, such as prostate volume, uroflowmetry, or postvoid residual volume, in the analysis.

2.1. Eligibility criteria

All studies that specifically evaluated the presence of MACE in patients with or without moderate or severe LUTS were included in the analysis. The following inclusion criteria were adopted: (1) The study reported original data and was published in a peer-reviewed journal (ie, not a meeting abstract or review article), (2) the study was a cohort study (prospective or historical cohort) consisting of male human adults or the male cohort data could be extrapolated from the paper, and (3) the authors reported the incidence or the prevalence of cardiovascular events in patients with and without LUTS.

For the definition of MACE, we used the International Classification of Primary Care (ICPC) definitions for CVDs, including K74, angina pectoris; K75, acute myocardial infarction; K76, other chronic ischemic heart disease; K77, congestive heart failure; K89, transient ischemic attack; and K90, cerebrovascular accident [12].

2.2. Information source and search strategy

An extensive search of Medline, Embase, and Cochrane was performed and included the following words: benign prostatic hyperplasia (BPH) and/or lower urinary tract symptoms (LUTS) and cardiovascular, cardio, major adverse cardiac events, MACE, heart disease, heart, myocardial infarction, myocardial, infarction, stroke, ischemic events, ischemic, cardiac death, coronary syndrome. The search through September 30, 2015, was restricted to English-language articles and studies of human participants. Bibliographies of retrieved papers were hand searched for additional references. Details of the literature search process are outlined in the flow chart. Identification of relevant abstracts, selection of studies based on the criteria described, and subsequent data extraction were performed independently by two authors (A.S., M.G.), and conflicts were resolved by a third investigator (G.C.). Full-text articles and meeting abstracts were included.

2.3. Study selection

We included all cross-sectional and longitudinal trials enrolling men and comparing the prevalence or incidence of MACE in patients with moderate to severe LUTS compared with those with mild or without LUTS [9], [10], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], and [24] (Fig. 1 and Table 1). Studies not specifically stating the occurrence or absence of cardiovascular-related events or the severity of LUTS were excluded from the analysis. In addition, articles in which only nocturia was evaluated were excluded from the analysis [25]. One study was excluded because only the OR was reported, not allowing proper data analysis.

2.4. Outcome and quality assessment

The principal outcome of this analysis was to evaluate the incidence of MACE in patients with moderate to severe LUTS in comparison to those with mild or no LUTS. A secondary outcome included the prevalence of MACE in participants with the same characteristics. The quality of the studies was assessed using the Cochrane criteria [26].

2.5. Statistical analysis

Heterogeneity of prevalence of MACE in patients with moderate to severe LUTS in comparison to those with mild or no LUTS was assessed using I² statistics. Even when low heterogeneity was detected, a random-effects model was applied because the validity of tests of heterogeneity can be limited with a small number of component studies. To estimate possible publication or disclosure bias, we used funnel plots, the Begg adjusted rank correlation test, and Egger's test. A metaregression analysis was performed to test the effect of age and associated morbidities (including diabetes and hypertension) on MACE prevalence. In addition, a linear regression analysis model weighting each study for the number of participants enrolled was carried out to verify the independent effect of diabetes on LUTS-related MACE after adjustment for age. All data were calculated using Comprehensive Meta-Analysis software version 2 (Biostat, Englewood, NJ, USA). Logistic multivariate analysis was done in SPSS for Windows 20.1 (IBM Corp, Armonk, NY, USA).

3.1. Results

Of 477 retrieved articles, 463 articles were excluded for different reasons. The flow of the meta-analysis is summarized in Fig. 1, and the characteristics of the trials included in the meta-analysis are summarized in Table 1[9], [10], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], and [24]. Thirty-one full-text articles were considered as potentially eligible for the meta-analysis; however, 17 of them were excluded because they did not include clear information about MACE or LUTS.

Among the 14 selected studies, 10 studies [9], [13], [14], [15], [16], [17], [18], [19], [20], and [21] reported the prevalence of history of MACE at enrollment in patients with moderate to severe LUTS compared with those with mild or no LUTS (Table 1). In addition, five trials longitudinally collected the incidence of MACE in participants with the same characteristics (Table 1) [10], [21], [22], [23], and [24].One of the trials [21] reported both the prevalence of MACE in the populations of interest and the incidence of MACE after the exclusion of men with a history of CVD.

Among the cross-sectional studies, 38 218 patients and 2527 MACE were included in the meta-analysis. The mean age of enrolled patients was 62.2 ± 8.0 yr. In addition, the mean prevalence of diabetes and hypertension was 18.6% and 33.8%, respectively. I² in cross-sectional trials was 79.9 (p < 0.0001). The funnel plot and Begg adjusted rank correlation test (Kendall τ: 0.0; p = 1.0) suggested no major publication bias. The presence of moderate to severe LUTS is significantly associated with an increased prevalence of reported history of MACE (OR: 2.38; 95% confidence interval [CI], 2.56–3.07; p < 0.001) (Fig. 2).

Metaregression analysis showed that the risk of MACE was lower in older patients and higher in participants with reported higher prevalence of diabetes (Fig. 3). Accordingly, a higher risk of LUTS-associated MACE was detected in those studies enrolling participants with a mean age < 60 yr at enrollment (OR: 3.40 [95% CI, 2.78–4.17] vs 1.96 [95% CI, 1.62–3.37]; Q = 15.12; p < 0.0001). A linear regression analysis model weighting each study for the number of participants enrolled was carried out to verify the independent effect of diabetes on moderate/severe LUTS–associated MACE. When age and diabetes were introduced as possible confounders and LUTS-related MACE was considered as a dependent variable, diabetes was significantly associated with MACE, even after adjustment for age (adjusted r = 0.498; p < 0.0001). Similar results were observed when trial follow-up was introduced in the model as a possible further confounder (adjusted r = 0.478; p < 0.0001). Conversely, no relationship between MACE and hypertension was observed (data not shown).

Longitudinal trials included 25 494 patients and 2291 MACE. The mean age of enrolled patients was 52.5 ± 5.5 yr, and the mean follow-up was 86.8 ± 22.1 mo. The presence of moderate to severe LUTS was associated with a significantly increased incidence of MACE compared with the rest of the sample (OR: 1.68; 95% CI, 1.13–2.50; p = 0.01) (Fig. 4).

3.2. Discussion

MACE and LUTS are health problems that are becoming more prevalent in the worldwide aging population. The association between CVD and micturition disorders has been investigated previously in white, black, and Chinese populations [19]. In the present meta-analytic systematic review, we showed that patients with moderate to severe LUTS have an increased risk of MACE, most probably because the two conditions share similar determinants. In particular, men with metabolic conditions, including diabetes, may present with alterations in vascular supply and innervation of several tissues (which may include bladder and prostate), finally leading to increased cardiovascular risk and MACE [27]. Accordingly, the present data indicated that the risk of LUTS-associated MACE increased as a function of diabetes prevalence.

In the EpiLUTS study, heart disease was independently associated with voiding symptoms combined with storage and postmicturition symptoms, suggesting that LUTS and MACE could share common pathophysiologic pathways [17]. MetS represents a constellation of pathologic conditions (hypertension, obesity, hyperglycemia, and dyslipidemia), reflecting insulin resistance and association with increased cardiovascular and metabolic risk [28]. Accordingly, in a meta-analysis of available evidence, Mottillo et al [8] reported a twofold increase in the risk of CVD in men with MetS. Interestingly, we previously reported that MetS was associated with a larger total and transitional prostatic volume, confirming a pathologic link between MetS and BPH [6]. In this review, we showed that BPH is significantly associated with higher cardiovascular risk tightly related to the prevalence of diabetes mellitus. The latter association was confirmed even when adjusted for age, suggesting that BPH is an early marker of cardiovascular risk rather than playing a direct pathogenic role in cardiovascular risk stratification. Accordingly, the risk was lower in older participants (particularly those aged >60 yr), for whom other major conventional cardiovascular risk might smooth the significance of LUTS. Our data are in line with the findings of the Boston Area Community Health (BACH) population-based epidemiological survey of urologic symptoms [18], [24], and [29], showing that, after a median follow-up of 4.8 yr, LUTS was associated with earlier diabetes onset in younger men (aged <50 yr). The association of moderate to severe LUTS with diabetes was evident only in men, not in women, and the increased risk of diabetes and heart disease occurred in obese LUTS patients, suggesting that LUTS could be considered as a sentinel marker for diabetes and heart disease.

In a recent study of MetS and prostate abnormalities in young men (mean age: 36 yr) of infertile couples, Lotti et al [30] reported that MetS was positively associated not only with biochemical (seminal interleukin 8) and ultrasound-derived signs of prostatic inflammation but also with prostate enlargement, suggesting that MetS could be a determinant of an early onset form of BPE, as a possible response of prostatic tissue to a chronic inflammation. The same authors reported that insulin levels were associated with prostate volume even after adjustment for confounders [30], suggesting that prostate overgrowth in young men may be due to a hyperinsulinemic state leading to chronic prostatic inflammation.

Although historically considered a “normal” consequence of the aging process, LUTS/BPH should be faced proactively as a preventable disorder in elderly patients. Type of diet and level of physical activity are emerging as other important factors affecting prostate health in aging men, most probably reducing risk factors such as MetS, hypogonadism, and inflammation [31] and [32]. Whether physical exercise, weight loss, and modifications of dietary habit can really alter the natural history of LUTS/BPH remains to be determined.

In a meta-analysis limited to five longitudinal trials, including 6027 men with LUTS and 18 993 men without LUTS, Bouwman et al [33] failed to find an association between LUTS severity and CVD. Some important differences in trial selection compared with the present study must be emphasized. Bouwman et al [33] excluded one trial [24] because the authors applied logistic regression analyses instead of survival analyses, whereas they included a trial targeted exclusively at nocturia [25] that was excluded from our analysis. We excluded four trials on nocturia from our review (Fig. 1) because nocturia can be caused by a variety of other factors, including aberrant behavior (fluid intake), systemic disorders (respiratory, neurogenic, endocrine, nephrologic), sleep disorders, and impaired storage function. Moreover, because lower urinary tract dysfunction is the cause of nocturia in only a minority of cases, it may be more appropriate to define nocturia as a systemic symptom rather than as LUTS [34]. Study selection, adjustment for different items, statistical methods, and different LUTS and MACE definitions could cause different outcomes between the reviews.

The findings of our review may have clinical implications (1) for prevention, by redefining the role of lifestyle (diet and physical activity) for men at risk of both urinary symptoms and heart disease; (2) for diagnosis, by suggesting screening for CVD in men with LUTS and vice versa; and (3) for treatment, by promoting a holistic approach (general practitioner, urologist, endocrinologist, cardiologist) to treat men with comorbid MetS, LUTS, and CVD. The main limitation of our review is surveillance bias, as men with LUTS or MACE might be more likely to seek health care and receive medical diagnoses. Moreover, none of the trials included in the review estimated the impact of behavioral education and medical treatment for MACE on LUTS and vice versa. Furthermore, our data were not adjusted for some relevant issues such as family history (of both LUTS and MACE), diet, smoking, physical activity, and pertinent comorbidities including depression. Finally, we were not able to include some other relevant urologic parameters, such as prostate volume, uroflowmetry, or postvoid residual volume, in the analysis.

2.1. Eligibility criteria

All studies that specifically evaluated the presence of MACE in patients with or without moderate or severe LUTS were included in the analysis. The following inclusion criteria were adopted: (1) The study reported original data and was published in a peer-reviewed journal (ie, not a meeting abstract or review article), (2) the study was a cohort study (prospective or historical cohort) consisting of male human adults or the male cohort data could be extrapolated from the paper, and (3) the authors reported the incidence or the prevalence of cardiovascular events in patients with and without LUTS.

For the definition of MACE, we used the International Classification of Primary Care (ICPC) definitions for CVDs, including K74, angina pectoris; K75, acute myocardial infarction; K76, other chronic ischemic heart disease; K77, congestive heart failure; K89, transient ischemic attack; and K90, cerebrovascular accident [12].

2.2. Information source and search strategy

An extensive search of Medline, Embase, and Cochrane was performed and included the following words: benign prostatic hyperplasia (BPH) and/or lower urinary tract symptoms (LUTS) and cardiovascular, cardio, major adverse cardiac events, MACE, heart disease, heart, myocardial infarction, myocardial, infarction, stroke, ischemic events, ischemic, cardiac death, coronary syndrome. The search through September 30, 2015, was restricted to English-language articles and studies of human participants. Bibliographies of retrieved papers were hand searched for additional references. Details of the literature search process are outlined in the flow chart. Identification of relevant abstracts, selection of studies based on the criteria described, and subsequent data extraction were performed independently by two authors (A.S., M.G.), and conflicts were resolved by a third investigator (G.C.). Full-text articles and meeting abstracts were included.

2.3. Study selection

We included all cross-sectional and longitudinal trials enrolling men and comparing the prevalence or incidence of MACE in patients with moderate to severe LUTS compared with those with mild or without LUTS [9], [10], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], and [24] (Fig. 1 and Table 1). Studies not specifically stating the occurrence or absence of cardiovascular-related events or the severity of LUTS were excluded from the analysis. In addition, articles in which only nocturia was evaluated were excluded from the analysis [25]. One study was excluded because only the OR was reported, not allowing proper data analysis.

2.4. Outcome and quality assessment

The principal outcome of this analysis was to evaluate the incidence of MACE in patients with moderate to severe LUTS in comparison to those with mild or no LUTS. A secondary outcome included the prevalence of MACE in participants with the same characteristics. The quality of the studies was assessed using the Cochrane criteria [26].

2.5. Statistical analysis

Heterogeneity of prevalence of MACE in patients with moderate to severe LUTS in comparison to those with mild or no LUTS was assessed using I² statistics. Even when low heterogeneity was detected, a random-effects model was applied because the validity of tests of heterogeneity can be limited with a small number of component studies. To estimate possible publication or disclosure bias, we used funnel plots, the Begg adjusted rank correlation test, and Egger's test. A metaregression analysis was performed to test the effect of age and associated morbidities (including diabetes and hypertension) on MACE prevalence. In addition, a linear regression analysis model weighting each study for the number of participants enrolled was carried out to verify the independent effect of diabetes on LUTS-related MACE after adjustment for age. All data were calculated using Comprehensive Meta-Analysis software version 2 (Biostat, Englewood, NJ, USA). Logistic multivariate analysis was done in SPSS for Windows 20.1 (IBM Corp, Armonk, NY, USA).

3.1. Results

Of 477 retrieved articles, 463 articles were excluded for different reasons. The flow of the meta-analysis is summarized in Fig. 1, and the characteristics of the trials included in the meta-analysis are summarized in Table 1[9], [10], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], and [24]. Thirty-one full-text articles were considered as potentially eligible for the meta-analysis; however, 17 of them were excluded because they did not include clear information about MACE or LUTS.

Among the 14 selected studies, 10 studies [9], [13], [14], [15], [16], [17], [18], [19], [20], and [21] reported the prevalence of history of MACE at enrollment in patients with moderate to severe LUTS compared with those with mild or no LUTS (Table 1). In addition, five trials longitudinally collected the incidence of MACE in participants with the same characteristics (Table 1) [10], [21], [22], [23], and [24].One of the trials [21] reported both the prevalence of MACE in the populations of interest and the incidence of MACE after the exclusion of men with a history of CVD.

Among the cross-sectional studies, 38 218 patients and 2527 MACE were included in the meta-analysis. The mean age of enrolled patients was 62.2 ± 8.0 yr. In addition, the mean prevalence of diabetes and hypertension was 18.6% and 33.8%, respectively. I² in cross-sectional trials was 79.9 (p < 0.0001). The funnel plot and Begg adjusted rank correlation test (Kendall τ: 0.0; p = 1.0) suggested no major publication bias. The presence of moderate to severe LUTS is significantly associated with an increased prevalence of reported history of MACE (OR: 2.38; 95% confidence interval [CI], 2.56–3.07; p < 0.001) (Fig. 2).

Metaregression analysis showed that the risk of MACE was lower in older patients and higher in participants with reported higher prevalence of diabetes (Fig. 3). Accordingly, a higher risk of LUTS-associated MACE was detected in those studies enrolling participants with a mean age < 60 yr at enrollment (OR: 3.40 [95% CI, 2.78–4.17] vs 1.96 [95% CI, 1.62–3.37]; Q = 15.12; p < 0.0001). A linear regression analysis model weighting each study for the number of participants enrolled was carried out to verify the independent effect of diabetes on moderate/severe LUTS–associated MACE. When age and diabetes were introduced as possible confounders and LUTS-related MACE was considered as a dependent variable, diabetes was significantly associated with MACE, even after adjustment for age (adjusted r = 0.498; p < 0.0001). Similar results were observed when trial follow-up was introduced in the model as a possible further confounder (adjusted r = 0.478; p < 0.0001). Conversely, no relationship between MACE and hypertension was observed (data not shown).

Longitudinal trials included 25 494 patients and 2291 MACE. The mean age of enrolled patients was 52.5 ± 5.5 yr, and the mean follow-up was 86.8 ± 22.1 mo. The presence of moderate to severe LUTS was associated with a significantly increased incidence of MACE compared with the rest of the sample (OR: 1.68; 95% CI, 1.13–2.50; p = 0.01) (Fig. 4).

3.2. Discussion

MACE and LUTS are health problems that are becoming more prevalent in the worldwide aging population. The association between CVD and micturition disorders has been investigated previously in white, black, and Chinese populations [19]. In the present meta-analytic systematic review, we showed that patients with moderate to severe LUTS have an increased risk of MACE, most probably because the two conditions share similar determinants. In particular, men with metabolic conditions, including diabetes, may present with alterations in vascular supply and innervation of several tissues (which may include bladder and prostate), finally leading to increased cardiovascular risk and MACE [27]. Accordingly, the present data indicated that the risk of LUTS-associated MACE increased as a function of diabetes prevalence.

In the EpiLUTS study, heart disease was independently associated with voiding symptoms combined with storage and postmicturition symptoms, suggesting that LUTS and MACE could share common pathophysiologic pathways [17]. MetS represents a constellation of pathologic conditions (hypertension, obesity, hyperglycemia, and dyslipidemia), reflecting insulin resistance and association with increased cardiovascular and metabolic risk [28]. Accordingly, in a meta-analysis of available evidence, Mottillo et al [8] reported a twofold increase in the risk of CVD in men with MetS. Interestingly, we previously reported that MetS was associated with a larger total and transitional prostatic volume, confirming a pathologic link between MetS and BPH [6]. In this review, we showed that BPH is significantly associated with higher cardiovascular risk tightly related to the prevalence of diabetes mellitus. The latter association was confirmed even when adjusted for age, suggesting that BPH is an early marker of cardiovascular risk rather than playing a direct pathogenic role in cardiovascular risk stratification. Accordingly, the risk was lower in older participants (particularly those aged >60 yr), for whom other major conventional cardiovascular risk might smooth the significance of LUTS. Our data are in line with the findings of the Boston Area Community Health (BACH) population-based epidemiological survey of urologic symptoms [18], [24], and [29], showing that, after a median follow-up of 4.8 yr, LUTS was associated with earlier diabetes onset in younger men (aged <50 yr). The association of moderate to severe LUTS with diabetes was evident only in men, not in women, and the increased risk of diabetes and heart disease occurred in obese LUTS patients, suggesting that LUTS could be considered as a sentinel marker for diabetes and heart disease.

In a recent study of MetS and prostate abnormalities in young men (mean age: 36 yr) of infertile couples, Lotti et al [30] reported that MetS was positively associated not only with biochemical (seminal interleukin 8) and ultrasound-derived signs of prostatic inflammation but also with prostate enlargement, suggesting that MetS could be a determinant of an early onset form of BPE, as a possible response of prostatic tissue to a chronic inflammation. The same authors reported that insulin levels were associated with prostate volume even after adjustment for confounders [30], suggesting that prostate overgrowth in young men may be due to a hyperinsulinemic state leading to chronic prostatic inflammation.

Although historically considered a “normal” consequence of the aging process, LUTS/BPH should be faced proactively as a preventable disorder in elderly patients. Type of diet and level of physical activity are emerging as other important factors affecting prostate health in aging men, most probably reducing risk factors such as MetS, hypogonadism, and inflammation [31] and [32]. Whether physical exercise, weight loss, and modifications of dietary habit can really alter the natural history of LUTS/BPH remains to be determined.

In a meta-analysis limited to five longitudinal trials, including 6027 men with LUTS and 18 993 men without LUTS, Bouwman et al [33] failed to find an association between LUTS severity and CVD. Some important differences in trial selection compared with the present study must be emphasized. Bouwman et al [33] excluded one trial [24] because the authors applied logistic regression analyses instead of survival analyses, whereas they included a trial targeted exclusively at nocturia [25] that was excluded from our analysis. We excluded four trials on nocturia from our review (Fig. 1) because nocturia can be caused by a variety of other factors, including aberrant behavior (fluid intake), systemic disorders (respiratory, neurogenic, endocrine, nephrologic), sleep disorders, and impaired storage function. Moreover, because lower urinary tract dysfunction is the cause of nocturia in only a minority of cases, it may be more appropriate to define nocturia as a systemic symptom rather than as LUTS [34]. Study selection, adjustment for different items, statistical methods, and different LUTS and MACE definitions could cause different outcomes between the reviews.

The findings of our review may have clinical implications (1) for prevention, by redefining the role of lifestyle (diet and physical activity) for men at risk of both urinary symptoms and heart disease; (2) for diagnosis, by suggesting screening for CVD in men with LUTS and vice versa; and (3) for treatment, by promoting a holistic approach (general practitioner, urologist, endocrinologist, cardiologist) to treat men with comorbid MetS, LUTS, and CVD. The main limitation of our review is surveillance bias, as men with LUTS or MACE might be more likely to seek health care and receive medical diagnoses. Moreover, none of the trials included in the review estimated the impact of behavioral education and medical treatment for MACE on LUTS and vice versa. Furthermore, our data were not adjusted for some relevant issues such as family history (of both LUTS and MACE), diet, smoking, physical activity, and pertinent comorbidities including depression. Finally, we were not able to include some other relevant urologic parameters, such as prostate volume, uroflowmetry, or postvoid residual volume, in the analysis.