Phosphodiesterase inhibitors for lower urinary tract symptoms consistent with benign prostatic hyperplasia

Smita Pattanaik; Ravimohan S Mavuduru; Arabind Panda; Joseph L Mathew; Mayank M Agarwal; Eu Chang Hwang; Jennifer A Lyon; Shrawan K Singh; Arup K Mandal

doi:10.1002/14651858.CD010060.pub2

Inhibidores de la fosfodiesterasa para los síntomas urinarios bajos compatibles con hiperplasia prostática benigna

Declaraciones de intereses de los autores

Versión publicada: 16 noviembre 2018 Historial de versiones

https://doi.org/10.1002/14651858.CD010060.pub2

Contraer todo Desplegar todo

Resumen

disponible en

Antecedentes

La hiperplasia prostática benigna (HPB) se refiere al aumento de volumen no maligno de la glándula prostática, que puede causar síntomas urinarios bajos (SUB) molestos. Los alfabloqueantes (AB) y los inhibidores de la 5‐alfa reductasa (5‐ARI) son la base del tratamiento médico. Recientemente, los inhibidores de la fosfodiesterasa (IFDE) que hasta el presente se han utilizado sobre todo para tratar la disfunción eréctil se introdujeron para el tratamiento de los SUB masculinos.

Objetivos

Evaluar los efectos de los IFDE en comparación con placebo y otros fármacos (AB y 5‐ARI) en hombres con SUB compatibles con HPB.

Métodos de búsqueda

Se realizó una búsqueda sistemática en el Registro Cochrane Central de Ensayos Controlados (Cochrane Central Register of Controlled Trials) (CENTRAL), MEDLINE, Embase, Web of Science, y en registros de ensayos clínicos de la World Health Organization (WHO) y en los National Institutes of Health (NIH) (actualizado 2 de agosto de 2018). Se realizó el seguimiento de citas y la búsqueda manual de resúmenes y actas de congresos. También se contactó con los autores de los estudios para solicitar información adicional.

Criterios de selección

Se consideraron para inclusión en esta revisión sistemática los ensayos controlados aleatorios (ECA) que compararon IFDE versus placebo, AB o 5‐ARI durante al menos cuatro semanas en pacientes con SUB‐HPB.

Obtención y análisis de los datos

Tres autores de la revisión, de forma independiente, examinaron la literatura y extrajeron los datos. Los resultados primarios fueron los efectos sobre los síntomas urinarios evaluados según la International Prostate Symptom Score (IPSS‐total, una puntuación que varía de 0 a 35, en la que los valores más altos indican más síntomas), las molestias urinarias evaluadas con el Benign Prostatic Hyperplasia Impact Index (BPHII, una puntuación que varía de 0 a 13, en la que los valores mayores indican más molestias) y los eventos adversos (EA). Para evaluar la calidad de la evidencia se utilizaron los criterios GRADE. Se consideraron los resultados a corto plazo (hasta 12 semanas) y a largo plazo (12 semanas o más) por separado.

Resultados principales

En esta revisión se incluyeron 16 ensayos aleatorios. Los hallazgos para los resultados primarios son los siguientes.

IFDE versus placebo: los IFDE pueden dar lugar a una mejoría pequeña en la puntuación IPSS‐total (diferencia de medias [DM] 1,89 menos; intervalo de confianza [IC] del 95%: 2,27 menos a 1,50 menos; n = 4293; evidencia de calidad baja) en comparación con placebo y pueden reducir ligeramente la puntuación BPHII (DM 0,52 menos; IC del 95%: 0,71 menos a 0,33 menos; n = 3646; evidencia de baja calidad). Las tasas de EA pueden aumentar (cociente de riesgos [CR] 1,42; IC del 95%: 1,21 a 1,67; n = 4386; evidencia de baja calidad). Este aumento corresponde a 95 EA más por 1000 participantes (IC del 95%: 47 más a 151 más por 1000). Los resultados de los estudios estuvieron limitados a una duración del tratamiento de seis a 12 semanas.

IFDE versus AB: los IFDE y los AB probablemente proporcionan una mejoría similar en la puntuación IPSS‐total (DM 0,22 más; IC del 95%: 0,49 menos a 0,93 más; n = 933; evidencia de calidad moderada) y pueden tener un efecto similar sobre la puntuación BPHII (DM 0,03 más; IC del 95%: 1,10 menos a 1,16 más; n = 550; evidencia de calidad baja) y los EA (CR 1,35; IC del 95%: 0,80 a 2,30; n = 936; evidencia de baja calidad). Este aumento corresponde a 71 EA más por 1000 participantes (IC del 95%: 41 más a 264 más por 1000). Los resultados de los estudios estuvieron limitados a una duración del tratamiento de seis a 12 semanas.

IFDE y AB versus AB solo: la combinación de IFDE y AB puede proporcionar una mejoría pequeña en la puntuación IPSS‐total (DM 2,56 menos; IC del 95%: 3,92 menos a 1,19 menos; n = 193; evidencia de calidad baja) en comparación con AB solo. No se encontró evidencia para las puntuaciones BPHII. Los EA pueden aumentar (CR 2,81; IC del 95%: 1,53 a 5,17; n = 194; evidencia de calidad moderada). Este aumento corresponde a 235 EA más por 1000 participantes (IC del 95%: 69 más a 542 más por 1000). Los resultados de los estudios estuvieron limitados a una duración del tratamiento de seis a 12 semanas.

IFDE y AB versus IFDE solo: la combinación de IFDE y AB puede proporcionar una mejoría pequeña en la IPSS‐total (DM 2,4 menos; IC del 95%: 6,47 menos a 1,67 menos; n = 40; evidencia de calidad baja) en comparación con IFDE solo. No se encontraron datos sobre BPHII ni sobre los EA. Los resultados de los estudios estuvieron limitados a una duración del tratamiento de cuatro semanas.

IFDE y 5‐ARI versus 5‐ARI solo: a corto plazo (hasta 12 semanas), la combinación de IFDE y 5‐ARI probablemente da lugar a una mejoría pequeña en la puntuación IPSS‐total (DM 1,40 menos; IC del 95%: 2,24 menos a 0,56 menos; n = 695; evidencia de calidad moderada) en comparación con 5‐ARI solo. No se encontró evidencia de las puntuaciones BPHII ni de los EA. A largo plazo (13 a 26 semanas), la combinación de IFDE y 5‐ARI probablemente resulte en una pequeña reducción de la puntuación IPSS‐total (DM 1,00 menos; IC del 95%: 1,83 menos a 0,17 menos; n = 695; evidencia de calidad moderada). No se encontró evidencia acerca de los efectos sobre las puntuaciones BPHII. Puede no haber diferencias en las tasas de EA (CR 1,07; IC del 95%: 0,84 a 1,36; n = 695; evidencia de baja calidad). Este aumento corresponde a 19 EA más por 1000 participantes (IC del 95%: 43 más a 98 más por 1000).

No se encontraron ensayos que compararan otras combinaciones de tratamientos o diferentes agentes IFDE.

Conclusiones de los autores

En comparación con placebo, los IFDE probablemente den lugar a una pequeña reducción de las puntuaciones IPSS‐total y BPHII, con un posible aumento de los EA. Puede que no haya diferencias entre IFDE y AB con respecto a la mejoría en la IPSS‐total, la BPHII ni la incidencia de EA. Tampoco parece que haya un beneficio agregado de IFDE combinado con AB, en comparación con IFDE o AB solo, o IFDE combinado con 5‐ARI en comparación con IRA solo con respecto a los síntomas urinarios. La mayor parte de la evidencia estuvo limitada al tratamiento a corto plazo hasta 12 semanas y fue de certeza moderada o baja.

PICO

Population

Intervention

Comparison

Outcome

El uso y la enseñanza del modelo PICO están muy extendidos en el ámbito de la atención sanitaria basada en la evidencia para formular preguntas y estrategias de búsqueda y para caracterizar estudios o metanálisis clínicos. PICO son las siglas en inglés de cuatro posibles componentes de una pregunta de investigación: paciente, población o problema; intervención; comparación; desenlace (outcome).

Para saber más sobre el uso del modelo PICO, puede consultar el Manual Cochrane.

Resumen en términos sencillos

disponible en

Inhibidores de la fosfodiesterasa para el tratamiento de los problemas urinarios en pacientes con hiperplasia prostática benigna

Pregunta de la revisión

¿Cuáles son los efectos de los inhibidores de la fosfodiesterasa en comparación con otros fármacos, en pacientes con síntomas urinarios, que se piensan se deben al agrandamiento de la próstata?

Antecedentes

Los trastornos relacionados con la micción, especialmente en hombres de edad avanzada, a menudo se deben a hiperplasia prostática benigna. Para tratar estos síntomas habitualmente se utilizan fármacos como los alfabloqueantes y los inhibidores de la 5‐alfa reductasa. Estos fármacos ayudan a controlar los síntomas, pero pueden tener efectos no deseados. Los inhibidores de la fosfodiesterasa son fármacos que se han utilizado durante mucho tiempo para mejorar la erección en los hombres. También pueden ayudar a mejorar los síntomas urinarios en pacientes con hiperplasia prostática benigna.

Fecha de la búsqueda

Se realizaron búsquedas de ensayos aleatorios, que estudiaron estos tratamientos, hasta el 2 de agosto de 2018.

Características de los estudios

Se encontraron 16 estudios. Incluyeron principalmente hombres, mayores de 60 años de edad, con síntomas urinarios moderados a graves. La mayoría de los estudios fueron financiados por empresas que fabrican estos fármacos.

Resultados clave

En comparación con placebo, los inhibidores de la fosfodiesterasa pueden mejorar los síntomas urinarios y reducir las molestias, pero también pueden causar más efectos no deseados.

Probablemente no haya diferencias entre los inhibidores de la fosfodiesterasa y los alfabloqueantes cuando se trata de mejorar los síntomas urinarios y puede no haber diferencias con respecto al grado de molestia de los síntomas o los efectos no deseados.

La administración de un inhibidor de la fosfodiesterasa con un alfabloqueante puede mejorar los síntomas urinarios en un grado ligeramente mayor que tomar los alfabloqueantes solos. No se encontró evidencia con respecto a las molestias urinarias. Sin embargo, el tratamiento combinado probablemente causa un mayor número de efectos no deseados.

La administración de un inhibidor de la fosfodiesterasa con un alfabloqueante puede mejorar los síntomas urinarios en un grado ligeramente mayor que tomar un inhibidor de la fosfodiesterasa solo. No se encontró evidencia con respecto a los efectos no deseados o las molestias.

A corto plazo (hasta 12 semanas), la combinación de un inhibidor de la fosfodiesterasa con un inhibidor de la 5‐alfa reductasa probablemente mejora los síntomas urinarios en un grado algo mayor que un inhibidor de la 5‐alfa reductasa solo, pero el efecto puede ser demasiado pequeño como para notarse. No se encontró evidencia sobre las molestias ni sobre las tasas de efectos no deseados. Cuando se administra por más tiempo (13 a 26 semanas), el tratamiento combinado probablemente también mejora ligeramente los síntomas urinarios, en un grado que puede no ser notable. No se encontró evidencia con respecto a las molestias urinarias. Los efectos no deseados pueden ser similares.

No se encontró evidencia de otros tratamientos combinados ni de una comparación de diferentes inhibidores de la fosfodiesterasa. La mayoría de los estudios investigaron sólo el uso a corto plazo de estos fármacos (hasta 12 semanas); por lo tanto, en gran parte se desconocen los efectos a largo plazo.

Calidad de la evidencia

La calidad de la evidencia en su mayoría se consideró moderada o baja, lo que significa que la seguridad sobre los resultados verdaderos es baja o inexistente. Los efectos verdaderos pueden ser similares o muy diferentes.

Conclusiones de los autores

disponible en

Implicaciones para la práctica

La mejor evidencia actual indica que la administración de un IFDE (tadalafil) para el tratamiento de SUB‐HPB durante un período corto de 12 semanas proporciona una mejoría relativamente muy pequeña en los síntomas urinarios y la calidad de vida en comparación con placebo, con un mayor riesgo adicional de EA. El grupo de evidencia disponible carece de información sobre si esta mejoría moderada se puede mantener por períodos más prolongados en comparación con un AB. La naturaleza y la incidencia de los EA a largo plazo con los IFDE no se conocen bien.

La eficacia y la seguridad de los IFDE pueden ser similares a los AB para los SUB‐HPB. Sin embargo, es baja la confianza en estos cálculos.

El IFDE combinado con un AB no parece agregar un beneficio en la mejoría de los síntomas urinarios, pero puede dar lugar a una mejoría pequeña cuando se combina con 5‐ARI en pacientes con hipertrofia prostática.

Implicaciones para la investigación

Se encontró evidencia de calidad moderada a baja de los resultados de eficacia y seguridad con el uso a corto plazo de un IFDE (tadalafil) en pacientes con SUB‐HPB. La calificación de la evidencia se disminuyó con frecuencia por problemas como las limitaciones de los estudios, la inconsistencia, la imprecisión clínicamente importante y la heterogeneidad, aunque no por sesgo de publicación. Los ensayos futuros deben tener mayor rigor metodológico y publicarse independientemente de la dirección de los resultados. Además, deben tener un poder estadístico adecuado, estratificarse para permitir la realización de análisis de subgrupos significativos y evaluar los fármacos durante períodos más prolongados para evaluar la eficacia así como la seguridad.

En ocho de 11 ECA incluidos en esta revisión sistemática, el grupo placebo cumplió con éxito la definición de la AUA 2010 de diferencia mínima clínicamente importante (DMCI), con una mejoría de los síntomas urinarios en más de 3 puntos en la escala IPSS‐total (única mejor medida de resultado universalmente aceptada para la evaluación y el tratamiento de los síntomas urinarios en la HPB). Por lo tanto, técnicamente, se cumplió con los criterios de la AUA de una intervención efectiva para los SUB‐HPB. La respuesta consistente al placebo según las puntuaciones IPSS‐total merece la consideración de la revisión de la DMCI para los SUB‐HPB. La DMCI para los SUB‐HPB se propuso hace más de dos décadas (Barry 1995) y ha sido respaldada por la AUA. Se pueden necesitar estudios de investigación adicionales para establecer nuevos criterios de DMCI, que deben tener en cuenta el grado de respuesta al placebo.

Debido a la preocupación creciente con respecto al uso de IFDE y la incidencia de melanoma (Deng 2018; Feng 2018; Han 2018; Lian 2016; Loeb 2015; Loeb 2017; Matthews 2016; Pottegard 2016; Tang 2017; Wang 2017; Wronski 2014), son importantes los estudios que sigan los resultados de salud a largo plazo de los pacientes que reciban IFDE para evaluar los EA menos frecuentes, aunque graves. Se deben considerar los estudios longitudinales y los registros específicos de fármacos, ya que es poco probable que los ECA tengan la duración o el poder estadístico adecuado para detectar estos EA.

Summary of findings

Open in table viewer

Summary of findings for the main comparison. PDEI compared to placebo for lower urinary tract symptoms consistent with benign prostatic hyperplasia

PDEI compared to placebo for LUTS consistent with BPH for short‐term use (up to 12 weeks) Patient: men with LUTS consistent with BPH Setting: outpatient Intervention: PDEI Comparator: placebo
Outcomes	No. of participants (studies) Follow‐up	Certainty of the evidence (GRADE)	Relative effect (95% CI)	*Anticipated absolute effects (95% CI)**
				Risk with placebo	Risk difference with PDEI
IPSS‐total score Assessed with International Prostate Symptom Score Scale from 0 (best: not at all) to 35 (worst: almost always) Follow up: range 6 to 12 weeks	4293 (11 RCTs)	⊕⊕⊕⊝ LOW^a,b	‐	Mean change in IPSS‐total scores ranged from ‐5.1 to ‐1.2	MD 1.89 lower (2.27 lower to 1.5 lower)
BPHII score Assessed with BPHII questionnaire Scale from 0 (best: none) to 13 (worst: a lot or all the time) Follow‐up: range 6 to 12 weeks	3646 (8 RCTs)	⊕⊕⊝⊝ LOW^a,c	‐	Mean change in BPHII scores ranged from ‐2 to ‐0.6	MD 0.52 lower (0.71 lower to 0.33 lower)
Adverse events Assessed with any event Follow‐up: range 6 to 12 weeks	4386 (11 RCTs)	⊕⊕⊝⊝ LOW^a,b,d	RR 1.42 (1.21 to 1.67)	Study population
				226 per 1000	95 more per 1000 (47 more to 151 more)
*The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). BPH: benign prostatic hyperplasia; BPHII: Benign Prostatic Hyperplasia Impact Index; CI: confidence interval; IPSS: International Prostate Symptom Score; LUTS: lower urinary tract symptoms; MCID: minimum clinically important difference; MD: mean difference; PDEI: phosphodiesterase inhibitor; RCT: randomised controlled trial; RR: risk ratio.
GRADE Working Group grades of evidence. High certainty: we are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect. Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect.
^aDowngraded for study limitations. ^bDowngraded for possible publication bias based on funnel plot asymmetry. ^cClinically important imprecision; 95% CI crosses MCID of 0.5. ^dClinically important heterogeneity with I² of 50%.

Open in table viewer

Summary of findings 2. PDEI compared to alpha‐blockers for LUTS associated with BPH

PDEI compared to alpha‐blockers for LUTS consistent with BPH for short‐term use (up to 12 weeks) Patient: men with LUTS consistent with BPH Setting: outpatient Intervention: PDEI Comparator: alpha‐blockers
Outcomes	No. of participants (studies) Follow‐up	Certainty of the evidence (GRADE)	Relative effect (95% CI)	*Anticipated absolute effects (95% CI)**
				Risk with alpha‐blockers	Risk difference with PDEI
IPSS‐total score Assessed with International Prostate Symptom Score Scale from 0 (best: not at all) to 35 (worst: almost always) Follow‐up: range 8 to 12 weeks	933 (4 RCTs)	⊕⊕⊕⊝ MODERATE^a	‐	Mean change in IPSS‐total scores ranged from ‐5.7 to ‐ 5.35	MD 0.22 higher (0.49 lower to 0.93 higher)
BPHII score Assessed with BPHII questionnaire Scale from 0 (best: none) to 13 (worst: a lot or all the time) Follow‐up: mean 12 weeks	550 (2 RCTs)	⊕⊕⊝⊝ LOW^a,b,c	‐	Mean change in BPHII score was ‐1.6	MD 0.03 higher (1.1 lower to 1.16 higher)
Adverse events Follow‐up: range 8 to 12 weeks	786 (4 RCTs)	⊕⊕⊝⊝ LOW^a,b,c	RR 1.35 (0.80 to 2.30)	Study population
				203 per 1000	71 more per 1000 (41 fewer to 264 more)
*The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). BPH: benign prostatic hyperplasia; BPHII: Benign Prostatic Hyperplasia Impact Index; CI: confidence interval; IPSS: International Prostate Symptom Score; LUTS: lower urinary tract symptoms; MD: mean difference; PDEI: phosphodiesterase inhibitor; RCT: randomised controlled trial; RR: risk ratio.
GRADE Working Group grades of evidence. High certainty: we are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect. Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect.
^aDowngraded for study limitations. ^bClinically important heterogeneity. ^cClinically important imprecision likely due to inconsistency.

Open in table viewer

Summary of findings 3. PDEI plus alpha‐blocker compared to alpha‐blockers alone for LUTS associated with BPH (short‐term)

Combination of PDEI with alpha‐blocker compared to alpha‐blockers alone for LUTS consistent with BPH for short‐term use (up to 12 weeks) Patient: men with LUTS consistent with BPH Setting: outpatient Intervention: PDEI + alpha‐blockers Comparator: alpha‐blockers
Outcomes	No. of participants (studies) Follow‐up	Certainty of the evidence (GRADE)	Relative effect (95% CI)	*Anticipated absolute effects (95% CI)**
				Risk with alpha‐blockers alone	Risk difference with PDEI plus alpha‐blocker
IPSS‐total score Assessed with International Prostate Symptom Score Scale from 0 (best: not at all) to 35 (worst: almost always) Follow‐up: range 4 to 12 weeks	193 (4 RCTs)	⊕⊕⊝⊝ LOW^a,b	‐	Mean change in IPSS‐total scores ranged from ‐6.7 to ‐3.62	MD 2.56 lower (3.92 lower to 1.19 lower)
BPHII score Assessed with BPHII questionnaire Scale from 0 (best: none) to 13 (worst: a lot or all the time)		‐	‐	‐	‐
Adverse effects Follow‐up: range 4 to 12 weeks	194 (4 RCTs)	⊕⊕⊕⊝ MODERATE^a	RR 2.81 (1.53 to 5.17)	Study population
				130 per 1000	235 more per 1000 (69 more to 542 more)
*The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). BPH: benign prostatic hyperplasia; BPHII: Benign Prostatic Hyperplasia Impact Index; CI: confidence interval; IPSS: International Prostate Symptom Score; LUTS: lower urinary tract symptoms; MCID: maximum clinically important difference; MD: mean difference; PDEI: phosphodiesterase inhibitor; RCT: randomised controlled trial; RR: risk ratio.
GRADE Working Group grades of evidence. High certainty: we are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect. Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect.
^aDowngraded for study limitations. ^bClinically important imprecision; 95% CI crosses MCID of 3.0

Open in table viewer

Summary of findings 4. PDEI plus an alpha‐blocker compared to PDEI alone for LUTS associated with BPH (short‐term)

Combination of PDEI with an alpha‐blocker compared to PDEI alone for LUTS consistent with BPH for short‐term use (up to 12 weeks) Patient: men with LUTS consistent with BPH Setting: outpatient Intervention: PDEI + alpha‐blockers Comparator: PDEI
Outcomes	No. of participants (studies) Follow‐up	Certainty of the evidence (GRADE)	Relative effect (95% CI)	*Anticipated absolute effects (95% CI)**
				Risk with PDEI alone	Risk difference with PDEI plus an alpha‐blocker
IPSS‐total score Assessed with International Prostate Symptom Score Scale from 0 (best: not at all) to 35 (worst: almost always) Follow‐up: mean 4 weeks	40 (1 RCT)	⊕⊕⊝⊝ LOW^a,b	‐	Mean change in IPSS‐total score was ‐3.95	MD 2.4 lower (6.47 lower to 1.67 higher)
BPHII score Assessed with BPHII questionnaire Scale from 0 (best: none) to 13 (worst: a lot or all the time)		‐	‐	‐	‐
Adverse events		‐	‐	‐
				‐	‐
*The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). BPH: benign prostatic hyperplasia; BPHII: Benign Prostatic Hyperplasia Impact Index; CI: confidence interval; IPSS: International Prostate Symptom Score; MD: mean difference; PDEI: phosphodiesterase inhibitor; RCT: randomised controlled trial.
GRADE Working Group grades of evidence. High certainty: we are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect. Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect.
^aDowngraded for study limitations. ^bDowngraded for clinically important imprecision; 95% CI crosses MCID of 3.

Open in table viewer

Summary of findings 5. PDEI + 5‐ARI compared to 5‐ARI alone for LUTS associated with BPH (short‐term)

Combination of PDEI with 5‐ARI compared to 5‐ARI alone for LUTS consistent with BPH for short‐term use (up to 12 weeks) Patient: men with LUTS consistent with BPH Setting: outpatient Intervention: PDEI + 5‐ARI Comparator: 5‐ARI
Outcomes	No. of participants (studies) Follow‐up	Certainty of the evidence (GRADE)	Relative effect (95% CI)	*Anticipated absolute effects (95% CI)**
				Risk with 5‐ARI alone	Risk difference with PDEI + 5‐ARI
IPSS‐total score Assessed with International Prostate Symptom Score Scale from 0 (best: not at all) to 35 (worst: almost always) Follow‐up: mean 12 weeks	695 (1 RCT)	⊕⊕⊕⊝ MODERATE^a,b	‐	Mean change in IPSS‐total score was ‐3.8	MD 1.4 lower (2.24 lower to 0.56 lower)
BPHII score Assessed with BPHII questionnaire Scale from 0 (best: none) to 13 (worst: a lot or all the time)		‐	‐	‐
					‐
Adverse events		‐	‐	‐
					‐
*The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). 5‐ARI: 5‐alpha reductase inhibitor; BPH: benign prostatic hyperplasia; BPHII: Benign Prostatic Hyperplasia Impact Index; CI: confidence interval; IPSS: International Prostate Symptom Score; LUTS: lower urinary tract symptoms; MD: mean difference; PDEI: phosphodiesterase inhibitor; RCT: randomised controlled trial.
GRADE Working Group grades of evidence. High certainty: we are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect. Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect.
^aDowngraded for study limitations. ^bDowngraded for clinically important imprecision.

Open in table viewer

Summary of findings 6. PDEI + 5‐ARI compared to 5‐ARI alone for LUTS associated with BPH (long‐term)

Combination of PDEI with 5‐ARI compared to 5‐ARI alone for LUTS consistent with BPH in long‐term use (26 weeks) Patient: men with LUTS consistent with BPH Setting: outpatient Intervention: PDEI + 5‐ARI Comparator: 5‐ARI
Outcomes	No. of participants (studies) Follow‐up	Certainty of the evidence (GRADE)	Relative effect (95% CI)	*Anticipated absolute effects (95% CI)**
				Risk with 5‐ARI alone	Risk difference with PDEI + 5‐ARI
IPSS‐total score Assessed with International Prostate Symptom Score Scale from 0 (best: not at all) to 35 (worst: almost always) Follow‐up: mean 26 weeks	695 (1 RCT)	⊕⊕⊕⊝ MODERATE^a,b	‐	Mean change in IPSS‐total score was ‐4.5	MD 1 lower (1.83 lower to 0.17 lower)
BPHII score Assessed with BPHII questionnaire Scale from 0 (best: none) to 13 (worst: a lot or all the time)		‐	‐

Adverse events Follow‐up: mean 26 weeks	695 (1 RCT)	⊕⊕⊝⊝ LOW^a,b	RR 1.07 (0.84 to 1.36)	Study population
				271 per 1000	19 more per 1000 (43 fewer to 98 more)
*The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). 5‐ARI: 5‐alpha reductase inhibitor; BPH: benign prostatic hyperplasia; BPHII: Benign Prostatic Hyperplasia Impact Index; CI: confidence interval; IPSS: International Prostate Symptom Score; LUTS: lower urinary tract symptoms; MD: mean difference; PDEI: phosphodiesterase inhibitor; RCT: randomised controlled trial; RR: risk ratio.
GRADE Working Group grades of evidence. High certainty: we are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect. Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect.
^aDowngraded for study limitations. ^bDowngraded for clinically important imprecision.

Antecedentes

disponible en

Descripción de la afección

La hiperplasia prostática benigna (HPB) habitualmente se presenta con síntomas urinarios bajos (SUB) (Gerraway 1991). La disfunción eréctil (DE) a menudo se asocia con SUB‐HPB y repercute negativamente en la calidad de vida de los pacientes de edad avanzada (Rosen 2003). En estudios epidemiológicos a gran escala realizados en pacientes con diversos orígenes étnicos se ha descrito una fuerte asociación entre los SUB y la DE (Braun 2000; Braun 2003; Li 2005; Martin‐Morales 2001; Mondul 2008; Rosen 2003; Rosen 2009). La presencia de factores de riesgo como la diabetes, la hipertensión, la obesidad, el síndrome metabólico, la enfermedad cardiovascular, la depresión y la cirugía pélvica anterior aumenta la prevalencia de SUB‐HPB y DE en pacientes de edad avanzada (Gerraway 1991). Sin embargo, los SUB se han propuesto como un factor de riesgo independiente de DE independiente de la edad y de otros factores de riesgo relacionados con la comorbilidad (Braun 2000; MacFarlane 1996; McVary 2005a; Ponholzer 2004; Rosen 2003). También se ha propuesto que la DE es más frecuente en los pacientes con SUB graves, y que la gravedad de ambas afecciones es a menudo congruente (McVary 2005a; McVary 2005b; Mondul 2008). Se han propuesto varias teorías para indicar una conexión biológica verosímil entre los SUB y la DE (Andersson 2011), aunque la asociación entre SUB‐HPB y DE en pacientes de edad avanzada no necesariamente implica una relación de causa‐efecto. La eficacia de los inhibidores de la fosfodiesterasa (IFDE) en el alivio de los SUB‐HPB también apoya la indicación de una fisiopatología común que afecta, de manera simultánea, los sistemas vasculares y autonómicos de las estructuras pélvicas, como la vejiga, el tejido prostático y la uretra (Guliano 2012). Los SUB‐HPB‐DE se consideran cada vez más como un complejo de enfermedades, un concepto popular en la comunidad de urólogos, con la posibilidad de tratamiento con un único fármaco (Calogero 2018; Hatzimouratidis 2014).

En las dos últimas décadas, la evidencia que surge de los estudios preclínicos indica una fisiopatología unificada y la presencia de objetivos moleculares en las estructuras anatómicas asociadas con ambas afecciones (Fibbi 2010; Filippi 2007; Hedlund 2005; Kedia 2009; Kedia 2011a; Kedia 2011b; Truss 1996; Uckert 2007; Uckert 2009). Los mecanismos fisiopatológicos comunes propuestos para los SUB‐HPB y la DE incluyen (1) la disponibilidad insuficiente de óxido nítrico (ON), (2) la hiperactividad de la Rho‐quinasa, (3) la desregulación autonómica, (4) la vasculopatía pélvica ateroesclerótica, (5) la inflamación del tejido de las vías urinarias bajas y (6) la alteración del ambiente androgénico. La menor disponibilidad del ON se podría deber a la reducción de la sintetasa del óxido nítrico endotelial (SONe) o la sintetasa del óxido nítrico neuronal (SONn) o una mayor degradación por presencia de especies de oxígeno reactivo (Barton 1997; Cai 2000). Se ha informado que la expresión de la SON y la reducción posterior del ON se reducen en la zona de transición de la próstata en la HPB, en comparación con la próstata normal (Bloch 1997). La reducción del ON se ha relacionado con la hiperplasia del músculo liso prostático, con aumento de la fuerza contráctil del músculo liso en el cuello vesical y la uretra prostática, la hipoxia de órganos pélvicos y la desregulación autonómica debido a la dilatación inadecuada de los vasa vasorum y los vasa nervorum (Guliano 2012; Orabi 2011). La hipoxia activa el sistema Rho‐A/Rho‐A de la quinasa sensible al calcio (ROCK, por sus siglas en inglés) sistema y se ha propuesto que aumenta la proliferación de músculo liso, aunque inhibe su contractilidad (Ree 2003). Se cree que muchos de estos mecanismos de convergencia causan una disfunción microvascular que altera la función endotelial y neural normal y, de ese modo, afectan la "fisiología pélvica global". Estos mecanismos se han propuesto como el denominador fisiopatológico común para los SUB y la DE (Celleck 2014). Sin embargo, se desconoce la contribución relativa de estos mecanismos propuestos de disfunción microvascular que causan SUB‐HPB y DE, aunque se ha observado una mejoría en la DE con el tratamiento de los SUB y viceversa (Fonari 2017).

Se ha demostrado que la disfunción endotelial que afecta al sistema ON/monofosfato de guanosina cíclico (GMPc) desempeña una función central en la fisiopatología de la DE. La prueba de concepto ha sido confirmada con los resultados consistentes del tratamiento con inhibidores de la fosfodiesterasa (McVary K 2006). La acción de la vía ON/GMPc en el detrusor, el cuello vesical, el tejido prostático, la uretra prostática, el tejido peniano, la uretra peniana y la vasculatura de las vías urinarias inferiores ha sido el objetivo propuesto de los IFDE en el contexto de los SUB‐HPB‐DE (Fibbi 2010; Filippi 2007; Hedlund 2005; Kedia 2009; Kedia 2011a; Kedia 2011b; Truss 1996; Uckert 2007; Uckert 2009).

Descripción de la intervención

La guía actual del tratamiento farmacológico para los SUB‐HPB incluye los bloqueantes alfa‐1 (de manera genérica denominados alfabloqueantes [AB] en esta revisión) con o sin inhibidores de la 5‐alfa reductasa (5‐ARI) (AUA 2010; EAU 2016). Los fármacos anticolinérgicos se prescriben en ocasiones para proporcionar un alivio adicional de síntomas similares a la vejiga hiperactiva (VHA).

Alfabloqueantes (AB)

Un objetivo importante del tratamiento de los SUB‐HPB incluye los receptores alfa‐1 (1A, 1B y 1D) presentes en el músculo liso del cuello vesical y el tejido prostático, que funciona como esfínter interno. Los bloqueadores alfa‐1 disminuyen el tono simpático en el músculo liso del cuello vesical y el estroma fibromuscular prostático y de este modo atenúa el componente dinámico de obstrucción y mejora el flujo urinario (Milani 2005). Mejoran la International Prostate Symptom Score total (IPSS‐total) del 30% al 40% en comparación con del 10% al 20% con placebo (Djavan 2004), y aumentan la tasa de flujo urinario máximo (Qmax) del 20% al 25%. La mejoría en la IPSS‐total es cercana al 50% y la mejoría del Qmax es de hasta el 40% en los estudios abiertos (EAU 2013). Los fármacos disponibles son: alfuzosina, doxazocina, tamsulosina, terazosina y silodosina. Estos agentes se consideran igual de efectivos y preferentemente se deben indicar los agentes más antiguos por las consideraciones de la relación entre costo y efectividad (AUA 2010; EAU 2013). Los eventos adversos (EA) debido a la administración de AB están mediados por mecanismos centrales y periféricos, e incluyen mareos, cefalea, astenia, hipotensión postural, síncope, edema periférico y eyaculación retrógrada (Djavan 1999). Las tasas de interrupción por efectos secundarios varían del 4% al 10% (Djavan 1999). Aunque se ha demostrado una mejoría satisfactoria en los SUB y el flujo urinario, con estos fármacos no se ha logrado una mejoría en la DE. Según los informes, la eficacia de estos fármacos se mantiene alrededor de cuatro años (McConnell 2003; Roehrborn 2006).

Inhibidores de la 5‐alfa reductasa (5‐ARI)

Estos fármacos inhiben la conversión de testosterona a 5‐hidroxi‐testosterona. Los 5‐ARI se utilizan en pacientes con hipertrofia de la próstata, junto con los AB para atenuar el crecimiento prostático inducido por los andrógenos. Ejercen sus efectos mediante la reducción de la masa de la próstata del 18% al 20% y de este modo actúan sobre el componente estático de la obstrucción (Djavan 2011; Shibata 2017). También se ha demostrado que reducen los niveles séricos de antígeno prostático específico (PSA) en el 50% durante seis meses a un año de tratamiento (EAU 2013). Estos agentes brindan una mejoría del 15% al 30% en la velocidad máxima del flujo urinario en comparación con del 10% al 15% con placebo (Roerborn 2002). Tienen un perfil de tolerancia variable. Los efectos secundarios principales incluyen: DE, disminución de la libido, problemas de eyaculación y ginecomastia (Andriole 2003), que se explica por la acción farmacológica del fármaco.

Los AB y los 5‐ARI solos, o en combinación, son en cierto modo satisfactorios para el alivio de los SUB, pero no producen mejoría y pueden provocar el empeoramiento de la disfunción sexual masculina coexistente (Brookes 2002). También se informó que los pacientes con tratamiento combinado han seguido el tratamiento médico para la HPB por más tiempo que los que recibieron el tratamiento con un solo fármaco (Rensing 2017).

Inhibidores de la fosfodiesterasa (IFDE)

Las fosfodiesterasas (FDE) comprenden un grupo de 11 enzimas (FDE₁‐₁₁) que se expresan en diversos tejidos. Aumentan el nivel intracelular de monofosfato de adenosina cíclico (AMPc) o monofosfato de guanosina cíclico (GMPc) mediante la reducción de su degradación. La FDE₅ es la enzima prevalente en la vasculatura peniana y otros tejidos pélvicos. Los inhibidores de la FDE₅ no tienen propiedades vasodilatadoras directas, pero aumentan el ON liberado en la unión neuromuscular (Bischoff 2004). La Food and Drug Administration (FDA) de los EE.UU. ha aprobado cuatro agentes IFDE, sildenafil (Viagra 1988), tadalafil (Cialis 2003), vardenafil (Levitra 2003; Staxyn 2010) y avanafil (Stendra 2012), para el tratamiento de la DE, pero ha aprobado sólo dos agentes para el uso en la hipertensión arterial pulmonar: sildenafil (según se describe en Revatio 1998) y tadalafil (según se ha informado en Adicirca 2009). Las dosis aprobadas de los fármacos para el uso en la DE "según y cuándo se necesite" son: sildenafil 25, 50, o 100 mg; tadalafil 10 o 20 mg; vardenafil 10 o 20 mg; y avanafil 25, 50 o 100 mg antes del coito. Estos fármacos parecen tener una eficacia equivalente en el tratamiento de la DE; por lo tanto, el efecto de clase está bien establecido (Corona 2011). Difieren en el perfil de inhibición de las isoenzimas FDE (y en consecuencia el perfil de efectos secundarios) y en la duración de la acción (Lue 2007). El tadalafil es superior a otros debido a la duración más prolongada de la acción (t₁/₂ = 17,5 horas; duración total de la acción = 36 horas) y causa menos efectos secundarios visuales (700 veces más selectividad para inhibir la FDE₅ que FDE₆ presente en la retina) y musculoesqueléticos (inhibición mínima de DE₁₁) (Meherotra 2007; Rezvanfar 2014). El tadalafil ha sido aprobado por la FDA (en octubre 2011) para el tratamiento de los SUB‐HPB con o sin DE, a una dosis de 2,5 o 5 mg diarios (FDA 2011).

De qué manera podría funcionar la intervención

Truss 1996 informó por primera vez de la presencia de FDE₅ en el músculo detrusor con métodos cromatográficos. Los estudios de inmunomarcación e immunohistoquímicos han revelado que la FDE se expresa en el músculo detrusor, en los vasos pélvicos que irrigan este músculo liso, en el músculo liso vascular del tejido prostático y en la uretra prostática (Ho 1998; Uckert 2006; Uckert 2011). Se ha demostrado que PDE₅ y PDE₁₁ se expresan en regiones glandulares y subglandulares de la próstata en humanos (Fibbi 2010). Gacci 2011 y Oger 2010 encontraron que la expresión de FDE₅‐ARN fue mucho menor en el tejido prostático que en el detrusor y el tejido peniano. El GMPc y el óxido nítrico son neurotransmisores inhibitorios importantes en el músculo liso de la uretra, y se ha demostrado una fuerte inmunorreactividad de la FDE₅ en las células del músculo liso de la uretra prostática y no prostática (Fibbi 2010; Kedia 2008). Para la función del detrusor vesical, el AMPc parece más importante que el GMPc, pero se ha observado inmunorreactividad de la FDE₅ en las células musculares endoteliales y lisas de la vasculatura y en las células del músculo liso de los haces musculares (Filippi 2007). Los experimentos tisulares aislados han demostrado el efecto relajante de la FDE₅ en el cuello vesical (Bittencourt 2009; Guiliano 2013b; Werkstrom 2009). Se ha demostrado la presencia de FDE₅ en la vasculatura de vías urinarias inferiores, localizado en células del músculo liso de las arterias más grandes, células endoteliales que revisten la luz arterial y el endotelio de las arterias más pequeñas (Fibbi 2010). Los FDE se encuentran en diversos sitios de las vías urinarias inferiores, como la vejiga (cúpula vesical que comprende el detrusor [el músculo liso del cuello vesical que forma el esfínter interno]); la próstata (músculo liso prostático de la zona de transición que rodea la uretra prostática [células epiteliales glandulares prostáticas y tejido estrómico]); la uretra (músculo liso uretral); y los vasos sanguíneos y terminaciones nerviosas que irrigan/inervan estas estructuras (Gacci 2011). Las expresiones de FDE en las vías urinarias inferiores parecen ser ubicuas, aunque la densidad de estas enzimas puede variar de un lugar a otro. La presencia significativa de las enzimas FDE, especialmente FDE₅, en el tejido de las vías urinarias inferiores, y los efectos sobre diversos músculos lisos son resultados favorables que indican un nuevo objetivo del fármaco para el tratamiento de los SUB compatible con la HPB. Sin embargo, no está claro si el predominio indica un predominio en el control de la fisiología de SUB y la fisiopatología posterior.

Sairam 2002 presentó la primera evidencia para la prueba del concepto en seres humanos en un estudio observacional. Desde entonces, se ha acumulado evidencia a favor de la inhibición de la FDE₅ en los SUB‐HPB de estudios preclínicos y clínicos. Los estudios clínicos se han realizado en poblaciones de origen étnico variado y han demostrado un efecto beneficioso de los IFDE en los SUB‐HPB con éxito variable.

Por qué es importante realizar esta revisión

La guía actual del tratamiento médico para los SUB compatible con HPB (AB con o sin 5‐ARI) es bastante satisfactoria con respecto a la eficacia, pero se tolera mal debido a los efectos secundarios vasculares (Laborde 2009). La eficacia de estos fármacos tiende a disminuir con el transcurso del tiempo (McConnell 2003; Roehrborn 2006). Además, causan efectos adversos en la función sexual con disminución de la libido (3% al 6% con 5‐ARI), DE (6% al 8% con 5‐ARI), disminución del volumen eyaculatorio (1% al 4% con 5‐ARI) y eyaculación retrógrada (8% al 28% con AB) (Laborde 2009).
Se ha demostrado que los IFDE tienen eficacia para el alivio de los SUB en diversos tipos de estudios clínicos incluidos los estudios observacionales, los ensayos clínicos no controlados, los ensayos aleatorios abiertos y los ensayos aleatorios controlados con placebo.
Los IFDE, los AB y los 5 IRA actúan por medio de diferentes componentes en la fisiopatología de los SUB compatibles con la HPB. El inhibidor de FDE₅, tadalafil, es relativamente nuevo y se necesita una revisión minuciosa de la eficacia para ayudar a los médicos a optar por el mejor fármaco y combinación medicamentosa para el tratamiento.
No están claros los efectos a largo plazo de los IFDE para mantener la eficacia en los SUB, los efectos sobre la evolución de la enfermedad y los resultados adversos.

Por lo tanto, se necesita una revisión sistemática de la evidencia emergente para permitir una evaluación general de la eficacia y la seguridad de estos agentes.

Objetivos

disponible en

Evaluar los efectos de los IFDE en comparación con placebo y otros fármacos (AB y 5‐ARI) en hombres con SUB compatibles con HPB.

Métodos

disponible en

Criterios de inclusión de estudios para esta revisión

Tipos de estudios

Se consideró la inclusión en la revisión de ensayos clínicos controlados aleatorios (ECA) en que los participantes y los investigadores del estudio y los evaluadores de resultado desconocían las intervenciones. Los estudios que evaluaron una (o más) de las medidas de resultado mencionadas a continuación se consideraron aptos para el análisis. Se incluyeron estudios elegibles independientemente de su estado de publicación o idioma de publicación.

Tipos de participantes

Se buscaron los estudios que incluyeron hombres con HPB y SUB significativos evaluados con una International Prostate Symptom Score‐total (IPSS‐total) de 12 o más (de la American Urologic Association [AUA]) con o sin DE concomitante.

Tipos de intervenciones

Los estudios realizaron las siguientes comparaciones.

IFDE versus placebo (administrado durante al menos cuatro semanas).
IFDE versus otro tratamiento médico estándar para los SUB‐HPB (AB y 5‐ARI).
IFDE en combinación con otro tratamiento médico estándar para los SUB‐HPB versus cualquiera de los tratamientos solo.

Los grupos de comparación específicos fueron los siguientes.

IFDE versus placebo.
IFDE versus AB.
IFDE versus 5‐ARI.
IFDE + AB versus IFDE solo.
IFDE + AB versus AB solo.
IFDE + 5‐ARI versus IFDE solo.
IFDE + 5‐ARI versus 5‐ARI solo.
IFDE + AB + 5‐ARI versus AB + 5‐ARI.
Comparación de diferentes agentes de IFDE para el tratamiento de los SUB‐HPB.

Tipos de medida de resultado

Resultados primarios

Total International Prostate Symptom (IPSS‐total) score
Benign Prostatic Hyperplasia Impact Index (BPHII) score
Eventos adversos (EA)

Resultados secundarios

IPSS‐quality of life score (IPSS‐QOL)
Retención urinaria aguda (RUA)
Intervenciones quirúrgicas
Otros efectos perjudiciales relacionados con el tratamiento

Todos los resultados se evaluaron a corto plazo (hasta 12 semanas) y a largo plazo (12 semanas y más). Se informaron diferencias mínimas clínicamente importantes (DMCI) con relevancia clínica. Para la IPSS‐total, se utilizó una DMCI de 3 (AUA 2010; Barry 1995); para BPHII, se utilizó una DMCI de 0,5 (Barry 1995).

Métodos de búsqueda para la identificación de los estudios

Búsquedas electrónicas

We first searched electronic databases in October 2014; we periodically updated the search during conduct of the review, conducting the final update in August 2018. Our search included the Cochrane Central Register of Controlled Trials (CENTRAL; updated 2018 August 2) in the Wiley Cochrane Library; MEDLINE (Ovid®) (updated 2 August 2018); Embase (Elsevier) (updated 2 August 2018); and Science Citation Index Expanded via Web of Science™ (updated 2 August 2018). We also searched the clinical trials registries ClinicalTrials.gov and the World Health Organization International Clinical Trials Registry Platform (WHO ICTRP) (updated 13 August 2018). We have detailed the search strategies used in Appendices 1 to 5 (Appendix 1; Appendix 2; Appendix 3; Appendix 4; Appendix 5).

Búsqueda de otros recursos

We inspected the references lists of all included RCTs and selected relevant reviews to look for more trials. We asked the authors of eligible RCTs for additional data from their study if these data were not reported in the published manuscripts. We asked investigators of unpublished trials for additional information. We handsearched relevant peer‐reviewed journals (Journal of Urology, BJU International, European Urology, Journal of Sexual Medicine, Urology Gold; from January 2010 to August 2, 2018) and abstract books including contributions presented at congresses of international societies (e.g. American Urological Association, International Continence Society; from January 2010 to August 2, 2018) for additional information.

Obtención y análisis de los datos

Selección de los estudios

Four review authors (SP, RM, AP, and MMA) independently assessed relevant titles and abstracts, identified by the literature search, against predefined inclusion criteria. The same four review authors investigated all potentially relevant records as full text, mapped records to studies, and classified studies as included studies or excluded studies. We resolved conflicts by discussion and by consensus.

We reviewed all English and non‐English papers. We corresponded with original study authors to request any relevant additional information needed to determine the eligibility of a study for inclusion. When we identified that more than one publication of a trial existed, we grouped reports together and used the most complete data set. We sought help from the Cochrane Urology Group in resolving issues when needed. We have outlined the results of our search in Figure 1.

Extracción y manejo de los datos

Three review authors (SP, RM, and AP) independently abstracted data using a standardised data abstraction form that had been pilot‐tested. Extracted data included study design; time period when the study was conducted; study setting; participant inclusion and exclusion criteria; participant age; sample size for each intervention group; details of interventions; details of outcomes relevant to the review; how outcomes were measured; time points at which they were measured; details of funding sources; declarations of interest among primary study authors; any published protocols; and any other study details relevant to risk of bias assessment. Data analysis was based on available trial data relevant to comparisons and outcomes of interest. We contacted authors of the primary studies for clarification of data and/or additional information if relevant data were missing from the published manuscripts. We finalised data analysis after discussion among the review authors (JLM, MMA, SKS, and AKM), and we resolved disagreements by consensus and in consultation with the Cochrane Urology Group.

Evaluación del riesgo de sesgo de los estudios incluidos

Two independent review authors (SP and JLM) assessed the quality and risk of bias (ROB) of included studies, without blinding to authorship or journal. We assessed the following items using the Cochrane method for assessing risk of bias (Higgins 2011), along with the 'Risk of bias' tool provided in RevMan 5.3.5 (Revman 2014).

Random sequence generation (selection bias).
Allocation concealment.
Blinding of participants and personnel.
Blinding of outcome assessment.
Incomplete outcome data.
Selective reporting.
Other sources of bias.

We judged risk of bias domains as low risk, high risk, or unclear risk, as described in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). We assessed risk of bias for each outcome within and across studies. We noted that the risk of bias for random sequence generation and allocation concealment was consistent across outcomes. However, we assessed risk of bias for blinding of participants and personnel, blinding of outcome assessment, incomplete outcome data, and selective reporting on an outcome‐specific basis, as the risk of bias of each of these domains is likely to vary by outcome.

Based on the nature of the outcomes, we grouped them into (1) IPSS‐total, BPHII, and IPSS‐QOL; (2) AEs and other treatment‐related harms; and (3) AUR and need for surgical intervention for assessment of risk of bias.

We have generated 'Risk of bias tables' for each included study and have presented a 'Risk of bias summary' in Figure 2.

Medidas del efecto del tratamiento

We used Review Manager 5.3 to analyse data (Revman 2014). We combined data from RCTs for meta‐analyses. We calculated the mean difference (MD) with 95% confidence interval (95% CI) for continuous outcomes and the risk ratio (RR) with 95% CI for dichotomous outcomes.

Cuestiones relativas a la unidad de análisis

We considered parallel‐design randomised controlled clinical trials as well as cross‐over studies for inclusion in the review. We performed data extraction and analysis and risk of bias assessments for cross‐over trials as per Section 16.4, in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). The individual participant is considered a unit in this type of analysis, hence the unit of analysis issue is not relevant for this review.

Manejo de los datos faltantes

We contacted the authors of papers with missing data (studies, outcomes, summary data, individuals, measures of variance, or study level characteristics) to request missing or unreported data. We performed analysis using the intention‐to‐treat (ITT) principle. When we could not obtain data from the authors of incompletely reported papers, we excluded them from the review, citing the specific reason for the same. However, if pooled data were available for such studies, we included them in the analysis as published literature.

Evaluación de la heterogeneidad

We regarded an I² value > 50% as showing considerable heterogeneity, and we dealt with this by using the random‐effects model. Alternatively if we judged that clinical or study design heterogeneity was present, we conducted subgroup analyses to assess the robustness of pooled outcomes and conclusions.

Evaluación de los sesgos de notificación

We attempted to obtain the protocols of included studies to assess selective outcome reporting. We constructed funnel plots (trial effect vs trial size) for comparisons including at least 10 studies to investigate the possibility of publication bias. We also looked at other relevant sources of bias.

Síntesis de los datos

We used a fixed‐effect model for analysis unless we noted statistically significant heterogeneity (I² > 50%) between studies. We employed a random‐effects model if heterogeneity existed. We used the inverse variance method for meta‐analysis of continuous outcomes and the Cochran‐Mantel‐Haenzsel method for meta‐analysis of dichotomous outcomes. When a meta‐analysis was not possible or was not appropriate, we prepared a narrative synthesis.

Análisis de subgrupos e investigación de la heterogeneidad

We conducted the following subgroup analyses.

Participants with or without ED.
Different PDEIs, namely, tadalafil versus vardenafil versus sildenafil.
Different doses of PDEIs.

Análisis de sensibilidad

We excluded RCTs without a run‐in period

Among the included RCTs, some had incorporated a two‐ to four‐week run‐in period before providing the randomised study intervention. The placebo and/or active drug run‐in period represents a potential source of bias because this may lead to selection of participants who are more compliant and/or less likely to experience AE. Therefore, we conducted a sensitivity analysis to explore the effect of the pre‐intervention run‐in period on effect size for various outcomes.

'Summary of findings' table

We rated the overall quality of evidence using the Grading of Recommendations Assessment, Development and Evaluation (GRADE) framework, which takes into account criteria related not only to internal validity: study limitations (risk of bias), inconsistency, imprecision, and publication bias; but also to external validity: indirectness of results (Guyatt 2008). For each comparison, two review authors (SP and RM) independently rated the quality of evidence for each outcome as high, moderate, low, or very low, using GRADEproGDT software. SP, RM, and JLM constructed 'Summary of findings' tables and resolved discrepancies by consensus or, if needed, via arbitration by other review authors (AP, SKS, and AKM). For each comparison, we have presented a summary of evidence for the main outcomes in a 'Summary of findings' table. This provides key information about the best estimate of the magnitude of effect in relative terms and absolute differences for each relevant comparison of alternative management strategies, numbers of participants and studies addressing each important outcome, and the rating of overall confidence in effect estimates for each outcome (Guyatt 2011; Schünemann 2011; Schünemann 2013).

Results

Description of studies

We have outlined search results in Figure 1. As per the protocol, we used nine comparison groups to evaluate the efficacy and safety of PDEI compared to placebo or existing treatments (alone or in combination). We segregated the included studies as per the comparison groups. Of the nine comparison groups, we could identify no studies for four groups.

PDEI versus 5‐ARI (group 3).
PDEI + 5‐ARI versus PDEI alone (group 6).
PDEI + AB + 5‐ARI versus AB + 5‐ARI (group 8).
Comparison of different PDEIs for treatment of BPH‐LUTS (group 9).

We have presented detailed descriptions of the interventions in Table 1, and we have described baseline characteristics of the included studies in Table 2.

Open in table viewer

Table 1. Description of interventions

Study (year)	Intervention(s) (route, frequency, total dose/d)	Intervention(s) appropriate as applied in a clinical practice setting^a (description)	Comparator(s) (route, frequency, total dose/d)	Comparator(s) appropriate as applied in a clinical practice setting^a (description)
Casabe (2014)	I1: tadalafil 5 mg/d + finasteride 5 mg/d oral	Tadalafil and finasteride were used in this study as per clinical practice setting	C1: placebo + finasteride 5 mg/d oral	Finasteride was used in this study as per clinical practice setting Placebo; N/CPS
Regadas (2013)	I1: tadalafil 5 mg/d + tamsulosin 0.4 mg/d oral	Tadalafil and tamsulosin were used in this study as per clinical practice setting	C1: tamsulosin 0.4 mg/d + placebo oral	Tamsulosin was used in this study as per clinical practice setting Placebo; N/CPS
Brock (2013a)	I1: tadalafil 5 mg/d oral	Tadalafil was used in this study as per clinical practice setting	C1: placebo once daily oral	N/CPS
Yokoyama (2013)	I1: tadalafil 2.5 mg/d oral I2: tadalafil 5 mg/d oral I3: tamsulosin 0.2 mg/d oral	Tadalafil was used in this study as per clinical practice setting	C1: placebo oral	N/CPS
Egerdie (2012)	I1: tadalafil 2.5 mg/d oral I2: tadalafil 5 mg/d oral	Tadalafil was used in this study as per clinical practice setting	C1: placebo oral	N/CPS
Gacci (2012)	I1: tamsulosin 0.4 mg/d + vardenafil 10 mg/d oral	Tadalafil and vardenafil were used in this study as per clinical practice setting	C1: tamsulosin 0.4 mg/d oral	Tamsulosin was used in this study as per clinical practice setting
Oelke (2012)	I1: tadalafil 5 mg/d oral I2: tamsulosin 0.4 mg/d oral	Tadalafil and tamsulosin were used in this study as per clinical practice setting	C1: placebo oral	N/CPS
Takeda (2012)	I1: tadalafil 2.5 mg/d oral I2: tadalafil 5 mg/d oral	Tadalafil was used in this study as per clinical practice setting	C1: placebo oral	N/CPS
Kim (2011)	I1: tadalafil 5 mg/d oral I2: tamsulosin 0.2 mg/d oral	Tadalafil and tamsulosin were used in this study as per clinical practice setting	C1: placebo oral	N/CPS
Dmochowski (2010)	I1: tamsulosin 20 mg/d oral	Tadalafil is approved for use in BPH‐LUTS at 2.5 mg and 5 mg per day Tadalafil dose used in this study is higher than that used in the clinical practice setting	C1: placebo oral	N/CPS
Prost (2011)	I1: tadalafil 5 mg/d oral	Tadalafil was used in this study as per clinical practice setting	C1: placebo oral	N/CPS
Tuncel (2010)	I1: sildenafil 25 mg/4 times weekly + tamsulosin 0.4 mg/d oral I2: sildenafil 25 mg/4 times weekly oral	Sildenafil and tamsulosin were used in this study as per clinical practice setting	C1: tamsulosin 0.4 mg/d oral	Tamsulosin was used in this study as per clinical practice setting
Bechara (2008)	I1: tadalafil 20 mg/d + tamsulosin 0.4 mg/d oral	Tadalafil is approved for use in BPH‐LUTS at 2.5 mg and 5 mg per day Tadalafil dose used in this study is higher than that used in clinical practice setting Tamsulosin was used in this study as per clinical practice setting	C1: tamsulosin 0.4 mg/d + placebo oral	Tamsulosin was used in this study as per clinical practice setting Placebo; N/CPS
Bechara (2008)	I2: tamsulosin 0.4 mg/d + placebo oral	Tamsulosin was used in this study as per clinical practice setting For placebo; N/CPS	C2: tadalafil 20 mg/d + tamsulosin 0.4 mg/d oral	Tadalafil is approved for use in BPH‐LUTS at 2.5 mg and 5 mg per day Tadalafil dose used in this study was higher than that used in clinical practice setting Tamsulosin was used in this study as per clinical practice setting
Roehrborn (2008)	I1: tadalafil 2.5 mg/d oral I2: tadalafil 5 mg/d oral I3: tadalafil 10 mg/d oral I4: tadalafil 20 mg/d oral	Tadalafil is approved for use in BPH‐LUTS at 2.5 mg and 5 mg per day This is a dose‐ranging study; tadalafil doses used in this study for I3 and I4 were higher than those used in clinical practice setting	C1: placebo oral	N/CPS
Stief (2008)	I1: vardenafil 10 mg	Vardenafil was used in this study as per clinical practice setting	C1: placebo oral	N/CPS
McVary (2007a)	I1: sildenafil 50 to 100 mg/d oral	Sildenafil was used in this study as per clinical practice setting	C1: placebo oral	N/CPS
McVary (2007b)	I1: tadalafil 5 to 20 mg/d oral	Tadalafil dose used in this study in the dose escalation phase was higher than that used in clinical practice setting	C1: placebo oral	N/CPS
^aThe term 'clinical practice setting' refers to the specification of the intervention/comparator as used in the course of a standard medical treatment (such as dose, dose escalation, dosing scheme, provision for contraindications, and other important features). BPH: benign prostatic hyperplasia; C: comparator; I: intervention; LUTS: lower urinary tract symptoms; N/CPS: no specification of clinical practice setting possible.

Open in table viewer

Table 2. Baseline characteristics

Study (year)	Intervention(s) and comparator(s)	Duration of intervention (duration of follow‐up) (days, months, years...)	Description of participants	Trial period (year to year)	Country	Setting	Ethnic groups (%)	Duration of disease (mean/range years (SD), or as reported)
Casabe (2014)	I1: tadalafil 5 mg/d + finasteride 5 mg/d	12 weeks (for primary endpoint) Study continued maintaining blinding until 26 weeks (for secondary endpoints)	Men > 45 years with BPH‐LUTS; IPSS ≥ 13 Qmax 4 to 15 mL/s Prostatic volume ≥ 30 cc Naive to 5‐ARI therapy	November 2010 to September 2012	USA, Canada, Latin America	‐	Whites (85.5%)	‐
							Black/African American (10.1%)
							Native American (2.5%)
	C1: placebo + finasteride 5 mg/d						Other (3.9%)
Regadas (2013)	I1: tadalafil 5 mg/d + tamsulosin 0.4 mg/d	30 days	Men > 45 years, BOOI > 20 IPSS > 14	October 2010 to September 2011	"‐"	‐	"‐"	‐
	C1: tamsulosin 0.4 mg/d + placebo
Brock (2013a)	I1: tadalafil 5 mg/d	12 weeks	Men ≥ 45 years BPH‐LUTS > 6 months IPSS ≥ 13 Qmax 4 to 15 mL/s	"‐"	‐	‐	‐	‐
	C1: placebo
Yokoyama (2013)	I1: tadalafil 2.5 mg/d	12 weeks	Men > 45 years; BPH‐LUTS > 6 months IPSS > 13 Qmax 4 to 15 mL/s Prostate volume > 20 cc	"‐"	Japan Korea Taiwan	‐	Japan (55.8%)	‐
	I2: tadalafil 5 mg/d						Korea (29.1%)
	I3: tamsulosin 0.2 mg/d						Taiwan (14.8%)
	C1: placebo
Egerdie (2012)	I1: tadalafil 2.5 mg/d	12 weeks	Men ≥ 45 years, ≥ 3 months h/o ED, ≥ 6 months h/o BPH‐LUTS (LUTS + ED)		Europe Mexico USA/Canada	‐	White (93%)	‐
	I2: tadalafil 5 mg/d						Black American (4%)
	C2: placebo						Asian (2.5%)
							Hispanic (15%)
							Other (0.6%)
Gacci (2012)	I1: tamsulosin 0.4 mg/d + vardenafil 10 mg/d	12 weeks	Men with persistent storage/irritative symptoms after tamsulosin 0.4 mg treatment for 2 weeks; IPSS‐storage ≥ 8	"‐"	"‐"	‐	"‐"	‐
	C1: tamsulosin 0.4 mg/d
Oelke (2012)	I1: tadalafil 5 mg/d	12 weeks	Men ≥ 45 years, BPH‐LUTS > 6 months, IPSS ≥ 13, Qmax 4 to 15 mL/s before placebo lead‐in period	December 2009 to January 2011	Australia, Austria, Belgium, France, Germany, Greece, Italy, Mexico, The Netherlands, Poland	‐	White (76%)	‐
	I2: tamsulosin 0.4 mg/d						Black or African American (0.3%)
	C1: placebo						American Indian/Alaska Native (23.1%)
Takeda (2012)	I1: tadalafil 2.5 mg/d	12 weeks, followed by a 42‐week open‐label extension phase	> 45 years, IPSS > 13, Qmax 4 to 15 mL/s, prostate volume > 20 cc	"‐"	Japan	‐	Asian (100%)	‐
	I2: tadalafil 5 mg/d
	C1: placebo
Kim (2011)	I1: tadalafil 5 mg/d	12 weeks	Men ≥ 45 years, h/o BPH‐LUTS for 6 months or longer, IPSS ≥ 13, Qmax > 5 and < 15 after the washout period	"‐"	Korea	‐	Asian (100%)	‐
	I2: tamsulosin 0.2 mg/d
	C1: placebo
Dmochowski (2010)	I1: tamsulosin 20 mg/d	12 weeks	Men > 40 years, longer than 6 months h/o BPH‐LUTS; IPSS ≥ 13, with or without BOO	"‐"	USA, Canada	‐	White (76%)	‐
	C1: placebo						Black American (12.5%)
							Hispanic (8%)
							Other (3.0%)
Prost (2011)	I1: tadalafil 5 mg/d	12 weeks	Men > 45 years, BPH‐LUTS > 6 months IPSS > 13 and Qmax 4 to 15 mL/s	"‐"	Argentina, Germany, Italy, Mexico, USA	‐	White (91%)	‐
	C1: placebo						Black or African American (2.5%)
							American Indian or Alaska Native (5.2%)
							Other (1%)
Tuncel (2010)	I1: sildenafil 25 mg/4 times weekly + tamsulosin 0.4 mg/d	12 weeks	IPSS > 12, IIEF < 21 (SIHM < 21) treatment‐naive and all participants had ED	‐	‐	‐	‐	‐
	I2: sildenafil 25 mg/4 times weekly
	C1: placebo
Bechara (2008)	I1: tadalafil 20 mg/d + tamsulosin 0.4 mg/d	45 days	Men ≥ 50 years, h/o BPH > 6 months, IPSS ≥ 12, ⁶PSA ≤ 4 ng/mL, ²Qmax > 5 and < 15 for a minimum voided volume of 125 mL	September 2007 to February 2008	Argentina	‐	‐	‐
	I2: tamsulosin 0.4 mg/d + placebo
Roehrborn (2008)	I1: tadalafil 2.5 mg/d	12 weeks	Men > 45 years, BPH‐LUTS > 6 months, IPSS > 13, Qmax 4 to 15 mL/s	‐	North America, Europe, Australia	‐	White (85%)	‐
	I2: tadalafil 5 mg/d						Hispanic (11.5%)
	I3: tadalafil 10 mg/d						Black American (2.42%)
	I4: tadalafil 20 mg/d						Other (0.5%)
Stief (2008)	I1: vardenafil 10 mg	12 weeks	Men > 45 to 64 years, BPH‐LUTS > 6 months, IPSS > 12	October 2005 to June 2006	Germany	‐	White (99%)	‐
							Black Americans (0.5%)
	C1: placebo						Other (0.5%)
McVary (2007b)	I1: tadalafil 5 mg/d	12 weeks	Men > 45 years, h/o BPH‐LUTS 6 months with IPSS > 13, Qmax > 4 to < 15 mL/s for a voided volume of 125 mL	November 2004 to July 2005	USA	‐	White (81%)	‐
	C1: placebo						Black American (9.5%)
							Hispanic (6.5%)
							Other (2.5%)
McVary (2007a)	I1: sildenafil 50 to 100 mg/d	12 weeks	Men ≥ 45 years, with IPSS > 12 and IIEF score < 25 All participants had ED	March 2004 to May 2005	USA	‐	White (82%)	‐
	C1: placebo						Black American (11.5%)
							Asian (2%)
							Other (4%)
‐ denotes not reported; "‐" denotes missing information. BOO: bladder outflow obstruction; BOOI: Bladder Outflow Obstruction Index; C: comparator; I: intervention; IIEF: International Index of Erectile Function; PSA: prostate‐specific antigen; Qmax: maximum urinary flow rate; SD: standard deviation; SHIM: Sexual Health Inventory for Males.

PDEI versus placebo

Through our search strategy, we identified 11 RCTs meeting the inclusion criteria (Dmochowski 2010; Egerdie 2012; Kim 2011; McVary 2007a; McVary 2007b; Oelke 2012; Porst 2011; Roehrborn 2008; Stief 2008; Takeda 2012; Tamimi 2010; Yokoyama 2013). We did not include one RCT, as we could not obtain data from study authors or could not impute data from available published literature (Tamimi 2010). We included in the analysis a total of 11 RCTs with 4293 participants (1576 in the placebo group, and 2717 in the PDEI treatment group). All RCTs were multi‐centric clinical trials. They included elderly men with LUTS consistent with BPH living in various parts of North America, Europe, South America, Japan, and Korea. Five of the 11 RCTs were multi‐national (Kim 2011; McVary 2007a; McVary 2007b; Stief 2008; Takeda 2012; Yokoyama 2013). Among the other studies, one was conducted in the USA (McVary 2007a; McVary 2007b), two in Japan (Takeda 2012; Yokoyama 2013), one in Korea (Kim 2011), and one in Germany (Stief 2008).

All included RCTs reported the racial origin of participants. About 85% to 87% of participants belonged to the white race, and 4% to 6% belonged to the black race, as reported in seven RCTs (Dmochowski 2010; Egerdie 2012; McVary 2007a; McVary 2007b; Oelke 2012; Porst 2011 Roehrborn 2008; Stief 2008). The remaining participants (7% to 11%) were Asian men included in three RCTs (Kim 2011;Takeda 2012; Yokoyama 2013).

RCTs included men older than 45 years with symptoms of BPH‐LUTS, irrespective of their ED status. The mean age of participants in PDEI and placebo groups was 69.1 ± 8.4 and 62.02 ± 8.8 years, respectively. Two RCTs included participants with both BPH‐LUTS and ED (Egerdie 2012; McVary 2007a). Included participants had moderate to severe LUTS at the qualifying visit. Mean IPSS‐total score in PDEI and placebo groups was 17.6 ± 5.5 and 17.5 ± 5.6, respectively. Baseline characteristics included baseline LUTS severity, ED severity, prior AB or 5‐ARI use, uroflow parameters, post‐void residue (PVR), PSA, and demographic features. Six of the 11 RCTs reported participant characteristics including LUTS severity, proportions of participants with ED, severity of ED, and prior AB use status. Six RCTs did not report any of these participant characteristics (Dmochowski 2010; McVary 2007a; Roehrborn 2008; Stief 2008; Takeda 2012; Yokoyama 2013). The mean duration of BPH‐LUTS and/or ED also was not described in many of the RCTs (Dmochowski 2010; Egerdie 2012; Kim 2011; McVary 2007b; Oelke 2012; Roehrborn 2008; Stief 2008; Takeda 2012; Yokoyama 2013). Men with a history of BPH‐LUTS longer than six months were considered for inclusion. RCTs had excluded participants with Qmax < 4 mL/s and > 15 mL/s, PSA > 10 ng/mL or positive prostatic biopsy with PSA between 4 and 10 ng/mL, use of 5‐ARIs within the last three to six months, and other pelvic diseases or history of pelvic surgery within the preceding six months.

Nine of the 11 RCTs used the long‐acting PDEI, tadalafil, whereas McVary 2007a used sildenafil and Stief 2008 used vardenafil. We were able to find one study that examined use of gisadenafil but were unable to retrieve data required for the meta‐analysis (Tamimi 2010). We identified no studies for avanafil or udenafil. The dose of tadalafil studied ranged from 2.5 to 20 mg once daily, for sildenafil 50 to 100 mg once daily, and vardenafil 10 mg twice daily. The dose of tadalafil used most commonly was 5 mg per day (i.e. the FDA‐approved highest dose for BPH‐LUTS) (Cialis 2011). Four RCTs included an additional arm given 2.5 mg per day (Egerdie 2012; Roehrborn 2008; Takeda 2012; Yokoyama 2013). Dmochowski 2010 studied only one dose of tadalafil ‐ 20 mg per day ‐ and the primary outcome was the urodynamic parameter. Roehrborn 2008, a dose‐ranging study, compared all dosages, namely, 2.5, 5, 10, or 20 mg per day, against placebo. McVary 2007a used sildenafil in a dose titration design but did not report the number of participants using 50 or 100 mg of the drug.

Most RCTs were short‐term efficacy studies. The maximum study duration was 12 weeks (in 10/11 RCTs) and the minimum duration was eight weeks (Stief 2008). Two studies had an open‐label extension (OLE) phase of up to 42 weeks ‐ in Takeda 2012 ‐and 64 weeks in Roehrborn 2008. These studies used tadalafil at a dose of 5 mg per day in the OLE. Seven of 11 RCTs included a treatment‐free washout period of four weeks for participants who were already receiving treatment at screening (Egerdie 2012; Kim 2011; McVary 2007b; Oelke 2012; Porst 2011; Roehrborn 2008; Yokoyama 2013), but Takeda 2012 reported a washout period of two weeks. After the washout, investigators subjected these participants to a four‐week single‐blind placebo run‐in period. However, the treatment‐naïve participants were not subjected to any wash‐out but only a four‐week single‐blind placebo run‐in period. Three of 11 RCTs did not follow this pattern for enrolment. Two of them used only a drug‐free washout period (Dmochowski 2010; Stief 2008), but they did not include a placebo run‐in period. McVary 2007a neither included a washout period nor a run‐in period. This study also did not specify whether participants were naïve or experienced to treatment.

The primary outcome in most RCTs was effect of PDEIs on LUTS, but for Dmochowski 2010, the primary outcome was change in the urodynamic parameter for bladder outlet obstruction (BOO), and change in erectile function was the secondary outcome.

Pharmaceutical manufacturers of the active drugs (PDEIs) sponsored eight of 11 included RCTs comparing PDEIs versus placebo (Egerdie 2012; McVary 2007a; McVary 2007b; Oelke 2012; Porst 2011; Roehrborn 2008; Stief 2008; Takeda 2012; Yokoyama 2013), and manufacturers of the investigational drug presumably sponsored the remaining two RCTs, although published reports included no statement on trial sponsorship (Dmochowski 2010; Kim 2011). Apart from providing the logistics, study sponsors made significant contributions to design and conduct of the study, data acquisition, data analysis, interpretation of results, and preparation and final approval of the manuscript.

PDEI versus AB

We included four RCTs comprising 933 participants for this comparison (Kim 2011; Oelke 2012; Tuncel 2010; Yokoyama 2013). Three of these RCTs were included in the previous comparison and had three arms, namely, PDEI, AB, and placebo (Kim 2011; Oelke 2012; Yokoyama 2013). The only remaining RCT compared PDEI, AB, and a combination of the two (Tuncel 2010). Three RCTs used tadalafil 5 mg as the PDEI (Kim 2011; Oelke 2012; Yokoyama 2013), and one used sildenafil 25 mg (Tuncel 2010). Yokoyama 2013 included an additional tadalafil 2.5 mg arm. All researchers used PDEI as once‐daily therapy, but the sildenafil study ‐ Tuncel 2010 ‐ used it four times weekly. The AB in all included studies was tamsulosin, although the dose was not uniform across studies. Two RCTs used 0.2 mg once daily (Kim 2011; Yokoyama 2013), whereas the other two used 0.4 mg once daily (Oelke 2012; Tuncel 2010). All participants included in Tuncel 2010 were treatment naïve; the other studies did not specify so and included a mixed cohort of previously treated and untreated participants. The tadalafil studies incorporated a drug washout period, followed by a placebo run‐in period (Kim 2011; Oelke 2012; Yokoyama 2013). All studies examined short‐term treatment provided over eight to 12 weeks.

PDEI versus 5‐ARI

We found no studies that met the inclusion criteria for this comparison.

PDEI + AB versus AB alone

For this comparison, we included four RCTs involving a total of 193 participants (Bechara 2008; Gacci 2012; Regadas 2013; Tuncel 2010). All RCTs compared treatments in parallel design, except Bechara 2008, which used a cross‐over design. The duration of studies varied from 30 days to 12 weeks. PDEIs used were not uniform across studies. Bechara 2008 used tadalafil 20 mg/d and Regadas 2013 5 mg/d, Tuncel 2010 used sildenafil 25 mg/4 times a week, and Gacci 2012 used vardenafil 10 mg/d. The AB used was uniform in type and dose across studies (tamsulosin 0.4 mg/d). Two of the four RCTs were placebo controlled and included a placebo arm along with the AB alone arm (Gacci 2012; Regadas 2013); the other two RCTs were not placebo controlled (Bechara 2008; Tuncel 2010). Two RCTs included a placebo run‐in period (Bechara 2008; Gacci 2012), and the other two did not (Regadas 2013; Tuncel 2010). Participants included in Tuncel 2010 were treatment naïve, whereas remaining RCTs did not specify this.

PDEI + AB versus PDEI alone

Only one RCT, conducted in Turkey, was eligible for inclusion (Tuncel 2010). It included a total of 60 participants and three arms (sildenafil 25 mg four times a week vs tamsulosin 0.4 mg once a day vs sildenafil 25 mg four times a week + tamsulosin 0.4 mg per day). The duration of the study was eight weeks, and it included all treatment‐naïve participants with both BPH‐LUTS and ED. This study did not include a placebo run‐in period, and participants directly entered the randomisation phase.

PDEI + 5‐ARI versus 5‐ARI alone

Only one study qualified for inclusion in this group (Casabe 2014). This study included 695 participants and was conducted in North America, Europe, and Latin America. The duration of the study was 26 weeks. Participants underwent a drug‐free washout period of four weeks, followed by a four‐week placebo run‐in period, before randomisation. Participants were randomised to tadalafil 5 mg plus finasteride 5 mg per day or placebo plus finasteride 5 mg per day. Included participants had a larger prostate volume of 30 mL or greater and were naïve to 5‐ARI therapy but not to other LUTS treatments, namely, alpha‐blockers or anticholinergics.

PDEI + 5‐ARI versus PDEI alone

We found no studies that met the inclusion criteria for this comparison.

PDEI + AB + 5‐ARI versus AB + 5‐ARI

We found no studies that met the inclusion criteria for this comparison.

Comparison of different PDEI agents for treatment of BPH‐LUTS

We found no studies that met the inclusion criteria for this comparison.

Results of the search

We conducted the original search in May 2014 and updated the search through 20 February 2018. We have provided detailed results of the search in Figure 1.

Included studies

We included a total of 16 RCTs in the review. We have described the characteristics of included studies in the Characteristics of included studies section and in Table 1 and Table 2. We grouped studies by comparison. Some RCTs had more than two arms and hence were eligible for inclusion in more than one comparison group. Analysis of PDEI versus placebo included 11 RCTs; PDEI versus alpha‐blockers four RCTs; combination of PDEI and AB versus AB alone four RCTs; combination of PDEI and alpha‐blockers versus PDEI alone one RCT; and combination of PDEI and 5‐ARI versus 5‐ARI alone one RCT.

Excluded studies

We excluded RCTs not meeting the inclusion criteria for this review. We described the characteristics of excluded studies and reasons for exclusion in the Characteristics of excluded studies section.

Risk of bias in included studies

We have described below risk of bias assessment by comparison. We have presented in Figure 2 a summary of this assessment.

PDEI versus placebo

Allocation (selection bias)

Random sequence generation

We judged seven of the 11 included RCTs as having unclear risk of bias (Dmochowski 2010; Kim 2011; McVary 2007b; Oelke 2012; Porst 2011; Roehrborn 2008; Yokoyama 2013).

Allocation concealment

We judged nine of 11 RCTs as having unclear risk of bias (Dmochowski 2010; Gacci 2012; Kim 2011; McVary 2007a; McVary 2007b; Oelke 2012; Porst 2011; Roehrborn 2008; Yokoyama 2013). Three RCTs used an interactive voice response system or opaque, sealed, and numbered envelopes; hence we rated them as having low risk of bias.

Blinding

Blinding of participants and personnel (performance bias)

We deemed risk of bias as unclear for six of 11 RCTs for IPSS‐total, BPHII, IPSS‐QOL, AEs, and treatment‐related harms (Dmochowski 2010; Kim 2011; McVary 2007a; McVary 2007b; Porst 2011; Yokoyama 2013). However we judged AUR and need for surgical interventions as having low risk of bias for all included RCTs.

Blinding of outcome assessment (detection bias)

Risk of bias for IPSS‐total, BPHII, IPSS‐QOL, AEs, and treatment‐related harms was unclear for eight of 11 RCTs (Dmochowski 2010; Kim 2011; McVary 2007a; McVary 2007b; Porst 2011; Roehrborn 2008; Stief 2008; Yokoyama 2013). However due to the nature of the outcomes of AUR and need for surgical intervention, we judged all included RCTs to have low risk of detection bias.

Incomplete outcome data (attrition bias)

Eight of 11 RCTs had unclear risk of bias for IPSS‐total, BPHII, IPSS‐QOL, AEs, and treatment‐related harms (Dmochowski 2010; Kim 2011; McVary 2007a; McVary 2007b; Porst 2011; Roehrborn 2008; Stief 2008; Yokoyama 2013). Seven of 11 RCTs had unclear risk of bias for AUR and need for surgical interventions (Dmochowski 2010; Egerdie 2012; McVary 2007a; Oelke 2012; Roehrborn 2008; Stief 2008; Yokoyama 2013) .

Selective reporting (reporting bias)

We judged one of the 11 included RCTs to have unclear risk of selective reporting bias for IPSS‐total, BPHII, and IPSS‐QOL (Dmochowski 2010). None of the included studies had selective reporting bias for AEs, treatment‐related harms, AUR, or need for surgical intervention.

Other bias

We judged seven of the 11 included RCTs to have unclear risk of other bias for all outcomes because they used study designs that employed a placebo run‐in period (Egerdie 2012; Kim 2011; McVary 2007a; McVary 2007b; Roehrborn 2008; Takeda 2012; Yokoyama 2013).

PDEI versus alpha‐blockers

Allocation (selection bias)

Random sequence generation

Three of the four included RCTs did not adequately describe the method used for random sequence generation; hence we judged them to have unclear risk of selection bias (Kim 2011; Tuncel 2010; Yokoyama 2013) .

Allocation concealment

Three of the four included RCTs did not adequately describe the method used for allocation concealment; hence we judged them to have unclear risk of selection bias (Kim 2011; Tuncel 2010; Yokoyama 2013).

Blinding

Blinding of participants and personnel (performance bias)

We judged three of four RCTs to have unclear risk of performance bias for IPSS‐total, BPHII, IPSS‐QOL, AEs, and treatment‐related harms (Kim 2011; Tuncel 2010; Yokoyama 2013). We determined that none of the RCTs had performance bias for AUR or need for surgical intervention.

Blinding of outcome assessment (detection bias)

We judged three of four RCTs to have unclear risk of detection bias for IPSS‐total, BPHII, IPSS‐QOL, AEs, and treatment‐related harms (Kim 2011; Tuncel 2010; Yokoyama 2013). We determined that none of the RCTs had detection bias for AUR or need for surgical intervention.

Incomplete outcome data (attrition bias)

Two of four included RCTs had unclear risk of attrition bias for all outcomes (Oelke 2012; Yokoyama 2013).

Selective reporting (reporting bias)

We judged all four included RCTs to have low risk of selective reporting bias for all outcomes.

Other bias

Three of four included RCTs employed placebo or an active drug run‐in period before randomisation; hence we judged them to have unclear risk of other bias (Kim 2011; Oelke 2012; Yokoyama 2013).

PDEI + AB versus AB alone

Allocation (selection bias)

Random sequence generation

Three of the four included RCTs did not adequately describe the method used for random sequence generation; hence we judged them to have unclear risk of bias (Bechara 2008; Regadas 2013; Tuncel 2010).

Allocation concealment

No RCTs adequately described the method used; hence we judged them to have unclear risk of bias (Bechara 2008; Gacci 2012; Regadas 2013; Tuncel 2010).

Blinding

Blinding of participants and personnel (performance bias)

We judged two of four RCTs to have unclear risk of performance bias for IPSS‐total, BPHII, IPSS‐QOL, AEs, and treatment‐related harms (Regadas 2013; Tuncel 2010). We determined that no RCTs had performance bias for AUR or need for surgical intervention.

Blinding of outcome assessment (detection bias)

We judged two of four included RCTs to have unclear risk of bias (Regadas 2013; Tuncel 2010). However we determined that risk of detection bias was low for AUR and need for surgical interventions for all four studies, as they are less susceptible to such bias.

Incomplete outcome data (attrition bias)

We determined that none of the four RCTs had attrition bias for any outcomes.

Selective reporting (reporting bias)

We judged two of the four included studies to have unclear risk of selective reporting bias for quality of life assessments (BPHII and IPSS‐QOL), AEs, and treatment‐related harms (Gacci 2012; Regadas 2013). We determined that all four RCTs had low risk of bias for AE, AUR, and need for surgical intervention.

Other bias

We judged two of the four included RCTs to have unclear risk of other bias for all outcomes: one RCT because of lack of a washout period between two cross‐over phases (Bechara 2008), and the other for including an active drug run‐in period (Gacci 2012).

PDEI + alpha‐blockers versus PDEI alone

We identified only one RCT for this comparison (Tuncel 2010).

Allocation (selection bias)

Random sequence generation

We judged this RCT to have unclear risk of selection bias for all outcomes because study authors did not describe the method used.

Allocation concealment

We judged risk of bias as unclear for allocation concealment for all outcomes.

Blinding

Blinding of participants and personnel (performance bias)

We judged the risk of performance bias as unclear for IPSS‐total, BPHII, IPSS‐QOL, and AE assessments, as study authors did not clearly describe whether participants and treatment administrators were masked to the interventions. However we judged the risk of performance bias to be low for AUR and need for surgical intervention, as they are less susceptible to such bias.

Blinding of outcome assessment (detection bias)

We judged the risk of detection bias as unclear for IPSS‐total, BPHII, IPSS‐QOL, and AE assessments, as study authors did not clearly describe whether outcome assessors were masked to the interventions. However we judged the risk of detection bias to be low for AUR and need for surgical intervention, as they are less susceptible to such bias.

Incomplete outcome data (attrition bias)

We judged the risk of attrition bias to be low for all outcomes, as study authors reported no loss to follow‐up and reported outcomes for all randomised participants.

Selective reporting (reporting bias)

We judged the risk of selective reporting bias to be low for all outcomes.

Other bias

We detected no significant risk of other bias.

PDEI + 5‐ARI versus 5‐ARI alone

We included only one RCT for this comparison (Casabe 2014).

Allocation (selection bias)

Random sequence generation

We judged this RCT to have unclear risk of bias.

Allocation concealment

Researchers used an interactive voice response system in allocating interventions to participants; therefore we judged allocation concealment to be adequate.

Blinding

Blinding of participants and personnel (performance bias)

Blinding of outcome assessment (detection bias)

Incomplete outcome data (attrition bias)

This RCT had significant attrition of participants in both PDEI and combination groups. Therefore, we judged the risk of bias to be unclear for IPSS‐total, BPHII, IPSS‐QOL, AEs, AUR, and need for surgical intervention.

Selective reporting (reporting bias)

This study reported all relevant outcomes; hence we judged the risk of selective reporting bias to be low.

Other bias

Participants underwent a placebo run‐in before randomisation. This may have led to selection of more compliant participants and hence may have compromised external validity. Therefore, we judged the risk of bias to be unclear for IPSS‐total, BPHII, IPSS‐QOL, AEs, treatment‐related harms, AUR, and need for surgical intervention.

Effects of interventions

See: Summary of findings for the main comparison PDEI compared to placebo for lower urinary tract symptoms consistent with benign prostatic hyperplasia; Summary of findings 2 PDEI compared to alpha‐blockers for LUTS associated with BPH; Summary of findings 3 PDEI plus alpha‐blocker compared to alpha‐blockers alone for LUTS associated with BPH (short‐term); Summary of findings 4 PDEI plus an alpha‐blocker compared to PDEI alone for LUTS associated with BPH (short‐term); Summary of findings 5 PDEI + 5‐ARI compared to 5‐ARI alone for LUTS associated with BPH (short‐term); Summary of findings 6 PDEI + 5‐ARI compared to 5‐ARI alone for LUTS associated with BPH (long‐term)

PDEI versus placebo

Short‐term (up to 12 weeks) efficacy and safety of PDEI compared to placebo

Eleven RCTs assessed the effects of PDEIs compared to placebo for short‐term use (eight to 12 weeks).

Primary outcomes

IPSS‐total score

PDEIs may result in a small improvement in IPSS (MD 1.89 lower, 95% CI 2.27 lower to 1.50 lower; 11 trials; low‐quality evidence; Analysis 1.1;Figure 3 Figure 4 summary of findings Table for the main comparison). We downgraded the quality of evidence for study limitations and possible publication bias. The duration of treatment ranged from eight to 12 weeks. All included trials reported the mean change in IPSS‐total score from baseline.

Figure 3

Forest plot of comparison: 1 PDEI versus placebo, outcome: 1.1 Change in IPSS‐total: main analysis.

Figure 4

Funnel plot of comparison: 1 PDEI versus placebo, outcome: 1.1 Change in IPSS‐total: main analysis.

BPHII score

PDEIs may result in a lower BPHII score (MD 0.52 lower, 95% CI 0.71 lower to 0.33 lower; eight trials; low‐quality evidence; Analysis 1.6; Figure 5; summary of findings Table for the main comparison). We downgraded the quality of evidence for study limitations and imprecision, assuming an MCID of 0.5 (Barry 1995).

Figure 5

Forest plot of comparison: 1 PDEI versus placebo, outcome: 1.6 Change in BPHII: main analysis.

Adverse events

PDEIs may increase AEs (RR 1.42, 95% CI 1.21 to 1.67; 11 trials; low‐quality evidence; Analysis 1.10; Figure 6; summary of findings Table for the main comparison). This corresponds to 95 more AEs per 1000 men (95% CI 47 more to 151 more). We downgraded the quality of evidence for study limitations and clinically important inconsistency. We were concerned about possible publication bias but decided not to downgrade further.

Figure 6

Forest plot of comparison: 1 PDEI versus placebo, outcome: 1.10 Any AE with short‐term use.

Secondary outcomes

IPSS‐QOL score

PDEI probably results in a small improvement in IPSS‐QOL (MD 0.26 lower, 95 CI 0.35 lower to 0.16 lower; nine trials; moderate‐quality evidence; Analysis 1.15; Figure 7). We downgraded the quality of evidence for study limitations.

Figure 7

Change in IPSS‐QOL with PDEI versus placebo

Acute urinary retention

Six of 11 RCTs did not report AUR. Reporting of AUR in five other RCTs was not uniform across PDEI and placebo groups (Egerdie 2012; Porst 2011; Roehrborn 2008; Takeda 2012; Yokoyama 2013). Egerdie 2012 reported no AUR observed in any of three treatment groups (tadalafil 2.5, 5 mg, and placebo). Porst 2011 reported no incidence of AUR in the tadalafil 5 mg treatment group, but they did not report the same for the placebo group. Roehrborn 2008 reported that one participant in the placebo group had AUR requiring catheterisation, but this RCT did not mention the incidence in any of the four tadalafil treatment groups. Takeda 2012 reported that one participant (74 years) in the tadalafil 5 mg group reported a severe adverse event (AUR) after 4.5 months of randomisation. This participant required catheterisation and had concomitant prostatitis. Yokoyama 2013 reported that one participant in the placebo group had urinary retention but did not mention the incidence in the tadalafil 2.5 and 5 mg groups.

Surgical interventions

None of the 11 RCTs, including 4386 participants in PDEI or placebo groups, reported the need for surgical intervention during the treatment period.

Other treatment‐related harms

All included RCTs reported AEs as patient‐reported safety outcomes. Treatment‐related AEs most commonly reported were headache, flushing, dizziness, syncope, body pain, myalgia, back pain, nasopharyngitis, dyspepsia, epigastric discomfort, abdominal pain, and diarrhoea.

AEs related to vasodilatation (headache, flushing, dizziness, syncope, orthostatic hypotension, nasopharyngitis)

All 11 RCTs reported these AEs, although the extent of reporting varied. Researchers reported a total of 252 AEs related to vasodilatation in 2757 participants in the PDEI group and 115 in 1629 participants in the placebo group. The risk of AE due to vasodilatation was significantly higher in the PDEI group (RR 1.44, 95% CI 1.18 to 1.76; Table 3).

Open in table viewer

Table 3. Other treatment‐related harms ‐ PDEI versus placebo

Name of the subgroup	Number of RCTs	Summary estimate RR (95% CI)
AEs related to vasodilatation	11	1.44 (1.18 to 1.76)
AEs related to body pain	10	3.44 (1.98 to 5.98)
AEs related to GI tract	8	14.19 (5.80 to 34.72)
AEs leading to treatment discontinuation	10	2.20 (1.38 to 3.51)

AE: adverse event; CI: confidence interval; GI: gastrointestinal; PDEI: phosphodiesterase inhibitor; RCT: randomised controlled trial.

AEs related to body pain (myalgia, body pain, back pain, leg pain)

Ten of 11 RCTs (Dmochowski 2010; Egerdie 2012; Kim 2011; McVary 2007b; Oelke 2012; Porst 2011; Roehrborn 2008; Stief 2008; Takeda 2012; Yokoyama 2013) reported AEs related to body pain. Researchers reported a total of 107 AEs related to body pain in 2658 participants in the PDEI group and 15 in 1449 participants in the placebo group. The RR of developing body pain was significantly higher with PDEI compared to placebo (RR 3.44, 95% CI 1.98 to 5.98; Table 3).

AEs related to GI tract (dyspepsia, gastroesophageal reflux disease (GERD), epigastric discomfort, abdominal pain, diarrhoea)

Eight of 11 RCTs reported AEs related to GI tract (Dmochowski 2010; Kim 2011; McVary 2007a; McVary 2007b; Oelke 2012; Roehrborn 2008; Stief 2008; Takeda 2012). The number of AEs related to the GI tract was 120 in 1884 participants treated with PDEI compared to 2 in 1111 participants treated with placebo. The RR of developing GI‐related AEs was significantly higher with PDEIs compared to placebo (RR 14.19, 95% CI 5.80 to 34.72; Table 3).

Number of treatment discontinuations possibly due to AEs

Ten out of 11 RCTs (Dmochowski 2010; Egerdie 2012; Kim 2011; McVary 2007a; McVary 2007b; Oelke 2012; Porst 2011; Roehrborn 2008; Takeda 2012; Yokoyama 2013) reported possible treatment discontinuation due to AEs. A total of 95 of 2649 participants in the PDEI group withdrew from treatment because of AEs, as did 21 of 1516 in the placebo group. The risk of treatment discontinuation due to AEs was higher in the PDEI group (RR 2.20, 95% CI 1.38 to 3.51; Table 3).

Subgroup analyses

In participants with or without ED

IPSS‐total score

Data were available from five RCTs for participants with ED (MD ‐2.55, 95% CI ‐3.33 to ‐1.77; n = 1486) (Egerdie 2012; McVary 2007a; Oelke 2012; Porst 2011; Roehrborn 2008), and three RCTs provided data for participants without ED (MD ‐2.10, 95% CI ‐3.39 to ‐0.81; n = 330) (Oelke 2012; Porst 2011; Roehrborn 2008). The test for interaction was not significant (P = 0.56; Analysis 1.2; Figure 8).

Figure 8

IPSS‐total subgroup analysis (PDEI vs placebo).

BPHII score

Data were available from five RCTs for participants with ED (MD ‐0.81, 95% CI ‐1.12 to ‐0.49; n = 1480) (Egerdie 2012; Oelke 2012; Porst 2011; Roehrborn 2008; McVary 2007a), and three RCTs provided data for participants without ED (MD ‐0.40, 95% CI ‐0.98 to 0.18; n = 329) (Oelke 2012; Porst 2011; Roehrborn 2008). The test for interaction was not significant (P = 0.23; Analysis 1.7; Figure 9).

Figure 9

Summary forest plot of subgroups for change in BPHII (PDEI Vs Placebo)

Adverse events

Data were available from four RCTs for participants with ED (McVary 2007a; Oelke 2012; Porst 2011; Roehrborn 2008), and three RCTs provided data for participants without ED (Oelke 2012; Porst 2011; Roehrborn 2008). The RR for AEs with PDEIs for participants with ED was 1.36 (95% CI 1.11 to 1.68; n = 1120), and the RR for AEs for participants without ED was 1.73 (95% CI 1.33 to 2.24; n = 338). The test for interaction was not significant (P = 0.16; Analysis 1.11; Figure 6).

Different PDEIs (tadalafil, sildenafil, vardenafil)

Nine of 11 RCTs compared tadalafil versus placebo (Dmochowski 2010; Egerdie 2012; Kim 2011; McVary 2007b; Oelke 2012; Porst 2011; Roehrborn 2008; Takeda 2012; Yokoyama 2013). One RCT used sildenafil (McVary 2007a), and one RCT used vardenafil (Stief 2008).

IPSS‐total score

The MD for improvement in IPSS‐total score for tadalafil was ‐1.80 (95% CI ‐2.20 to ‐1.39; n = 3738), for sildenafil ‐4.40 (95% CI ‐6.96 to ‐1.84; n = 341), and for vardenafil ‐2.20 (95% CI ‐3.57 to ‐0.83; n = 315). The test for interaction was not significant (P = 0.13; Analysis 1.3;Figure 9).

BPHII score

The MD for improvement in BPHII was ‐0.50 (95% CI ‐0.67 to ‐0.33; n = 3738; seven RCT) with tadalafil, and ‐1.10 (95% CI ‐2.11 to ‐0.09; n = 341; one RCT) with sildenafil. No data were available for vardenafil. The test for interaction was not significant (P = 0.25; Analysis 1.8;Figure 9).

Adverse events

The MD in the RR of any AE for tadalafil was 1.44 (95% CI 1.19 to 1.75; n = 3796; nine RCT), for sildenafil 1.22 (95% CI 0.99 to 1.51; n = 341), and for vardenafil 1.86 (95% CI 1.11 to 3.11; n = 315). The test for interaction was not significant (P = 0.25; Analysis 1.12;Figure 9).

Different doses of PDEIs

Nine RCTs compared tadalafil to placebo (Dmochowski 2010; Egerdie 2012; Kim 2011; McVary 2007b; Oelke 2012; Porst 2011; Roehrborn 2008; Takeda 2012; Yokoyama 2013). Researchers administered tadalafil in doses of 2.5, 5, 10, or 20 mg daily; the most commonly used dose of tadalafil was 5 mg per day (seven of nine trials used this dose). Dmochowski 2010 used 20 mg daily, McVary 2007b used a dose escalation pattern starting with 5 mg and reaching up to 20 mg tadalafil but did not report the median dose used. McVary 2007a used sildenafil at a dose of 50 mg per day, titrated to 100 mg per day for participants who could tolerate this. however, this study did not report the median dose of sildenafil. Stief 2008 used vardenafil at a dose of 10 mg twice daily.

Study authors analysed dose‐related effects only for tadalafil. Available data were insufficient for assessment of the same for sildenafil and vardenafil (McVary 2007a; Stief 2008).

IPSS‐total score

The MD for improvement in IPSS‐total with tadalafil 2.5 mg/d was ‐1.31 (95% CI ‐2.19 to ‐0.42]; n = 1396) (Egerdie 2012; Roehrborn 2008; Takeda 2012; Yokoyama 2013), with 5 mg/d ‐1.88 (95% CI ‐2.33 to ‐1.42; n = 2452) (Egerdie 2012; Kim 2011; McVary 2007b; Oelke 2012; Porst 2011; Roehrborn 2008; Takeda 2012; Yokoyama 2013), with 10 mg/d ‐3.43 (95% CI ‐4.79 to ‐2.07; n = 426) (Roehrborn 2008), and with 20 mg/d ‐3.29 (95% CI ‐4.44 to ‐2.14; n = 589) (Dmochowski 2010; Roehrborn 2008). The test of interaction was significant (P = 0.008). Investigators observed a dose‐dependent increase in improvement in IPSS‐total score from 2.5 mg/d to 10 mg/d and observed the ceiling in effect size by increasing the dose up to 20 mg/d (Analysis 1.4; Figure 8).

BPHII score

The MD for improvement in IPSS‐total with tadalafil 2.5 mg/d was ‐0.28 (95% CI ‐0.61 to 0.04; n = 997) (Egerdie 2012; Roehrborn 2008; Yokoyama 2013), with 5 mg/d ‐0.47 (95% CI ‐0.70 to ‐0.25; n = 2139) (Egerdie 2012; Kim 2011; McVary 2007b; Oelke 2012; Porst 2011; Roehrborn 2008; Takeda 2012; Yokoyama 2013), with 10 mg/d ‐0.58 (95% CI ‐1.15 to ‐0.01; n = 426) (Roehrborn 2008), and with 20 mg/d ‐0.62 (95% CI ‐1.2 to ‐0.09) (Dmochowski 2010; Roehrborn 2008). Researchers noted an indication of a dose‐dependent improvement in BPHII for tadalafil from 2.5 mg/d through 20 mg/d; but the test for interaction was non‐significant (P = 0.64; Analysis 1.9; Figure 9).

Adverse events

The RR of any AE with tadalafil 2.5 mg/d was 1.23 (95% CI 1.03 to 1.46; n = 1405), with 5 mg/d 1.35 (95% CI 1.12 to 1.63; n = 2471), with 10 mg/d 1.63 (95% CI 1.19 to 2.24; n = 427), and with 20 mg/d 1.93 (95% CI 1.53 to 2.44; n = 620). The test for interaction was significant (P = 0.02), suggesting an increase in AE with increasing doses of tadalafil.

Sensitivity analyses

We conducted sensitivity analysis for RCTs with placebo or an active drug run‐in period before the randomised intervention was provided. We included eight RCTs in the sensitivity analysis for reduction in IPSS‐total (Egerdie 2012; Kim 2011; McVary 2007b; Oelke 2012; Porst 2011; Roehrborn 2008; Takeda 2012; Yokoyama 2013), and the MD between PDEI and placebo was similar to the main analysis (‐1.84, 95% CI ‐2.30 to ‐1.39; n = 3567; Analysis 1.5).

We did not conduct a sensitivity analysis for BPHII, as it was the same as in the main analysis (all studies included a run‐in period).

Sensitivity analysis for AEs included eight RCTs (Egerdie 2012; Kim 2011; McVary 2007b; Oelke 2012; Porst 2011; Roehrborn 2008; Takeda 2012; Yokoyama 2013), and the RR of AEs with PDEIs versus placebo was similar to the main analysis (1.36, 95% CI 1.13 to 1.65; n = 3956; Analysis 1.14).

Overall, we detected no differences between the effect sizes obtained in sensitivity analyses and in the main analyses.

Long‐term (up to one year) efficacy and safety of PDEI compared to placebo

The maximum duration of RCTs comparing PDEI to placebo was 12 weeks. No long‐term comparative effectiveness data were available for any outcomes.

PDEI versus AB

Four RCTs comprising 936 participants compared PDEI monotherapy versus AB monotherapy (Kim 2011; Oelke 2012; Tuncel 2010; Yokoyama 2013). Three RCTs used tadalafil, and one used sildenafil (Tuncel 2010). Duration of treatment ranged from eight to 12 weeks.

Short‐term (up to 12 weeks) efficacy and safety of PDEI compared to AB

Primary outcomes

IPSS‐total score

There is probably no difference between PDEI and AB for improvement in IPSS scores (MD 0.35 higher, 95% CI 0.28 lower to 0.98 higher; four trials; moderate‐quality evidence; Analysis 2.1; summary of findings Table 2). We downgraded the quality of evidence for study limitations.

BPHII score

There may be no difference between PDEI and AB for improvement in BPHII scores (MD 0.03 higher, 95% CI 1.1 lower to 1.16 higher; two trials; low‐quality evidence; Analysis 2.5; summary of findings Table 2). We downgraded the quality of evidence for study limitations and clinically important heterogeneity. We assumed the apparent imprecision to be secondary to observed inconsistency and did not downgrade further. Two trials with 403 participants reported this outcome (Kim 2011; Yokoyama 2013).

Adverse events

Researchers most commonly reported AEs such as headache, myalgia, back pain, dizziness, dyspepsia, and nasopharyngitis. Trial results show 100 AEs reported in 397 participants on PDEIs compared to 79 AEs reported in 391 participants in the AB group. Rates of AEs may be similar (RR 1.35, 95% CI 0.80 to 2.30; four trials; low‐quality evidence; Analysis 2.9; summary of findings Table 2). This corresponds to 71 more AEs per 100 men (95% CI 41 fewer to 264 more). We downgraded the quality of evidence for study limitations and clinically important inconsistency. We assumed the apparent imprecision to be secondary to observed inconsistency and did not downgrade further.

Secondary outcomes

IPSS‐QOL score

PDEIs and ABs may have similar effects on IPSS‐QOL (MD 0.1 lower, 95% CI 0.48 lower to 0.29 higher; four RCT; low‐quality evidence; Analysis 2.10). We downgraded the quality of evidence for study limitations and clinically important heterogeneity.

Acute urinary retention

None of the four RCTs comprising 936 participants reported AUR as an event.

Surgical intervention

None of the four RCTs comprising 936 participants reported if there was any need for surgical intervention during the treatment period.

Other treatment‐related harms

AEs related to vasodilatation

Researchers reported a total of 18 AEs related to vasodilatation in the PDEI group among 397 participants and seven AEs in the AB group among 391 participants. Risk of reporting vasodilatation‐related AEs was significantly higher with tadalafil than with tamsulosin (RR 2.42, 95% CI 1.05 to 5.56; n = 746; Analysis 2.14; Table 4).

Open in table viewer

Table 4. Other treatment‐related harms ‐ PDEI versus alpha‐blockers

Name of the subgroup	Number of RCTs	Summary estimate RR (95% CI)
AEs related to vasodilatation	3	2.42 (1.05 to 5.56)
AEs related to body pain	3	1.75 (0.38 to 8.18)
AEs related to GI tract	3	4.87 (0.58 to 41.22)
AEs leading to retreatment discontinuation	3	1.26 (0.31 to 5.05)

AE: adverse event; CI: confidence interval; GI: gastrointestinal; PDEI: phosphodiesterase inhibitor; RCT: randomised controlled trial.

AEs related to body pain
The number of AEs related to body pain was four among 378 participants treated with PDEIs compared to two among 369 participants treated with ABs. The risk of developing body pain was similar in PDEI and AB treatment groups (RR 1.75, 95% CI 0.38 to 8.18; n = 747; Analysis 2.15; Table 4).

AEs related to GI tract

Three of the four RCTs reported AEs related to the upper GI tract (dyspepsia, GERD, epigastric discomfort, diarrhoea). The number of AEs related to the GI tract was four among 397 participants treated with PDEIs compared to zero among 389 participants treated with placebo. The odds of developing GI AEs were similar in PDEI and AB treatment groups (RR 4.87, 95% CI 0.58 to 41.22; n = 746; Analysis 2.16; Table 4).

AEs leading to treatment discontinuation

A total of four of 397 participants in the PDEI group discontinued treatment compared to three of 391 in the AB group. The risk of treatment discontinuation was similar with PDEIs or ABs (RR 1.26, 95% CI 0.31 to 5.05; n = 786; Analysis 2.17; Table 4).

Subgroup analyses

Participants with and without ED

Tuncel 2010 enrolled participants with BPH‐LUTS and ED. The MD obtained in this study was 1.40 (95% CI ‐2.24 to 5.04). The other studies included BPH‐LUTS participants irrespective of their ED status. Therefore, data were insufficient for analysis of any outcomes.

Different PDEI

Results show appreciable heterogeneity among included RCTs with regard to doses of PDEI and AB used. Three RCTs used tadalafil 5 mg daily (Kim 2011; Oelke 2012; Yokoyama 2013), and one used sildenafil four times a week (Tuncel 2010). The dose of AB used was not uniform. Oelke 2012 used 0.4 mg per day of tamsulosin, whereas Kim 2011 and Yokoyama 2013 used 0.2 mg per day of tamsulosin.

Results show no differences between tadalafil and tamsulosin treatment groups for IPSS‐total (MD ‐0.13, 95% CI ‐0.65 to 0.91; n = 742) nor between sildenafil and tamsulosin groups (MD ‐1.40, 95% CI ‐2.24 to 5.04; n = 40; Analysis 2.2).

Findings show no differences between tadalafil and tamsulosin for BPHII (MD 0.23, 95% CI ‐0.23 to 0.69; n = 403; Analysis 2.6) nor for AE (RR 1.35, 95% CI 0.80 to 2.30; n = 742; Analysis 2.9).

Dose effects of PDEIs

Four comparison groups were available.

Tadalafil 5 mg per day versus tamsulosin 0.4 mg per day

We included only one RCT with 336 participants in this analysis (Oelke 2012). Results show no differences between tadalafil 5 mg/d and tamsulosin 0.4 mg/d treatment groups for improvement in IPSS‐total (MD ‐0.60, 95% CI ‐1.98 to 0.78; n = 336; Analysis 2.3), BPHII (MD 0.24, 95% CI ‐0.40 to 0.88; n = 700; Analysis 2.7), and IPSS‐QOL (MD ‐0.30, 95% CI ‐0.02 to 0.58; n = 336; Analysis 2.12).

Tadalafil 5 mg per day versus tamsulosin 0.2 mg per day

We included two RCTs in this comparison (Kim 2011; Yokoyama 2013). Results show no differences between tadalafil 5 mg/d and tamsulosin 0.2 mg/d for improvement in IPSS‐total (MD ‐0.42, 95% CI ‐0.68 to 1.51; n = 406; Analysis 2.3), BPHII (MD 0.04, 95% CI ‐1.13 to 1.21; n = 403; Analysis 2.7), AE (RR 2.41, 95% CI 0.51 to 11.50; n = 406; Analysis 2.9), and IPSS‐QOL (MD 0.11, 95% CI ‐0.37 to 0.59; n = 406; Analysis 2.12).

Tadalafil 2.5 mg per day versus tamsulosin 0.2 mg per day

We included only one RCT in this analysis (Yokoyama 2013). Results show no differences between tadalafil 2.5 mg/d and tamsulosin 0.2 mg/d treatment groups for improvement in IPSS‐total (MD ‐0.70, 95% CI ‐0.41 to 1.81; n = 303; Analysis 2.3), BPHII (MD 0.50, 95% CI ‐0.06 to 1.06; n = 297; Analysis 2.7), AE (RR 1.27, 95% CI 0.91 to 1.76; n = 306; Analysis 2.9), and IPSS‐QOL (MD 0.30, 95% CI 0.02 to 0.58; n = 303; Analysis 2.12).

Sildenafil 25 mg 4 times a week versus tamsulosin 0.4 mg per day

We included only one RCT in this analysis (Tuncel 2010). Results show no differences between sildenafil 25 mg/d and tamsulosin 0.4 mg/d treatment groups for improvement in IPSS‐total (MD ‐1.40, 95% CI ‐2.24 to 5.04; n = 40; Analysis 2.3) nor in IPSS‐QOL (MD ‐0.80, 95% CI ‐1.78 to 0.18; n = 40; Analysis 2.12).

The dosages used across studies were not uniform, and the therapeutic equivalence status of dosages used is not clear. We maintained the standard tadalafil dose (5 mg/d) as a constant factor for comparison of groups 2 and 3, which suggests no differences in any outcome measures.

Sensitivity analyses for trials with run‐in period

Sensitivity analysis included three trials (Kim 2011; Oelke 2012; Yokoyama 2013). Similar to the main analysis, results show no differences between PDEIs and ABs for IPSS‐total (MD 0.07, 95% CI ‐0.88 to 1.02; n = 742; Analysis 2.4); BPHII (MD 0.03, 95% CI ‐1.10 to 1.16; n = 554; Analysis 2.8; Kim 2011; Yokoyama 2013), AEs (RR 1.35, 95% CI 0.80 to 2.30; n = 896), and IPSS‐QOL (MD ‐0.01, 95% CI ‐0.38 to 0.36; n = 742; Analysis 2.13). We found no data for AUR and need for surgical intervention.

Long‐term (up to one year) efficacy and safety of PDEIs compared to ABs

RCTs comparing PDEIs and ABs were 12 weeks in maximum duration. We found no available data for comparative long‐term efficacy and safety of these two classes of drugs used for BPH‐LUTS.

PDEI + AB versus AB alone

Four RCTs comprising 193 participants compared PDEI plus AB combination therapy versus AB monotherapy (Bechara 2008; Gacci 2012; Regadas 2013; Tuncel 2010). Two of these RCTs used tadalafil (Bechara 2008; Regadas 2013), one used sildenafil (Tuncel 2010), and one used vardenafil (Gacci 2012).

Short‐term (up to 12 weeks) efficacy and safety of PDEI combination with AB compared to AB alone

Primary outcomes

IPSS‐total

PDEI in combination with an AB may result in a small improvement in IPSS‐total (MD 2.56 lower, 95% CI 3.92 lower to 1.19 lower; four trials, low‐quality evidence; Analysis 3.1; summary of findings Table 3), but we are unsure of this finding. We downgraded the quality of evidence for study limitations and clinically important imprecision, assuming an MCID of 3.

BPHII score

We could not assess this score because none of the four RCTs included BPHII as an outcome.

Adverse events

AE are probably increased when combination treatment is compared with AB alone (RR 2.81, 95% CI 1.53 to 5.17; four trials; moderate‐quality evidence). We downgraded the quality of evidence for study limitations (Analysis 3.5; summary of findings Table 3).

Secondary outcomes

IPSS‐QOL score

Two of the four RCTs reported IPSS‐QOL scores (Bechara 2008; Tuncel 2010), noting that combination therapy may improve IPSS‐QOL scores slightly (MD ‐1.08, 95% CI ‐1.92 to ‐0.23; n = 94). The quality of evidence was low. We downgraded for study limitations and imprecision (Analysis 3.9).

Acute urinary retention

None of the four RCTs comprising 193 participants reported AUR as an event.

Surgical intervention

None of the four RCTs comprising 193 participants reported if there was any need for surgical intervention for any participants during conduct of the RCT.

Other treatment‐related harms

AEs related to vasodilatation

Study authors reported a total of 27 AEs related to vasodilatation among 77 participants in the PDEI‐AB combination group and four AEs among 77 participants in the AB alone group. Risk of vasodilatation‐related AEs was higher with the combination treatment than with tamsulosin alone (RR 6.58, 95% CI 2.36 to 18.31; n = 154; Analysis 3.13; Table 5).

Open in table viewer

Table 5. Other treatment‐related harms: PDEI + alpha‐blockers versus alpha‐blockers alone

Name of the subgroup	Number of RCTs	Summary estimate RR (95% CI)
AEs related to vasodilatation	3	4.80 (1.93 to 11.91)
AEs related to GI tract	4	6.00 (0.75 to 48.30)
AEs leading to retreatment discontinuation	3	3.00 (0.13 to 70.53)

AE: adverse event; CI: confidence interval; GI: gastrointestinal; PDEI: phosphodiesterase inhibitor; RCT: randomised controlled trial.

AEs related to body pain

None of the trials reported AEs related to body pain (Table 5).

AEs related to upper GI tract

Researchers reported a total of five AEs related to vasodilatation among 77 participants in the PDEI plus AB combination group and no AEs among 77 participants in the AB alone group. The risk of developing GI tract‐related AEs was similar with PDEI and AB combination treatment compared to AB alone (RR 6.57, 95% CI 0.75 to 56.34; n = 154; Analysis 3.14; Table 5).

AEs leading to treatment discontinuation

One participant in the combination treatment group discontinued treatment compared to none in the AB only treatment. Differences between groups were not significant (RR 3.00, 95% CI 0.13 to 70.53; n = 154; Analysis 3.15; Table 5).

Subgroup analyses

Participants with and without ED

Tuncel 2010 enrolled participants with BPH‐LUTS and ED. The other RCTs included BPH‐LUTS participants irrespective of their ED status (Bechara 2008; Regadas 2013; Gacci 2012). However, these studies neither stratified participants based on the presence or absence of ED nor reported outcomes by this subgroup. Therefore, available study data were insufficient for meaningful analysis of any outcomes.

Different PDEIs

Bechara 2008 used tadalafil 20 mg/d and Regadas 2013 5 mg/d; Tuncel 2010 used sildenafil 25 mg 4 times/week, and Gacci 2012.used vardenafil 10 mg/d. All four RCTs used tamsulosin 0.4 mg/d as the AB for comparison. The test for interaction showed no differences between PDEI and tamsulosin combination groups versus tamsulosin alone (P = 0.67).

Dose effects of PDEIs

Researchers observed no dose effect for tadalafil (5 mg/d and 20 mg/d) in combination with tamsulosin (0.4 mg/d) versus tamsulosin (0.4 mg/d) alone (P = 0.52) for improvement in IPSS‐total and AE. We could not assess the dose effect relationship for sildenafil and vardenafil, as available data were insufficient with only a single RCT for each (Analysis 3.3).

Sensitivity analysis

Of the four RCTs, two included a placebo run‐in period for all participants included in the study preceding the randomisation phase, in addition to a drug for the washout period for participants who were previously treated and were not naïve to the study drug (Bechara 2008; Gacci 2012). The magnitude of improvement in IPSS‐total was similar to findings in the main analysis (MD ‐2.34, 95% CI ‐4.02 to ‐0.66; n = 114; Analysis 3.4). The RR of AEs was found to be similar (MD 2.80, 95% CI 1.50 to 5.21; n = 114). Only one study reported improvement in IPSS‐QOL with MD similar to the main analysis (‐0.70, 95% CI ‐2.40 to 1.00; n = 59; Analysis 3.12; Bechara 2008).

Long‐term (up to one year) efficacy and safety of PDEI combination with AB versus AB alone

None of the RCT continued beyond 12 weeks; therefore no data are available for assessing the long‐term efficacy and safety of the combination of PDEI and AB versus AB alone for BPH‐LUTS.

PDEI + AB versus PDEI alone

Only one RCTs qualified for this comparison, with three intervention arms: tamsulosin 0.4 mg daily (n = 20), sildenafil 50 mg four times weekly plus tamsulosin 0.4 mg daily (n = 20), and sildenafil 50 mg four times weekly (Tuncel 2010). For analysis, we included group data from two arms comprising only 40 participants. The duration of the study was eight weeks, and all participants were naïve to the study drugs.

Short‐term (up to 12 weeks) efficacy and safety of PDEI combination with AB compared to PDEI alone

Primary outcomes

IPSS‐total score

PDEI in combination with an AB may result in small improvement in IPSS‐total compared to PDEI alone (MD 2.4 lower, 95% CI 6.47 lower to 1.67 higher; one trial; low‐quality evidence; Analysis 4.1;summary of findings Table 4). We downgraded the quality of evidence for study limitations and clinically important imprecision, assuming an MCID of 3.

BPHII score

Tuncel 2010, did not report this outcome.

Adverse events

Tuncel 2010 reported no withdrawals due to AEs but did not report whether any AEs occurred.

Secondary outcomes

IPSS‐QOL score

PDEI in combination with an AB may result in similar improvement in IPSS‐QOL (MD 0.4 lower, 95% CI 1.8 lower to 1.0 higher; one trial; low‐quality evidence) compared to PDEI alone. We downgraded for study limitations and clinically important imprecision, assuming an MCID of 0.5.

Acute urinary retention

Tuncel 2010 did not report this outcome.

Surgical intervention

Tuncel 2010 also did not report this outcome.

Subgroup analyses

We conducted no subgroup analyses, as the sample size was small and researchers did not report outcomes as per the subgroups.

Sensitivity analysis

We performed no sensitivity analysis, as we included only one study for this comparison.

Long‐term (up to one year) efficacy and safety of PDEI combination with AB compared to PDEI alone

For this comparison, we identified only one RCT that was eight weeks in duration. No other long‐term data were available for this analysis. Hence the evidence is insufficient to determine comparative long‐term efficacy and safety of the PDEI combination with AB versus AB alone.

PDEI + 5‐ARI versus 5‐ARI alone

Short‐term (up to 12 weeks) efficacy and safety of PDEI combination with 5‐ARI compared to 5‐ARI alone

Only one RCT qualified for inclusion in this analysis (Casabe 2014). This study compared tadalafil 5 mg plus finasteride 5 mg daily (n = 345) versus placebo plus finasteride 5 mg daily (n = 350) in a parallel‐group design. All participants were naïve to finasteride therapy. They underwent a four‐week drug‐free washout period followed by a four‐week placebo run‐in period before enrolment. The study was 26 weeks in duration. Study authors assessed the efficacy of treatments for IPSS‐total score at four, 12, and 26 weeks.

Primary outcomes

IPSS‐total score

PDEI in combination with a 5‐ARI is likely to result in a small improvement in IPSS (MD 1.4 lower, 95% CI 2.84 lower to 1.16 lower; one RCT; moderate‐quality evidence) at 12 weeks. We downgraded for study limitations.

BPHII score

The only study identified for this comparison did not evaluate BPHII scores; hence we found no data pertaining to this outcome (Casabe 2014).

Adverse events

Casabe 2014 did not report AEs at 12 weeks.

Secondary outcomes

IPSS‐QOL score

PDEI plus 5‐ARI probably has a similar effect on IPSS‐QOL (MD 0.2 lower, 95% CI 0.48 lower to 0.06 lower; one RCT; moderate‐quality evidence) at 12 weeks compared to 5‐ARI alone. We downgraded the quality of evidence for study limitations.

Acute urinary retention

Casabe 2014 did not report AUR as an event during the study in either of the treatment groups at 12 weeks.

Surgical intervention

Casabe 2014 also did not report if there was any need for surgical intervention for any participants at 12 weeks during conduct of the RCT.

Long‐term (up to one year) efficacy and safety of PDEI combination with 5‐ARI compared to 5‐ARI alone

Primary outcomes

IPSS‐total

The combination of PDEI and 5‐ARI probably leads to minor improvement in IPSS‐total (MD 1.0 lower, 95% CI 1.84 lower to 0.86 lower; one trial; moderate‐quality evidence) compared to 5‐ARI alone after 24 weeks of treatment. We downgraded the quality of evidence for study limitations.

BPHII score

The only study identified for this comparison did not evaluate BPHII scores; hence we found no data pertaining to this outcome (Casabe 2014).

Adverse events

AEs may be similar in the long term (RR 1.07, 95% CI 0.84 to 1.36; one trial; low‐quality evidence). This corresponds to 19 more AEs per 1000 men (95% CI 43 fewer to 98 more). We downgraded the quality of evidence for study limitations and clinically important imprecision.

Secondary outcomes

IPSS‐QOL score

The combination of PDEI and 5‐ARI probably improves IPSS‐QOL to a small degree (MD 0.2 lower, 95% CI 0.48 lower to 0.08 lower; one trial; moderate‐quality evidence) at 24 weeks compared to 5‐ARI alone. We downgraded the quality of evidence for study limitations.

Acute urinary retention

Casabe 2014 did not report AUR as an event during the study in either treatment group.

Surgical intervention

Casabe 2014 also did not report if there was any need for surgical intervention for any participants during conduct of the RCT.

Subgroup analyses

We were unable to conduct any of the preplanned subgroup analyses, as no data were available.

Sensitivity analysis

We did not perform the sensitivity analyses, as we identified only a single RCT for this comparison.

Other treatment‐related harms

Results show a higher number of body pain‐related AEs (RR 2.70, 95% CI 1.07 to 6.81) in the combination treatment group (Analysis 5.5), but show no difference in AEs related to vasodilatation (RR 1.01, 95% CI 0.46 to 2.22), upper GI symptoms (RR 4.05, 95% CI 0.87 to 18.92), and treatment discontinuation (RR 0.63, 95% CI 0.21 to 1.91) between groups (Analysis 5.4; Analysis 5.6; Analysis 5.7). Data also show no difference in AEs related to sexual function (RR 0.51, 95% CI 0.17 to 1.4) between the finasteride and tadalafil combination group and the finasteride and placebo group (Analysis 5.8;Table 6).

Open in table viewer

Table 6. Other treatment‐related harms: PDEI + 5‐ARI versus 5‐ARI alone

Name of the subgroup	Number of RCTs	Summary estimate RR (95% CI)
AEs related to vasodilatation	1	1.01 (0.46 to 2.22)
AEs related to body pain	1	2.70 (1.07 to 6.81)
AEs related to GI tract	1	4.05 (0.87 to 18.92)
AEs leading to retreatment discontinuation	1	0.63 (0.21 to 1.91)
AEs related to sexual function	1	0.51 (0.17 to 1.46)

5‐ARI: 5‐alpha reductase inhibitor; AE: adverse event; CI: confidence interval; GI: gastrointestinal; PDEI: phosphodiesterase inhibitor; RCT: randomised controlled trial.

Discusión

disponible en

Resumen de los resultados principales

Se encontraron datos de ensayos controlados aleatorios (ECA) elegibles para realizar el análisis de cinco comparaciones, con datos limitados en gran parte a un límite de tiempo breve de 12 semanas de tratamiento. El predominio de la evidencia concierne al uso de los inhibidores de la fosfodiesterasa (IFDE) en comparación con placebo.

En comparación con placebo, los IFDE probablemente mejoran la International Prostate Symptom Score (IPSS)‐total en un grado ligeramente superior a placebo con el uso a corto plazo durante seis a 12 semanas. También pueden mejorar ligeramente la calidad de vida (CdV) y dar lugar a una tasa más alta de eventos adversos (EA). No se encontraron datos a largo plazo sobre el tratamiento después de las 12 semanas.

El uso de IFDE y alfabloqueantes (AB) en la hiperplasia prostática benigna (HPB) probablemente proporciona una mejoría similar en la puntuación IPSS‐total y puede dar lugar a una mejoría similar en la CdV. La incidencia de EA puede ser similar. No se encontraron datos a largo plazo sobre el tratamiento después de las 12 semanas.

La combinación de IFDE y AB puede dar lugar a una mejoría pequeña en la puntuación IPSS‐total en comparación con cualquier fármaco solo. No se encontraron datos sobre el Benign Prostatic Hyperplasia Impact Index (BPHII) ni sobre los EA. Los efectos sobre la IPSS‐CdV pueden ser similares. Los resultados de los estudios estuvieron limitados a una duración del tratamiento de cuatro a 12 semanas.

Para el uso a corto plazo (hasta 12 semanas), la combinación de un IFDE y un inhibidor de la 5‐alfa reductasa (5‐ARI) probablemente da lugar a una mejoría muy pequeña en la puntuación IPSS‐total en comparación con un 5‐ARI solo. No se encontró evidencia de las puntuaciones BPHII ni de los EA. Los efectos sobre la IPSS‐CdV probablemente son similares. Con el uso a largo plazo (13 a 26 semanas), la combinación de IFDE y 5 IRA probablemente da lugar a una reducción muy pequeña en la puntuación IPSS‐total. No se encontró evidencia de los efectos sobre las puntuaciones BPHII. Puede no haber diferencias en la tasa de EA y es probable que los efectos sobre la IPSS‐CdV sean similares.

Compleción y aplicabilidad general de las pruebas

Se incluyeron 16 ensayos controlados aleatorios (ECA) que analizaron cinco comparaciones. Las limitaciones son las siguientes.

La mayoría de los estudios fueron estudios de eficacia de duración corta (máximo de 12 semanas) y no aportaron datos a más largo plazo. Este tema es relevante ya que los síntomas urinarios bajos (SUB) por lo general son crónicos y los fármacos habitualmente se utilizan durante años.
Esta revisión se centró en los resultados más importantes para los pacientes que se podrían evaluar de forma significativa en ECA. No se incluyeron estudios no aleatorios, que se consideran importantes para evaluar los resultados a largo plazo y menos frecuentes. La evidencia reciente indica que el uso a largo plazo de la fosfodiesterasa (FDE) puede aumentar el riesgo de melanoma (Deng 2018; Feng 2018; Han 2018; Li 2014; Lian 2016; Loeb 2015; Loeb 2017; Matthews 2016; Pottegard 2016; Tang 2017; Wrishko 2009; Wronski 2014), pero este debate excede los objetivos de esta revisión.
Más del 80% de los participantes en los estudios (8/11 ECA) eran hombres de origen caucásico; del 4% al 6% eran de raza negra y del 8% al 11% asiáticos (en tres ECA que reclutaron pacientes sobre todo en Japón y Corea). Los resultados pueden diferir en los hombres de otros orígenes étnicos.

Calidad de la evidencia

A pesar del gran número de ECA en las cinco comparaciones analizadas, no se encontró evidencia de alta calidad para ningún resultado. La calidad de la evidencia varió principalmente de moderada a baja, y la evidencia de calidad baja representa la categoría más frecuente. Las razones más frecuentes para disminuir la calificación de la calidad fueron las limitaciones de los estudios, aunque ningún dominio se consideró con alto riesgo de sesgo; predominó el riesgo incierto de sesgo relacionado con la selección, la realización, la detección y a veces el sesgo de informe selectivo, que en conjunto dio lugar a la disminución de la calidad en un nivel. Otras razones para disminuir la calificación de calidad fueron la imprecisión clínicamente importante (cuando se cruzó la diferencia mínima clínicamente importante [DMCI] u otro umbral adoptado de relevancia clínica) y la asimetría del gráfico en embudo, lo que plantea problemas de sesgo de publicación. En general, lo anterior indica que la confianza en la mayoría de las estimaciones del efecto es limitada; el efecto verdadero puede ser (o es probable que sea) significativamente diferente de la estimación del efecto informada.

Sesgos potenciales en el proceso de revisión

Durante la realización de la revisión, hubo varias desviaciones conscientes del protocolo con respecto a la inclusión de solamente los ECA que informaran el uso de algún cegamiento y se desestimaron algunos resultados secundarios y análisis de subgrupos preplanificados. Estas modificaciones se efectuaron para reducir la duración de la revisión, mejorar la legibilidad y centrarse en aspectos que son muy importantes para la práctica clínica. Se piensa que es poco probable que esta decisión haya introducido sesgo en el proceso de esta revisión sistemática.

Se realizó una búsqueda bibliográfica exhaustiva y durante la realización de la revisión esporádicamente se actualizó la búsqueda. Aunque el énfasis en el idioma inglés podría ser una limitación potencial, también se buscaron estudios registrados en ClinicalTrials.gov y los registros de ensayos clínicos de la Organización Mundial de la Salud (OMS) para identificar estudios potencialmente elegibles. Por lo tanto, es poco probable que el proceso de revisión hubiera estado sesgado por la omisión de estudios. Se observaron ciertas limitaciones durante el proceso de revisión, como el reto de excluir los estudios y literatura gris en idioma no inglés de la revisión debido a la identificación incompleta de estos estudios. Sin embargo, se trata de una limitación genérica asociada con las revisiones sistemáticas. Otra limitación fue la detección de EA graves y poco frecuentes. Esta revisión sólo incluyó ECA, y sus autores excluyeron otros estudios no aleatorios. Este hecho compromete el alcance para la detección exhaustiva de EA en los estudios poblacionales. La duración más corta de los estudios incluidos también contribuye a la no detección de EA a largo plazo relacionados con el uso.

Esta revisión incluye datos de los ECA incluidos y se observa que falta cierta información de algunos de los estudios incluidos. Pese a los intentos repetidos, no se obtuvieron los datos faltantes de un ECA (Tamimi 2010). No obstante, este hecho no afectó el tamaño del efecto de las estimaciones, ya que en el análisis primario se incluyeron datos de 11 de los 12 ECA.

Acuerdos y desacuerdos con otros estudios o revisiones

En los últimos años se han publicado varias revisiones sistemáticas y descriptivas (narrativas) sobre la eficacia y el uso clínico de los IFDE (específicamente tadalafil) para los SUB‐HPB. Estas revisiones han superado los estudios originales realizados durante la última década (Alsaikhan 2014; Anderson 2007; Brock 2014; Fullhase 2014; Gacci 2011; Govorov 2014; Guliano 2012; Hatzimouratidis 2014; Jiang 2013; Martinez‐Salamanca 2011; Miller 2013; Radomski 2014; Rosso 2015; Silva 2014; Wronski 2014; Ukcert 2013; Yan 2014), y muchos de ellos recomiendan el tratamiento con IFDE para los SUB‐HPB, y pocos manifiestan escepticismo acerca de este tratamiento. Por otro lado, no es la intención de esta revisión realizar recomendaciones acerca del uso clínico apropiado.

Se realizaron cuatro metanálisis para examinar la evidencia que mostró los efectos de los IFDE versus placebo para los SUB‐HPB.

La revisión sistemática más antigua (Laydner 2011) incluyó cuatro ECA (McVary 2007a; McVary 2007b; Roehrborn 2008; Stief 2008; n = 1880). Sólo se trataba de un análisis narrativo y los autores de la revisión señalaron que no fue posible realizar un metanálisis debido a los problemas de heterogeneidad. (Sin embargo, no se encontró heterogeneidad estadística cuando se analizaron los datos; I² = 21%.) Los autores de la revisión concluyeron que los IFDE son útiles en los SUB‐HPB para mejorar los SUB, la función eréctil y la calidad de vida. No se realizó metanálisis y los autores de la revisión estuvieron estrechamente relacionados con el proceso de desarrollo de los fármacos y las pruebas clínicas iniciales.
Otra revisión sistemática (Liu 2011) incluyó cinco ECA (Dmochowski 2010; McVary 2007a; McVary 2007b; Roehrborn 2008; Stief 2008, n = 2074) de los que cuatro compararon tadalafil con placebo (Dmochowski 2010; McVary 2007b; Roehrborn 2008; Stief 2008). Esta revisión sistemática informó la DM para IPSS‐total en ‐2,57 (IC del 95%: ‐3,15 a ‐1,98) para tadalafil y tuvo deficiencias metodológicas con respecto a la realización. Surgió un problema con respecto a la "unidad de análisis" porque el grupo placebo se incluyó cuatro veces, lo que aumentó de manera errónea el tamaño de la muestra y en consecuencia se comprometió la exactitud de los cálculos. El análisis corregido mostró una DM de ‐2,29 (IC del 95%: ‐2,91 a ‐1,68). El tamaño del efecto obtenido en el estudio sobre un tamaño de la muestra más grande (11 ECA; n = 4293) fue menor que el observado en este análisis (‐1,80 versus ‐2,57).
Gacci 2012 b fue una revisión sistemática que incluyó datos de cuatro ensayos clínicos (McVary 2007a; McVary 2007b; Roehrborn 2008; Stief 2008; Tamimi 2010; n = 2749). La DM en la mejoría de IPSS‐total con IFDE sobre placebo informada en este estudio fue ‐2,8 (IC del 95%: ‐3,6 a ‐2,1), mientras que los análisis de una población más grande de pacientes mostró una DM de ‐1,89 (IC del 95%: ‐2,28 a ‐1,64; n = 4293).
El cuarto metanálisis de revisiones sistemáticas (Dong 2013) incluyó siete ECA de tadalafil o placebo (Dmochowski 2010; Egerdie 2012; Kim 2011; McVary 2007b; Oelke 2012; Porst 2011; Roehrborn 2008; n = 2913). Se encontró una media del tamaño del efecto en la IPSS–total de –2,35 (IC del 95%: ‐2,83 a 1,88). Un grupo más grande de participantes del estudio mostró un tamaño más pequeño del efecto.

La presente revisión destaca por el rigor metodológico, que incluye un protocolo publicado a priori; una búsqueda bibliográfica exhaustiva de los trabajos publicados y no publicados, independientemente del idioma de publicación; el uso de los criterios GRADE según resultados; y un énfasis en los resultados y las diferencias clínicamente importantes para los pacientes. Los autores también quieren destacar la política estricta de conflicto de intereses Cochrane, que se asumió, mientras que muchas otras revisiones han sido publicadas por autores de revisiones con conflictos de intereses financieros e intelectuales reconocidos.

Figure 1

Study flow diagram.

Ir a la figura de la revisiónAbrir en una pestaña nueva

Figure 2

Risk of bias summary: review authors' judgements about each risk of bias item for each included study.

Ir a la figura de la revisiónAbrir en una pestaña nueva

Figure 3

Forest plot of comparison: 1 PDEI versus placebo, outcome: 1.1 Change in IPSS‐total: main analysis.

Ir a la figura de la revisiónAbrir en una pestaña nueva

Figure 4

Funnel plot of comparison: 1 PDEI versus placebo, outcome: 1.1 Change in IPSS‐total: main analysis.

Ir a la figura de la revisiónAbrir en una pestaña nueva

Figure 5

Forest plot of comparison: 1 PDEI versus placebo, outcome: 1.6 Change in BPHII: main analysis.

Ir a la figura de la revisiónAbrir en una pestaña nueva

Figure 6

Forest plot of comparison: 1 PDEI versus placebo, outcome: 1.10 Any AE with short‐term use.

Ir a la figura de la revisiónAbrir en una pestaña nueva

Figure 7

Change in IPSS‐QOL with PDEI versus placebo

Ir a la figura de la revisiónAbrir en una pestaña nueva

Figure 8

IPSS‐total subgroup analysis (PDEI vs placebo).

Ir a la figura de la revisiónAbrir en una pestaña nueva

Figure 9

Summary forest plot of subgroups for change in BPHII (PDEI Vs Placebo)

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 1.1

Comparison 1 PDEI versus placebo, Outcome 1 IPSS‐total.

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 1.2

Comparison 1 PDEI versus placebo, Outcome 2 IPSS‐total: subgroup analysis (with or without ED).

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 1.3

Comparison 1 PDEI versus placebo, Outcome 3 IPSS‐total: subgroup analysis (different PDEI).

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 1.4

Comparison 1 PDEI versus placebo, Outcome 4 IPSS‐total: subgroup analysis (different doses of PDEI).

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 1.5

Comparison 1 PDEI versus placebo, Outcome 5 IPSS‐total: sensitivity analysis (RCT without run‐in period excluded).

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 1.6

Comparison 1 PDEI versus placebo, Outcome 6 BPHII.

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 1.7

Comparison 1 PDEI versus placebo, Outcome 7 BPHII: subgroup analysis (with or without ED).

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 1.8

Comparison 1 PDEI versus placebo, Outcome 8 BPHII: subgroup analysis (different PDEI).

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 1.9

Comparison 1 PDEI versus placebo, Outcome 9 BPHII: subgroup analysis (different doses of PDEI).

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 1.10

Comparison 1 PDEI versus placebo, Outcome 10 Any AE with short‐term use.

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 1.11

Comparison 1 PDEI versus placebo, Outcome 11 Any AE with short‐term use: subgroup analysis (with or without ED).

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 1.12

Comparison 1 PDEI versus placebo, Outcome 12 Any AE with short‐term use: subgroup analysis (different PDEI).

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 1.13

Comparison 1 PDEI versus placebo, Outcome 13 Any AE with short‐term use: subgroup analysis (different doses of PDEI).

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 1.14

Comparison 1 PDEI versus placebo, Outcome 14 Any AE with short‐term use: sensitivity analysis (RCT without run‐in period excluded).

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 1.15

Comparison 1 PDEI versus placebo, Outcome 15 IPSS‐QOL.

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 1.16

Comparison 1 PDEI versus placebo, Outcome 16 IPSS‐QOL: subgroup analysis (with or without ED).

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 1.17

Comparison 1 PDEI versus placebo, Outcome 17 IPSS‐QOL: subgroup analysis (different PDEI).

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 1.18

Comparison 1 PDEI versus placebo, Outcome 18 IPSS‐QOL: subgroup analysis (different doses of PDEI).

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 1.19

Comparison 1 PDEI versus placebo, Outcome 19 IPSS‐QOL: sensitivity analysis (RCT without run‐in period excluded).

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 1.20

Comparison 1 PDEI versus placebo, Outcome 20 Other treatment‐related harms: AE with short‐term use related to vasodilatation.

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 1.21

Comparison 1 PDEI versus placebo, Outcome 21 Other treatment‐related harms: AE with short‐term use related to body pain.

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 1.22

Comparison 1 PDEI versus placebo, Outcome 22 Other treatment‐related harms: AE with short‐term use related to upper GI tract.

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 1.23

Comparison 1 PDEI versus placebo, Outcome 23 Other treatment‐related harms: AE with short‐term use leading to treatment discontinuation.

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 2.1

Comparison 2 PDEI versus AB, Outcome 1 IPSS‐total.

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 2.2

Comparison 2 PDEI versus AB, Outcome 2 IPSS‐total: subgroup analysis (different PDEI).

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 2.3

Comparison 2 PDEI versus AB, Outcome 3 IPSS‐total: subgroup analysis (different doses of PDEI).

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 2.4

Comparison 2 PDEI versus AB, Outcome 4 IPSS‐total: sensitivity analysis (RCT without run‐in period excluded).

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 2.5

Comparison 2 PDEI versus AB, Outcome 5 BPHII.

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 2.6

Comparison 2 PDEI versus AB, Outcome 6 BPHII: subgroup analysis (different PDEI).

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 2.7

Comparison 2 PDEI versus AB, Outcome 7 BPHII: subgroup analysis (different doses of PDEI).

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 2.8

Comparison 2 PDEI versus AB, Outcome 8 BPHII: sensitivity analysis (RCT without run‐in period excluded).

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 2.9

Comparison 2 PDEI versus AB, Outcome 9 Any AE with short‐term use.

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 2.10

Comparison 2 PDEI versus AB, Outcome 10 IPSS‐QOL.

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 2.11

Comparison 2 PDEI versus AB, Outcome 11 IPSS‐QOL: subgroup analysis (different PDEIs).

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 2.12

Comparison 2 PDEI versus AB, Outcome 12 IPSS‐QOL: subgroup analysis (different doses of PDEI).

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 2.13

Comparison 2 PDEI versus AB, Outcome 13 IPSS‐QOL: sensitivity analysis (RCT without run‐in period excluded).

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 2.14

Comparison 2 PDEI versus AB, Outcome 14 Other treatment‐related harms: AE with short‐term use related to vasodilatation.

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 2.15

Comparison 2 PDEI versus AB, Outcome 15 Other treatment‐related harms: AE with short‐term use related to body pain.

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 2.16

Comparison 2 PDEI versus AB, Outcome 16 Other treatment‐related harms: AE with short‐term use related to upper GI tract.

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 2.17

Comparison 2 PDEI versus AB, Outcome 17 Other treatment‐related harms: AE with short‐term use leading to treatment discontinuation.

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 3.1

Comparison 3 PDEI + AB versus AB alone, Outcome 1 IPSS‐total.

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 3.2

Comparison 3 PDEI + AB versus AB alone, Outcome 2 IPSS‐total: subgroup analysis (different PDEI).

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 3.3

Comparison 3 PDEI + AB versus AB alone, Outcome 3 IPSS‐total: subgroup analysis (different doses of PDEI).

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 3.4

Comparison 3 PDEI + AB versus AB alone, Outcome 4 IPSS‐total: sensitivity analysis (RCTs without placebo run‐in period excluded).

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 3.5

Comparison 3 PDEI + AB versus AB alone, Outcome 5 Any AE in short‐term use.

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 3.6

Comparison 3 PDEI + AB versus AB alone, Outcome 6 Any AE in short‐term use: subgroup analysis (different PDEI).

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 3.7

Comparison 3 PDEI + AB versus AB alone, Outcome 7 Any AE with short‐term use: subgroup analysis (different doses of PDEI).

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 3.8

Comparison 3 PDEI + AB versus AB alone, Outcome 8 Any AE with short‐term use: sensitivity analysis (RCT without run‐in period excluded).

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 3.9

Comparison 3 PDEI + AB versus AB alone, Outcome 9 IPSS‐QOL.

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 3.10

Comparison 3 PDEI + AB versus AB alone, Outcome 10 IPSS‐QOL: subgroup analysis (different PDEIs).

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 3.11

Comparison 3 PDEI + AB versus AB alone, Outcome 11 IPSS‐QOL: subgroup analysis (different doses of PDEI).

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 3.12

Comparison 3 PDEI + AB versus AB alone, Outcome 12 IPSS‐QOL: sensitivity analysis (RCT without placebo run‐in period excluded).

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 3.13

Comparison 3 PDEI + AB versus AB alone, Outcome 13 Other treatment‐related harms: AE with short‐term use related to vasodilatation.

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 3.14

Comparison 3 PDEI + AB versus AB alone, Outcome 14 Other treatment‐related harms: AE with short‐term use related to upper GI tract.

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 3.15

Comparison 3 PDEI + AB versus AB alone, Outcome 15 Other treatment‐related harms: AE with short‐term use leading to treatment discontinuation.

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 4.1

Comparison 4 PDEI + AB versus PDEI alone, Outcome 1 IPSS‐total.

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 4.2

Comparison 4 PDEI + AB versus PDEI alone, Outcome 2 IPSS‐QOL.

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 5.1

Comparison 5 PDEI + 5‐ARI versus 5‐ARI alone, Outcome 1 IPSS‐total.

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 5.2

Comparison 5 PDEI + 5‐ARI versus 5‐ARI alone, Outcome 2 IPSS‐QOL.

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 5.3

Comparison 5 PDEI + 5‐ARI versus 5‐ARI alone, Outcome 3 Any AE with long‐term use (26 weeks).

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 5.4

Comparison 5 PDEI + 5‐ARI versus 5‐ARI alone, Outcome 4 Other treatment‐related harms: AE with long‐term use (26 weeks) related to vasodilatation.

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 5.5

Comparison 5 PDEI + 5‐ARI versus 5‐ARI alone, Outcome 5 Other treatment‐related harms: AE with long‐term use (26 weeks) related to body pain.

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 5.6

Comparison 5 PDEI + 5‐ARI versus 5‐ARI alone, Outcome 6 Other treatment‐related harms: AE with long‐term use (26 weeks) related to upper GI tract.

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 5.7

Comparison 5 PDEI + 5‐ARI versus 5‐ARI alone, Outcome 7 Other treatment‐related harms: AE with long‐term use (26 weeks) leading to treatment discontinuation.

Ir a la figura de la revisiónAbrir en una pestaña nueva

Analysis 5.8

Comparison 5 PDEI + 5‐ARI versus 5‐ARI alone, Outcome 8 Other treatment‐related harms: AE with long‐term use (26 weeks) related to sexual function.

Ir a la figura de la revisiónAbrir en una pestaña nueva

Summary of findings for the main comparison. PDEI compared to placebo for lower urinary tract symptoms consistent with benign prostatic hyperplasia

PDEI compared to placebo for LUTS consistent with BPH for short‐term use (up to 12 weeks) Patient: men with LUTS consistent with BPH Setting: outpatient Intervention: PDEI Comparator: placebo
Outcomes	No. of participants (studies) Follow‐up	Certainty of the evidence (GRADE)	Relative effect (95% CI)	*Anticipated absolute effects (95% CI)**
				Risk with placebo	Risk difference with PDEI
IPSS‐total score Assessed with International Prostate Symptom Score Scale from 0 (best: not at all) to 35 (worst: almost always) Follow up: range 6 to 12 weeks	4293 (11 RCTs)	⊕⊕⊕⊝ LOW^a,b	‐	Mean change in IPSS‐total scores ranged from ‐5.1 to ‐1.2	MD 1.89 lower (2.27 lower to 1.5 lower)
BPHII score Assessed with BPHII questionnaire Scale from 0 (best: none) to 13 (worst: a lot or all the time) Follow‐up: range 6 to 12 weeks	3646 (8 RCTs)	⊕⊕⊝⊝ LOW^a,c	‐	Mean change in BPHII scores ranged from ‐2 to ‐0.6	MD 0.52 lower (0.71 lower to 0.33 lower)
Adverse events Assessed with any event Follow‐up: range 6 to 12 weeks	4386 (11 RCTs)	⊕⊕⊝⊝ LOW^a,b,d	RR 1.42 (1.21 to 1.67)	Study population
				226 per 1000	95 more per 1000 (47 more to 151 more)
*The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). BPH: benign prostatic hyperplasia; BPHII: Benign Prostatic Hyperplasia Impact Index; CI: confidence interval; IPSS: International Prostate Symptom Score; LUTS: lower urinary tract symptoms; MCID: minimum clinically important difference; MD: mean difference; PDEI: phosphodiesterase inhibitor; RCT: randomised controlled trial; RR: risk ratio.
GRADE Working Group grades of evidence. High certainty: we are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect. Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect.
^aDowngraded for study limitations. ^bDowngraded for possible publication bias based on funnel plot asymmetry. ^cClinically important imprecision; 95% CI crosses MCID of 0.5. ^dClinically important heterogeneity with I² of 50%.

Summary of findings for the main comparison. PDEI compared to placebo for lower urinary tract symptoms consistent with benign prostatic hyperplasia

Ir a la tabla de la revisión

Summary of findings 2. PDEI compared to alpha‐blockers for LUTS associated with BPH

PDEI compared to alpha‐blockers for LUTS consistent with BPH for short‐term use (up to 12 weeks) Patient: men with LUTS consistent with BPH Setting: outpatient Intervention: PDEI Comparator: alpha‐blockers
Outcomes	No. of participants (studies) Follow‐up	Certainty of the evidence (GRADE)	Relative effect (95% CI)	*Anticipated absolute effects (95% CI)**
				Risk with alpha‐blockers	Risk difference with PDEI
IPSS‐total score Assessed with International Prostate Symptom Score Scale from 0 (best: not at all) to 35 (worst: almost always) Follow‐up: range 8 to 12 weeks	933 (4 RCTs)	⊕⊕⊕⊝ MODERATE^a	‐	Mean change in IPSS‐total scores ranged from ‐5.7 to ‐ 5.35	MD 0.22 higher (0.49 lower to 0.93 higher)
BPHII score Assessed with BPHII questionnaire Scale from 0 (best: none) to 13 (worst: a lot or all the time) Follow‐up: mean 12 weeks	550 (2 RCTs)	⊕⊕⊝⊝ LOW^a,b,c	‐	Mean change in BPHII score was ‐1.6	MD 0.03 higher (1.1 lower to 1.16 higher)
Adverse events Follow‐up: range 8 to 12 weeks	786 (4 RCTs)	⊕⊕⊝⊝ LOW^a,b,c	RR 1.35 (0.80 to 2.30)	Study population
				203 per 1000	71 more per 1000 (41 fewer to 264 more)
*The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). BPH: benign prostatic hyperplasia; BPHII: Benign Prostatic Hyperplasia Impact Index; CI: confidence interval; IPSS: International Prostate Symptom Score; LUTS: lower urinary tract symptoms; MD: mean difference; PDEI: phosphodiesterase inhibitor; RCT: randomised controlled trial; RR: risk ratio.
GRADE Working Group grades of evidence. High certainty: we are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect. Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect.
^aDowngraded for study limitations. ^bClinically important heterogeneity. ^cClinically important imprecision likely due to inconsistency.

Summary of findings 2. PDEI compared to alpha‐blockers for LUTS associated with BPH

Ir a la tabla de la revisión

Summary of findings 3. PDEI plus alpha‐blocker compared to alpha‐blockers alone for LUTS associated with BPH (short‐term)

Combination of PDEI with alpha‐blocker compared to alpha‐blockers alone for LUTS consistent with BPH for short‐term use (up to 12 weeks) Patient: men with LUTS consistent with BPH Setting: outpatient Intervention: PDEI + alpha‐blockers Comparator: alpha‐blockers
Outcomes	No. of participants (studies) Follow‐up	Certainty of the evidence (GRADE)	Relative effect (95% CI)	*Anticipated absolute effects (95% CI)**
				Risk with alpha‐blockers alone	Risk difference with PDEI plus alpha‐blocker
IPSS‐total score Assessed with International Prostate Symptom Score Scale from 0 (best: not at all) to 35 (worst: almost always) Follow‐up: range 4 to 12 weeks	193 (4 RCTs)	⊕⊕⊝⊝ LOW^a,b	‐	Mean change in IPSS‐total scores ranged from ‐6.7 to ‐3.62	MD 2.56 lower (3.92 lower to 1.19 lower)
BPHII score Assessed with BPHII questionnaire Scale from 0 (best: none) to 13 (worst: a lot or all the time)		‐	‐	‐	‐
Adverse effects Follow‐up: range 4 to 12 weeks	194 (4 RCTs)	⊕⊕⊕⊝ MODERATE^a	RR 2.81 (1.53 to 5.17)	Study population
				130 per 1000	235 more per 1000 (69 more to 542 more)
*The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). BPH: benign prostatic hyperplasia; BPHII: Benign Prostatic Hyperplasia Impact Index; CI: confidence interval; IPSS: International Prostate Symptom Score; LUTS: lower urinary tract symptoms; MCID: maximum clinically important difference; MD: mean difference; PDEI: phosphodiesterase inhibitor; RCT: randomised controlled trial; RR: risk ratio.
GRADE Working Group grades of evidence. High certainty: we are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect. Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect.
^aDowngraded for study limitations. ^bClinically important imprecision; 95% CI crosses MCID of 3.0

Summary of findings 3. PDEI plus alpha‐blocker compared to alpha‐blockers alone for LUTS associated with BPH (short‐term)

Ir a la tabla de la revisión

Summary of findings 4. PDEI plus an alpha‐blocker compared to PDEI alone for LUTS associated with BPH (short‐term)

Combination of PDEI with an alpha‐blocker compared to PDEI alone for LUTS consistent with BPH for short‐term use (up to 12 weeks) Patient: men with LUTS consistent with BPH Setting: outpatient Intervention: PDEI + alpha‐blockers Comparator: PDEI
Outcomes	No. of participants (studies) Follow‐up	Certainty of the evidence (GRADE)	Relative effect (95% CI)	*Anticipated absolute effects (95% CI)**
				Risk with PDEI alone	Risk difference with PDEI plus an alpha‐blocker
IPSS‐total score Assessed with International Prostate Symptom Score Scale from 0 (best: not at all) to 35 (worst: almost always) Follow‐up: mean 4 weeks	40 (1 RCT)	⊕⊕⊝⊝ LOW^a,b	‐	Mean change in IPSS‐total score was ‐3.95	MD 2.4 lower (6.47 lower to 1.67 higher)
BPHII score Assessed with BPHII questionnaire Scale from 0 (best: none) to 13 (worst: a lot or all the time)		‐	‐	‐	‐
Adverse events		‐	‐	‐
				‐	‐
*The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). BPH: benign prostatic hyperplasia; BPHII: Benign Prostatic Hyperplasia Impact Index; CI: confidence interval; IPSS: International Prostate Symptom Score; MD: mean difference; PDEI: phosphodiesterase inhibitor; RCT: randomised controlled trial.
GRADE Working Group grades of evidence. High certainty: we are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect. Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect.
^aDowngraded for study limitations. ^bDowngraded for clinically important imprecision; 95% CI crosses MCID of 3.

Summary of findings 4. PDEI plus an alpha‐blocker compared to PDEI alone for LUTS associated with BPH (short‐term)

Ir a la tabla de la revisión

Summary of findings 5. PDEI + 5‐ARI compared to 5‐ARI alone for LUTS associated with BPH (short‐term)

Combination of PDEI with 5‐ARI compared to 5‐ARI alone for LUTS consistent with BPH for short‐term use (up to 12 weeks) Patient: men with LUTS consistent with BPH Setting: outpatient Intervention: PDEI + 5‐ARI Comparator: 5‐ARI
Outcomes	No. of participants (studies) Follow‐up	Certainty of the evidence (GRADE)	Relative effect (95% CI)	*Anticipated absolute effects (95% CI)**
				Risk with 5‐ARI alone	Risk difference with PDEI + 5‐ARI
IPSS‐total score Assessed with International Prostate Symptom Score Scale from 0 (best: not at all) to 35 (worst: almost always) Follow‐up: mean 12 weeks	695 (1 RCT)	⊕⊕⊕⊝ MODERATE^a,b	‐	Mean change in IPSS‐total score was ‐3.8	MD 1.4 lower (2.24 lower to 0.56 lower)
BPHII score Assessed with BPHII questionnaire Scale from 0 (best: none) to 13 (worst: a lot or all the time)		‐	‐	‐
					‐
Adverse events		‐	‐	‐
					‐
*The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). 5‐ARI: 5‐alpha reductase inhibitor; BPH: benign prostatic hyperplasia; BPHII: Benign Prostatic Hyperplasia Impact Index; CI: confidence interval; IPSS: International Prostate Symptom Score; LUTS: lower urinary tract symptoms; MD: mean difference; PDEI: phosphodiesterase inhibitor; RCT: randomised controlled trial.
GRADE Working Group grades of evidence. High certainty: we are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect. Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect.
^aDowngraded for study limitations. ^bDowngraded for clinically important imprecision.

Summary of findings 5. PDEI + 5‐ARI compared to 5‐ARI alone for LUTS associated with BPH (short‐term)

Ir a la tabla de la revisión

Summary of findings 6. PDEI + 5‐ARI compared to 5‐ARI alone for LUTS associated with BPH (long‐term)

Combination of PDEI with 5‐ARI compared to 5‐ARI alone for LUTS consistent with BPH in long‐term use (26 weeks) Patient: men with LUTS consistent with BPH Setting: outpatient Intervention: PDEI + 5‐ARI Comparator: 5‐ARI
Outcomes	No. of participants (studies) Follow‐up	Certainty of the evidence (GRADE)	Relative effect (95% CI)	*Anticipated absolute effects (95% CI)**
				Risk with 5‐ARI alone	Risk difference with PDEI + 5‐ARI
IPSS‐total score Assessed with International Prostate Symptom Score Scale from 0 (best: not at all) to 35 (worst: almost always) Follow‐up: mean 26 weeks	695 (1 RCT)	⊕⊕⊕⊝ MODERATE^a,b	‐	Mean change in IPSS‐total score was ‐4.5	MD 1 lower (1.83 lower to 0.17 lower)
BPHII score Assessed with BPHII questionnaire Scale from 0 (best: none) to 13 (worst: a lot or all the time)		‐	‐

Adverse events Follow‐up: mean 26 weeks	695 (1 RCT)	⊕⊕⊝⊝ LOW^a,b	RR 1.07 (0.84 to 1.36)	Study population
				271 per 1000	19 more per 1000 (43 fewer to 98 more)
*The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). 5‐ARI: 5‐alpha reductase inhibitor; BPH: benign prostatic hyperplasia; BPHII: Benign Prostatic Hyperplasia Impact Index; CI: confidence interval; IPSS: International Prostate Symptom Score; LUTS: lower urinary tract symptoms; MD: mean difference; PDEI: phosphodiesterase inhibitor; RCT: randomised controlled trial; RR: risk ratio.
GRADE Working Group grades of evidence. High certainty: we are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect. Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect.
^aDowngraded for study limitations. ^bDowngraded for clinically important imprecision.

Summary of findings 6. PDEI + 5‐ARI compared to 5‐ARI alone for LUTS associated with BPH (long‐term)

Ir a la tabla de la revisión

Table 1. Description of interventions

Study (year)	Intervention(s) (route, frequency, total dose/d)	Intervention(s) appropriate as applied in a clinical practice setting^a (description)	Comparator(s) (route, frequency, total dose/d)	Comparator(s) appropriate as applied in a clinical practice setting^a (description)
Casabe (2014)	I1: tadalafil 5 mg/d + finasteride 5 mg/d oral	Tadalafil and finasteride were used in this study as per clinical practice setting	C1: placebo + finasteride 5 mg/d oral	Finasteride was used in this study as per clinical practice setting Placebo; N/CPS
Regadas (2013)	I1: tadalafil 5 mg/d + tamsulosin 0.4 mg/d oral	Tadalafil and tamsulosin were used in this study as per clinical practice setting	C1: tamsulosin 0.4 mg/d + placebo oral	Tamsulosin was used in this study as per clinical practice setting Placebo; N/CPS
Brock (2013a)	I1: tadalafil 5 mg/d oral	Tadalafil was used in this study as per clinical practice setting	C1: placebo once daily oral	N/CPS
Yokoyama (2013)	I1: tadalafil 2.5 mg/d oral I2: tadalafil 5 mg/d oral I3: tamsulosin 0.2 mg/d oral	Tadalafil was used in this study as per clinical practice setting	C1: placebo oral	N/CPS
Egerdie (2012)	I1: tadalafil 2.5 mg/d oral I2: tadalafil 5 mg/d oral	Tadalafil was used in this study as per clinical practice setting	C1: placebo oral	N/CPS
Gacci (2012)	I1: tamsulosin 0.4 mg/d + vardenafil 10 mg/d oral	Tadalafil and vardenafil were used in this study as per clinical practice setting	C1: tamsulosin 0.4 mg/d oral	Tamsulosin was used in this study as per clinical practice setting
Oelke (2012)	I1: tadalafil 5 mg/d oral I2: tamsulosin 0.4 mg/d oral	Tadalafil and tamsulosin were used in this study as per clinical practice setting	C1: placebo oral	N/CPS
Takeda (2012)	I1: tadalafil 2.5 mg/d oral I2: tadalafil 5 mg/d oral	Tadalafil was used in this study as per clinical practice setting	C1: placebo oral	N/CPS
Kim (2011)	I1: tadalafil 5 mg/d oral I2: tamsulosin 0.2 mg/d oral	Tadalafil and tamsulosin were used in this study as per clinical practice setting	C1: placebo oral	N/CPS
Dmochowski (2010)	I1: tamsulosin 20 mg/d oral	Tadalafil is approved for use in BPH‐LUTS at 2.5 mg and 5 mg per day Tadalafil dose used in this study is higher than that used in the clinical practice setting	C1: placebo oral	N/CPS
Prost (2011)	I1: tadalafil 5 mg/d oral	Tadalafil was used in this study as per clinical practice setting	C1: placebo oral	N/CPS
Tuncel (2010)	I1: sildenafil 25 mg/4 times weekly + tamsulosin 0.4 mg/d oral I2: sildenafil 25 mg/4 times weekly oral	Sildenafil and tamsulosin were used in this study as per clinical practice setting	C1: tamsulosin 0.4 mg/d oral	Tamsulosin was used in this study as per clinical practice setting
Bechara (2008)	I1: tadalafil 20 mg/d + tamsulosin 0.4 mg/d oral	Tadalafil is approved for use in BPH‐LUTS at 2.5 mg and 5 mg per day Tadalafil dose used in this study is higher than that used in clinical practice setting Tamsulosin was used in this study as per clinical practice setting	C1: tamsulosin 0.4 mg/d + placebo oral	Tamsulosin was used in this study as per clinical practice setting Placebo; N/CPS
Bechara (2008)	I2: tamsulosin 0.4 mg/d + placebo oral	Tamsulosin was used in this study as per clinical practice setting For placebo; N/CPS	C2: tadalafil 20 mg/d + tamsulosin 0.4 mg/d oral	Tadalafil is approved for use in BPH‐LUTS at 2.5 mg and 5 mg per day Tadalafil dose used in this study was higher than that used in clinical practice setting Tamsulosin was used in this study as per clinical practice setting
Roehrborn (2008)	I1: tadalafil 2.5 mg/d oral I2: tadalafil 5 mg/d oral I3: tadalafil 10 mg/d oral I4: tadalafil 20 mg/d oral	Tadalafil is approved for use in BPH‐LUTS at 2.5 mg and 5 mg per day This is a dose‐ranging study; tadalafil doses used in this study for I3 and I4 were higher than those used in clinical practice setting	C1: placebo oral	N/CPS
Stief (2008)	I1: vardenafil 10 mg	Vardenafil was used in this study as per clinical practice setting	C1: placebo oral	N/CPS
McVary (2007a)	I1: sildenafil 50 to 100 mg/d oral	Sildenafil was used in this study as per clinical practice setting	C1: placebo oral	N/CPS
McVary (2007b)	I1: tadalafil 5 to 20 mg/d oral	Tadalafil dose used in this study in the dose escalation phase was higher than that used in clinical practice setting	C1: placebo oral	N/CPS
^aThe term 'clinical practice setting' refers to the specification of the intervention/comparator as used in the course of a standard medical treatment (such as dose, dose escalation, dosing scheme, provision for contraindications, and other important features). BPH: benign prostatic hyperplasia; C: comparator; I: intervention; LUTS: lower urinary tract symptoms; N/CPS: no specification of clinical practice setting possible.

Table 1. Description of interventions

Ir a la tabla de la revisión

Table 2. Baseline characteristics

Study (year)	Intervention(s) and comparator(s)	Duration of intervention (duration of follow‐up) (days, months, years...)	Description of participants	Trial period (year to year)	Country	Setting	Ethnic groups (%)	Duration of disease (mean/range years (SD), or as reported)
Casabe (2014)	I1: tadalafil 5 mg/d + finasteride 5 mg/d	12 weeks (for primary endpoint) Study continued maintaining blinding until 26 weeks (for secondary endpoints)	Men > 45 years with BPH‐LUTS; IPSS ≥ 13 Qmax 4 to 15 mL/s Prostatic volume ≥ 30 cc Naive to 5‐ARI therapy	November 2010 to September 2012	USA, Canada, Latin America	‐	Whites (85.5%)	‐
							Black/African American (10.1%)
							Native American (2.5%)
	C1: placebo + finasteride 5 mg/d						Other (3.9%)
Regadas (2013)	I1: tadalafil 5 mg/d + tamsulosin 0.4 mg/d	30 days	Men > 45 years, BOOI > 20 IPSS > 14	October 2010 to September 2011	"‐"	‐	"‐"	‐
	C1: tamsulosin 0.4 mg/d + placebo
Brock (2013a)	I1: tadalafil 5 mg/d	12 weeks	Men ≥ 45 years BPH‐LUTS > 6 months IPSS ≥ 13 Qmax 4 to 15 mL/s	"‐"	‐	‐	‐	‐
	C1: placebo
Yokoyama (2013)	I1: tadalafil 2.5 mg/d	12 weeks	Men > 45 years; BPH‐LUTS > 6 months IPSS > 13 Qmax 4 to 15 mL/s Prostate volume > 20 cc	"‐"	Japan Korea Taiwan	‐	Japan (55.8%)	‐
	I2: tadalafil 5 mg/d						Korea (29.1%)
	I3: tamsulosin 0.2 mg/d						Taiwan (14.8%)
	C1: placebo
Egerdie (2012)	I1: tadalafil 2.5 mg/d	12 weeks	Men ≥ 45 years, ≥ 3 months h/o ED, ≥ 6 months h/o BPH‐LUTS (LUTS + ED)		Europe Mexico USA/Canada	‐	White (93%)	‐
	I2: tadalafil 5 mg/d						Black American (4%)
	C2: placebo						Asian (2.5%)
							Hispanic (15%)
							Other (0.6%)
Gacci (2012)	I1: tamsulosin 0.4 mg/d + vardenafil 10 mg/d	12 weeks	Men with persistent storage/irritative symptoms after tamsulosin 0.4 mg treatment for 2 weeks; IPSS‐storage ≥ 8	"‐"	"‐"	‐	"‐"	‐
	C1: tamsulosin 0.4 mg/d
Oelke (2012)	I1: tadalafil 5 mg/d	12 weeks	Men ≥ 45 years, BPH‐LUTS > 6 months, IPSS ≥ 13, Qmax 4 to 15 mL/s before placebo lead‐in period	December 2009 to January 2011	Australia, Austria, Belgium, France, Germany, Greece, Italy, Mexico, The Netherlands, Poland	‐	White (76%)	‐
	I2: tamsulosin 0.4 mg/d						Black or African American (0.3%)
	C1: placebo						American Indian/Alaska Native (23.1%)
Takeda (2012)	I1: tadalafil 2.5 mg/d	12 weeks, followed by a 42‐week open‐label extension phase	> 45 years, IPSS > 13, Qmax 4 to 15 mL/s, prostate volume > 20 cc	"‐"	Japan	‐	Asian (100%)	‐
	I2: tadalafil 5 mg/d
	C1: placebo
Kim (2011)	I1: tadalafil 5 mg/d	12 weeks	Men ≥ 45 years, h/o BPH‐LUTS for 6 months or longer, IPSS ≥ 13, Qmax > 5 and < 15 after the washout period	"‐"	Korea	‐	Asian (100%)	‐
	I2: tamsulosin 0.2 mg/d
	C1: placebo
Dmochowski (2010)	I1: tamsulosin 20 mg/d	12 weeks	Men > 40 years, longer than 6 months h/o BPH‐LUTS; IPSS ≥ 13, with or without BOO	"‐"	USA, Canada	‐	White (76%)	‐
	C1: placebo						Black American (12.5%)
							Hispanic (8%)
							Other (3.0%)
Prost (2011)	I1: tadalafil 5 mg/d	12 weeks	Men > 45 years, BPH‐LUTS > 6 months IPSS > 13 and Qmax 4 to 15 mL/s	"‐"	Argentina, Germany, Italy, Mexico, USA	‐	White (91%)	‐
	C1: placebo						Black or African American (2.5%)
							American Indian or Alaska Native (5.2%)
							Other (1%)
Tuncel (2010)	I1: sildenafil 25 mg/4 times weekly + tamsulosin 0.4 mg/d	12 weeks	IPSS > 12, IIEF < 21 (SIHM < 21) treatment‐naive and all participants had ED	‐	‐	‐	‐	‐
	I2: sildenafil 25 mg/4 times weekly
	C1: placebo
Bechara (2008)	I1: tadalafil 20 mg/d + tamsulosin 0.4 mg/d	45 days	Men ≥ 50 years, h/o BPH > 6 months, IPSS ≥ 12, ⁶PSA ≤ 4 ng/mL, ²Qmax > 5 and < 15 for a minimum voided volume of 125 mL	September 2007 to February 2008	Argentina	‐	‐	‐
	I2: tamsulosin 0.4 mg/d + placebo
Roehrborn (2008)	I1: tadalafil 2.5 mg/d	12 weeks	Men > 45 years, BPH‐LUTS > 6 months, IPSS > 13, Qmax 4 to 15 mL/s	‐	North America, Europe, Australia	‐	White (85%)	‐
	I2: tadalafil 5 mg/d						Hispanic (11.5%)
	I3: tadalafil 10 mg/d						Black American (2.42%)
	I4: tadalafil 20 mg/d						Other (0.5%)
Stief (2008)	I1: vardenafil 10 mg	12 weeks	Men > 45 to 64 years, BPH‐LUTS > 6 months, IPSS > 12	October 2005 to June 2006	Germany	‐	White (99%)	‐
							Black Americans (0.5%)
	C1: placebo						Other (0.5%)
McVary (2007b)	I1: tadalafil 5 mg/d	12 weeks	Men > 45 years, h/o BPH‐LUTS 6 months with IPSS > 13, Qmax > 4 to < 15 mL/s for a voided volume of 125 mL	November 2004 to July 2005	USA	‐	White (81%)	‐
	C1: placebo						Black American (9.5%)
							Hispanic (6.5%)
							Other (2.5%)
McVary (2007a)	I1: sildenafil 50 to 100 mg/d	12 weeks	Men ≥ 45 years, with IPSS > 12 and IIEF score < 25 All participants had ED	March 2004 to May 2005	USA	‐	White (82%)	‐
	C1: placebo						Black American (11.5%)
							Asian (2%)
							Other (4%)
‐ denotes not reported; "‐" denotes missing information. BOO: bladder outflow obstruction; BOOI: Bladder Outflow Obstruction Index; C: comparator; I: intervention; IIEF: International Index of Erectile Function; PSA: prostate‐specific antigen; Qmax: maximum urinary flow rate; SD: standard deviation; SHIM: Sexual Health Inventory for Males.

Table 2. Baseline characteristics

Ir a la tabla de la revisión

Table 3. Other treatment‐related harms ‐ PDEI versus placebo

Name of the subgroup	Number of RCTs	Summary estimate RR (95% CI)
AEs related to vasodilatation	11	1.44 (1.18 to 1.76)
AEs related to body pain	10	3.44 (1.98 to 5.98)
AEs related to GI tract	8	14.19 (5.80 to 34.72)
AEs leading to treatment discontinuation	10	2.20 (1.38 to 3.51)
AE: adverse event; CI: confidence interval; GI: gastrointestinal; PDEI: phosphodiesterase inhibitor; RCT: randomised controlled trial.

Table 3. Other treatment‐related harms ‐ PDEI versus placebo

Ir a la tabla de la revisión

Table 4. Other treatment‐related harms ‐ PDEI versus alpha‐blockers

Name of the subgroup	Number of RCTs	Summary estimate RR (95% CI)
AEs related to vasodilatation	3	2.42 (1.05 to 5.56)
AEs related to body pain	3	1.75 (0.38 to 8.18)
AEs related to GI tract	3	4.87 (0.58 to 41.22)
AEs leading to retreatment discontinuation	3	1.26 (0.31 to 5.05)
AE: adverse event; CI: confidence interval; GI: gastrointestinal; PDEI: phosphodiesterase inhibitor; RCT: randomised controlled trial.

Table 4. Other treatment‐related harms ‐ PDEI versus alpha‐blockers

Ir a la tabla de la revisión

Table 5. Other treatment‐related harms: PDEI + alpha‐blockers versus alpha‐blockers alone

Name of the subgroup	Number of RCTs	Summary estimate RR (95% CI)
AEs related to vasodilatation	3	4.80 (1.93 to 11.91)
AEs related to GI tract	4	6.00 (0.75 to 48.30)
AEs leading to retreatment discontinuation	3	3.00 (0.13 to 70.53)
AE: adverse event; CI: confidence interval; GI: gastrointestinal; PDEI: phosphodiesterase inhibitor; RCT: randomised controlled trial.

Table 5. Other treatment‐related harms: PDEI + alpha‐blockers versus alpha‐blockers alone

Ir a la tabla de la revisión

Table 6. Other treatment‐related harms: PDEI + 5‐ARI versus 5‐ARI alone

Name of the subgroup	Number of RCTs	Summary estimate RR (95% CI)
AEs related to vasodilatation	1	1.01 (0.46 to 2.22)
AEs related to body pain	1	2.70 (1.07 to 6.81)
AEs related to GI tract	1	4.05 (0.87 to 18.92)
AEs leading to retreatment discontinuation	1	0.63 (0.21 to 1.91)
AEs related to sexual function	1	0.51 (0.17 to 1.46)
5‐ARI: 5‐alpha reductase inhibitor; AE: adverse event; CI: confidence interval; GI: gastrointestinal; PDEI: phosphodiesterase inhibitor; RCT: randomised controlled trial.

Table 6. Other treatment‐related harms: PDEI + 5‐ARI versus 5‐ARI alone

Ir a la tabla de la revisión

Comparison 1. PDEI versus placebo

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 IPSS‐total Show forest plot	11	4293	Mean Difference (IV, Fixed, 95% CI)	‐1.89 [‐2.27, ‐1.50]

2 IPSS‐total: subgroup analysis (with or without ED) Show forest plot	3	1816	Mean Difference (IV, Random, 95% CI)	‐2.42 [‐3.04, ‐1.80]

2.1 IPSS‐total in participants with LUTS and ED	3	1486	Mean Difference (IV, Random, 95% CI)	‐2.55 [‐3.33, ‐1.77]
2.2 IPSS‐total in participants with LUTS but without ED	1	330	Mean Difference (IV, Random, 95% CI)	‐2.10 [‐3.39, ‐0.81]
3 IPSS‐total: subgroup analysis (different PDEI) Show forest plot	11	4293	Mean Difference (IV, Fixed, 95% CI)	‐1.89 [‐2.27, ‐1.50]

3.1 Tadalafil	9	3738	Mean Difference (IV, Fixed, 95% CI)	‐1.80 [‐2.20, ‐1.39]
3.2 Sildenafil	1	341	Mean Difference (IV, Fixed, 95% CI)	‐4.4 [‐6.96, ‐1.84]
3.3 Vardenafil	1	214	Mean Difference (IV, Fixed, 95% CI)	‐2.20 [‐3.57, ‐0.83]
4 IPSS‐total: subgroup analysis (different doses of PDEI) Show forest plot	9	4863	Mean Difference (IV, Random, 95% CI)	‐1.96 [‐2.41, ‐1.52]

4.1 Tadalafil 2.5 mg/d	4	1396	Mean Difference (IV, Random, 95% CI)	‐1.31 [‐2.19, ‐0.42]
4.2 Tadalafil 5 mg/d	8	2452	Mean Difference (IV, Random, 95% CI)	‐1.88 [‐2.33, ‐1.42]
4.3 Tadalafil 10 mg/d	1	426	Mean Difference (IV, Random, 95% CI)	‐3.43 [‐4.79, ‐2.07]
4.4 Tadalafil 20 mg/d	2	589	Mean Difference (IV, Random, 95% CI)	‐3.29 [‐4.44, ‐2.14]
5 IPSS‐total: sensitivity analysis (RCT without run‐in period excluded) Show forest plot	8	3567	Mean Difference (IV, Fixed, 95% CI)	‐1.71 [‐2.12, ‐1.29]

6 BPHII Show forest plot	7	3646	Mean Difference (IV, Fixed, 95% CI)	‐0.52 [‐0.71, ‐0.33]

7 BPHII: subgroup analysis (with or without ED) Show forest plot	3	1809	Mean Difference (IV, Fixed, 95% CI)	‐0.71 [‐0.99, ‐0.43]

7.1 IPSS‐BPHII score ‐ in participants with LUTS and ED	3	1480	Mean Difference (IV, Fixed, 95% CI)	‐0.81 [‐1.12, ‐0.49]
7.2 IPSS‐BPHII score ‐ in participants with LUTS but no ED	1	329	Mean Difference (IV, Fixed, 95% CI)	‐0.40 [‐0.98, 0.18]
8 BPHII: subgroup analysis (different PDEI) Show forest plot	7	3878	Mean Difference (IV, Fixed, 95% CI)	‐0.52 [‐0.71, ‐0.33]

8.1 Tadalafil	6	3527	Mean Difference (IV, Fixed, 95% CI)	‐0.50 [‐0.70, ‐0.31]
8.2 Sildenafil	1	351	Mean Difference (IV, Fixed, 95% CI)	‐1.1 [‐2.11, ‐0.09]
8.3 Vardenafil	0	0	Mean Difference (IV, Fixed, 95% CI)	0.0 [0.0, 0.0]
9 BPHII: subgroup analysis (different doses of PDEI) Show forest plot	6	4502	Mean Difference (IV, Fixed, 95% CI)	‐0.45 [‐0.61, ‐0.29]

9.1 Tadalafil 2.5 mg/d	3	1097	Mean Difference (IV, Fixed, 95% CI)	‐0.28 [‐0.61, 0.04]
9.2 Tadalafil 5 mg/d	6	2561	Mean Difference (IV, Fixed, 95% CI)	‐0.48 [‐0.69, ‐0.28]
9.3 Tadalafil 10 mg/d	1	426	Mean Difference (IV, Fixed, 95% CI)	‐0.58 [‐1.15, ‐0.01]
9.4 Tadalafil 20 mg/d	1	418	Mean Difference (IV, Fixed, 95% CI)	‐0.62 [‐1.20, ‐0.04]
10 Any AE with short‐term use Show forest plot	11	4386	Risk Ratio (M‐H, Random, 95% CI)	1.42 [1.21, 1.67]

11 Any AE with short‐term use: subgroup analysis (with or without ED) Show forest plot	2	1458	Risk Ratio (M‐H, Random, 95% CI)	1.46 [1.21, 1.76]

11.1 Participants with ED	2	1120	Risk Ratio (M‐H, Random, 95% CI)	1.36 [1.11, 1.68]
11.2 Participants without ED	1	338	Risk Ratio (M‐H, Random, 95% CI)	1.73 [1.33, 2.24]
12 Any AE with short‐term use: subgroup analysis (different PDEI) Show forest plot	11	4386	Risk Ratio (M‐H, Random, 95% CI)	1.42 [1.21, 1.67]

12.1 Tadalafil	9	3796	Risk Ratio (M‐H, Random, 95% CI)	1.44 [1.19, 1.75]
12.2 Sildenafil	1	369	Risk Ratio (M‐H, Random, 95% CI)	1.22 [0.99, 1.51]
12.3 Vardenafil	1	221	Risk Ratio (M‐H, Random, 95% CI)	1.86 [1.11, 3.11]
13 Any AE with short‐term use: subgroup analysis (different doses of PDEI) Show forest plot	9	4923	Risk Ratio (M‐H, Random, 95% CI)	1.42 [1.25, 1.62]

13.1 Tadalafil 2.5 mg/d	4	1405	Risk Ratio (M‐H, Random, 95% CI)	1.23 [1.03, 1.46]
13.2 Tadalafil 5 mg/d	8	2471	Risk Ratio (M‐H, Random, 95% CI)	1.35 [1.12, 1.63]
13.3 Tadalafil 10 mg/d	1	427	Risk Ratio (M‐H, Random, 95% CI)	1.63 [1.19, 2.24]
13.4 Tadalafil 20 mg/d	2	620	Risk Ratio (M‐H, Random, 95% CI)	1.93 [1.53, 2.44]
14 Any AE with short‐term use: sensitivity analysis (RCT without run‐in period excluded) Show forest plot	8	3596	Risk Ratio (M‐H, Random, 95% CI)	1.36 [1.13, 1.65]

15 IPSS‐QOL Show forest plot	9	3920	Mean Difference (IV, Fixed, 95% CI)	‐0.26 [‐0.35, ‐0.16]

16 IPSS‐QOL: subgroup analysis (with or without ED) Show forest plot	3	1815	Mean Difference (IV, Fixed, 95% CI)	‐0.41 [‐0.60, ‐0.21]

16.1 IPSS‐QOL in participants with LUTS and ED	3	1486	Mean Difference (IV, Fixed, 95% CI)	‐0.41 [‐0.62, ‐0.20]
16.2 IPSS‐QOL in participants with LUTS but no ED	1	329	Mean Difference (IV, Fixed, 95% CI)	‐0.40 [‐0.98, 0.18]
17 IPSS‐QOL: subgroup analysis (different PDEI) Show forest plot	9	3920	Mean Difference (IV, Fixed, 95% CI)	‐0.26 [‐0.35, ‐0.16]

17.1 Tadalafil	8	3569	Mean Difference (IV, Fixed, 95% CI)	‐0.24 [‐0.34, ‐0.15]
17.2 Sildenafil	1	351	Mean Difference (IV, Fixed, 95% CI)	‐0.68 [‐1.18, ‐0.18]
17.3 Vardenafil	0	0	Mean Difference (IV, Fixed, 95% CI)	0.0 [0.0, 0.0]
18 IPSS‐QOL: subgroup analysis (different doses of PDEI) Show forest plot	8	4686	Mean Difference (IV, Fixed, 95% CI)	‐0.27 [‐0.35, ‐0.19]

18.1 Tadalafil 2.5 mg/d	4	1390	Mean Difference (IV, Fixed, 95% CI)	‐0.18 [‐0.34, ‐0.02]
18.2 Tadalafil 5 mg/d	8	2452	Mean Difference (IV, Fixed, 95% CI)	‐0.28 [‐0.38, ‐0.18]
18.3 Tadalafil 10 mg/d	1	426	Mean Difference (IV, Fixed, 95% CI)	‐0.42 [‐0.71, ‐0.13]
18.4 Tadalafil 20 mg/d	1	418	Mean Difference (IV, Fixed, 95% CI)	‐0.39 [‐0.69, ‐0.09]
19 IPSS‐QOL: sensitivity analysis (RCT without run‐in period excluded) Show forest plot	8	3569	Mean Difference (IV, Random, 95% CI)	‐0.24 [‐0.34, ‐0.15]

20 Other treatment‐related harms: AE with short‐term use related to vasodilatation Show forest plot	11	4386	Risk Ratio (M‐H, Fixed, 95% CI)	1.44 [1.18, 1.76]

21 Other treatment‐related harms: AE with short‐term use related to body pain Show forest plot	10	4017	Risk Ratio (M‐H, Fixed, 95% CI)	3.44 [1.98, 5.98]

22 Other treatment‐related harms: AE with short‐term use related to upper GI tract Show forest plot	8	2995	Risk Ratio (M‐H, Fixed, 95% CI)	14.19 [5.80, 34.72]

23 Other treatment‐related harms: AE with short‐term use leading to treatment discontinuation Show forest plot	10	4165	Risk Ratio (M‐H, Fixed, 95% CI)	2.20 [1.38, 3.51]

Comparison 1. PDEI versus placebo

Ir a la tabla de la revisión

Comparison 2. PDEI versus AB

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 IPSS‐total Show forest plot	4	933	Mean Difference (IV, Fixed, 95% CI)	0.22 [‐0.49, 0.93]

2 IPSS‐total: subgroup analysis (different PDEI) Show forest plot	4	782	Mean Difference (IV, Fixed, 95% CI)	0.19 [‐0.58, 0.95]

2.1 Tadalafil	3	742	Mean Difference (IV, Fixed, 95% CI)	0.13 [‐0.65, 0.91]
2.2 Sildenafil	1	40	Mean Difference (IV, Fixed, 95% CI)	1.40 [‐2.24, 5.04]
3 IPSS‐total: subgroup analysis (different doses of PDEI) Show forest plot	4	1085	Mean Difference (IV, Random, 95% CI)	0.35 [‐0.28, 0.98]

3.1 Tadalafil 5 mg/d vs tamsulosin 0.4 mg/d	1	336	Mean Difference (IV, Random, 95% CI)	‐0.60 [‐1.98, 0.78]
3.2 Tadalafil 5 mg/d vs tamsulosin 0.2 mg/d	2	406	Mean Difference (IV, Random, 95% CI)	0.42 [‐0.68, 1.51]
3.3 Tadalafil 2.5 mg/d vs tamsulosin 0.2 mg/d	1	303	Mean Difference (IV, Random, 95% CI)	0.70 [‐0.41, 1.81]
3.4 Sildenafil 25 mg administered 4 times a week vs tamsulosin 0.4 mg/d	1	40	Mean Difference (IV, Random, 95% CI)	1.40 [‐2.24, 5.04]
4 IPSS‐total: sensitivity analysis (RCT without run‐in period excluded) Show forest plot	3	742	Mean Difference (IV, Random, 95% CI)	0.07 [‐0.88, 1.02]

5 BPHII Show forest plot	2	550	Mean Difference (IV, Random, 95% CI)	0.03 [‐1.10, 1.16]

6 BPHII: subgroup analysis (different PDEI) Show forest plot	2	403	Mean Difference (IV, Fixed, 95% CI)	0.23 [‐0.23, 0.69]

6.1 Tadalafil	2	403	Mean Difference (IV, Fixed, 95% CI)	0.23 [‐0.23, 0.69]
7 BPHII: subgroup analysis (different doses of PDEI) Show forest plot	2	700	Mean Difference (IV, Random, 95% CI)	0.24 [‐0.40, 0.88]

7.1 Tadalafil 5 mg/d vs tamsulosin 0.2 mg/d	2	403	Mean Difference (IV, Random, 95% CI)	0.04 [‐1.13, 1.21]
7.2 Tadalafil 2.5 mg/d vs tamsulosin 0.2 mg/d	1	297	Mean Difference (IV, Random, 95% CI)	0.5 [‐0.06, 1.06]
8 BPHII: sensitivity analysis (RCT without run‐in period excluded) Show forest plot	2	550	Mean Difference (IV, Random, 95% CI)	0.03 [‐1.10, 1.16]

9 Any AE with short‐term use Show forest plot	4	936	Risk Ratio (M‐H, Random, 95% CI)	1.35 [0.80, 2.30]

10 IPSS‐QOL Show forest plot	4	933	Mean Difference (IV, Random, 95% CI)	‐0.10 [‐0.48, 0.29]

11 IPSS‐QOL: subgroup analysis (different PDEIs) Show forest plot	4	782	Mean Difference (IV, Fixed, 95% CI)	‐0.01 [‐0.19, 0.17]

11.1 Tadalafil	3	742	Mean Difference (IV, Fixed, 95% CI)	0.02 [‐0.16, 0.21]
11.2 Sildenafil	1	40	Mean Difference (IV, Fixed, 95% CI)	‐0.8 [‐1.78, 0.18]
12 IPSS‐QOL: subgroup analysis (different doses of PDEI) Show forest plot	4	1085	Mean Difference (IV, Random, 95% CI)	0.01 [‐0.29, 0.32]

12.1 Tadalafil 5 mg/d vs tamsulosin 0.4 mg/d	1	336	Mean Difference (IV, Random, 95% CI)	‐0.20 [‐0.48, 0.08]
12.2 Tadalafil 5 mg/d vs tamsulosin 0.2 mg/d	2	406	Mean Difference (IV, Random, 95% CI)	0.11 [‐0.37, 0.59]
12.3 Tadalafil 2.5 mg/d vs tamsulosin 0.2 mg/d	1	303	Mean Difference (IV, Random, 95% CI)	0.30 [0.02, 0.58]
12.4 Sildenafil 25 mg/d vs tamsulosin 0.4 mg/d	1	40	Mean Difference (IV, Random, 95% CI)	‐0.8 [‐1.78, 0.18]
13 IPSS‐QOL: sensitivity analysis (RCT without run‐in period excluded) Show forest plot	3	742	Mean Difference (IV, Random, 95% CI)	‐0.01 [‐0.38, 0.36]

14 Other treatment‐related harms: AE with short‐term use related to vasodilatation Show forest plot	3	746	Risk Ratio (M‐H, Fixed, 95% CI)	2.42 [1.05, 5.56]

15 Other treatment‐related harms: AE with short‐term use related to body pain Show forest plot	3	747	Risk Ratio (M‐H, Fixed, 95% CI)	1.75 [0.38, 8.18]

16 Other treatment‐related harms: AE with short‐term use related to upper GI tract Show forest plot	3	746	Risk Ratio (M‐H, Fixed, 95% CI)	4.87 [0.58, 41.22]

17 Other treatment‐related harms: AE with short‐term use leading to treatment discontinuation Show forest plot	4	786	Risk Ratio (M‐H, Fixed, 95% CI)	1.26 [0.31, 5.05]

Comparison 2. PDEI versus AB

Ir a la tabla de la revisión

Comparison 3. PDEI + AB versus AB alone

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 IPSS‐total Show forest plot	4	193	Mean Difference (IV, Fixed, 95% CI)	‐2.56 [‐3.92, ‐1.19]

2 IPSS‐total: subgroup analysis (different PDEI) Show forest plot	4	193	Mean Difference (IV, Fixed, 95% CI)	‐2.56 [‐3.92, ‐1.19]

2.1 Tadalafil	2	94	Mean Difference (IV, Fixed, 95% CI)	‐3.02 [‐4.83, ‐1.22]
2.2 Sildenafil	1	40	Mean Difference (IV, Fixed, 95% CI)	‐1.0 [‐5.56, 3.56]
2.3 Vardenafil	1	59	Mean Difference (IV, Fixed, 95% CI)	‐2.18 [‐4.53, 0.17]
3 IPSS‐total: subgroup analysis (different doses of PDEI) Show forest plot	4	193	Mean Difference (IV, Random, 95% CI)	‐2.56 [‐3.92, ‐1.19]

3.1 Tadalafil 5 mg + tamsulosin 0.4 mg/d vs tamsulosin 0.4 mg/d	1	40	Mean Difference (IV, Random, 95% CI)	‐3.70 [‐6.44, ‐0.96]
3.2 Tadalafil 20 mg + tamsulosin 0.4 mg/d vs tamsulosin 0.4 mg/d	1	54	Mean Difference (IV, Random, 95% CI)	‐2.50 [‐4.91, ‐0.09]
3.3 Vardenafil 10 mg + tamsulosin 0.4 mg/d vs tamsulosin 0.4 mg/d	1	59	Mean Difference (IV, Random, 95% CI)	‐2.18 [‐4.53, 0.17]
3.4 Sildenafil 25 mg + tamsulosin 0.4 mg 4 times weekly vs tamsulosin 0.4 mg/d	1	40	Mean Difference (IV, Random, 95% CI)	‐1.0 [‐5.56, 3.56]
4 IPSS‐total: sensitivity analysis (RCTs without placebo run‐in period excluded) Show forest plot	2	113	Mean Difference (IV, Random, 95% CI)	‐2.34 [‐4.02, ‐0.66]

5 Any AE in short‐term use Show forest plot	4	194	Risk Ratio (M‐H, Fixed, 95% CI)	2.81 [1.53, 5.17]

6 Any AE in short‐term use: subgroup analysis (different PDEI) Show forest plot	4	194	Risk Ratio (M‐H, Fixed, 95% CI)	2.81 [1.53, 5.17]

6.1 Tadalafil	2	94	Risk Ratio (M‐H, Fixed, 95% CI)	3.0 [1.30, 6.90]
6.2 Sildenafil	1	40	Risk Ratio (M‐H, Fixed, 95% CI)	0.0 [0.0, 0.0]
6.3 Vardenafil	1	60	Risk Ratio (M‐H, Fixed, 95% CI)	2.6 [1.06, 6.39]
7 Any AE with short‐term use: subgroup analysis (different doses of PDEI) Show forest plot	4	194	Risk Ratio (M‐H, Fixed, 95% CI)	2.81 [1.53, 5.17]

7.1 Tadalafil 5 mg + tamsulosin 0.4 mg/d vs tamsulosin 0.4 mg/d	1	40	Risk Ratio (M‐H, Fixed, 95% CI)	3.0 [0.13, 69.52]
7.2 Tadalafil 20 mg + tamsulosin 0.4 mg/d vs tamsulosin 0.4 mg/d	1	54	Risk Ratio (M‐H, Fixed, 95% CI)	3.0 [1.27, 7.09]
7.3 Vardenafil 5 mg + tamsulosin 0.4 mg/d vs tamsulosin 0.4 mg/d	1	60	Risk Ratio (M‐H, Fixed, 95% CI)	2.6 [1.06, 6.39]
7.4 Sildenafil 25 mg + tamsulosin 0.4 mg 4 times weekly vs tamsulosin 0.4 mg daily	1	40	Risk Ratio (M‐H, Fixed, 95% CI)	0.0 [0.0, 0.0]
8 Any AE with short‐term use: sensitivity analysis (RCT without run‐in period excluded) Show forest plot	2	114	Risk Ratio (M‐H, Fixed, 95% CI)	2.8 [1.50, 5.21]

9 IPSS‐QOL Show forest plot	2	94	Mean Difference (IV, Fixed, 95% CI)	‐1.08 [‐1.92, ‐0.23]

10 IPSS‐QOL: subgroup analysis (different PDEIs) Show forest plot	2	94	Mean Difference (IV, Fixed, 95% CI)	‐1.08 [‐1.92, ‐0.23]

10.1 Tadalafil	1	54	Mean Difference (IV, Fixed, 95% CI)	‐0.7 [‐2.40, 1.00]
10.2 Sildenafil	1	40	Mean Difference (IV, Fixed, 95% CI)	‐1.2 [‐2.18, ‐0.22]
10.3 Vardenafil	0	0	Mean Difference (IV, Fixed, 95% CI)	0.0 [0.0, 0.0]
11 IPSS‐QOL: subgroup analysis (different doses of PDEI) Show forest plot	2	94	Mean Difference (IV, Random, 95% CI)	‐1.08 [‐1.92, ‐0.23]

11.1 Sildenafil 25 mg + tamsulosin 0.4 mg 4 times weekly vs tamsulosin 0.4 mg/d	1	40	Mean Difference (IV, Random, 95% CI)	‐1.2 [‐2.18, ‐0.22]
11.2 Tadalafil 20 mg + tamsulosin 0.4 mg/d vs tamsulosin 0.4 mg/d	1	54	Mean Difference (IV, Random, 95% CI)	‐0.7 [‐2.40, 1.00]
12 IPSS‐QOL: sensitivity analysis (RCT without placebo run‐in period excluded) Show forest plot	1	54	Mean Difference (IV, Random, 95% CI)	‐0.7 [‐2.40, 1.00]

13 Other treatment‐related harms: AE with short‐term use related to vasodilatation Show forest plot	3	154	Risk Ratio (M‐H, Fixed, 95% CI)	4.8 [1.93, 11.91]

14 Other treatment‐related harms: AE with short‐term use related to upper GI tract Show forest plot	4	194	Risk Ratio (M‐H, Fixed, 95% CI)	6.0 [0.75, 48.30]

15 Other treatment‐related harms: AE with short‐term use leading to treatment discontinuation Show forest plot	3	154	Risk Ratio (M‐H, Fixed, 95% CI)	3.0 [0.13, 70.53]

Comparison 3. PDEI + AB versus AB alone

Ir a la tabla de la revisión

Comparison 4. PDEI + AB versus PDEI alone

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 IPSS‐total Show forest plot	1	40	Mean Difference (IV, Random, 95% CI)	‐2.40 [‐6.47, 1.67]

2 IPSS‐QOL Show forest plot	1	40	Mean Difference (IV, Fixed, 95% CI)	‐0.40 [‐1.80, 1.00]

Comparison 4. PDEI + AB versus PDEI alone

Ir a la tabla de la revisión

Comparison 5. PDEI + 5‐ARI versus 5‐ARI alone

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 IPSS‐total Show forest plot	1		Mean Difference (IV, Random, 95% CI)	Subtotals only

1.1 at 4 weeks	1	695	Mean Difference (IV, Random, 95% CI)	‐2.0 [‐2.84, ‐1.16]
1.2 at 12 weeks	1	695	Mean Difference (IV, Random, 95% CI)	‐1.40 [‐2.24, ‐0.56]
1.3 at 26 weeks	1	695	Mean Difference (IV, Random, 95% CI)	‐1.0 [‐1.83, ‐0.17]
2 IPSS‐QOL Show forest plot	1		Mean Difference (IV, Fixed, 95% CI)	Subtotals only

2.1 at 4 weeks	1	695	Mean Difference (IV, Fixed, 95% CI)	‐0.3 [‐0.58, ‐0.02]
2.2 at 12 weeks	1	695	Mean Difference (IV, Fixed, 95% CI)	‐0.20 [‐0.48, 0.08]
2.3 at 26 weeks	1	695	Mean Difference (IV, Fixed, 95% CI)	‐0.20 [‐0.48, 0.08]
3 Any AE with long‐term use (26 weeks) Show forest plot	1	695	Risk Ratio (M‐H, Fixed, 95% CI)	1.07 [0.84, 1.36]

4 Other treatment‐related harms: AE with long‐term use (26 weeks) related to vasodilatation Show forest plot	1	696	Risk Ratio (M‐H, Fixed, 95% CI)	1.01 [0.46, 2.22]

5 Other treatment‐related harms: AE with long‐term use (26 weeks) related to body pain Show forest plot	1	696	Risk Ratio (M‐H, Fixed, 95% CI)	2.70 [1.07, 6.81]

6 Other treatment‐related harms: AE with long‐term use (26 weeks) related to upper GI tract Show forest plot	1	696	Risk Ratio (M‐H, Fixed, 95% CI)	4.05 [0.87, 18.92]

7 Other treatment‐related harms: AE with long‐term use (26 weeks) leading to treatment discontinuation Show forest plot	1	696	Risk Ratio (M‐H, Fixed, 95% CI)	0.63 [0.21, 1.91]

8 Other treatment‐related harms: AE with long‐term use (26 weeks) related to sexual function Show forest plot	1	696	Risk Ratio (M‐H, Fixed, 95% CI)	0.51 [0.17, 1.46]

Comparison 5. PDEI + 5‐ARI versus 5‐ARI alone

Ir a la tabla de la revisión

	Idioma de la Revisión Cochrane Escoja su idioma de preferencia para las revisiones Cochrane y otros contenidos. Las secciones sin una traducción aparecerán en inglés.

	Idioma de la web Escoja su idioma de preferencia para la web de la Biblioteca Cochrane.

Idioma de la Revisión Cochrane

Idioma de la web

Resumen

Antecedentes

Objetivos

Métodos de búsqueda

Criterios de selección

Obtención y análisis de los datos

Resultados principales

Conclusiones de los autores

PICO

PICO

Population

Intervention

Comparison

Outcome

Resumen en términos sencillos

Inhibidores de la fosfodiesterasa para el tratamiento de los problemas urinarios en pacientes con hiperplasia prostática benigna

Resumen visual

Conclusiones de los autores

Implicaciones para la práctica

Implicaciones para la investigación

Summary of findings

Antecedentes

Descripción de la afección

Descripción de la intervención

Alfabloqueantes (AB)

Inhibidores de la 5‐alfa reductasa (5‐ARI)

Inhibidores de la fosfodiesterasa (IFDE)

De qué manera podría funcionar la intervención

Por qué es importante realizar esta revisión

Objetivos

Métodos

Criterios de inclusión de estudios para esta revisión

Tipos de estudios

Tipos de participantes

Tipos de intervenciones

Tipos de medida de resultado

Resultados primarios

Resultados secundarios

Métodos de búsqueda para la identificación de los estudios

Búsquedas electrónicas

Búsqueda de otros recursos

Obtención y análisis de los datos

Selección de los estudios

Extracción y manejo de los datos

Evaluación del riesgo de sesgo de los estudios incluidos

Medidas del efecto del tratamiento

Cuestiones relativas a la unidad de análisis

Manejo de los datos faltantes

Evaluación de la heterogeneidad

Evaluación de los sesgos de notificación

Síntesis de los datos

Análisis de subgrupos e investigación de la heterogeneidad

Análisis de sensibilidad

'Summary of findings' table

Results

Description of studies

PDEI versus placebo

PDEI versus AB

PDEI versus 5‐ARI

PDEI + AB versus AB alone

PDEI + AB versus PDEI alone

PDEI + 5‐ARI versus 5‐ARI alone

PDEI + 5‐ARI versus PDEI alone

PDEI + AB + 5‐ARI versus AB + 5‐ARI

Comparison of different PDEI agents for treatment of BPH‐LUTS

Results of the search

Included studies

Excluded studies

Risk of bias in included studies

PDEI versus placebo

Allocation (selection bias)

Random sequence generation

Allocation concealment

Blinding

Blinding of participants and personnel (performance bias)

Blinding of outcome assessment (detection bias)

Incomplete outcome data (attrition bias)

Selective reporting (reporting bias)