Efficacy and Safety of Hexanic Lipidosterolic Extract of Serenoa repens (Permixon) in the Treatment of Lower Urinary Tract Symptoms Due to Benign Prostatic Hyperplasia: Systematic Review and Meta-analysis of Randomized Controlled Trials

3.1. Quality of evidence

Figure 1 summarizes the literature review process that led to the identification of the 12 RCTs used in the meta-analysis. Specifically, seven RCTs compared Permixon with placebo [15], [16], [17], [18], [19], [20], and [21], two RCTs compared Permixon with tamsulosin [22] and [23], two RCTs compared Permixon plus tamsulosin with placebo plus tamsulosin [24] and with tamsulosin alone [25], and a single RCT compared Permixon with finasteride [26]. Among these publications, there were four RCTs of good methodological quality (level of evidence 2) [22], [23], [24], and [26] and eight RCTs of poor methodological quality (level of evidence 3) [15], [16], [17], [18], [19], [20], [21], and [25]. Clinically speaking, all studies but three [22], [24], and [25] evaluated short-term treatment schedules; standard outcome measures, such as the International Prostate Symptom Score (IPSS) and the American Urological Association symptom index, were used in only five studies [22], [23], [24], [25], and [26].

3.2. Permixon versus placebo

Table 1 summarizes the studies reporting the efficacy and safety of Permixon in comparison to placebo [15], [16], [17], [18], [19], [20], and [21].

The number of nocturnal voids at study end were significantly lower with Permixon (WMD −0.31; 95% CI, −0.59 to −0.03; p = 0.03). Moreover, maximum urinary flow rate (Q_max) at study end was significantly higher in the patients treated with Permixon (WMD 3.37; 95% CI, 1.71–5.03; p < 0.0001). With regard to safety, the overall adverse event rates were similar for Permixon and placebo (OR 1.12; 95% CI, 0.13–9.75; p = 0.92). Finally, withdrawal rates were similar following Permixon or placebo (OR 1.52; 95% CI, 0.32–7.33; p = 0.60).

Figure 2 shows the forest plots for efficacy and safety data of Permixon in comparison to placebo.

3.3. Permixon or combination therapy versus tamsulosin

Table 2 summarizes the studies reporting efficacy and safety of Permixon or combination therapy compared with tamsulosin.

In the studies comparing Permixon with tamsulosin [22] and [23], no statistically significant difference was found in terms of mean change in IPSS from baseline to study end (WMD 1.15; 95% CI, −1.11 to 3.40; p = 0.32) and mean change in Q_max from baseline to study end (WMD −0.16; 95% CI, −0.60 to 0.28; p = 0.48) (Fig. 3a and 3b). With regard to safety, prevalence of adverse events (OR 0.95; 95% CI, 0.72–1.26; p = 0.72) and withdrawal rate (OR 0.96; 95% CI, 0.66–1.40; p = 0.83) were lower with Permixon than with tamsulosin, albeit not significantly (Fig. 3c and 3d). In one study [22], however, ejaculatory dysfunction was significantly less common with Permixon than with tamsulosin (0.5% vs 4%; p = 0.007). Figure 3 shows the forest plots for efficacy and safety data of Permixon compared with tamsulosin.

Fig. 3

In the studies comparing Permixon plus tamsulosin with tamsulosin alone [24] and [25], mean IPSS significantly decreased from baseline to study end with the combination therapy versus tamsulosin alone (WMD 0.31; 95% CI 0.13–0.48; p < 0.01). A non–statistically significant trend in favor of the combination therapy was demonstrated for Q_max improvement (WMD 0.10; 95% CI, −0.02 to 0.21; p = 0.10). With regard to rate of adverse events, a non–statistically significant trend favoring tamsulosin monotherapy was shown (OR 1.63; 95% CI, 0.94–2.84; p = 0.08). Figure 4 shows the forest plots concerning efficacy and safety data of Permixon plus tamsulosin compared with tamsulosin.

3.4. Permixon versus finasteride

A single study compared efficacy and safety of Permixon and finasteride [26]. Specifically, Carraro et al. randomized 1098 patients with moderate to severe LUTS/BPH to 26-wk treatment with Permixon (n = 553) or finasteride (n = 484). Mean prostate volume was 43 and 44 ml in the Permixon and finasteride arms, respectively. Both drugs showed similar efficacy with regard to improvement in IPSS (−5.8 with Permixon vs −6.2 with finasteride; p = 0.17) and improvement in the IPSS item for quality of life (−1.5 vs −1.4, respectively; p = 0.14), although increase in Q_max was slightly higher with finasteride (+2.7 vs +3.2, respectively; p = 0.035). However, decreased libido and impotence were less common in the Permixon arm (2.2% and 1.5%, respectively) than in the finasteride arm (3% and 2.8%, respectively). Sexual function score was significantly lower (corresponding to better function) with Permixon than with finasteride (7.9 vs 9.3; p < 0.01).

3.5. Publication bias

Funnel plots of all studies used in this meta-analysis were generated for all evaluated comparisons. Only a single study lies outside the 95% CI with an even distribution about the vertical, suggesting little evidence of publication bias.

3.6. Discussion

α-Blockers, 5-ARIs, or their combination are among the standard drug therapies for bothersome LUTS/BPH [1] and [2]. Both categories of drugs are effective for improving symptoms, with 5-ARIs also able to reduce the risk of disease progression [1] and [2]. However, such therapies are associated with significant side effects, including postural hypotension and ejaculatory dysfunction for α-blockers and erectile dysfunction and loss of libido for 5-ARIs, that may reduce patient adherence to therapies [27] and [28]. Impairment of sexual function is of particular concern, considering that concomitant sexual dysfunction and LUTS are highly prevalent in adult men [29].

Phytotherapy is a common therapeutic option for LUTS/BPH and is used in approximately 17% of patients [3]. Specifically, Tacklind et al. reported on behalf of the Cochrane Collaboration a large meta-analysis of all RCTs evaluating Serenoa repens, which is the most common phytoterapeutic drug, and demonstrated no advantage for Serenoa repens in comparisons with placebo in relieving LUTS [7]. However, the latest version of the Cochrane meta-analysis suffers from some limitations. All of the different brands of Serenoa repens were pooled together, which is a questionable choice, as Bilia et al. clearly stated in the feedback on the Cochrane meta-analysis [7]. Specifically, some differences exist among the different brands of Serenoa repens, as reported by some preclinical studies, with Permixon seeming to be the brand with the highest percentage of free fatty acids and the highest in vitro efficacy [8], [9], and [10]. Consequently, the conclusion of the Cochrane meta-analysis might be affected by the fact that various kinds of extracts were pooled in the analysis. Indeed, this conclusion changed when the Cochrane Collaboration included in its meta-analysis well-designed RCTs but with different extracts of Serenoa repens obtained through different extraction processes [30] and [31]. Specifically, Barry et al. used Prosta-Urgenin Uno, an ethanolic Serenoa repens extract produced by Rottapharm Madaus (Selangor, Malaysia) [30], whereas Bent et al. evaluated a carbon dioxide Serenoa repens extract (Indena, Milan, Italy) [31]. That issue may explain why the Cochrane Collaboration concluded in a previous meta-analysis that Serenoa repens improves urologic symptoms and flow measures compared with placebo [32]. In fact, the same authors of the Cochrane meta-analysis stated that they did not know if their conclusions were generalizable to proprietary products of Serenoa repens such as Permixon, inviting investigation of the issue [7] and [33], which was the objective of the current meta-analysis. Moreover, some inaccuracies are evident in their data analysis and extraction. For example, Q_max values for two studies [21] and [34] reported in the Cochrane meta-analysis were not consistent with those reported in the original papers.

Consequently, to assess the validity of the Cochrane meta-analysis data for all brands of Serenoa repens with regard to Permixon, we performed a systematic review and meta-analysis of the RCTs assessing efficacy and safety of Permixon. We found that Permixon was more effective than placebo in reducing the number of nocturnal voids and Q_max and as effective as tamsulosin and short-term finasteride in improving LUTS. Moreover, combination therapy of Permixon and tamsulosin might provide some benefits in comparison to tamsulosin monotherapy in terms of LUTS improvement. In all comparisons with the other drugs, Permixon was associated with a favorable profile of adverse events, especially with regard to ejaculatory and erectile functions.

The findings of the present study were partially similar to those of Boyle et al, who reported a systematic review of RCTs and open-label studies on Permixon in 2004 [35]. Specifically, Boyle et al. demonstrated that Permixon was able to increase Q_max by roughly 2 ml/s, to reduce the mean number of nocturnal voids by roughly one, and to cause a five-point decrease in IPSS in comparison to placebo. Our analysis, which included studies comparing Permixon with tamsulosin or short-term finasteride, reconfirmed the same figures with regard to improvement in number of nocturnal voids and Q_max.

The present data showed stronger evidence for Permixon, the hexanic extract of Serenoa repens, either as monotherapy or in combination with tamsulosin, than the extract suggested by current guidelines for the whole category of phytoterapeutic drugs and for other Serenoa repens brands. This finding is supported by the recent assessment report on Serenoa repens released by the European Medicines Agency [36]. In this report, only the hexanic extract of Serenoa repens has been recognized as a well-established medicinal product thanks to its proven efficacy and tolerability in controlled trials. Conversely, ethanolic extracts have been categorized as traditional use products because of the lack of clinical trials, and supercritical CO₂ extracts have not been assigned any particular status because clinical studies have not provided sufficiently reliable data and because the extracts have been marketed for <15 yr. Moreover, the anti-inflammatory properties of Serenoa repens may represent a further potential advantage of this drug to improve storage and voiding LUTS. Interestingly, this direct anti-inflammatory effect has not been demonstrated for other drugs commonly prescribed for the treatment of LUTS/BPH [6].

We acknowledge that the overall quality of the available studies was limited, with the majority of RCTs being of low methodological quality. Specifically, most studies enrolled a limited number of patients, lacked sample size calculation, and did not use validated tools to measure LUTS (eg, IPSS and frequency–volume charts). Moreover, most studies assessed short-term treatment schedules, especially regarding finasteride. High-quality RCTs comparing Permixon with placebo, α-blockers, and 5-ARIs in the long term and large clinical trials testing the anti-inflammatory properties of Permixon are needed to corroborate our findings. Furthermore, how long Permixon should be continued, who will profit most from this therapy, and whether Permixon may reduce the risk of BPH progression and related complications remain unanswered questions.

3.1. Quality of evidence

Figure 1 summarizes the literature review process that led to the identification of the 12 RCTs used in the meta-analysis. Specifically, seven RCTs compared Permixon with placebo [15], [16], [17], [18], [19], [20], and [21], two RCTs compared Permixon with tamsulosin [22] and [23], two RCTs compared Permixon plus tamsulosin with placebo plus tamsulosin [24] and with tamsulosin alone [25], and a single RCT compared Permixon with finasteride [26]. Among these publications, there were four RCTs of good methodological quality (level of evidence 2) [22], [23], [24], and [26] and eight RCTs of poor methodological quality (level of evidence 3) [15], [16], [17], [18], [19], [20], [21], and [25]. Clinically speaking, all studies but three [22], [24], and [25] evaluated short-term treatment schedules; standard outcome measures, such as the International Prostate Symptom Score (IPSS) and the American Urological Association symptom index, were used in only five studies [22], [23], [24], [25], and [26].

3.2. Permixon versus placebo

Table 1 summarizes the studies reporting the efficacy and safety of Permixon in comparison to placebo [15], [16], [17], [18], [19], [20], and [21].

The number of nocturnal voids at study end were significantly lower with Permixon (WMD −0.31; 95% CI, −0.59 to −0.03; p = 0.03). Moreover, maximum urinary flow rate (Q_max) at study end was significantly higher in the patients treated with Permixon (WMD 3.37; 95% CI, 1.71–5.03; p < 0.0001). With regard to safety, the overall adverse event rates were similar for Permixon and placebo (OR 1.12; 95% CI, 0.13–9.75; p = 0.92). Finally, withdrawal rates were similar following Permixon or placebo (OR 1.52; 95% CI, 0.32–7.33; p = 0.60).

Figure 2 shows the forest plots for efficacy and safety data of Permixon in comparison to placebo.

3.3. Permixon or combination therapy versus tamsulosin

Table 2 summarizes the studies reporting efficacy and safety of Permixon or combination therapy compared with tamsulosin.

In the studies comparing Permixon with tamsulosin [22] and [23], no statistically significant difference was found in terms of mean change in IPSS from baseline to study end (WMD 1.15; 95% CI, −1.11 to 3.40; p = 0.32) and mean change in Q_max from baseline to study end (WMD −0.16; 95% CI, −0.60 to 0.28; p = 0.48) (Fig. 3a and 3b). With regard to safety, prevalence of adverse events (OR 0.95; 95% CI, 0.72–1.26; p = 0.72) and withdrawal rate (OR 0.96; 95% CI, 0.66–1.40; p = 0.83) were lower with Permixon than with tamsulosin, albeit not significantly (Fig. 3c and 3d). In one study [22], however, ejaculatory dysfunction was significantly less common with Permixon than with tamsulosin (0.5% vs 4%; p = 0.007). Figure 3 shows the forest plots for efficacy and safety data of Permixon compared with tamsulosin.

Fig. 3

In the studies comparing Permixon plus tamsulosin with tamsulosin alone [24] and [25], mean IPSS significantly decreased from baseline to study end with the combination therapy versus tamsulosin alone (WMD 0.31; 95% CI 0.13–0.48; p < 0.01). A non–statistically significant trend in favor of the combination therapy was demonstrated for Q_max improvement (WMD 0.10; 95% CI, −0.02 to 0.21; p = 0.10). With regard to rate of adverse events, a non–statistically significant trend favoring tamsulosin monotherapy was shown (OR 1.63; 95% CI, 0.94–2.84; p = 0.08). Figure 4 shows the forest plots concerning efficacy and safety data of Permixon plus tamsulosin compared with tamsulosin.

3.4. Permixon versus finasteride

A single study compared efficacy and safety of Permixon and finasteride [26]. Specifically, Carraro et al. randomized 1098 patients with moderate to severe LUTS/BPH to 26-wk treatment with Permixon (n = 553) or finasteride (n = 484). Mean prostate volume was 43 and 44 ml in the Permixon and finasteride arms, respectively. Both drugs showed similar efficacy with regard to improvement in IPSS (−5.8 with Permixon vs −6.2 with finasteride; p = 0.17) and improvement in the IPSS item for quality of life (−1.5 vs −1.4, respectively; p = 0.14), although increase in Q_max was slightly higher with finasteride (+2.7 vs +3.2, respectively; p = 0.035). However, decreased libido and impotence were less common in the Permixon arm (2.2% and 1.5%, respectively) than in the finasteride arm (3% and 2.8%, respectively). Sexual function score was significantly lower (corresponding to better function) with Permixon than with finasteride (7.9 vs 9.3; p < 0.01).

3.5. Publication bias

Funnel plots of all studies used in this meta-analysis were generated for all evaluated comparisons. Only a single study lies outside the 95% CI with an even distribution about the vertical, suggesting little evidence of publication bias.

3.6. Discussion

α-Blockers, 5-ARIs, or their combination are among the standard drug therapies for bothersome LUTS/BPH [1] and [2]. Both categories of drugs are effective for improving symptoms, with 5-ARIs also able to reduce the risk of disease progression [1] and [2]. However, such therapies are associated with significant side effects, including postural hypotension and ejaculatory dysfunction for α-blockers and erectile dysfunction and loss of libido for 5-ARIs, that may reduce patient adherence to therapies [27] and [28]. Impairment of sexual function is of particular concern, considering that concomitant sexual dysfunction and LUTS are highly prevalent in adult men [29].

Phytotherapy is a common therapeutic option for LUTS/BPH and is used in approximately 17% of patients [3]. Specifically, Tacklind et al. reported on behalf of the Cochrane Collaboration a large meta-analysis of all RCTs evaluating Serenoa repens, which is the most common phytoterapeutic drug, and demonstrated no advantage for Serenoa repens in comparisons with placebo in relieving LUTS [7]. However, the latest version of the Cochrane meta-analysis suffers from some limitations. All of the different brands of Serenoa repens were pooled together, which is a questionable choice, as Bilia et al. clearly stated in the feedback on the Cochrane meta-analysis [7]. Specifically, some differences exist among the different brands of Serenoa repens, as reported by some preclinical studies, with Permixon seeming to be the brand with the highest percentage of free fatty acids and the highest in vitro efficacy [8], [9], and [10]. Consequently, the conclusion of the Cochrane meta-analysis might be affected by the fact that various kinds of extracts were pooled in the analysis. Indeed, this conclusion changed when the Cochrane Collaboration included in its meta-analysis well-designed RCTs but with different extracts of Serenoa repens obtained through different extraction processes [30] and [31]. Specifically, Barry et al. used Prosta-Urgenin Uno, an ethanolic Serenoa repens extract produced by Rottapharm Madaus (Selangor, Malaysia) [30], whereas Bent et al. evaluated a carbon dioxide Serenoa repens extract (Indena, Milan, Italy) [31]. That issue may explain why the Cochrane Collaboration concluded in a previous meta-analysis that Serenoa repens improves urologic symptoms and flow measures compared with placebo [32]. In fact, the same authors of the Cochrane meta-analysis stated that they did not know if their conclusions were generalizable to proprietary products of Serenoa repens such as Permixon, inviting investigation of the issue [7] and [33], which was the objective of the current meta-analysis. Moreover, some inaccuracies are evident in their data analysis and extraction. For example, Q_max values for two studies [21] and [34] reported in the Cochrane meta-analysis were not consistent with those reported in the original papers.

Consequently, to assess the validity of the Cochrane meta-analysis data for all brands of Serenoa repens with regard to Permixon, we performed a systematic review and meta-analysis of the RCTs assessing efficacy and safety of Permixon. We found that Permixon was more effective than placebo in reducing the number of nocturnal voids and Q_max and as effective as tamsulosin and short-term finasteride in improving LUTS. Moreover, combination therapy of Permixon and tamsulosin might provide some benefits in comparison to tamsulosin monotherapy in terms of LUTS improvement. In all comparisons with the other drugs, Permixon was associated with a favorable profile of adverse events, especially with regard to ejaculatory and erectile functions.

The findings of the present study were partially similar to those of Boyle et al, who reported a systematic review of RCTs and open-label studies on Permixon in 2004 [35]. Specifically, Boyle et al. demonstrated that Permixon was able to increase Q_max by roughly 2 ml/s, to reduce the mean number of nocturnal voids by roughly one, and to cause a five-point decrease in IPSS in comparison to placebo. Our analysis, which included studies comparing Permixon with tamsulosin or short-term finasteride, reconfirmed the same figures with regard to improvement in number of nocturnal voids and Q_max.

The present data showed stronger evidence for Permixon, the hexanic extract of Serenoa repens, either as monotherapy or in combination with tamsulosin, than the extract suggested by current guidelines for the whole category of phytoterapeutic drugs and for other Serenoa repens brands. This finding is supported by the recent assessment report on Serenoa repens released by the European Medicines Agency [36]. In this report, only the hexanic extract of Serenoa repens has been recognized as a well-established medicinal product thanks to its proven efficacy and tolerability in controlled trials. Conversely, ethanolic extracts have been categorized as traditional use products because of the lack of clinical trials, and supercritical CO₂ extracts have not been assigned any particular status because clinical studies have not provided sufficiently reliable data and because the extracts have been marketed for <15 yr. Moreover, the anti-inflammatory properties of Serenoa repens may represent a further potential advantage of this drug to improve storage and voiding LUTS. Interestingly, this direct anti-inflammatory effect has not been demonstrated for other drugs commonly prescribed for the treatment of LUTS/BPH [6].

We acknowledge that the overall quality of the available studies was limited, with the majority of RCTs being of low methodological quality. Specifically, most studies enrolled a limited number of patients, lacked sample size calculation, and did not use validated tools to measure LUTS (eg, IPSS and frequency–volume charts). Moreover, most studies assessed short-term treatment schedules, especially regarding finasteride. High-quality RCTs comparing Permixon with placebo, α-blockers, and 5-ARIs in the long term and large clinical trials testing the anti-inflammatory properties of Permixon are needed to corroborate our findings. Furthermore, how long Permixon should be continued, who will profit most from this therapy, and whether Permixon may reduce the risk of BPH progression and related complications remain unanswered questions.

3.1. Quality of evidence

Figure 1 summarizes the literature review process that led to the identification of the 12 RCTs used in the meta-analysis. Specifically, seven RCTs compared Permixon with placebo [15], [16], [17], [18], [19], [20], and [21], two RCTs compared Permixon with tamsulosin [22] and [23], two RCTs compared Permixon plus tamsulosin with placebo plus tamsulosin [24] and with tamsulosin alone [25], and a single RCT compared Permixon with finasteride [26]. Among these publications, there were four RCTs of good methodological quality (level of evidence 2) [22], [23], [24], and [26] and eight RCTs of poor methodological quality (level of evidence 3) [15], [16], [17], [18], [19], [20], [21], and [25]. Clinically speaking, all studies but three [22], [24], and [25] evaluated short-term treatment schedules; standard outcome measures, such as the International Prostate Symptom Score (IPSS) and the American Urological Association symptom index, were used in only five studies [22], [23], [24], [25], and [26].

3.2. Permixon versus placebo

Table 1 summarizes the studies reporting the efficacy and safety of Permixon in comparison to placebo [15], [16], [17], [18], [19], [20], and [21].

The number of nocturnal voids at study end were significantly lower with Permixon (WMD −0.31; 95% CI, −0.59 to −0.03; p = 0.03). Moreover, maximum urinary flow rate (Q_max) at study end was significantly higher in the patients treated with Permixon (WMD 3.37; 95% CI, 1.71–5.03; p < 0.0001). With regard to safety, the overall adverse event rates were similar for Permixon and placebo (OR 1.12; 95% CI, 0.13–9.75; p = 0.92). Finally, withdrawal rates were similar following Permixon or placebo (OR 1.52; 95% CI, 0.32–7.33; p = 0.60).

Figure 2 shows the forest plots for efficacy and safety data of Permixon in comparison to placebo.

3.3. Permixon or combination therapy versus tamsulosin

Table 2 summarizes the studies reporting efficacy and safety of Permixon or combination therapy compared with tamsulosin.

In the studies comparing Permixon with tamsulosin [22] and [23], no statistically significant difference was found in terms of mean change in IPSS from baseline to study end (WMD 1.15; 95% CI, −1.11 to 3.40; p = 0.32) and mean change in Q_max from baseline to study end (WMD −0.16; 95% CI, −0.60 to 0.28; p = 0.48) (Fig. 3a and 3b). With regard to safety, prevalence of adverse events (OR 0.95; 95% CI, 0.72–1.26; p = 0.72) and withdrawal rate (OR 0.96; 95% CI, 0.66–1.40; p = 0.83) were lower with Permixon than with tamsulosin, albeit not significantly (Fig. 3c and 3d). In one study [22], however, ejaculatory dysfunction was significantly less common with Permixon than with tamsulosin (0.5% vs 4%; p = 0.007). Figure 3 shows the forest plots for efficacy and safety data of Permixon compared with tamsulosin.

Fig. 3

In the studies comparing Permixon plus tamsulosin with tamsulosin alone [24] and [25], mean IPSS significantly decreased from baseline to study end with the combination therapy versus tamsulosin alone (WMD 0.31; 95% CI 0.13–0.48; p < 0.01). A non–statistically significant trend in favor of the combination therapy was demonstrated for Q_max improvement (WMD 0.10; 95% CI, −0.02 to 0.21; p = 0.10). With regard to rate of adverse events, a non–statistically significant trend favoring tamsulosin monotherapy was shown (OR 1.63; 95% CI, 0.94–2.84; p = 0.08). Figure 4 shows the forest plots concerning efficacy and safety data of Permixon plus tamsulosin compared with tamsulosin.

3.4. Permixon versus finasteride

A single study compared efficacy and safety of Permixon and finasteride [26]. Specifically, Carraro et al. randomized 1098 patients with moderate to severe LUTS/BPH to 26-wk treatment with Permixon (n = 553) or finasteride (n = 484). Mean prostate volume was 43 and 44 ml in the Permixon and finasteride arms, respectively. Both drugs showed similar efficacy with regard to improvement in IPSS (−5.8 with Permixon vs −6.2 with finasteride; p = 0.17) and improvement in the IPSS item for quality of life (−1.5 vs −1.4, respectively; p = 0.14), although increase in Q_max was slightly higher with finasteride (+2.7 vs +3.2, respectively; p = 0.035). However, decreased libido and impotence were less common in the Permixon arm (2.2% and 1.5%, respectively) than in the finasteride arm (3% and 2.8%, respectively). Sexual function score was significantly lower (corresponding to better function) with Permixon than with finasteride (7.9 vs 9.3; p < 0.01).

3.5. Publication bias

Funnel plots of all studies used in this meta-analysis were generated for all evaluated comparisons. Only a single study lies outside the 95% CI with an even distribution about the vertical, suggesting little evidence of publication bias.

3.6. Discussion

α-Blockers, 5-ARIs, or their combination are among the standard drug therapies for bothersome LUTS/BPH [1] and [2]. Both categories of drugs are effective for improving symptoms, with 5-ARIs also able to reduce the risk of disease progression [1] and [2]. However, such therapies are associated with significant side effects, including postural hypotension and ejaculatory dysfunction for α-blockers and erectile dysfunction and loss of libido for 5-ARIs, that may reduce patient adherence to therapies [27] and [28]. Impairment of sexual function is of particular concern, considering that concomitant sexual dysfunction and LUTS are highly prevalent in adult men [29].

Phytotherapy is a common therapeutic option for LUTS/BPH and is used in approximately 17% of patients [3]. Specifically, Tacklind et al. reported on behalf of the Cochrane Collaboration a large meta-analysis of all RCTs evaluating Serenoa repens, which is the most common phytoterapeutic drug, and demonstrated no advantage for Serenoa repens in comparisons with placebo in relieving LUTS [7]. However, the latest version of the Cochrane meta-analysis suffers from some limitations. All of the different brands of Serenoa repens were pooled together, which is a questionable choice, as Bilia et al. clearly stated in the feedback on the Cochrane meta-analysis [7]. Specifically, some differences exist among the different brands of Serenoa repens, as reported by some preclinical studies, with Permixon seeming to be the brand with the highest percentage of free fatty acids and the highest in vitro efficacy [8], [9], and [10]. Consequently, the conclusion of the Cochrane meta-analysis might be affected by the fact that various kinds of extracts were pooled in the analysis. Indeed, this conclusion changed when the Cochrane Collaboration included in its meta-analysis well-designed RCTs but with different extracts of Serenoa repens obtained through different extraction processes [30] and [31]. Specifically, Barry et al. used Prosta-Urgenin Uno, an ethanolic Serenoa repens extract produced by Rottapharm Madaus (Selangor, Malaysia) [30], whereas Bent et al. evaluated a carbon dioxide Serenoa repens extract (Indena, Milan, Italy) [31]. That issue may explain why the Cochrane Collaboration concluded in a previous meta-analysis that Serenoa repens improves urologic symptoms and flow measures compared with placebo [32]. In fact, the same authors of the Cochrane meta-analysis stated that they did not know if their conclusions were generalizable to proprietary products of Serenoa repens such as Permixon, inviting investigation of the issue [7] and [33], which was the objective of the current meta-analysis. Moreover, some inaccuracies are evident in their data analysis and extraction. For example, Q_max values for two studies [21] and [34] reported in the Cochrane meta-analysis were not consistent with those reported in the original papers.

Consequently, to assess the validity of the Cochrane meta-analysis data for all brands of Serenoa repens with regard to Permixon, we performed a systematic review and meta-analysis of the RCTs assessing efficacy and safety of Permixon. We found that Permixon was more effective than placebo in reducing the number of nocturnal voids and Q_max and as effective as tamsulosin and short-term finasteride in improving LUTS. Moreover, combination therapy of Permixon and tamsulosin might provide some benefits in comparison to tamsulosin monotherapy in terms of LUTS improvement. In all comparisons with the other drugs, Permixon was associated with a favorable profile of adverse events, especially with regard to ejaculatory and erectile functions.

The findings of the present study were partially similar to those of Boyle et al, who reported a systematic review of RCTs and open-label studies on Permixon in 2004 [35]. Specifically, Boyle et al. demonstrated that Permixon was able to increase Q_max by roughly 2 ml/s, to reduce the mean number of nocturnal voids by roughly one, and to cause a five-point decrease in IPSS in comparison to placebo. Our analysis, which included studies comparing Permixon with tamsulosin or short-term finasteride, reconfirmed the same figures with regard to improvement in number of nocturnal voids and Q_max.

The present data showed stronger evidence for Permixon, the hexanic extract of Serenoa repens, either as monotherapy or in combination with tamsulosin, than the extract suggested by current guidelines for the whole category of phytoterapeutic drugs and for other Serenoa repens brands. This finding is supported by the recent assessment report on Serenoa repens released by the European Medicines Agency [36]. In this report, only the hexanic extract of Serenoa repens has been recognized as a well-established medicinal product thanks to its proven efficacy and tolerability in controlled trials. Conversely, ethanolic extracts have been categorized as traditional use products because of the lack of clinical trials, and supercritical CO₂ extracts have not been assigned any particular status because clinical studies have not provided sufficiently reliable data and because the extracts have been marketed for <15 yr. Moreover, the anti-inflammatory properties of Serenoa repens may represent a further potential advantage of this drug to improve storage and voiding LUTS. Interestingly, this direct anti-inflammatory effect has not been demonstrated for other drugs commonly prescribed for the treatment of LUTS/BPH [6].

We acknowledge that the overall quality of the available studies was limited, with the majority of RCTs being of low methodological quality. Specifically, most studies enrolled a limited number of patients, lacked sample size calculation, and did not use validated tools to measure LUTS (eg, IPSS and frequency–volume charts). Moreover, most studies assessed short-term treatment schedules, especially regarding finasteride. High-quality RCTs comparing Permixon with placebo, α-blockers, and 5-ARIs in the long term and large clinical trials testing the anti-inflammatory properties of Permixon are needed to corroborate our findings. Furthermore, how long Permixon should be continued, who will profit most from this therapy, and whether Permixon may reduce the risk of BPH progression and related complications remain unanswered questions.

3.1. Quality of evidence

Figure 1 summarizes the literature review process that led to the identification of the 12 RCTs used in the meta-analysis. Specifically, seven RCTs compared Permixon with placebo [15], [16], [17], [18], [19], [20], and [21], two RCTs compared Permixon with tamsulosin [22] and [23], two RCTs compared Permixon plus tamsulosin with placebo plus tamsulosin [24] and with tamsulosin alone [25], and a single RCT compared Permixon with finasteride [26]. Among these publications, there were four RCTs of good methodological quality (level of evidence 2) [22], [23], [24], and [26] and eight RCTs of poor methodological quality (level of evidence 3) [15], [16], [17], [18], [19], [20], [21], and [25]. Clinically speaking, all studies but three [22], [24], and [25] evaluated short-term treatment schedules; standard outcome measures, such as the International Prostate Symptom Score (IPSS) and the American Urological Association symptom index, were used in only five studies [22], [23], [24], [25], and [26].

3.2. Permixon versus placebo

Table 1 summarizes the studies reporting the efficacy and safety of Permixon in comparison to placebo [15], [16], [17], [18], [19], [20], and [21].

The number of nocturnal voids at study end were significantly lower with Permixon (WMD −0.31; 95% CI, −0.59 to −0.03; p = 0.03). Moreover, maximum urinary flow rate (Q_max) at study end was significantly higher in the patients treated with Permixon (WMD 3.37; 95% CI, 1.71–5.03; p < 0.0001). With regard to safety, the overall adverse event rates were similar for Permixon and placebo (OR 1.12; 95% CI, 0.13–9.75; p = 0.92). Finally, withdrawal rates were similar following Permixon or placebo (OR 1.52; 95% CI, 0.32–7.33; p = 0.60).

Figure 2 shows the forest plots for efficacy and safety data of Permixon in comparison to placebo.

3.3. Permixon or combination therapy versus tamsulosin

Table 2 summarizes the studies reporting efficacy and safety of Permixon or combination therapy compared with tamsulosin.

In the studies comparing Permixon with tamsulosin [22] and [23], no statistically significant difference was found in terms of mean change in IPSS from baseline to study end (WMD 1.15; 95% CI, −1.11 to 3.40; p = 0.32) and mean change in Q_max from baseline to study end (WMD −0.16; 95% CI, −0.60 to 0.28; p = 0.48) (Fig. 3a and 3b). With regard to safety, prevalence of adverse events (OR 0.95; 95% CI, 0.72–1.26; p = 0.72) and withdrawal rate (OR 0.96; 95% CI, 0.66–1.40; p = 0.83) were lower with Permixon than with tamsulosin, albeit not significantly (Fig. 3c and 3d). In one study [22], however, ejaculatory dysfunction was significantly less common with Permixon than with tamsulosin (0.5% vs 4%; p = 0.007). Figure 3 shows the forest plots for efficacy and safety data of Permixon compared with tamsulosin.

Fig. 3

In the studies comparing Permixon plus tamsulosin with tamsulosin alone [24] and [25], mean IPSS significantly decreased from baseline to study end with the combination therapy versus tamsulosin alone (WMD 0.31; 95% CI 0.13–0.48; p < 0.01). A non–statistically significant trend in favor of the combination therapy was demonstrated for Q_max improvement (WMD 0.10; 95% CI, −0.02 to 0.21; p = 0.10). With regard to rate of adverse events, a non–statistically significant trend favoring tamsulosin monotherapy was shown (OR 1.63; 95% CI, 0.94–2.84; p = 0.08). Figure 4 shows the forest plots concerning efficacy and safety data of Permixon plus tamsulosin compared with tamsulosin.

3.4. Permixon versus finasteride

A single study compared efficacy and safety of Permixon and finasteride [26]. Specifically, Carraro et al. randomized 1098 patients with moderate to severe LUTS/BPH to 26-wk treatment with Permixon (n = 553) or finasteride (n = 484). Mean prostate volume was 43 and 44 ml in the Permixon and finasteride arms, respectively. Both drugs showed similar efficacy with regard to improvement in IPSS (−5.8 with Permixon vs −6.2 with finasteride; p = 0.17) and improvement in the IPSS item for quality of life (−1.5 vs −1.4, respectively; p = 0.14), although increase in Q_max was slightly higher with finasteride (+2.7 vs +3.2, respectively; p = 0.035). However, decreased libido and impotence were less common in the Permixon arm (2.2% and 1.5%, respectively) than in the finasteride arm (3% and 2.8%, respectively). Sexual function score was significantly lower (corresponding to better function) with Permixon than with finasteride (7.9 vs 9.3; p < 0.01).

3.5. Publication bias

Funnel plots of all studies used in this meta-analysis were generated for all evaluated comparisons. Only a single study lies outside the 95% CI with an even distribution about the vertical, suggesting little evidence of publication bias.

3.6. Discussion

α-Blockers, 5-ARIs, or their combination are among the standard drug therapies for bothersome LUTS/BPH [1] and [2]. Both categories of drugs are effective for improving symptoms, with 5-ARIs also able to reduce the risk of disease progression [1] and [2]. However, such therapies are associated with significant side effects, including postural hypotension and ejaculatory dysfunction for α-blockers and erectile dysfunction and loss of libido for 5-ARIs, that may reduce patient adherence to therapies [27] and [28]. Impairment of sexual function is of particular concern, considering that concomitant sexual dysfunction and LUTS are highly prevalent in adult men [29].

Phytotherapy is a common therapeutic option for LUTS/BPH and is used in approximately 17% of patients [3]. Specifically, Tacklind et al. reported on behalf of the Cochrane Collaboration a large meta-analysis of all RCTs evaluating Serenoa repens, which is the most common phytoterapeutic drug, and demonstrated no advantage for Serenoa repens in comparisons with placebo in relieving LUTS [7]. However, the latest version of the Cochrane meta-analysis suffers from some limitations. All of the different brands of Serenoa repens were pooled together, which is a questionable choice, as Bilia et al. clearly stated in the feedback on the Cochrane meta-analysis [7]. Specifically, some differences exist among the different brands of Serenoa repens, as reported by some preclinical studies, with Permixon seeming to be the brand with the highest percentage of free fatty acids and the highest in vitro efficacy [8], [9], and [10]. Consequently, the conclusion of the Cochrane meta-analysis might be affected by the fact that various kinds of extracts were pooled in the analysis. Indeed, this conclusion changed when the Cochrane Collaboration included in its meta-analysis well-designed RCTs but with different extracts of Serenoa repens obtained through different extraction processes [30] and [31]. Specifically, Barry et al. used Prosta-Urgenin Uno, an ethanolic Serenoa repens extract produced by Rottapharm Madaus (Selangor, Malaysia) [30], whereas Bent et al. evaluated a carbon dioxide Serenoa repens extract (Indena, Milan, Italy) [31]. That issue may explain why the Cochrane Collaboration concluded in a previous meta-analysis that Serenoa repens improves urologic symptoms and flow measures compared with placebo [32]. In fact, the same authors of the Cochrane meta-analysis stated that they did not know if their conclusions were generalizable to proprietary products of Serenoa repens such as Permixon, inviting investigation of the issue [7] and [33], which was the objective of the current meta-analysis. Moreover, some inaccuracies are evident in their data analysis and extraction. For example, Q_max values for two studies [21] and [34] reported in the Cochrane meta-analysis were not consistent with those reported in the original papers.

Consequently, to assess the validity of the Cochrane meta-analysis data for all brands of Serenoa repens with regard to Permixon, we performed a systematic review and meta-analysis of the RCTs assessing efficacy and safety of Permixon. We found that Permixon was more effective than placebo in reducing the number of nocturnal voids and Q_max and as effective as tamsulosin and short-term finasteride in improving LUTS. Moreover, combination therapy of Permixon and tamsulosin might provide some benefits in comparison to tamsulosin monotherapy in terms of LUTS improvement. In all comparisons with the other drugs, Permixon was associated with a favorable profile of adverse events, especially with regard to ejaculatory and erectile functions.

The findings of the present study were partially similar to those of Boyle et al, who reported a systematic review of RCTs and open-label studies on Permixon in 2004 [35]. Specifically, Boyle et al. demonstrated that Permixon was able to increase Q_max by roughly 2 ml/s, to reduce the mean number of nocturnal voids by roughly one, and to cause a five-point decrease in IPSS in comparison to placebo. Our analysis, which included studies comparing Permixon with tamsulosin or short-term finasteride, reconfirmed the same figures with regard to improvement in number of nocturnal voids and Q_max.

The present data showed stronger evidence for Permixon, the hexanic extract of Serenoa repens, either as monotherapy or in combination with tamsulosin, than the extract suggested by current guidelines for the whole category of phytoterapeutic drugs and for other Serenoa repens brands. This finding is supported by the recent assessment report on Serenoa repens released by the European Medicines Agency [36]. In this report, only the hexanic extract of Serenoa repens has been recognized as a well-established medicinal product thanks to its proven efficacy and tolerability in controlled trials. Conversely, ethanolic extracts have been categorized as traditional use products because of the lack of clinical trials, and supercritical CO₂ extracts have not been assigned any particular status because clinical studies have not provided sufficiently reliable data and because the extracts have been marketed for <15 yr. Moreover, the anti-inflammatory properties of Serenoa repens may represent a further potential advantage of this drug to improve storage and voiding LUTS. Interestingly, this direct anti-inflammatory effect has not been demonstrated for other drugs commonly prescribed for the treatment of LUTS/BPH [6].

We acknowledge that the overall quality of the available studies was limited, with the majority of RCTs being of low methodological quality. Specifically, most studies enrolled a limited number of patients, lacked sample size calculation, and did not use validated tools to measure LUTS (eg, IPSS and frequency–volume charts). Moreover, most studies assessed short-term treatment schedules, especially regarding finasteride. High-quality RCTs comparing Permixon with placebo, α-blockers, and 5-ARIs in the long term and large clinical trials testing the anti-inflammatory properties of Permixon are needed to corroborate our findings. Furthermore, how long Permixon should be continued, who will profit most from this therapy, and whether Permixon may reduce the risk of BPH progression and related complications remain unanswered questions.

3.1. Quality of evidence

Figure 1 summarizes the literature review process that led to the identification of the 12 RCTs used in the meta-analysis. Specifically, seven RCTs compared Permixon with placebo [15], [16], [17], [18], [19], [20], and [21], two RCTs compared Permixon with tamsulosin [22] and [23], two RCTs compared Permixon plus tamsulosin with placebo plus tamsulosin [24] and with tamsulosin alone [25], and a single RCT compared Permixon with finasteride [26]. Among these publications, there were four RCTs of good methodological quality (level of evidence 2) [22], [23], [24], and [26] and eight RCTs of poor methodological quality (level of evidence 3) [15], [16], [17], [18], [19], [20], [21], and [25]. Clinically speaking, all studies but three [22], [24], and [25] evaluated short-term treatment schedules; standard outcome measures, such as the International Prostate Symptom Score (IPSS) and the American Urological Association symptom index, were used in only five studies [22], [23], [24], [25], and [26].

3.2. Permixon versus placebo

Table 1 summarizes the studies reporting the efficacy and safety of Permixon in comparison to placebo [15], [16], [17], [18], [19], [20], and [21].

The number of nocturnal voids at study end were significantly lower with Permixon (WMD −0.31; 95% CI, −0.59 to −0.03; p = 0.03). Moreover, maximum urinary flow rate (Q_max) at study end was significantly higher in the patients treated with Permixon (WMD 3.37; 95% CI, 1.71–5.03; p < 0.0001). With regard to safety, the overall adverse event rates were similar for Permixon and placebo (OR 1.12; 95% CI, 0.13–9.75; p = 0.92). Finally, withdrawal rates were similar following Permixon or placebo (OR 1.52; 95% CI, 0.32–7.33; p = 0.60).

Figure 2 shows the forest plots for efficacy and safety data of Permixon in comparison to placebo.

3.3. Permixon or combination therapy versus tamsulosin

Table 2 summarizes the studies reporting efficacy and safety of Permixon or combination therapy compared with tamsulosin.

In the studies comparing Permixon with tamsulosin [22] and [23], no statistically significant difference was found in terms of mean change in IPSS from baseline to study end (WMD 1.15; 95% CI, −1.11 to 3.40; p = 0.32) and mean change in Q_max from baseline to study end (WMD −0.16; 95% CI, −0.60 to 0.28; p = 0.48) (Fig. 3a and 3b). With regard to safety, prevalence of adverse events (OR 0.95; 95% CI, 0.72–1.26; p = 0.72) and withdrawal rate (OR 0.96; 95% CI, 0.66–1.40; p = 0.83) were lower with Permixon than with tamsulosin, albeit not significantly (Fig. 3c and 3d). In one study [22], however, ejaculatory dysfunction was significantly less common with Permixon than with tamsulosin (0.5% vs 4%; p = 0.007). Figure 3 shows the forest plots for efficacy and safety data of Permixon compared with tamsulosin.

Fig. 3

In the studies comparing Permixon plus tamsulosin with tamsulosin alone [24] and [25], mean IPSS significantly decreased from baseline to study end with the combination therapy versus tamsulosin alone (WMD 0.31; 95% CI 0.13–0.48; p < 0.01). A non–statistically significant trend in favor of the combination therapy was demonstrated for Q_max improvement (WMD 0.10; 95% CI, −0.02 to 0.21; p = 0.10). With regard to rate of adverse events, a non–statistically significant trend favoring tamsulosin monotherapy was shown (OR 1.63; 95% CI, 0.94–2.84; p = 0.08). Figure 4 shows the forest plots concerning efficacy and safety data of Permixon plus tamsulosin compared with tamsulosin.

3.4. Permixon versus finasteride

A single study compared efficacy and safety of Permixon and finasteride [26]. Specifically, Carraro et al. randomized 1098 patients with moderate to severe LUTS/BPH to 26-wk treatment with Permixon (n = 553) or finasteride (n = 484). Mean prostate volume was 43 and 44 ml in the Permixon and finasteride arms, respectively. Both drugs showed similar efficacy with regard to improvement in IPSS (−5.8 with Permixon vs −6.2 with finasteride; p = 0.17) and improvement in the IPSS item for quality of life (−1.5 vs −1.4, respectively; p = 0.14), although increase in Q_max was slightly higher with finasteride (+2.7 vs +3.2, respectively; p = 0.035). However, decreased libido and impotence were less common in the Permixon arm (2.2% and 1.5%, respectively) than in the finasteride arm (3% and 2.8%, respectively). Sexual function score was significantly lower (corresponding to better function) with Permixon than with finasteride (7.9 vs 9.3; p < 0.01).

3.5. Publication bias

Funnel plots of all studies used in this meta-analysis were generated for all evaluated comparisons. Only a single study lies outside the 95% CI with an even distribution about the vertical, suggesting little evidence of publication bias.

3.6. Discussion

α-Blockers, 5-ARIs, or their combination are among the standard drug therapies for bothersome LUTS/BPH [1] and [2]. Both categories of drugs are effective for improving symptoms, with 5-ARIs also able to reduce the risk of disease progression [1] and [2]. However, such therapies are associated with significant side effects, including postural hypotension and ejaculatory dysfunction for α-blockers and erectile dysfunction and loss of libido for 5-ARIs, that may reduce patient adherence to therapies [27] and [28]. Impairment of sexual function is of particular concern, considering that concomitant sexual dysfunction and LUTS are highly prevalent in adult men [29].

Phytotherapy is a common therapeutic option for LUTS/BPH and is used in approximately 17% of patients [3]. Specifically, Tacklind et al. reported on behalf of the Cochrane Collaboration a large meta-analysis of all RCTs evaluating Serenoa repens, which is the most common phytoterapeutic drug, and demonstrated no advantage for Serenoa repens in comparisons with placebo in relieving LUTS [7]. However, the latest version of the Cochrane meta-analysis suffers from some limitations. All of the different brands of Serenoa repens were pooled together, which is a questionable choice, as Bilia et al. clearly stated in the feedback on the Cochrane meta-analysis [7]. Specifically, some differences exist among the different brands of Serenoa repens, as reported by some preclinical studies, with Permixon seeming to be the brand with the highest percentage of free fatty acids and the highest in vitro efficacy [8], [9], and [10]. Consequently, the conclusion of the Cochrane meta-analysis might be affected by the fact that various kinds of extracts were pooled in the analysis. Indeed, this conclusion changed when the Cochrane Collaboration included in its meta-analysis well-designed RCTs but with different extracts of Serenoa repens obtained through different extraction processes [30] and [31]. Specifically, Barry et al. used Prosta-Urgenin Uno, an ethanolic Serenoa repens extract produced by Rottapharm Madaus (Selangor, Malaysia) [30], whereas Bent et al. evaluated a carbon dioxide Serenoa repens extract (Indena, Milan, Italy) [31]. That issue may explain why the Cochrane Collaboration concluded in a previous meta-analysis that Serenoa repens improves urologic symptoms and flow measures compared with placebo [32]. In fact, the same authors of the Cochrane meta-analysis stated that they did not know if their conclusions were generalizable to proprietary products of Serenoa repens such as Permixon, inviting investigation of the issue [7] and [33], which was the objective of the current meta-analysis. Moreover, some inaccuracies are evident in their data analysis and extraction. For example, Q_max values for two studies [21] and [34] reported in the Cochrane meta-analysis were not consistent with those reported in the original papers.

Consequently, to assess the validity of the Cochrane meta-analysis data for all brands of Serenoa repens with regard to Permixon, we performed a systematic review and meta-analysis of the RCTs assessing efficacy and safety of Permixon. We found that Permixon was more effective than placebo in reducing the number of nocturnal voids and Q_max and as effective as tamsulosin and short-term finasteride in improving LUTS. Moreover, combination therapy of Permixon and tamsulosin might provide some benefits in comparison to tamsulosin monotherapy in terms of LUTS improvement. In all comparisons with the other drugs, Permixon was associated with a favorable profile of adverse events, especially with regard to ejaculatory and erectile functions.

The findings of the present study were partially similar to those of Boyle et al, who reported a systematic review of RCTs and open-label studies on Permixon in 2004 [35]. Specifically, Boyle et al. demonstrated that Permixon was able to increase Q_max by roughly 2 ml/s, to reduce the mean number of nocturnal voids by roughly one, and to cause a five-point decrease in IPSS in comparison to placebo. Our analysis, which included studies comparing Permixon with tamsulosin or short-term finasteride, reconfirmed the same figures with regard to improvement in number of nocturnal voids and Q_max.

The present data showed stronger evidence for Permixon, the hexanic extract of Serenoa repens, either as monotherapy or in combination with tamsulosin, than the extract suggested by current guidelines for the whole category of phytoterapeutic drugs and for other Serenoa repens brands. This finding is supported by the recent assessment report on Serenoa repens released by the European Medicines Agency [36]. In this report, only the hexanic extract of Serenoa repens has been recognized as a well-established medicinal product thanks to its proven efficacy and tolerability in controlled trials. Conversely, ethanolic extracts have been categorized as traditional use products because of the lack of clinical trials, and supercritical CO₂ extracts have not been assigned any particular status because clinical studies have not provided sufficiently reliable data and because the extracts have been marketed for <15 yr. Moreover, the anti-inflammatory properties of Serenoa repens may represent a further potential advantage of this drug to improve storage and voiding LUTS. Interestingly, this direct anti-inflammatory effect has not been demonstrated for other drugs commonly prescribed for the treatment of LUTS/BPH [6].

We acknowledge that the overall quality of the available studies was limited, with the majority of RCTs being of low methodological quality. Specifically, most studies enrolled a limited number of patients, lacked sample size calculation, and did not use validated tools to measure LUTS (eg, IPSS and frequency–volume charts). Moreover, most studies assessed short-term treatment schedules, especially regarding finasteride. High-quality RCTs comparing Permixon with placebo, α-blockers, and 5-ARIs in the long term and large clinical trials testing the anti-inflammatory properties of Permixon are needed to corroborate our findings. Furthermore, how long Permixon should be continued, who will profit most from this therapy, and whether Permixon may reduce the risk of BPH progression and related complications remain unanswered questions.

3.1. Quality of evidence

Figure 1 summarizes the literature review process that led to the identification of the 12 RCTs used in the meta-analysis. Specifically, seven RCTs compared Permixon with placebo [15], [16], [17], [18], [19], [20], and [21], two RCTs compared Permixon with tamsulosin [22] and [23], two RCTs compared Permixon plus tamsulosin with placebo plus tamsulosin [24] and with tamsulosin alone [25], and a single RCT compared Permixon with finasteride [26]. Among these publications, there were four RCTs of good methodological quality (level of evidence 2) [22], [23], [24], and [26] and eight RCTs of poor methodological quality (level of evidence 3) [15], [16], [17], [18], [19], [20], [21], and [25]. Clinically speaking, all studies but three [22], [24], and [25] evaluated short-term treatment schedules; standard outcome measures, such as the International Prostate Symptom Score (IPSS) and the American Urological Association symptom index, were used in only five studies [22], [23], [24], [25], and [26].

3.2. Permixon versus placebo

Table 1 summarizes the studies reporting the efficacy and safety of Permixon in comparison to placebo [15], [16], [17], [18], [19], [20], and [21].

The number of nocturnal voids at study end were significantly lower with Permixon (WMD −0.31; 95% CI, −0.59 to −0.03; p = 0.03). Moreover, maximum urinary flow rate (Q_max) at study end was significantly higher in the patients treated with Permixon (WMD 3.37; 95% CI, 1.71–5.03; p < 0.0001). With regard to safety, the overall adverse event rates were similar for Permixon and placebo (OR 1.12; 95% CI, 0.13–9.75; p = 0.92). Finally, withdrawal rates were similar following Permixon or placebo (OR 1.52; 95% CI, 0.32–7.33; p = 0.60).

Figure 2 shows the forest plots for efficacy and safety data of Permixon in comparison to placebo.

3.3. Permixon or combination therapy versus tamsulosin

Table 2 summarizes the studies reporting efficacy and safety of Permixon or combination therapy compared with tamsulosin.

In the studies comparing Permixon with tamsulosin [22] and [23], no statistically significant difference was found in terms of mean change in IPSS from baseline to study end (WMD 1.15; 95% CI, −1.11 to 3.40; p = 0.32) and mean change in Q_max from baseline to study end (WMD −0.16; 95% CI, −0.60 to 0.28; p = 0.48) (Fig. 3a and 3b). With regard to safety, prevalence of adverse events (OR 0.95; 95% CI, 0.72–1.26; p = 0.72) and withdrawal rate (OR 0.96; 95% CI, 0.66–1.40; p = 0.83) were lower with Permixon than with tamsulosin, albeit not significantly (Fig. 3c and 3d). In one study [22], however, ejaculatory dysfunction was significantly less common with Permixon than with tamsulosin (0.5% vs 4%; p = 0.007). Figure 3 shows the forest plots for efficacy and safety data of Permixon compared with tamsulosin.

Fig. 3

In the studies comparing Permixon plus tamsulosin with tamsulosin alone [24] and [25], mean IPSS significantly decreased from baseline to study end with the combination therapy versus tamsulosin alone (WMD 0.31; 95% CI 0.13–0.48; p < 0.01). A non–statistically significant trend in favor of the combination therapy was demonstrated for Q_max improvement (WMD 0.10; 95% CI, −0.02 to 0.21; p = 0.10). With regard to rate of adverse events, a non–statistically significant trend favoring tamsulosin monotherapy was shown (OR 1.63; 95% CI, 0.94–2.84; p = 0.08). Figure 4 shows the forest plots concerning efficacy and safety data of Permixon plus tamsulosin compared with tamsulosin.

3.4. Permixon versus finasteride

A single study compared efficacy and safety of Permixon and finasteride [26]. Specifically, Carraro et al. randomized 1098 patients with moderate to severe LUTS/BPH to 26-wk treatment with Permixon (n = 553) or finasteride (n = 484). Mean prostate volume was 43 and 44 ml in the Permixon and finasteride arms, respectively. Both drugs showed similar efficacy with regard to improvement in IPSS (−5.8 with Permixon vs −6.2 with finasteride; p = 0.17) and improvement in the IPSS item for quality of life (−1.5 vs −1.4, respectively; p = 0.14), although increase in Q_max was slightly higher with finasteride (+2.7 vs +3.2, respectively; p = 0.035). However, decreased libido and impotence were less common in the Permixon arm (2.2% and 1.5%, respectively) than in the finasteride arm (3% and 2.8%, respectively). Sexual function score was significantly lower (corresponding to better function) with Permixon than with finasteride (7.9 vs 9.3; p < 0.01).

3.5. Publication bias

Funnel plots of all studies used in this meta-analysis were generated for all evaluated comparisons. Only a single study lies outside the 95% CI with an even distribution about the vertical, suggesting little evidence of publication bias.

3.6. Discussion

α-Blockers, 5-ARIs, or their combination are among the standard drug therapies for bothersome LUTS/BPH [1] and [2]. Both categories of drugs are effective for improving symptoms, with 5-ARIs also able to reduce the risk of disease progression [1] and [2]. However, such therapies are associated with significant side effects, including postural hypotension and ejaculatory dysfunction for α-blockers and erectile dysfunction and loss of libido for 5-ARIs, that may reduce patient adherence to therapies [27] and [28]. Impairment of sexual function is of particular concern, considering that concomitant sexual dysfunction and LUTS are highly prevalent in adult men [29].

Phytotherapy is a common therapeutic option for LUTS/BPH and is used in approximately 17% of patients [3]. Specifically, Tacklind et al. reported on behalf of the Cochrane Collaboration a large meta-analysis of all RCTs evaluating Serenoa repens, which is the most common phytoterapeutic drug, and demonstrated no advantage for Serenoa repens in comparisons with placebo in relieving LUTS [7]. However, the latest version of the Cochrane meta-analysis suffers from some limitations. All of the different brands of Serenoa repens were pooled together, which is a questionable choice, as Bilia et al. clearly stated in the feedback on the Cochrane meta-analysis [7]. Specifically, some differences exist among the different brands of Serenoa repens, as reported by some preclinical studies, with Permixon seeming to be the brand with the highest percentage of free fatty acids and the highest in vitro efficacy [8], [9], and [10]. Consequently, the conclusion of the Cochrane meta-analysis might be affected by the fact that various kinds of extracts were pooled in the analysis. Indeed, this conclusion changed when the Cochrane Collaboration included in its meta-analysis well-designed RCTs but with different extracts of Serenoa repens obtained through different extraction processes [30] and [31]. Specifically, Barry et al. used Prosta-Urgenin Uno, an ethanolic Serenoa repens extract produced by Rottapharm Madaus (Selangor, Malaysia) [30], whereas Bent et al. evaluated a carbon dioxide Serenoa repens extract (Indena, Milan, Italy) [31]. That issue may explain why the Cochrane Collaboration concluded in a previous meta-analysis that Serenoa repens improves urologic symptoms and flow measures compared with placebo [32]. In fact, the same authors of the Cochrane meta-analysis stated that they did not know if their conclusions were generalizable to proprietary products of Serenoa repens such as Permixon, inviting investigation of the issue [7] and [33], which was the objective of the current meta-analysis. Moreover, some inaccuracies are evident in their data analysis and extraction. For example, Q_max values for two studies [21] and [34] reported in the Cochrane meta-analysis were not consistent with those reported in the original papers.

Consequently, to assess the validity of the Cochrane meta-analysis data for all brands of Serenoa repens with regard to Permixon, we performed a systematic review and meta-analysis of the RCTs assessing efficacy and safety of Permixon. We found that Permixon was more effective than placebo in reducing the number of nocturnal voids and Q_max and as effective as tamsulosin and short-term finasteride in improving LUTS. Moreover, combination therapy of Permixon and tamsulosin might provide some benefits in comparison to tamsulosin monotherapy in terms of LUTS improvement. In all comparisons with the other drugs, Permixon was associated with a favorable profile of adverse events, especially with regard to ejaculatory and erectile functions.

The findings of the present study were partially similar to those of Boyle et al, who reported a systematic review of RCTs and open-label studies on Permixon in 2004 [35]. Specifically, Boyle et al. demonstrated that Permixon was able to increase Q_max by roughly 2 ml/s, to reduce the mean number of nocturnal voids by roughly one, and to cause a five-point decrease in IPSS in comparison to placebo. Our analysis, which included studies comparing Permixon with tamsulosin or short-term finasteride, reconfirmed the same figures with regard to improvement in number of nocturnal voids and Q_max.

The present data showed stronger evidence for Permixon, the hexanic extract of Serenoa repens, either as monotherapy or in combination with tamsulosin, than the extract suggested by current guidelines for the whole category of phytoterapeutic drugs and for other Serenoa repens brands. This finding is supported by the recent assessment report on Serenoa repens released by the European Medicines Agency [36]. In this report, only the hexanic extract of Serenoa repens has been recognized as a well-established medicinal product thanks to its proven efficacy and tolerability in controlled trials. Conversely, ethanolic extracts have been categorized as traditional use products because of the lack of clinical trials, and supercritical CO₂ extracts have not been assigned any particular status because clinical studies have not provided sufficiently reliable data and because the extracts have been marketed for <15 yr. Moreover, the anti-inflammatory properties of Serenoa repens may represent a further potential advantage of this drug to improve storage and voiding LUTS. Interestingly, this direct anti-inflammatory effect has not been demonstrated for other drugs commonly prescribed for the treatment of LUTS/BPH [6].

We acknowledge that the overall quality of the available studies was limited, with the majority of RCTs being of low methodological quality. Specifically, most studies enrolled a limited number of patients, lacked sample size calculation, and did not use validated tools to measure LUTS (eg, IPSS and frequency–volume charts). Moreover, most studies assessed short-term treatment schedules, especially regarding finasteride. High-quality RCTs comparing Permixon with placebo, α-blockers, and 5-ARIs in the long term and large clinical trials testing the anti-inflammatory properties of Permixon are needed to corroborate our findings. Furthermore, how long Permixon should be continued, who will profit most from this therapy, and whether Permixon may reduce the risk of BPH progression and related complications remain unanswered questions.

3.1. Quality of evidence

Figure 1 summarizes the literature review process that led to the identification of the 12 RCTs used in the meta-analysis. Specifically, seven RCTs compared Permixon with placebo [15], [16], [17], [18], [19], [20], and [21], two RCTs compared Permixon with tamsulosin [22] and [23], two RCTs compared Permixon plus tamsulosin with placebo plus tamsulosin [24] and with tamsulosin alone [25], and a single RCT compared Permixon with finasteride [26]. Among these publications, there were four RCTs of good methodological quality (level of evidence 2) [22], [23], [24], and [26] and eight RCTs of poor methodological quality (level of evidence 3) [15], [16], [17], [18], [19], [20], [21], and [25]. Clinically speaking, all studies but three [22], [24], and [25] evaluated short-term treatment schedules; standard outcome measures, such as the International Prostate Symptom Score (IPSS) and the American Urological Association symptom index, were used in only five studies [22], [23], [24], [25], and [26].

3.2. Permixon versus placebo

Table 1 summarizes the studies reporting the efficacy and safety of Permixon in comparison to placebo [15], [16], [17], [18], [19], [20], and [21].

The number of nocturnal voids at study end were significantly lower with Permixon (WMD −0.31; 95% CI, −0.59 to −0.03; p = 0.03). Moreover, maximum urinary flow rate (Q_max) at study end was significantly higher in the patients treated with Permixon (WMD 3.37; 95% CI, 1.71–5.03; p < 0.0001). With regard to safety, the overall adverse event rates were similar for Permixon and placebo (OR 1.12; 95% CI, 0.13–9.75; p = 0.92). Finally, withdrawal rates were similar following Permixon or placebo (OR 1.52; 95% CI, 0.32–7.33; p = 0.60).

Figure 2 shows the forest plots for efficacy and safety data of Permixon in comparison to placebo.

3.3. Permixon or combination therapy versus tamsulosin

Table 2 summarizes the studies reporting efficacy and safety of Permixon or combination therapy compared with tamsulosin.

In the studies comparing Permixon with tamsulosin [22] and [23], no statistically significant difference was found in terms of mean change in IPSS from baseline to study end (WMD 1.15; 95% CI, −1.11 to 3.40; p = 0.32) and mean change in Q_max from baseline to study end (WMD −0.16; 95% CI, −0.60 to 0.28; p = 0.48) (Fig. 3a and 3b). With regard to safety, prevalence of adverse events (OR 0.95; 95% CI, 0.72–1.26; p = 0.72) and withdrawal rate (OR 0.96; 95% CI, 0.66–1.40; p = 0.83) were lower with Permixon than with tamsulosin, albeit not significantly (Fig. 3c and 3d). In one study [22], however, ejaculatory dysfunction was significantly less common with Permixon than with tamsulosin (0.5% vs 4%; p = 0.007). Figure 3 shows the forest plots for efficacy and safety data of Permixon compared with tamsulosin.

Fig. 3

In the studies comparing Permixon plus tamsulosin with tamsulosin alone [24] and [25], mean IPSS significantly decreased from baseline to study end with the combination therapy versus tamsulosin alone (WMD 0.31; 95% CI 0.13–0.48; p < 0.01). A non–statistically significant trend in favor of the combination therapy was demonstrated for Q_max improvement (WMD 0.10; 95% CI, −0.02 to 0.21; p = 0.10). With regard to rate of adverse events, a non–statistically significant trend favoring tamsulosin monotherapy was shown (OR 1.63; 95% CI, 0.94–2.84; p = 0.08). Figure 4 shows the forest plots concerning efficacy and safety data of Permixon plus tamsulosin compared with tamsulosin.

3.4. Permixon versus finasteride

A single study compared efficacy and safety of Permixon and finasteride [26]. Specifically, Carraro et al. randomized 1098 patients with moderate to severe LUTS/BPH to 26-wk treatment with Permixon (n = 553) or finasteride (n = 484). Mean prostate volume was 43 and 44 ml in the Permixon and finasteride arms, respectively. Both drugs showed similar efficacy with regard to improvement in IPSS (−5.8 with Permixon vs −6.2 with finasteride; p = 0.17) and improvement in the IPSS item for quality of life (−1.5 vs −1.4, respectively; p = 0.14), although increase in Q_max was slightly higher with finasteride (+2.7 vs +3.2, respectively; p = 0.035). However, decreased libido and impotence were less common in the Permixon arm (2.2% and 1.5%, respectively) than in the finasteride arm (3% and 2.8%, respectively). Sexual function score was significantly lower (corresponding to better function) with Permixon than with finasteride (7.9 vs 9.3; p < 0.01).

3.5. Publication bias

Funnel plots of all studies used in this meta-analysis were generated for all evaluated comparisons. Only a single study lies outside the 95% CI with an even distribution about the vertical, suggesting little evidence of publication bias.

3.6. Discussion

α-Blockers, 5-ARIs, or their combination are among the standard drug therapies for bothersome LUTS/BPH [1] and [2]. Both categories of drugs are effective for improving symptoms, with 5-ARIs also able to reduce the risk of disease progression [1] and [2]. However, such therapies are associated with significant side effects, including postural hypotension and ejaculatory dysfunction for α-blockers and erectile dysfunction and loss of libido for 5-ARIs, that may reduce patient adherence to therapies [27] and [28]. Impairment of sexual function is of particular concern, considering that concomitant sexual dysfunction and LUTS are highly prevalent in adult men [29].

Phytotherapy is a common therapeutic option for LUTS/BPH and is used in approximately 17% of patients [3]. Specifically, Tacklind et al. reported on behalf of the Cochrane Collaboration a large meta-analysis of all RCTs evaluating Serenoa repens, which is the most common phytoterapeutic drug, and demonstrated no advantage for Serenoa repens in comparisons with placebo in relieving LUTS [7]. However, the latest version of the Cochrane meta-analysis suffers from some limitations. All of the different brands of Serenoa repens were pooled together, which is a questionable choice, as Bilia et al. clearly stated in the feedback on the Cochrane meta-analysis [7]. Specifically, some differences exist among the different brands of Serenoa repens, as reported by some preclinical studies, with Permixon seeming to be the brand with the highest percentage of free fatty acids and the highest in vitro efficacy [8], [9], and [10]. Consequently, the conclusion of the Cochrane meta-analysis might be affected by the fact that various kinds of extracts were pooled in the analysis. Indeed, this conclusion changed when the Cochrane Collaboration included in its meta-analysis well-designed RCTs but with different extracts of Serenoa repens obtained through different extraction processes [30] and [31]. Specifically, Barry et al. used Prosta-Urgenin Uno, an ethanolic Serenoa repens extract produced by Rottapharm Madaus (Selangor, Malaysia) [30], whereas Bent et al. evaluated a carbon dioxide Serenoa repens extract (Indena, Milan, Italy) [31]. That issue may explain why the Cochrane Collaboration concluded in a previous meta-analysis that Serenoa repens improves urologic symptoms and flow measures compared with placebo [32]. In fact, the same authors of the Cochrane meta-analysis stated that they did not know if their conclusions were generalizable to proprietary products of Serenoa repens such as Permixon, inviting investigation of the issue [7] and [33], which was the objective of the current meta-analysis. Moreover, some inaccuracies are evident in their data analysis and extraction. For example, Q_max values for two studies [21] and [34] reported in the Cochrane meta-analysis were not consistent with those reported in the original papers.

Consequently, to assess the validity of the Cochrane meta-analysis data for all brands of Serenoa repens with regard to Permixon, we performed a systematic review and meta-analysis of the RCTs assessing efficacy and safety of Permixon. We found that Permixon was more effective than placebo in reducing the number of nocturnal voids and Q_max and as effective as tamsulosin and short-term finasteride in improving LUTS. Moreover, combination therapy of Permixon and tamsulosin might provide some benefits in comparison to tamsulosin monotherapy in terms of LUTS improvement. In all comparisons with the other drugs, Permixon was associated with a favorable profile of adverse events, especially with regard to ejaculatory and erectile functions.

The findings of the present study were partially similar to those of Boyle et al, who reported a systematic review of RCTs and open-label studies on Permixon in 2004 [35]. Specifically, Boyle et al. demonstrated that Permixon was able to increase Q_max by roughly 2 ml/s, to reduce the mean number of nocturnal voids by roughly one, and to cause a five-point decrease in IPSS in comparison to placebo. Our analysis, which included studies comparing Permixon with tamsulosin or short-term finasteride, reconfirmed the same figures with regard to improvement in number of nocturnal voids and Q_max.

The present data showed stronger evidence for Permixon, the hexanic extract of Serenoa repens, either as monotherapy or in combination with tamsulosin, than the extract suggested by current guidelines for the whole category of phytoterapeutic drugs and for other Serenoa repens brands. This finding is supported by the recent assessment report on Serenoa repens released by the European Medicines Agency [36]. In this report, only the hexanic extract of Serenoa repens has been recognized as a well-established medicinal product thanks to its proven efficacy and tolerability in controlled trials. Conversely, ethanolic extracts have been categorized as traditional use products because of the lack of clinical trials, and supercritical CO₂ extracts have not been assigned any particular status because clinical studies have not provided sufficiently reliable data and because the extracts have been marketed for <15 yr. Moreover, the anti-inflammatory properties of Serenoa repens may represent a further potential advantage of this drug to improve storage and voiding LUTS. Interestingly, this direct anti-inflammatory effect has not been demonstrated for other drugs commonly prescribed for the treatment of LUTS/BPH [6].

We acknowledge that the overall quality of the available studies was limited, with the majority of RCTs being of low methodological quality. Specifically, most studies enrolled a limited number of patients, lacked sample size calculation, and did not use validated tools to measure LUTS (eg, IPSS and frequency–volume charts). Moreover, most studies assessed short-term treatment schedules, especially regarding finasteride. High-quality RCTs comparing Permixon with placebo, α-blockers, and 5-ARIs in the long term and large clinical trials testing the anti-inflammatory properties of Permixon are needed to corroborate our findings. Furthermore, how long Permixon should be continued, who will profit most from this therapy, and whether Permixon may reduce the risk of BPH progression and related complications remain unanswered questions.

3.1. Quality of evidence

Figure 1 summarizes the literature review process that led to the identification of the 12 RCTs used in the meta-analysis. Specifically, seven RCTs compared Permixon with placebo [15], [16], [17], [18], [19], [20], and [21], two RCTs compared Permixon with tamsulosin [22] and [23], two RCTs compared Permixon plus tamsulosin with placebo plus tamsulosin [24] and with tamsulosin alone [25], and a single RCT compared Permixon with finasteride [26]. Among these publications, there were four RCTs of good methodological quality (level of evidence 2) [22], [23], [24], and [26] and eight RCTs of poor methodological quality (level of evidence 3) [15], [16], [17], [18], [19], [20], [21], and [25]. Clinically speaking, all studies but three [22], [24], and [25] evaluated short-term treatment schedules; standard outcome measures, such as the International Prostate Symptom Score (IPSS) and the American Urological Association symptom index, were used in only five studies [22], [23], [24], [25], and [26].

3.2. Permixon versus placebo

Table 1 summarizes the studies reporting the efficacy and safety of Permixon in comparison to placebo [15], [16], [17], [18], [19], [20], and [21].

The number of nocturnal voids at study end were significantly lower with Permixon (WMD −0.31; 95% CI, −0.59 to −0.03; p = 0.03). Moreover, maximum urinary flow rate (Q_max) at study end was significantly higher in the patients treated with Permixon (WMD 3.37; 95% CI, 1.71–5.03; p < 0.0001). With regard to safety, the overall adverse event rates were similar for Permixon and placebo (OR 1.12; 95% CI, 0.13–9.75; p = 0.92). Finally, withdrawal rates were similar following Permixon or placebo (OR 1.52; 95% CI, 0.32–7.33; p = 0.60).

Figure 2 shows the forest plots for efficacy and safety data of Permixon in comparison to placebo.

3.3. Permixon or combination therapy versus tamsulosin

Table 2 summarizes the studies reporting efficacy and safety of Permixon or combination therapy compared with tamsulosin.

In the studies comparing Permixon with tamsulosin [22] and [23], no statistically significant difference was found in terms of mean change in IPSS from baseline to study end (WMD 1.15; 95% CI, −1.11 to 3.40; p = 0.32) and mean change in Q_max from baseline to study end (WMD −0.16; 95% CI, −0.60 to 0.28; p = 0.48) (Fig. 3a and 3b). With regard to safety, prevalence of adverse events (OR 0.95; 95% CI, 0.72–1.26; p = 0.72) and withdrawal rate (OR 0.96; 95% CI, 0.66–1.40; p = 0.83) were lower with Permixon than with tamsulosin, albeit not significantly (Fig. 3c and 3d). In one study [22], however, ejaculatory dysfunction was significantly less common with Permixon than with tamsulosin (0.5% vs 4%; p = 0.007). Figure 3 shows the forest plots for efficacy and safety data of Permixon compared with tamsulosin.

Fig. 3

In the studies comparing Permixon plus tamsulosin with tamsulosin alone [24] and [25], mean IPSS significantly decreased from baseline to study end with the combination therapy versus tamsulosin alone (WMD 0.31; 95% CI 0.13–0.48; p < 0.01). A non–statistically significant trend in favor of the combination therapy was demonstrated for Q_max improvement (WMD 0.10; 95% CI, −0.02 to 0.21; p = 0.10). With regard to rate of adverse events, a non–statistically significant trend favoring tamsulosin monotherapy was shown (OR 1.63; 95% CI, 0.94–2.84; p = 0.08). Figure 4 shows the forest plots concerning efficacy and safety data of Permixon plus tamsulosin compared with tamsulosin.

3.4. Permixon versus finasteride

A single study compared efficacy and safety of Permixon and finasteride [26]. Specifically, Carraro et al. randomized 1098 patients with moderate to severe LUTS/BPH to 26-wk treatment with Permixon (n = 553) or finasteride (n = 484). Mean prostate volume was 43 and 44 ml in the Permixon and finasteride arms, respectively. Both drugs showed similar efficacy with regard to improvement in IPSS (−5.8 with Permixon vs −6.2 with finasteride; p = 0.17) and improvement in the IPSS item for quality of life (−1.5 vs −1.4, respectively; p = 0.14), although increase in Q_max was slightly higher with finasteride (+2.7 vs +3.2, respectively; p = 0.035). However, decreased libido and impotence were less common in the Permixon arm (2.2% and 1.5%, respectively) than in the finasteride arm (3% and 2.8%, respectively). Sexual function score was significantly lower (corresponding to better function) with Permixon than with finasteride (7.9 vs 9.3; p < 0.01).

3.5. Publication bias

Funnel plots of all studies used in this meta-analysis were generated for all evaluated comparisons. Only a single study lies outside the 95% CI with an even distribution about the vertical, suggesting little evidence of publication bias.

3.6. Discussion

α-Blockers, 5-ARIs, or their combination are among the standard drug therapies for bothersome LUTS/BPH [1] and [2]. Both categories of drugs are effective for improving symptoms, with 5-ARIs also able to reduce the risk of disease progression [1] and [2]. However, such therapies are associated with significant side effects, including postural hypotension and ejaculatory dysfunction for α-blockers and erectile dysfunction and loss of libido for 5-ARIs, that may reduce patient adherence to therapies [27] and [28]. Impairment of sexual function is of particular concern, considering that concomitant sexual dysfunction and LUTS are highly prevalent in adult men [29].

Phytotherapy is a common therapeutic option for LUTS/BPH and is used in approximately 17% of patients [3]. Specifically, Tacklind et al. reported on behalf of the Cochrane Collaboration a large meta-analysis of all RCTs evaluating Serenoa repens, which is the most common phytoterapeutic drug, and demonstrated no advantage for Serenoa repens in comparisons with placebo in relieving LUTS [7]. However, the latest version of the Cochrane meta-analysis suffers from some limitations. All of the different brands of Serenoa repens were pooled together, which is a questionable choice, as Bilia et al. clearly stated in the feedback on the Cochrane meta-analysis [7]. Specifically, some differences exist among the different brands of Serenoa repens, as reported by some preclinical studies, with Permixon seeming to be the brand with the highest percentage of free fatty acids and the highest in vitro efficacy [8], [9], and [10]. Consequently, the conclusion of the Cochrane meta-analysis might be affected by the fact that various kinds of extracts were pooled in the analysis. Indeed, this conclusion changed when the Cochrane Collaboration included in its meta-analysis well-designed RCTs but with different extracts of Serenoa repens obtained through different extraction processes [30] and [31]. Specifically, Barry et al. used Prosta-Urgenin Uno, an ethanolic Serenoa repens extract produced by Rottapharm Madaus (Selangor, Malaysia) [30], whereas Bent et al. evaluated a carbon dioxide Serenoa repens extract (Indena, Milan, Italy) [31]. That issue may explain why the Cochrane Collaboration concluded in a previous meta-analysis that Serenoa repens improves urologic symptoms and flow measures compared with placebo [32]. In fact, the same authors of the Cochrane meta-analysis stated that they did not know if their conclusions were generalizable to proprietary products of Serenoa repens such as Permixon, inviting investigation of the issue [7] and [33], which was the objective of the current meta-analysis. Moreover, some inaccuracies are evident in their data analysis and extraction. For example, Q_max values for two studies [21] and [34] reported in the Cochrane meta-analysis were not consistent with those reported in the original papers.

Consequently, to assess the validity of the Cochrane meta-analysis data for all brands of Serenoa repens with regard to Permixon, we performed a systematic review and meta-analysis of the RCTs assessing efficacy and safety of Permixon. We found that Permixon was more effective than placebo in reducing the number of nocturnal voids and Q_max and as effective as tamsulosin and short-term finasteride in improving LUTS. Moreover, combination therapy of Permixon and tamsulosin might provide some benefits in comparison to tamsulosin monotherapy in terms of LUTS improvement. In all comparisons with the other drugs, Permixon was associated with a favorable profile of adverse events, especially with regard to ejaculatory and erectile functions.

The findings of the present study were partially similar to those of Boyle et al, who reported a systematic review of RCTs and open-label studies on Permixon in 2004 [35]. Specifically, Boyle et al. demonstrated that Permixon was able to increase Q_max by roughly 2 ml/s, to reduce the mean number of nocturnal voids by roughly one, and to cause a five-point decrease in IPSS in comparison to placebo. Our analysis, which included studies comparing Permixon with tamsulosin or short-term finasteride, reconfirmed the same figures with regard to improvement in number of nocturnal voids and Q_max.

The present data showed stronger evidence for Permixon, the hexanic extract of Serenoa repens, either as monotherapy or in combination with tamsulosin, than the extract suggested by current guidelines for the whole category of phytoterapeutic drugs and for other Serenoa repens brands. This finding is supported by the recent assessment report on Serenoa repens released by the European Medicines Agency [36]. In this report, only the hexanic extract of Serenoa repens has been recognized as a well-established medicinal product thanks to its proven efficacy and tolerability in controlled trials. Conversely, ethanolic extracts have been categorized as traditional use products because of the lack of clinical trials, and supercritical CO₂ extracts have not been assigned any particular status because clinical studies have not provided sufficiently reliable data and because the extracts have been marketed for <15 yr. Moreover, the anti-inflammatory properties of Serenoa repens may represent a further potential advantage of this drug to improve storage and voiding LUTS. Interestingly, this direct anti-inflammatory effect has not been demonstrated for other drugs commonly prescribed for the treatment of LUTS/BPH [6].

We acknowledge that the overall quality of the available studies was limited, with the majority of RCTs being of low methodological quality. Specifically, most studies enrolled a limited number of patients, lacked sample size calculation, and did not use validated tools to measure LUTS (eg, IPSS and frequency–volume charts). Moreover, most studies assessed short-term treatment schedules, especially regarding finasteride. High-quality RCTs comparing Permixon with placebo, α-blockers, and 5-ARIs in the long term and large clinical trials testing the anti-inflammatory properties of Permixon are needed to corroborate our findings. Furthermore, how long Permixon should be continued, who will profit most from this therapy, and whether Permixon may reduce the risk of BPH progression and related complications remain unanswered questions.