Effects of pumpkin on benign prostatic hyperplasia: a systematic review and meta-analysis

Introduction: Benign prostatic hyperplasia (BPH) is a common condition affecting aging men worldwide. Pumpkin and its extracts are enriched with antioxidant, anti-androgenic and anti-inflammatory properties. Objectives : To systematically review the effect of pumpkin on the reduction of lower urinary tract symptoms (LUTS) due to bladder outflow obstruction in BPH Methods : The review (PROSPERO ID: CRD42020181395) perused PubMed, EMBASE, Cochrane Library, CINAHL and AGRICOLA databases in two rounds independently by two reviewers according to three screening questions: being a pumpkin related human study, having BPH related outcomes and including quantifiable measurements. Risk of bias assessments were done with GRADE and ROBINS-I Criteria. After assessing the clinical, methodological and statistical heterogeneity, meta-analysis was performed for one outcome with mean difference (MD) using Inverse Variance Method with Review Manager 5.4 software. Results : Out of the 242 articles with de-duplication and screening questions, 10 based on interventional studies among adult males with LUTS were selected in ??third round (8 on pumpkin seeds alone; 2 on pumpkin combined products). International Prostate Symptoms Score (IPSS) was described in nine articles. Other outcomes assessed were quality of life (QOL), prostate volume, Prostate Specific Antigen (PSA), uroflow rate, post-voidal residual volume, frequency of micturition during day and night, nocturia and erectile dysfunction. Meta-analysis of three studies (650 and 631 subjects in experimental and control groups) was performed for IPSS reduction. Pooled MD (I2=0%) was -1.41 (95% CI: -1.97, -0.85) with random effect assumption. In all relevant studies, the intervention group demonstrated significant IPSS reduction and increased QOL. One study each demonstrated significant reduction in prostate volume and residual volume; three studies in uroflow rate; and none on PSA improvement. Conclusions & Recommendations : Pumpkin seeds may be beneficial in reducing IPSS and to some extent in improving prostate and residual volume, uroflow rate and QOL in BPH. Further studies are needed to strengthen the hypothesis.


Introduction
Benign prostatic hyperplasia (BPH) is one of the commonest benign neoplasms of males (1), specifically affecting the aging men worldwide, with nearly 1 in 4 men suffering from this condition over their lifetime (2).The prevalence of BPH is more than 40% among men aged 50-60 years and 70% among men aged 61-70 years and 90% among men in the ninth decade (3)(4).
The BPH is a histological diagnosis based on the proliferation of both smooth muscle and epithelial cells in the prostatic transition zone.It causes bladder outlet obstruction giving rise to associated LUTS comprising voiding symptoms (hesitancy, weak stream, intermittency, terminal dribbling, feeling of incomplete emptying) and storage symptoms (frequency, urgency, nocturia) (5), caused either by thickening of the prostate leading to narrowing of the urethra (static component) or increasing smooth muscle tone (dynamic component) (3,(5)(6).Though not a life-threatening condition, BPH may also lead to poor QOL (3).IPSS, developed by the American Urological Association in 1992, is a scoring system that assesses the severity of LUTS and urological QOL.Patients with 0-7 points are categorized as 'mild', 7-19 as 'moderate' and 20-35 as 'severe' symptomatic patients.The treatment for BPH mainly depends on the stage of the disease and range from medical therapy (alpha-adrenergic blockers, 5-alphareductase inhibitors, combination therapy) to minimally invasive and surgical therapies (6).
Pumpkin (Cucurbita maxima L.) belongs to the Cucurbitaceae family and is considered as a popular vegetable all over the world (4).Pumpkin seeds and seed oil are used as common snacks in several cultures.The seed oil has culinary and pharmaceutical applications (7) and is rich in many bioactive components such as fatty acids, essential amino acids, vitamins, phytosterols, tocopherols, squalene and carotenoid pigments (7)(8)(9)(10).Pumpkin seed has many properties such as antioxidant, antiandrogenic and anti-inflammatory.In addition, it has a positive effect on bladder contractility and reduction of prostate gland growth (10).Therefore, the use of pumpkin seeds in the management of patients with LUTS/BPH seems to be effective for improving their symptoms and QOL.
Many of the research findings show that pumpkin seeds are effective in treating patients with BPH.A pilot study on phytotherapy found that an oral combination of pumpkin seed extract, soy germ isoflavonoids and cranberry could be beneficial for BPH patients with mild to moderate LUTS, through a significant decrease in IPSS score and improving QOL (11).Significant improvement of BPH symptoms have also been reported with the use of oil free hydroethanolic pumpkin seed extract (12) and of BPH/LUTS and IPSS related QOL with the use of pumpkin seeds (13).Another study done in Australia with phytotherapy containing pumpkin seed oil reported a significant positive effect on the management of LUTS with significant improvements (14).A randomized clinical trial conducted in Iran using both prazosin and prostafit (contains pumpkin seed oil) also revealed a similar effect of pumpkin, however, the effect was not found to be much effective as prazosin (15).Thus, though many studies have been conducted to justify the use of pumpkin seed to treat BPH, the findings need to be substantiated in a confirmatory study or systematic review (13).
Although a few reviews had been done with regards to pumpkin and BPH, the effect of pumpkin on BPH has not been systematically reviewed comprehensively (10,16).Such an evaluation would enable establishing more precise interpretations on the effectiveness of pumpkin seed extract as a complementary treatment or preventive mode for BPH.Pumpkin can be easily grown and is consumed as a cheap and freely available vegetable in tropical countries.Therefore, the results of this meta-analysis will be beneficial not only for the patients with BPH but also for the economy of such countries through the reduction of expenses on medication.Thus, we conducted a systematic review with potential metaanalysis to establish the effectiveness of pumpkin on BPH.The primary outcome of interest was IPSS reduction while the secondary outcomes were changes of post void residual volume, QOL, prostate volume, PSA level and urinary flow.

Protocol and registration
The registration of the review was done in the International Prospective Register of Systemic Reviews (PROSPERO ID: CRD42020181395).Preferred Reporting Items for Systemic reviews and Meta-analysis (PRISMA) guidelines were referred to when preparing the article.

Eligibility criteria, information sources and study selection
The review question was developed according to PICOS sequence as, "is there an effect of intake of any form of pumpkin on benign prostatic hyperplasia compared to non-intake of it?".The participants included patients already diagnosed with BPH.The intervention included those with consumption of pumpkin or pumpkin containing product.Comparators were those who did not receive any pumpkin or pumpkin containing product or those who received an alternate product.Reduction of LUTS (hesitancy, intermittency, poor stream, terminal dribbling and incomplete emptying) related to BPH was considered as the main outcome and it was assessed using IPSS reduction.Randomized, quasi-experimental and observational studies (cohort, case control and cross sectional analytical) were utilized for the review.
Electronic bibliographic databases of PubMed, EMBASE, The Cochrane Library, CINAHL and AGRICOLA were used to search for the relevant articles (search strategies used attached as supplementary files).Studies published since the inception of the databases up to the date of completion of searches were selected.The search was limited to articles in English or with an English translation.The searches were re-run just before the final analysis and further studies were retrieved for inclusion.
The screening of articles for inclusion criteria were independently done by two reviewers and a third reviewer's opinion was sought when there was disagreement between individual judgments.The study selection was completed in two rounds, with the titles and available abstracts being screened in the first round and the full articles in the second round.For this purpose, three screening questions were used; whether the article was on a human study; whether the intervention was related to pumpkin; and whether prostate related quantitative outcomes were there.

Data extraction process, risk of bias in individual studies
Study design, exposure/intervention, results of outcomes including methodology, results and effect measures were extracted from the articles (Table 1).Data were independently extracted by two reviewers, while a third reviewer was involved in resolving the disagreements between individual judgments.The study investigators were contacted for unreported data or additional details if needed.
The risk of bias of the selected randomized studies was assessed using Grading of Recommendations Assessment Development and Evaluation (GRADE) using the Review Manager software.This assessment was based on seven criteria, namely random sequence generation, allocation concealment, performance bias, detection bias, attrition bias, reporting bias and any other bias.Nonrandomized studies were assessed using Risk of Bias in Non-Randomized Studies of Interventions (ROBINS-I) Criteria (17)(18)(19)(20)(21).

Summary measures and synthesis of results
After assessing the clinical, methodological and statistical heterogeneity (assessed using ᵡ 2 test and I² test), the meta-analysis was performed for one outcome using Review Manager Software version 5.4.The pooling method used was Inverse Variance Method and mean difference was the effect measure.
A narrative review was done for other outcomes found in studies, which were not included in the meta-analysis due to heterogeneity.The quality of evidence was assessed using the Criteria of GRADE Working Group, with the help of GRADEproGDT application.

Selection of studies
Of the 242 articles screened, 91 experimental studies were selected in the first round, out of which 11 studies were excluded in the second round and 67 studies rejected in the third round.Out of the remaining 13 studies, three could not be accessed as full articles, thus only 10 studies (seven randomised and three non-randomised) were selected for the final review (Table 1).Three of them were included in the meta-analysis of the primary outcome of interest (13,22,24).

Study characteristics
As given in Table 2, majority of the participants included in the 10 selected studies were more than 50 years.In eight studies, pumpkinseeds alone was the intervention and in two, it was a combined product containing pumpkin seeds.IPSS reduction was the primary outcome described in nine articles, while other outcomes included QOL in eight studies (seven assessed urological QOL using the IPSS related QOL question and one using the Life Quality Index); prostate volume, PSA, uroflow rate and post-voidal residual volume in five studies; frequency of micturition during day and night in four studies; nocturia in one study; and erectile dysfunction in one study.

Risk of bias within studies
All seven randomized studies were with low risk for reporting bias.Blinding of participants was done only in four studies.High risk of bias was due to issues related to selection and attrition bias in one study each and performance bias in two studies.Information related to the blinding of the outcome assessment was not given in all studies.All three non-randomized studies demonstrated low risk for selection of the reported results.Serious risk was identified only in relation to bias in the classification of intervention.Moderate risk was identified in relation to bias due to confounding and measurement of the outcome in one study each (Figure 2 and Table 3).

Findings of the meta-analysis
Meta-analysis of the three studies on post intervention IPSS value included 650 in the experimental group (pumpkin seed involved in the experiment) and 631 in the control group (pumpkin seed not involved in the experiment).The studies were not significantly heterogeneous (I 2 =0%; p=0.96).The pooled mean difference was -1.41 (95% CI: -1.97, -0.85) with random-effect assumption.(11)(12)(13)(14)(15)(22)(23)(24)(25) described a significant reduction of IPSS in the intervention group.The IPSS median score was observed to be reduced from 15 to 9 within a period of three months (p<0.001) in a one-arm trial (11); by 36% in the active group (p<0.05) in a randomized doubleblinded clinical trial (14); from 15.7 to 10.8 within a period of 12 weeks (p<0.0001)IPSS median score by 36% in the active group (p<0.05(12).In addition, in another randomized clinical trial, the mean IPSS score was 14.5, 11.9 and 9.24 at the baseline, 3 months and 6 months respectively following treatment (15).Further, another placebo controlled study showed a statistically significant difference in five-point reduction of IPSS score between the placebo group (54.2%) and treatment group (64.8%) after 12 months (24).
QOL -Eight studies demonstrated QOL as an outcome.A randomized placebo-controlled study demonstrated an improvement of QOL by 36% in the treatment group compared to 29.2% in the placebo group (13).In another study, the life quality score was reduced by 46.1% (23).Urological QOL was significantly decreased from a median of 4 at baseline to 3 at one month and 2 at 3 months (p<0.001)(11).An improvement in QOL was observed in 69% of the treatment group compared to 31% of the control group (25).
Prostate volume -Although five studies (13,15,(24)(25) demonstrated prostate volume as an outcome, one study showed a significant reduction.Median prostate volume was significantly reduced in the treatment group (p=0.002)compared to no significant reduction in the control group (p=0.438)(24).
Uroflow rate -Out of the five studies (13,15,22,24,35), uroflow rate was significantly increased in three studies (15,22,35).A randomized clinical trial stated an improvement in the mean uroflow rate from 14.5 ml/s to 19.0 ml/s in the treatment group (15).Another study showed a significant improvement by 14.9% following a period of 12 months (22).Furthermore, another study stated a statistically significant improvement in uroflow rate (p<0.001)(35).
Post-voidal residual volume -Out of the five studies with the outcome of post-voidal residual volume (12-13, 24-25, 35), three studies showed a significant reduction.One study showed a significant reduction from 83.67 ml at baseline to 63.11 ml at the end of intervention.(p=0.0394)(12).In addition, another study demonstrated a significant improvement in residual volume (p=0.01)(35).
Erectile dysfunction -One study (11) revealed a positive effect on the International Index of Erectile Dysfunction Score.The median score was improved from 15 at baseline to 17 after 3 months (p<0.001)(12).Out of the five studies with the outcome of PSA (12-13, 15, 22, 24), none showed a significant difference.

Discussion
This was the first comprehensive systematic review which included a meta-analysis component on the effectiveness of pumpkin on BPH.The results revealed that pumpkin seeds contribute to the reduction of symptoms of BPH.
Several steps were taken to ensure the quality of this systematic review and meta-analysis.Firstly, the data search was done without restricting to a time period.It is mentioned that searching all available data is a better strategy in systematic reviews (27).Risk of bias estimates were not done by averaging the several components of the bias estimates but by utilizing recommended guidelines (17,(28)(29).Blinding was not done in one study and there was a high risk for performance bias in two studies.Strict categorizing criteria were adhered to in the estimation of bias.It is recommended to consider the statistical, clinical and methodological heterogeneity in interpreting the quality of meta-analysis (30)(31)(32).The statistical heterogeneity of the studies was seemingly acceptable.Measures to minimize methodological bias were adhered to when selecting the studies and expressing the risk of bias of each study.Furthermore, the control groups of the studies had been recruited from similar settings.
The primary outcome of this study was IPSS reduction with a pooled mean difference of -1.41 for post intervention IPSS value between intervention group and control group.Out of nine studies where the IPSS reduction was described, all showed a significant reduction in IPSS in the intervention group.This is comparable with the study done by Leibbrand et al. (2019) where a reduction of IPSS of 30.1% was seen between the baseline and end of intervention and in the placebo group, the effect observed was a reduction from IPSS 17.7 to 12.2 (12).The findings are in line with other investigations with pumpkin seed extract.In the patient group treated with pumpkin seed extract, there was a continuous improvement in IPSS score over the entire treatment period, while there were no significant changes in the placebo group after the 6th month of treatment.All laboratory and vital signs showed no negative changes over the entire study period.In addition, no significant adverse events restricting the tolerance were found (24).Vahlensieck et el.( 2015) have revealed that treatment with pumpkin seed results in a substantial improvement in LUTS related to BPH (13).
Overproduction of 5-α-reductase followed by a hormonal imbalance between androgens and oestrogens leads to symptoms related to IPSS such as frequent micturition and urinary incontinence (17).In vitro as well as in vivo studies have shown that pumpkin seed has an inhibitory effect on inhibition of 5-α reductase and aromatase enzymes leading to inhibition of prostate tissue growth (12,33).Of the LUTS symptoms, irritative symptoms have the most profound effects on the patient's QOL.Of these, the reduction in nocturia would have the most positive effect (34).None of the studies revealed a positive effect on PSA level.Uroflow rate was increased in three studies postulated by reduction in size of the prostate gland.Post-voidal and prostate volume were improved in one study each.Even though the alpha blockers and 5-alpha reductase inhibitors were given as treatment for BPH, side effects such as erectile dysfunction and orthostatic hypotension were observed.However, pumpkin seeds did not show significant side effects according to most of the studies, while one study revealed a positive effect on the International Index of Erectile Function score (11,13).
Though pumpkin is a commonly found cheaper vegetable in Sri Lanka, pumpkin seeds are not utilized or commercially available.By promoting production and consumption of pumpkin seeds will not only have the health benefits described above but also indirectly aid on country's economy.This systematic review and the meta-analysis act as a foundation for the development of health policies.Utilization of pumpkin seed extracts can be promoted hand in hand with the clinical treatments.As a result, it may contribute to achieve a significant improvement in the quality of life in BPH patients through reduction of symptoms.Furthermore, commercially prepared pumpkin seeds can be promoted to use in the regular diet.In addition, new health interventions with pumpkin seed extracts can be encouraged with this review.This systematic review has several limitations.The studies included showed substantial heterogeneity with respect to the analysed population size, followup length and baseline age.Due to the low number of studies available and the heterogeneity, metaanalysis was performed only for post intervention IPSS value and three studies were used for the metaanalysis.In line with this, it is also not possible to assess publication bias and small study effects.

Conclusions & Recommendations
The review suggests that pumpkin may be beneficial for the reduction of IPSS, and to some extent for the improvement of QOL, the reduction of prostate volume, residual volume and uroflow rate in relation to the control of BPH.Further studies are recommended to strengthen the hypothesis.