SEARCH

SEARCH BY CITATION

Keywords:

  • benign prostatic hyperplasia;
  • community-based study;
  • incidence;
  • Japanese male;
  • longitudinal study

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgment
  8. Conflict of interest
  9. References

Objectives

To determine the incidence of and the risk factors for treatment in Japanese men with benign prostatic hyperplasia/lower urinary tract symptoms enrolled into a longitudinal community-based study.

Methods

A total of 267 of 319 men aged 40–79 years were eligible for this study, with nearly 15 years of follow up. Their medical records were checked to look for any medical treatment for benign prostatic hyperplasia/lower urinary tract symptoms. The risk of treatment for benign prostatic hyperplasia/lower urinary tract symptoms was determined by calculating the hazard ratio using the Cox proportional hazards model. Five baseline parameters were considered: the International Prostate Symptom Score, the peak urinary flow rate, the prostate volume, the serum prostate-specific antigen and the internal prostatic architecture on transrectal ultrasonography.

Results

Data were successfully collected for 171 men (64%; 121 survivors and 50 deceased). During approximately 1900 person-years of follow up, the overall incidence of treatment for benign prostatic hyperplasia/lower urinary tract symptoms was 15.4/1000 person-years. All five parameters were statistically significant predictors of future treatment for benign prostatic hyperplasia/lower urinary tract symptoms: International Prostate Symptom Score greater than 7 (hazard ratio 6.2, P < 0.001), prostate volume greater than 30 mL (hazard ratio 4.3, P = 0.002), peak urinary flow rate less than 12 mL/s (hazard ratio 4.4, P < 0.001), prostate-specific antigen greater than 1.4 ng/mL (hazard ratio 4.0, P < 0.001) and internal prostatic architecture group 3 (hazard ratio 3.2, P = 0.002).

Conclusions

Severity of lower urinary tract symptoms, decreased peak urinary flow rate, enlarged prostate volume, high prostate-specific antigen value and internal prostatic architecture at baseline are independent risk factors for treatment in Japanese men presenting with benign prostatic hyperplasia/lower urinary tract symptoms.


Abbreviations & Acronyms
BPH/LUTS

benign prostatic hyperplasia

HR

hazard ratio

IPA

internal prostatic architecture

IPSS

International Prostate Symptom Score

IQR

interquartile range

LUTS

lower urinary tract symptoms

PSA

prostate-specific antigen

PV

prostate volume

Qmax

peak urinary flow rate

TURP

transurethral resection of the prostate

TZ

transition zone

Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgment
  8. Conflict of interest
  9. References

Men are often annoyed by LUTS when they reach an age around the 50s. The prostates in older men enlarge with advancing age. A randomized control trial associated with BPH and LUTS showed that an enlarged prostate was a risk factor for symptomatic progression, acute urinary retention and BPH surgery.[1-4] However, it is difficult to perfectly understand the disease, because there is no clear definition of BPH and insufficient data are available on the natural history of PV and LUTS in elderly men. There have been some reports on the natural history of BPH/LUTS, such as the Olmsted County study[5-7] and Krimpen study.[8, 9] The previous data showed that there was variability in PV among different races.[10] Furthermore, LUTS were also variable among men in different countries because of the differences in their linguistic or ethnic backgrounds.[11, 12] Therefore, as we need to determine the natural history of BPH/LUTS of Japanese men, we reported the natural history of BPH/LUTS in them based on cross-sectional[10, 11] and longitudinal[13-15] community-based studies that were carried out in Shimamaki-mura in Hokkaido, Japan.

It is not easy to predict how much treatment for BPH/LUTS men will require in the future. If we can calculate the probability of the risk using the data at baseline, the information will be useful in clinical practice. In the USA, the risk of treatment for BPH/LUTS was analyzed using a community-based cohort in Olmsted County.[16] It was concluded that men with moderate to severe symptoms, an impaired urinary flow rate, enlarged prostate and high level of PSA were more likely to undergo treatment. In the current study, we investigated the incidence and risk of any treatment or surgery for BPH/LUTS in Japanese men using the same method, and compared their risk with that of USA men.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgment
  8. Conflict of interest
  9. References

From October 1992 to May 1993, we carried out a cross-sectional community-based study to determine the prevalence of BPH/LUTS in men aged 40–79 years in Shimamaki-mura in Hokkaido, Japan.[10, 11] From February 2007 to February 2008, we carried out a longitudinal survey using the same methods as in the initial survey. A part of the results of the natural history of BPH/LUTS was previously reported.[14, 15] In the current study, we investigated the medical histories during follow up of the participants in the initial study. We collected this information by interviewing survivors and checking medical charts from clinics and hospitals they visited with their consent. When initial participants had already died, we asked their families about their medical histories, and checked their medical charts after receiving consent. The present study was approved by the Sapporo Medical University Ethics Committee in February 2007 (#18-9).

Participants

In our previous study, 319 of the 682 men in the community-based population (47%) participated.[11] Of the 319 men who did not have any medical treatment for BPH/LUTS at baseline or any history of disease at baseline that affected LUTS such as back surgery, apoplexy, prostate cancer, transurethral resection of bladder cancer, or neurogenic bladder, and completed all examinations in the initial study, 267 were eligible to participate in the current study.

Of these 267 men, 156 were still alive, 84 had died and 27 had moved away during the period of approximately 15 years. Finally, we collected data for 121 survivors and 50 deceased men (58%).

Measurements

In the initial study, we investigated the IPSS, Qmax, PSA value, PV and IPA. We also collected information about histories of diabetes mellitus and hypertension. Qmax was measured using a Urodyn 1000 System (Mediwatch, Rugby, UK). All patients voided while standing. If voided volume was less than 150 mL, we requested them to retry. If voided volume did not reach 150 mL despite repeated attempts, we used the flow rate from the maximal voided volume as Qmax. Serum PSA was measured using an IMX PSA kit (Abbott Laboratories, Abbott Park, IL, USA). PV was measured by a single examiner (NM) using transrectal ultrasonography and was estimated by the formula reported previously.[10, 17] We classified prostates into three groups according to the IPA on transrectal ultrasonography: group 1, an invisible TZ; group 2, a visible TZ with an unclear border; and group 3, a visible TZ with a clear border.[17] We previously reported this classification and verified the correlation between the IPA and future prostatic growth.[14]

Statistical analysis

We divided participants into three groups according to their treatment during follow up: no treatment, medication and TURP. Men who had medication therapy with an alpha-blocker and/or anti-androgen were included in the medication group. We did not define phytotherapy or Chinese herbal therapy as a treatment for BPH/LUTS and categorized them into the no treatment group. The TURP group also included men who underwent TURP after medication therapy. Differences in age, IPSS, Qmax, PSA and PV at baseline among the three groups were analyzed by the Kruskal–Wallis test. If there was any significance, the Mann–Whitney U-test with Bonferroni's correction was used as a post-hoc test for the Kruskal–Wallis test.

We calculated the incidences of treatment for BPH/LUTS according to age decade and demonstrated them by the person-year method. The cumulative incidence of treatment for BPH/LUTS according to age decade was estimated by the Kaplan–Meier method. The difference in the time between the initial study and TURP was analyzed using the Mann–Whitney U-test.

The risk of treatment for BPH/LUTS was calculated using the Cox proportional hazards model. If any treatment, such as medication or TURP for BPH/LUTS, occurred during the follow up as a first event, it was regarded as an uncensored event (end-point) when analyzing the risk of any treatment. In contrast, only TURP was regarded as an uncensored event when we calculated the risk of TURP. Antecedent events, such as death and the occurrence of other diseases that affected LUTS, were regarded as censored. We carried out this analysis using two models: one included age, IPSS, Qmax, PV, PSA and IPA as risk factors, and the other added histories of diabetes mellitus and hypertension to the six aforementioned factors.

We used the same cut-off points for baseline variables as used in the Olmsted County study.[16] We divided the participants into four categories according to age decade. Other variables were divided into two groups, IPSS (≥8 vs <8), Qmax (<12 mL/s vs ≥12 mL/s), PV (≥30 mL vs <30 mL), PSA (≥1.4 ng/mL vs <1.4 ng/mL) and IPA (group 3 vs groups 1 and 2).

We made three multivariate models for adjusting risk factors to calculate HR of the risk of treatment for BPH/LUTS. Because PV, PSA and IPA were correlated with each other (data not shown),[17, 18] each model used one of three factors that were related to PV.

P < 0.05 was regarded as statistically significant. Statistical analyses were carried out using SPSS 15.0 (SPSS, Chicago, IL, USA).

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgment
  8. Conflict of interest
  9. References

During 15-years follow up of the 171 men, 29 had treatments for BPH/LUTS (Fig. 1a), 86 had no event, 36 died without treatment for BPH/LUTS and 20 developed other diseases that were likely to affect their LUTS, including apoplexy (8), prostate cancer (4), back surgery (3), bladder cancer (3), hemodialysis (1) and neurogenic bladder (1). Of the 29 having treatment for BPH/LUTS, 16 required TURP (Fig. 1b). Of these 16 men, 10 directly underwent TURP as the first treatment and the remaining six had medication as a prior therapy. The median times between the initial study and TURP were 11.1, 12.2, 2.3 and 2.3 years for men in their 40s, 50s, 60s and 70s at baseline, respectively. The median time to TURP in men in their 40s and 50s was significantly longer than that for those in their 60s and 70s (P = 0.007).

figure

Figure 1. Cumulative incidences of treatment for BPH/LUTS during 15 years according to age decade (a) any treatment and (b) TURP. (image), 40s; (image), 50s; (image), 60s; (image), 70s.

Download figure to PowerPoint

Table 1 presents the baseline characteristics of the three groups according to the treatment results. Although no significant difference in age was observed among the three groups (P = 0.050), there were significant differences in the IPSS (P < 0.001), Qmax (P < 0.001), PSA (P = 0.007) and PV (P = 0.002). Post-hoc analysis showed that the men who had TURP had a significantly greater IPSS, PSA and PV, and lower Qmax than those without any treatment. However, the differences in these four parameters between the groups with no treatment and medication, or medication and TURP were not significant.

Table 1. Participants' characteristics at baseline according to follow-up treatment
 Total (n = 171)No treatment (n = 142)Medication (n = 13)TURP (n = 16)
Age (years)    
Median (IQR) 60 (51, 68)64 (62, 71)65.5 (58, 69.5)
No. participants (%)    
40–493331 (94)0 (0)2 (6)
50–594135 (85)2 (5)4 (10)
60–696148 (79)7 (12)6 (10)
70–793628 (78)4 (11)4 (11)
IPSS    
Median (IQR) 7 (4, 9)9 (6, 17)11.5 (9, 17)
No. participants (%)    
0–78681 (94)4 (5)1 (1)
8–358561 (72)9 (11)15 (18)
Qmax (mL/s)    
Median (IQR) 18.4 (12.6, 23.8)11.6 (9.4, 20.5)10.4 (9.0, 14.0)
No. participants (%)    
≥12120108 (90)6 (5)6 (5)
<125134 (67)7 (14)10 (20)
PSA (ng/mL)    
Median (IQR) 0.8 (0.5, 1.3)1.0 (0.5, 1.7)2.4 (0.9, 3.6)
No. participants (%)    
≤1.3122108 (89)8 (7)6 (5)
≥1.44934 (69)5 (10)10 (20)
PV (mL)    
Median (IQR) 17.2 (15.1, 21.6)17.5 (15.9, 21.5)22.9 (20.5, 31.0)
No. participants (%)    
<30157134 (85)12 (8)11 (7)
≥30148 (57)1 (7)5 (36)
IPA (group)    
No. participants (%)    
1 and 2116103 (89)9 (8)4 (3)
35539 (71)4 (7)12 (22)

Of the 171 men, 46 (27%) and 12 (7%) had histories of hypertension and diabetes mellitus, respectively.

Incidence of treatment for BPH/LUTS

During 1887 person-years, 29 men had treatment for BPH/LUTS, with an overall incidence of 15.4/1000 person-years. The incidences in men aged in their 40s, 50s, 60s and 70s were 4.4, 12.0, 20.3 and 27.1/1000 person-years, respectively. In contrast, 16 men had TURP during 1949 person-years. The overall incidence rate of men who had TURP was 8.2/1000 person-years. The incidence of men who had TURP also increased with age, being 4.4, 7.6, 9.1 and 12.9/1000 person-years for men aged in their 40s, 50s, 60s and 70s, respectively.

Risk of treatment for BPH/LUTS in the future

We separately calculated the risk of any treatment for BPH/LUTS and that of TURP with the Cox proportional hazards model (Tables 2, 3). In univariate analysis for the risk of any treatment, men with an IPSS of 8 or greater, PV of 30 mL or greater, PSA of 1.4 ng/mL or greater, Qmax of less than 12 mL/s or IPA group 3 had significantly higher risks (approximately three- to sixfold) than those without these factors. In the multivariate analysis, all of these five factors were still statistically significant and the HR ranged from 2.4 to 5.0. In addition, men aged in their 60s and 70s had significantly higher risks than those aged in their 40s in the univariate model. However, the significant association between age decade and the risk of any treatment was eliminated when adjusting for all factors. The same results were obtained with regard to the risk for TURP, except for age decade, which was not associated with the risk even in univariate analysis. In multivariate analysis, HR of the five factors, except for age decade, ranged from 3.2 to 13.7. Among all the parameters, the HR for the IPSS had the highest risk for any treatment or TURP.

Table 2. The risk of any treatment for BPH/LUTS in the future according to baseline measurements
 nUnivariate analysisMultivariate analysis including PVMultivariate analysis including PSAMultivariate analysis including IPA
HR (95% CI)P-valueHR (95% CI)P-valueHR (95% CI)P-valueHR (95% CI)P-value
Age (years)40–49331.0 1.0 1.0 1.0 
50–59412.8 (0.6–13.9)0.2073.6 (0.7–17.8)0.1213.5 (0.7–17.7)0.1233.4 (0.7–16.8)0.138
60–69614.7 (1.1–21.0)0.0413.0 (0.6–14.2)0.1673.3 (0.7–14.8)0.1242.9 (0.6–13.6)0.187
70–79366.5 (1.4–31.0)0.0184.2 (0.9–20.4)0.0733.6 (0.7–17.7)0.1112.5 (0.5–13.4)0.287
IPSS<8861.0 1.0 1.0 1.0 
≥8856.2 (2.4–16.2)<0.0014.8 (1.8–13.0)0.0024.7 (1.8–12.6)0.0025.0 (1.9–13.6)0.001
Qmax (mL/s)≥121201.0 1.0 1.0 1.0 
<12514.4 (2.1–9.2)<0.0012.5 (1.1–5.5)0.0272.5 (1.1–5.4)0.0212.4 (1.1–5.5)0.032
PV (mL)<301571.0 1.0     
≥30144.3 (1.7–10.5)0.0022.8 (1.1–7.5)0.034    
PSA (ng/mL)<1.41221.0   1.0   
≥1.4494.0 (1.9–8.4)<0.001  2.7 (1.3–5.8)0.008  
IPA1 or 21161.0     1.0 
3553.2 (1.5–6.6)0.002    2.5 (1.1–5.8)0.031
Table 3. The risk of TURP for BPH/LUTS in the future according to baseline measurements
 nUnivariate analysisMultivariate analysis including PVMultivariate analysis including PSAMultivariate analysis including IPA
HR (95% CI)P-valueHR (95% CI)P-valueHR (95% CI)P-valueHR (95% CI)P-value
Age (years)40–49331.0 1.0 1.0 1.0 
50–59411.8 (0.3–9.6)0.5132.1 (0.4–11.4)0.4001.9 (0.3–10.8)0.4491.3 (0.2–7.7)0.741
60–69612.0 (0.4–10.1)0.3850.7 (0.1–4.6)0.7481.2 (0.2–6.4)0.7890.4 (0.1–3.1)0.418
70–79362.9 (0.5–15.9)0.2281.4 (0.2–8.1)0.7101.4 (0.2–7.9)0.7200.3 (0.0–2.7)0.313
IPSS<8861.0 1.0 1.0 1.0 
≥88517.9 (2.4–135.4)0.00512.9 (1.7–100.3)0.01511.8 (1.5–90.8)0.01813.7 (1.8–106.2)0.012
Qmax (mL/s)≥121201.0 1.0 1.0 1.0 
<12514.9 (1.8–13.5)0.0023.2 (1.1–9.7)0.0373.2 (1.1–9.2)0.0354.1 (1.1–14.6)0.032
PV (mL)<301571.0 1.0     
≥30146.8 (2.4–19.6)<0.0017.5 (2.1–26.4)0.002    
PSA (ng/mL)<1.41221.0   1.0   
≥1.4495.7 (2.0–15.6)0.001  4.1 (1.5–11.6)0.007  
IPA1 or 21161.0     1.0 
3557.2 (2.3–22.5)0.001    9.9 (2.5–38.3)0.001

Although we calculated the risk of any treatment for BPH with another model that additionally included histories of diabetes mellitus and hypertension as risk factors, these two factors were not significant (data not shown).

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgment
  8. Conflict of interest
  9. References

Jacobsen et al. reported that the incidences of any treatment for BPH/LUTS were 3.3, 11.7, 29.3 and 30.2/1000 person-years in population-based USA men in their 40s, 50s, 60s and 70s, respectively. The incidences of TURP were 0.3, 1.4, 8.6 and 10.9/1000 person-years in men aged in their 40s, 50s, 60s and 70s, respectively.[16] The incidence of any treatment for BPH/LUTS in the Japanese men was similar to that in the USA men. However, the incidences of TURP in men aged in their 40s and 50s were relatively higher in the Japanese men than in the USA men. We thought that the follow-up period was associated with this result. Although it was 6 years in the USA cohort, it was approximately 14 years in our cohort. The cumulative incidence (Fig. 1b) showed that TURP events in men aged in their 40s and 50s occurred late during follow up. Therefore, it seemed that the TURP incidence in Japanese men was high because of the long follow-up period.

We previously reported the prediction of treatment for BPH in the future in which we analyzed just 107 men who participated in both the initial and follow-up studies.[19] In that report, we concluded that the IPSS was a significant predictor of any treatment for BPH/LUTS in the future. In addition, not only the IPSS, but also PV and Qmax as objective values were important to decide whether men who complained of LUTS underwent TURP. The current study including men who only participated in the initial study or had already died showed similar results derived from the Cox proportional hazards models. An IPSS of 8 or greater at baseline was the most important factor for any treatment for both BPH/LUTS and TURP in the future. Although the HR of PV for any treatment was approximately 3, which was similar to Qmax, the HR of PV for TURP was higher than that of Qmax. The same tendency was seen in the HR of the IPA and PSA. Thus, we found that the parameters associated with prostate size at baseline were related to BPH surgery in the future.

In our longitudinal community-based study on BPH/LUTS, we verified our hypothesis in the initial study; that is, “the IPA on transrectal ultrasonography predicts future prostatic growth”.[14, 17] Therefore, we also included the IPA in this analysis as a risk factor of treatment for BPH/LUTS in the future and verified again that it was applicable. The HR of IPA group 3 was equivalent to that of PV greater than 30 mL. However, it seemed that IPA group 3 was practically superior to PV greater than 30 mL as a predictor for BPH surgery in the future, because there were a small number of men with PV greater than 30 mL in our cohort. PV is, of course, an important factor in BPH/LUTS progression,[1, 4, 20, 21] but there is a possibility that the IPA can detect men who will require BPH surgery in the future despite having a PV of less than 30 mL.

We compared the present results with those of the Olmsted County study.[16] In both Japanese and USA men, the HR of treatment for an IPSS of 8 or greater, Qmax of less than 12 mL/s, PV greater than 30 mL or PSA greater than 1.4 ng/mL were all significantly higher than those in referents. Although Tsukamoto et al. reported that Japanese men had lower symptom bother scores in comparison with their complaint of LUTS,[11] it was interesting that the most important risk factor of any treatment for BPH/LUTS was the IPSS in men in both countries with almost same the HR. In multivariate analysis without PV or PSA data in the Olmsted County study, age was the strongest risk factor for TURP in the future. Contrary to their result, age was not a significant risk factor for TURP in Japanese men. This difference was likely associated with the relatively higher incidences of TURP in men aged in their 40s and 50s at baseline.

There were several limitations in the current study. First, the number of participants was small. Of the 319 initial participants, just 171 (54%) were included, and 50 (29%) were already deceased. However, we thought that we could collect the follow-up data precisely because the number of clinics or hospitals the participants visited was relatively limited. Second, there was no definitive indication for any treatment for BPH/LUTS or for TURP. Thus, it seemed that there was high variability in the IPSS, Qmax and PV in men who had any treatment for BPH/LUTS. Finally, it was unclear whether the cut-off points of the parameters that we used in the current study were appropriate for Japanese men. In particular, it seemed that the cut-off point of 30 mL for PV was unbalanced, because there were few men who had a PV greater than 30 mL in the general population. Furthermore, Gupta et al. suggested that the prostates of Japanese men might produce and release more PSA per unit than those of Caucasian men.[22] Although a PV of 40 mL is consistent with PSA 1.4 ng/mL in those men, a PV of 40 mL in Japanese men corresponds with PSA 5.2 ng/mL. Therefore, it is likely that the ideal PSA cut-off for the risk of BPH treatment in the future might be greater than 1.4 ng/mL. However, we had to use the same cut-off point to compare the results from the two countries.

In conclusion, in a 15-year-longitudinal community-based study of Japanese men, impaired LUTS, decreased Qmax, enlarged PV, a high PSA value and IPA group 3 at baseline were independent risk factors for any treatment or TURP for BPH/LUTS in the future.

Acknowledgment

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgment
  8. Conflict of interest
  9. References

This study was partly supported by grants-in-aid from the Japan Society for the Promotion of Science (19591862). The sponsor of the study had no role in the study design, collection, analysis and interpretation of data. We thank Koji Yamada and Yuichi Ogawa from the Welfare Division of Shimamaki-mura who helped us to organize prostate examinations there. We thank the urologists Toshihiro Maeda, Masatoshi Muto, Manabu Igarashi and Shintaro Miyamoto from Sapporo Medical University who assisted us in implementing the survey.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgment
  8. Conflict of interest
  9. References
  • 1
    Roehrborn CG, McConnell JD, Lieber M et al. Serum prostate-specific antigen concentration is a powerful predictor of acute urinary retention and need for surgery in men with clinical benign prostatic hyperplasia. PLESS Study Group. Urology 1999; 53: 473480.
  • 2
    McConnell JD, Roehrborn CG, Bautista OM et al. The long-term effect of doxazosin, finasteride, and combination therapy on the clinical progression of benign prostatic hyperplasia. N. Engl. J. Med. 2003; 349: 23872398.
  • 3
    Roehrborn CG, Siami P, Barkin J et al. The effects of combination therapy with dutasteride and tamsulosin on clinical outcomes in men with symptomatic benign prostatic hyperplasia: 4-year results from the CombAT study. Eur. Urol. 2010; 57: 123131.
  • 4
    Roehrborn CG, Barkin J, Siami P et al. Clinical outcomes after combined therapy with dutasteride plus tamsulosin or either monotherapy in men with benign prostatic hyperplasia (BPH) by baseline characteristics: 4-year results from the randomized, double-blind Combination of Avodart and Tamsulosin (CombAT) trial. BJU Int. 2011; 107: 946954.
  • 5
    Chute CG, Panser LA, Girman CJ et al. The prevalence of prostatism: a population-based survey of urinary symptoms. J. Urol. 1993; 150: 8589.
  • 6
    Sarma AV, Jacobsen SJ, Girman CJ et al. Concomitant longitudinal changes in frequency of and bother from lower urinary tract symptoms in community dwelling men. J. Urol. 2002; 168: 14461452.
  • 7
    Lieber MM, Rhodes T, Jacobson DJ et al. Natural history of benign prostatic enlargement: long-term longitudinal population-based study of prostate volume doubling times. BJU Int. 2009; 105: 214219.
  • 8
    Bosch JL, Tilling K, Bohnen AM, Bangma CH, Donovan JL. Establishing normal reference ranges for prostate volume change with age in the population-based Krimpen-study: prediction of future prostate volume in individual men. Prostate 2007; 67: 18161824.
  • 9
    Bosch JL, Bangma CH, Groeneveld FP, Bohnen AM. The long-term relationship between a real change in prostate volume and a significant change in lower urinary tract symptom severity in population-based men: the Krimpen study. Eur. Urol. 2008; 53: 819825; discussion 825–7.
  • 10
    Masumori N, Tsukamoto T, Kumamoto Y et al. Japanese men have smaller prostate volumes but comparable urinary flow rates relative to American men: results of community based studies in 2 countries. J. Urol. 1996; 155: 13241327.
  • 11
    Tsukamoto T, Kumamoto Y, Masumori N et al. Prevalence of prostatism in Japanese men in a community-based study with comparison to a similar American study. J. Urol. 1995; 154: 391395.
  • 12
    Sagnier PP, Girman CJ, Garraway M et al. International comparison of the community prevalence of symptoms of prostatism in four countries. Eur. Urol. 1996; 29: 1520.
  • 13
    Masumori N, Tsukamoto T, Rhodes T, Girman CJ. Natural history of lower urinary tract symptoms in men – result of a longitudinal community-based study in Japan. Urology 2003; 61: 956960.
  • 14
    Fukuta F, Masumori N, Mori M, Tsukamoto T. Internal prostatic architecture on transrectal ultrasonography predicts future prostatic growth: natural history of prostatic hyperplasia in a 15-year longitudinal community-based study. Prostate 2011; 71: 597603.
  • 15
    Fukuta F, Masumori N, Mori M, Tsukamoto T. Natural history of lower urinary tract symptoms in Japanese men from a 15-year longitudinal community-based study. BJU Int. 2012; 110: 10231029.
  • 16
    Jacobsen SJ, Jacobson DJ, Girman CJ et al. Treatment for benign prostatic hyperplasia among community dwelling men: the Olmsted County study of urinary symptoms and health status. J. Urol. 1999; 162: 13011306.
  • 17
    Masumori N, Tsukamoto T, Kumamoto Y et al. Age-related differences in internal prostatic architecture on transrectal ultrasonography: results of a community based survey in Japan. J. Urol. 1997; 157: 17181722.
  • 18
    Oesterling JE, Kumamoto Y, Tsukamoto T et al. Serum prostate-specific antigen in a community-based population of healthy Japanese men: lower values than for similarly aged white men. Br. J. Urol. 1995; 75: 347353.
  • 19
    Fukuta F, Masumori N, Mori M, Tsukamoto T. Prediction of treatment for benign prostatic hyperplasia in the future: results from a 15-year longitudinal community-based study in Shimamaki-mura, Hokkaido. Nihon Hainyokino Gakkaishi 2009; 20: 312316. (In Japanese.)
  • 20
    Roehrborn CG, McConnell J, Bonilla J et al. Serum prostate specific antigen is a strong predictor of future prostate growth in men with benign prostatic hyperplasia. PROSCAR long-term efficacy and safety study. J. Urol. 2000; 163: 1320.
  • 21
    Crawford ED, Wilson SS, McConnell JD et al. Baseline factors as predictors of clinical progression of benign prostatic hyperplasia in men treated with placebo. J. Urol. 2006; 175: 14221426.
  • 22
    Gupta A, Aragaki C, Gotoh M et al. Relationship between prostate specific antigen and indexes of prostate volume in Japanese men. J. Urol. 2005; 173: 503506.