A pilot study evaluating a new questionnaire for prostatic symptom scoring, the SPSS, and its sensitivity as constructed to objective measures of outflow obstruction

Authors


Kosaku Yasuda md, Koshigaya Hospital, Dokkyo University School of Medicine, 2-1-50 Minami-Koshigaya, Koshigay-shi, Saitama 343–8555, Japan. Email: chiaki-n@dokkyomed.ac.jp

Abstract

Aim: To evaluate the extent to which our newly developed questionnaire, the Saitama Prostate Symptom Score (SPSS), for prostatic symptom scoring reflects objective findings in benign prostatic hyperplasia (clinical BPH) and to compare it with the International Prostate Symptom Score (IPSS) with regard to diagnostic sensitivity in clinical BPH.

Methods: In this study, both the SPSS and the IPSS were self-administered by patients. Free uroflowmetry, a pressure-flow study and the measurement of prostatic volume were carried out.

Results: There was no significant correlation between the results of the IPSS questionnaire and the urethral obstruction grade estimated by Schaefer or Abrams-Griffiths nomograms. The total score of the SPSS was correlated with these nomograms (P = 0.0487 and P = 0.0413, respectively). There was no significant correlation between the results of the IPSS questionnaire and the total volume or transition zone volume of the prostate, whereas the total score of the SPSS correlated with the total volume of the gland and transition zone volume (P = 0.0044 and P= 0.0051, respectively).

Conclusion: This study revealed the SPSS to correlate with objective findings satisfactorily. However, there are still several aspects of the SPSS which need to be improved upon, and the questionnaire should be studied in larger numbers of patients suffering from lower urinary tract symptoms.

Introduction

Clinical benign prostatic hyperplasia (clinical BPH) is a condition where the prostatic adenomatous tissue in the transition zone compresses the urethra (benign prostatic enlargement, BPE), resulting in an obstruction to the bladder outlet (benign prostatic obstruction, BPO), which causes various lower urinary tract symptoms (LUTS). Benign prostatic obstruction is found in 66–75% of patients with clinical BPH, suggesting that BPO is not responsible for all of the symptoms of clinical BPH.1 There must be multiple factors in the pathogenesis of LUTS. At least five causative factors contributing to LUTS need to be considered: BPO, impaired detrusor contractility, detrusor overactivity, sensory urgency and BPE.2–7

In order to quantify the severity of clinical BPH, several prostatic symptom scores have been developed.8–10 These include the American Urological Association prostate symptom score,11 which was renamed the International Prostatic Symptom Score (IPSS) and adopted as an international standard at the 1st International Consultation on Benign Prostatic Hyperplasia (Paris, June, 1991). The IPSS has been found useful in assessing therapeutic outcomes in BPH. However, in Europe, the International Continence Society ‘Benign Prostatic Hyperplasia’ (ICS-BPH) study group conducted a large-scale investigation in response to the perceived underutilization of urodynamics in the diagnosis of patients presenting with clinical BPH and they emphasized the importance of the diagnosis of BPO.12 As a result, it became clear that there was little or no correlation between a wide range of LUTS derived from the International Continence Society test bed questionnaire which was used and the results of free uroflowmetry and pressure-flow studies. Furthermore, there was considerable doubt as to the feasibility of developing a diagnostic symptom score, particularly bearing in mind that with increasing age, a wide variety of lower urinary tract dysfunctions develop.12,13

According to the report of ICS-BPH study group, the IPSS is useful for judging the severity of BPH and evaluating therapeutic outcomes compared to base line in individual subjects. It is not, however, a diagnostic tool for BPH because there is a poor or no correlation with BPO or BPE.13 Thus, we have devised a BPH scoring tool which, we believe, may well allow the diagnosis of BPO and/or BPE with more sensitivity and specificity than the IPSS. We also believe that this is more likely to be user-friendly, particularly for non-specialist practitioners, such as primary care physicians.

Our prostatic symptom score, which we have named the Saitama Prostatic Symptom Score (SPSS), consists of 10 indices, including symptoms that have been reported to be the most frequent in the ICS-BPH report13 and which are urodynamically related to BPO, such as terminal dribbling and hesitancy.14,15 In the present preliminary pilot study, the SPSS and the IPSS were compared with regard to their sensitivity to diagnose BPO and BPE.

Patients and methods

Patients

The present study enrolled a total of 59 men who, on the basis of uroflowmetry, were believed to have BPO. None of them had complications of BPO, including urinary infection, bladder stones, acute or chronic retention or renal impairment. The age range of the group was 51–80 years (mean 67 years). They were selected from a group who were aged over 50 years and diagnosed to have BPE of 20 mL or greater with an adenoma in the transition zone as estimated by transrectal ultrasonography. None of the patients had undergone prostate surgery, had prostate cancer or other conditions interfering with voiding. Informed consent was obtained in written form from all participants prior to entry in the study.

Methods

Examination at first visit

Urinalysis, digital rectal examination, the measurement of prostate volume, including total volume (TV) of the gland and transition zone volume (TZ) by transrectal ultrasonography (TRUS) were all performed. Free uroflowmetry (UFM) tests were also performed, including voided volume (VV), maximum flow rate (Qmax), average flow rate (Qave), residual urine (RU) and residual urine ratio (RU-ratio). Uroflowmetry was repeated at the second visit. The SPSS and the IPSS questionnaires were completed by the patients themselves.

Urodynamic study

A pressure-flow study was performed in patients with LUTS thought to be due to BPE, based on the initial assessment at their first visit.

A 14 Fr catheter and a 4.6 Fr catheter were inserted into the bladder through the urethra, the former for the instillation of physiological saline (at 50 mL/min) and the latter for cystometry to examine the intravesical pressure during storage. The 4.6 Fr catheter was left in situ for the measurement of vesical pressure during voiding.

Patients with obstructions graded III or more according to the Schäfer nomogram and defined according to the Abrams-Griffiths (A-G) nomogram were considered to have BPO.

Statistics

The Mann–Whitney U-test was used for two-group comparisons and the Kruskal–Wallis test for three-group comparisons. Spearman rank correlation coefficients were calculated. P-values less than 0.05 were considered significant.

Scoring

The intensity of each symptom was self-graded by individual patients on a four-point scale (0–3) in the SPSS (Appendix I).

Results

IPSS vs SPSS comparison

There was no significant correlation between the results of the IPSS questionnaire and TRUS TV or TZ, whereas the sum of the SPSS scores correlated with TRUS TV and TZ (P = 0.0044 and 0.0051, respectively). The SPSS voiding score correlated with TRUS TV and TZ (P = 0.0104 and 0.0155, respectively). The SPSS storage indexes also correlated with TRUS TV and TZ (P = 0.0301 and 0.0254, respectively) (Table 1). There was no significant correlation between the results of the IPSS questionnaire and the urethral obstruction grade of the Schaefer nomogram or A-G nomogram. The total score of the SPSS correlated with the Schaefer nomogram (P = 0.0487) and with the A-G nomogram (P = 0.0413). The storage score of the SPSS also correlated with the Schaefer nomogram (P = 0.0234) and with the A-G nomogram (P = 0.0029) (Table 2).

Table 1.   Correlation between prostate volume and symptom scores
  SPSSIPSS
rsP-valuersP-value
  1. IPSS, international prostate symptom score; NS, not significant; SPSS, Saitama prostate symptom score; TRUS TV, total volume of the prostate as measured by transrectal ultrasonography; TRUS TZ, transition zone volume as measured by ultrasonography.  IPSS results were not correlated with prostate volume or transition zone volume.

Total score vsTRUS TV0.3950.0040.265NS
TRUS TZ0.3890.0050.261NS
Voiding score vsTRUS TV0.3560.0100.245NS
TRUS TZ0.3660.0160.214NS
Storage score vsTRUS TV0.3010.0300.220NS
TRUS TZ0.3100.0250.247NS
Table 2.   Pressure-flow study parameters vs symptom scores
 CategorySPSSIPSS
Mean ± SDP-valueMean ± SDP-value
  1. A-G, Abrams-Griffiths nomogram; IPSS, international prostate symptom score; NS, not significant; SPSS, Saitama prostate symptom score; UOG, urethral obstruction grade (≤2 = unobstructive).  IPSS was not correlated with Schäfer nomogram or with A-G nomogram.

Total score vs Schäfer UOGObstructive13.1 ± 6.20.04917.6 ± 9.2NS
Unobstructive 9.9 ± 4.7 14.7 ± 8.8 
Voiding score vs Schäfer UOGObstructive 8.2 ± 4.4NS 9.6 ± 5.8NS
Unobstructive 6.6 ± 3.6  8.7 ± 5.9 
Storage score vs Schäfer UOGObstructive 5.0 ± 2.70.023 8.0 ± 4.4NS
Unobstructive 3.3 ± 1.8  6.0 ± 4.0 
Total score vs A-GObstructive13.6 ± 6.20.04118.4 ± 9.4NS
Equivocal 9.8 ± 4.8 14.3 ± 7.1 
Unobstructive 9.4 ± 4.3 14.2 ± 6.8 
Voiding score vs A-GObstructive 8.5 ± 4.4NS10.1 ± 5.8NS
Equivocal 6.9 ± 3.6  8.8 ± 6.0 
Unobstructive 5.0 ± 3.5  7.3 ± 5.2 
Storage score vs A-GObstructive 5.1 ± 2.60.003 8.3 ± 4.2NS
Equivocal 2.8 ± 1.6  5.5 ± 4.0 
Unobstructive 4.4 ± 2.2  6.9 ± 4.3 

The total IPSS score significantly correlated with some of the data from free UFM: VV (P = 0.0033, rs = –0.443), Qmax (P = 0.0106, rs = –0.385) and Qave (P = 0.0030, rs = –0.448). The total SPSS score showed a signicantly increased correlation with the respective urodynamic parameters (P = 0.0043; rs = –0.431, P = 0.0005, rs = –0.527 and P < 0.0001, rs =–0.630, respectively; Table 3).

Table 3.   Index showing significant differences in free uroflowmetric parameters
  SPSSIPSS
rsP-valuersP-value
  1. IPSS, international prostate symptom score; NS, not significant; Qave, average flow rate; Qmax, maximum flow rate; RU, residual urine; SPSS, Saitama prostate symptom score; VV, voided volume.

Total score vsUFM VV− 0.4310.004− 0.4430.003
UFM RU0.3810.040 NS
UFM Qmax− 0.5270.001− 0.3850.011
UFM Qave− 0.630< 0.001− 0.4480.003
Voiding score vsUFM VV− 0.3800.012− 0.3380.025
UFM RU NS NS
UFM Qmax− 0.4910.001 NS
UFM Qave− 0.582< 0.001− 0.3740.013
Storage score vsUFM VV− 0.3800.017− 0.5090.001
UFM RU0.4760.002 0.3960.009
UFM Qmax− 0.3880.010− 0.4640.002
UFM Qave− 0.4580.002− 0.4710.002

Discussion

A self-scoring system seems to be a reasonable way to quantify the severity of BPH specific LUTS. For this purpose, several symptom questionnaires have been developed.8–10 Specifically, the IPSS, proposed by the American Urological Association in 1992, has been shown to have internal consistency and ability to powerfully discriminate between BPH and control patients.11 However, it has been shown that the IPSS score is poorly correlated with BPO or BPE.16–18 Additionally, the IPSS score is not sex-specific, it is effective in assessing the severity of LUTS in women and it is also reported that the  severity  of  LUTS,  as  assessed  by  the  IPSS,  tends to increase in an age-related manner in the elderly population.19–22

Considering the advantages and disadvantages of the IPSS, and the results of the large-scale international study conducted by the ICS-BPH group, what is expected from a prostate symptom score is: (i) to quantify LUTS in terms of both prevalence and severity; (ii) to verify the influence of LUTS (BPO and/or BPE) upon quality of life, including erectile function; (iii) to predict the degree of therapeutic efficacy and the risks of adverse effects; (iv) wide applicability across a number of different countries; (v) the ability to quantify lower urinary tract dysfunction, such as BPO or detrusor overactivity; (vi) the diagnosis of BPE; and (vii) the inclusion of indices suggestive of complications due to BPO.

As a result of the present comparison between the SPSS and the IPSS, we believe that the SPSS has shown sensitivity in identifying both BPO and BPE in BPH patients.

This seems to be because the SPSS includes symptom indices closely related with BPO. Abrams has suggested that the only absolute way of diagnosing bladder outlet obstruction (BOO) is via a pressure flow study.23 Reynard et al. stated that while the symptom of terminal dribbling was probably not related to BOO or prostatic enlargement, objective evidence of terminal dribbling on a flow curve recording is specific to BOO and as such, its presence could potentially be of value in the assessment of men with LUTS.14 Schurch et al. previously identified a correlation between hesitancy and detrusor-bladder neck dyssynergia (DBND) in clinical BPH patients by conducting a urodynamic study with multiple microtip transducer sensors. Schurch et al. define DBND as BPO which occurs not only during voiding, but also at the start or end of voiding and consider this falling within the umbrella term BOO.24 Kitahara et al. have also previously reported a close correlation between hesitancy and DBND (P = 0.002).15

According to the report of ICS-BPH study group, the symptoms and the urodynamic measures which correlated with the Lin-PURR obstruction category were urgency (P < 0.01), urge incontinence (P < 0.01), burning (P < 0.01), strength of stream (P < 0.001), frequency (P < 0.01) and urinary retention (P < 0.01).13 In other words, many symptoms associated with clinical BPH can be attributed to bladder overactivity. Abrams et al. found many symptoms that are attributed to obstruction are also attributed to bladder instability.4 de la Rosette et al. stated that, although there was a statistically significant correlation between the specific questions of the IPSS and objective grade of obstruction, the clinical significance of this finding is doubtful because none of the Spearman rank correlation coefficients were above 0.23, indicating a very weak correlation.13 This may mean that the filling symptoms they reported are not indicative of BPO or BPE.

It is interesting that, when considering the correlation of the SPSS score with urodynamic parameters, the SPSS total score was more closely related with these parameters than the SPSS voiding score and filling scores. On the other hand, it was found in the present series of patients that the IPSS filling score was more closely related with UFM parameters than the IPSS total and voiding scores (Table 3).

Barry et al. pointed out that there was no clinical usefulness to calculating separate filling and voiding subscores for the American Urological Association symptom index apart from its psychometric validity.25 However, at least in comparison with the SPSS, the IPSS may be much more useful in assessing filling phase symptoms than the SPSS. However, this means that the IPSS reflects filling phase symptoms which are not directly  associated  with  BPO  and  BPE,  and  this  is why the IPSS is often useful in evaluating female LUTS.19–21 In short, the comparison of the IPSS with the SPSS led to the classification of the roles of the respective prostate symptom scores, the former reflects filling phase symptoms and the latter may more sensitively detect BPO.

Our questionnaire incorporates 10 symptom questions: four filling symptoms (40%) and six symptoms relating to the voiding phase, while the IPSS includes seven symptom questions (four relating to voiding symptoms and three to filling symptoms) and the Danish Prostate Symptom Score (DANPSS-1) has 12 questions (six relating to filling symptoms and six to voiding symptoms). Painful voiding, stress incontinence and overflow/sleeping incontinence, which are included in DANPSS-1, were not included in our SPSS, because their prevalence is low not only in Japan but also in other countries.26 It would appear that ‘strain at the start of voiding’ is regarded as synonymous with ‘hesitancy’ in the IPSS,11,25 but the ICS-BPH study group clearly distinguishes one from the other.26 We categorize ‘hesitancy’ as ‘delay in starting the stream’ and, in our opinion, ‘strain’ includes not only ‘strain to start’ but also ‘strain during the process of voiding’. Intermittent stream is included in the IPSS as the third most frequent symptom in the ICS-BPH study group report.26 To some extent, an intermittent stream leads unconsciously to strain and, as a consequence, the SPSS has the question ‘do you have to strain to start and/or maintain urination?’. Urge incontinence is an important symptom, particularly in regard to therapeutic outcomes in BPH patients. The prevalence of this symptom was also reported as common by the ICS-BPH study group. Bearing this in mind, and taking into consideration the balance between the number of filling symptoms and the number of voiding symptoms (4 and 6, respectively), ‘urge incontinence’ was included in the SPSS. The prevalence of frequency is relatively low, being the 13th in the descending order of symptoms in the ICS-BPH study group report.26 This is however, an extremely bothersome symptom and is a major reason for visiting an out-patient service at a hospital in Japan.27 We think it is an important problem which may be related to the coexistence of detrusor overactivity and for this reason we included it in the SPSS.

In the IPSS questionnaire, each symptom is self-scored by the patient on a 6-grade scale. The IPSS aims at assessing therapeutic outcomes and grading the severity of individual urinary problems. On the other hand, the SPSS mainly aims at quantifying the severity of BPO and BPE, both of which slowly advance with age. With this aim in mind, a 4-grade scale was used in the SPSS.

The use of α-blockers to treat clinical BPH has been rapidly increasing in Japan since 1995, partly because of a significant increase in their use by primary care practitioners,28 who might often use the IPSS as a diagnostic tool. The consequence of this has been the importance of the rapid development of guidelines for the treatment of clinical BPH directed at primary care practitioners. It is essential that such guidelines should include a prostate symptom score which accurately reflects BPO and/or BPE.

The SPSS was demonstrated to significantly reflect BPO and BPE. There are however, several points that need to be further verified. First, its reproducibility and reliability need to be confirmed. The DANPSS-1 was found to correlate with BPO and BPE when it was examined in a series of 50 cases, and both reproducibility and reliability were verified.10 However, when it was re-tested in 188 cases, it was not found to correlate with either BPO or BPE.29 In order to verify the diagnostic value of the SPSS, members of Saitama Group of the Japanese Urological Association have evaluated the reproducibility, reliability, and internal consistency of the SPSS.

Lastly, we feel that we should emphasize the importance of linguistic problems, particularly in regard to difficulties in detailing and expressing the nuances of language. Even in a country like Japan, there are many dialects which sometimes lead to a misunderstanding or lack of comprehension and hence discrepancy and variation in symptom scores. In order to minimize these negative factors related to self-scoring of symptoms, it is helpful that doctors or nurses help patients to understand the content of each item that is being queried, as pointed out by Bozlu et al.30 From an international point of view, the results of studies in particular countries may not be generally applicable to other countries, necessitating the development of country specific scoring systems. Our present report is just a preliminary report but we believe that the SPSS still remains to be improved upon, particularly in the way in which questions are posed to ensure that the SPSS will be able to be widely understood by patients.

Appendix I

Content of the SPSS Questionnaire

 1 Do you have to wait for urination to start?

0-No

1-Rarely

2-Daily

3-Every time

 2 Is your urinary stream weak or does it dribble?

0-Normal

1-Weak

2-Very weak

3-Dripping

 3 Do you feel that you empty your bladder completely?

0-Always

1-Occasionally

2-Rarely

3-Never

 4 Do you have to strain to start and/or maintain urination?

0-No

1-Rarely

2-Daily

3-Every time

 5 What is the longest interval between each urination, from when you wake up until you go to bed?

0-More than three hours

1-Two to three hours

2-One to two hours

3-Less than one hour

 6 How many times do you have to urinate during the night?

0-None

1-Once or twice

2-Three to four times

3-Five times or more

 7 Do you experience an imperative (strong) urge to urinate?

0-Never

1-Rarely

2-Daily

3-Always

 8 Is the urge to urinate so strong that urine starts to flow before you reach the toilet?

0-Never

1-Rarely

2-Daily

3-Always

 9 Do you feel that the duration of your urination is excessively long?

0-None

1-Occasionally

2-Often

3-Every time

10 Do you experience dribbling after voiding, when you feel you have finished urination?

0-Never

1-Small amount just after

2-A small amount wetting the under wear

3-Sufficient to wet the trousers

Ancillary