To investigate the correlation between the visual prostate symptom score, the international prostate symptom score, and uroflowmetry parameters in adult Thai males of different educational levels.

BACKGROUND: To evaluate the correlation between the International Prostate Symptom Score (IPSS), Visual Prostate Symptom Score (VPSS), and uroflowmetry parameters in Thai males and to examine the possibility in establishing a severity cut-off point for VPSS.
METHODS: Between 1st February and 31st May 2016 a total of 200 men were enrolled onto the study and divided into high and low educated groups. All of them were requested to complete paperwork including their personal data, and then to complete a VPSS and IPSS questionnaire. Uroflowmetry, residual urine and prostate size were measured. The relationship between the answers to the VPSS and IPSS together with the other objective parameters was assessed using Spearman's rank test.
RESULTS: The mean age of the patients was 61.2 years. 69.9% of subjects were highly educated. There was a statistically significant correlation between VPSS and IPSS in total, and any individual scores except frequency score. There was weak correlation between the VPSS and the uroflowmetry parameters and prostate size. The low educated group had a statistically significant lower completion rate of both the VPSS (41.6% vs. 79.8%) and IPSS (61.2% vs. 81.2%) without assistance than the higher educated group and most of the lower educated group felt that the IPSS was easier to understand than the VPSS (51.2% vs. 48.8%, P < 0.001). A VPSS severity score ≤6 or ≥14 had a very high specificity that predicted the patients would have mild or severe symptoms (94.7% and 98.6%) while a VPSS between 7 and 13 had a high sensitivity (90.8%) but a low specificity (16.9%) when it came to the prediction of moderate symptoms.
CONCLUSION: VPSS showed a significant correlation to the IPSS and uroflowmetry parameters. A VPSS score ≤6, 7 to 13 and ≥14 may indicate mild, moderate, and severe symptoms respectively.

Introduction

Lower urinary tract symptoms (LUTS) are common urologic complaints in male patients with a benign prostate obstruction. An epidemiologic study showed that LUTS could be found in 72.3% of Western men aged between 40 and 95 years. The international prostate symptom score (IPSS), first published in 1992 by The American Urologic Association, is a globally used questionnaire. The aims of this questionnaire are to evaluate the symptoms and to determine the severity of LUTS by converting subjective symptoms into an objective scoring system. The total calculated as a result of the IPSS can also be used in longitudinal follow-up to assess treatment efficacy.3, 4 To date, the IPSS is translated and validated in multiple languages across the world.

Theoretically, the IPSS was designed to be an easy, self-administered questionnaire. However, some studies showed that successful completion of an IPSS would require at least grade 6 reading level to understand the questions, and up to 77.5% of low-educated patients were unable to completely fill in an IPSS. Therefore, many patients would often need assistance to complete the questionnaire which could result in a bias in the patients' interpretation and responses to the IPSS questionnaire.

The visual prostate symptom score (VPSS) is a new alternative scoring system which was developed by A. E Groeneveld et al from South Africa. Recent studies demonstrated that the results of a VPSS show a significant correlation with the IPSS and can be completed without assistance, requiring less time especially in low literacy patients. The VPSS also correlates more significantly with urine flow rate than an IPSS. One study showed that the VPSS might be utilized in the longitudinal follow-up assessment of changes in symptom severity and evaluation of treatment efficacy.

The IPSS had already been translated and validated into the Thai language. However, due to the fact that Thailand is a developing country, many patients, especially the elderly, are low educated. Views have been expressed by these patients that the IPSS is difficult to understand and that they are unable to complete it by themselves. Thus, the VPSS may be a good alternative to overcome this problem and ideally it should be able to replace the IPSS in clinical practice. Our authors believe that the VPSS could replace the IPSS but only if it could assess the symptoms and severity of LUTS by dividing them into mild, moderate, and severe symptoms. The rationale behind this reasoning is that treatment options differ between each severity subgroup, therefore only proving a correlation to the IPSS results may not fulfill this objective. Unfortunately, there is no data to support the ability of the VPSS, and the cut-off point of VPSS for each severity group is not yet established although it is available for the IPSS. Also limited data are available regarding the applicability of the VPSS in Thai population.

The objectives of this study have primary and secondary endpoints. The primary endpoints are (1) to evaluate the correlation between the VPSS, IPSS, and uroflowmetry parameters in Thai men and (2) to compare the accessibility of self-completion and understanding of each symptom score system between low and high-educated patients. The secondary endpoint is to establish a VPSS severity cut-off point by comparing results to the IPSS severity score system.

Materials and methods

This multicenter study was conducted at Lerdsin hospital in Bangkok and Maharaj Nakorn Chiang Mai Hospital in Chiang Mai, northern Thailand. The Institutional Ethics Committee of Lerdsin Hospital approved this prospective cohort study. From 1st February to May 2016, 200 male patients who visited the two urologic departments were enrolled to the study.

The sample size for this study was calculated using a 47.9% prevalence of male LUTS in the general population and 95% confidence interval. From this calculation, a minimum of 153 patients must be enrolled. Two hundred male patients enrolled to this study ensured the sample size was maintained despite the possibility of any lost data for whatever reason.

All of the patients were asked to fill in both an IPSS and a VPSS questionnaire independently. Assistance was provided only when the patient requested it. The VPSS consists of 4 questions: Q1: force of urine stream; Q2: frequency; Q3: nocturia; and Q4: quality of life. The VPSS severity score is defined using the sum of the Q1 to Q3 score. In addition to the completion of both questionnaires, an additional question was added: “Which questionnaire do you feel that is easier to understand, the IPSS or the VPSS?”

Prostate size was measured by transrectal ultrasonography or transabdominal ultrasound after digital rectal examination. Maximal flow rate, mean flow rate, and voided volume were accessed by uroflowmetry (Danflow 1100, Dantec Dynamics Ltd, Bristol, UK). Postvoided residual volume was also measured by bladder scan (Cubescan Biocon-500, Mcube Technology, Seoul, Korea). Demographical details include age, level of education, occupation, and monthly income were collected. Patients were divided into low education (≤primary school) and high education (≥high school) group.

Patient demographic characteristics were analyzed using descriptive statistics. The χ2/Fisher's exact test was performed to compare categorical variables such as level of education, which may be important factors related to completing and understanding the IPSS and VPSS questionnaires. Pearson correlation coefficient (r)/Spearman's rank correlation coefficient was used to evaluate the correlation between the VPSS and IPSS, urodynamic parameters, and prostate size. A P < 0.05 was accepted as statistically significant.

The sensitivity and specificity of presumed VPSS severity score ranges which may be used to define the severity of symptoms were evaluated by comparing results to the IPSS data as a gold standard reference.

Results

A total of 200 men were enrolled from 1st February to 31st May 2016. The mean age of patients was 61.2 years, the range being 18–98 years. Patients were subdivided into low and high-educated subgroups according to their educational status. Patients who had no formal schooling or only went to primary school (grade 6 education) were defined as low education while others were classed as highly educated. Numbers, income, and mean age of each subgroup are shown in Table 1. 99.5% of patients were able to read and understand Thai language. The majority of patients in this study were well-educated (69.9% vs. 30.1%). Patients in the low-educated group were older than the highly educated patients (65.1 vs. 60.2), while monthly income is significantly higher in the highly educated group. The mean scores for IPSS and VPSS were 14.4 and 9.61, respectively. 70.2% of patients in this study had moderate symptoms while 34 (19.4%) and 18 (10.3%) of patients had mild and severe symptoms, respectively, according to the IPSS data. The quality of life (QoL) and symptom subscores are also shown in Table 2. The results of the uroflowmetry parameters and prostate size of study population are reported in Table 3.

Table 1

Demographic characteristics of the patients.

Level of education	Low-education level		High-education level				P
	No education	Primary school	High school	Bachelor degree	Master degree	Doctoral degree
Number of patients (%)	1 (0.50)	59 (29.65)	95 (47.74)	29 (14.57)	13 (6.52)	2 (1.01)	–
Mean age (years)	81.0	64.8	60.3	60.5	60.0	52.0	0.013
Median income per month (USD)	None reported	300	500	666	1300	1333	0.0001

Table 2

Mean total IPSS, total VPSS, symptom subscore of the patients and correlation coefficient (R) between IPSS and VPSS.

Symptom score	Mean IPSS	Mean VPSS	Coefficient	P
Total score	14.4	9.61	0.493	<0.001
QoL score	2.50	2.43	0.738	<0.001
Voiding symptom score (IPSS Q1, 3, 5, 6 vs. VPSS Q1)	7.14	3.01	0.410	<0.001
Storage symptom score (IPSS Q2, 4, 7 vs. VPSS Q2, 3)	6.23	6.61	0.417	<0.001
Frequency score (IPSS Q2 vs. VPSS Q2)	1.94	4.12	0.183	0.015
Nocturia score (IPSS Q7 vs. VPSS Q3)	2.57	2.49	0.745	<0.001

IPSS, international prostate symptom score; VPSS, visual prostate symptom score; QoL, quality of life; Q, question.

Table 3

Results of uroflowmetry parameters and prostate size of the patients.

Uroflowmetry parameters and prostate size	Mean value (range)
Maximal flow rate (ml/sec)	10.7 (1.0–29.8)
Average flow rate (ml/sec)	5.60 (0.5–18.0)
Voided volume (ml)	177.2 (38.0–630.0)
Postvoided residual (ml)	63.6 (0.0–440.0)
Prostate volume (ml)	35.8 (11.4–65.0)

There was a statistically significant correlation between the total scores of the IPSS and VPSS, the QoL score, storage symptom score, voiding symptom score, frequency, and nocturia (P < 0.05). The strength of the correlation of the total symptom score (R = 0.493), storage (R = 0.41), and voiding symptom score (R = 0.417) is moderate. There were strong correlations in both the QoL and nocturia score (R = 0.738 and 0.745). In contrast, a very poor correlation between the frequency scores in the questionnaires were found (R = 0.183) (Table 2). There was a significant correlation between the VPSS and the uroflowmetry parameters and prostate size; in all sections apart for the voided volume, this correlation strength is poor (Table 4). Unlike the VPSS, no statistical correlation between IPSS and any uroflowmetry parameters and prostate size was detected.

Table 4

Correlation coefficient between IPSS or VPSS using uroflowmetry parameters and prostate size.

Parameter	IPSS		VPSS
	Correlation coefficient (R)	P	Correlation coefficient (R)	P
Maximal flow rate (ml/sec)	−0.046	0.548	−0.159	0.035∗
Average flow rate (ml/sec)	−0.048	0.528	−0.158	0.037∗
Voided volume (ml)	−0.026	0.734	−0.146	0.054
Postvoided residual (ml)	0.061	0.425	0.167	0.027∗
Prostate size (ml)	0.125	0.101	0.241	0.001∗

IPSS, international prostate symptom score; VPSS, visual prostate symptom score.

Statistically significant.

Seventy-five percent of all patients were able to complete the IPSS without any assistance while only 61.3% completed the VPSS unaided (P < 0.001). Comparing the low and high education group, the low-educated group had a statistically significant lower self-completion rate of both the IPSS and VPSS (Table 5) (P < 0.001). Thirty-seven out of 60 (61.7%) of the low-educated patients could complete the IPSS without any assistance whereas only 25 of them (41.6%) could do the same for the VPSS. A higher proportion of the study population felt the IPSS was easier to understand than the VPSS (51.2% vs. 48.8%, P < 0.001), especially for low-education patients (54.9% vs. 45.1%, P < 0.001). On the other hand, most high-educated patients (50.4%) felt VPSS was easier to understand.

Table 5

Number of patients able to complete questionnaire without any assistances.

Questionnaire	Total (%)	Low education (%)	High education (%)	P
IPSS	150 (75.00)	37 (61.70)	113 (81.20)	0.001∗
VPSS	122 (61.30)	25 (41.60)	97 (79.80)	<0.001∗

IPSS, international prostate symptom score; VPSS, visual prostate symptom score.

Statistically significant.

As regards the secondary endpoint, the sensitivity and specificity of a presumed severity score cut-off point of the VPSS by comparing this to the standard IPSS value are shown as Table 6. VPSS ≤6 or ≥14 had very high specificity (94.7% and 98.58%) when it came to predicting that patients would have mild or severe symptoms, whereas a score between 7 and 13 had a high sensitivity (90.82%) but low specificity (16.88%) for the prediction of moderate symptoms as described by the IPSS.

Table 6

The sensitivity and specificity of VPSS severity cut-off points compared to standard IPSS cut-off points.

VPSS severity cut-off point	Sensitivity	95% CI	Specificity	95% CI
Mild (≤6)	13.95	5.30–27.93	94.70	89.38–97.84
Moderate (7–13)	90.82	83.28–95.71	16.88	9.31–27.14
Severe (14–17)	20.59	8.70–37.90	98.58	94.97–99.83

CI, confidence interval; IPSS, international prostate symptom score; VPSS, visual prostate symptom score.

Discussion

The IPSS is a globally accepted standard questionnaire that is used for the evaluation of LUTS in men. It is able to assist physicians in the design of a treatment strategy by defining what symptoms the patient has and how severe the symptoms. However, the IPSS is quite difficult for patients to complete, especially when the patients have a low level of education. A new scoring system, the VPSS with the use of pictograms was developed to overcome this obstacle. One early study showed that the results from the VPSS correlates significantly with those calculated using the IPSS (R = 0.6227, P < 0.001), Qmax (R = −0.42, P < 0.001), and Qave (R = −041,P < 0.001). Furthermore, the VPSS can be completed without assistance by a greater proportion of illiterate men (68% in VPSS vs. 13% in IPSS). This pictogram was modified to make it easier for patients to understand in a subsequent study. Several recent studies from different countries confirmed that the results from the VPSS has a statistically significant correlation to the IPSS, even in urethral stricture patients, and may be useful for longitudinal follow-up.7, 10, 11, 12, 13, 14, 15

Similar to the findings of previous studies, our study confirmed a significantly high correlation between the IPSS and VPSS in total and all individual scores apart from the frequency score. In contrast, however, several previous studies showed a fair correlation of frequency score between questionnaires9, 11, 12, 13, 14, 15, whereas in our study there is a very poor correlation of frequency score (r = 0.183, P = 0.015). This may be because there are imperfections in both questionnaires when it comes to assessing frequency. We noted that Q2 of the VPSS may overestimate frequency symptoms since urinating six times during the day is generally accepted as normal but would result in the highest score (a score of 6) which would be interpreted as a very bad symptom. Additionally, in clinical practice, a single daytime urination would not be considered normal. There was only one patient in our study who reported “1” in Q2 of the VPSS. For these reasons, we felt this question should be modified to better reflect real life situations.

Our study showed that the VPSS has a higher correlation with some objective uroflowmetry parameters than IPSS although these correlations are weak. They include uroflow rate, residual urine, and prostate size. These results are comparable with previous studies7, 9, 10, 11, 12, 13, 14, 15. This phenomenon may be the result of the Q1 pictogram in the VPSS which helps patients to recognize the urine flow more precisely than does the IPSS.

Unlike many previous studies which reported that a greater percentage of lower educated patients could complete the VPSS than IPSS without assistance,9, 11, 12, 13, 14 surprisingly our study reported an inverse result. A lower proportion of low-educated patients could complete the VPSS without any assistance than the IPSS (41.6% vs. 61.2%) this was also found to be the case in high-educated patients. This may be the result of many low-educated patients not understanding what the pictograms in the VPSS meant. This uncertainty meant they asked for assistance. Meanwhile highly educated patients felt the VPSS is easier to understand than the IPSS. This result may be due to Thai language being the only single national language in Thailand. This may explain the fact that low-educated patients who are literate can understand the meaning of each IPSS question easily. In contrast, for pictograms in VPSS, it may be difficult to interpret the meaning of each picture. So the VPSS may be more difficult to be understood by low-educated patients in Thailand. Nevertheless, the alteration of responses to VPSS was significantly less than alterations in the IPSS when receiving assistance. Therefore, the VPSS may still be useful in a country with a single national language and a more literate population.

To date our study is the first research to examine the possibility of establishing a VPSS severity score classification by comparing the outcomes to the gold standard IPSS severity system. We found that a VPSS of less than 6 or greater than 14 had a high specificity but a low sensitivity to correlate with mild (≤7) or severe IPSS (≥20). Meanwhile a VPSS between 7 and 13 had a very high sensitivity but low specificity when it came to a correlation with a moderate IPSS (8–19). The reason for this result may be because our study was not primarily designed for this diagnostic study. So further studies, specific for this purpose should be conducted.

In conclusion, there was a statistically significant correlation between the VPSS and the IPSS as regards the overall score and also between the individual scores apart from the score specific to frequency. The VPSS also had a higher statistically significant correlation to the uroflow rate, residual urine, and prostate size than the IPSS. However a higher proportion of the low-educated patients were unable to complete VPSS without any assistance compared to the number who could not complete the IPSS. This may be because the VPSS pictograms are more difficult to understand for low-educated patients. As regards the severity assessment, items ≤6, 7 to 13, and ≥14 from the VPSS may be used to indicate mild, moderate, and severe symptoms but further modification would to improve its accessibility and therefore increase its reliability and feasibility in clinical practice.

Conflicts of interest

No conflict of interest relevant to this article was declared.