ISSN: (Print) (Online) Journal homepage:

Archives of Andrology Journal of Reproductive Systems ISSN: 0148-5016 (Print) (Online) Journal homepage: http://www.tandfonline.com/loi/iaan19 CHANGE IN INTERNATIONAL PROSTATE SYMPTOM SCORE AFTER TRANSURETHRAL PROSTATECTOMY IN TAIWANESE MEN WITH BENIGN PROSTATE HYPERPLASIA: USE OF THESE CHANGES TO PREDICT THE OUTCOME F.-P. CHUANG, S.-S. LEE, S.-T. WU, D.-S. YU, H.-I. CHEN, S.-Y. CHANG & G.-H. SUN To cite this article: F.-P. CHUANG, S.-S. LEE, S.-T. WU, D.-S. YU, H.-I. CHEN, S.-Y. CHANG & G.-H. SUN (2003) CHANGE IN INTERNATIONAL PROSTATE SYMPTOM SCORE AFTER TRANSURETHRAL PROSTATECTOMY IN TAIWANESE MEN WITH BENIGN PROSTATE HYPERPLASIA: USE OF THESE CHANGES TO PREDICT THE OUTCOME, Archives of Andrology, 49:2, 129-137, DOI: 10.1080/01480510390129250 To link to this article: https://doi.org/10.1080/01480510390129250 Published online: 09 Jul 2009. Submit your article to this journal Article views: 54 View related articles Citing articles: 1 View citing articles Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalinformation?journalcode=iaan19

ARCHIVES OF ANDROLOGY 49:129 137 (2003) Copyright # 2003 Taylor & Francis 0148-5016/03 $12.00+.00 DOI: 10.1080=01480510390129250 CHANGE IN INTERNATIONAL PROSTATE SYMPTOM SCORE AFTER TRANSURETHRAL PROSTATECTOMY IN TAIWANESE MEN WITH BENIGN PROSTATE HYPERPLASIA: USE OF THESE CHANGES TO PREDICT THE OUTCOME F.-P. CHUANG S.-S. LEE S.-T. WU D.-S. YU H.-I. CHEN S.-Y. CHANG G.-H. SUN Division of Urology, Department of Surgery, Tri-Service General Hospital, National Defense Medical Center, National Defense College, Taipei, Taiwan, Republic of China To investigate whether the pre- and postoperative International Prostate Symptom Score (IPSS) change predicts the outcome of transurethral prostatectomy in a Taiwanese population, 99 patients (transurethral prostatectomy candidates) were assessed with the IPSS before and 6 12 months after surgery. All symptoms improved significantly postoperatively. Patients with a greater preoperative IPSS benefited the most. Improvements in preoperative obstructive symptoms (incomplete emptying, intermittency, straining, and weak stream) were greater than those in irritable symptoms (urgency, frequency, and nocturia). A significant correlation was found between IPSS and quality of life (QOL) before and after transurethral prostatectomy. A change of 1 unit on the IPSS scale was found to decrease the QOL score 0.282 units. The positive predictive value of a 7-IPSS point decrease depended on the predictive IPSS criteria applied. When the preoperative IPSS was more than 17, the sensitivity was 83.5% and specificity was 30%. Postoperative improvement did not differ significantly between acute urinary retention (AUR) and non-aur patients. Change in IPSS of more than 7 points predicted symptomatic improvement with high sensitivity. The predictive value depends on the definition of significant improvement (magnitude of IPSS change) and on the level of IPSS symptoms (sufficient to warrant transurethral prostatectomy). Keywords IPSS, prostatectomy, prostatic hyperplasia The authors acknowledge Meei-Horng Yang, who diligently analyzed all completed questionnaire data. Address correspondence to Guang-Huan Sun, Division of Urology, Department of Surgery, Tri-Service General Hospital No. 325, Section 2, Cheng-Gung Road, Neihu 114, Taipei, Taiwan, ROC. E-mail: ghsun@ndmctsgh.edu.tw 129

130 F.-P. Chuang et al. The main feature of benign prostatic hyperplasia (BPH) is benign prostatic enlargement, which may lead to benign prostatic obstruction (BPO) and lower urinary tract symptoms (LUTS). Patients with benign prostatic enlargement seek medical treatment for bothersome lower urinary tract symptoms and the relief of symptoms is the most frequent indication for intervention [15, 16]. Several differential instruments have been developed to quantitate the severity of BPH symptoms, such as the Boyarsky score [5, 15], Madsen Iversen score [11, 15], American Urological Association (AUA) symptom index, and Danish prostatic symptom score [2, 7, 15]. The AUA symptom index was developed specifically as an outcome measure in the study of different BPH therapies [15, 16]. The first World Health Organization consultation on benign prostatic hyperplasia adopted the AUA symptom index with the addition of one quality-of-life question, and called it the International Prostate Symptom Score (IPSS) [13, 15]. IPSS is widely used as a tool to quantify symptoms related to benign prostatic enlargement. Reporting changes in symptom scores following treatment has not been standardized. When the baseline symptom scores are low or intermediate, small and clinically insignificant changes may yield large percentage changes in scores, and when baseline symptom scores are high, the resulting relatively large absolute changes may not be clinically significant. Therefore, the significance of postoperative change in subjective symptoms and the use of IPSS as a basis for predicting outcome and treatment decisions warrants further investigation. MATERIALS AND METHODS A total of 99 consecutive transurethral prostatectomy patients were retrospectively studied. These patients sought medical treatment for bothersome lower urinary tract symptoms (30 patients with acute urinary retention) and were assessed by staff urologists in the conventional manner by history taking and digital rectal examination. Exclusion criteria were previous prostatic surgery and known carcinoma of the prostate. Patients were assessed with the IPSS before and 6 12 months after surgery. The IPSS questionnaire, which includes one quality of life question, was selfadministered on both occasions. IPSS scores (range 0 35) and quality of life scores (range 0 6) are expressed as means plus or minus standard deviation of the mean. Completed questionnaire data were analyzed using the statistical package for social sciences (SPSS) for Windows software package. Means were compared by independent t test. Pearson s correlation coefficient was used to measure associations between continuous variables and the one-way ANOVA-Scheffe s method was used to measure associations between nominal variables. The level of significance for all tests was set at p <.05. RESULTS Mean patient age was 73 6 years (range 55 93). The mean preoperative IPSS was 22 6.9 (range 7 35) and median quality of life score was 5 (range 2 6). At follow-up of 6 12 months postoperatively, mean IPSS was 5.7 5.3 (range 0 23) and median quality of life score was 2 (range 0 5). There was a significant mean decrease (improvement) in IPSS of 16.3 for all questions ( p <.005 for all symptoms). The greatest improvement was found for weak stream, incomplete emptying, and intermittency. Of the symptom scores,

Change in IPSS After TURP 131 the highest mean scores were for frequency, weak stream, and incomplete emptying preoperatively, and for nocturia and frequency postoperatively (Figure 1). Patients with a greater preoperative IPSS improved the most, and the preoperative total IPSS correlated with postoperative quality of life (QOL) score (Spearman s correlation coefficient r ¼.322 and p <.005 preoperative total IPSS, and r ¼.56 and p <.001 postoperative QOL score). In addition, postoperative total IPSS correlated with postoperative QOL score (r ¼.632 and p <.001). The similar and strong preoperative and postoperative correlations seem to indicate a constant relationship, that is, a change of 1 unit on the IPSS scale corresponded to an improvement of 0.282 units in the QOL score (Figure 2). Obstructive symptoms (emptying, intermittency, straining, and weak stream) improved more than irritable symptoms (urgency, frequency, and nocturia) ( p <.001). The symptoms of all patients selected for transurethral prostatectomy were categorized preoperatively by conventional clinical assessment as mild, moderate, or severe (which corresponded to IPSS 0 7, 8 19, and 20 35, respectively). Among these 3 subgroups, there were no statistically differences in age, prostate-specific antigen (PSA) level, postoperative IPSS, and quality of life score. However, these groups differed in the amount of postoperative IPSS improvement ( p <.001). The magnitude of postoperative improvement increased with the severity of preoperative symptoms. Additionally, the severity of postoperative symptoms was not significantly different among these groups. Receiver operating characteristic (ROC) curves were calculated to estimate the value of the preoperative IPSS assessment in predicting significant symptomatic postoperative improvement (Table 1 and Figure 3) [8, 12, 15]. We arbitrarily defined significant postoperative symptomatic improvement as a decrease of 10 or more symptom score points, which correlated approximately with an improvement of 3 quality of life points, although symptom improvements of less than 7 10 IPSS points were still subjectively perceptible [4, 15]. ROC analysis indicated that if only patients with a preoperative IPSS of 17 or more underwent surgery, we would have selected 83.5% of those who might achieve an improvement of 10 or more IPSS points (true positives), and only 30% of those who might not (false positives). Preoperative total IPSS was not correlated with prostate weight, serum PSA level, and age (prostate weight: r ¼.088 and p >.05; serum PSA level: r ¼.003 and p >.05; and age: r ¼ 7.032 and p >.05). Postoperative improvement did not differ significantly between AUR and non-aur patients. DISCUSSION The IPSS was designed to assess symptoms in men with benign prostatic enlargement. It is not gender specific, since age-matched men and women have similar IPSSs [10, 15]. It is based on the AUA symptom index and consists of questions on seven LUTS, scored from 0 to 5, with a total score ranging from 0 to 35. In addition, one question concerning the impact of BPH on QOL, scored on a scale from 0 to 6, has been added. The quality of life score was well correlated with the symptom score [9]. IPSS cannot be disease-specific for benign prostatic enlargement. IPSS does not correlate with urodynamic assessment of obstruction and detrusor contractility, serum PSA level, prostate weight, and age [1, 9, 15]. Although IPSS can be useful in evaluating patients with lower urinary tract symptoms and

132 F.-P. Chuang et al. Figure 1. (A) Symptom profile before transurethral prostatectomy (TURP). (B) Symptom profile after TURP. (C) Difference in mean pre- and post-turp IPSSs.

Change in IPSS After TURP 133 Figure 1. Continued. possible obstruction, is it reasonable to use IPSS as an indication for prostatectomy and to measure the outcome? Benign prostatic hyperplasia is a benign enlargement of the prostate with consequent benign prostatic obstruction and lower urinary tract symptoms. Transurethral prostatectomy is a common procedure, which 20% of all men are likely to undergo eventually [14, 15]. There are many measures of outcome in the treatment of LUTS=BPH, such as urodynamic parameters (flow rate and post void residual volumes), prostate size, symptom scores, the rates of acute urinary retention, and the rates of prostatectomy [3, 17]. In our study, IPSS and QOL scores were strongly and significantly correlated before and after surgery. The similar and strong preoperative and postoperative correlation seems to indicate a constant relationship, that is, a decrease of 1 unit on the IPSS scale corresponds to an improvement of 0.282 units in the QOL score. We found significant improvements in overall IPSS, all IPSS symptoms, and quality of life 6 12 months after transurethral prostatectomy (Figures 1 and 2). Patients with a greater preoperative IPSS gained the most symptomatic benefit. Patients with mild symptoms (IPSS 0 7) do not seem to benefit from the transurethral prostatectomy. In our series, the classical obstructive symptoms (incomplete emptying, intermittency, weak stream, and straining) improved the most, while frequency and nocturia improved less, remaining main symptoms postoperatively and indicating that nocturia is less responsive

134 F.-P. Chuang et al. Figure 2. Total IPSS improvement versus quality of life (QOL) improvement (r ¼.282, p <.01). A change of 1 unit on the IPSS scale corresponded to an improvement of 0.282 units in the QOL score. Table 1. True-positive and false-positive percentages for a predicted postoperative IPSS improvement of at least 7 or at least 10 points % IPSS improvement 7 or more points 10 or more points IPSS threshold True positive False positive True positive False positive 6 100 100 100 100 8 100 76 100 85 10 96 61 100 70 13 95 46 98 50 15 87 30 91 35 17 80 23 83 30 19 74 23 77 30 21 59 15 60 25 24 46 7.7 46 20 29 15.1 0 16 0 34 4.7 0 5 0 Note. IPSS thresholds signify the preoperative IPSS above which the defined outcome is predicted. The percentage of true positives is a measure of sensitivity and that of false positives, of specificity. Specificity and sensitivity of predicting the specified IPSS improvement depend on the given preoperative IPSS threshold.

Change in IPSS After TURP 135 Figure 3. ROC curve for predictive value of preoperative IPSS. Good outcome was defined as improvement of 10 or more points in IPSS. Percentage of patients who achieved good outcome was calculated for cutoff points of preoperative IPSS. True-positive and false-positive values represent percentages of patients with greater than defined preoperative IPSS cutoff values who did and did not achieve postoperative IPSS improvement, respectively. to transurethral prostatectomy than other symptoms (Figure 1). Bruskewitz et al. assessed the long-term (3 years) effect of TURP on several obstructive and irritable symptoms in 84 patients. Most patients experienced an improvement in obstructive symptoms after TURP, although there was a small deterioration after 3 years. In addition, the improvement in irritable symptoms was smaller than for voiding symptoms [6]. These data confirm the fact that TURP has a larger effect on voiding than on filling symptoms. Since no single preoperative IPSS symptom strongly correlated with postoperative maximum flow rate change or IPSS improvement, the preoperative total IPSS score seems the best predictor of postoperative symptomatic outcome [15]. Preoperative total IPSS was not correlated with prostate weight, serum PSA level, or age, which was same as in previous reports [1, 9, 15]. Postoperative improvements in AUR and non-aur patients did not differ significantly. Is it useful to use preoperative IPSS as a predictor of postoperative outcome? Oliver et al. [15] reported that a threshold for transurethral prostatectomy of a preoperative IPSS of 17 would be more appropriate. The calculated ROC curves indicate that

136 F.-P. Chuang et al. preoperative IPSS can predict symptomatic outcome with good sensitivity and moderate specificity. We chose a postoperative decrease of 10 IPSS points, which corresponded to an improvement of approximately 3 points in quality of life, as indicative of a clinically significant symptomatic improvement. It seems clear that preoperative IPSS can be used as a good predictor of symptomatic improvement at the expense of a moderate number of patients who will not improve significantly despite high IPSSs. If a low threshold is chosen, a relatively large number not achieving the defined level of improvement will be included. For example, if all patients with an IPSS of 6 or more undergo transurethral prostatectomy, this would include 100% of those who would benefit, that is, IPSS improvement of 10 points, but also 90% of those who would not (Table 1). It would appear that a threshold for transurethral prostatectomy of an IPSS of 17 would be more appropriate, since this would include 83.5% of patients who will and only 30% of those who will not benefit. CONCLUSION The IPSS is valuable in assessing the symptoms complex of lower urinary tract dysfunction with benign prostatic enlargement. Transurethral prostatectomy is commonly performed to achieve improvement in symptoms and quality of life. Most predictive of outcome after transurethral prostatectomy is the preoperative total IPSS. Symptomatic improvement after transurethral prostatectomy can be predicted by the preoperative IPSS with a high sensitivity, depending on the preoperative IPSS cutoff point chosen as an indication for surgery. Therefore, change in IPSS (a composite measure of symptomatic and quality of life improvement) can be used as a measure of outcome. EDITORIAL COMMENT In this study, patients with greater preoperative IPSS benefited the most. Obstructive symptom was relieved more readily than bladder instability, which is not unexpected. There was no difference in response between those with or without preoperative acute urinary retention (AUR). However, adjustments for age usually indicate that patients older than 80 years old do not fair as well when they present with AUR. REFERENCES 1. Barry M, Cockett A, Holtgrewe H, McConnel J, Sihelnik S, Winfield H (1993): Relationship of symptoms of prostatism of commonly used physiological and anatomical measures of the severity of benign prostatic hyperplasia. J Urol 150:351 358. 2. Barry M, Fowler F, O Leary M, Bruskewitz R, Holtgrewe H, Mebust W, Cockett A, the Measurement Committee of the American Urological Association (1992): The American Urological Association symptom index for benign prostatic hyperplasia. J Urol 148:1549 1557. 3. Barry M, Roehrborn C (2000): Benign prostatic hyperplasia. Clin Evidence 4:453 461.

Change in IPSS After TURP 137 4. Barry M, Williford W, Chang Y, Machi M, Jones K, Walker-Corkery E, Lepor H (1995): Benign prostatic hyperplasia specific health status measures in clinical research: how much change in the American Urological Association symptom index and the benign prostatic hyperplasia impact index is perceptible to patients? J Urol 154:1770 1774. 5. Boyarsky S, Jones G, Paulsen D, Prout G Jr (1976): A new look at bladder neck obstruction by the Food and Drug Administration regulators: guidelines for investigation of benign prostatic hypertrophy. Trans Am Assoc Genitourin Surg 68:29 32. 6. Bruskewitz R, Larsen E, Madsen P, Dorflinger T (1986): 3-year follow-up of urinary symptoms after transurethral resection of the prostate. J Urol 136:613 615. 7. Hansen B, Flyger H, Brasso K, Schou J, Nordling J, Thorup J, Mortensen S, Meyhoff H, Walter S, Hald T (1995): Validation of the self-administered Danish Prostatic Symptom Score (DAN-PSS-1) system for use in benign prostatic hyperplasia. Br J Urol 76:451 458. 8. Hanley J, McNeil B (1982): The meaning and use of the area under a receiver operating characteristic (ROC) curve. Radiology 143:29 36. 9. Javle P, Jenkins SA, Machin DG, Parsons KF (1998): Grading of benign prostatic obstruction can predict the outcome of transurethral prostatectomy. J Urol 160:1713 1717. 10. Lepor H Machi G (1993): Comparison of AUA symptom index in unselected males and females between fifty-five and seventy-nine years of age. Urology 42:36 40. 11. Madsen P, Iversen P (1983): A point system for selecting operative candidates. In: Benign Prostatic Hypertrophy. Hinman F Jr, Boyarsky S (Eds). New York: Springer-Verlag, pp 763 765. 12. McNeil B, Keller E, Adelstein S (1975): Primer on certain elements of medical decision making. New Engl J Med 293:211 215. 13. Mebust W, Roizo R, Schroeder F, Villers A (1991): Correlations between pathology, clinical symptoms and the course of the disease. In: Proceedings of the International Consultation on Benign Prostatic Hyperplasia, pp 53 62. 14. Mebust W, Holtgrewe H, Cockett A, Peters P, Writing Committee (1989): Transurethral prostatectomy: immediate and postoperative complication: a cooperative study of 13 participating institutions evaluating 3885 patients. J Urol 141: 243 247. 15. Oliver W, Carole B, Villis R (1997): Does evaluation with the international prostate symptom score predict the outcome of transurethral resection of the prostate? J Urol 158:94 99. 16. Pierre T (1998): Relief of BPO or improvement in quality of life? Eur Urol 34(suppl 2):3 9. 17. Speakman MJ (2001): Initial choices and final outcomes in lower urinary tract symptoms. Eur Urol 40(suppl 4):21 30.