- Email: [email protected]
Transition zone volume and transition zone ratio: predictor of uroflow response to finasteride therapy in benign prostatic hyperplasia patients
TRANSITION ZONE VOLUME AND TRANSITION ZONE RATIO: PREDICTOR OF UROFLOW RESPONSE TO FINASTERIDE THERAPY IN BENIGN PROSTATIC HYPERPLASIA PATIENTS ASHUTOSH TEWARI, M.D. KATSUTO SHINOHARA, M.D. PERINCHERY NARAYAN, M.D. From the University of Florida, Gainesville, Florida, and the University of California School of Medicine and the Department of Veterans Affairs Medical Center, San Francisco, California
ABSTRACT-Objectives. The aim of this study was to: (1) determine effects of finasteride on transition zone (TZ) volume, TZ ratio (TZ volume/total prostate volume), and total prostate volume; (2) analyze differences in TZ and total volume reduction among patients who improve peak urinary flow rates following finasteride therapy with those who do not; (3) investigate which parameters correlate with improvement in peak urinary flow rate and urinary symptom score; and (4) establish if there is any predictive value of these parameters for response to therapy. Methods. Twenty-three patients with symptomatic benign prostatic hyperplasia (BPH) were treated with finasteride (5 mg/d) for 12 months and underwent transrectal ultrasound (TRUS) evaluation of total and TZ volume of prostate and measurement of peak flow rate and modified Boyarsky symptom score at baseline and at 12 months. Statistical analysis was done by unpaired t, Mann-Whitney, and Spearman rank correlation tests among responders (more than 3 cc/s improvement in peak flow rate) and nonresponders (less than 3 cc/s improvement in peak flow rate) to therapy. Results. (1) Responders had substantial reduction in TZ volume (44.8% versus 16.05%; PC 0.03) and TZ ratio (25% versus 5% increase, P-c 0.02) compared with nonresponders. (2) There was a significant correlation between reduction in TZ volume (r = 0.50; PC 0.03) and TZ ratio (r = 0.60; PC 0.006) with improvement in peak flow rates. No similar correlation was seetr with total prostate volume changes. (3) Pretreatment TZ ratio helped in predicting peak flow improvement following finasteride therapy (r = 0.52; P-c 0.01) and there was a 2.5-fold increased chance of improvement if baseline TZ ratio was more than 0.51. The modified Boyarsky symptom score decreased by 3.1 (mean), but there was no correlation with changes in peak urinary flow rate, total prostate volume, TZ volume, and TZ ratio. TZ ratio did not have significant predictive value for improvement in symptom score. Conclusions. This study provides data that simple measurable parameters are available which may be used prior to therapy to predict uroflow response to finasteride and similar agents.
Benign prostatic hyperplasia (BPH) is the most common neoplasm in the aging human male. It is estimated that 1 in every 4 men in the United Submitted (Rapid Communication): (with revisions): October 18, 1994
258
July 22, 1994, accepted
States will be treated for relief of symptomatic BPH by the age of 80 years.‘,* Because of aging, the number of patients eligible for treatment in the United States is expected to increase fivefold from 450,000 in 1990 to 2.1 million by 1995.3 Transurethral resection of prostate (TURP) is the
UROLOGY@ I
FEBRUARY 1995
1 VOLUME45,
NLMBER2
most common modality used to treat BPH. TURP however, is also one of the most expensive health care items, costing about $4 billion annually4 The high morbidity (16%), mortality (2.01%), and cost attributed to the procedure5-’ have resulted in an intense search for alternative treatments. Among medical agents, two that have been approved for treatment of BPH in the United States are the alpha-blocker terazosin and the 5-alpha-reductase inhibitor finasteride. Finasteride (Proscar, Merck and Co., Inc., West Point, Pa) works by inhibiting conversion of testosterone to dihydrotestosterone (DHT). By its action, finasteride causes a decrease in intraprostatic and serum DHT, which eventually results in atrophy of prostate cells. Both epithelial and stromal cells are affected, and after 12 months of treatment, there is an approximately 20% reduction in total prostate volume.8 Results of therapy however, have not correlated with total prostate volume measurements, and there are patients with improvement in urinary flow rates and symptom relief who have not had any significant reduction in total prostate volume, and vice versa.9 Part of the reason for this may be that total prostate volume is a poor parameter to measure response to finasteride therapy Enlargement due to BPH selectively affects only the transition zone (TZ) of the prostate, and changes in TZ volume may be a more accurate indicator of response to finasteride. A careful analysis of the published literature revealed that only one report, by Tempany and associates,1° specifically analyzed differential zonal volumes and their changes after finasteride. These authors concluded that, indeed, the TZ volume was more affected by finasteride than peripheral zone volume. A drawback of this study, however, was that the authors made no attempt to correlate TZ volume changes with urinary flow rates. In this report we analyzed the effects of finasteride on TZ volume, TZ ratio (TZ volumejtotal prostate volume), and total prostate volume. Ourfindings suggest that TZ volume and TZ ratio are more important than total prostate volume in assessment of uroflow response to finasteride. Our findings also suggest that an individuals TZ ratio at baseline may provide valuable information on potential for therapeutic response to finasteride. Since there is a time lag of 3 to 6 months for efficacy of finasteride to be clinically apparent, measurement of TZ ratio may be cost-effective in BPH management. MATERIAL AND METHODS The clinical material consisted of 36 patients who were part of a double-blind, placebo-controlled study of long-term efficacy of finasteride in men
UROLOGY@ / FEBRUARY 1995 I
VOLIWE 45, NUMBER2
with symptoms of BPH. Symptoms and signs of BPH were assessed by a modified Boyarski symptom score, uroflow measurements (peak flow rate less than 15 cc/s>, and residual urine determinations (upper limits of which was 300 cc). Additional inclusion criteria were serum creatinine less than 2 mg/dL, normal liver function test, absence of prostate cancer as judged by a normal digital rectal examination (DRE), prostate-specific antigen (PSA) less than 40 ng/mL (considered reasonable when study was begun in 19891, and a normal transrectal ultrasound (TRUS). Thirty-six patients started the study and 30 patients completed the 24month follow-up period. Patients were randomized to receive finasteride 5 mg/d (n = lo), 1 mg/d (n = lo), or placebo (n = 10). At 12 months, the study was converted to open label, and all patients were switched to 5 mg/d of finasteride. Six patients opted out of the study for various reasons, as discussed in Results section. In 23 patients complete data were available on total and TZ volumes. Symptom score, uroflow, residual urine, serum chemistry, PSA, and prostatic acid phosphatase estimations were performed at the screening visit and at 3-month intervals. TRUS and magnetic resonance imaging (MRI) were performed at baseline and every 6 months thereafter. All TRUS examinations were performed by the same examiner (K.S.) using a biplanar mechanical sector scanner with variable frequency of 5.0 MHz, 6.0 MHz, and 7.5 MHz (Siemens SI-200). Images were obtained with the patient in the left lateral decubitus position. A lubricated transrectal probe was inserted into the rectum and a coupling balloon was inflated with 40 cc of water while in the rectum. Transition zone was marked during the study and multiple transverse and sagittal sections were recorded on videotape and hard copies. The transverse and anteroposterior diameters of total prostate and TZ at the largest cross-sectional area were used for calculations. Superoinferior diameter of the prostate was measured on the midline sagittal images and that of the TZ was measured at the point of its largest diameter on the sagittal image. The transverse plane in this method was almost perpendicular to the posterior surface of the prostate gland (Figs. 1 and 2). Total and TZ prostate volume were calculated using the formula for the volume estimation of an ellipsoid: [volume = 0.52 X transverse diameter X anteroposterior diameter X superoinferior diameter]. _ Total volume, TZ volume, and TZ ratio were recorded for each patient at baseline and at 12 months after finasteride therapy (5 mg/d).
259
FIGURE 1. Transition zone ratio changes on finasteride therapy.
3 8*r. 0.40
-H-
RESPONDERS
-.-
NONRESPONDERS 0.35 I OMONTH
12 MONTHS n&In
m lmo
between improvement in peak urinary flow rates and symptom score and the various parameters. We further calculated relative risk of peak flow improvement based on baseline TZ ratio. The statistical software used in this study was Statview (Abascus Concepts) and PC SAS (SAS, Cary, NC). RESULTS
0.0 + -100
0
~lMPR0vEMENT
100
2
INPEAKFwwRATE5
FIGURE 2. Nomogram showing a direct correlation between baseline transition zone ratio and peak urinary flow rates at 12 months.
Similarly, peak urinary flow rates were recorded at baseline and at 12 months. Patients were then grouped into responders (having improvement of peak urinary flow rate more than 3 cc/s) and nonresponders (peak urinary flow rate improvement less than 3 c~/s).~ To evaluate if there was a difference in reduction of total versus TZ volume and to determine if TZ ratio would be a useful measure to predict uroflow response, data were analyzed among responders and nonresponders by unpaired t and Mann-Whitney tests. To determine which, among the three variables of total volume, TZ volume, and TZ ratio, would best correlate with changes in peak urinary flow rates and symptom score, data were analyzed by Spearman rank correlation and Pearson correlation. This was calculated in each patient by analysis of variance
260
Thirty patients completed follow-up of 24 months. Six patients dropped out of the study for the following reasons: 2 patients had difficulty with traveling to the medical center at the required intervals, 1 developed prostate cancer, 1 had a cerebral stroke (unrelated to medications), and 2 others who did not respond to the medications sought other therapy In 7 patients TRUS data were not obtained due to scheduling problems. Data are presented here on 23 patients who received finasteride 5 mg/d for at least 12 months and in whom complete data on the total and TZ volume of the prostate were available. The baseline total and TZ mean volumes were 40.1 and 20.7 cc, respectively, which were reduced to a mean of 36.9 and 15.0 cc, respectively, at 24 months. The mean reduction in total prostate volume and TZ volume was 3.2 (8%) and 5.6 cc (27%), respectively (P ~0.03) (Table I). In comparison to the baseline values, there was a statistically significant increase in peak urinary flow rates at 12 months (mean increase, 3.2 cc/s; 32%; P ~0.0001) (Table II). The mean modified Boyarsky symptom score at baseline was 19.4, and it was reduced significantly at 12 months (mean reduction, 3.1; 32.5%; P
TABLE
I.
Changes in prostatic volume Difference
Group *
No. (%)
All patients
23
Responders
14 (61) 9
Nonresponders
12-Month Volume (cc)
Baseline Volume (cc) Total Volume TZ Volume Mean (SD) Mean (SD)
139)
40.1 (22.1) 42.4 (25.8) 36.7 (15.4)
Total Volume
Total Volume Mean (SD)
TZ Volume Mean (SD)
36.9 (18.2) 3.61 (19.61 38.2 (16.7)
15.0 (1 1.5) 13.4 (1 1.61 17.7 (1 1 .8)
20.7 (20.6) 24.2 (1 1.6) 15.2 (8.2)
TZ Volume
Mean A (cc) 3.2
Reduction (%) 8
Mean A
6.2
15
10.9
-4
-2.4
-1.6
5.6
Reduction (%) 27 45 P = 0.03+ -16
*Responder:peakflow improvement more than 3 mUs; tnonresponder; peakflow improvements less than 3 mUs; +Reductionin transition zone (TZJ volume: responder versus nonresponder. KEY: TZ = transition zone; SD = standard deviation; A = change; - = increase in volume.
TABLE
II.
Changes in peak flow rates
Baseline Peak Flow (mus) Mean SD
Group
12-Month Peak Flow (mus) Mean SD
All patients
10.1
3.2
13.3
3.4
Responders
9.9
30
14.2
3.4
10.6
3.5
11.9
3.0
Nonresponders
Differences in Peak Flow Mean A 3.2 (P <0.0001) 4.4 (P <0.0001)
1.3
%A 32 45 12
(P <0.0001) KEY:SD = standard deviation; A = change.
TABLE
III.
Changes in transition zone ratio volume
Baseline TZR
12-Month TZR
Group
Mean
SD
Mean
SD
All patients Responders Nonresponders
0.45 0.48 0.41
0.18 0.19 0.14
0.39 0.36 0.43
0.19 0.19 0.21
Differences in TZR* Mean A 0.06 0.12 -0.02
%A
P Value
14 25 -5
0.23 0.04 0.32
*P < 0.0235for percent change (A) in TZR in responders versus nonresponders. KEY: TZR = transitional zone volume/total prostate volume; A = chan,ee;- = increase in TZR
TOTALAND TRANSITIONZONEVOLUME IN UROFLOW RESPONDERS ANDNONRESPONDERS Based on the improvement in peak urinary flow rates, the patients were grouped in one of two categories*: responders were those patients who had a peak flow rate improvement of more than 3 ccjs (n = 14; 61%); nonresponders had a peak flow rate improvement of less than 3 cc/s (n = 9; 39.2%) (Tables I, II, and III). The mean improvement in peak urinary flow rate was 4.4 cc/s (48.7%) in the responders and 1.3 cc/s (27.0%) in nonresponders (P
UROLOGY@
/
FEBRUARY 1995I VOLUME 45, NUMBER 2
in total prostate volume was statistically insignificant among the two groups. TRANSITIONZONE RATIOIN UROFLOWRESPONDERS ANDNONRESPONDERS Mean TZ ratio was reduced significantly (25%) in the responders, whereas it increased (5%) in nonresponders (P ~0.02) (Fig. 1).
CORRELATION OF PEAKURINARY FLOWRATE
WITH
CHANGESIN VARIOUSVOLUMESAND TRANSITION
ZONE RATIO
Statistically significant correlations between peak urinary flow rate and baseline TZ ratio (r = 0.52; P
261
found even when all patients were considered together rather than being grouped as responders and nonresponders. However, there was no similar correlation between peak urinary flow rates and total prostate volume. Figure 2 shows a nomogram depicting correlation of percentage improvement in peak urinary flow rate with baseline TZ ratio when all patients were considered together. RELATIVE RISKOF UROFLOWIMPROVEMENT BASEDON INITIALTRANSITION ZONE RATIO
We further calculated relative risk of improvement based on initial TZ ratio. Any patient having a TZ ratio less than 0.51 had only a 38% chance of improvement (relative risk = 0.38), whereas patients having a TZ ratio greater than 0.51 had a 100% chance (relative risk = 2.67) of improving peak urinary flow rate following finasteride therapy (P ~0.005). CORRELATION OF SYMPTOMSCOREIMPROVEMENT WITH VARIOUS PARAMETERS No statistically significant correlation was noted between the modified Boyarsky symptom score and total prostate volume (r = 0.08; P = 0.719), TZ volume (r = 0.07; P = 0.755), and TZ ratio (r = 0.280; P = 0.198). Additionally, no correlation was noted with peak urinary flow rates and symptom score (r = 0.355; P = 0.0965), and there was no statistically significant correlation and predictive value of baseline TZ ratio (r = 0.94; P = 0.378) for improvement in symptom score.
COMMENT The major findings of this study were: (1) patients who had significant (more than 3 mVs> improvement in peak urinary flow rate had substantial reduction in TZ volume and TZ ratio, whereas patients who had less than 3 cc/s improvement in peak urinary flow rate had insignificant reduction or an actual increase in TZ volume and TZ ratio (P co.03 for TZ volume and P co.02 for TZ ratio); (2) irrespective of outcome after finasteride therapy (significant or insignificant reduction in peak urinary flow rate), all patients had a significant direct correlation between reduction in TZ volume (r = 0.500; P ~0.03) and TZ ratio (r = 0.60, P ~0.007) with improvement in peak urinary flow rates. No similar correlation was seen with changes in total prostate volume; (3) it was also noted that pretreatment calculation of the TZ ratio helped to predict improvement in peak urinary flow rates (r = 0.52; P ~0.01). As seen in Table III, mean TZ ratio was higher among responders, and it was reduced by a mean of 25% after therapy compared
262
with a mean increase of 5% in nonresponders. Furthermore, patients having TZ ratio more than 0.51 prior to therapy had a 100% chance of improving peak urinary flow rates following finasteride therapy (relative risk = 2.67; P ~0.005) compared with only a 38% chance in patients with pretreatment TZ ratio less than 0.51. However, there was a total lack of correlation between symptom score with peak urinary flow rates and various prostate volume measurements, and calculation of TZ ratio did not help in predicting symptomatic response to finasteride therapy. The mechanism of BPH volume diminution after finasteride is based on inhibition of S-alphareductase. Finasteride is a 4-azasteroid inhibitor of 5-alpha-reductase, which catalyzes conversion of testosterone to DHT. DHT is the major androgen utilized by the prostate. By blocking DHT formation, finasteride causes prostate cells to die and eventually this results in a decreased prostate size. In clinical trials, finasteride has been noted to cause a 17% to 23% reduction in total prostate volume. lo211 However, there has been no correlation between improvement in peak urinary flow rates and reduction in total prostate volume.9 Furthermore, the prostate volume reduction after finasteride takes several months to occur. One of the challenges facing the clinician is development of parameters to predict finasteride efficacy prior to starting therapy Since BPH predominantly affects the TZ, a logical question arises as to whether finasteride has any selective effect on the TZ and, secondly, whether measurement of TZ volume can predict response to finasteride therapy Among published data, we could find only one study in which Tempany and associates,” using MRI in 20 patients demonstrated that following 12 months of finasteride therapy (1 and 5 mg/d), there was a significant reduction in TZ volume (6.2 + 3 cc) rather than peripheral zone volume. These investigators, however, made no attempt to correlate TZ volume changes with improvement in peak urinary flow rates. Our study clearly demonstrates that there is a significant difference in TZ volume reduction among responders when compared with nonresponders after finasteride therapy (45% versus 6%, P ~0.0298). It is known that the symptomatic improvement in BPH does not always correlate with uroflow because of the complexity and subjective nature of BPH symptoms. On the other hand, there is no satisfactory explanation as to why uroflow does not correlate with prostate volume changes in all patients who are treated by agents that reduce prostate volume. Heretofore, it was thought that factors other than prostate volume, such as bladder
U ROLOCY@ / FEBRUARY 1995 I
VOLUME45, NUMBER2
FIGURE 3. (A) Transrectal ultrasound (TRW) image depicting small transition prostate. (B) TRUS image depicting greater TZ volume in a 30 cc prostate.
tone and compliance, were responsible for this lack of correlation. Our observations, however, suggest that when TZ volume is measured, there is a direct relationship between volume reduction and uroflow improvement. However, this would not be observed if total prostate volumes were measured. In this study, measurement of total prostate volume had no significant correlation with improvement in peak urinary flow rate. This was not surprising, since mechanical obstruction to the prostate urethra in BPH occurs due to the TZ mass, and changes in TZ volume may not be reflected in total prostate volume. In the large multicenter trial of finasteride therapy for BPH, it was found that almost 63% of patients developed more than 20% reduction in total prostate volume but only 40% developed more than 3 cc/s improvement in peak flow rates.8 Similar lack of correlation between flow rates and prostate volume was found in a trial of BPH patients treated with buserelin (luteinizing hormone-releasing hormone agonist) and cyproterone acetate (antiandrogen). l2 In these patients, the total prostate volume decreased by 29%, but there was no correlation between volume reduction and improvement in peak urinary flow rates. The mechanism for lack of total prostate volume reduction with antiandrogen and finasteride therapies may be related to expansion of the compressed peripheral zone as the shrinkage of TZ occurs.1o Total prostate volume measurements can be misleading in patients with BPH. For example, two prostate glands may be the same size but may have substantially different TZ volumes (Fig. 3). Considering these factors, one can appreciate that UROLOGY@ / FEBRUARY 1995 I VOLUME45, NUMBER2
zone (TZ) volume in a 40 cc
even in the absence of reduction of total volume the obstructive effect of TZ on the prostatic urethra can be relieved and thus the peak flow rate can improve. This, in fact, has been the working hypothesis for finasteride efficacy despite lack of correlation with total prostate volume changes. Another finding of this study was the significant correlation of peak urinary flow rate with TZ ratio in patients with untreated BPH (r = 0.60; P >0.006). TZ ratio may also provide relative risk data for assessment of improvement following finasteride therapy Patients having initial TZ ratios more than 0.51 had a 25fold greater chance of improving peak urinary flow following finasteride therapy than those having TZ ratio less than 0.51 (100% versus 38%). Although calculation of TZ volume takes into account the various sizes of the prostate, it does not take into account the fact that TZ volume in some patients may actually increase while on therapy with finasteride. Heretofore, this has been an unrecognized fact in patients undergoing therapy with finasteride. The explanation for this may be that the rate of BPH cell growth in some patients is greater than the rate of cell death due to finasteride. Calculation of TZ ratio takes into account both changes in TZ volume and expansion of the peripheral zone, thereby making TZ ratio a more sensitive parameter both to select patients for finasteride therapy and to predict degree of response. As can be seen in Tables I and III, the TZ volume and ratio actually increase in patients not responding to therapy at 12 months. The use of more than 3 cc/s improvement as a measure for response to finasteride therapy has 263
caused some concerns because a patient improving from 4 to 7 cc/s peak urinary flow rate may not be equivalent to a patient improving from 12 to 15 cc/s. Use of a nomogram (Fig. 2) with both TZ ratio and peak flow rate may be more beneficial in determining response to therapy. Even though TZ volume and TZ ratio correlated significantly with changes in peak urinary flow rates, we did not find similar correlations with symptom score. This lack of correlation between symptoms and urodynamic findings has been noted in several prior studies.13-15 Lower urinary voiding symptoms in men in the age groups in which BPH occurs are commonly referred to as “prostatism.” The term is a misnomer because in a significant number of patients the source of these symptoms is not the prostate, and similar symptoms have been noted in women of comparable age groups, who obviously do not have prostates.13v15 Various poorly understood vesical, neural, prostatic, and urethral factors are responsible for a patient’s perception of this symptom complex. Invasive urodynamic tests (synchronous pressure flow studies) are required to differentiate true prostatic obstruction along with associated detrusor instability or underactivity. Detrusor instability has been noted in 50% and hypotonicity in 20% of older men.13 We did not perform invasive urodynamic studies in these patients and that may explain lack of correlation between symptom score with peak urinary flow rates and various volume parameters. CONCLUSION In this report we provide simple measurable parameters that may be used prior to treatment to predict efficacy of finasteride and similar agents in improving urinary flow rates. Estimation of TZ volume and calculation of TZ ratio may be the preferred method to monitor uroflow response to finasteride therapy in BPH. Primary reduction in prostate volume occurs in the TZ, and changes in this zone have more pronounced effect on the relief of prostate urethral compression. Measurement of TZ ratio will take into account variations in proportions of TZ versus peripheral zones and thus simplify the monitoring of BPH patients on finasteride therapy. On the basis of this study, we believe that potential BPH patients being considered for finasteride therapy should undergo estimation of TZ volume and TZ ratio. This will help in identifying uroflow responders and nonresponders to finasteride therapy.
264
Perinchery Narayan, M.D. University ofFlorida Division of Urology P.O. Box 100247 Gainesville. FL 32610-0247 ACKNOWLEDGMENT.To Anitha Toke, M.D., for her assistance in collection of data; Dia Hertz and Susan Goldfien for statistical analysis; and Elaine Patarini for her editorial assistance.
REFERENCES 1. Barry MJ: Medical outcomes research and benign prostatic hyperplasia. Prostate (Suppl) 3: 61-74, 1990. 2. Barry MJ: Physicians: taking the lead. Health Manage Q 13: 20-23, 1991. 3. McCarthy ME, and Willard M: Health care prostate therapies. Shearson, Lehman, and Hutton, Jun 5 1990, pp 1-12. 4. Holtgrewe HL, Mebust WK, Dowd JB, Cockett AT, Peters PC, and Proctor C: Transurethral prostatectomy: practice aspects of the dominant operation in American urology. J Urol 141: 248-253, 1989. 5. Mebust WK, Holtgrewe HL, Cockett AT, and Peters PC: Transurethral prostatectomy: immediate and postoperative complications. A cooperative study of 13 participating institutions evaluating 3,885 patients. J Urol 141: 243-247, 1989. 6. Lepor H, and Rigaud G: The efficacy of transurethral resection of the prostate in men with moderate symptoms of prostatism. J Urol 143: 533-537, 1990. 7. McConnell JD. Barry MJ, and Bruskewitz RC: Benign prostatic hyperplasia: diagnosis and treatment. Agency for Health Care Policy and Research. Clin Pratt Guidl Quick Ref Guide Clin 8: 1-17, 1994. 8. Stoner E: Three-year safety and efficacy data on the use of finasteride in the treatment of benign prostatic hyperplasia. Urology 43: 284-294, 1994. 9. Lepor H. Editorial comment to: Three-year safety and efficacy data on the use of finasteride in the treatment of benign prostatic hyperplasia. Stoner E and the Members of the Finasteride Study Group. Urology 43: 292-293, 1994. 10. Tempany CM, Partin AW, Zerhouni EA. Zinreich SJ, and Walsh PC: The influence of finasteride on the volume of the peripheral and periurethral zones of the prostate in men with benign prostatic hyperplasia. Prostate 22: 39-42, 1993. 11. Gormley GJ, Stoner E, Bruskewitz RC, ImperatoMcGinley J, Walsh PC, McConnell JD, Andriole GL, Geller J, Brachen BR, Tenover JS, et al: The effect of finasteride in men with benign prostatic hyperplasia. The Finasteride Study Group. N Engl J Med 327: 1185-1191, 1992. 12. Bosch RJ, Griffiths DJ, Blom JH, and Schroeder FH: Treatment of benign prostatic hyperplasia by androgen deprivation: effects on prostate size and urodynamic parameters. J Urol 141: 68-72, 1989. 13. Abrams P: In support of pressure-flow studies for evaluating men with lower urinary tract symptoms. Urology 44: 153-155, 1994. 14. Lepor H, and Machi G: Comparison of the AUA symptom index in unselected males and females between fiftyfive and seventy-nine years of age. Urology 42: 36-40, 1993. 15. Blaivas JG: Multichannel urodynamic studies in men with benign prostatic hyperplasia. Indications and interpretation. Ural Clin North Am 13: 543-551, 1990.
UROLOGY@
I
FEBRUARY 1995 I VOLUME45, NUMBER2
EDITORIAL COMMENT Urologists have long placed considerable importance on overall prostate size when deciding when to intervene in benign prostatic hyperplasia (BPH). Numerous publications have indicated that overall prostate size, although a factor for determining which type of treatment is appropriate, is not helpful in establishing the presence of obstruction or the need to treat. One of finasteride’s attributes is its ability to reduce prostate volume over time, but even the large North American trial that was instrumental in the approval of finasteride did not identify size as a predictor of response to the drug. The authors indicate that it may be the transitional zone volume (as determined by transrectal ultrasound) that holds the key to predicting which patients will respond to finasteride. They suggest that if this is confirmed in further studies, transrectal ultrasound may be an appropriate maneuver for urologists to use prior to initiating the trial of a drug that may take 3 months, 6 months, or longer to manifest its benefits. They suggest that the number of patients eligible for medical management may increase fivefold in the short run because of aging of the population. Although aging alone cannot be expected to make this large an impact on the number of men eligible for BPH management, attitudes about what level of symptoms and bother from urinary symptoms constitute a level in need of treatment will. The authors identify a correlation with transition zone volume and urine flow rate but not with urinary symptoms. In the future most men receiving treatment for BPH will do so under the direction of their primary care physician. Uroflow meters
UROLOGY@ /FEBRUARY 1995 I VOLUME45, NUMBER2
and transrectal ultrasound transducers are not readily available to these clinicians. Urologists should give their opinions regarding what level of symptoms and discomfort related to BPH are appropriate for initiation of therapy. In my opinion, minimal bother or minimal symptoms are not. If patients believe their symptoms are bothersome enough to warrant medical intervention, then a medical therapeutic trial is warranted irrespective of uroflow and transrectal ultrasound considerations. Reginald C. Bruskewitz, M.D. Centerfor Health Sciences University of Wisconsin-Madison Madison, WI53792 REPLY BY AUTHORS Although we agree the benign prostatic hyperplasia should be treated based on symptoms and that many patients will be receiving treatment from their primary care physicians, it is our contention that a significant number of such patients will have a transrectal ultrasound at some point due to considerations such as elevated prostate-specific antigen or prostate size. It would be a simple task in these patients to calculate a TZ ratio to allow the physician a means to prognosticate which type of medical agent may be beneficial. Perinchery
Narayan,
M.D.
Gainesville, Fla.
265
ABSTRACT-Objectives. The aim of this study was to: (1) determine effects of finasteride on transition zone (TZ) volume, TZ ratio (TZ volume/total prostate volume), and total prostate volume; (2) analyze differences in TZ and total volume reduction among patients who improve peak urinary flow rates following finasteride therapy with those who do not; (3) investigate which parameters correlate with improvement in peak urinary flow rate and urinary symptom score; and (4) establish if there is any predictive value of these parameters for response to therapy. Methods. Twenty-three patients with symptomatic benign prostatic hyperplasia (BPH) were treated with finasteride (5 mg/d) for 12 months and underwent transrectal ultrasound (TRUS) evaluation of total and TZ volume of prostate and measurement of peak flow rate and modified Boyarsky symptom score at baseline and at 12 months. Statistical analysis was done by unpaired t, Mann-Whitney, and Spearman rank correlation tests among responders (more than 3 cc/s improvement in peak flow rate) and nonresponders (less than 3 cc/s improvement in peak flow rate) to therapy. Results. (1) Responders had substantial reduction in TZ volume (44.8% versus 16.05%; PC 0.03) and TZ ratio (25% versus 5% increase, P-c 0.02) compared with nonresponders. (2) There was a significant correlation between reduction in TZ volume (r = 0.50; PC 0.03) and TZ ratio (r = 0.60; PC 0.006) with improvement in peak flow rates. No similar correlation was seetr with total prostate volume changes. (3) Pretreatment TZ ratio helped in predicting peak flow improvement following finasteride therapy (r = 0.52; P-c 0.01) and there was a 2.5-fold increased chance of improvement if baseline TZ ratio was more than 0.51. The modified Boyarsky symptom score decreased by 3.1 (mean), but there was no correlation with changes in peak urinary flow rate, total prostate volume, TZ volume, and TZ ratio. TZ ratio did not have significant predictive value for improvement in symptom score. Conclusions. This study provides data that simple measurable parameters are available which may be used prior to therapy to predict uroflow response to finasteride and similar agents.
Benign prostatic hyperplasia (BPH) is the most common neoplasm in the aging human male. It is estimated that 1 in every 4 men in the United Submitted (Rapid Communication): (with revisions): October 18, 1994
258
July 22, 1994, accepted
States will be treated for relief of symptomatic BPH by the age of 80 years.‘,* Because of aging, the number of patients eligible for treatment in the United States is expected to increase fivefold from 450,000 in 1990 to 2.1 million by 1995.3 Transurethral resection of prostate (TURP) is the
UROLOGY@ I
FEBRUARY 1995
1 VOLUME45,
NLMBER2
most common modality used to treat BPH. TURP however, is also one of the most expensive health care items, costing about $4 billion annually4 The high morbidity (16%), mortality (2.01%), and cost attributed to the procedure5-’ have resulted in an intense search for alternative treatments. Among medical agents, two that have been approved for treatment of BPH in the United States are the alpha-blocker terazosin and the 5-alpha-reductase inhibitor finasteride. Finasteride (Proscar, Merck and Co., Inc., West Point, Pa) works by inhibiting conversion of testosterone to dihydrotestosterone (DHT). By its action, finasteride causes a decrease in intraprostatic and serum DHT, which eventually results in atrophy of prostate cells. Both epithelial and stromal cells are affected, and after 12 months of treatment, there is an approximately 20% reduction in total prostate volume.8 Results of therapy however, have not correlated with total prostate volume measurements, and there are patients with improvement in urinary flow rates and symptom relief who have not had any significant reduction in total prostate volume, and vice versa.9 Part of the reason for this may be that total prostate volume is a poor parameter to measure response to finasteride therapy Enlargement due to BPH selectively affects only the transition zone (TZ) of the prostate, and changes in TZ volume may be a more accurate indicator of response to finasteride. A careful analysis of the published literature revealed that only one report, by Tempany and associates,1° specifically analyzed differential zonal volumes and their changes after finasteride. These authors concluded that, indeed, the TZ volume was more affected by finasteride than peripheral zone volume. A drawback of this study, however, was that the authors made no attempt to correlate TZ volume changes with urinary flow rates. In this report we analyzed the effects of finasteride on TZ volume, TZ ratio (TZ volumejtotal prostate volume), and total prostate volume. Ourfindings suggest that TZ volume and TZ ratio are more important than total prostate volume in assessment of uroflow response to finasteride. Our findings also suggest that an individuals TZ ratio at baseline may provide valuable information on potential for therapeutic response to finasteride. Since there is a time lag of 3 to 6 months for efficacy of finasteride to be clinically apparent, measurement of TZ ratio may be cost-effective in BPH management. MATERIAL AND METHODS The clinical material consisted of 36 patients who were part of a double-blind, placebo-controlled study of long-term efficacy of finasteride in men
UROLOGY@ / FEBRUARY 1995 I
VOLIWE 45, NUMBER2
with symptoms of BPH. Symptoms and signs of BPH were assessed by a modified Boyarski symptom score, uroflow measurements (peak flow rate less than 15 cc/s>, and residual urine determinations (upper limits of which was 300 cc). Additional inclusion criteria were serum creatinine less than 2 mg/dL, normal liver function test, absence of prostate cancer as judged by a normal digital rectal examination (DRE), prostate-specific antigen (PSA) less than 40 ng/mL (considered reasonable when study was begun in 19891, and a normal transrectal ultrasound (TRUS). Thirty-six patients started the study and 30 patients completed the 24month follow-up period. Patients were randomized to receive finasteride 5 mg/d (n = lo), 1 mg/d (n = lo), or placebo (n = 10). At 12 months, the study was converted to open label, and all patients were switched to 5 mg/d of finasteride. Six patients opted out of the study for various reasons, as discussed in Results section. In 23 patients complete data were available on total and TZ volumes. Symptom score, uroflow, residual urine, serum chemistry, PSA, and prostatic acid phosphatase estimations were performed at the screening visit and at 3-month intervals. TRUS and magnetic resonance imaging (MRI) were performed at baseline and every 6 months thereafter. All TRUS examinations were performed by the same examiner (K.S.) using a biplanar mechanical sector scanner with variable frequency of 5.0 MHz, 6.0 MHz, and 7.5 MHz (Siemens SI-200). Images were obtained with the patient in the left lateral decubitus position. A lubricated transrectal probe was inserted into the rectum and a coupling balloon was inflated with 40 cc of water while in the rectum. Transition zone was marked during the study and multiple transverse and sagittal sections were recorded on videotape and hard copies. The transverse and anteroposterior diameters of total prostate and TZ at the largest cross-sectional area were used for calculations. Superoinferior diameter of the prostate was measured on the midline sagittal images and that of the TZ was measured at the point of its largest diameter on the sagittal image. The transverse plane in this method was almost perpendicular to the posterior surface of the prostate gland (Figs. 1 and 2). Total and TZ prostate volume were calculated using the formula for the volume estimation of an ellipsoid: [volume = 0.52 X transverse diameter X anteroposterior diameter X superoinferior diameter]. _ Total volume, TZ volume, and TZ ratio were recorded for each patient at baseline and at 12 months after finasteride therapy (5 mg/d).
259
FIGURE 1. Transition zone ratio changes on finasteride therapy.
3 8*r. 0.40
-H-
RESPONDERS
-.-
NONRESPONDERS 0.35 I OMONTH
12 MONTHS n&In
m lmo
between improvement in peak urinary flow rates and symptom score and the various parameters. We further calculated relative risk of peak flow improvement based on baseline TZ ratio. The statistical software used in this study was Statview (Abascus Concepts) and PC SAS (SAS, Cary, NC). RESULTS
0.0 + -100
0
~lMPR0vEMENT
100
2
INPEAKFwwRATE5
FIGURE 2. Nomogram showing a direct correlation between baseline transition zone ratio and peak urinary flow rates at 12 months.
Similarly, peak urinary flow rates were recorded at baseline and at 12 months. Patients were then grouped into responders (having improvement of peak urinary flow rate more than 3 cc/s) and nonresponders (peak urinary flow rate improvement less than 3 c~/s).~ To evaluate if there was a difference in reduction of total versus TZ volume and to determine if TZ ratio would be a useful measure to predict uroflow response, data were analyzed among responders and nonresponders by unpaired t and Mann-Whitney tests. To determine which, among the three variables of total volume, TZ volume, and TZ ratio, would best correlate with changes in peak urinary flow rates and symptom score, data were analyzed by Spearman rank correlation and Pearson correlation. This was calculated in each patient by analysis of variance
260
Thirty patients completed follow-up of 24 months. Six patients dropped out of the study for the following reasons: 2 patients had difficulty with traveling to the medical center at the required intervals, 1 developed prostate cancer, 1 had a cerebral stroke (unrelated to medications), and 2 others who did not respond to the medications sought other therapy In 7 patients TRUS data were not obtained due to scheduling problems. Data are presented here on 23 patients who received finasteride 5 mg/d for at least 12 months and in whom complete data on the total and TZ volume of the prostate were available. The baseline total and TZ mean volumes were 40.1 and 20.7 cc, respectively, which were reduced to a mean of 36.9 and 15.0 cc, respectively, at 24 months. The mean reduction in total prostate volume and TZ volume was 3.2 (8%) and 5.6 cc (27%), respectively (P ~0.03) (Table I). In comparison to the baseline values, there was a statistically significant increase in peak urinary flow rates at 12 months (mean increase, 3.2 cc/s; 32%; P ~0.0001) (Table II). The mean modified Boyarsky symptom score at baseline was 19.4, and it was reduced significantly at 12 months (mean reduction, 3.1; 32.5%; P
TABLE
I.
Changes in prostatic volume Difference
Group *
No. (%)
All patients
23
Responders
14 (61) 9
Nonresponders
12-Month Volume (cc)
Baseline Volume (cc) Total Volume TZ Volume Mean (SD) Mean (SD)
139)
40.1 (22.1) 42.4 (25.8) 36.7 (15.4)
Total Volume
Total Volume Mean (SD)
TZ Volume Mean (SD)
36.9 (18.2) 3.61 (19.61 38.2 (16.7)
15.0 (1 1.5) 13.4 (1 1.61 17.7 (1 1 .8)
20.7 (20.6) 24.2 (1 1.6) 15.2 (8.2)
TZ Volume
Mean A (cc) 3.2
Reduction (%) 8
Mean A
6.2
15
10.9
-4
-2.4
-1.6
5.6
Reduction (%) 27 45 P = 0.03+ -16
*Responder:peakflow improvement more than 3 mUs; tnonresponder; peakflow improvements less than 3 mUs; +Reductionin transition zone (TZJ volume: responder versus nonresponder. KEY: TZ = transition zone; SD = standard deviation; A = change; - = increase in volume.
TABLE
II.
Changes in peak flow rates
Baseline Peak Flow (mus) Mean SD
Group
12-Month Peak Flow (mus) Mean SD
All patients
10.1
3.2
13.3
3.4
Responders
9.9
30
14.2
3.4
10.6
3.5
11.9
3.0
Nonresponders
Differences in Peak Flow Mean A 3.2 (P <0.0001) 4.4 (P <0.0001)
1.3
%A 32 45 12
(P <0.0001) KEY:SD = standard deviation; A = change.
TABLE
III.
Changes in transition zone ratio volume
Baseline TZR
12-Month TZR
Group
Mean
SD
Mean
SD
All patients Responders Nonresponders
0.45 0.48 0.41
0.18 0.19 0.14
0.39 0.36 0.43
0.19 0.19 0.21
Differences in TZR* Mean A 0.06 0.12 -0.02
%A
P Value
14 25 -5
0.23 0.04 0.32
*P < 0.0235for percent change (A) in TZR in responders versus nonresponders. KEY: TZR = transitional zone volume/total prostate volume; A = chan,ee;- = increase in TZR
TOTALAND TRANSITIONZONEVOLUME IN UROFLOW RESPONDERS ANDNONRESPONDERS Based on the improvement in peak urinary flow rates, the patients were grouped in one of two categories*: responders were those patients who had a peak flow rate improvement of more than 3 ccjs (n = 14; 61%); nonresponders had a peak flow rate improvement of less than 3 cc/s (n = 9; 39.2%) (Tables I, II, and III). The mean improvement in peak urinary flow rate was 4.4 cc/s (48.7%) in the responders and 1.3 cc/s (27.0%) in nonresponders (P
UROLOGY@
/
FEBRUARY 1995I VOLUME 45, NUMBER 2
in total prostate volume was statistically insignificant among the two groups. TRANSITIONZONE RATIOIN UROFLOWRESPONDERS ANDNONRESPONDERS Mean TZ ratio was reduced significantly (25%) in the responders, whereas it increased (5%) in nonresponders (P ~0.02) (Fig. 1).
CORRELATION OF PEAKURINARY FLOWRATE
WITH
CHANGESIN VARIOUSVOLUMESAND TRANSITION
ZONE RATIO
Statistically significant correlations between peak urinary flow rate and baseline TZ ratio (r = 0.52; P
261
found even when all patients were considered together rather than being grouped as responders and nonresponders. However, there was no similar correlation between peak urinary flow rates and total prostate volume. Figure 2 shows a nomogram depicting correlation of percentage improvement in peak urinary flow rate with baseline TZ ratio when all patients were considered together. RELATIVE RISKOF UROFLOWIMPROVEMENT BASEDON INITIALTRANSITION ZONE RATIO
We further calculated relative risk of improvement based on initial TZ ratio. Any patient having a TZ ratio less than 0.51 had only a 38% chance of improvement (relative risk = 0.38), whereas patients having a TZ ratio greater than 0.51 had a 100% chance (relative risk = 2.67) of improving peak urinary flow rate following finasteride therapy (P ~0.005). CORRELATION OF SYMPTOMSCOREIMPROVEMENT WITH VARIOUS PARAMETERS No statistically significant correlation was noted between the modified Boyarsky symptom score and total prostate volume (r = 0.08; P = 0.719), TZ volume (r = 0.07; P = 0.755), and TZ ratio (r = 0.280; P = 0.198). Additionally, no correlation was noted with peak urinary flow rates and symptom score (r = 0.355; P = 0.0965), and there was no statistically significant correlation and predictive value of baseline TZ ratio (r = 0.94; P = 0.378) for improvement in symptom score.
COMMENT The major findings of this study were: (1) patients who had significant (more than 3 mVs> improvement in peak urinary flow rate had substantial reduction in TZ volume and TZ ratio, whereas patients who had less than 3 cc/s improvement in peak urinary flow rate had insignificant reduction or an actual increase in TZ volume and TZ ratio (P co.03 for TZ volume and P co.02 for TZ ratio); (2) irrespective of outcome after finasteride therapy (significant or insignificant reduction in peak urinary flow rate), all patients had a significant direct correlation between reduction in TZ volume (r = 0.500; P ~0.03) and TZ ratio (r = 0.60, P ~0.007) with improvement in peak urinary flow rates. No similar correlation was seen with changes in total prostate volume; (3) it was also noted that pretreatment calculation of the TZ ratio helped to predict improvement in peak urinary flow rates (r = 0.52; P ~0.01). As seen in Table III, mean TZ ratio was higher among responders, and it was reduced by a mean of 25% after therapy compared
262
with a mean increase of 5% in nonresponders. Furthermore, patients having TZ ratio more than 0.51 prior to therapy had a 100% chance of improving peak urinary flow rates following finasteride therapy (relative risk = 2.67; P ~0.005) compared with only a 38% chance in patients with pretreatment TZ ratio less than 0.51. However, there was a total lack of correlation between symptom score with peak urinary flow rates and various prostate volume measurements, and calculation of TZ ratio did not help in predicting symptomatic response to finasteride therapy. The mechanism of BPH volume diminution after finasteride is based on inhibition of S-alphareductase. Finasteride is a 4-azasteroid inhibitor of 5-alpha-reductase, which catalyzes conversion of testosterone to DHT. DHT is the major androgen utilized by the prostate. By blocking DHT formation, finasteride causes prostate cells to die and eventually this results in a decreased prostate size. In clinical trials, finasteride has been noted to cause a 17% to 23% reduction in total prostate volume. lo211 However, there has been no correlation between improvement in peak urinary flow rates and reduction in total prostate volume.9 Furthermore, the prostate volume reduction after finasteride takes several months to occur. One of the challenges facing the clinician is development of parameters to predict finasteride efficacy prior to starting therapy Since BPH predominantly affects the TZ, a logical question arises as to whether finasteride has any selective effect on the TZ and, secondly, whether measurement of TZ volume can predict response to finasteride therapy Among published data, we could find only one study in which Tempany and associates,” using MRI in 20 patients demonstrated that following 12 months of finasteride therapy (1 and 5 mg/d), there was a significant reduction in TZ volume (6.2 + 3 cc) rather than peripheral zone volume. These investigators, however, made no attempt to correlate TZ volume changes with improvement in peak urinary flow rates. Our study clearly demonstrates that there is a significant difference in TZ volume reduction among responders when compared with nonresponders after finasteride therapy (45% versus 6%, P ~0.0298). It is known that the symptomatic improvement in BPH does not always correlate with uroflow because of the complexity and subjective nature of BPH symptoms. On the other hand, there is no satisfactory explanation as to why uroflow does not correlate with prostate volume changes in all patients who are treated by agents that reduce prostate volume. Heretofore, it was thought that factors other than prostate volume, such as bladder
U ROLOCY@ / FEBRUARY 1995 I
VOLUME45, NUMBER2
FIGURE 3. (A) Transrectal ultrasound (TRW) image depicting small transition prostate. (B) TRUS image depicting greater TZ volume in a 30 cc prostate.
tone and compliance, were responsible for this lack of correlation. Our observations, however, suggest that when TZ volume is measured, there is a direct relationship between volume reduction and uroflow improvement. However, this would not be observed if total prostate volumes were measured. In this study, measurement of total prostate volume had no significant correlation with improvement in peak urinary flow rate. This was not surprising, since mechanical obstruction to the prostate urethra in BPH occurs due to the TZ mass, and changes in TZ volume may not be reflected in total prostate volume. In the large multicenter trial of finasteride therapy for BPH, it was found that almost 63% of patients developed more than 20% reduction in total prostate volume but only 40% developed more than 3 cc/s improvement in peak flow rates.8 Similar lack of correlation between flow rates and prostate volume was found in a trial of BPH patients treated with buserelin (luteinizing hormone-releasing hormone agonist) and cyproterone acetate (antiandrogen). l2 In these patients, the total prostate volume decreased by 29%, but there was no correlation between volume reduction and improvement in peak urinary flow rates. The mechanism for lack of total prostate volume reduction with antiandrogen and finasteride therapies may be related to expansion of the compressed peripheral zone as the shrinkage of TZ occurs.1o Total prostate volume measurements can be misleading in patients with BPH. For example, two prostate glands may be the same size but may have substantially different TZ volumes (Fig. 3). Considering these factors, one can appreciate that UROLOGY@ / FEBRUARY 1995 I VOLUME45, NUMBER2
zone (TZ) volume in a 40 cc
even in the absence of reduction of total volume the obstructive effect of TZ on the prostatic urethra can be relieved and thus the peak flow rate can improve. This, in fact, has been the working hypothesis for finasteride efficacy despite lack of correlation with total prostate volume changes. Another finding of this study was the significant correlation of peak urinary flow rate with TZ ratio in patients with untreated BPH (r = 0.60; P >0.006). TZ ratio may also provide relative risk data for assessment of improvement following finasteride therapy Patients having initial TZ ratios more than 0.51 had a 25fold greater chance of improving peak urinary flow following finasteride therapy than those having TZ ratio less than 0.51 (100% versus 38%). Although calculation of TZ volume takes into account the various sizes of the prostate, it does not take into account the fact that TZ volume in some patients may actually increase while on therapy with finasteride. Heretofore, this has been an unrecognized fact in patients undergoing therapy with finasteride. The explanation for this may be that the rate of BPH cell growth in some patients is greater than the rate of cell death due to finasteride. Calculation of TZ ratio takes into account both changes in TZ volume and expansion of the peripheral zone, thereby making TZ ratio a more sensitive parameter both to select patients for finasteride therapy and to predict degree of response. As can be seen in Tables I and III, the TZ volume and ratio actually increase in patients not responding to therapy at 12 months. The use of more than 3 cc/s improvement as a measure for response to finasteride therapy has 263
caused some concerns because a patient improving from 4 to 7 cc/s peak urinary flow rate may not be equivalent to a patient improving from 12 to 15 cc/s. Use of a nomogram (Fig. 2) with both TZ ratio and peak flow rate may be more beneficial in determining response to therapy. Even though TZ volume and TZ ratio correlated significantly with changes in peak urinary flow rates, we did not find similar correlations with symptom score. This lack of correlation between symptoms and urodynamic findings has been noted in several prior studies.13-15 Lower urinary voiding symptoms in men in the age groups in which BPH occurs are commonly referred to as “prostatism.” The term is a misnomer because in a significant number of patients the source of these symptoms is not the prostate, and similar symptoms have been noted in women of comparable age groups, who obviously do not have prostates.13v15 Various poorly understood vesical, neural, prostatic, and urethral factors are responsible for a patient’s perception of this symptom complex. Invasive urodynamic tests (synchronous pressure flow studies) are required to differentiate true prostatic obstruction along with associated detrusor instability or underactivity. Detrusor instability has been noted in 50% and hypotonicity in 20% of older men.13 We did not perform invasive urodynamic studies in these patients and that may explain lack of correlation between symptom score with peak urinary flow rates and various volume parameters. CONCLUSION In this report we provide simple measurable parameters that may be used prior to treatment to predict efficacy of finasteride and similar agents in improving urinary flow rates. Estimation of TZ volume and calculation of TZ ratio may be the preferred method to monitor uroflow response to finasteride therapy in BPH. Primary reduction in prostate volume occurs in the TZ, and changes in this zone have more pronounced effect on the relief of prostate urethral compression. Measurement of TZ ratio will take into account variations in proportions of TZ versus peripheral zones and thus simplify the monitoring of BPH patients on finasteride therapy. On the basis of this study, we believe that potential BPH patients being considered for finasteride therapy should undergo estimation of TZ volume and TZ ratio. This will help in identifying uroflow responders and nonresponders to finasteride therapy.
264
Perinchery Narayan, M.D. University ofFlorida Division of Urology P.O. Box 100247 Gainesville. FL 32610-0247 ACKNOWLEDGMENT.To Anitha Toke, M.D., for her assistance in collection of data; Dia Hertz and Susan Goldfien for statistical analysis; and Elaine Patarini for her editorial assistance.
REFERENCES 1. Barry MJ: Medical outcomes research and benign prostatic hyperplasia. Prostate (Suppl) 3: 61-74, 1990. 2. Barry MJ: Physicians: taking the lead. Health Manage Q 13: 20-23, 1991. 3. McCarthy ME, and Willard M: Health care prostate therapies. Shearson, Lehman, and Hutton, Jun 5 1990, pp 1-12. 4. Holtgrewe HL, Mebust WK, Dowd JB, Cockett AT, Peters PC, and Proctor C: Transurethral prostatectomy: practice aspects of the dominant operation in American urology. J Urol 141: 248-253, 1989. 5. Mebust WK, Holtgrewe HL, Cockett AT, and Peters PC: Transurethral prostatectomy: immediate and postoperative complications. A cooperative study of 13 participating institutions evaluating 3,885 patients. J Urol 141: 243-247, 1989. 6. Lepor H, and Rigaud G: The efficacy of transurethral resection of the prostate in men with moderate symptoms of prostatism. J Urol 143: 533-537, 1990. 7. McConnell JD. Barry MJ, and Bruskewitz RC: Benign prostatic hyperplasia: diagnosis and treatment. Agency for Health Care Policy and Research. Clin Pratt Guidl Quick Ref Guide Clin 8: 1-17, 1994. 8. Stoner E: Three-year safety and efficacy data on the use of finasteride in the treatment of benign prostatic hyperplasia. Urology 43: 284-294, 1994. 9. Lepor H. Editorial comment to: Three-year safety and efficacy data on the use of finasteride in the treatment of benign prostatic hyperplasia. Stoner E and the Members of the Finasteride Study Group. Urology 43: 292-293, 1994. 10. Tempany CM, Partin AW, Zerhouni EA. Zinreich SJ, and Walsh PC: The influence of finasteride on the volume of the peripheral and periurethral zones of the prostate in men with benign prostatic hyperplasia. Prostate 22: 39-42, 1993. 11. Gormley GJ, Stoner E, Bruskewitz RC, ImperatoMcGinley J, Walsh PC, McConnell JD, Andriole GL, Geller J, Brachen BR, Tenover JS, et al: The effect of finasteride in men with benign prostatic hyperplasia. The Finasteride Study Group. N Engl J Med 327: 1185-1191, 1992. 12. Bosch RJ, Griffiths DJ, Blom JH, and Schroeder FH: Treatment of benign prostatic hyperplasia by androgen deprivation: effects on prostate size and urodynamic parameters. J Urol 141: 68-72, 1989. 13. Abrams P: In support of pressure-flow studies for evaluating men with lower urinary tract symptoms. Urology 44: 153-155, 1994. 14. Lepor H, and Machi G: Comparison of the AUA symptom index in unselected males and females between fiftyfive and seventy-nine years of age. Urology 42: 36-40, 1993. 15. Blaivas JG: Multichannel urodynamic studies in men with benign prostatic hyperplasia. Indications and interpretation. Ural Clin North Am 13: 543-551, 1990.
UROLOGY@
I
FEBRUARY 1995 I VOLUME45, NUMBER2
EDITORIAL COMMENT Urologists have long placed considerable importance on overall prostate size when deciding when to intervene in benign prostatic hyperplasia (BPH). Numerous publications have indicated that overall prostate size, although a factor for determining which type of treatment is appropriate, is not helpful in establishing the presence of obstruction or the need to treat. One of finasteride’s attributes is its ability to reduce prostate volume over time, but even the large North American trial that was instrumental in the approval of finasteride did not identify size as a predictor of response to the drug. The authors indicate that it may be the transitional zone volume (as determined by transrectal ultrasound) that holds the key to predicting which patients will respond to finasteride. They suggest that if this is confirmed in further studies, transrectal ultrasound may be an appropriate maneuver for urologists to use prior to initiating the trial of a drug that may take 3 months, 6 months, or longer to manifest its benefits. They suggest that the number of patients eligible for medical management may increase fivefold in the short run because of aging of the population. Although aging alone cannot be expected to make this large an impact on the number of men eligible for BPH management, attitudes about what level of symptoms and bother from urinary symptoms constitute a level in need of treatment will. The authors identify a correlation with transition zone volume and urine flow rate but not with urinary symptoms. In the future most men receiving treatment for BPH will do so under the direction of their primary care physician. Uroflow meters
UROLOGY@ /FEBRUARY 1995 I VOLUME45, NUMBER2
and transrectal ultrasound transducers are not readily available to these clinicians. Urologists should give their opinions regarding what level of symptoms and discomfort related to BPH are appropriate for initiation of therapy. In my opinion, minimal bother or minimal symptoms are not. If patients believe their symptoms are bothersome enough to warrant medical intervention, then a medical therapeutic trial is warranted irrespective of uroflow and transrectal ultrasound considerations. Reginald C. Bruskewitz, M.D. Centerfor Health Sciences University of Wisconsin-Madison Madison, WI53792 REPLY BY AUTHORS Although we agree the benign prostatic hyperplasia should be treated based on symptoms and that many patients will be receiving treatment from their primary care physicians, it is our contention that a significant number of such patients will have a transrectal ultrasound at some point due to considerations such as elevated prostate-specific antigen or prostate size. It would be a simple task in these patients to calculate a TZ ratio to allow the physician a means to prognosticate which type of medical agent may be beneficial. Perinchery
Narayan,
M.D.
Gainesville, Fla.
265