- Email: [email protected]
TEST-RETEST VARIATION OF PRESSURE FLOW PARAMETERS IN MEN WITH BLADDER OUTLET OBSTRUCTION
0022-5347/01/1654-1188/0 THE JOURNAL OF UROLOGY® Copyright © 2001 by AMERICAN UROLOGICAL ASSOCIATION, INC.®
Vol. 165, 1188 –1192, April 2001 Printed in U.S.A.
Urological Neurology and Urodynamics TEST-RETEST VARIATION OF PRESSURE FLOW PARAMETERS IN MEN WITH BLADDER OUTLET OBSTRUCTION LARS M. ERI,* NICOLAI WESSEL
AND
VIKTOR BERGE
From the Departments of Urology, Ullevaal University Hospital, Oslo and Central Hospital of Akershus, Nordbyhagen, Norway
ABSTRACT
Purpose: We assess short-term (5 to 10 minutes) and long-term (24 weeks) test-retest changes of repeated pressure flow examinations. Materials and Methods: The pressure flow charts of 84 patients with benign prostatic enlargement and bladder outlet obstruction who had received either androgen suppressive therapy or placebo were reviewed retrospectively. Pressure flow examinations were performed at baseline, and at weeks 24 and 48. Each pressure flow session included 3 sequential voids. Results: Median detrusor opening pressure, maximum detrusor pressure, detrusor pressure at maximum flow rate and minimum voiding pressure decreased statistically significantly from void 1 to 2, ranging from 9.5% to 15.8%. From void 2 to 3 during the same pressure flow session there was a further reduction in obstruction parameters. Median Abrams/Griffiths number was 10.7% lower at void 2 compared to void 1 (p ⬍0.0001) and the urethral resistance algorithm was 3.2% lower (p ⬍0.0001). Long-term test-retest changes from baseline to week 24 and from week 24 to week 48 for the pressure flow parameters studied were negligible. Conclusions: Changes in pressure flow parameters at short-term test-retesting are considerable and probably of clinical significance. The standard pressure flow nomograms, which are based on single void pressure flow studies, might need modification when applied to repeat void studies. KEY WORDS: prostatic hyperplasia, urodynamics, urethra, reproducibility of results
Repeat voids are often performed during a pressure flow session, for example when it is assumed that the first void is not representative, or as a routine to increase the validity of the result.1, 2 Due to random patient variation and inaccuracy of the measurement method, registration from a repeat will always be different from the first void but a systematic trend towards lower detrusor pressures at short-term repeat testing, or so called bladder fatigue, has been observed.3– 6 Specially designed nomograms are often used to classify pressure flow studies in regard to obstruction. The limits of defining obstruction on these nomograms are based on single void studies, and modifications might be necessary when the nomograms are applied to repeat void pressure flow examinations.3 However, at long-term retesting pressure flow results appear to be more stable.7, 8 We performed multiple pressure flow examinations at short-term (5 to 10 minutes) and long-term (24 weeks) intervals on 84 patients with benign prostatic enlargement and bladder outlet obstruction. We determined possible systematic changes in these parameters with time and discuss the clinical significance. We analyzed random variation of pressure flow variables at repeat examination based on the same database in a separate report (unpublished). PATIENTS AND METHODS
We performed a retrospective study based on a complete review of charts of pressure flow studies of 84 patients who
had participated in a study of 24 weeks of hormonal treatment, using either the luteinizing hormone releasing hormone agonist leuprolide depot (27) or the nonsteroidal antiandrogen bicalutamide (15), or matched placebos (42).9, 10 Although the 2 drugs were chemically and pharmacologically different, both provided androgen suppression, thereby reducing prostate size and infravesical obstruction. All patients had moderate to severe lower urinary tract symptoms, a prostate larger than 30 ml., maximum urinary flow rate less than 12 ml. per second, residual urine less than 300 ml. and a detrusor pressure at maximum flow rate greater than 45 cm. H2O at pressure flow examination. Pressure flow examinations were performed at baseline, and at weeks 24 and 48 after therapy. At each session the patient was scheduled to perform 3 sequential voids. The bladder pressure transducer was zeroed with the tip of an 8F water filled single lumen catheter at the level of the symphysis pubis. Intra-abdominal pressure was measured by a water filled balloon in the rectum. The intravesical pressure before filling was considered the true estimate of the resting intra-abdominal pressure. The catheter was used for bladder filling before the voids and for bladder pressure recording during voiding. To save time, a filling rate of 80 ml. per minute was used, which was the maximum filling rate obtained by the urethral catheter we used. Filling and voiding were performed with the patient standing, and the recording of pressures was routinely continued for 1 to 2 minutes after completion of the voids to ensure that pressures were stabilized. The voids were repeated twice with 7 to 8-minute interval.
Accepted for publication November 22, 2000. * Financial interest and/or other relationship with Abbott Pharmaceutical and AstraZeneca Co. 1188
1189
TEST-RETEST VARIATION OF PRESSURE FLOW PARAMETERS
All examinations were performed and interpreted by the same investigator (L. M. E.). While reading the flow curves, a flow increase of less than 2 seconds in duration was ignored.11 A flow delay of 1 second between the recording of flow rate and corresponding bladder pressure was assumed. The maximum flow rate (Qmax) was corrected for spikes lasting for less than 2 seconds by smoothing the curve to a straight line. When determining minimum urethral opening pressure the tail of the curve, representing the last 5 to 10 ml. of the flow recording, was discarded and the pressure was usually read 2 to 5 seconds before flow actually ended.11 Parameters. Detrusor opening pressure, maximum detrusor pressure, detrusor pressure at maximum flow rate and minimum voiding pressure are the parameters proposed by the International Continence Society to be most relevant for describing detrusor function during micturition.12 In addition, all voids were classified by the 3 established parameters of bladder outlet obstruction, the Abrams/Griffiths number,13 linearized passive urethral resistance relation2 and urethral resistance algorithm.1 The Abrams/Griffiths number is a continuous obstruction parameter, incorporating detrusor pressure at maximum flow rate and maximum flow rate into a single number, and was calculated by the formula, pdet.Qmax-(2 ⫻ Qmax).12 Voidings were also classified as obstructed, equivocal or nonobstructed according to the Abrams/Griffiths nomogram.13 Linearized passive urethral resistance relation was read manually by identifying a point on the Scha¨fer nomogram, which was defined by detrusor pressure at maximum flow rate and the corresponding maximum flow rate. Linearized passive urethral resistance relation classifies the severity of obstruction in 7 steps from normal (0) to mild or moderate (I to III) to severe (IV to VI).2 Urethral resistance algorithm quantifies bladder outlet obstruction using a continuous scale. Like the Abrams/Griffiths number and linearized passive urethral resistance relation, urethral resistance algorithm is determined by maximum pressure at maximum flow rate and the corresponding maximum flow rate. It might be read manually on a graph or calculated using the formula, [1 ⫹ 4d Q2 pdet.Qmax)1/2 ⫺ 1]/(2d Qmax2), where the constant d ⫽ 3.8 10⫺4.1 Long-term and short-term test-retest variation. We first performed a preliminary study by assessing possible longterm test-retest differences among placebo cases for examinations performed at the 24-week intervals using voids 1 to 3 separately. Data on each void were similar and, therefore, they were added together to facilitate statistical analysis and ease of presentation. Correspondingly, we tested for systematic differences in short-term test-retest changes. Results were similar for examinations performed at weeks 0, 24 and 48 and, therefore, data for these time points were added together. Analysis of short-term test-retest changes was performed on patients on placebo as well as on those receiving active drug. For all data we had fewer observations for voids 2 and 3 during the same session and for weeks 24 and 48 compared to week 0. To omit bias, short-term and long-term changes were determined by calculating change for those patients who had data at both time points. Most data had a nonsymmetrical distribution and are presented as medians and 95% confidence interval (CI). The statistical calculations were done using Mann-Whitney 2-sample rank test and Wilcoxon’s nonparametric test for matched pairs. The methods, definitions and units conform to the standards proposed by the International Continence Society.12 RESULTS
Baseline characteristics of the patients are summarized in table 1. We analyzed a total of 567 voids, including 219 first, 205 second and 143 third voids. Of those voids 220 were
TABLE 1. Baseline characteristics based on first void No. participants Mean pt. age ⫾ SD Mean prostate vol. (ml.) ⫾ SD Detrusor opening pressure (cm. H2O): No. observations/median Mean ⫾ SD Max. detrusor pressure (cm. H2O): No. observations/median Mean ⫾ SD Min. voiding pressure (cm. H2O): No. observations/median Mean ⫾ SD Detrusor pressure at max. flow rate (cm. H2O): No. observations/median Mean ⫾ SD Max. flow rate (ml./sec.): No. observations/median Mean ⫾ SD Urethral resistance algorithm: No. observations/median Mean ⫾ SD Abrams/Griffiths No.: No. observations/median Mean ⫾ SD No. Abrams/Griffiths nomogram category: Nonobstructed Equivocal Obstructed No. linearized passive urethral resistance relation category: 0 I II III IV V VI
84 69.8 ⫾ 5.8 58.7 ⫾ 26.0 83/80 83.1 ⫾ 30.3 84/90 98.2 ⫾ 35.2 83/47 53.2 ⫾ 24.4 84/77 83.1 ⫾ 28.9 84/6.1 6.2 ⫾ 2.6 84/46.1 49.3 ⫾ 16.5 84/64.6 70.6 ⫾ 28.5 0 15 69 0 1 8 19 32 18 6
performed at baseline, 185 at week 24 and 162 at week 48. All pressure flow variables were unchanged from baseline to week 24 and from baseline to week 48 for placebo cases, except for a statistically significant (p ⫽ 0.021) decline in median minimum urethral opening pressure from 44 cm. H2O at week 0 to 39 cm. H2O at week 48 (table 2). Compared to void 1 of the repeat void pressure flow examinations, median detrusor opening pressure of void 2 was 15.8% lower, median maximum detrusor pressure was 12.5% lower, median detrusor pressure at maximum flow rate was
TABLE 2. Pressure flow parameters at long-term repeat testing for 3 successive voids at each time point for placebo cases Week 0 Detrusor opening pressure (cm. H2O): No. observations/median 110/66 Mean ⫾ SD 69.6 ⫾ 30.1 Max. detrusor pressure (cm. H2O): No. observations/median 111/82 Mean ⫾ SD 88.2 ⫾ 34.1 Min. voiding pressure (cm. H2O): No. observations/median 109/44 Mean ⫾ SD 47.3 ⫾ 21.1 Detrusor pressure at max. flow rate (cm. H2O): No. observations/median 111/72 Mean ⫾ SD 75.2 ⫾ 28.7 Mean linearized passive urethral resistance relation: No. observations/median 111/4 Mean ⫾ SD 3.60 ⫾ 2.21 Abrams/Griffiths No.: No. observations/median 111/59 Mean ⫾ SD 63.6 ⫾ 27.8 Urethral resistance algorithm: No. observations/median 111/44 Mean ⫾ SD 45.9 ⫾ 13.6 * Decrease from week 0 p ⬍0.05.
Week 24
Week 48
88/66 67.1 ⫾ 30.0
76/66 68.7 ⫾ 26.3
91/80 85.3 ⫾ 34.2
77/80 86.0 ⫾ 33.2
91/42 77/39* 45.0 ⫾ 19.9 42.6* ⫾ 15.9 91/72 75.1 ⫾ 31.3
76/68 72.5 ⫾ 28.5
92/4 3.58 ⫾ 1.25
76/4 3.53 ⫾ 1.24
91/60 64.1 ⫾ 30.3
76/53 59.9 ⫾ 28.4
91/45 44.9 ⫾ 13.6
76/42 45.4 ⫾ 14.5
1190
TEST-RETEST VARIATION OF PRESSURE FLOW PARAMETERS
FIG. 1. Short-term (5 to 10 minutes) test-retest changes (median and 95% CI) of detrusor pressure parameters at sequential voids. Data for 84 patients are added together for weeks 0, 24 and 48, and represent 219 pressure flow sessions, consisting of 219 first, 205 second and 143 third voids. Pdet.max, maximum detrusor pressure. Pdet.Qmax, detrusor pressure at maximum flow rate. Pdet.open, detrusor opening pressure. Pdet.min.void, minimum voiding pressure.
9.5% lower and median minimum voiding pressure was 10.9% lower (fig. 1). For all parameters there was a further reduction from void 2 to 3. All changes were statistically significant (p ⬍0.001 to p ⫽ 0.008). The relative decrease of median detrusor opening pressure from void 1 to 3 (26.3%) was more pronounced than corresponding decreases for median maximum detrusor pressure (15.9%), median detrusor pressure at maximum flow rate (16.2%) and median minimum voiding pressure (13.0%) (p ⫽ 0.0001 to 0.0019, fig. 1). Mean maximum flow rate plus or minus standard deviation was 6.4 ⫾ 2.5 ml. per second for all first voids added together for weeks 0, 24 and 48, 6.1 ⫾ 2.1 for all second voids and 6.0 ⫾ 2.0 for all third voids. Median Abrams/Griffiths number was 10.7% lower at void 2 compared to void 1 during the same pressure flow examination (p ⬍0.0001) and urethral resistance algorithm was 3.2% lower (p ⬍0.0001), again indicating a less obstructed state (fig. 2). These parameters decreased further from the void 2 to 3 (p ⬍0.0001). The relative decrease in Abrams/ Griffiths number from void 1 to 3 (19.2%) was statistically significantly (p ⫽ 0.0001) more pronounced than the relative decrease for urethral resistance algorithm (10.8%). No patients were unobstructed according to the Abrams/ Griffiths nomogram at baseline, 18% were in the equivalent zone and 82% were obstructed (table 1). From void 1 to 2 or from void 2 to 3 during the same pressure flow session 10% of the voids represented a change in Abrams/Griffiths category from obstructed to equivocal, 88% remained in the same category and 2% changed from equivocal to obstructed (table 3). The distribution in linearized passive urethral resistance relation categories of the 84 patients at baseline are summarized in table 1. During the repeat void pressure flow session 35% of the voids changed from void 1 to 2 or from void 2 to 3
FIG. 2. Short term (5 to 10 minutes) test-retest changes (median and 95% CI) of pressure flow parameters at sequential voids. Data for 84 patients are added together for weeks 0, 24 and 48, and represent 219 pressure flow sessions, consisting of 219 first, 205 second and 143 third voids. AG number, Abrams/Griffiths number. URA, urethral resistance algorithm.
TABLE 3. Short-term test-retest changes in Abrams/Griffiths category Category
No. From Void 1-2 No. From Void 2-3
From obstructed to equivocal 28 9 Unchanged 176 136 From equivocal to obstructed 3 4 Data for examinations at weeks 0, 24 and 48 are combined.
TABLE 4. Short-term test-retest changes in linearized passive urethral resistance relation category Decrease/Increase Categories ⫺3 ⫺2 ⫺1 0 ⫹1 ⫹2
No. Void 1-2
No. Void 2-3
0 7 60 118 22 0
1 2 54 81 11 0
Totals 207 Data for voids at weeks 0, 24 and 48 are combined.
149
to a lower obstruction category, 56% remained in the same category and 9% moved to a more obstructed category (table 4). We divided the 219 pressure flow sessions at void 1 into the 3 obstruction categories of low pressure—maximum detrusor pressure less than 70 cm. H2O, medium pressure—70 to 110 and high pressure—110 or greater. Median percent decrease in detrusor pressure at maximum flow rate from void 1 to 3 during a repeat void pressure flow examination was almost similar in the 3 obstruction categories (17.6%, 15.4% and 17.3%, respectively). We defined a pronounced bladder fatigue group that had a more than 25% decrease in maximum detrusor pressure from void 1 to 3 during the same pressure flow session at baseline (20 of 61 patients). These patients were not at increased risk for a similar decrease at repeat void pressure flow examinations at weeks 24 or 48. DISCUSSION
Long term test-retest variation. A pressure flow study is considered the most objective parameter of bladder outlet obstruction,12 and we found largely unchanged median val-
TEST-RETEST VARIATION OF PRESSURE FLOW PARAMETERS
ues for the parameters studied during our 1-year study period (table 2). Experience from placebo cases in drug studies also suggests unchanged bladder outlet obstruction at longterm test-retesting at least within 1 year.14 However, a slight decrease in bladder outlet obstruction has also been described, possibly because the patients become accustomed to the testing procedure or due to regression to mean for patients who are selected because of high voiding pressures.15 Short-term test-retest variation. We found successively lower values for all pressure flow parameters studied at sequential voids during the same pressure flow session (figs. 1 and 2). Maximum flow rate had a tendency to be lower for voids 2 and 3 compared to void 1, which would tend to increase the level of obstruction as described by the Abrams/ Griffiths number and linearized passive urethral resistance relation. Nevertheless, these obstruction parameters also changed statistically significantly to a lower obstruction level from void 1 to 2 and again from void 2 to 3. This finding is in agreement with most published studies but the magnitude of the decrease was greater in our study than in most other reports. Some authors have found stable detrusor pressure at short-term repeat testing.7, 8 In animal experiments Bross et al found a lower bladder pressure at repeated bladder neurostimulations at 5-minute intervals or less in the dog, and ascribed this finding to reversible and short-lived bladder smooth muscle fatigue.16 This phenomenon was also described in rhesus monkeys but a 30-minute waiting period was necessary for the bladder to recover.17 The mechanisms for a decrease in bladder pressure at repeat examination is probably the same for humans. However, if the patient becomes more relaxed during the repeat voids, with less pelvic floor and urethral sphincter contraction, and less obstruction, another possible mechanism is that the bladder might provide the same (acceptable?) flow rate using less force. The percent decrease from void 1 to 3 was more pronounced for detrusor opening pressure than for the other detrusor parameters (p ⫽ 0.000 to 0.0019). Detrusor opening pressure is often read when the detrusor pressure is in a steep increase and a small forward shift in the time for onset of flow due to a lowered urethral pressure will result in a lowering of detrusor opening pressure. We also found a more pronounced relative decrease in Abrams/Griffiths number from void 1 to 3 than that in urethral resistance algorithm. We have no obvious explanation for this discrepancy. The relationship between the level of detrusor pressure and degree of bladder fatigue was studied by Rosier et al who found that the pressure decrease at retesting was relatively most pronounced for patients without or with pronounced bladder outlet obstruction.5 We found the same percentage decrease in detrusor pressure at maximum flow rate at short-term testretesting, irrespective of the level of maximum detrusor pressure. Animal studies also suggest that the degree of bladder fatigue is related to the level of bladder pressure, and is more pronounced when the interval until retesting is short.16, 17 The patients who were defined as having pronounced bladder fatigue based on the repeat void pressure flow examination at baseline showed no tendency to have more bladder fatigue than others at weeks 24 and 48. This finding suggests that bladder fatigue is a normal physiological feature and not a phenomenon limited to a patient subgroup. In the linearized passive urethral resistance relation nomogram, which divides voids into 7 obstruction categories, moving 1 category is considered to represent a sufficiently large change in bladder outlet obstruction and is of clinical significance.2 Due to random variation 9% of the voids were in a more obstructed category than the preceding void in our study. However, 35% of the voids changed category to a less obstructed state (table 4). When reading the charts of the pressure flow examinations, voids were classified as first, second or third. In a few
1191
cases, if a patient had performed a first void, which for some technical reason was not recorded, void 2 was read as void 1. Furthermore, in a few instances void 1 was impossible to analyze and was replaced by void 2. This practice can be criticized but because the changes from void 1 to 2 were basically the same as those from void 2 to 3, it is unlikely that this inaccuracy influenced our results. CONCLUSIONS
We found stable pressure flow parameters in patients on placebo examined at 24-week intervals (long-term test-retest variation). However, at repeat voids during the same pressure flow session there was a statistically significant decrease in the 7 obstruction parameters (short-term testretest repeatability). A modification of the limits of defining obstruction should be considered for repeat void pressure flow studies. To describe more accurately the effects of repeat testing, more studies are required based on large, high quality databases that include cases representing a large range of obstruction grades. REFERENCES
1. Griffiths, D., Mastrigt, R. and Bosch, R.: Quantification of urethral resistance and bladder function during voiding, with special reference to the effects of prostate size reduction on urethral obstruction due to benign prostatic hyperplasia. Neurourol Urodyn, 8: 17, 1989 2. Scha¨fer, W.: Principles and clinical application of advanced urodynamic analysis of voiding function. Urol Clin North Am, 17: 553, 1990 3. Abrams, P., Scha¨fer, W., Tammela, T. L. J. et al: Improvement of pressure flow parameters with finasteride is greater in men with large prostates. J Urol, 161: 1513, 1999 4. Hansen, F., Olsen, L., Atan, A. et al: Pressure-flow studies: short-time repeatability. Neurourol Urodynam, 18: 205, 1999 5. Rosier, P. F. W. M., de la Rosette, J. J. M. C. H., Koldewijn, E. L. et al: Variability of pressure-flow analysis parameters in repeated cystometry in patients with benign prostatic hyperplasia. J Urol, 153: 1520, 1995 6. Tammela, T. L. J., Scha¨fer, W., Barrett, D. M. et al: Repeated pressure-flow studies in the evaluation of bladder outlet obstruction due to benign prostatic enlargement. Neurourol Urodynam, 18: 17, 1999 7. Lose, G. and Thyssen, H.: Reproducibility of cystometry and pressure flow parameters in female patients. Neurourol Urodynam, 15: 302, 1996 8. Madsen, F. A., Rhodes, P. R. and Bruskewitz, R. C.: Reproducibility of pressure-flow variables in patients with symptomatic benign prostatic hyperplasia. Urology, 46: 816, 1995 9. Eri, L. M. and Tveter, K. J.: A prospective, placebo-controlled study of the luteinizing hormone-releasing hormone agonist leuprolide as treatment for patients with benign prostatic hyperplasia. J Urol, 150: 359, 1993 10. Eri, L. M. and Tveter, K. J.: A prospective, placebo-controlled study of the antiandrogen Casodex as treatment for patients with benign prostatic hyperplasia. J Urol, 150: 90, 1993 11. Scha¨fer, W., Thu¨roff, J. W. and the ICS-BPH-Study Group: Comparing different obstruction parameters. Neurourol Urodynam, 14: 564, 1995 12. Griffiths, D., Ho¨fner, K., van Mastrigt, R. et al: Standardization of terminology of lower urinary tract function: pressure-flow studies of voiding, urethral resistance, and urethral obstruction. Neurourol Urodynam, 16: 1, 1997 13. Lin, C. S. and Abrams, P.: The Abrams-Griffiths nomogram. World J Urol, 13: 34, 1995 14. deWildt, M. J. A. M., Rosier, P. F. W. M., Witjes, W. P. J. et al: Watchful waiting in the treatment of prostatism complaints: do urodynamic parameters change? Neurourol Urodynam, 13: 394, 1994 15. Witjes, W. P. J., de Wildt, M. J. A. M., Rosier, P. F. W. M. et al: Variability of clinical and pressure-flow study variables after 6 months of watchful waiting in patients with lower urinary tract symptoms and benign prostatic enlargement. J Urol, 156: 1026, 1996
1192
TEST-RETEST VARIATION OF PRESSURE FLOW PARAMETERS
16. Bross, S., Schumacher, S., Scheepe, J. R. et al: Smooth muscle fatigue due to repeated urinary bladder neurostimulation: an in vivo study. Neurourol Urodynam, 18: 41, 1999 17. Shoukry, M. S. and Ghoniem, G. M.: Effect of time interval and overdistension on repeated urodynamic studies. J Urol, 147: 185, 1992 EDITORIAL COMMENT The authors analyze again the reproducibility of pressure flow studies, which is the prerequisite for acceptance of the measurement routine for patients with lower urinary tract symptoms and bladder outlet obstruction. Overall, there is little short-term variability for an invasive measurement without anesthesia. However, the authors determine a significant decrease in relevant voiding pressure values from measurement 1 to 3, and in principle these results were confirmed by Rosier and Hansen et al (references 4 and 5 in article). At short-term repeat measurement Rosier et al reported that variations of maximum flow rate were not significant and values for detrusor pressure at maximum flow rate were significant lower. The appropriate detrusor pressure at maximum flow rate was less than 15 cm. water in 80% of the patients. Overall, rarely were the intraindividual fluctuations great enough to effect a change in the grade of obstruction on the Scha¨fter nomogram. Hansen et al found similar results with a systematic variation in detrusor pressure at maximum flow rate and a lack of statistically significant systematic variations in maximum flow rate in 22 patients. Apparently, it appears to be physiological that the values of voiding pressures decrease on repeated studies without a significant variation of maximum flow rate being observed. However, in contrast to previous publications, the authors report a decrease in maximum flow rate but information on concrete values of the decrease or its significance is lacking. It would be important to attempt to explain the cause of the decrease in voiding parameters on repeated measurements. Generally, a decrease in pressure values (mainly detrusor pressure at maximum flow rate as the most accepted obstruction value) without relevant maximum flow rate changes indicates a less obstructed second voiding. The discussion by the authors that “bladder fatigue,” as observed by Bross et al during quick repetition of electrostimulation (reference 16 in article) or detrusor over distention (reference 17 in article)1 is not likely the cause of the decrease in voiding pressures without maximum flow rate variation. In the case of electrostimulation the conditions for initiation and quality of detrusor contraction are totally different. Bladder over distention is only evident when a significantly higher filled volume is used during the second or third filling of the bladder. In the article by Rosier et al the filled volumes did not vary significantly during the second repeat measurement but even so detrusor pressure at maximum flow rate was significantly lower. Furthermore, the authors provide no data for bladder fatigue
during a significant reduction of detrusor contractility. A second but coarse method is the variation of the detrusor pressure at maximum flow rate point on the Scha¨fer nomogram into an area of lower detrusor contractility. In view of the pressure values published this fact is also not convincing. The only explanation remaining is the variation of the grade of obstruction, which the authors present in the reduction of Abrams/Griffiths number and urethral resistance algorithm, which is in agreement with others (references 4 to 6 in article). The reproducibility of urodynamic measurements becomes a decisive question when they are the basis for diagnosis and therapeutic decision. Complete relaxation of the urethral muscles is a precondition for the diagnosis of mechanical obstruction, as unfavorable investigation conditions (unfavorable position of the patient during voiding or disturbance during investigation) make complete relaxation of the sphincter muscles and, thus, detection of the mechanical resistance (resistance remaining after complete relaxation of the urethral muscles) impossible. Therefore, measurements should be repeated in any case to achieve relaxation of the urethral muscles during voiding. Each investigation showing the lowest value of mechanical obstruction is considered correct during repeated measurement, independent of the type of analysis or classification concept. In principle the lowest detrusor pressure at maximum flow rate value registered on repeated measurements is valid, if only single points of the pressure flow plot are used for classification. If during multiple measurements equal maximum flow rate values are registered, the investigation with the lowest detrusor pressure during maximum flow rate should be used. The same procedure applies for minimal voiding pressure. Klaus Ho¨fner Department of Urology Hannover Medical School Hannover, Germany 1. Bross, S., Schumacher, S., Scheepe, J. R. et al: Effects of acute urinary bladder overdistension on bladder response during sacral neurostimulation. Eur Urol, 36: 354, 1999 REPLY BY AUTHORS We found a decrease in bladder outlet obstruction at short-term retesting, which did not result in improved flow rate because repeat voids were performed with a lower detrusor pressure. We used the somewhat simplistic term bladder fatigue for this phenomenon. However, the decrease in detrusor pressure at repeat voiding might also have been a result of adaptation of the detrusor to a lower level of bladder outlet obstruction, and it might have been more appropriate for us to use a more generic term.
Vol. 165, 1188 –1192, April 2001 Printed in U.S.A.
Urological Neurology and Urodynamics TEST-RETEST VARIATION OF PRESSURE FLOW PARAMETERS IN MEN WITH BLADDER OUTLET OBSTRUCTION LARS M. ERI,* NICOLAI WESSEL
AND
VIKTOR BERGE
From the Departments of Urology, Ullevaal University Hospital, Oslo and Central Hospital of Akershus, Nordbyhagen, Norway
ABSTRACT
Purpose: We assess short-term (5 to 10 minutes) and long-term (24 weeks) test-retest changes of repeated pressure flow examinations. Materials and Methods: The pressure flow charts of 84 patients with benign prostatic enlargement and bladder outlet obstruction who had received either androgen suppressive therapy or placebo were reviewed retrospectively. Pressure flow examinations were performed at baseline, and at weeks 24 and 48. Each pressure flow session included 3 sequential voids. Results: Median detrusor opening pressure, maximum detrusor pressure, detrusor pressure at maximum flow rate and minimum voiding pressure decreased statistically significantly from void 1 to 2, ranging from 9.5% to 15.8%. From void 2 to 3 during the same pressure flow session there was a further reduction in obstruction parameters. Median Abrams/Griffiths number was 10.7% lower at void 2 compared to void 1 (p ⬍0.0001) and the urethral resistance algorithm was 3.2% lower (p ⬍0.0001). Long-term test-retest changes from baseline to week 24 and from week 24 to week 48 for the pressure flow parameters studied were negligible. Conclusions: Changes in pressure flow parameters at short-term test-retesting are considerable and probably of clinical significance. The standard pressure flow nomograms, which are based on single void pressure flow studies, might need modification when applied to repeat void studies. KEY WORDS: prostatic hyperplasia, urodynamics, urethra, reproducibility of results
Repeat voids are often performed during a pressure flow session, for example when it is assumed that the first void is not representative, or as a routine to increase the validity of the result.1, 2 Due to random patient variation and inaccuracy of the measurement method, registration from a repeat will always be different from the first void but a systematic trend towards lower detrusor pressures at short-term repeat testing, or so called bladder fatigue, has been observed.3– 6 Specially designed nomograms are often used to classify pressure flow studies in regard to obstruction. The limits of defining obstruction on these nomograms are based on single void studies, and modifications might be necessary when the nomograms are applied to repeat void pressure flow examinations.3 However, at long-term retesting pressure flow results appear to be more stable.7, 8 We performed multiple pressure flow examinations at short-term (5 to 10 minutes) and long-term (24 weeks) intervals on 84 patients with benign prostatic enlargement and bladder outlet obstruction. We determined possible systematic changes in these parameters with time and discuss the clinical significance. We analyzed random variation of pressure flow variables at repeat examination based on the same database in a separate report (unpublished). PATIENTS AND METHODS
We performed a retrospective study based on a complete review of charts of pressure flow studies of 84 patients who
had participated in a study of 24 weeks of hormonal treatment, using either the luteinizing hormone releasing hormone agonist leuprolide depot (27) or the nonsteroidal antiandrogen bicalutamide (15), or matched placebos (42).9, 10 Although the 2 drugs were chemically and pharmacologically different, both provided androgen suppression, thereby reducing prostate size and infravesical obstruction. All patients had moderate to severe lower urinary tract symptoms, a prostate larger than 30 ml., maximum urinary flow rate less than 12 ml. per second, residual urine less than 300 ml. and a detrusor pressure at maximum flow rate greater than 45 cm. H2O at pressure flow examination. Pressure flow examinations were performed at baseline, and at weeks 24 and 48 after therapy. At each session the patient was scheduled to perform 3 sequential voids. The bladder pressure transducer was zeroed with the tip of an 8F water filled single lumen catheter at the level of the symphysis pubis. Intra-abdominal pressure was measured by a water filled balloon in the rectum. The intravesical pressure before filling was considered the true estimate of the resting intra-abdominal pressure. The catheter was used for bladder filling before the voids and for bladder pressure recording during voiding. To save time, a filling rate of 80 ml. per minute was used, which was the maximum filling rate obtained by the urethral catheter we used. Filling and voiding were performed with the patient standing, and the recording of pressures was routinely continued for 1 to 2 minutes after completion of the voids to ensure that pressures were stabilized. The voids were repeated twice with 7 to 8-minute interval.
Accepted for publication November 22, 2000. * Financial interest and/or other relationship with Abbott Pharmaceutical and AstraZeneca Co. 1188
1189
TEST-RETEST VARIATION OF PRESSURE FLOW PARAMETERS
All examinations were performed and interpreted by the same investigator (L. M. E.). While reading the flow curves, a flow increase of less than 2 seconds in duration was ignored.11 A flow delay of 1 second between the recording of flow rate and corresponding bladder pressure was assumed. The maximum flow rate (Qmax) was corrected for spikes lasting for less than 2 seconds by smoothing the curve to a straight line. When determining minimum urethral opening pressure the tail of the curve, representing the last 5 to 10 ml. of the flow recording, was discarded and the pressure was usually read 2 to 5 seconds before flow actually ended.11 Parameters. Detrusor opening pressure, maximum detrusor pressure, detrusor pressure at maximum flow rate and minimum voiding pressure are the parameters proposed by the International Continence Society to be most relevant for describing detrusor function during micturition.12 In addition, all voids were classified by the 3 established parameters of bladder outlet obstruction, the Abrams/Griffiths number,13 linearized passive urethral resistance relation2 and urethral resistance algorithm.1 The Abrams/Griffiths number is a continuous obstruction parameter, incorporating detrusor pressure at maximum flow rate and maximum flow rate into a single number, and was calculated by the formula, pdet.Qmax-(2 ⫻ Qmax).12 Voidings were also classified as obstructed, equivocal or nonobstructed according to the Abrams/Griffiths nomogram.13 Linearized passive urethral resistance relation was read manually by identifying a point on the Scha¨fer nomogram, which was defined by detrusor pressure at maximum flow rate and the corresponding maximum flow rate. Linearized passive urethral resistance relation classifies the severity of obstruction in 7 steps from normal (0) to mild or moderate (I to III) to severe (IV to VI).2 Urethral resistance algorithm quantifies bladder outlet obstruction using a continuous scale. Like the Abrams/Griffiths number and linearized passive urethral resistance relation, urethral resistance algorithm is determined by maximum pressure at maximum flow rate and the corresponding maximum flow rate. It might be read manually on a graph or calculated using the formula, [1 ⫹ 4d Q2 pdet.Qmax)1/2 ⫺ 1]/(2d Qmax2), where the constant d ⫽ 3.8 10⫺4.1 Long-term and short-term test-retest variation. We first performed a preliminary study by assessing possible longterm test-retest differences among placebo cases for examinations performed at the 24-week intervals using voids 1 to 3 separately. Data on each void were similar and, therefore, they were added together to facilitate statistical analysis and ease of presentation. Correspondingly, we tested for systematic differences in short-term test-retest changes. Results were similar for examinations performed at weeks 0, 24 and 48 and, therefore, data for these time points were added together. Analysis of short-term test-retest changes was performed on patients on placebo as well as on those receiving active drug. For all data we had fewer observations for voids 2 and 3 during the same session and for weeks 24 and 48 compared to week 0. To omit bias, short-term and long-term changes were determined by calculating change for those patients who had data at both time points. Most data had a nonsymmetrical distribution and are presented as medians and 95% confidence interval (CI). The statistical calculations were done using Mann-Whitney 2-sample rank test and Wilcoxon’s nonparametric test for matched pairs. The methods, definitions and units conform to the standards proposed by the International Continence Society.12 RESULTS
Baseline characteristics of the patients are summarized in table 1. We analyzed a total of 567 voids, including 219 first, 205 second and 143 third voids. Of those voids 220 were
TABLE 1. Baseline characteristics based on first void No. participants Mean pt. age ⫾ SD Mean prostate vol. (ml.) ⫾ SD Detrusor opening pressure (cm. H2O): No. observations/median Mean ⫾ SD Max. detrusor pressure (cm. H2O): No. observations/median Mean ⫾ SD Min. voiding pressure (cm. H2O): No. observations/median Mean ⫾ SD Detrusor pressure at max. flow rate (cm. H2O): No. observations/median Mean ⫾ SD Max. flow rate (ml./sec.): No. observations/median Mean ⫾ SD Urethral resistance algorithm: No. observations/median Mean ⫾ SD Abrams/Griffiths No.: No. observations/median Mean ⫾ SD No. Abrams/Griffiths nomogram category: Nonobstructed Equivocal Obstructed No. linearized passive urethral resistance relation category: 0 I II III IV V VI
84 69.8 ⫾ 5.8 58.7 ⫾ 26.0 83/80 83.1 ⫾ 30.3 84/90 98.2 ⫾ 35.2 83/47 53.2 ⫾ 24.4 84/77 83.1 ⫾ 28.9 84/6.1 6.2 ⫾ 2.6 84/46.1 49.3 ⫾ 16.5 84/64.6 70.6 ⫾ 28.5 0 15 69 0 1 8 19 32 18 6
performed at baseline, 185 at week 24 and 162 at week 48. All pressure flow variables were unchanged from baseline to week 24 and from baseline to week 48 for placebo cases, except for a statistically significant (p ⫽ 0.021) decline in median minimum urethral opening pressure from 44 cm. H2O at week 0 to 39 cm. H2O at week 48 (table 2). Compared to void 1 of the repeat void pressure flow examinations, median detrusor opening pressure of void 2 was 15.8% lower, median maximum detrusor pressure was 12.5% lower, median detrusor pressure at maximum flow rate was
TABLE 2. Pressure flow parameters at long-term repeat testing for 3 successive voids at each time point for placebo cases Week 0 Detrusor opening pressure (cm. H2O): No. observations/median 110/66 Mean ⫾ SD 69.6 ⫾ 30.1 Max. detrusor pressure (cm. H2O): No. observations/median 111/82 Mean ⫾ SD 88.2 ⫾ 34.1 Min. voiding pressure (cm. H2O): No. observations/median 109/44 Mean ⫾ SD 47.3 ⫾ 21.1 Detrusor pressure at max. flow rate (cm. H2O): No. observations/median 111/72 Mean ⫾ SD 75.2 ⫾ 28.7 Mean linearized passive urethral resistance relation: No. observations/median 111/4 Mean ⫾ SD 3.60 ⫾ 2.21 Abrams/Griffiths No.: No. observations/median 111/59 Mean ⫾ SD 63.6 ⫾ 27.8 Urethral resistance algorithm: No. observations/median 111/44 Mean ⫾ SD 45.9 ⫾ 13.6 * Decrease from week 0 p ⬍0.05.
Week 24
Week 48
88/66 67.1 ⫾ 30.0
76/66 68.7 ⫾ 26.3
91/80 85.3 ⫾ 34.2
77/80 86.0 ⫾ 33.2
91/42 77/39* 45.0 ⫾ 19.9 42.6* ⫾ 15.9 91/72 75.1 ⫾ 31.3
76/68 72.5 ⫾ 28.5
92/4 3.58 ⫾ 1.25
76/4 3.53 ⫾ 1.24
91/60 64.1 ⫾ 30.3
76/53 59.9 ⫾ 28.4
91/45 44.9 ⫾ 13.6
76/42 45.4 ⫾ 14.5
1190
TEST-RETEST VARIATION OF PRESSURE FLOW PARAMETERS
FIG. 1. Short-term (5 to 10 minutes) test-retest changes (median and 95% CI) of detrusor pressure parameters at sequential voids. Data for 84 patients are added together for weeks 0, 24 and 48, and represent 219 pressure flow sessions, consisting of 219 first, 205 second and 143 third voids. Pdet.max, maximum detrusor pressure. Pdet.Qmax, detrusor pressure at maximum flow rate. Pdet.open, detrusor opening pressure. Pdet.min.void, minimum voiding pressure.
9.5% lower and median minimum voiding pressure was 10.9% lower (fig. 1). For all parameters there was a further reduction from void 2 to 3. All changes were statistically significant (p ⬍0.001 to p ⫽ 0.008). The relative decrease of median detrusor opening pressure from void 1 to 3 (26.3%) was more pronounced than corresponding decreases for median maximum detrusor pressure (15.9%), median detrusor pressure at maximum flow rate (16.2%) and median minimum voiding pressure (13.0%) (p ⫽ 0.0001 to 0.0019, fig. 1). Mean maximum flow rate plus or minus standard deviation was 6.4 ⫾ 2.5 ml. per second for all first voids added together for weeks 0, 24 and 48, 6.1 ⫾ 2.1 for all second voids and 6.0 ⫾ 2.0 for all third voids. Median Abrams/Griffiths number was 10.7% lower at void 2 compared to void 1 during the same pressure flow examination (p ⬍0.0001) and urethral resistance algorithm was 3.2% lower (p ⬍0.0001), again indicating a less obstructed state (fig. 2). These parameters decreased further from the void 2 to 3 (p ⬍0.0001). The relative decrease in Abrams/ Griffiths number from void 1 to 3 (19.2%) was statistically significantly (p ⫽ 0.0001) more pronounced than the relative decrease for urethral resistance algorithm (10.8%). No patients were unobstructed according to the Abrams/ Griffiths nomogram at baseline, 18% were in the equivalent zone and 82% were obstructed (table 1). From void 1 to 2 or from void 2 to 3 during the same pressure flow session 10% of the voids represented a change in Abrams/Griffiths category from obstructed to equivocal, 88% remained in the same category and 2% changed from equivocal to obstructed (table 3). The distribution in linearized passive urethral resistance relation categories of the 84 patients at baseline are summarized in table 1. During the repeat void pressure flow session 35% of the voids changed from void 1 to 2 or from void 2 to 3
FIG. 2. Short term (5 to 10 minutes) test-retest changes (median and 95% CI) of pressure flow parameters at sequential voids. Data for 84 patients are added together for weeks 0, 24 and 48, and represent 219 pressure flow sessions, consisting of 219 first, 205 second and 143 third voids. AG number, Abrams/Griffiths number. URA, urethral resistance algorithm.
TABLE 3. Short-term test-retest changes in Abrams/Griffiths category Category
No. From Void 1-2 No. From Void 2-3
From obstructed to equivocal 28 9 Unchanged 176 136 From equivocal to obstructed 3 4 Data for examinations at weeks 0, 24 and 48 are combined.
TABLE 4. Short-term test-retest changes in linearized passive urethral resistance relation category Decrease/Increase Categories ⫺3 ⫺2 ⫺1 0 ⫹1 ⫹2
No. Void 1-2
No. Void 2-3
0 7 60 118 22 0
1 2 54 81 11 0
Totals 207 Data for voids at weeks 0, 24 and 48 are combined.
149
to a lower obstruction category, 56% remained in the same category and 9% moved to a more obstructed category (table 4). We divided the 219 pressure flow sessions at void 1 into the 3 obstruction categories of low pressure—maximum detrusor pressure less than 70 cm. H2O, medium pressure—70 to 110 and high pressure—110 or greater. Median percent decrease in detrusor pressure at maximum flow rate from void 1 to 3 during a repeat void pressure flow examination was almost similar in the 3 obstruction categories (17.6%, 15.4% and 17.3%, respectively). We defined a pronounced bladder fatigue group that had a more than 25% decrease in maximum detrusor pressure from void 1 to 3 during the same pressure flow session at baseline (20 of 61 patients). These patients were not at increased risk for a similar decrease at repeat void pressure flow examinations at weeks 24 or 48. DISCUSSION
Long term test-retest variation. A pressure flow study is considered the most objective parameter of bladder outlet obstruction,12 and we found largely unchanged median val-
TEST-RETEST VARIATION OF PRESSURE FLOW PARAMETERS
ues for the parameters studied during our 1-year study period (table 2). Experience from placebo cases in drug studies also suggests unchanged bladder outlet obstruction at longterm test-retesting at least within 1 year.14 However, a slight decrease in bladder outlet obstruction has also been described, possibly because the patients become accustomed to the testing procedure or due to regression to mean for patients who are selected because of high voiding pressures.15 Short-term test-retest variation. We found successively lower values for all pressure flow parameters studied at sequential voids during the same pressure flow session (figs. 1 and 2). Maximum flow rate had a tendency to be lower for voids 2 and 3 compared to void 1, which would tend to increase the level of obstruction as described by the Abrams/ Griffiths number and linearized passive urethral resistance relation. Nevertheless, these obstruction parameters also changed statistically significantly to a lower obstruction level from void 1 to 2 and again from void 2 to 3. This finding is in agreement with most published studies but the magnitude of the decrease was greater in our study than in most other reports. Some authors have found stable detrusor pressure at short-term repeat testing.7, 8 In animal experiments Bross et al found a lower bladder pressure at repeated bladder neurostimulations at 5-minute intervals or less in the dog, and ascribed this finding to reversible and short-lived bladder smooth muscle fatigue.16 This phenomenon was also described in rhesus monkeys but a 30-minute waiting period was necessary for the bladder to recover.17 The mechanisms for a decrease in bladder pressure at repeat examination is probably the same for humans. However, if the patient becomes more relaxed during the repeat voids, with less pelvic floor and urethral sphincter contraction, and less obstruction, another possible mechanism is that the bladder might provide the same (acceptable?) flow rate using less force. The percent decrease from void 1 to 3 was more pronounced for detrusor opening pressure than for the other detrusor parameters (p ⫽ 0.000 to 0.0019). Detrusor opening pressure is often read when the detrusor pressure is in a steep increase and a small forward shift in the time for onset of flow due to a lowered urethral pressure will result in a lowering of detrusor opening pressure. We also found a more pronounced relative decrease in Abrams/Griffiths number from void 1 to 3 than that in urethral resistance algorithm. We have no obvious explanation for this discrepancy. The relationship between the level of detrusor pressure and degree of bladder fatigue was studied by Rosier et al who found that the pressure decrease at retesting was relatively most pronounced for patients without or with pronounced bladder outlet obstruction.5 We found the same percentage decrease in detrusor pressure at maximum flow rate at short-term testretesting, irrespective of the level of maximum detrusor pressure. Animal studies also suggest that the degree of bladder fatigue is related to the level of bladder pressure, and is more pronounced when the interval until retesting is short.16, 17 The patients who were defined as having pronounced bladder fatigue based on the repeat void pressure flow examination at baseline showed no tendency to have more bladder fatigue than others at weeks 24 and 48. This finding suggests that bladder fatigue is a normal physiological feature and not a phenomenon limited to a patient subgroup. In the linearized passive urethral resistance relation nomogram, which divides voids into 7 obstruction categories, moving 1 category is considered to represent a sufficiently large change in bladder outlet obstruction and is of clinical significance.2 Due to random variation 9% of the voids were in a more obstructed category than the preceding void in our study. However, 35% of the voids changed category to a less obstructed state (table 4). When reading the charts of the pressure flow examinations, voids were classified as first, second or third. In a few
1191
cases, if a patient had performed a first void, which for some technical reason was not recorded, void 2 was read as void 1. Furthermore, in a few instances void 1 was impossible to analyze and was replaced by void 2. This practice can be criticized but because the changes from void 1 to 2 were basically the same as those from void 2 to 3, it is unlikely that this inaccuracy influenced our results. CONCLUSIONS
We found stable pressure flow parameters in patients on placebo examined at 24-week intervals (long-term test-retest variation). However, at repeat voids during the same pressure flow session there was a statistically significant decrease in the 7 obstruction parameters (short-term testretest repeatability). A modification of the limits of defining obstruction should be considered for repeat void pressure flow studies. To describe more accurately the effects of repeat testing, more studies are required based on large, high quality databases that include cases representing a large range of obstruction grades. REFERENCES
1. Griffiths, D., Mastrigt, R. and Bosch, R.: Quantification of urethral resistance and bladder function during voiding, with special reference to the effects of prostate size reduction on urethral obstruction due to benign prostatic hyperplasia. Neurourol Urodyn, 8: 17, 1989 2. Scha¨fer, W.: Principles and clinical application of advanced urodynamic analysis of voiding function. Urol Clin North Am, 17: 553, 1990 3. Abrams, P., Scha¨fer, W., Tammela, T. L. J. et al: Improvement of pressure flow parameters with finasteride is greater in men with large prostates. J Urol, 161: 1513, 1999 4. Hansen, F., Olsen, L., Atan, A. et al: Pressure-flow studies: short-time repeatability. Neurourol Urodynam, 18: 205, 1999 5. Rosier, P. F. W. M., de la Rosette, J. J. M. C. H., Koldewijn, E. L. et al: Variability of pressure-flow analysis parameters in repeated cystometry in patients with benign prostatic hyperplasia. J Urol, 153: 1520, 1995 6. Tammela, T. L. J., Scha¨fer, W., Barrett, D. M. et al: Repeated pressure-flow studies in the evaluation of bladder outlet obstruction due to benign prostatic enlargement. Neurourol Urodynam, 18: 17, 1999 7. Lose, G. and Thyssen, H.: Reproducibility of cystometry and pressure flow parameters in female patients. Neurourol Urodynam, 15: 302, 1996 8. Madsen, F. A., Rhodes, P. R. and Bruskewitz, R. C.: Reproducibility of pressure-flow variables in patients with symptomatic benign prostatic hyperplasia. Urology, 46: 816, 1995 9. Eri, L. M. and Tveter, K. J.: A prospective, placebo-controlled study of the luteinizing hormone-releasing hormone agonist leuprolide as treatment for patients with benign prostatic hyperplasia. J Urol, 150: 359, 1993 10. Eri, L. M. and Tveter, K. J.: A prospective, placebo-controlled study of the antiandrogen Casodex as treatment for patients with benign prostatic hyperplasia. J Urol, 150: 90, 1993 11. Scha¨fer, W., Thu¨roff, J. W. and the ICS-BPH-Study Group: Comparing different obstruction parameters. Neurourol Urodynam, 14: 564, 1995 12. Griffiths, D., Ho¨fner, K., van Mastrigt, R. et al: Standardization of terminology of lower urinary tract function: pressure-flow studies of voiding, urethral resistance, and urethral obstruction. Neurourol Urodynam, 16: 1, 1997 13. Lin, C. S. and Abrams, P.: The Abrams-Griffiths nomogram. World J Urol, 13: 34, 1995 14. deWildt, M. J. A. M., Rosier, P. F. W. M., Witjes, W. P. J. et al: Watchful waiting in the treatment of prostatism complaints: do urodynamic parameters change? Neurourol Urodynam, 13: 394, 1994 15. Witjes, W. P. J., de Wildt, M. J. A. M., Rosier, P. F. W. M. et al: Variability of clinical and pressure-flow study variables after 6 months of watchful waiting in patients with lower urinary tract symptoms and benign prostatic enlargement. J Urol, 156: 1026, 1996
1192
TEST-RETEST VARIATION OF PRESSURE FLOW PARAMETERS
16. Bross, S., Schumacher, S., Scheepe, J. R. et al: Smooth muscle fatigue due to repeated urinary bladder neurostimulation: an in vivo study. Neurourol Urodynam, 18: 41, 1999 17. Shoukry, M. S. and Ghoniem, G. M.: Effect of time interval and overdistension on repeated urodynamic studies. J Urol, 147: 185, 1992 EDITORIAL COMMENT The authors analyze again the reproducibility of pressure flow studies, which is the prerequisite for acceptance of the measurement routine for patients with lower urinary tract symptoms and bladder outlet obstruction. Overall, there is little short-term variability for an invasive measurement without anesthesia. However, the authors determine a significant decrease in relevant voiding pressure values from measurement 1 to 3, and in principle these results were confirmed by Rosier and Hansen et al (references 4 and 5 in article). At short-term repeat measurement Rosier et al reported that variations of maximum flow rate were not significant and values for detrusor pressure at maximum flow rate were significant lower. The appropriate detrusor pressure at maximum flow rate was less than 15 cm. water in 80% of the patients. Overall, rarely were the intraindividual fluctuations great enough to effect a change in the grade of obstruction on the Scha¨fter nomogram. Hansen et al found similar results with a systematic variation in detrusor pressure at maximum flow rate and a lack of statistically significant systematic variations in maximum flow rate in 22 patients. Apparently, it appears to be physiological that the values of voiding pressures decrease on repeated studies without a significant variation of maximum flow rate being observed. However, in contrast to previous publications, the authors report a decrease in maximum flow rate but information on concrete values of the decrease or its significance is lacking. It would be important to attempt to explain the cause of the decrease in voiding parameters on repeated measurements. Generally, a decrease in pressure values (mainly detrusor pressure at maximum flow rate as the most accepted obstruction value) without relevant maximum flow rate changes indicates a less obstructed second voiding. The discussion by the authors that “bladder fatigue,” as observed by Bross et al during quick repetition of electrostimulation (reference 16 in article) or detrusor over distention (reference 17 in article)1 is not likely the cause of the decrease in voiding pressures without maximum flow rate variation. In the case of electrostimulation the conditions for initiation and quality of detrusor contraction are totally different. Bladder over distention is only evident when a significantly higher filled volume is used during the second or third filling of the bladder. In the article by Rosier et al the filled volumes did not vary significantly during the second repeat measurement but even so detrusor pressure at maximum flow rate was significantly lower. Furthermore, the authors provide no data for bladder fatigue
during a significant reduction of detrusor contractility. A second but coarse method is the variation of the detrusor pressure at maximum flow rate point on the Scha¨fer nomogram into an area of lower detrusor contractility. In view of the pressure values published this fact is also not convincing. The only explanation remaining is the variation of the grade of obstruction, which the authors present in the reduction of Abrams/Griffiths number and urethral resistance algorithm, which is in agreement with others (references 4 to 6 in article). The reproducibility of urodynamic measurements becomes a decisive question when they are the basis for diagnosis and therapeutic decision. Complete relaxation of the urethral muscles is a precondition for the diagnosis of mechanical obstruction, as unfavorable investigation conditions (unfavorable position of the patient during voiding or disturbance during investigation) make complete relaxation of the sphincter muscles and, thus, detection of the mechanical resistance (resistance remaining after complete relaxation of the urethral muscles) impossible. Therefore, measurements should be repeated in any case to achieve relaxation of the urethral muscles during voiding. Each investigation showing the lowest value of mechanical obstruction is considered correct during repeated measurement, independent of the type of analysis or classification concept. In principle the lowest detrusor pressure at maximum flow rate value registered on repeated measurements is valid, if only single points of the pressure flow plot are used for classification. If during multiple measurements equal maximum flow rate values are registered, the investigation with the lowest detrusor pressure during maximum flow rate should be used. The same procedure applies for minimal voiding pressure. Klaus Ho¨fner Department of Urology Hannover Medical School Hannover, Germany 1. Bross, S., Schumacher, S., Scheepe, J. R. et al: Effects of acute urinary bladder overdistension on bladder response during sacral neurostimulation. Eur Urol, 36: 354, 1999 REPLY BY AUTHORS We found a decrease in bladder outlet obstruction at short-term retesting, which did not result in improved flow rate because repeat voids were performed with a lower detrusor pressure. We used the somewhat simplistic term bladder fatigue for this phenomenon. However, the decrease in detrusor pressure at repeat voiding might also have been a result of adaptation of the detrusor to a lower level of bladder outlet obstruction, and it might have been more appropriate for us to use a more generic term.