- Email: [email protected]
NONINVASIVE MEASUREMENT OF BLADDER PRESSURE BY CONTROLLED INFLATION OF A PENILE CUFF
0022-5347/02/1673-1344/0 THE JOURNAL OF UROLOGY® Copyright © 2002 by AMERICAN UROLOGICAL ASSOCIATION, INC.®
Vol. 167, 1344 –1347, March 2002 Printed in U.S.A.
NONINVASIVE MEASUREMENT OF BLADDER PRESSURE BY CONTROLLED INFLATION OF A PENILE CUFF C. J. GRIFFITHS,* D. RIX, A. M. MACDONALD, M. J. DRINNAN, R. S. PICKARD P. D. RAMSDEN
AND
From the Regional Medical Physics Department and Department of Urology, Freeman Hospital, Newcastle upon Tyne, United Kingdom
ABSTRACT
Purpose: A noninvasive test providing reliable objective quantification of bladder pressure during the voiding cycle would make an important contribution to the management of lower urinary tract symptoms. We developed a new noninvasive test to measure bladder pressure in males based on controlled inflation of a penile cuff during voiding. We compared the new technique with simultaneous invasive bladder pressure measurement. Materials and Methods: We evaluated 7 volunteers and 32 patients. A conventional pressure flow study was performed first. The bladder was refilled, a penile cuff was fitted and after voiding commenced the cuff was inflated in steps of 10 cm. water every 0.75 seconds until urine flow was interrupted. The cuff was rapidly deflated, allowing flow to resume, and the cycle was repeated until the end of voiding. The flow rate was graphed against cuff pressure for each interruption cycle to determine the pressure at which flow was interrupted. This pressure was compared with simultaneous invasive isovolumetric bladder pressure. Results: Invasive and noninvasive pressure measurements agreed well. Average cuff pressure at interruption of flow exceeded mean simultaneous isovolumetric bladder pressure plus or minus standard deviation by 14.5 ⫾ 14.0 cm. water. Conclusions: The new method provides noninvasive quantitative information on voiding bladder pressure in males. Further study is required to assess whether the technique can contribute to the management of lower urinary tract symptoms. KEY WORDS: bladder, urination, bladder neck obstruction, urodynamics, prostatic hyperplasia
A poor urine flow rate in men may be due to decreased bladder contractility or bladder outlet obstruction. To ascertain the cause information on bladder pressure during voiding can be acquired from an invasive pressure flow study.1, 2 However, a pressure flow study is expensive, timeconsuming, uncomfortable for the patient and causes a greater than 5% risk of symptomatic urinary infection.3 Consequently it is not routinely performed. There has been considerable interest in the noninvasive estimation of bladder pressure during voiding.4 –9 In each method flow is blocked artificially at the penis. Provided the urethra remains open, it acts as a fluid filled catheter and bladder pressure is transmitted to the location of the blockage. Even when there is restriction above the blockage, pressure builds to equal source pressure, as when a dripping tap is blocked with a thumb. Some groups applied a condom catheter with an outlet blocked midstream to interrupt urine flow.4, 5, 7, 8 Urine pressure is measured just proximal to the blockage via a side port. Despite some success the test is awkward, prone to leakage and is less effective at low flow rates.8, 10 McRae et al modeled their test on conventional blood pressure measurement.6 The patient attempts to void against an inflated penile cuff. The cuff is then deflated slowly until flow begins and cuff pressure is measured, followed by rapid deflation. The test is easily applied and is not susceptible to leakage. However, only a single measurement is obtained and false starts can occur.11 We developed a new technique using controlled inflation of a cuff, as described previously with preliminary results.12 After voiding begins we inflate the cuff until flow is interAccepted for publication October 19, 2001. Supported by Mediplus Ltd. for a patent application. * Financial interest and/or other relationship with Mediplus.
rupted. We measure corresponding cuff pressure and then rapidly deflate the cuff, allowing voiding to resume. This cycle can be repeated until voiding ends. We compared the new noninvasive measurement with bladder pressure measured in a simultaneous invasive pressure flow study in volunteers and patients. MATERIALS AND METHODS
Ethics approval was obtained. We recruited 39 subjects, including 7 volunteers 34 to 53 years old from the research group and 32 patients 50 to 75 years old referred for a pressure flow study to investigate lower urinary tract symptoms. All participants provided informed consent. Each subject underwent conventional pressure flow study. The bladder was then refilled, the cuff was applied, and invasive and noninvasive measurements of bladder pressure were made simultaneously during the next voiding cycle. For the pressure flow study an 8Fr transurethral double lumen catheter (Mediplus Ltd., High Wycombe, United Kingdom) was used for filling and bladder pressure measurement. A 6Fr manometer line covered with a vented finger cot was positioned in the rectum for recording abdominal pressure. The fluid filled manometer lines were connected to external transducers at the level of the public symphysis and zero referenced to atmospheric pressure. The bladder was filled with saline at a rate of 50 ml. per minute. We used a load cell flow meter (Gaeltec, Isle of Skye, United Kingdom). We then performed the noninvasive technique. In patients the cuff technique was explained (fig. 1). A pediatric blood pressure cuff (Critikon, Johnson & Johnson, Arlington, Texas) was fitted around the penis, using the largest size suitable (3.7, 4.6 or 5.4 cm.)13 After refilling the bladder the patient was instructed to void, if possible without abdominal
1344
NONINVASIVE BLADDER PRESSURE MEASUREMENT
FIG. 1. Principle of cuff inflation technique. After flow commences controlled inflation of cuff is started at steps of 10 cm. water every 0.75 second. When cuff pressure exceeds bladder pressure, flow is interrupted. Cuff is then deflated, allowing flow to resume. Cycle can be repeated until end of voiding.
straining. With voiding under way the pressure in the cuff was increased by 10 cm. water every 0.75 seconds using a purposely built pneumatic pressure controller (Regional Medical Physics Department, Newcastle upon Tyne, United Kingdom). When urine flow was interrupted or a limit of 200 cm. water was achieved, the cuff was rapidly deflated to allow resumption of voiding. The inflation-deflation cycle was repeated until the end of voiding. In volunteers the same procedure was followed except 3 fill-void cycles were also performed to assess the effect of the inflation rate. We used 0.5, 0.75 or 1-second step intervals in random order. We analyzed only interruption cycles followed by flow rate recovery, which excluded recordings at the termination of voiding. For each interruption cycle a graph was plotted of the flow rate (vertical axis) versus cuff pressure (horizontal axis) after compensating for an experimentally determined 0.9-second delay in the flow rate due to mechanical and electronic delays in the flowmeter. The cuff pressure at which flow ceased was measured from this graph and simultaneous isovolumetric bladder pressure was measured from the invasive pressure flow study. Cuff pressure at which flow ceased and isovolumetric bladder pressure were compared for all analyzed interruption cycles using the method of Bland and Altman,14 including calculation of the mean plus or minus standard deviation (SD) of the differences. The effect of the step interval of 0.5, 0.75 or 1 second in volunteers was examined using analysis of variance. Abdominal pressure during voiding per subject was measured, and the overall mean and SD was calculated. All values are shown as the mean plus or minus SD unless otherwise stated. RESULTS
Data were successfully recorded on all 7 volunteers and 27 patients. A single patient was unable to void, 2 strained excessively, preventing analysis, and in 2 the flow rate did not recover after any of the inflation-deflation cycles. A total of 115 inflation cycles were included in analysis, that is 63 in patients and 52 in volunteers. There were 15, 26 and 11 cycles at 0.5, 0.75 and 1 second steps, respectively. Figure 2 shows a typical voiding cycle in a volunteer with 3 flow interruptions. The flow rate decreased to zero when cuff pressure exceeded bladder pressure. The invasive recording confirmed that the bladder contraction was maintained during the interruption and the resumption of flow each time that cuff pressure was released indicates that the urethra remained open. Figure 3 shows 2 examples of flow rate graphed against cuff pressure, which were used to measure the pressure at which flow was decreased to zero. Overall cuff pressure at which flow ceased exceeded isovolumetric bladder pressure by an average of 14.5 ⫾ 14.0 cm. water with a value of 15.6 ⫾ 10.0 and 13.5 ⫾ 16.5 cm. water
1345
FIG. 2. Invasive and noninvasive data on voiding cycle of volunteer with 3 flow interruptions when cuff pressure (Pcuff) exceeded bladder pressure (Pves). Note how bladder presure increased to isovolumetric bladder pressure as interruption approached. Pabd, abdominal pressure. ml/s, ml. per second.
FIG. 3. Typical graphs of flow rate versus cuff pressure in volunteer and in symptomatic patient show relatively unaffected flow until cuff pressure achieved certain level. ml/s, ml. per second.
in the volunteer and patient groups, respectively (figs. 4 and 5). The mean inflation rate in the volunteer group was 15.2 ⫾ 8.7, 15.9 ⫾ 10.8 and 15.6 ⫾ 10.6 cm. water for 0.5, 0.75 and 1 second, respectively. One-way analysis of variance revealed no significant difference among the 3 inflation step durations in volunteers (p ⫽ 0.98). Figures 4 and 5 show pooled data. Average abdominal pressure in all subjects was 37 ⫾ 10 cm. water. DISCUSSION
Does the new method work? Figures 4 and 5 show good agreement between cuff pressure at which flow ceased with isovolumetric bladder pressure. Most of the average overestimation in volunteers and patients was accounted for by the height difference in the bladder and cuff. The SD error of ⫾ 14 cm. water compares favorably with other published noninvasive data and it is similar to the error observed when measuring penile urethral pressure directly under the cuff.13 Further investigation is required to identify whether it is an inherent limitation of the technique or whether cuff design optimization would help. There is evidence of a systematic component of the error in individuals.13 Why does it work? Several postulates underlie the technique. Their validity is supported by the effectiveness of the
1346
NONINVASIVE BLADDER PRESSURE MEASUREMENT
FIG. 4. Scattergram of cuff pressure at interruption (Pcuff, int) versus simultaneously measured isovolumetric bladder pressure (Pves, isv).
FIG. 5. Figure 4 data at flow interruption plotted using Bland and Altman method14 shows mean of 14.5 ⫾ 14 cm. water. Pcuff,int, cuff pressure at interruption. Pves,isv, isovolumetric bladder pressure.
method but there is additional supporting evidence. There is published evidence that bladder contraction is maintained during flow interruption.8, 15 Our recordings confirm that pressure just after interruption equaled pressure before interruption started.16 Also in many subjects bladder pressure increased during interruption, as predicted from theory (fig. 2).17, 18 The technique requires the urethra upstream from the interruption to remain open when flow stops, so that it acts as a fluid filled manometer tube. Radiographic imaging confirms this finding.7, 8 We assumed that flow was interrupted when pressure in the urethra within the cuff exceeded bladder pressure, allowing for the height difference. This assumption equates to that made for noninvasive blood pressure measurement and it is consistent with the Griffiths definition of urethral closure pressure.17 Thus, the accuracy of the method depends on how well cuff pressure is transmitted to the penile urethra. At the mid point of wider cuffs cuff pressure agreed well with pressure inside the penile urethra when measured using a solid state sensor.13 Pressure profile measurements showed that toward the cuff edges the pressure decreased. For narrow
cuffs, pressure was low even at the mid point. For blood pressure measurement cuff width should be 120% to 150% of the diameter of the arm. Data indicate that equivalent advice applies to the penis, although it is not always practicable when the penis is short. As the cuff is inflated, the upstream urethra experiences increasing pressure and because it is compliant, it expands. If cuff pressure is raised too rapidly, particularly at a low flow rate, bladder flow may be completely diverted to the expanding urethra, so that flow stops prematurely at the external meatus, leading to error. In addition, the flow meter has a limited ability to respond to a rapidly changing flow rate. These reasons may explain why van Mastrigt and Pel reported that interrupting flow by rapid inflation of a penile cuff provided unreliable results.9 In practice we noted that steps of 10 cm. water every 0.75 second were a reasonable compromise. Optimization of inflation characteristics requires further investigation. A linear increase in pressure at a similar overall rate may prove to be suitable. Controlled deflation after interruption may be useful to for assessing the pressure at which flow resumes. Our analysis assumed that cuff pressure at which flow ceased was not affected by the double lumen catheter. It would be helpful to evaluate our assumption by making simultaneous measurements with a suprapubic catheter. Flowmeter damping causes difficulty when assessing the point of flow interruption from the flow rate trace. Since the graph of flow rate versus cuff pressure typically decreased in approximately linear fashion as interruption approached, extrapolation of the graph confirmed interruption pressure. We compared this method with other noninvasive methods. Most published studies of noninvasive bladder pressure measurement describe use of a condom catheter with a tap attached to interrupt flow and a side port to measure the pressure increase.4, 5, 7, 8 At higher flow rates the technique provided results comparable to those of the cuff inflation technique but at low flow rates it underestimates and is less reliable.8, 10 In our study cuff pressure at which flow ceased minus isovolumetric bladder pressure was 8.9 ⫾ 13.6 cm. water for flow rates below 5.6 ml. per second10 for 21 inflations and 15.7 ⫾ 13.7 cm. water in the remaining 94, implying that the new technique does not have a particular problem at low flow rates. The cuff is rapidly and easily applied and the problem of leakage associated with a condom catheter is completely avoided. McRae6 and Gleason11 et al pioneered the use of a penile cuff with a deflation technique modeled on blood pressure measurement. A cuff fitted around the penis before voiding is inflated. After voiding should have commenced, it is slowly deflated. When flow begins, pressure is measured and the cuff is then rapidly deflated, providing a single measurement at the start of voiding. The mean overestimation compared with simultaneous invasive data was 36.5 ⫾ 16.2 cm. water. This variability was similar to that in our series but the systematic error was higher, possibly due to using a cuff that was only 2 cm. wide, which our results indicate transmits pressure less efficiently to the penile urethra.13 The deflation method also has some possible disadvantages.11 It depends on good patient understanding and cooperation. Starting early before bladder contraction is well established or the outlet is relaxed may lead to underestimation. Initiating voiding in the presence of maximum penile compression may not be conducive to normal voiding. In contrast, the inflation technique obtains measurements after normal voiding is established and multiple measurements can usually be made within each void, providing some indication of repeatability. We also assessed clinical applicability. Volunteers and patients reported either no discomfort during the test or brief discomfort at the final stage of interruption. All considered the new test preferable to catheterization. In this study with the bladder filled to capacity via a
NONINVASIVE BLADDER PRESSURE MEASUREMENT
catheter 1 patient did not void and 2 did not sustain satisfactory flow. The success rate of the test under natural filling conditions remains to be determined. It may prove advisable to determine bladder volume on ultrasound beforehand to improve the prospect of a satisfactory test. A disadvantage of all proposed noninvasive techniques is the lack of abdominal pressure measurement. In this study we measured a mean abdominal pressure of 37 ⫾ 10 cm. water, and so in principle an approximate correction may be applied. All except 2 patients understood and complied with the request not to strain during voiding. As indicated by invasive pressure recording, irregular flow rate data on those who strained indicated that the noninvasive data were suspect. We determined the potential relevance of this technique for clinical management. Urologists who practice urodynamics are familiar with using detrusor pressure at maximum flow as a measure of bladder contraction. Noninvasive techniques estimate isovolumetric pressure. According to well established theory isovolumetric pressure is a true measure of bladder contractility because it is by definition measured under the same conditions of zero flow.4, 9, 17 At low flow good bladder contractility on an invasive pressure flow study objectively diagnoses bladder outlet obstruction.1, 2 Preliminary application of noninvasive measurements in a similar way is encouraging.19 A further prospect for future research is the characteristic plateau region on the graphs, where the flow rate remained relatively constant during initial inflation (fig. 3). This plateau was evident in most subjects and is consistent with the flow controlling zone, as predicted from theoretical considerations.17, 18 It requires further study to investigate whether the pressure at which flow starts to decrease provides additional urodynamic information.20 CONCLUSIONS
The new method provides quantitative information on bladder pressure during voiding without catheterization. The pressure measured is isovolumetric bladder pressure, which is an accepted measure of bladder contractility. We recommend further research to assess whether the noninvasive measurement of bladder contractility differentiates poor bladder contraction from bladder outlet obstruction as the cause of low flow rate in an individual. Srs. Wendy Robson and Linda Kelly provided assistance, and Steve Fisher designed and developed the equipment. REFERENCES
1. Abrams, P.: Objective evaluation of bladder outlet obstruction. Br J Urol, suppl., 76: 11, 1995 2. Griffiths, D., Hofner, K., van Mastrigt, R. et al: Standardization of terminology of lower urinary tract funtion: pressure-flow studies of voiding, urethral resistance, and urethral obstruction. Neurourol Urodyn, 16: 1, 1997
1347
3. Klinger, H. C., Madersbacher, S., Djavan, B. et al: Morbidity of the evaluation of the lower urinary tract with transurethral multichannel pressure-flow studies. J Urol, 159: 191, 1998 4. Schafer, W., Kirschner-Hermans, R. and Jaske, G.: Non-invasive pressure/flow measurement for precise grading of bladder outflow obstruction. J Urol, part 2, 151: 323A, abstract 1994 5. van Mastrigt, R.: Non-invasive bladder pressure measurement. Methodology and reproducibility. Neurourol Urodyn, 14: 480, 1995 6. McRae, L. P., Bottaccini, M. R. and Gleason, D. M.: Non-invasive quantitative method for measuring isovolumetric bladder pressure and urethral resistance in the male: 1. Experimental validation of the theory. Neurourol, Urodyn, 14: 101, 1995 7. McCahy, P. J., Tweedie, R. J., Griffiths, C. J. et al: Clinical applicability of non-invasive pressure flow. Klin Fys, 4: 4, 1997 8. Gommer, E. D., Vanspauwen, T. J., Miklosi, M., et al: Validity of a non-invasive determination of the isovolumetric bladder pressure during voiding in men with LUTS. Neurourol Urodyn, 18: 477, 1999 9. van Mastrigt, R. and Pel, J. J.: Towards a non-invasive urodynamic diagnosis of infravesical obstruction. BJU Int, 84: 195, 1999 10. Pel, J. J. and van Mastrigt, R.: The accuracy of a non-invasive bladder pressure measurement with an external catheter. Neurourol Urodyn, 18: 251, 1999 11. Gleason, D. M., Bottaccini, M. R. and McRae, L. P.: Noninvasive urodynamics: a study of male voiding dysfunction. Neurourol Urodyn, 16: 93, 1997 12. Griffiths, C., Pickard, R., Robson, W. et al.: A new method for non-invasive assessment of bladder pressure during voiding compared with simultaneous invasive urodynamics. Neurourol Urodyn, 18: 253, 1999 13. Drinnan, M. J., Robson, W., Reddy, M. et al: Transmission of penile cuff pressure to the penile urethra. J Urol, 166: 2545, 2001 14. Bland, J. M. and Altman, D. G.: Statistical methods for assessing agreement between two methods of clinical measurement. Lancet, I: 307, 1986 15. Rikken, B., Pel, J. J. and van Mastrigt R.: Repeat noninvasive bladder pressure measurements with an external catheter. J Urol, 162: 474, 1999 16. McIntosh, S. L., Drinnan, M. J., Pickard R. S. et al: Non-invasive bladder pressure monitoring - how does interrupting the urinary stream affect intra-vesicle pressure? Neurourol Urodyn, 20: 382, 2001 17. Griffiths, D. J.: Hydrodynamics and mechanics of the bladder and urethra. In: Urodynamics: Principles, Practice and Application, 2nd ed. Edited by A. R. Mundy, T. P. Stephenson and A. J. Wein. New York: Churchill Livingstone, chapt. 5, p. 71, 1994 18. Schafer, W.: Urethral resistance? Urodynamic concepts of physiological and pathological bladder outlet function during voiding. Neurourol Urodyn, 4: 161, 1985 19. Griffiths, C. J., Rix, D., Macdonald, A. et al: Can non-invasive bladder measurements identify men with bladder outflow obstruction? Neurourol Urodyn, 19: 429, 2000 20. Drinnan, M. J., Pickard, R. S., Ramsden, P. D. et al: An experimental model reproducing the response of urine flow rate to penile cuff pressure observed in men. Neurourol Urodyn, 19: 486, 2000
Vol. 167, 1344 –1347, March 2002 Printed in U.S.A.
NONINVASIVE MEASUREMENT OF BLADDER PRESSURE BY CONTROLLED INFLATION OF A PENILE CUFF C. J. GRIFFITHS,* D. RIX, A. M. MACDONALD, M. J. DRINNAN, R. S. PICKARD P. D. RAMSDEN
AND
From the Regional Medical Physics Department and Department of Urology, Freeman Hospital, Newcastle upon Tyne, United Kingdom
ABSTRACT
Purpose: A noninvasive test providing reliable objective quantification of bladder pressure during the voiding cycle would make an important contribution to the management of lower urinary tract symptoms. We developed a new noninvasive test to measure bladder pressure in males based on controlled inflation of a penile cuff during voiding. We compared the new technique with simultaneous invasive bladder pressure measurement. Materials and Methods: We evaluated 7 volunteers and 32 patients. A conventional pressure flow study was performed first. The bladder was refilled, a penile cuff was fitted and after voiding commenced the cuff was inflated in steps of 10 cm. water every 0.75 seconds until urine flow was interrupted. The cuff was rapidly deflated, allowing flow to resume, and the cycle was repeated until the end of voiding. The flow rate was graphed against cuff pressure for each interruption cycle to determine the pressure at which flow was interrupted. This pressure was compared with simultaneous invasive isovolumetric bladder pressure. Results: Invasive and noninvasive pressure measurements agreed well. Average cuff pressure at interruption of flow exceeded mean simultaneous isovolumetric bladder pressure plus or minus standard deviation by 14.5 ⫾ 14.0 cm. water. Conclusions: The new method provides noninvasive quantitative information on voiding bladder pressure in males. Further study is required to assess whether the technique can contribute to the management of lower urinary tract symptoms. KEY WORDS: bladder, urination, bladder neck obstruction, urodynamics, prostatic hyperplasia
A poor urine flow rate in men may be due to decreased bladder contractility or bladder outlet obstruction. To ascertain the cause information on bladder pressure during voiding can be acquired from an invasive pressure flow study.1, 2 However, a pressure flow study is expensive, timeconsuming, uncomfortable for the patient and causes a greater than 5% risk of symptomatic urinary infection.3 Consequently it is not routinely performed. There has been considerable interest in the noninvasive estimation of bladder pressure during voiding.4 –9 In each method flow is blocked artificially at the penis. Provided the urethra remains open, it acts as a fluid filled catheter and bladder pressure is transmitted to the location of the blockage. Even when there is restriction above the blockage, pressure builds to equal source pressure, as when a dripping tap is blocked with a thumb. Some groups applied a condom catheter with an outlet blocked midstream to interrupt urine flow.4, 5, 7, 8 Urine pressure is measured just proximal to the blockage via a side port. Despite some success the test is awkward, prone to leakage and is less effective at low flow rates.8, 10 McRae et al modeled their test on conventional blood pressure measurement.6 The patient attempts to void against an inflated penile cuff. The cuff is then deflated slowly until flow begins and cuff pressure is measured, followed by rapid deflation. The test is easily applied and is not susceptible to leakage. However, only a single measurement is obtained and false starts can occur.11 We developed a new technique using controlled inflation of a cuff, as described previously with preliminary results.12 After voiding begins we inflate the cuff until flow is interAccepted for publication October 19, 2001. Supported by Mediplus Ltd. for a patent application. * Financial interest and/or other relationship with Mediplus.
rupted. We measure corresponding cuff pressure and then rapidly deflate the cuff, allowing voiding to resume. This cycle can be repeated until voiding ends. We compared the new noninvasive measurement with bladder pressure measured in a simultaneous invasive pressure flow study in volunteers and patients. MATERIALS AND METHODS
Ethics approval was obtained. We recruited 39 subjects, including 7 volunteers 34 to 53 years old from the research group and 32 patients 50 to 75 years old referred for a pressure flow study to investigate lower urinary tract symptoms. All participants provided informed consent. Each subject underwent conventional pressure flow study. The bladder was then refilled, the cuff was applied, and invasive and noninvasive measurements of bladder pressure were made simultaneously during the next voiding cycle. For the pressure flow study an 8Fr transurethral double lumen catheter (Mediplus Ltd., High Wycombe, United Kingdom) was used for filling and bladder pressure measurement. A 6Fr manometer line covered with a vented finger cot was positioned in the rectum for recording abdominal pressure. The fluid filled manometer lines were connected to external transducers at the level of the public symphysis and zero referenced to atmospheric pressure. The bladder was filled with saline at a rate of 50 ml. per minute. We used a load cell flow meter (Gaeltec, Isle of Skye, United Kingdom). We then performed the noninvasive technique. In patients the cuff technique was explained (fig. 1). A pediatric blood pressure cuff (Critikon, Johnson & Johnson, Arlington, Texas) was fitted around the penis, using the largest size suitable (3.7, 4.6 or 5.4 cm.)13 After refilling the bladder the patient was instructed to void, if possible without abdominal
1344
NONINVASIVE BLADDER PRESSURE MEASUREMENT
FIG. 1. Principle of cuff inflation technique. After flow commences controlled inflation of cuff is started at steps of 10 cm. water every 0.75 second. When cuff pressure exceeds bladder pressure, flow is interrupted. Cuff is then deflated, allowing flow to resume. Cycle can be repeated until end of voiding.
straining. With voiding under way the pressure in the cuff was increased by 10 cm. water every 0.75 seconds using a purposely built pneumatic pressure controller (Regional Medical Physics Department, Newcastle upon Tyne, United Kingdom). When urine flow was interrupted or a limit of 200 cm. water was achieved, the cuff was rapidly deflated to allow resumption of voiding. The inflation-deflation cycle was repeated until the end of voiding. In volunteers the same procedure was followed except 3 fill-void cycles were also performed to assess the effect of the inflation rate. We used 0.5, 0.75 or 1-second step intervals in random order. We analyzed only interruption cycles followed by flow rate recovery, which excluded recordings at the termination of voiding. For each interruption cycle a graph was plotted of the flow rate (vertical axis) versus cuff pressure (horizontal axis) after compensating for an experimentally determined 0.9-second delay in the flow rate due to mechanical and electronic delays in the flowmeter. The cuff pressure at which flow ceased was measured from this graph and simultaneous isovolumetric bladder pressure was measured from the invasive pressure flow study. Cuff pressure at which flow ceased and isovolumetric bladder pressure were compared for all analyzed interruption cycles using the method of Bland and Altman,14 including calculation of the mean plus or minus standard deviation (SD) of the differences. The effect of the step interval of 0.5, 0.75 or 1 second in volunteers was examined using analysis of variance. Abdominal pressure during voiding per subject was measured, and the overall mean and SD was calculated. All values are shown as the mean plus or minus SD unless otherwise stated. RESULTS
Data were successfully recorded on all 7 volunteers and 27 patients. A single patient was unable to void, 2 strained excessively, preventing analysis, and in 2 the flow rate did not recover after any of the inflation-deflation cycles. A total of 115 inflation cycles were included in analysis, that is 63 in patients and 52 in volunteers. There were 15, 26 and 11 cycles at 0.5, 0.75 and 1 second steps, respectively. Figure 2 shows a typical voiding cycle in a volunteer with 3 flow interruptions. The flow rate decreased to zero when cuff pressure exceeded bladder pressure. The invasive recording confirmed that the bladder contraction was maintained during the interruption and the resumption of flow each time that cuff pressure was released indicates that the urethra remained open. Figure 3 shows 2 examples of flow rate graphed against cuff pressure, which were used to measure the pressure at which flow was decreased to zero. Overall cuff pressure at which flow ceased exceeded isovolumetric bladder pressure by an average of 14.5 ⫾ 14.0 cm. water with a value of 15.6 ⫾ 10.0 and 13.5 ⫾ 16.5 cm. water
1345
FIG. 2. Invasive and noninvasive data on voiding cycle of volunteer with 3 flow interruptions when cuff pressure (Pcuff) exceeded bladder pressure (Pves). Note how bladder presure increased to isovolumetric bladder pressure as interruption approached. Pabd, abdominal pressure. ml/s, ml. per second.
FIG. 3. Typical graphs of flow rate versus cuff pressure in volunteer and in symptomatic patient show relatively unaffected flow until cuff pressure achieved certain level. ml/s, ml. per second.
in the volunteer and patient groups, respectively (figs. 4 and 5). The mean inflation rate in the volunteer group was 15.2 ⫾ 8.7, 15.9 ⫾ 10.8 and 15.6 ⫾ 10.6 cm. water for 0.5, 0.75 and 1 second, respectively. One-way analysis of variance revealed no significant difference among the 3 inflation step durations in volunteers (p ⫽ 0.98). Figures 4 and 5 show pooled data. Average abdominal pressure in all subjects was 37 ⫾ 10 cm. water. DISCUSSION
Does the new method work? Figures 4 and 5 show good agreement between cuff pressure at which flow ceased with isovolumetric bladder pressure. Most of the average overestimation in volunteers and patients was accounted for by the height difference in the bladder and cuff. The SD error of ⫾ 14 cm. water compares favorably with other published noninvasive data and it is similar to the error observed when measuring penile urethral pressure directly under the cuff.13 Further investigation is required to identify whether it is an inherent limitation of the technique or whether cuff design optimization would help. There is evidence of a systematic component of the error in individuals.13 Why does it work? Several postulates underlie the technique. Their validity is supported by the effectiveness of the
1346
NONINVASIVE BLADDER PRESSURE MEASUREMENT
FIG. 4. Scattergram of cuff pressure at interruption (Pcuff, int) versus simultaneously measured isovolumetric bladder pressure (Pves, isv).
FIG. 5. Figure 4 data at flow interruption plotted using Bland and Altman method14 shows mean of 14.5 ⫾ 14 cm. water. Pcuff,int, cuff pressure at interruption. Pves,isv, isovolumetric bladder pressure.
method but there is additional supporting evidence. There is published evidence that bladder contraction is maintained during flow interruption.8, 15 Our recordings confirm that pressure just after interruption equaled pressure before interruption started.16 Also in many subjects bladder pressure increased during interruption, as predicted from theory (fig. 2).17, 18 The technique requires the urethra upstream from the interruption to remain open when flow stops, so that it acts as a fluid filled manometer tube. Radiographic imaging confirms this finding.7, 8 We assumed that flow was interrupted when pressure in the urethra within the cuff exceeded bladder pressure, allowing for the height difference. This assumption equates to that made for noninvasive blood pressure measurement and it is consistent with the Griffiths definition of urethral closure pressure.17 Thus, the accuracy of the method depends on how well cuff pressure is transmitted to the penile urethra. At the mid point of wider cuffs cuff pressure agreed well with pressure inside the penile urethra when measured using a solid state sensor.13 Pressure profile measurements showed that toward the cuff edges the pressure decreased. For narrow
cuffs, pressure was low even at the mid point. For blood pressure measurement cuff width should be 120% to 150% of the diameter of the arm. Data indicate that equivalent advice applies to the penis, although it is not always practicable when the penis is short. As the cuff is inflated, the upstream urethra experiences increasing pressure and because it is compliant, it expands. If cuff pressure is raised too rapidly, particularly at a low flow rate, bladder flow may be completely diverted to the expanding urethra, so that flow stops prematurely at the external meatus, leading to error. In addition, the flow meter has a limited ability to respond to a rapidly changing flow rate. These reasons may explain why van Mastrigt and Pel reported that interrupting flow by rapid inflation of a penile cuff provided unreliable results.9 In practice we noted that steps of 10 cm. water every 0.75 second were a reasonable compromise. Optimization of inflation characteristics requires further investigation. A linear increase in pressure at a similar overall rate may prove to be suitable. Controlled deflation after interruption may be useful to for assessing the pressure at which flow resumes. Our analysis assumed that cuff pressure at which flow ceased was not affected by the double lumen catheter. It would be helpful to evaluate our assumption by making simultaneous measurements with a suprapubic catheter. Flowmeter damping causes difficulty when assessing the point of flow interruption from the flow rate trace. Since the graph of flow rate versus cuff pressure typically decreased in approximately linear fashion as interruption approached, extrapolation of the graph confirmed interruption pressure. We compared this method with other noninvasive methods. Most published studies of noninvasive bladder pressure measurement describe use of a condom catheter with a tap attached to interrupt flow and a side port to measure the pressure increase.4, 5, 7, 8 At higher flow rates the technique provided results comparable to those of the cuff inflation technique but at low flow rates it underestimates and is less reliable.8, 10 In our study cuff pressure at which flow ceased minus isovolumetric bladder pressure was 8.9 ⫾ 13.6 cm. water for flow rates below 5.6 ml. per second10 for 21 inflations and 15.7 ⫾ 13.7 cm. water in the remaining 94, implying that the new technique does not have a particular problem at low flow rates. The cuff is rapidly and easily applied and the problem of leakage associated with a condom catheter is completely avoided. McRae6 and Gleason11 et al pioneered the use of a penile cuff with a deflation technique modeled on blood pressure measurement. A cuff fitted around the penis before voiding is inflated. After voiding should have commenced, it is slowly deflated. When flow begins, pressure is measured and the cuff is then rapidly deflated, providing a single measurement at the start of voiding. The mean overestimation compared with simultaneous invasive data was 36.5 ⫾ 16.2 cm. water. This variability was similar to that in our series but the systematic error was higher, possibly due to using a cuff that was only 2 cm. wide, which our results indicate transmits pressure less efficiently to the penile urethra.13 The deflation method also has some possible disadvantages.11 It depends on good patient understanding and cooperation. Starting early before bladder contraction is well established or the outlet is relaxed may lead to underestimation. Initiating voiding in the presence of maximum penile compression may not be conducive to normal voiding. In contrast, the inflation technique obtains measurements after normal voiding is established and multiple measurements can usually be made within each void, providing some indication of repeatability. We also assessed clinical applicability. Volunteers and patients reported either no discomfort during the test or brief discomfort at the final stage of interruption. All considered the new test preferable to catheterization. In this study with the bladder filled to capacity via a
NONINVASIVE BLADDER PRESSURE MEASUREMENT
catheter 1 patient did not void and 2 did not sustain satisfactory flow. The success rate of the test under natural filling conditions remains to be determined. It may prove advisable to determine bladder volume on ultrasound beforehand to improve the prospect of a satisfactory test. A disadvantage of all proposed noninvasive techniques is the lack of abdominal pressure measurement. In this study we measured a mean abdominal pressure of 37 ⫾ 10 cm. water, and so in principle an approximate correction may be applied. All except 2 patients understood and complied with the request not to strain during voiding. As indicated by invasive pressure recording, irregular flow rate data on those who strained indicated that the noninvasive data were suspect. We determined the potential relevance of this technique for clinical management. Urologists who practice urodynamics are familiar with using detrusor pressure at maximum flow as a measure of bladder contraction. Noninvasive techniques estimate isovolumetric pressure. According to well established theory isovolumetric pressure is a true measure of bladder contractility because it is by definition measured under the same conditions of zero flow.4, 9, 17 At low flow good bladder contractility on an invasive pressure flow study objectively diagnoses bladder outlet obstruction.1, 2 Preliminary application of noninvasive measurements in a similar way is encouraging.19 A further prospect for future research is the characteristic plateau region on the graphs, where the flow rate remained relatively constant during initial inflation (fig. 3). This plateau was evident in most subjects and is consistent with the flow controlling zone, as predicted from theoretical considerations.17, 18 It requires further study to investigate whether the pressure at which flow starts to decrease provides additional urodynamic information.20 CONCLUSIONS
The new method provides quantitative information on bladder pressure during voiding without catheterization. The pressure measured is isovolumetric bladder pressure, which is an accepted measure of bladder contractility. We recommend further research to assess whether the noninvasive measurement of bladder contractility differentiates poor bladder contraction from bladder outlet obstruction as the cause of low flow rate in an individual. Srs. Wendy Robson and Linda Kelly provided assistance, and Steve Fisher designed and developed the equipment. REFERENCES
1. Abrams, P.: Objective evaluation of bladder outlet obstruction. Br J Urol, suppl., 76: 11, 1995 2. Griffiths, D., Hofner, K., van Mastrigt, R. et al: Standardization of terminology of lower urinary tract funtion: pressure-flow studies of voiding, urethral resistance, and urethral obstruction. Neurourol Urodyn, 16: 1, 1997
1347
3. Klinger, H. C., Madersbacher, S., Djavan, B. et al: Morbidity of the evaluation of the lower urinary tract with transurethral multichannel pressure-flow studies. J Urol, 159: 191, 1998 4. Schafer, W., Kirschner-Hermans, R. and Jaske, G.: Non-invasive pressure/flow measurement for precise grading of bladder outflow obstruction. J Urol, part 2, 151: 323A, abstract 1994 5. van Mastrigt, R.: Non-invasive bladder pressure measurement. Methodology and reproducibility. Neurourol Urodyn, 14: 480, 1995 6. McRae, L. P., Bottaccini, M. R. and Gleason, D. M.: Non-invasive quantitative method for measuring isovolumetric bladder pressure and urethral resistance in the male: 1. Experimental validation of the theory. Neurourol, Urodyn, 14: 101, 1995 7. McCahy, P. J., Tweedie, R. J., Griffiths, C. J. et al: Clinical applicability of non-invasive pressure flow. Klin Fys, 4: 4, 1997 8. Gommer, E. D., Vanspauwen, T. J., Miklosi, M., et al: Validity of a non-invasive determination of the isovolumetric bladder pressure during voiding in men with LUTS. Neurourol Urodyn, 18: 477, 1999 9. van Mastrigt, R. and Pel, J. J.: Towards a non-invasive urodynamic diagnosis of infravesical obstruction. BJU Int, 84: 195, 1999 10. Pel, J. J. and van Mastrigt, R.: The accuracy of a non-invasive bladder pressure measurement with an external catheter. Neurourol Urodyn, 18: 251, 1999 11. Gleason, D. M., Bottaccini, M. R. and McRae, L. P.: Noninvasive urodynamics: a study of male voiding dysfunction. Neurourol Urodyn, 16: 93, 1997 12. Griffiths, C., Pickard, R., Robson, W. et al.: A new method for non-invasive assessment of bladder pressure during voiding compared with simultaneous invasive urodynamics. Neurourol Urodyn, 18: 253, 1999 13. Drinnan, M. J., Robson, W., Reddy, M. et al: Transmission of penile cuff pressure to the penile urethra. J Urol, 166: 2545, 2001 14. Bland, J. M. and Altman, D. G.: Statistical methods for assessing agreement between two methods of clinical measurement. Lancet, I: 307, 1986 15. Rikken, B., Pel, J. J. and van Mastrigt R.: Repeat noninvasive bladder pressure measurements with an external catheter. J Urol, 162: 474, 1999 16. McIntosh, S. L., Drinnan, M. J., Pickard R. S. et al: Non-invasive bladder pressure monitoring - how does interrupting the urinary stream affect intra-vesicle pressure? Neurourol Urodyn, 20: 382, 2001 17. Griffiths, D. J.: Hydrodynamics and mechanics of the bladder and urethra. In: Urodynamics: Principles, Practice and Application, 2nd ed. Edited by A. R. Mundy, T. P. Stephenson and A. J. Wein. New York: Churchill Livingstone, chapt. 5, p. 71, 1994 18. Schafer, W.: Urethral resistance? Urodynamic concepts of physiological and pathological bladder outlet function during voiding. Neurourol Urodyn, 4: 161, 1985 19. Griffiths, C. J., Rix, D., Macdonald, A. et al: Can non-invasive bladder measurements identify men with bladder outflow obstruction? Neurourol Urodyn, 19: 429, 2000 20. Drinnan, M. J., Pickard, R. S., Ramsden, P. D. et al: An experimental model reproducing the response of urine flow rate to penile cuff pressure observed in men. Neurourol Urodyn, 19: 486, 2000