HD The Corporation (2003) | Deluded Mind-CODEX | Daemonen.der.Seele.1972.German.DL.1080p.BluRa

Urethral Pressure Profile in Female Stress Incontinence

Urethral Pressure Profile in Female Stress Incontinence

0022-5347 /W/1202-8...

225KB Sizes 2 Downloads 20 Views

0022-5347 /W/1202-8<75$02 00/0 Vol. 120, OctcDe:r



Copyright © 1978 by The V.lilliams & VVilkins Co.






From the Department of Urology and Biomedical Engineering Unit, Queen's University, Kingston, Ontario, Canada


The accuracy of the urethral pressure profile as a measure of sphincteric competence was examined in female subjects. Most profile measurements selected proved to be significantly different in patients with stress incontinence from those in controls. However, the measurement that seemed to have the highest potential for diagnostic accuracy was the maximum closure pressure in the continence zone, recorded with the bladder full and the patient standing. The concept of the continence zone and incorporating the effect of standing were believed to be the main reasons for this high accuracy. The second best measurement was the maximum closure pressure with the bladder full and the patient supine. To lessen the chances of a diagnostic error it was recommended that both of these measurements should be obtained. The physiological implications of these findings and the clinical role of the urethral pressure profile examination in the assessment of female patients with urinary incontinence are discussed. A precise determination of the competence of the sphincteric mechanism may be essential for the successful management of female patients with urinary incontinence. In the presence of a typical clinical history a positive Marshall-Marchetti (Bonney) test, abnormal angles on the stress cystogram and a normal cystometrogram the diagnosis of stress (sphincteric) incontinence can be made confidently and treatment usually is successful. However, it is not uncommon to encounter patients with urinary incontinence in whom these findings are inconclusive or not uniformly positive. In these patients a clear-cut diagnosis cannot be made and inappropriate treatment might be instituted. Thus, it is generally accepted that there is a need for a method with a high degree of accuracy to measure the competence of the sphincteric mechanism. In recent years the accuracy of the urethral pressure profile as a measure of urethral sphincter competence has been examined. It was reported that in patients with stress incontinence the maximum urethral and closure pressures (as defined by the International Continence Society)' were significantly lower than in controls. 2- 5 However, it also was indicated that there was an overlap between the measurements of the 2 groups, which significantly limited the diagnostic value of the test. Gleason and associates also examined 2 other measurements: 1) the area under the profile curve and 2) the mean pressure (area divided by urethral length), and found that a similar limitation existed. 5 However, the was minimal when these same measurements were taken from the continence zone alone. During the clinical assessment of patients with urinary incontinence the effect of physical stress or activity on the symptoms usually is determined. Also, during the pelvic examination the effect of bearing down or coughing is tested regularly. Coughing is an integral part of the Marshall test and the results are more often positive when the patient is standing than supine. When the stress cystogram is performed the effect of stress on the urethrovesical angle and the angle of inclination are measured routinely. Some investigators have examined the effect of coughing as a physical stress on the urethral pressure profile 2 • 6 and found that the increase in intraurethral pressure was significantly lower in patients with stress incontinence than in controls and, in some instances, a decrease rather than an increase in pressure was obtained. In a previous study it was observed that standing Accepted for publication January 27, 1978. Read at annual meeting of American Urological Association, Chicago, Illinois, April 24-28, 1977. 475

produces a steady and significant increase in intravesical pressure' but the effect of standing on the urethral profile pressures was not examined. The objectives of our study were to determine which of the profile measurements have the greatest diagnostic accuracy and to find out if there is any advantage in doing the profile with patients in the standing position. MATERIAL AND METHOD

Twenty women with stress incontinence and 10 controls were studied. The patients had typical histories of stress incontinence, positive Marshall tests, abnormal angles on the 8 normal neurological stress cystograms (Green type I or examinations and normal cystometrograms. The controls were continent and neurologically normal. This group consisted largely of patients who presented with frequency or recurrent lower urinary tract infection. The ranges and means of the ages of the 2 groups were comparable, the range in patients with stress incontinence being 32 to 73 years (mean 55) and in controls 22 to 78 years (mean 47). The parity of these patients also was similar. At the time of evaluation all patients had negative urine cultures, normal excretory urograms and negative cystoscopic examinations. The urethral pressure profile was done the Brown and Wickham method,9 modified by using an apparatus that withdrew the catheter at a of 1 cm. per minute. 10 The catheter was perfused vvith sterile water at a rate of 4. 7 ml. per minute using a Harvard pump. The catheters consisted of a regular latex 12F catheter with a 5F feeding tube inserted through its lumen. The feeding tube protruded approximately 8 cm. beyond the catheter tip. Radiopaque paste was used to fill the lumen of the tip of the latex catheter around the feeding tube for approximately 1 cm. This obliterated the most distal side hole of the latex catheter, thus leaving only 1 oval side hole (1 x 2 mm.) approximately 2 cm. from the tip. The radiopaque paste allowed fluoroscopic monitoring of the position of the tip of the catheter during examination. The intravesical and urethral pressures were recorded through Statham pressure transducers on a Grass polygraph. 1 The latex catheter recorded the urethral pressure and the feeding tube recorded the bladder pressure. The chart speed was matched to the catheter withdrawal speed to obtain a 1:1 graphic representation of urethral length. The profile examinations were done first with the patient in the supine position with an empty bladder. After the cystometrogram and with the bladder containing between 350 and



400 ml. the profile was repeated with the patient in the supine position and then standing. These 3 sets of profiles will be referred to as: 1) empty supine, 2) full supine and 3) full standing, respectively. In each instance the profile examination was done twice and the averages of the 2 sets of measurements were used. Adequate stability of the catheter system during standing was made possible by using an x-ray table that can be tilted vertically and to which the catheter withdrawal apparatus was attached (fig. 1). The continence line was determined in each examination as suggested by Gleason and associates. 5 Technically, however, we found that the determination of the position of the continence line, which is based on the prompt detection of the water leak, was awkward and inaccurate with the patient standing. Therefore, in the standing profiles the continence line was arbitrarily drawn at the peak pressure point and when the maximum pressure was sustained (that is a plateau), the line was drawn when the pressure began to decrease. The continence zone was, thus, demarcated on all the profile curves. The following continence zone measurements were obtained: 1) length, 2) maximum closure pressure by subtracting the intravesical pressure from the maximum urethral pressure, 3) integrated pressure by measuring the surface area and 4) mean pressure by dividing the surface area by length. STATISTICAL TREATMENT OF DATA

Arithmetic means and standard deviations of all parameters

for the 2 groups were calculated (table 1). The following statistical tests were done: 1) The means of length in the patients with stress incontinence and control groups were subjected to a t test, which indicated a significant difference (p < 0.05) in full standing profiles only. 2) Comparison of means of the other parameters was complicated by significant differences in the variance (table 1 shows the values obtained for the standard deviations). After a logarithmic transformation was done, 11 the variances were stabilized and a t test could then be done to examine the difference between the means. (It was necessary to eliminate 2 values of zero from the maximum closure, integrated and mean pressures in full standing profiles before the transformation.) The difference was highly significant (p < 0.001). 3) The logarithmically transformed data for each parameter in each group did not deviate significantly from a normal distribution. Therefore, it was assumed that a normal distribution was appropriate for the populations from which these samples were drawn. These distributions were defined by the means and standard deviations of the samples. Since small samples were involved the values of the mean and standard deviation have large standard errors themselves. Therefore, the normal distribution will require review as more data become available. 4) To compare the potential of each parameter as a tool in the diagnosis of stress incontinence a measure of the parameter's efficiency in predicting such a· diagnosis is required. Histograms show that there is an overlap range with each parameter (figs. 2 and 3). Presumably, a small overlap indicates a high efficiency. This overlap was defined using the normal distribution established earlier. 5) Since the 2 populations are mutually exclusive any patient value obtained must be in one population or the other. We were prepared to accept a 1 per cent falsely positive or negative result with these parameters. Therefore, the 99 per cent confidence level was used to define the overlap range. Since all the uncertainty exists in 1 tail of the distribution only the 1 tail 99 per cent limits are appropriate. 6) The overlap range for each parameter is between the lower 99 per cent level of the controls A and the upper 99 per cent of the patients with stress incontinence B (fig. 4). The area outside the overlap range for each curve is shown in table 2 as a percentage. RESULTS

FIG. 1. Mechanical puller (white box) attached to x-ray table that can be tilted vertically. Puller is placed between patient's legs and patient rests feet on board. This setting provides adequate stability for standing profile.

Although the length measurements in the supine position (empty or full bladder) were lower in patients with stress incontinence than in the controls the difference was not statistically significant. These 2 measurements were excluded from further analysis. On the other hand, a significant difference was found in length- full standing measurements (p < 0.05) and in the remaining measurements (p < 0.001), which


1 Controls

Pts. With Stress Incontinence Mean Length (mm.): Empty supine* Full supine* Full standingt Maximum closure pressure (cm. water)::j: Empty supine Full supine Full standing Integrated pressure (cm. water mm.)::j: Empty supine Full supine Full standing Mean pressure (cm. water)::j: Empty supine Full supine Full standing

Standard Deviation

Standard Error


Standard Deviation

Standard Error

12.1 9.8 8.7

5.1 4.6 5.2

1.2 1.0 1.2

16.1 14.6 13.8

4.7 3.2 4.6

1.5 1.0 1.5

35.9 28.2 26.9

16.3 9.1 14.8

3.6 2.0 3.4

101 87.5 102.9

52.0 43.5 35.7

16.4 13.7 11.3

70.1 44.6 54.7

58.8 30.0 40.4

13.2 6.7 9.3

194.4 163.1 215.6

120.5 99.7 112.6

38.1 31.5 35.6

5.1 4.3 5.4

2.5 1.8 2.8

0.6 0.4 0.6

11.7 10.9 15.8

5.8 5.0 7.8

1.8 1.6 2.5

* Difference of means not significant. t Difference of means significant (p < 0.05). :j: Difference of means significant (p < 0.001) after logarithmic transformation.










200 190 180 170

26 25 24 23 22 21 20 19 18 17 16 15 14 13 12










000 000

7 00 6 0 5 4

0 0 00 0


3 2


[email protected] [email protected]® 0 0 '[email protected]® 000

000 000 000®€11 II 0 10 0

130 0


O® ©




0 @@



100 @


00®[email protected]



[email protected]



[email protected]® 0 0 000 00) 00







[email protected]

60 50




[email protected]

[email protected]











20 0







Fm. 2. Histograms of length and maximum closure pressure to show degree of overlap between 2 groups. 0 -stress incontinence. '111controls.



465 •50 035 420 405 390 [email protected] 375 360 345 330 315 300

180 165 150 ® 135 0 @ 120 OG 105 00© 90 75 0 60 0


[email protected]

45 0000 30 15 0000 0 0







37.~ @

e e @


. -






r ®'1!€1

0 [email protected]


00 0000


0 00

[email protected] 00000 0000 0 0 0000

36 "4.5 33 31.5 30 28.5 27

"" -


24 22 5 21




15 !3.5 12 10.5 9

[email protected]

®@ @


7.5 [email protected]!\

4~roooooo 3 oooooe 1.5- 000 0






[email protected] @


6® 0 0 000(')<:!l
[email protected]® 00000 [email protected] 00000 000 0 00

FIG. 3. Histograms of integrated pressure and mean pressure to show degree of overlap between 2 groups. 0- stress incontinence. controls.

indicates that these parameters correlate with the presence of stress incontinence. The overlaps between the 2 groups of the individual measurements were demonstrated. The overlap was smallest with the maximum closure pressure - full standing measurements and largest with the length measurements. Also, in general, the measurements obtained with the bladder empty appeared to have a larger overlap than those obtained with the bladder full (figs. 2 and 3). The potential value of the measurements in detecting stress incontinence in the population at large was determined as



described previously (see statistical analysis). The percentages listed in table 2 are the respective probabilities of detecting the presence of stress incontinence in patients with this condition and its absence in controls. The higher these percentages are the more valuable is the measurement as a diagnostic parameter. The maximum closure pressure-full standing appears to be superior to the rest and next to it the maximum closure pressure - full supine. No advantage is indicated in the length, integrated pressure or mean pressure measurements over the maximum closure pressure measurements.









ct -~






log parameter

FIG. 4. Normal distribution curves of controls (C) and patients with stress incontinence (SI) are shown with 99 per cent confidence limits A and B. Overlap range is defined by these 2 points.


2. Probability of measurement being outside overlap range Pts. With Stress Incontinence (%)

Length (full standing) Maximum closure pressure: Empty supine Full supine Full standing Integrated pressure: Empty supine Full supine Full standing Mean pressure: Empty supine Full supine Full standing

Controls (%)



38 45 83

50 75 65

31 54 57

13 31

15 45 37

35 41 47



Technical factors are significant in any urethral profile technique. We believe that a steady withdrawal of the catheter is essential. The speed we chose is similar to the one reported by Gleason and associates, that is 1 cm. per minute. 5 The perfusion rate was within the range described by other investigators. 12 • "' The reliability and reproducibility of our technique have been reported previously. 10 Although the urethral pressure profile has been used frequently as a research tool its value in clinical diagnosis has had limited testing. The present work indicates that the means of all the maximum closure pressure, integrated pressure, mean pressure and length-full standing correlate with the presence of stress incontinence. However, as a diagnostic test, a difference in means is not sufficient and it is essential to establish the degree of overlap between the measurements of patients with stress incontinence and controls, and the likelihood of a measurement outside this zone in the population at large. The present study demonstrates that the maximum closure pressure with a full bladder and the patient standing is the most useful for diagnostic purposes and the second most useful measurement is the maximum closure pressure with a full bladder and the patient supine. It appears that there is no benefit in doing the profile with an empty bladder. Therefore, the procedure we recommend is to do a cystometrogram initially to exclude an uninhibited (or small capacity) bladder and then proceed with the profile examination with the bladder full. In contrast to Gleason and associates5 our results did not demonstrate that the integrated pressure (surface area) or the mean pressure is more valuable diagnostically than the maximum closure pressure. We believe that the reason for this is that any delay by the examiner in detecting the water appearing at the meatus will definitely affect the integrated and mean pressures but may not affect

the maximum closure pressure. Thus, the integrated and mean pressures are more liable to subjective error than the maximum closure pressure. It should be stressed that the maximum closure pressures in this study were measured within the continence zone rather than from the whole profile curve, as defined by the International Continence Society. 1 We believe that the former is superior because it tends to eliminate the risk of obtaining falsely high measurements from the distal part of the urethra, which has no sphincteric function. We concur with Gleason and associates 5 that the continence zone concept adds to the validity of the profile studies. Despite the fact that the end of the continence zone was arbitrarily demarcated in the standing profiles the maximum closure pressure - full standing proved to have a potentially higher diagnostic accuracy than the maximum closure pressure - full supine. The probable reason for this is that the former measurement incorporates the response of urethral closure to physical stress. We believe that this adds to the diagnostic value of the test. However, since the degree of accuracy of the maximum closure pressure - full standing was still less than ideal (table 2) we recommend that until more experience is acquired supine and standing profiles should be done to reduce the chances of error. Any test that is used for the diagnosis of stress incontinence is not a substitute for a thorough clinical assessment and good clinical judgment. The stress cystogram is simpler to perform than the urethral pressure profile. However, falsely positive or negative results are frequent. Kitzmiller and associates found that although women with stress incontinence tended to have wider posterior urethrovesical angles there was a large overlap between them and a control group. 14 Unfortunately, a statistical analysis of the degree of overlap was not done. They reported that if the arbitrary limit of150 degrees of a normal posterior urethrovesical angles was selected 28 per cent of the continent women had abnormal angles and 26 per cent of the patients with stress incontinence had normal angles. By comparison, if for example we select from our data 50 cm. water as the limit in the maximum closure pressurefull standing group the respective figures would be O and 5 per cent. Statistically, however, one cannot apply such figures to the population at large and they do not reflect the diagnostic accuracy of the measurement. We believe that the change in urethrovesical angle on stress as determined by the stress cystogram is a reflection of the efficacy of the supporting structures and does not take into consideration the intrinsic closure property of the urethra, which is a serious limitation. The Marshall-Marchetti test is a practical method to detect stress incontinence. The incidence of falsely negative results has never been established. In our experience a higher yield of positive tests will be obtained in the standing position. However, in the presence of an uninhibited bladder a Marshall test in the standing position may provoke a mild detrusor contraction and a falsely positive diagnosis of stress incontinence could be made. Also, the value of raising the bladder neck intravaginally during the Marshall test is questionable, since this maneuver could probably stop a urinary leak whether the patient has stress or detrusor incontinence. In our experience this maneuver has never failed to stop the leakage of urine. Therefore, we believe that the Marshall test is only reliable if it is positive in patients with an unequivocal history of stress incontinence and with normal cystometrograms. The significance of the various urethral changes as measured by the urethral pressure profile in the pathophysiology of stress incontinence remains the subject of debate. Our results confirm that the measurements of urethral pressure (maximum closure pressure, integrated pressure and mean pressure) and length-full standing, are significantly reduced. Whether the reduction in length is just a sign or has functional importance15 is controversial. Doubt also has been expressed


on the physiological relevance of the urethral pressure measurements. It has been argued that they are artifactual. However, since these pressure changes correlate with the presence of stress incontinence 2-<, they cannot be considered spurious. It also has been stated that the significant part of the urethral sphincteric mechanism is at the internal urethral orifice (that is the bladder neck) and features below this level cannot be a basic cause of urethral incompetence. 16 However, there is no reason to refute the concept that although the sphincteric mechanism begins at the bladder neck its maximum closure point is more distal. Profile studies in stress incontinence demonstrate that the whole mechanism is affected. In the present series the integrated pressure, mean pressure and maximum closure pressure were all significantly reduced. Comparable results were obtained by previous investigators.4·5 Although the maximum closure pressure has been emphasized as a useful diagnostic measurement it is not intended to imply that its site is the only physiological point of urethral closure. The change in maximum closure pressure on standing is an interesting phenomenon and may have functional and therapeutic implications. This will be the subject of a future report. It is impractical and unwarranted to recommend the routine use of the urethral pressure profile in the evaluation of all female patients with urinary incontinence. In general, the test should be done when the clinical findings, the Marshall test, the cystometric and stress cystogram results are ambiguous and do not permit a clear-cut diagnosis of stress incontinence. Specifically, the indications are 1) when the history and cystometric findings suggest the possibility of a mixed type of incontinence (that is detrusor and stress), 2) when the history is characteristic of stress incontinence but the Marshall test is negative (in these cases the pressure profile examination is probably more useful than the stress cystogram) and 3) in patients who have had an operation that failed to correct the incontinence. The role of the urethral pressure profile in these situations is being evaluated presently in our institution. REFERENCES

1. International Continence Society: First report on the standardi-



4. 5.

sation of terminology of lower urinary tract function. Brit. J. Urol., 48: 39, 1976. Enhiirning, G.: Simultaneous recording of intravesical and intraurethral pressure. A study on urethral closure in normal and stress incontinent women. Acta Chir. Scand., suppl. 276, p. 1, 1961. Toews, H. A.: Intraurethral and intravesical pressures in normal and stress incontinent women. Obst. Gynec., 29: 613, 1967. Low, J. A. and Kao, M.: Intravesical and intraurethral pressure as a measure of urethral sphincter function. Obst. Gynec., 40: 627, 1972. Gleason, D. M., Reilly, R. J., Bottaccini, M. R. and Pierce, M. J.: The urethral continence zone and its relation to stress incontinence. J. Urol., 112: 81, 1974.


6. l\foGuire, E. J., Lytton, B., Pepe, V. and Kohorn, E. I.: Stress urinary incontinence. Obst. Gynec., 47: 255, 1976. 7. Bjerle, P.: Relationship between perivesical and intravesical urinary bladder pressures and intragastric pressure. Acta Physiol. Scand., 92: 465, 1974. 8. Green, T. H., Jr.: Urinary stress incontinence: differential diagnosis, pathophysiology, and management. Amer. J. Obst. Gynec., 122: 368, 1975. 9. Brown, M. and Wickham, J.: The urethral pressure profile. Brit. J. Urol., 41: 211, 1969. 10. Lowe, P. J., Saunders, G. A. B., Downie, J. W. and Awad, S. A.: Catheter withdrawing apparatus for clinical urethral pressure profile studies. J. Urol., lHi: 626, 1976. 11. Snedecor, G. W. and Cochran, W. G.: Statistical Methods, 6th ed. Ames: Iowa State University Press, p. 329, 1967. 12. Edwards, L. and Malvern, J.: The urethral pressure profile: theoretical considerations and clinical application. Brit. J. Urol., 46: 325, 1974. 13. Ghoneim, M.A., Rottembourg, J. L., Fretin, J. and Susset, J. C.: Urethral pressure profile: standardization of technique and study ofreproducibility. Urology, 5: 632, 1975. 14. Kitzmiller, J. L., Manzer, G. A., Nebel, W. A. and Lucas, W. E.: Chain cystourethrogram and stress incontinence. Obst. Gynec., 39: 333, 1972. 15. Lapides, J., Ajemian, E. P., Stewart, B. H., Breakey, B. H. and Lichtwardt, J. R.: Further observations on the kinetics of the urethrovesical sphincter. J. Urol., 84: 86, 1960. 16. Jeffcoate, T. N. A.: The principles governing the treatment of stress incontinence of urine in the female. Brit. J. Urol., 37: 633, 1965. EDITORIAL COMMENT Pressure profile measurement is valuable in evaluating sphincteric activity at various levels along the urethral length, relating this activity to the 2 muscular components. However, a pressure profile should be considered as only 1 of several parameters and useful if all of these parameters are given attention. Alone, its value might prove to be limited. In stress incontinence, particularly, the pressure profile is usually low in comparison to the normal and the loss in pressure is generally attributable to both sphincter elements: the smooth component in the proximal segment of the urethra and the voluntary component in the mid segment. In mild cases and with the bladder relatively empty, a profile obtained in the sitting or supine position might be normal. However, it will decrease with bladder filling and also when the patient assumes the upright position. Most important is what happens when the patient is under stress-with the stress of beardown, coughing or straining the closure pressure usually decreases. The profile is valuable not only in measuring maximum urethral pressure or closure pressure but in defining the functional length of the urethra, which is generally short in stress incontinent patients. Finally, the urethral pressure profile is important in differentiating stress incontinence from other types of incontinence, particularly urge incontinence. In these latter cases the closure pressure as well as the response to stress •Nill be normal. Emil A. Tanagho Division of Urology University of California School of Medicine San Francisco, California