In 1962, Enhörning introduced the urethral pressure profile as a new method into the field of urodynamics.1 Principles and algorithm of interpretation have not changed since then, even though electronic transducers and computer‐assisted urodynamics have been introduced into clinical practice for registration and analysis of urethral pressure profile measurements. The procedure itself has been standardized by the International Continence Society requiring exact description of patient position, bladder volume, placement and direction of the transducer, and velocity of catheter withdrawal.2 However, urethral pressure profile measurements have distinct limits. A big overlap of urethral pressure profile measurements at rest between continent and stress incontinent women has been demonstrated previously, and thus the clinical validity of this procedure to assess genuine stress incontinence still remains a matter of debate.2,3 The term “hypotonic urethra” is defined when the urethral closure pressure (Pclo) does not exceed 20 cm H2O, but this is an arbitrary cutoff value. In these patients, an increased risk of recurrent genuine stress incontinence has been observed after anti‐incontinence surgery, such as the Burch colposuspension.4
Urethral pressure profile measurements during stress are a matter of an even more controversial discussion. To date, exact identification of patients with genuine stress incontinence by conclusive evaluation of the urethral pressure profile during stress was impossible. Moreover, a sensitivity of 93.3% and specificity of 82.5% of the urethral pressure profile during stress have been observed in a previous study of 911 patients with clinically demonstrable genuine stress incontinence.5 A clinical stress test or pad test with direct vision of urinary loss through the urethra remains the gold standard so far.
Many factors may influence urethral pressure profile measurements during stress. Patient position, extent of the Valsalva or coughing maneuver, transducer position, and bladder volume are some of the many variables responsible for the outcome of the investigation. The force of the cough impulse and the resulting intravesical pressure (Pves) cannot be standardized unless simultaneous abdominal pressure is evaluated. Repetitive coughing may vary, and thus again influences the outcome of measurements. Moreover, movements of the pelvic floor with the urethra against the transurethral catheter carrying the measuring transducer may influence urethral pressure profile measurements contributing to reduced reproducibility.
As a consequence, urethral pressure profile measurements at rest and during stress are helpful to identify urethral occlusive forces either by extrinsic or intrinsic factors. However, without other tests, these techniques have remained unreliable to identify genuine stress incontinence.
Because of these findings, there definitely is a need for a more sensitive urodynamic method to assess the urethral competence mechanism in continent women and patients with genuine stress incontinence. We developed an alternative method for analysis of the urethral pressure profile under stress referred to as computer‐assisted virtual urethral pressure profile. This method considers the aspects of the limited value of urethral pressure profile measurements during stress and takes advantage of the extensive mathematical possibilities of modern computer technology.
MATERIALS AND METHODS
A total of 31 women complaining of urinary loss during stress and seen in the urogynecologic unit of the Department of Gynecology, Martin‐Luther‐University Halle‐Wittenberg between July 1999 and December 1999 were enrolled in our study. Moreover, 30 healthy and symptom‐free women admitted to our hospital because of other gynecologic diseases (uterine bleeding, n = 17; breast cancer, n = 13) served as controls. All subjects gave written informed consent for the urogynecologic investigations. The study was approved by the Institutional Review Board of the Medical Faculty of Martin‐Luther‐University Halle. Urogynecologic assessment comprised patient's history, a clinical gynecologic investigation, residual urine measurements, and multichannel urodynamics (filling cystometry, urethral pressure profile measurements at rest and during stress). A clinical stress test in the sitting and standing position at a bladder volume of 300 mL was carried out in all patients. All patients in our study had grossly normal neurologic functions by reflex testing in the sensomotor region L1‐S5. All urodynamic data were recorded, processed, and analyzed with a multichannel urodynamic system (Ellipse 4, Andromeda Medizintechnik, Taufkirchen, Germany) using standard criteria of the International Continence Society, except where specifically noted.6
Patients with genital prolapse, diabetes, neurologic disorders, residual bladder volumes (greater than 50 mL), urinary tract infections, or detrusor instabilities (increase in detrusor pressure to more than 15 cm H2O during filling cystometry, filling rate 90 mL per minute) were excluded from our study. Computer‐assisted virtual urethral pressure profile uses the same parameters as those for conventional urethral pressure profile measurements during stress, with the only change being that withdrawal of the catheter is stopped at 3 mm increments on the catheter when the patient coughs.6 Three increasing and measurable forceful coughs at each catheter stop are requested (Figure 1). As a consequence, three pairs of values of Pves and urethral pressure (Pura) are obtained, and Pura is determined by Pves. The value of Pves depends on the force of the cough impulse and is not reproducible. With the computer program Statgraphics (Manugistics Inc., Rockville, MD) and a self‐produced software, a spline interpolation is performed. This is a mathematical method frequently used in modern techniques (eg, in medicine for calculation of computed tomography images).7–11 Virtual interim values are deduced from actually measured value pairs allowing to construct a graph containing all possible values of the independent variable (Pves in the presented model) and the corresponding ones for the dependent variable (Pura) within the span where a mathematical relationship exists in between the two. This allows selection of value pairs of Pves and Pura arbitrarily from those that have been measured as well as from those that have been calculated. The values of Pura to be expected with 0, 10, 20, 30, …, and 100 cm H2O of Pves can also be determined. This means that the “standardized cough impulses” have been created.
The entire procedure is repeated eight times at each investigation. The software is presently set up for nine measuring sequences, but this figure can be varied as desired. As measurements are taken with the catheter arrested, movements of the pelvic floor become irrelevant because the location indicated relates to the catheter and not to the urethra. The catheter thus represents a rigid referral scale. Accuracy of the procedure requires that the “system” composed of measuring catheter and urethra swings back into an identical starting position before each measuring sequence. A conscientious measuring technique is an imperative condition. Based on the calculated value pairs of Pves and Pura for the standardized cough impulses, the virtual pressure profiles are subsequently worked out. A second spline interpolation ensues: profiles of Pura are now computed in function of the position of the measuring transducer along the referral axis keeping Pura at a constant level. In this way, individual graphs of Pura as to be expected for Pves = 0, 10, 20, 30, …, and 100 cm H2O are established. Drawing the graphs for Pura in a coordinate system allows realization of the extent and impact of pelvic floor movements. The variability of functional urethral length can be quantified, but this is irrelevant for further interpretation (Figure 2).
Finally, Pclo is calculated by subtraction of Pves from Pura. The graphs obtained are mathematical functions, which can be related to the well‐known parameters of the resting urethral pressure profile (Figure 3). In addition to functional urethral length and the maximal urethral closure pressure (Pclomax), the area under each curve is determined.
A total of 30 continent women (mean age 40, range 20–68 years) and 31 patients complaining of urinary loss (mean age 56.6, range 39–74 years) were submitted to conventional filling cystometry, urethral pressure measurements, and computer‐assisted virtual urethral pressure profile, as described above. Virtual pressure profiles were established. We then evaluated the mean values of Pclomax, functional urethral length, the area underneath the urethral pressure profile curve, and the percentual change of these variables to the initial value at rest.
Student t test for paired samples was used to compare the two groups. P < .05 was considered significant. All analyses were performed using the SPSS 8.0 Statistical Software package (SPSS Inc., Chicago, IL).
Group‐specific demographic descriptions of continent women and patients with genuine stress incontinence are presented in Table 1. According to the results of our urogynecologic evaluation, we identified two groups of women. As was to be expected, no incontinence was present during urogynecologic investigation in 30 symptom‐free women (group A). Genuine stress incontinence was present in 31 symptomatic patients (group B). No difference between group A and B patients was found for Pclo and for functional urethral length, measured during conventional urethral pressure profile at rest (Table 2).
In contrast, absolute values as well as relative changes compared with the initial pressure profile curve at rest (Pves = 0) of all parameters of computer‐assisted virtual urethral pressure profile showed significant differences between patients with genuine stress incontinence and continent subjects (Table 2). Although in some incontinent women Pclomax may rise (two of 31) or functional urethral length does not shorten (five of 31), during computer‐assisted virtual urethral pressure profile, the area under the urethral pressure profile curve always gets smaller with increasing Pves.
Compared with conventional urethral pressure profile measurements at rest, significant differences for all parameters of computer‐assisted virtual urethral pressure profile were found in our study between patients with genuine stress incontinence and continent women. The famous physician Heisenberg stated in 1927 that no object in nature can be measured without being deranged and transformed.12 Urethral pressure measurements by means of a transurethral catheter take place in a totally unphysiologic situation. This has to be kept in mind when interpreting urethral pressure profile measurements. The obtained profile can only be considered as the result of the registration procedure and its conditions, but not as a reflection of an undisturbed physiologic or pathophysiologic situation. The urethra reacts to the catheter.
Enhörning's assumption that Pura must always exceed Pves to ascertain continence applies, strictly speaking, only for the undisturbed urethra.1 Registration under stress introduces additional methodological errors. The resting urethral pressure profile is taken as a baseline onto which the stress urethral pressure profile is being projected. Both parts of the resulting curve are then compared and analyzed mathematically to obtain transmission pressures. The two situations that are being compared—resting urethral pressure profile in between cough impulses and stress urethral pressure profile during coughing—are, however, completely different. Furthermore, it is impossible to standardize the rise of Pves during stress simulation. Relative movements in between the catheter and the urethra have an impact as well.13–15 Sequence and number of cough impulses are chosen arbitrarily. All this limits validity, accuracy, reliability, and reproducibility of the method.
How many cough impulses are needed to be evaluated? Which of these cough impulses are relevant for quantification—and even more basic identification—of incontinence? Why do several patients have a positive urethral closure pressure paralleling a positive clinical stress test under identical conditions? All these questions remain unanswered so far. On the other hand, we know that the urethral pressure profile under stress is well apt to evaluate urethral closure function.16 Only further refinement of the method and innovative interpretation of the results can lead to improvements. Computer‐assisted virtual urethral pressure profile could bear such progress. All above listed methodological pitfalls are considered in this new method. In our study, the area under the curve seems to represent a highly significant parameter to confirm stress incontinence. This is a result of immediate relevance for urodynamic practice. However, the main advantage of computer‐assisted virtual urethral pressure profile seems to be that new questions can be asked and answered. Especially when considering the paradigm of the urethral closure mechanism established by Papa Petros and Ulmsten,17,18 this opens aspects for further research.
1. Enhörning GE. Simultaneous recording of the intravesical and intraurethral pressure. Acta Chir Scand 1962;276:1–68.
2. Abrams P, Blaivas JG, Stanton SL, Anderson JT. The standardization of terminology of lower urinary tract function recommended by the International Continence Society. Int Urogynecol J 1990;1:45–58.
3. Ramsay IN, Hilton P, Cox TF. Time-series analysis of urethral electrical conductance measurements in the assessment of unstable urethral pressure: Results in normal patients and in those with genuine stress incontinence. Neurourol Urodyn 1993;12:23–31.
4. Sand PK, Bowen LW, Panganiban R, Ostergard DR. The low pressure urethra as a factor in failed retropubic urethropexy. Obstet Gynecol 1987;69:399–402.
5. Hanzal E, Berger E, Koelbl H. Reliability of the urethral closure pressure profile during stress in the diagnosis of genuine stress incontinence. Br J Urol 1991;68:369–71.
6. The standardization of terminology of lower urinary tract function. Br J Obstet Gynaecol 1990;97:1–16.
7. Barillot C, Gibaud B, Lis O, Min LL, Bouliou A, Le Certen G, et al. Computer graphics in medicine: A survey. Crit Rev Biomed Eng 1988;15:269–307.
8. Larsen KH, Burch SE. Generation of dose calculation data tables using cubic spline interpolation. Med Dosim 1991; 16:147–51.
9. Lehmann TM, Gonner C, Spitzer K. Survey: Interpolation methods in medical image processing. IEEE Trans Med Imaging 1999;18:1049–75.
10. Ostuni JL, Santha AK, Mattay VS, Weinberger DR, Levin RL, Frank JA. Analysis of interpolation effects in the reslicing of functional MR images. J Comput Assist Tomogr 1997;21:803–10.
11. Schoenberg IJ. Contributions to the problem of approximation of equidistant data by analytic functions. Part A and Part B. Q Appl Math 1946;1:4, 45–99, 112–41.
12. Heisenberg W. Der Teil und das Ganze. Munich, Germany: Piper, 1996.
13. Schüssler B, Hesse U, Lentsch P, Anthuber C. Artefacts in urometry caused by marked genital prolapse. Neurourol Urodyn 1987;6:154–5.
14. Carey MP, Dwyer PL, Glenning PP. The sign of stress incontinence—Should we believe what we see? Aust N Z J Obstet Gynaecol 1997;37:436–9.
15. Clarke B. The role of urodynamic assessment in the diagnosis of lower urinary tract disorders. Int Urogynecol J Pelvic Floor Dysfunct 1997;8:196–9.
16. Eberhard J. Standardized measurement of urethral pressure with normal values in the diagnosis of stress incontinence. Geburtsh Frauenheilk 1986;46:145–50.
17. Papa Petros PE, Ulmsten U. An integral theory and its method for the diagnosis and management of female urinary incontinence. Scand J Urol Nephrol 1993;153:3–93.
© 2002 by The American College of Obstetricians and Gynecologists.
18. Papa Petros PE, Ulmsten U. An anatomical classification—A new paradigm for management of urinary dysfunction in the female. Int Urogynecol J Pelvic Floor Dysfunct 1999;10:29–35.