Objective
The objective of the study was to compare the diagnostic accuracy of a visual urethral mobility examination (VUME) with a Q-Tip test in the assessment of urethral hypermobility.
Study Design
Subjects were randomized to a VUME or Q-Tip test first, followed by the alternate assessment. Outcomes included hypermobile, not hypermobile, or indeterminate. Hypermobility was defined as 30° or greater from the horizontal. Pain perception and test preference were assessed.
Results
For the 54 subjects, 61.1% demonstrated hypermobility on VUME vs 72.2% on the Q-Tip test ( P = .39). Assessment of hypermobility was correlated between the 2 tools (ρ = 0.47, P = .001). Positive and negative predictive value, sensitivity, and specificity for VUME were 88%, 56%, 78%, and 71%, respectively, with a Q-Tip test as the reference standard. Mean pain score on visual analog scale was 0.72 for VUME and 3.15 for Q-tip test ( P < .001). VUME was preferred by 83% of subjects.
Conclusion
VUME is a diagnostic alternative to the Q-Tip test for the assessment of urethral hypermobility and is preferred by subjects.
Stress urinary incontinence (SUI), the involuntary leakage of urine on effort or exertion or on sneezing or coughing, affects 4-35% of women worldwide. SUI is classified based on urethral mobility, with urethral hypermobility associated with greater operative success after surgical placement of a suburethral sling. Thus, accurate assessment of urethral hypermobility is essential when assessing treatment options and providing preoperative patient counseling.
In 1971, Crystle et al identified the Q-Tip test as an inexpensive and simple method to assess urethral mobility. The authors demonstrated that when a Q-Tip is inserted into the urethra to the urethrovesical junction, the degree of rotation of the free end of the Q-Tip correlated well with posterior rotation of the urethrovesical junction on imaging studies, specifically bead chain cystourethography. Since that time the Q-Tip test has been shown to be a reliable tool and is commonly used to determine the presence of urethral hypermobilty. A straining angle of 30° or greater relative to the horizontal during Valsalva or cough has been published in the literature as the definition of urethral hypermobility and is accepted as clinically relevant.
Although a useful tool, the Q-Tip test requires the insertion of a rigid cotton-tipped swab into the urethra, which is uncomfortable for many women and may increase the chance of urinary tract infections. Identification of a visual examination tool to determine urethral hypermobility would help eliminate these issues with the Q-Tip test. Several studies have attempted to use the measured Aa point of the Pelvic Organ Prolapse Quantification (POP-Q) examination as a proxy for the Q-Tip examination. However, these studies failed to show a correlation between anterior vaginal wall prolapse and urethral hypermobility.
Identification of movement of the urethral meatus with Valsalva, rather than prolapse of the anterior vaginal wall, may be a better proxy for the urethrovesical junction. During the Q-Tip test, the proximal end of the Q-Tip is placed at the urethrovesical junction and the angle of the distal end of the Q-Tip relative to the horizontal is measured while the patient is straining. Because the Q-Tip is a rigid device, the degree of rotation from the horizontal at the distal end of the Q-Tip should theoretically represent the degree of rotation at the urethrovesical junction. Likewise, as the Q-Tip rotates posteriorly at the urethrovesical junction, the urethral meatus theoretically moves anteriorly. Thus, anterior movement of the meatus, which can be identified by visual examination alone, may be a visual proxy for the Q-Tip test.
The primary aim of our study was to compare the diagnostic accuracy of visual urethral mobility examination (VUME) with the Q-Tip test for the assessment of urethral hypermobility. Secondary aims were to compare patient pain associated with the 2 diagnostic methods, to assess patient test preference, and to determine whether the level of clinician experience affects the accuracy of visual examination of urethral hypermobility.
Materials and Methods
We conducted a prospective, randomized, cross-over trial assessing the diagnostic accuracy of VUME compared with the standard Q-Tip test for the assessment of urethral hypermobility. Prior to patient recruitment, the study was approved by the University of North Carolina (UNC) at Chapel Hill Biomedical Institutional Review Board. The Standards for Reporting of Diagnostic Accuracy checklist was used because this was an assessment of diagnostic accuracy. Women were recruited from the UNC Female Pelvic Medicine Clinic at 2 locations from December 2010 through March 2011.
Subjects with a chief complaint of urinary incontinence were invited to participate. Exclusion criteria were age less than 18 years, current urinary tract infection, history of recurrent urinary tract infections, or the unwillingness to tolerate 2 consecutive examinations.
After enrollment subjects were randomized to undergo either the VUME or Q-Tip test first, followed by the alternate method. All subjects underwent both methods of assessment. The 2 examinations were performed consecutively by 2 different examiners who were blinded to the results of the other examiner. Group assignment was determined by randomization through computer-generated number blocks of 6.
Allocation concealment using a sequentially numbered, opaque, sealed envelope maintained blinding until the envelope was opened for each subject by the examiner just before starting the first examination. Randomization determined which examination was performed first, and each subject underwent both assessments: the primary provider performed the first assessment and a different provider performed the second assessment. Examiners included 5 attending physicians and 3 fellow physicians in the Division of Female Pelvic Medicine and Reconstructive Surgery.
Subjects were not be blinded to their group assignment or examination because this would be technically difficult, given the nature of the differences in performing the examinations. We chose a randomized crossover design to avoid any influence of 1 examination over the other and to minimize the risk of performance bias with a repeat examination. Patients may have a different level of effort on the first examination as compared with the second examination and vice versa. Thus, by having 1 group of patients undergo the VUME first followed by the Q-Tip test and the other group doing the opposite, in a randomized fashion, we sought to overcome any influence of either the order of performance or patient or examiner perception of the 2 examinations.
Three different positions were measured for each test: resting angle, straining angle, and angle with levator contraction. For the VUME, the urethral meatus was used as a proxy for the urethrovesical junction (UVJ). The patient was asked to perform 3 Valsalva maneuvers and to cough strongly 3 times. The patient was then asked to perform a levator contraction 3 times. The examiner visualized the urethral meatus during each maneuver and, based on the strongest straining effort, subjectively determined whether the urethral meatus rotated less than or greater than 30° from the horizontal to determine hypermobility. Urethral rotation was characterized as indeterminate if the examiner could not determine the degree of movement of the urethra.
The Q-Tip test was performed in standard fashion. A cotton-tip applicator lubricated with 2% lidocaine jelly was inserted into the bladder via the urethral meatus and withdrawn until resistance was met, representing the UVJ. The patient was asked to perform 3 Valsalva maneuvers and then cough strongly 3 times. The patient was then asked to perform a levator contraction. The maximum straining angle (ie, the angle at which there is no further movement of the Q-tip) was measured in degrees from the horizontal plane using a paper protractor. Urethral hypermobility was defined as a maximum straining angle greater than or equal to 30° above the horizontal plane.
Upon completion of both assessments, subjects were asked to rate their pain during each intervention using a 10 cm visual analog scale (VAS) and to indicate which assessment tool they preferred. The medical records were reviewed and the following information was collected: age, race, height, weight, body mass index (BMI), prior pelvic surgeries, and obstetrical history.
The study was performed as a pilot; thus, an a priori power calculation was not performed. We planned to randomize a total of 56 patients with 28 patients undergoing the VUME first and 28 undergoing the Q-Tip test first. χ 2 and Fisher exact tests were used for categorical data. Pearson correlation coefficient was also assessed. Sensitivity, specificity, and positive and negative predictive values with 95% confidence intervals were calculated. The Q-Tip test was used as a reference standard against which to compare the VUME. Statistical analysis was performed using SPSS 18.0 (SPSS Inc, Chicago, IL).
Results
A total of 56 women were enrolled and randomized, and 54 completed both interventions ( Figure 1 ) . One subject declined any urethral examination, and the other subject was unable to tolerate the exam because of severe shortness of breath. Although randomized, neither patient underwent evaluation by either the visual examination or the Q-Tip test. Thus, they did not contribute data for the final analysis. No patients were lost to follow-up.
The majority of the women were white (70.4%), with a mean age of 57.2 ± 14.9 years. Mean BMI was 34.2 ±16.1, with 3 (5.6%) having undergone prior sling placement, and 6 (11.1%) having undergone prior prolapse surgery. During the randomization process, 1 envelope (#46) was inadvertently skipped; however, randomization continued sequentially as originally planned. Despite this event, there were no differences in demographics between the 2 randomization groups ( Table 1 ).
| Demographic | VUME first (n = 24) | Q-Tip test first (n = 30) | P value |
|---|---|---|---|
| Age, y | 58.4 ± 15.2 | 56.27 ± 14.8 | .602 |
| BMI, kg/m 2 | 35.9 ± 16.5 | 32.9 ± 16.0 | .497 |
| Race | .169 | ||
| White | 19 (79.2) | 19 (63.3) | |
| African American | 4 (16.7) | 3 (10.0) | |
| Other | 1 (4.2) | 7 (23.3) | |
| Gravidity | 2.7 ± 1.2 | 2.8 ± 2.0 | .832 |
| Parity | 2.3 ± 0.92 | 2.2 ± 1.4 | .616 |
| POP-Q | |||
| Aa | –1.3 | –0.6 | .245 |
| Ba | –1.2 | –0.6 | .283 |
| C | –5.7 | –5.8 | .925 |
| D | –7.8 | –7.1 | .651 |
| Ap | –2.0 | –2.1 | .450 |
| Bp | –2.0 | –2.1 | .450 |
| GH | 3.2 | 3.6 | .285 |
| PB | 3.2 | 2.7 | .139 |
| TVL | 8.9 | 8.9 | .901 |
| Menopausal | 14 (25.9) | 15 (27.8) | .785 |
| Current smoker | 3 (12.5) | 4 (13.3) | .928 |
| Prior sling procedure | 0 | 3 (10) | .067 |
| Prior prolapse surgery | 2 (8.3) | 4 (13.3) | .561 |
| Prior hysterectomy | 10 (41.7) | 10 (33.3) | .529 |
The rate of urethral hypermobility was similar between the 2 examinations: a total of 33 subjects (61.1%) were hypermobile per the VUME compared with 39 (72.2%) per the Q-Tip test ( P = .39) ( Figure 2 ). These findings correlated moderately well with a Pearson correlation coefficient of 0.47 ( P = .001). Hypermobility was labeled indeterminate on the VUME for only 3 of the subjects (5.6%). The VUME had a good positive predictive value of 88% when compared with the Q-Tip test as reference standard ( Table 2 ).
