- Email: [email protected]
Urodynamic Differences Between Dysfunctional Voiding and Primary Bladder Neck Obstruction in Women
Female Urology Urodynamic Differences Between Dysfunctional Voiding and Primary Bladder Neck Obstruction in Women Benjamin M. Brucker, Eva Fong, Sagar Shah, Christopher Kelly, Nirit Rosenblum, and Victor W. Nitti OBJECTIVE
METHODS
RESULTS
CONCLUSION
To determine the clinical and urodynamic differences in the presentation and the value of simultaneous fluoroscopy in dysfunctional voiding (DV) and primary bladder neck obstruction (PBNO); the 2 most common causes of non-neurogenic “functional” bladder outlet obstruction in women. A review of our urodynamic study database (March 2003 to August 2009) was conducted. DV was diagnosed when increased external sphincter activity was found during voluntary voiding on electromyography (EMG) or fluoroscopy. PBNO was diagnosed when a failure of bladder neck opening was noted on fluoroscopy during voiding. The demographics, symptoms, and urodynamic study parameters were collected. Comparisons were done using chi-square and 2-tailed t-tests. DV was diagnosed in 34 women and PBNO in 16. The patients with DV were younger than those with PBNO (40.9 vs 59.2 years, P ⬍ .001). Women with DV showed a clinical trend toward having more storage symptoms than those with PBNO and fewer voiding symptoms. Patients with DV had a greater mean maximal flow rate (12 vs 7 mL/s, P ⫽ .027) and lower mean postvoid residual urine volume (125 vs 400 mL, P ⫽ .012). No significant differences were found in maximal detrusor pressure, detrusor pressure at maximal flow rate, or detrusor overactivity. EMG showed increased activity during voiding in 79.4% of those with DV and 14.3% of those with PBNO (P ⬍ .001). Clinically, women with DV and PBNO had similar presentations, although those with PBNO had poorer emptying. The flow rates and patterns seemed to differ between those with DV and PBNO, although the voiding pressures were similar. EMG alone would have given the wrong diagnosis in 20.6% of those with DV (false negative) and 14.3% of those with PBNO (false positive). When fluoroscopy is used to define these entities, the accuracy of EMG to differentiate them is questionable. UROLOGY 80: 55– 60, 2012. © 2012 Elsevier Inc.
B
ladder outlet obstruction (BOO) in women can be categorized into 2 larger divisions: those caused from a fixed anatomic obstruction and those that are functional.1 Anatomic obstruction can often be predicted from the symptoms, history, and physical examination findings (eg, emptying symptoms and/or incomplete emptying in a woman with high-grade pelvic organ prolapse or urethral stricture or after recent anti-incontinence surgery). However, functional obstruction can be diagnosed only during the act of voiding, because no
Financial Disclosure: The authors declare that they have no relevant financial interests. From the Department of Urology, New York University Langone Medical Center, New York, New York Reprint requests: Benjamin M. Brucker, M.D., Department of Urology, New York University LangoneMedical Center, 150 East 32nd Street, Second Floor, New York, NY 10016. E-mail: [email protected] Submitted: March 5, 2012, accepted (with revisions): April 9, 2012
obvious anatomic abnormality will be associated with the patient’s symptoms. The most common causes of functional BOO in neurologically intact women are dysfunctional voiding (DV) and primary bladder outlet obstruction (PBNO).2,3 DV occurs when there is improper relaxation or contraction of the external striated sphincter during voiding. In PBNO, a failure of the bladder neck (or proximal sphincter) to open and funnel normally during voiding occurs. The accurate diagnosis of these entities is critical because the endpoints of the treatment algorithms are vastly divergent. The first-line treatment of DV involves education with biofeedback to “unlearn the learned behavior,” and PBNO is often initially treated with ␣-blockers and sometimes requires surgical incision of the bladder neck.4,5 Physical therapy with biofeedback can be time intensive and costly, and incision of the bladder neck is invasive with a small, but real, risk of incontinence. These facts make empiric therapy of suspected voiding phase dysfunction less desirable.
© 2012 Elsevier Inc. All Rights Reserved
0090-4295/12/$36.00 http://dx.doi.org/10.1016/j.urology.2012.04.011
55
Many attempts have been made to better define what constitutes BOO in women; however, very little has been done to define the urodynamic difference between the specific etiologies that result in BOO.3,6-9 Obviously, the least invasive and most cost-effective method to differentiate between the causes would be preferable. The present study sought to determine whether urodynamic and clinical differences exist between neurologically intact women with functional obstruction. The diagnosis of PBNO and DV has historically been made by urodynamic studies (UDS).10 Surface electromyography (EMG) activity has also been used in an attempt to separate these diagnoses.11 However, surface EMG is prone to artifacts, causing apparent increased activity during voiding.12 Needle EMG is uncomfortable, and many women are adverse to its use. Thus, we have come to rely on fluoroscopic imaging during UDS (videourodynamics) to localize any obstruction at the bladder neck or external sphincter. In addition, one can often see spasmodic activity of the sphincter during fluoroscopic assessment of voiding in patients with DV. We believe this to be the most accurate assessment of the voiding phase, and this has become our “standard” for diagnosing women with functional obstruction and differentiating its cause. However, we questioned how valuable videourodynamics truly is for this purpose and whether DV and PBNO in women can be accurately differentiated from each other without the use of fluoroscopy.
MATERIAL AND METHODS We performed a retrospective review of all female videourodynamic studies at a single institution from March 2003 to August 2009. It has been our practice to perform videourodynamics in all women with suspected or possible voiding phase dysfunction and other selected conditions undergoing UDS. The reasons for the evaluation, determined from the clinical history and presenting symptoms, included frequency and/or urgency refractory to conservative or medical therapy, urgency incontinence, elevated postvoid residual (PVR) urine volume, urinary retention, slowed urinary flow, recurrent urinary tract infection, stress urinary incontinence not demonstrated on physical examination, mixed incontinence, unaware incontinence, and before pelvic organ prolapse surgery (to assess for stress urinary incontinence with the reduction of prolapse, if not seen at the examination or with a supine cough stress test). All videourodynamics were performed with 7F transurethral and rectal balloon catheters according to the International Continence Society standards.13,14 EMG of the pelvic floor was done with surface electrodes. Filling cystometry was performed at a rate of 50 mL/min with radiographic contrast material. In cases of severe detrusor overactivity or known small functional capacity, the rate was decreased to 30 mL/min. Fluoroscopic images were obtained at select times as determined by the clinical scenario (ie, filling, Valsalva, void/attempt to void, after voiding). All studies were performed and interpreted by an experienced urologist. Data were collected on pressure-flow parameters, EMG activity during voluntary voiding, and radiographic images. Only patients with a final diagnosis of BOO were included in the present analysis. BOO was diagnosed using the videouro56
dynamic criteria of a sustained detrusor contraction of any magnitude with radiographic evidence of obstruction.3 Using the International Continence Society definition, DV was characterized by an intermittent and/or a fluctuating flow rate due to intermittent contractions of the periurethral striated muscles during a sustained voluntary detrusor contraction in a neurologically normal woman.15 This was diagnosed using a combination of increased EMG activity during voiding in the absence of abdominal straining or radiographic evidence of sphincteric contraction/failure to relax even in the absence of increased EMG activity during voiding. The clinical history was used to differentiate this entity from detrusor external sphincter dyssynergia where a known neurologic cause must exist. For cases in which DV was suspected, the uroflow from UDS was correlated with the noninvasive uroflow findings. If the intubated uroflow suggested DV, but the noninvasive uroflow did not, the DV was thought to be test induced, and such patients were not diagnosed with DV and thus were excluded from the present analysis. PBNO was defined as failure of the bladder neck to open adequately (remained closed or narrow on fluoroscopic images) during voiding, resulting in obstruction of urinary flow in the absence of other causes of obstruction or increased striated sphincter activity.3 In cases in which increased EMG activity was determined to be secondary to artifact (ie, poor contact, leading to persistent increased activity, movement during voiding, abdominal straining), EMG activity alone was not considered as increased striated sphincter activity. Urodynamic tracings and the diagnosis when UDS was performed were reviewed. The data and diagnosis recorded in the initial interpretation of the UDS were captured. The filling phase parameters included the first desire to void (the feeling when the woman first becomes aware of bladder filling), normal desire to void (the feeling that led the woman to pass urine in the next convenient moment but can delay if necessary), presence/absence of detrusor overactivity, maximum cystometric capacity, and compliance. The voiding phase parameters included the maximal flow rate (Qmax), detrusor pressure at Qmax, maximal detrusor pressure, and PVR. The maximal Pdet was recorded for cases in which no urine flow was recorded but a sustained detrusor contraction was present or if artifact (ie, a spike) of the uroflow made interpretation of the maximal detrusor pressure recorded at Qmax erroneous. EMG activity during voiding was noted as increased if the average amplitude of the EMG was increased from baseline during flow. The fluoroscopic images were reviewed for evidence of bladder neck relaxation or narrowing at the external urethral sphincter during voiding. The charts were reviewed for patient age, the presence of storage symptoms (frequency, urgency, and urgency incontinence), the presence of voiding symptoms (slow urinary stream, hesitancy, the sensation of incomplete emptying, and dysuria in the absence of infection). Patients were excluded if no flow was present and a sustained detrusor contraction was not, pressure flow data were missing, radiographic evidence of obstruction was missing, or if they had voided uncharacteristically during UDS compared with normal voiding. This included noncharacteristic straining or flow rates that the urodynamacist believed were not commensurate with noninvasive uroflow. The categorical variables were compared using chi-square and contingency tables. Continuous variables were evaluated using analysis of variance and t tests. A 2 ⫻ 2 table was used to calculate the sensitivity and specificity. UROLOGY 80 (1), 2012
Table 1. Urodynamic differences between DV and PBNO Variable First desire to void (mL) Normal desire to void (mL) Cystometric capacity (mL) Compliance (mL/cm H2O) Qmax (mL/s) PdetQmax (cm H2O) PdetMax (cm H2O) PVR (mL)
DV
PBNO
P Value
185.78 (26-720) 302.79 (56-1401) 395.32 (147-941) 118.75 (21.11-470.50) 12.06 (1-35.7) 38.94 (11-90) 46.11 (18-102) 124.50 (0-700)
226.19 (41-552) 341.38 (51-720) 534.56 (172-1500) 112.16 (57.33-253.67) 7.07 (0-16.0) 42.00 (18-95) 47.63 (24-100) 399.56 (0-1413)
.403 .612 .072 .821 .027* .608 .595 .012*
DV, dysfunctional voiding; PBNO, primary bladder neck obstruction; Qmax, maximal urinary flow rate; PdetQmax, detrusor pressure at Qmax; PdetMax, maximal Pdet; PVR, postvoid residual (urine volume). Data presented as mean, with ranges in parentheses. * Statistically significant.
RESULTS The analysis included 16 women diagnosed with PBNO and 34 women diagnosed with DV. The women with PBNO were, on average, older than those with DV (59.2 and 40.9 years, respectively, P ⬍ .001). The women with PBNO were also more likely to present with voiding symptoms than the women with DV but this failed to reach statistical significance (75% and 53%, respectively, P ⫽ .13). In contrast, a prevalence of storage symptoms was more likely in those with DV than in those with PBNO but, again, did not reach statistical significance (61% and 38%, respectively, P ⫽ .13). The presence of detrusor overactivity was similar in the 2 groups (26.3% DV and 31.3% PBNO, P ⫽ .71). A summary of the urodynamic findings is listed in Table 1. Women with PBNO were noted to have a mean Qmax significantly lower statistically than the women with DV (Qmax 7.07 and 12.06 mL/s respectively; P ⫽ .027). Also, a significantly greater average PVR urine volume was found in women with PBNO (399.56 vs 124.5 mL, P ⫽ .012). These results were confirmed by examination of the noninvasive uroflow/PVR data. The remainder of the pressure flow data did not show statistically significant differences. The EMG showed increased activity during voiding in 79.4% of the women diagnosed with DV (according to the standard of videourodynamics). Also, increased EMG activity with voiding was seen in 14.3% of patients with a final diagnosis of PBNO (videourodynamics; P ⬍ .001). In neurologically intact women with the diagnosis of BOO, EMG had a sensitivity of 79% and specificity of 85% to diagnose DV when videourodynamics was used as the standard. Thus, the positive predictive value of increased EMG activity was 93% and the negative predictive value was 63%. In all case of DV, the patients’ bladder neck was fluoroscopically seen to funnel appropriately.
COMMENT PBNO as a cause of BOO in women has been identified as the underlying cause in 9%-16% of obstructed women.2,3 Our study findings were consistent, with UROLOGY 80 (1), 2012
10.2% of obstructed women diagnosed with PBNO. This is a condition in which the bladder neck fails to open adequately during voiding, resulting in obstruction of urinary flow in the absence of increased striated sphincter activity or another anatomic obstruction. First described in 1933 by Marion,16 in 1973, Turner-Warwick et al17 advocated the use of UDS and fluoroscopy to diagnose bladder neck dysfunction in men. The entity was later described in women.18 The precise cause of PBNO has not been clearly elucidated.19,20 The accurate and timely diagnosis of PBNO remains the primary challenge.21 Failure of the bladder neck to open with a detrusor contraction can only be diagnosed by simultaneous imaging of the bladder outlet during voiding. Figure 1A shows the classic videourodynamic findings of PBNO. Involuntary contraction of the external urethral sphincter during voiding in the absence of neurologic injury or disease is known as DV. It has also been described as learned voiding dysfunction and by some as non-neurogenic, neurogenic bladder.22 DV, as defined by the International Continence Society, is an intermittent and/or fluctuating flow rate due to involuntary intermittent contractions of the periurethral striated muscle during voiding, in neurologically normal subjects.15 Figure 1B illustrates the classic videourodynamic findings of DV. It can be difficult to diagnose using surface EMG electrodes because of other factors that can produce increased EMG activity on a tracing, including attempts at augmenting bladder contractions by abdominal straining, movement, guarding reflex, and wet electrodes.23 Using fluoroscopy has proved valuable in the diagnosis because one can see a dilation of the urethra to the level of the striated sphincter (spinning top urethra) and/or intermittent contractions of the striated sphincter. The average age of women with PBNO in the present study was 59.2 years, consistent with contemporary series.24 In the study by Athanasopoulos et al,24 the mean Qmax of 9.74 mL/s was commensurate with the mean Qmax seen in our study (7.07 mL/s). The detrusor pressure at Qmax was greater in their series (99.72 vs 46.11 cm H2O), but they required obstruction to be present on the Blaivas-Groutz nomogram, possibly accounting for the difference. 57
Figure 1. A (top)- This represents the classic fluoroscopic, pressure flow and EMG findings of PBNO. B (bottom)- This is another pressure flow tracing showing high pressure/low flow voiding. The EMG in this case is clearly increased during voiding. The inset fluoroscopic image shows a narrowing at the external urethral sphincter. This is an excellent example of DV. (Color version available online.)
Carlson et al25 characterized 26 women with DV and found a mean age of 39.2 years comparable to the mean age in our series (40.9 years). The mean detrusor pressure at Qmax and PVR urine volume (50.3 cm H2O and 103.4 mL) were also very similar. The mean Qmax (10.4 mL/s) was between the 2 mean values in our series. Although the Qmax values were significantly different statistically, there was likely a large overlap between the 2 diagnoses. We assumed that videourodynamics would be the most accurate method to diagnose voiding phase dysfunction in women; thus, for the present comparative study, the videourodynamic diagnosis was considered the standard. Our analysis of the UDS findings revealed that had we relied on the surface EMG findings alone, we would have derived the incorrect diagnosis in 20.6% of the women with DV (eg, women with high sustained detrusor pressure during voluntary voiding, resulting in a low flow rate with a bladder neck that opens, but with narrowing at the external urethral sphincter without an appreciable increase in EMG activity during voiding and no anatomic obstruction such as stricture or periurethral fibrosis). In contrast, increased EMG activity during voiding was seen in 14.3% of women with PBNO (eg, a patient voiding with high pressure and low flow without any funneling or opening of the bladder neck and during voiding a substantial increase occurs EMG activity that would have suggested DV if fluoroscopy had not been used). Surface EMG use during UDS was first described by Barrett26 to determine perineal relaxation as an indirect 58
indicator of simultaneous external sphincter relaxation during the voiding phase. Compared with needle electrodes, surface electrodes are favored by many clinicians because of their noninvasive nature and ease of placement. Although a good surrogate for sphincter activity, the argument can be made that the signal source is inferior.27 EMG activity during UDS is affected by various factors, including patient factors, technology limitations, and operator factors. Patient factors include straining during voiding and guarding. Technological limitations include failure to specifically depict external urethral sphincter activity, poor skin contact, and wet leads.12 Finally, operator factors include failure to properly place the leads, improper interpretation, and failure to recognizing poor lead contact. A recent multicenter study of women with predominant stress incontinence found on preoperative UDS that 50% of the women had EMG amplitudes greater during voiding than during filling.28 The investigators noted that none of these women would have been classified as having DV, supporting our finding that the EMG results need to be cautiously interpreted. We suggest that the additional information provided by fluoroscopy is valuable if the diagnosis of DV is considered. A study comparing surface patch electrodes and concentric needle electrodes found unanimous agreement among the external reviewer interpretations of individual EMG in only 27% of tracings.29 Quiescence of EMG during voiding was seen in only 28% of the interpretable UROLOGY 80 (1), 2012
tracings, although 70% of the women had stress incontinence and no other voiding phase pathologic features. Some have advocated the use of the interval between pelvic floor relaxations and the start of urine flow to screen for PBNO in children.11,30 Obvious differences exists in the body habitus that might make EMG more accurate in children. Furthermore, the techniques used in these pediatric studies (ie, testing direct sphincter stimulation, sensitivity adjustment, audio monitoring) might not be applicable in the settings in which most women undergo testing. Our study was retrospective and thus had limitations. We did not have a specific protocol in place during UDS to standardize EMG accuracy (eg, testing EMG recording, troubleshooting poor signals, standard bandwidth). However, while performing and interpreting the UDS, the clinicians had no knowledge that these variables would be studied in the future. Therefore, the results depict UDS as they were routinely done and were not altered by a study protocol or preconceived hypothesis. Furthermore, our studies were all performed at a referral center by physicians specializing in voiding dysfunction; thus, we might have overestimated the utility of EMG by those less familiar with these less common diagnostic and clinical findings. Finally, the study reported urinary symptoms according to the patients’ presenting symptoms and not validated questionnaires. Though this practical approach might have clinical relevance, a standardized assessment of lower urinary tract symptoms might give more precise insight into the nature of urinary symptoms in these women. Although fluoroscopic imaging during UDS is widely used in many specialty centers, it is by no means the most commonly practiced method. EMG remains a valuable part of UDS, especially where fluoroscopy is not available or the patient and/or physician decide they would prefer to avoid fluoroscopy. However, in women in whom the suspicion of BOO is high, fluoroscopy in extremely useful to differentiate and diagnose DV and/or PBNO.
CONCLUSIONS DV and PBNO are both classified as functional disorders. These entries have a similar presentation. Our results showed that women with PBNO tended to have poorer emptying. The flow rates and patterns differed, but the voiding pressures did not. The greater flow rate seen in DV might be due to the intermittent nature of the obstruction compared with the fixed nature of PBNO. When fluoroscopy is used to define these entities, the accuracy of EMG to differentiate them is questionable. References 1. Patel R, Nitti VW. Bladder outlet obstruction in women: prevalence, recognition, and management. Curr Urol Rep. 2001;2:379387. 2. Kuo HC. Videourodynamic characteristics and lower urinary tract symptoms of female bladder outlet obstruction. Urology. 2005;66: 1005-1009.
UROLOGY 80 (1), 2012
3. Nitti VW, Tu LM, Gitlin J. Diagnosing bladder outlet obstruction in women. J Urol. 1999;161:1535-1540. 4. Minardi D, d’Anzeo G, Parri G, et al. The role of uroflowmetry biofeedback and biofeedback training of the pelvic floor muscles in the treatment of recurrent urinary tract infections in women with dysfunctional voiding: a randomized controlled prospective study. Urology. 2010;75:1299-1304. 5. Blaivas JG, Flisser AJ, Tash JA. Treatment of primary bladder neck obstruction in women with transurethral resection of the bladder neck. J Urol. 2004;171:1172-1175. 6. Chassagne S, Bernier PA, Haab F, et al. Proposed cutoff values to define bladder outlet obstruction in women. Urology. 1998;51:408411. 7. Defreitas GA, Zimmern PE, Lemack GE, et al. Refining diagnosis of anatomic female bladder outlet obstruction: comparison of pressure-flow study parameters in clinically obstructed women with those of normal controls. Urology. 2004;64:675-679. 8. Lemack GE, Zimmern PE. Pressure flow analysis may aid in identifying women with outflow obstruction. J Urol. 2000;163:18231828. 9. Blaivas JG, Groutz A. Bladder outlet obstruction nomogram for women with lower urinary tract symptomatology. Neurourol Urodyn. 2000;19:553-564. 10. Dupont M, Raz S. Urodynamics: contemporary evolution. In: Nitti VW, editor. Practical Urodynamics. Philadelphia: WB Saunders; 1998:2-4. 11. Combs AJ, Grafstein N, Horowitz M, et al. Primary bladder neck dysfunction in children and adolescents I: pelvic floor electromyography lag time—a new noninvasive method to screen for and monitor therapeutic response. J Urol. 2005;173:207-210. 12. Bradley CS, Smith KE, Kreder KJ. Urodynamic evaluation of the bladder and pelvic floor. Gastroenterol Clin North Am. 2008;37:539552. 13. Groutz A, Blaivas JG, Chaikin DC. Bladder outlet obstruction in women: definition and characteristics. Neurourol Urodyn. 2000;19: 213-220. 14. Carlson KV, Fiske J, Nitti VW. Value of routine evaluation of the voiding phase when performing urodynamic testing in women with lower urinary tract symptoms. J Urol. 2000;164:1614-1618. 15. Abrams P, Cardozo L, Fall M, et al. The standardisation of terminology of lower urinary tract function: report from the standardisation Sub-Committee of the International Continence Society. Neurourol Urodyn. 2002;21:167-178. 16. Marion G. Surgery of the neck of the bladder. Br J Urol. 1933;5: 351-380. 17. Turner-Warwick R, Whiteside CG, Worth PH, et al. A urodynamic view of the clinical problems associated with bladder neck dysfunction and its treatment by endoscopic incision and transtrigonal posterior prostatectomy. Br J Urol. 1973;45:44-59. 18. Diokno AC, Hollander JB, Bennett CJ. Bladder neck obstruction in women: a real entity. J Urol. 1984;132:294-298. 19. Leadbetter GW, Leadbetter WF. Diagnosis and treatment of congenital bladder-neck obstruction in children. N Engl J Med. 1959; 260:633-637. 20. Awad SA, Downie JW, Lywood DW, et al. Sympathetic activity in the proximal urethra in patients with urinary obstruction. J Urol. 1976;115:545-547. 21. Padmanabhan P, Nitti VW. Primary bladder neck obstruction in men, women, and children. Curr Urol Rep. 2007;8:379-384. 22. Hinman F. Nonneurogenic neurogenic bladder (the Hinman syndrome)—15 years later. J Urol. 1986;136:769-777. 23. Wein A, Barrett DM. Etiologic possibilities for increased pelvic floor electromyography activity during cystometry. J Urol. 1982; 127:949-952. 24. Athanasopoulos A, Gyftopoulos K, Giannitsas K, et al. Effect of alfuzosin on female primary bladder neck obstruction. Int Urogynecol J Pelvic Floor Dysfunct. 2009;20:217-222. 25. Carlson KV, Rome S, Nitti VW. Dysfunctional voiding in women. J Urol. 2001;165:143-147.
59
26. Barrett DM. Disposable (infant) surface electrocardiogram electrodes in urodynamics: a simultaneous comparative study of electrodes. J Urol. 1980;124:663-665. 27. O’Donnell PD. Electromyography. In: Nitti VW, editor. Practical Urodynamics. Philadelphia: WB Saunders; 1998:65-71. 28. Kirby AC, Nager CW, Litman HJ, et al. Perineal surface electromyography does not typically demonstrate expected relaxation during normal voiding. Neurourol Urodyn. 2011;30:1561-1596.
60
29. Mahajan ST, Fitzgerald MP, Kenton K, et al. Concentric needle electrodes are superior to perineal surface-patch electrodes for electromyographic documentation of urethral sphincter relaxation during voiding. BJU Int. 2006;97:117-120. 30. Glassberg KI, Combs AJ, Horowitz M. Nonneurogenic voiding disorders in children and adolescents: clinical and videourodynamic findings in 4 specific conditions. J Urol. 2010;184:21232127.
UROLOGY 80 (1), 2012
METHODS
RESULTS
CONCLUSION
To determine the clinical and urodynamic differences in the presentation and the value of simultaneous fluoroscopy in dysfunctional voiding (DV) and primary bladder neck obstruction (PBNO); the 2 most common causes of non-neurogenic “functional” bladder outlet obstruction in women. A review of our urodynamic study database (March 2003 to August 2009) was conducted. DV was diagnosed when increased external sphincter activity was found during voluntary voiding on electromyography (EMG) or fluoroscopy. PBNO was diagnosed when a failure of bladder neck opening was noted on fluoroscopy during voiding. The demographics, symptoms, and urodynamic study parameters were collected. Comparisons were done using chi-square and 2-tailed t-tests. DV was diagnosed in 34 women and PBNO in 16. The patients with DV were younger than those with PBNO (40.9 vs 59.2 years, P ⬍ .001). Women with DV showed a clinical trend toward having more storage symptoms than those with PBNO and fewer voiding symptoms. Patients with DV had a greater mean maximal flow rate (12 vs 7 mL/s, P ⫽ .027) and lower mean postvoid residual urine volume (125 vs 400 mL, P ⫽ .012). No significant differences were found in maximal detrusor pressure, detrusor pressure at maximal flow rate, or detrusor overactivity. EMG showed increased activity during voiding in 79.4% of those with DV and 14.3% of those with PBNO (P ⬍ .001). Clinically, women with DV and PBNO had similar presentations, although those with PBNO had poorer emptying. The flow rates and patterns seemed to differ between those with DV and PBNO, although the voiding pressures were similar. EMG alone would have given the wrong diagnosis in 20.6% of those with DV (false negative) and 14.3% of those with PBNO (false positive). When fluoroscopy is used to define these entities, the accuracy of EMG to differentiate them is questionable. UROLOGY 80: 55– 60, 2012. © 2012 Elsevier Inc.
B
ladder outlet obstruction (BOO) in women can be categorized into 2 larger divisions: those caused from a fixed anatomic obstruction and those that are functional.1 Anatomic obstruction can often be predicted from the symptoms, history, and physical examination findings (eg, emptying symptoms and/or incomplete emptying in a woman with high-grade pelvic organ prolapse or urethral stricture or after recent anti-incontinence surgery). However, functional obstruction can be diagnosed only during the act of voiding, because no
Financial Disclosure: The authors declare that they have no relevant financial interests. From the Department of Urology, New York University Langone Medical Center, New York, New York Reprint requests: Benjamin M. Brucker, M.D., Department of Urology, New York University LangoneMedical Center, 150 East 32nd Street, Second Floor, New York, NY 10016. E-mail: [email protected] Submitted: March 5, 2012, accepted (with revisions): April 9, 2012
obvious anatomic abnormality will be associated with the patient’s symptoms. The most common causes of functional BOO in neurologically intact women are dysfunctional voiding (DV) and primary bladder outlet obstruction (PBNO).2,3 DV occurs when there is improper relaxation or contraction of the external striated sphincter during voiding. In PBNO, a failure of the bladder neck (or proximal sphincter) to open and funnel normally during voiding occurs. The accurate diagnosis of these entities is critical because the endpoints of the treatment algorithms are vastly divergent. The first-line treatment of DV involves education with biofeedback to “unlearn the learned behavior,” and PBNO is often initially treated with ␣-blockers and sometimes requires surgical incision of the bladder neck.4,5 Physical therapy with biofeedback can be time intensive and costly, and incision of the bladder neck is invasive with a small, but real, risk of incontinence. These facts make empiric therapy of suspected voiding phase dysfunction less desirable.
© 2012 Elsevier Inc. All Rights Reserved
0090-4295/12/$36.00 http://dx.doi.org/10.1016/j.urology.2012.04.011
55
Many attempts have been made to better define what constitutes BOO in women; however, very little has been done to define the urodynamic difference between the specific etiologies that result in BOO.3,6-9 Obviously, the least invasive and most cost-effective method to differentiate between the causes would be preferable. The present study sought to determine whether urodynamic and clinical differences exist between neurologically intact women with functional obstruction. The diagnosis of PBNO and DV has historically been made by urodynamic studies (UDS).10 Surface electromyography (EMG) activity has also been used in an attempt to separate these diagnoses.11 However, surface EMG is prone to artifacts, causing apparent increased activity during voiding.12 Needle EMG is uncomfortable, and many women are adverse to its use. Thus, we have come to rely on fluoroscopic imaging during UDS (videourodynamics) to localize any obstruction at the bladder neck or external sphincter. In addition, one can often see spasmodic activity of the sphincter during fluoroscopic assessment of voiding in patients with DV. We believe this to be the most accurate assessment of the voiding phase, and this has become our “standard” for diagnosing women with functional obstruction and differentiating its cause. However, we questioned how valuable videourodynamics truly is for this purpose and whether DV and PBNO in women can be accurately differentiated from each other without the use of fluoroscopy.
MATERIAL AND METHODS We performed a retrospective review of all female videourodynamic studies at a single institution from March 2003 to August 2009. It has been our practice to perform videourodynamics in all women with suspected or possible voiding phase dysfunction and other selected conditions undergoing UDS. The reasons for the evaluation, determined from the clinical history and presenting symptoms, included frequency and/or urgency refractory to conservative or medical therapy, urgency incontinence, elevated postvoid residual (PVR) urine volume, urinary retention, slowed urinary flow, recurrent urinary tract infection, stress urinary incontinence not demonstrated on physical examination, mixed incontinence, unaware incontinence, and before pelvic organ prolapse surgery (to assess for stress urinary incontinence with the reduction of prolapse, if not seen at the examination or with a supine cough stress test). All videourodynamics were performed with 7F transurethral and rectal balloon catheters according to the International Continence Society standards.13,14 EMG of the pelvic floor was done with surface electrodes. Filling cystometry was performed at a rate of 50 mL/min with radiographic contrast material. In cases of severe detrusor overactivity or known small functional capacity, the rate was decreased to 30 mL/min. Fluoroscopic images were obtained at select times as determined by the clinical scenario (ie, filling, Valsalva, void/attempt to void, after voiding). All studies were performed and interpreted by an experienced urologist. Data were collected on pressure-flow parameters, EMG activity during voluntary voiding, and radiographic images. Only patients with a final diagnosis of BOO were included in the present analysis. BOO was diagnosed using the videouro56
dynamic criteria of a sustained detrusor contraction of any magnitude with radiographic evidence of obstruction.3 Using the International Continence Society definition, DV was characterized by an intermittent and/or a fluctuating flow rate due to intermittent contractions of the periurethral striated muscles during a sustained voluntary detrusor contraction in a neurologically normal woman.15 This was diagnosed using a combination of increased EMG activity during voiding in the absence of abdominal straining or radiographic evidence of sphincteric contraction/failure to relax even in the absence of increased EMG activity during voiding. The clinical history was used to differentiate this entity from detrusor external sphincter dyssynergia where a known neurologic cause must exist. For cases in which DV was suspected, the uroflow from UDS was correlated with the noninvasive uroflow findings. If the intubated uroflow suggested DV, but the noninvasive uroflow did not, the DV was thought to be test induced, and such patients were not diagnosed with DV and thus were excluded from the present analysis. PBNO was defined as failure of the bladder neck to open adequately (remained closed or narrow on fluoroscopic images) during voiding, resulting in obstruction of urinary flow in the absence of other causes of obstruction or increased striated sphincter activity.3 In cases in which increased EMG activity was determined to be secondary to artifact (ie, poor contact, leading to persistent increased activity, movement during voiding, abdominal straining), EMG activity alone was not considered as increased striated sphincter activity. Urodynamic tracings and the diagnosis when UDS was performed were reviewed. The data and diagnosis recorded in the initial interpretation of the UDS were captured. The filling phase parameters included the first desire to void (the feeling when the woman first becomes aware of bladder filling), normal desire to void (the feeling that led the woman to pass urine in the next convenient moment but can delay if necessary), presence/absence of detrusor overactivity, maximum cystometric capacity, and compliance. The voiding phase parameters included the maximal flow rate (Qmax), detrusor pressure at Qmax, maximal detrusor pressure, and PVR. The maximal Pdet was recorded for cases in which no urine flow was recorded but a sustained detrusor contraction was present or if artifact (ie, a spike) of the uroflow made interpretation of the maximal detrusor pressure recorded at Qmax erroneous. EMG activity during voiding was noted as increased if the average amplitude of the EMG was increased from baseline during flow. The fluoroscopic images were reviewed for evidence of bladder neck relaxation or narrowing at the external urethral sphincter during voiding. The charts were reviewed for patient age, the presence of storage symptoms (frequency, urgency, and urgency incontinence), the presence of voiding symptoms (slow urinary stream, hesitancy, the sensation of incomplete emptying, and dysuria in the absence of infection). Patients were excluded if no flow was present and a sustained detrusor contraction was not, pressure flow data were missing, radiographic evidence of obstruction was missing, or if they had voided uncharacteristically during UDS compared with normal voiding. This included noncharacteristic straining or flow rates that the urodynamacist believed were not commensurate with noninvasive uroflow. The categorical variables were compared using chi-square and contingency tables. Continuous variables were evaluated using analysis of variance and t tests. A 2 ⫻ 2 table was used to calculate the sensitivity and specificity. UROLOGY 80 (1), 2012
Table 1. Urodynamic differences between DV and PBNO Variable First desire to void (mL) Normal desire to void (mL) Cystometric capacity (mL) Compliance (mL/cm H2O) Qmax (mL/s) PdetQmax (cm H2O) PdetMax (cm H2O) PVR (mL)
DV
PBNO
P Value
185.78 (26-720) 302.79 (56-1401) 395.32 (147-941) 118.75 (21.11-470.50) 12.06 (1-35.7) 38.94 (11-90) 46.11 (18-102) 124.50 (0-700)
226.19 (41-552) 341.38 (51-720) 534.56 (172-1500) 112.16 (57.33-253.67) 7.07 (0-16.0) 42.00 (18-95) 47.63 (24-100) 399.56 (0-1413)
.403 .612 .072 .821 .027* .608 .595 .012*
DV, dysfunctional voiding; PBNO, primary bladder neck obstruction; Qmax, maximal urinary flow rate; PdetQmax, detrusor pressure at Qmax; PdetMax, maximal Pdet; PVR, postvoid residual (urine volume). Data presented as mean, with ranges in parentheses. * Statistically significant.
RESULTS The analysis included 16 women diagnosed with PBNO and 34 women diagnosed with DV. The women with PBNO were, on average, older than those with DV (59.2 and 40.9 years, respectively, P ⬍ .001). The women with PBNO were also more likely to present with voiding symptoms than the women with DV but this failed to reach statistical significance (75% and 53%, respectively, P ⫽ .13). In contrast, a prevalence of storage symptoms was more likely in those with DV than in those with PBNO but, again, did not reach statistical significance (61% and 38%, respectively, P ⫽ .13). The presence of detrusor overactivity was similar in the 2 groups (26.3% DV and 31.3% PBNO, P ⫽ .71). A summary of the urodynamic findings is listed in Table 1. Women with PBNO were noted to have a mean Qmax significantly lower statistically than the women with DV (Qmax 7.07 and 12.06 mL/s respectively; P ⫽ .027). Also, a significantly greater average PVR urine volume was found in women with PBNO (399.56 vs 124.5 mL, P ⫽ .012). These results were confirmed by examination of the noninvasive uroflow/PVR data. The remainder of the pressure flow data did not show statistically significant differences. The EMG showed increased activity during voiding in 79.4% of the women diagnosed with DV (according to the standard of videourodynamics). Also, increased EMG activity with voiding was seen in 14.3% of patients with a final diagnosis of PBNO (videourodynamics; P ⬍ .001). In neurologically intact women with the diagnosis of BOO, EMG had a sensitivity of 79% and specificity of 85% to diagnose DV when videourodynamics was used as the standard. Thus, the positive predictive value of increased EMG activity was 93% and the negative predictive value was 63%. In all case of DV, the patients’ bladder neck was fluoroscopically seen to funnel appropriately.
COMMENT PBNO as a cause of BOO in women has been identified as the underlying cause in 9%-16% of obstructed women.2,3 Our study findings were consistent, with UROLOGY 80 (1), 2012
10.2% of obstructed women diagnosed with PBNO. This is a condition in which the bladder neck fails to open adequately during voiding, resulting in obstruction of urinary flow in the absence of increased striated sphincter activity or another anatomic obstruction. First described in 1933 by Marion,16 in 1973, Turner-Warwick et al17 advocated the use of UDS and fluoroscopy to diagnose bladder neck dysfunction in men. The entity was later described in women.18 The precise cause of PBNO has not been clearly elucidated.19,20 The accurate and timely diagnosis of PBNO remains the primary challenge.21 Failure of the bladder neck to open with a detrusor contraction can only be diagnosed by simultaneous imaging of the bladder outlet during voiding. Figure 1A shows the classic videourodynamic findings of PBNO. Involuntary contraction of the external urethral sphincter during voiding in the absence of neurologic injury or disease is known as DV. It has also been described as learned voiding dysfunction and by some as non-neurogenic, neurogenic bladder.22 DV, as defined by the International Continence Society, is an intermittent and/or fluctuating flow rate due to involuntary intermittent contractions of the periurethral striated muscle during voiding, in neurologically normal subjects.15 Figure 1B illustrates the classic videourodynamic findings of DV. It can be difficult to diagnose using surface EMG electrodes because of other factors that can produce increased EMG activity on a tracing, including attempts at augmenting bladder contractions by abdominal straining, movement, guarding reflex, and wet electrodes.23 Using fluoroscopy has proved valuable in the diagnosis because one can see a dilation of the urethra to the level of the striated sphincter (spinning top urethra) and/or intermittent contractions of the striated sphincter. The average age of women with PBNO in the present study was 59.2 years, consistent with contemporary series.24 In the study by Athanasopoulos et al,24 the mean Qmax of 9.74 mL/s was commensurate with the mean Qmax seen in our study (7.07 mL/s). The detrusor pressure at Qmax was greater in their series (99.72 vs 46.11 cm H2O), but they required obstruction to be present on the Blaivas-Groutz nomogram, possibly accounting for the difference. 57
Figure 1. A (top)- This represents the classic fluoroscopic, pressure flow and EMG findings of PBNO. B (bottom)- This is another pressure flow tracing showing high pressure/low flow voiding. The EMG in this case is clearly increased during voiding. The inset fluoroscopic image shows a narrowing at the external urethral sphincter. This is an excellent example of DV. (Color version available online.)
Carlson et al25 characterized 26 women with DV and found a mean age of 39.2 years comparable to the mean age in our series (40.9 years). The mean detrusor pressure at Qmax and PVR urine volume (50.3 cm H2O and 103.4 mL) were also very similar. The mean Qmax (10.4 mL/s) was between the 2 mean values in our series. Although the Qmax values were significantly different statistically, there was likely a large overlap between the 2 diagnoses. We assumed that videourodynamics would be the most accurate method to diagnose voiding phase dysfunction in women; thus, for the present comparative study, the videourodynamic diagnosis was considered the standard. Our analysis of the UDS findings revealed that had we relied on the surface EMG findings alone, we would have derived the incorrect diagnosis in 20.6% of the women with DV (eg, women with high sustained detrusor pressure during voluntary voiding, resulting in a low flow rate with a bladder neck that opens, but with narrowing at the external urethral sphincter without an appreciable increase in EMG activity during voiding and no anatomic obstruction such as stricture or periurethral fibrosis). In contrast, increased EMG activity during voiding was seen in 14.3% of women with PBNO (eg, a patient voiding with high pressure and low flow without any funneling or opening of the bladder neck and during voiding a substantial increase occurs EMG activity that would have suggested DV if fluoroscopy had not been used). Surface EMG use during UDS was first described by Barrett26 to determine perineal relaxation as an indirect 58
indicator of simultaneous external sphincter relaxation during the voiding phase. Compared with needle electrodes, surface electrodes are favored by many clinicians because of their noninvasive nature and ease of placement. Although a good surrogate for sphincter activity, the argument can be made that the signal source is inferior.27 EMG activity during UDS is affected by various factors, including patient factors, technology limitations, and operator factors. Patient factors include straining during voiding and guarding. Technological limitations include failure to specifically depict external urethral sphincter activity, poor skin contact, and wet leads.12 Finally, operator factors include failure to properly place the leads, improper interpretation, and failure to recognizing poor lead contact. A recent multicenter study of women with predominant stress incontinence found on preoperative UDS that 50% of the women had EMG amplitudes greater during voiding than during filling.28 The investigators noted that none of these women would have been classified as having DV, supporting our finding that the EMG results need to be cautiously interpreted. We suggest that the additional information provided by fluoroscopy is valuable if the diagnosis of DV is considered. A study comparing surface patch electrodes and concentric needle electrodes found unanimous agreement among the external reviewer interpretations of individual EMG in only 27% of tracings.29 Quiescence of EMG during voiding was seen in only 28% of the interpretable UROLOGY 80 (1), 2012
tracings, although 70% of the women had stress incontinence and no other voiding phase pathologic features. Some have advocated the use of the interval between pelvic floor relaxations and the start of urine flow to screen for PBNO in children.11,30 Obvious differences exists in the body habitus that might make EMG more accurate in children. Furthermore, the techniques used in these pediatric studies (ie, testing direct sphincter stimulation, sensitivity adjustment, audio monitoring) might not be applicable in the settings in which most women undergo testing. Our study was retrospective and thus had limitations. We did not have a specific protocol in place during UDS to standardize EMG accuracy (eg, testing EMG recording, troubleshooting poor signals, standard bandwidth). However, while performing and interpreting the UDS, the clinicians had no knowledge that these variables would be studied in the future. Therefore, the results depict UDS as they were routinely done and were not altered by a study protocol or preconceived hypothesis. Furthermore, our studies were all performed at a referral center by physicians specializing in voiding dysfunction; thus, we might have overestimated the utility of EMG by those less familiar with these less common diagnostic and clinical findings. Finally, the study reported urinary symptoms according to the patients’ presenting symptoms and not validated questionnaires. Though this practical approach might have clinical relevance, a standardized assessment of lower urinary tract symptoms might give more precise insight into the nature of urinary symptoms in these women. Although fluoroscopic imaging during UDS is widely used in many specialty centers, it is by no means the most commonly practiced method. EMG remains a valuable part of UDS, especially where fluoroscopy is not available or the patient and/or physician decide they would prefer to avoid fluoroscopy. However, in women in whom the suspicion of BOO is high, fluoroscopy in extremely useful to differentiate and diagnose DV and/or PBNO.
CONCLUSIONS DV and PBNO are both classified as functional disorders. These entries have a similar presentation. Our results showed that women with PBNO tended to have poorer emptying. The flow rates and patterns differed, but the voiding pressures did not. The greater flow rate seen in DV might be due to the intermittent nature of the obstruction compared with the fixed nature of PBNO. When fluoroscopy is used to define these entities, the accuracy of EMG to differentiate them is questionable. References 1. Patel R, Nitti VW. Bladder outlet obstruction in women: prevalence, recognition, and management. Curr Urol Rep. 2001;2:379387. 2. Kuo HC. Videourodynamic characteristics and lower urinary tract symptoms of female bladder outlet obstruction. Urology. 2005;66: 1005-1009.
UROLOGY 80 (1), 2012
3. Nitti VW, Tu LM, Gitlin J. Diagnosing bladder outlet obstruction in women. J Urol. 1999;161:1535-1540. 4. Minardi D, d’Anzeo G, Parri G, et al. The role of uroflowmetry biofeedback and biofeedback training of the pelvic floor muscles in the treatment of recurrent urinary tract infections in women with dysfunctional voiding: a randomized controlled prospective study. Urology. 2010;75:1299-1304. 5. Blaivas JG, Flisser AJ, Tash JA. Treatment of primary bladder neck obstruction in women with transurethral resection of the bladder neck. J Urol. 2004;171:1172-1175. 6. Chassagne S, Bernier PA, Haab F, et al. Proposed cutoff values to define bladder outlet obstruction in women. Urology. 1998;51:408411. 7. Defreitas GA, Zimmern PE, Lemack GE, et al. Refining diagnosis of anatomic female bladder outlet obstruction: comparison of pressure-flow study parameters in clinically obstructed women with those of normal controls. Urology. 2004;64:675-679. 8. Lemack GE, Zimmern PE. Pressure flow analysis may aid in identifying women with outflow obstruction. J Urol. 2000;163:18231828. 9. Blaivas JG, Groutz A. Bladder outlet obstruction nomogram for women with lower urinary tract symptomatology. Neurourol Urodyn. 2000;19:553-564. 10. Dupont M, Raz S. Urodynamics: contemporary evolution. In: Nitti VW, editor. Practical Urodynamics. Philadelphia: WB Saunders; 1998:2-4. 11. Combs AJ, Grafstein N, Horowitz M, et al. Primary bladder neck dysfunction in children and adolescents I: pelvic floor electromyography lag time—a new noninvasive method to screen for and monitor therapeutic response. J Urol. 2005;173:207-210. 12. Bradley CS, Smith KE, Kreder KJ. Urodynamic evaluation of the bladder and pelvic floor. Gastroenterol Clin North Am. 2008;37:539552. 13. Groutz A, Blaivas JG, Chaikin DC. Bladder outlet obstruction in women: definition and characteristics. Neurourol Urodyn. 2000;19: 213-220. 14. Carlson KV, Fiske J, Nitti VW. Value of routine evaluation of the voiding phase when performing urodynamic testing in women with lower urinary tract symptoms. J Urol. 2000;164:1614-1618. 15. Abrams P, Cardozo L, Fall M, et al. The standardisation of terminology of lower urinary tract function: report from the standardisation Sub-Committee of the International Continence Society. Neurourol Urodyn. 2002;21:167-178. 16. Marion G. Surgery of the neck of the bladder. Br J Urol. 1933;5: 351-380. 17. Turner-Warwick R, Whiteside CG, Worth PH, et al. A urodynamic view of the clinical problems associated with bladder neck dysfunction and its treatment by endoscopic incision and transtrigonal posterior prostatectomy. Br J Urol. 1973;45:44-59. 18. Diokno AC, Hollander JB, Bennett CJ. Bladder neck obstruction in women: a real entity. J Urol. 1984;132:294-298. 19. Leadbetter GW, Leadbetter WF. Diagnosis and treatment of congenital bladder-neck obstruction in children. N Engl J Med. 1959; 260:633-637. 20. Awad SA, Downie JW, Lywood DW, et al. Sympathetic activity in the proximal urethra in patients with urinary obstruction. J Urol. 1976;115:545-547. 21. Padmanabhan P, Nitti VW. Primary bladder neck obstruction in men, women, and children. Curr Urol Rep. 2007;8:379-384. 22. Hinman F. Nonneurogenic neurogenic bladder (the Hinman syndrome)—15 years later. J Urol. 1986;136:769-777. 23. Wein A, Barrett DM. Etiologic possibilities for increased pelvic floor electromyography activity during cystometry. J Urol. 1982; 127:949-952. 24. Athanasopoulos A, Gyftopoulos K, Giannitsas K, et al. Effect of alfuzosin on female primary bladder neck obstruction. Int Urogynecol J Pelvic Floor Dysfunct. 2009;20:217-222. 25. Carlson KV, Rome S, Nitti VW. Dysfunctional voiding in women. J Urol. 2001;165:143-147.
59
26. Barrett DM. Disposable (infant) surface electrocardiogram electrodes in urodynamics: a simultaneous comparative study of electrodes. J Urol. 1980;124:663-665. 27. O’Donnell PD. Electromyography. In: Nitti VW, editor. Practical Urodynamics. Philadelphia: WB Saunders; 1998:65-71. 28. Kirby AC, Nager CW, Litman HJ, et al. Perineal surface electromyography does not typically demonstrate expected relaxation during normal voiding. Neurourol Urodyn. 2011;30:1561-1596.
60
29. Mahajan ST, Fitzgerald MP, Kenton K, et al. Concentric needle electrodes are superior to perineal surface-patch electrodes for electromyographic documentation of urethral sphincter relaxation during voiding. BJU Int. 2006;97:117-120. 30. Glassberg KI, Combs AJ, Horowitz M. Nonneurogenic voiding disorders in children and adolescents: clinical and videourodynamic findings in 4 specific conditions. J Urol. 2010;184:21232127.
UROLOGY 80 (1), 2012