Levator co-activation is a significant confounder of pelvic organ descent on Valsalva maneuver

Transcription

1 Ultrasound Obstet Gynecol 2007; 30: Published online 14 August 2007 in Wiley InterScience ( DOI: /uog.4082 Levator co-activation is a significant confounder of pelvic organ descent on Valsalva maneuver A. K. ÖRNÖ*andH.P.DIETZ *Department of Obstetrics and Gynecology, Clinical Sciences Lund, University of Lund, Lund, Sweden and Nepean Clinical School, University of Sydney, Sydney, Australia KEYWORDS: biofeedback; bladder neck descent; levator ani; levator hiatus; pelvic floor; pelvic organ prolapse; ultrasound; Valsalva maneuver ABSTRACT Objective A Valsalva maneuver is used clinically and on imaging in order to determine female pelvic organ prolapse. We have examined the potential confounding effect of levator co-activation at the time of a Valsalva maneuver and the impact of repetition with biofeedback instruction. Methods Fifty nulliparous women at weeks gestation received 3D/4D translabial ultrasound investigation in the dorsal resting position after bladder emptying. Valsalva maneuvers were recorded initially and after repeated attempts with visual biofeedback both during the maneuver and after, with the operator demonstrating findings on the ultrasound monitor, in order to abolish levator coactivation. Offline analysis was subsequently undertaken. Results Significant differences between first and optimal Valsalva maneuver were found for bladder neck position, bladder neck descent, hiatal sagittal diameter and hiatal area on Valsalva. In a minority of women (22/50) we observed a reduction in the sagittal hiatal diameter on first Valsalva maneuver, indicating levator co-activation. A reduction in sagittal diameter was seen in only 11/50 after instruction. Levator co-activation was associated with significantly lower bladder neck descent. Conclusion The Valsalva maneuver is frequently accompanied by a pelvic floor muscle contraction. Levator co-activation may be a substantial confounder, reducing pelvic organ descent. Without repetition and digital, auditory or visual biofeedback, women may not perform a correct Valsalva maneuver. Biofeedback markedly reduces the likelihood of levator co-activation but does not abolish it completely. Copyright 2007 ISUOG. Published by John Wiley & Sons, Ltd. INTRODUCTION A Valsalva maneuver is defined as forced expiration against a closed glottis, requiring contraction of the diaphragm and abdominal muscles, in order to obtain markedly increased intra-abdominal pressure. To maintain continence, a contraction of the levator ani muscle is probably a physiological accompaniment of a Valsalva maneuver. In other situations, such as defecation and voiding but also during vaginal childbirth, relaxation of the levator ani muscle is mandatory. The maneuver is used clinically and on imaging in order to determine female pelvic organ prolapse 1. Performing a Valsalva maneuver with a relaxed pelvic floor can be embarrassing for the patient, since involuntary passage of urine, wind or even stool may occur. As a consequence, subjects often involuntarily or voluntarily co-activate the levator ani muscle. This may be the main reason why normal values for pelvic organ descent in nulliparous women vary markedly in the published literature 2,3. Scoring systems used to determine the extent of pelvic organ prolapse 4 are based on the assumption of a subject s best (i.e. correctly performed) Valsalva maneuver, an assumption which may be incorrect in many cases. The development of imaging techniques has made it possible to grade the effectiveness of a Valsalva maneuver by measuring bladder neck descent, the increase in hiatal area and the increase in sagittal diameter between the posterior rim of the symphysis pubis and the pubovisceral muscle 5 7, and to detect levator co-activation by observing a decrease in the mid-sagittal hiatal diameter on Valsalva. The null hypothesis of this study was that visual and auditory biofeedback does not improve Valsalva performance. The effectiveness of Valsalva maneuvers was checked by quantitation of biometric changes in three-dimensional (3D)/4D ultrasound volume datasets, Correspondence to: Dr A. K. Örnö, Obstetrics and Gynecology, Department of Clinical Science, University of Lund, S Lund, Sweden ( ann-kristin.orno@med.lu.se) Accepted: 16 April 2007 Copyright 2007 ISUOG. Published by John Wiley & Sons, Ltd. ORIGINAL PAPER

2 Levator activation on Valsalva maneuver 347 with bladder neck descent on Valsalva the main outcome parameter. SUBJECTS AND METHODS In a prospective observational study, we examined 50 nulliparous, pregnant women at weeks gestation as part of a larger study on pelvic floor function and anatomy in pregnancy. 3D/4D translabial ultrasound was performed, using a Voluson 730 expert system (GE Kretz Ultrasound, Zipf, Austria) with an 8 4-MHz volume probe. Data volumes were obtained with the woman in the dorsal resting position, after bladder emptying. The women were first instructed orally to perform a strong push or bearing-down effort, and this first Valsalva was recorded as a cine-loop of volume datasets and stored for later analysis. Repeated trials were then performed to optimize the effort. Visual biofeedback was provided by demonstrating dynamic changes in both the midsagittal and the axial planes on the ultrasound monitor, both during the maneuver and on replaying the stored volumes to explain optimal performance. The women were encouraged to avoid a levator contraction (levator co-activation) during pushing, and appearances suggestive of a concomitant pelvic floor contraction, that is, a narrowing of the hiatus, were demonstrated in real time and on playback. Multiple Valsalva maneuvers were stored for later analysis, until the operator felt that no further improvement was to be expected. In order to show that a narrowing of hiatal diameters is a general feature of levator activation, we also analyzed volumes obtained on maximal pelvic floor contraction. Offline analysis was undertaken using the software 4D View version 5.0 (GE Kretz Ultrasound), in order to obtain standard measures of pelvic organ descent. Bladder neck descent and levator hiatal area were compared before and after biofeedback. Co-activation of the levator ani muscle was postulated to be present when the distance between the inferoposterior rim of the symphysis pubis and the pubovisceral muscle component of the levator ani muscle was less than that in the resting state. The operator performing the analysis (A.K.O.) was blinded to all clinical data and had not been present at the time of data acquisition. This project was performed as part of a larger study on pelvic floor function before and after childbirth that was approved by the local Human Research Ethics Committee. All participants had provided written informed consent. As there were no pilot data available to inform planning for this project, we did not perform a power calculation. Statistical analysis was performed after Normality testing, using Minitab version 13 (Minitab Inc., State College, PA, USA). We used Student s t-test for paired observations and for between-group comparisons. P < 0.05 was considered statistically significant. RESULTS The mean age of participants included in this study was 27 (range 18 38) years and they were seen at a mean gestation of 36.6 (range 36 38) weeks. All were nulliparous, with none having carried a previous pregnancy beyond 12 weeks. Using offline analysis, we assessed both the first volume dataset (documenting the first Valsalva maneuver) and a second dataset documenting the most effective of at least three maneuvers, as well as a volume dataset obtained on pelvic floor muscle contraction. All but four patients reduced the mid-sagittal diameter of the levator hiatus on pelvic floor muscle contraction. The effect of a contraction on this parameter was highly significant (57 mm (SD 8) at rest vs. 49 mm (SD 7) on contraction; P < 0.001), proving that reduction of the mid-sagittal diameter of the hiatus is strongly associated with contraction of the levator ani. Table 1 shows a comparison of results for the first and the optimal Valsalva maneuver recorded for each patient. Significant differences were found for position of the bladder neck on Valsalva maneuver, total bladder neck descent, hiatal sagittal diameter and hiatal area on Valsalva maneuver. Figure 1 demonstrates a case of moderate levator co-activation, resulting in much less bladder neck descent on first Valsalva compared to that on optimal Valsalva. Figure 2 demonstrates an increased hiatal area as a result of biofeedback training. As is clearly visible in Figure 2, some women effected a reduction in the sagittal hiatal diameter on first Valsalva maneuver, indicating levator co-activation. This was common on first Valsalva (22/50) and still seen in 11/50 after instruction. This change in co-activation before and after biofeedback training was statistically significant (P = 0.03 on Fisher s exact test). Table 2 gives bladder neck descent and changes in levator hiatal area on Valsalva between those women coactivating and those relaxing their pelvic floor on Valsalva before and after training. Differences between the coactivating and the relaxing groups were significant for both the first attempt and the best attempt at producing Table 1 Comparison of parameters of pelvic organ descent and hiatal dimensions, obtained on first and maximal (best) Valsalva Parameter First Valsalva Best Valsalva P Position of BN* at 29.4 (3.7) 28.8 (3.9) 0.4 rest (mm) Position of BN* on 19.0 (9.9) 12.8 (9.9) Valsalva (mm) BND (mm) 10.4 (9.9) 16.0 (8.7) Hiatal diameter (sagittal) 55.8 (10.3) 56.6 (7.6) 0.6 at rest (mm) Hiatal diameter (sagittal) 57.2 (12) 62.7 (10.6) on Valsalva (mm) Change in hiatal 1.4 (6.7) 6.0 (9.0) diameter (mm) Hiatal area on Valsalva (cm 2 ) 20.4 (5.9) 23.1 (6.5) Figures are given as mean (SD); paired Student s t-tests. *Bladder neck position was measured as the distance from the inferior margin of the symphysis pubis to the bladder neck. BN, bladder neck; BND, bladder neck descent.

3 348 Örnö anddietz Figure 1 The effect of levator co-activation on bladder neck descent. Mid-sagittal views at rest (a), on first Valsalva maneuver, confounded by levator co-activation (b) and on optimal Valsalva after biofeedback teaching (c). The examination was performed using translabial ultrasound with an 8 4-MHz transducer. The horizontal line signifies the inferior margin of the symphysis pubis. The vertical line indicates the bladder neck. It is evident that there is increasing bladder neck descent with improved Valsalva performance. LA, levator ani muscle. Figure 2 The effect of levator co-activation on hiatal dimensions. Axial views in the plane of minimal dimensions, at rest (a), on suboptimal Valsalva maneuver (b) and on optimal Valsalva (c). The examination was performed using translabial ultrasound with an 8 4-MHz transducer. The levator ani muscle is seen as a hyperechoic U-shaped structure. During Valsalva with relaxed levator ani muscle there was an increase in the sagittal hiatal diameter as well as in the hiatal area. Table 2 Bladder neck descent and increase in levator hiatal area in women co-activating the levator ani and in women able to relax the levator ani during Valsalva maneuver. At the first try, 22/50 women contracted the pelvic floor; after biofeedback, 11/50 contracted and 39/50 relaxed First Valsalva Best Valsalva Parameter BND (mm) Increase in area (cm 2 ) Parameter BND (mm) Increase in area (cm 2 ) Relaxing (n = 28) 13.3 (11.2) 5.7 (5.5) Relaxing (n = 39) 17.6 (8.8) 8.5 (8.7) Co-activating (n = 22) 6.6 (6.6) 2.8 (3.6) Co-activating (n = 11) 10.5 (5.9) 1.3 (3.8) P P Measurements are given as mean (SD); Student s t-test. BND, bladder neck descent. a Valsalva maneuver. The magnitude of the effect caused by co-activation can be estimated from the Best Valsalva columns of Table 2 (bladder neck descent of 17.6 mm in the relaxing group vs mm in the co-activating group; P = 0.016). DISCUSSION Levator co-activation is likely to influence the effect of a Valsalva maneuver on pelvic organ mobility, at least in nulliparous women. This is of importance for the

4 Levator activation on Valsalva maneuver 349 evaluation of female pelvic organ prolapse 4, since we rely on this maneuver for assessing pelvic organ descent. To our knowledge, there is no mention of levator coactivation as a confounder in the literature on clinical prolapse assessment, and very little mention of this phenomenon in the literature generally. In our study we have been able to show that levator co-activation is common in nulliparous women and associated with significantly reduced bladder neck descent and lower hiatal diameter and area measurements. We have also shown that in some women biofeedback training can abolish levator co-activation on Valsalva and result in higher measurements for most examined parameters. These findings again demonstrate the utility of ultrasound as a biofeedback tool, together with oral instructions and repetition, as shown previously in the context of pelvic floor muscle exercise training 8. A second potential confounder of the effect of a Valsalva maneuver is intra-abdominal pressure, a factor that has received more attention in the literature. While we did not measure intra-abdominal pressures for this study, and while intra-abdominal pressure will of course influence pelvic organ descent, this parameter is very unlikely to have any bearing on our results regarding the existence and the effect of levator co-activation. No variation of Valsalva pressure would be expected to reduce hiatal diameters, unless intra-abdominal pressures were to fall below zero. It is understood that the population examined here, pregnant nulliparous women, may not be representative of the general population. In nulliparous women, the levator ani muscle and the nerve supply of the pelvic floor can be assumed to be intact, as opposed to those in about 20% of the parous population Effective levator co-activation may therefore be more likely than in the general population. It would be useful to repeat this study in parous symptomatic women, many of whom would have learned how to perform an effective Valsalva maneuver during childbirth, reducing the likelihood of co-activation. On the other hand, many women symptomatic of pelvic floor dysfunction will inadvertently or voluntarily contract the levator ani to prevent urinary or anal incontinence. Despite its limitations, however, this study has significant implications for prolapse assessment, whether by clinical examination or by imaging methods. We have documented a previously unquantified confounder which needs to be controlled for, both clinically and on imaging. On clinical examination levator relaxation is likely to be achieved by distension of the hiatus on using a speculum or digitally. However, it is evident from the clinical experience of the authors that levator coactivation often occurs on clinical prolapse assessment and may at times prevent adequate assessment, in particular in women with a strong, intact levator shelf. This phenomenon may explain why findings in the operating theater often are at variance with the preoperative clinical assessment. In practical terms, our results underscore the need to consider levator function when assessing for pelvic organ prolapse. This is particularly true on imaging, when it is impossible to distend the hiatus digitally or with an instrument, and carries particular importance for prolapse assessment using magnetic resonance imaging (MRI). While real-time ultrasound investigation allows for observation of the effect of maneuvers and correction by visual biofeedback, this is not currently the case for MRI, as true real-time imaging is impossible due to technical limitations. Consequently, any patient undergoing MRI prolapse assessment should be taught how to perform an adequate Valsalva maneuver prior to the imaging assessment, either by digital examination or by ultrasound scanning. As the issue of levator co-activation has not been raised in any of the literature dealing with pelvic floor MRI and the majority of pelvic floor ultrasound studies, data obtained in those studies must now appear suspect. However, even with significant biofeedback teaching, it appears that a minority of nulliparous women will continue to co-activate the levator ani. After training, a total of 39 of 50 women were able to perform a Valsalva without levator co-activation, but 11 remained unable to do so. This implies that clinicians will always have to consider the possibility of a false-negative assessment, in particular in women with a strong, intact pubovisceral muscle. Assessment of the levator ani should be an integral part of any clinical assessment for female pelvic organ prolapse. In conclusion, this investigation describes and quantifies a major confounder of the assessment for pelvic organ prolapse in young nulliparous women. Levator coactivation at the time of a Valsalva maneuver may significantly reduce pelvic organ descent. Biofeedback instruction and repetition may overcome this effect in some women and reduce the likelihood of false-negative findings. This seems particularly important when prolapse is assessed by imaging. Further comparative studies are required to assess whether biofeedback can improve upon current assessment methods. ACKNOWLEDGMENTS We would like to thank Dr Orawan Lekskulchai, Pelvic Floor Fellow, Nepean Hospital, Sydney, for help with data acquisition. REFERENCES 1. DeLancey JOL. Functional anatomy of the pelvic floor. In Imaging Pelvic Floor Disorders, Bartram CI, DeLancey JOL (eds). Springer Verlag: Berlin, 2003; Reed H, Waterfield A, Freeman RM, Adekanmi OA. Bladder neck mobility in continent nulliparous women: normal references. Int Urogynecol J 2002; 13: S4. 3. Peschers UM, Fanger G, Schaer GN, Vodusek DB, DeLancey JOL, Schuessler B. Bladder neck mobility in continent nulliparous women. BJOG 2001; 108: Bump RC, Mattiasson A, Bø K, Brubaker LP, DeLancey JO, Klarskov P, Shull BL, Smith AR. The standardization of

5 350 Örnö anddietz terminology of female pelvic organ prolapse and pelvic floor dysfunction. Am J Obstet Gynecol 1996; 175: Hol M, van Bolhuis C, Vierhout ME. Vaginal ultrasound studies of bladder neck mobility. BrJObstetGynaecol1995; 102: Dietz HP. Ultrasound imaging of the pelvic floor. Part II: threedimensional or volume imaging. Ultrasound Obstet Gynecol 2004; 23: Troeger C, Gugger M, Holzgreve W, Wight E. Correlation of perineal ultrasound and lateral chain urethrocystography in the anatomical evaluation of the bladder neck. Int Urogynecol J 2003; 14: Dietz HP, Clarke B, Wilson PD. The use of translabial ultrasound to quantify levator activity and teach pelvic floor muscle exercises. Int Urogynecol J 2001; 12: Dannecker C, Lienemann A, Fischer T, Anthuber C. Influence of spontaneous and instrumental vaginal delivery on objective measures of pelvic organ support: assessment with the pelvic organ prolapse quantification (POPQ) technique and functional cine magnetic resonance imaging. Eur J Obstet Gynecol Reprod Biol 2004; 15: Dietz HP and Steensma A. The prevalence and clinical significance of major morphological abnormalities of the levator ani. BJOG 2006; 113: Kearney R, Miller J, Ashton-Miller J, DeLancey J. Obstetric factors associated with levator ani muscle injury after vaginal birth. Obstet Gynecol 2006; 107: