Association between ICS POP-Q coordinates and translabial ultrasound findings: implications for definition of normal pelvic organ support

Transcription

1 Ultrasound Obstet Gynecol 216; 47: Published online 29 January 216 in Wiley Online Library (wileyonlinelibrary.com). DOI: 1.12/uog Association between ICS POP-Q coordinates and translabial ultrasound findings: implications for definition of normal pelvic organ support H. P. DIETZ, I. KAMISAN ATAN and A. SALITA Department of Obstetrics and Gynecology, Sydney Medical School Nepean, University of Sydney, Penrith, Australia KEYWORDS: cystocele; female pelvic organ prolapse; rectocele; ultrasound; uterine prolapse ABSTRACT Objectives Female pelvic organ prolapse is quantified on clinical examination using the pelvic organ prolapse quantification system of the International Continence Society (ICS POP-Q). Pelvic organ descent on ultrasound is strongly associated with symptoms of prolapse, but associations between clinical and ultrasound findings remain unclear. This study was designed to compare clinical examination and imaging findings, especially regarding cut-offs for the distinction between normal pelvic organ support and prolapse. Methods This was a retrospective study using 839 archived datasets of women referred to a tertiary urogynecological center for symptoms of lower urinary tract and pelvic floor dysfunction between June 211 and May 213. The main outcome measures were the maximum downward displacement of the anterior vaginal wall (point Ba), the cervix (point C) and the posterior vaginal wall (point Bp), the length of the genital hiatus (Gh) and the length of the perineal body (Pb), as defined by the ICS POP-Q; explanatory parameters were measures of pelvic organ descent on translabial ultrasound, ascertained by offline volume data analysis at a later date, by an operator blinded to all other data. Results Full datasets were available for 825 women. On clinical examination, 646 (78.3%) were found to have prolapse of at least POP-Q Stage 2. All coordinates on clinical examination were strongly associated with the ultrasound measurements of pelvic organ descent (P <.1). These relationships were almost linear, especially for the anterior compartment. Conclusions There is a near linear relationship between sonographic and clinical measures of prolapse. Previously proposed cut-offs to define significant prolapse on ultrasound and POP-Q (Ba.5 and cystocele 1 mm below the symphysis pubis, C 5 and uterine position of 15 mm above the symphysis pubis, Bp.5 and rectocele 15 mm below the symphysis pubis) are plausible and mutually consistent. Copyright 215 ISUOG. Published by John Wiley & Sons Ltd. INTRODUCTION Female pelvic organ prolapse is a common condition that significantly impacts on quality of life, and affects 1 2% of the parous female population to such a degree that they seek surgical intervention 1,2. Remarkably, there is very little information available on what prolapse actually is, i.e. what constitutes normal or abnormal pelvic organ support. It is generally quantified on clinical examination using the pelvic organ prolapse quantification system of the International Continence Society (ICS POP-Q), published in , although, to our knowledge, there has been no attempt to define normality on ICS POP-Q examination until recently 4. The POP-Q describes descent of the anterior and posterior vaginal wall and uterine cervix or, after hysterectomy, the vaginal vault, relative to the hymen, without reference to what is deemed normal. In clinical practice, the POP-Q is often used to stratify findings into stages Stage (normal) and Stages 1 4 (prolapse) although it has recently become clear that Stage 1 anterior and posterior compartment descent is likely to be within the normal range 5.Pelvic organ mobility in young nulliparous women seems to be largely genetically determined 6, with substantial variability between individuals. The main environmental factor influencing pelvic organ mobility appears to be vaginal childbirth 7, notably due to disruption of the levator ani muscle at the time of crowning of the fetal head 8. Correspondence to: Prof. H. P. Dietz, Department of Obstetrics and Gynaecology, Sydney Medical School Nepean, Nepean Hospital, Penrith, NSW 275, Australia ( hpdietz@bigpond.com) Accepted: 7 April 215 Copyright 215 ISUOG. Published by John Wiley & Sons Ltd. ORIGINAL PAPER

2 364 Dietz et al. There are two fundamentally different approaches to determining the normality of a clinical finding. One is purely mathematical; any finding that is within the range of mean ± 2 SD is regarded as normal. The other uses symptoms to define abnormal, which seems safe enough in a condition that is benign and not highly progressive 9. We are unaware of the first approach having been used in the context of clinical prolapse examination, although such data are available for imaging prolapse 1. The second approach has been utilized for both ICS POP-Q and ultrasound imaging 11,12, and, in all instances, the data presented enable the formulation of cut-offs for defining normality. Although sonographic assessment of prolapse was first described in 21, together with limited information on comparative findings 13, we have identified only one other study in the literature that provides a direct comparison of ICS POP-Q coordinates and sonographic findings (MEDLINE search, 1 November 214, keywords prolapse and ultrasound ). In this study, Lone et al. 14 excluded women with prolapse beyond the hymen, limiting the utility of their findings. In the present study we set out to compare POP-Q coordinates and translabial ultrasound findings for the entire range of measurements for assessing prolapse, by determining the imaging equivalents of given POP-Q coordinate measurements. This would enable direct comparison of measurements and cut-offs obtained by both methods and should allow for better communication between imaging specialists and clinicians, help establish definitions of normal pelvic organ support and avoid false-negative results. METHODS This was a retrospective study using 839 archived datasets of women examined in a tertiary urogynecological center for symptoms of lower urinary tract and pelvic floor dysfunction, between June 211 and May 213. All women had undergone a local standardized interview, with symptoms of prolapse defined as the sensation of a vaginal lump or bulge or a dragging sensation. Approval for this retrospective study was obtained from the local human research ethics committee (ref. NBMLHD HREC 13 11). The clinical examination included ICS POP-Q assessment 3, performed by one of four subspecialty trainees and two subspecialists, and four-dimensional (4D) translabial ultrasound in the supine position and after voiding 15, the latter performed within 15 min of the POP-Q assessment by one of approximately 2 different ultrasound trainees under the direct supervision of H.P.D. The ICS POP-Q evaluates anterior, central and posterior compartment descent by measuring, on Valsalva maneuver, the maximum downward displacement of the anterior vaginal wall as point Ba, of the cervix as point C and of the posterior vaginal wall as point Bp, all relative to the hymen, with measurements given in cm. Positions above the hymen are given as negative values and positions below are given as positive values 3. Gh Figure 1 Translabial ultrasound image in the midsagittal plane on maximal Valsalva maneuver, showing descent of bladder (B), uterus (U) and rectal ampulla (R), relative to pubic symphysis (S). Measurements represent maximal caudad organ position on Valsalva, without reference to position at rest. Measurements below inferior symphyseal margin (reference line) are negative and those above line are positive. describes the length of the genital hiatus, measured as the distance from the external urethral meatus to the posterior fourchette, and Pb describes the length of the perineal body, measured as the distance from the fourchette to the anus, with both coordinates recorded in cm on maximal Valsalva maneuver 3. Pelvic floor ultrasound has been routine practice at our unit since 24 and in this study it was performed using Voluson 73 Expert and Voluson S6 systems (GE Medical Systems, Zipf, Austria). The acquisition of ultrasound data was achieved as a cine loop (succession) of volumes, i.e. as 4D ultrasound. The analysis was undertaken at a later date by manipulating volume datasets with software (4D View v 1., GE Medical Systems) that is similar to a DICOM viewer for computed tomography or magnetic resonance imaging data, so that measurements can be obtained in two-dimensional planes; in this case, the midsagittal plane. All assessments were standardized, in the sense that all Valsalva maneuvers were required to last at least 6 s 16. Standardization for pressure was not attempted, as this is probably unnecessary 17. The main outcome measures were Ba, C, Bp, Gh and Pb, as defined by the ICS POP-Q. Offline analysis for pelvic organ descent (Figure 1) was performed by A.S. at a later date, using 4D View v.1. on a desktop PC, enabling blinding of the examiner against all clinical data. Ultrasound measurements of organ descent (bladder, uterus, rectal ampulla or enterocele) were obtained against a horizontal line positioned through the inferior symphyseal margin, giving the maximal caudad position on Valsalva, without any reference to position at rest (Figure 1). Measurements below the symphysis pubis are given as negative values and those above as positive values 15.The most effective Valsalva maneuver, in terms of producing organ descent, of at least three attempts was used for analysis. Posterior compartment descent was defined as maximal descent either of the rectal ampulla, or of an enterocele (descent of the small bowel, sigmoid colon or Copyright 215 ISUOG. Published by John Wiley & Sons Ltd. Ultrasound Obstet Gynecol 216; 47:

3 Comparison of POP-Q and ultrasound findings 365 omentum) if the latter descended further than the rectal ampulla. Statistical analysis Statistical analysis was undertaken using the software Minitab version 13 (Minitab Inc., State College, PA, USA) and SAS V9.2 (SAS Institute Inc., Cary, NC, USA). Pearson s correlations and t-tests were used for univariate analysis of normally distributed data. ANOVA analysis and graphs were used to define visually and mathematically the relationship between individual POP-Q coordinates (Ba, C, Bp) and ultrasound measurements of bladder and uterine descent, as well as maximal descent of either the rectal ampulla or a rectocele/enterocele. Multivariate regression modeling was performed to estimate the potential impact of confounders identified on univariate analysis, such as age, parity, menopausal status, body mass index (BMI), history of instrumental delivery, hysterectomy, incontinence or prolapse surgery and levator avulsion. Agreement between sonographic and clinical definitions of significant prolapse was tested using Cohen s kappa (κ). RESULTS During the study period, 839 women were seen in our unit for symptoms of lower urinary tract and pelvic floor dysfunction. Valsalva volumes were missing in six women owing to the unavailability of ultrasound equipment, which was required elsewhere. In two women, ultrasound volume data were technically suboptimal, and six women had either no or only a suboptimal clinical examination owing to severe vaginal stenosis or vaginismus, leaving a study population of 825 women. Table 1 provides the demographic data of the study population and summarizes their presenting symptoms. On clinical examination, 646 (78.3%) women were found to have prolapse of ICS Stage 2 or higher, in the form of a cystocele in 57.7% (n = 476), uterine prolapse in 8.7% (n = 72), vault prolapse in 8.8% (n = 73) and a rectocele in 56.2% (n = 464). An enterocele was diagnosed clinically in 21 (2.5%) women. The mean measurements of ICS POP-Q coordinates are given in Table 2. The equivalent data for ultrasound findings of organ descent for bladder, uterus and posterior compartment are provided in Table 3. All clinical and sonographic measurements were normally or near normally distributed on Kolmogorov Smirnov testing. On univariate analysis, Ba, C and Bp were strongly associated with prolapse symptoms (P <.1 for all on t-test) and with bladder, uterine and rectocele/enterocele descent on ultrasound (r =.752 for Ba vs bladder descent,.52 for C vs uterine descent and.42 for Bp vs rectocele/enterocele descent; P <.1 for all). Figure 2 shows ANOVA graphs of POP-Q coordinates Ba, C and Bp and their association with corresponding ultrasound findings (P <.1 for all), and Figure 3 provides dot plots for the three POP-Q coordinates showing regression Table 1 Characteristics of 825 women with symptoms of lower urinary tract and pelvic floor dysfunction Characteristic Value Age (years) 56 ± 13.5 (18 88) Menopausal 59 (61.7) Body mass index (kg/m 2 ) 29± 6.2 (15 55) Vaginally parous 74 (89.7) Vaginal parity 2 ( 9) Age at first vaginal birth (years) 24 ± 5.1 (15 42) Instrumental birth 224 (27.2) Previous hysterectomy 257 (31.2) Previous incontinence/prolapse surgery 181 (21.9) Presenting symptoms Stress urinary incontinence 611 (74.1) Urge urinary incontinence 62 (73.) Urinary frequency 276 (33.5) Nocturia 385 (46.7) Symptoms of voiding dysfunction 289 (35.) Symptoms of obstructed defecation 428 (51.9) Data are given as mean ± SD (range), median (range) or n (%). Table 2 Pelvic organ prolapse assessment (POP-Q) findings on clinical examination of 825 women with symptoms of lower urinary tract and pelvic floor dysfunction POP-Q coordinate Mean ± SD Range Ba.7 ± to5 C 4.4 ± 3. 9 to8 Bp 1. ± 1. 3 to5 Gh 4.1 ± to 8.5 Pb 3.8 ± to 7. Gh + Pb 7.9 ± to 12.5 Ba, maximum downward displacement of anterior vaginal wall; Bp, maximum downward displacement of posterior vaginal wall; C, maximum downward displacement of cervix; Gh, length of genital hiatus; Pb, length of perineal body. Table 3 Measures of pelvic organ descent on translabial ultrasound in 825 women with symptoms of lower urinary tract and pelvic floor dysfunction Parameter Mean ± SD Range Bladder descent (mm) 6.2 ± to 38.2 Uterine descent (mm) (n = 568) 13.8 ± to 48.4 Rectocele/enterocele descent (mm) 8.7 ± to 38. Measurements are relative to inferior symphyseal margin, with negative values implying descent below the symphysis pubis. line, 95% CIs and 95% prediction intervals. Bladder, uterine and rectocele/enterocele descent on ultrasound were also strongly associated with prolapse symptoms (P <.1, P <.1 and P =.1, respectively, on t-test). All associations between POP-Q and ultrasound measurements remained highly significant when controlling for potential confounders on multivariate analysis, such as age, parity, menopausal status, BMI, history of instrumental delivery, hysterectomy, incontinence or prolapse surgery and levator avulsion. Agreement between sonographic and clinical definitions of significant prolapse was best for the anterior compartment (κ,.51) but poorer for the uterus and posterior compartment (κ,.37 and.28, respectively; Table 4). Copyright 215 ISUOG. Published by John Wiley & Sons Ltd. Ultrasound Obstet Gynecol 216; 47:

4 366 Dietz et al. (a) Ba on ICS POP-Q (b) 5 C on ICS POP-Q (c) Bladder descent on US (mm) Uterine descent on US (mm) Rectocele/enterocele descent on US (mm) Bp on ICS POP-Q Figure 2 Analysis of variance graphs showing association between pelvic organ descent on translabial ultrasound (US) and coordinates: (a) Ba (maximum downward displacement of anterior vaginal wall), (b) C (maximum downward displacement of cervix) and (c) Bp (maximum downward displacement of posterior vaginal wall), on International Continence Society pelvic organ prolapse assessment (ICS POP-Q) (n = 825). Mean and 95% CIs of pelvic organ descent on US are plotted (P <.1). (a) Bladder descent on US (mm) (b) Uterine descent on US (mm) Ba on ICS POP-Q (c) Rectocele/enterocele descent on US (mm) C on ICS POP-Q Bp on ICS POP-Q Figure 3 Regression plots ( ) showing the association between coordinates on International Continence Society pelvic organ prolapse assessment (ICS POP-Q) and translabial ultrasound (US) findings for diagnosis of significant pelvic organ prolapse: (a) Ba (maximum downward displacement of anterior vagina wall) and bladder descent (R 2 =.57); (b) C (maximum downward displacement of cervix) and uterine descent (R 2 =.27); (c) Bp (maximum downward displacement of posterior vaginal wall) and rectocele/enterocele descent (R 2 =.16). All relationships P <.1. 95% CI ( ) and 95% prediction intervals ( ) are shown. DISCUSSION In this large observational study, we were able to show a strong association between POP-Q and ultrasound findings of prolapse, and both were strongly associated with symptoms of prolapse ascertained on direct questioning. This is consistent with data obtained by others 14 over part of the scale of measurement and confirms that ultrasound assessment of prolapse is possible over the entire scale, i.e. not just for prolapse to the hymen, as claimed previously. On comparing equivalent measurements for ultrasound and clinical examination, it is apparent that previously proposed cut-offs for significant prolapse on ultrasound and POP-Q are mutually compatible and consistent (Figure 2). Although there is a good correlation between ultrasound and clinical measurements, the agreement between the two methods for diagnosing genital prolapse was moderate to poor and, as such, they should not be used interchangeably, especially for the central and posterior compartments (Table 4). This is not surprising, as ultrasound measurements are obtained using the symphysis pubis as a point of reference while clinical measurements use the hymen. This may introduce increasing variability at greater distances from the symphysis pubis. However, even if reference lines were positioned identically, one would still not expect perfect agreement for anterior and posterior compartments, as the ICS POP-Q describes surface anatomy, while sonographic prolapse assessment documents organ descent. In this context, it should also be pointed out that the two systems, for historical reasons, use different scales. On ultrasound, prolapse below the symphysis pubis is given a negative reading, while the ICS POP-Q provides positive measurements. Hence, bladder descent to 2.7 cm below the symphysis pubis ( 2.7 cm) is equivalent to a Ba of +2. For the anterior compartment, the recently published cut-off of Ba.5 4 is equivalent to a bladder position of approximately 1 mm below the symphysis pubis 11, and the association between measurements is very strong and almost linear. For the central compartment, the association is less strong than for anterior and posterior compartments and is not quite linear, but published cut-offs of C 5 and a uterine position of 15 mm above the symphysis pubis 12 are again mutually consistent, even if somewhat counterintuitive, given that a C of 5 would commonly be regarded as normal. For the posterior compartment, the association is stronger, even if not quite linear, with higher SDs at greater degrees of prolapse. Again, previously published clinical 4 and sonographic 11 cut-offs Copyright 215 ISUOG. Published by John Wiley & Sons Ltd. Ultrasound Obstet Gynecol 216; 47:

5 Comparison of POP-Q and ultrasound findings 367 Table 4 Agreement between sonographic and clinical definitions of significant prolapse based on findings from pelvic organ prolapse assessment (POP-Q) and ultrasound examination (US) in 825 women with symptoms of lower urinary tract and pelvic floor dysfunction Significant POP on US Parameter No Yes Agreement (%) Cohen s kappa (95% CI) PPV (%) NPV (%) Clinically significant POP No (.37.49) Yes Specificity 64% Sensitivity 9% Anterior compartment POP No (.46.55) Yes Specificity 64% Sensitivity 9% Uterine descent No (.29.45) Yes Specificity 77% Sensitivity 6% Posterior compartment POP No (.21.33) 49 8 Yes Specificity 47% Sensitivity 93% NPV, negative predictive value; PPV, positive predictive value. for significant prolapse (Bp.5 and rectocele 15 mm below the symphysis pubis) are highly consistent. There are several weaknesses of this study that have to be mentioned. It was a retrospective study on a dataset of consecutively assessed urogynecological patients, the majority of whom were of Caucasian ethnicity. This implies that our conclusions are limited to similar populations. It is likely that different results will be obtained in other ethnic groups or in the general population. It may also be argued that the consistency of findings is not surprising, given the partial overlap in the populations between this current dataset and those of two studies providing receiver operating characteristics curve statistics and cut-offs for clinical and ultrasonographic prolapse. However, all ultrasound measurements were obtained from archived volume datasets, with the operator blinded to all clinical data, and all ICS POP-Q measurements were obtained blinded to clinical symptoms. Hence, the introduction of significant bias appears unlikely. In addition, sonographic cut-offs for anterior and posterior compartment descent were obtained in a different population with very similar characteristics, assessed several years previously 11. Finally, the Valsalva maneuver utilized for all assessments was standardized for duration only (maneuver lasting 6 s) and not for pressure, owing to the substantial technical effort required, and assessments were performed by a large number of junior staff, which adds an element of variability. Irrespective of these potential limitations, however, all tests, both the clinical examination and the sonographic volume data acquisition, were performed using a method supported by more than 15 years of clinical research. We have been able to show that standardization of Valsalva duration is more important than standardization of pressure 16,17, and the strong associations observed between symptoms and signs, both clinical and sonographic, appear to validate our methodology for the assessment of prolapse. Together with the large dataset, this should provide a high degree of consistency. While there have been attempts to compare clinical findings with the results of fluoroscopic and magnetic resonance imaging 18,19, yielding much poorer agreement between methods and poorer associations with symptoms, ultrasound is likely to be superior to radiological methods owing to the simplicity of the assessment, its low cost and its lack of exposure to radiation. Our results may be of considerable clinical utility. False-negative results are likely to be quite common in prolapse assessment, given that some authors report a poor correlation between symptoms and signs 2 and that discrepancies between clinical findings and prolapse detected on examination under anesthesia are not uncommon. Since ultrasound imaging is increasingly used for the assessment of pelvic organ descent, our data will provide others with the opportunity to check clinical and sonographic findings for congruity, serving to avoid false-negative assessment results. Finally, the data presented in this study provide further support for the contention that the current ICS POP-Q staging is in need of revision, as suggested recently 4. Cut-offs suggested for the diagnosis of significant prolapse, i.e. prolapse that is likely to be symptomatic, seem to be mutually consistent and plausible, as proposed sonographic cut-offs 11,12 are shown to be identical or very similar to proposed clinical cut-offs 4 in the data presented here. Ba and Bp values of 2 and 1.5 (corresponding to Stage I anterior and posterior compartment prolapse ) should be regarded as normal. This is consistent with data obtained by others, using clinical examination 5.On the other hand, a cut-off of or lower, as suggested by Gutman et al. 21, seems too distal. This discrepancy may be explained by the fact that this study predicted the symptom of a visible or palpable bulge, not the sensation of a lump or bulge used in our study. In addition, Gutman et al. did not consider compartments separately, and there is no mention of central compartment prolapse 21. Findings regarding central compartment prolapse are more likely to be contentious, given that a cut-off of 5, as published recently, seems excessively cranial. In Copyright 215 ISUOG. Published by John Wiley & Sons Ltd. Ultrasound Obstet Gynecol 216; 47:

6 368 Dietz et al. our data, the previously established cut-off of 15 mm above the symphysis pubis for significant uterine descent would be consistent with a C value of This is closer to the generally accepted clinical cut-off of 4, and argues for the retention of this value. Whether vaginal length should be considered part of the definition of normal, as suggested by the original ICS POP-Q definition 3, is a matter that will require further investigation, and such a study is currently in progress at our unit. In conclusion, in this large retrospective observational study we have been able to show that there is a strong near linear association between POP-Q values and sonographic measures of prolapse. Previously proposed cut-offs for significant prolapse on ultrasound and POP-Q (Ba.5/cystocele 1 mm below the symphysis pubis, C 5/uterine descent of 15 mm above the symphysis pubis and Bp.5/rectocele 15 mm below the symphysis pubis) are mutually consistent and plausible. DISCLOSURE H.P.D. received unrestricted educational grants from GE Medical Systems. REFERENCES 1. Smith F, Holman D, Moorin R, Tsokos N. Lifetime risk of undergoing surgery for pelvic organ prolapse. Obstet Gynecol 21; 116: Olsen AL, Smith VJ, Bergstrom JO, Colling JC, Clark AL. Epidemiology of surgically managed pelvic organ prolapse and urinary incontinence. Obstet Gynecol 1997; 89: Bump RC, Mattiasson A, Bo K, Brubaker LP, DeLancey JO, Klarskov P, Shull BL, Smith AR. The standardization of terminology of female pelvic organ prolapse and pelvic floor dysfunction. Am J Obstet Gynecol 1996; 175: Dietz HP, Mann KP. What is clinically relevant prolapse? An attempt at defining cutoffs for the clinical assessment of pelvic organ descent. Int Urogynecol J 214; 25: Swift S, Woodman P, O Boyle A, Kahn M, Valley M, Bland D, Wang W, Schaffer J. Pelvic organ support study (POSST): the distribution, clinical definition, and epidemiologic condition of pelvic organ support defects. Am J Obstet Gynecol 25; 192: Dietz HP, Hansell NK, Grace ME, Eldridge AM, Clarke B, Martin NG. Bladder neck mobility is a heritable trait. BJOG 25; 112: Dietz HP. The aetiology of prolapse. Int Urogynecol J Pelvic Floor Dysfunct 28; 19: Dietz H, Lanzarone V. Levator trauma after vaginal delivery. Obstet Gynecol 25; 16: Dietz H. Prolapse worsens with age, doesn t it? Aust NZ J Obstet Gynaecol 28; 48: Dietz HP, Eldridge A, Grace M, Clarke B. Pelvic organ descent in young nulligravid women. Am J Obstet Gynecol 24; 191: Dietz HP, Lekskulchai O. Ultrasound assessment of prolapse: the relationship between prolapse severity and symptoms. Ultrasound Obstet Gynecol 27; 29: Shek K, Dietz H. What is significant descent of the uterus on pelvic floor ultrasound? Ultrasound Obstet Gynecol 214; 44(S1): Dietz HP, Haylen BT, Broome J. Ultrasound in the quantification of female pelvic organ prolapse. Ultrasound Obstet Gynecol 21; 18: Lone FW, Thakar R, Sultan AH, Stankiewicz A. Accuracy of assessing Pelvic Organ Prolapse Quantification points using dynamic 2D transperineal ultrasound in women with pelvic organ prolapse. Int Urogynecol J 212; 23: Dietz H. Ultrasound imaging of the pelvic floor. Part 1: two-dimensional aspects. Ultrasound Obstet Gynecol 24; 23: Orejuela F, Shek K, Dietz H. The time factor in the assessment of prolapse and levator ballooning. Int Urogynecol J 212; 23: Mulder F, Shek K, Dietz H. The pressure factor in the assessment of pelvic organ mobility. Aust NZ J Obstet Gynaecol 212; 52: Healy JC, Halligan S, Reznek RH, Watson S, Phillips RK, Armstrong P. Patterns of prolapse in women with symptoms of pelvic floor weakness: assessment with MR imaging. Radiology 1997; 23: Kenton K, Shott S, Brubaker L. Vaginal topography does not correlate well with visceral position in women with pelvic organ prolapse. Int Urogynecol J Pelvic Floor Dysfunct 1997; 8: Ghetti C, Gregory WT, Edwards SR, Otto LN, Clark AL. Pelvic organ descent and symptoms of pelvic floor disorders. Am J Obstet Gynecol 25; 193: Gutman RE, Ford DE, Quiroz LH, Shippey SH, Handa VL. Is there a pelvic organ prolapse threshold that predicts pelvic floor symptoms? Am J Obstet Gynecol 28; 199: 683.e1 7. Copyright 215 ISUOG. Published by John Wiley & Sons Ltd. Ultrasound Obstet Gynecol 216; 47: