| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Gastrointestinal Imaging |
1 From the Departments of Radiology (A.C.D., M.P.T., J.S.), Surgery (J.F.M.S.), and Clinical Epidemiology and Biostatistics (M.D., P.M.M.B.), Academic Medical Center, Meibergdreef 9, 1105 AZ Amsterdam, the Netherlands; Department of Surgery, Onze Lieve Vrouwe Gasthuis, Amsterdam, the Netherlands (M.F.G.); and Department of Radiology, University Hospital Maastricht, Maastricht, the Netherlands (R.G.H.B.). Received September 21, 2005; revision requested November 14; revision received January 31, 2006; accepted March 3; final version accepted May 3. Supported by the Netherlands Organization for Health Research and Development ZON MW (grant 945-01-013). Address correspondence to A.C.D. (e-mail: a.c.dobben{at}amc.uva.nl).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCES IN KNOWLEDGEReferences |
|---|
Materials and Methods: The study was approved by the medical ethics committee of all participating centers. A total of 237 consenting patients (214 women, 23 men; mean age, 58.6 years ± 13 [standard deviation]) with fecal incontinence were examined from 13 different hospitals by using endoanal MR imaging and endoanal US. Patients with an anterior EAS defect depicted on endoanal MR images and/or endoanal US scans underwent anal sphincter repair. Surgical findings were used as the reference standard in the determination of anterior EAS defects. The Cohen 
statistic and McNemar test were used to calculate agreement and differences between diagnostic techniques.
Results: Agreement between endoanal MR imaging and endoanal US was fair for the depiction of sphincter defects ( = 0.24 [95% confidence interval: 0.12, 0.36]). At surgery, EAS defects were found in 31 (86%) of 36 patients. There was no significant difference between MR imaging and US in the depiction of sphincter defects (P = .23). Sensitivity and positive predictive value were 81% and 89%, respectively, for endoanal MR imaging and 90% and 85%, respectively, for endoanal US.
Conclusion: In the selection of patients for anal sphincter repair, both endoanal MR imaging and endoanal US are sensitive tools for preoperative assessment, and both techniques can be used to depict surgically repairable anterior EAS defects.
© RSNA, 2007
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCES IN KNOWLEDGEReferences |
|---|
Currently, endoanal US is the preferred diagnostic technique to select patients for surgery (3,4). The advantages of endoanal US are its availability and limited cost, as well as the fact that radiologists typically have more experience with this modality (5). In contrast, endoanal MR imaging may allow clear visualization of the external anal sphincter (EAS) because there is a large difference in contrast between the EAS muscle and the surrounding fat, and endoanal MR imaging is capable of demonstrating EAS atrophy (5).
In previous studies of these imaging techniques, researchers have concluded that these two techniques should be considered comparable in the selection of patients for surgery (3,6–13), but all of these studies were single-center studies. Thus, the purpose of our study was to prospectively compare in a multicenter study the agreement between endoanal MR imaging and endoanal US in depicting EAS defects in patients with fecal incontinence.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCES IN KNOWLEDGEReferences |
|---|
Between December 2001 and May 2005, consecutive patients with fecal incontinence were included from 13 medical centers in the Netherlands. Details of the study design will be reported elsewhere.
All eligible patients were referred for standardized specialized pelvic floor rehabilitation after standardized diagnostic work-up, including imaging. If pelvic floor rehabilitation failed, overlapping anterior anal sphincter repair was considered as the next available treatment option for patients with an EAS defect. An EAS defect was defined as a solitary EAS defect that comprised 30° of the circumference of the sphincteric ring and that was detected at endoanal MR imaging and/or endoanal US. Patients with severe generalized EAS atrophy, which was defined as extensive thinning of the EAS muscle or diffuse replacement of EAS muscle by fat (14) at endoanal MR imaging, were excluded from overlapping anterior anal sphincter repair.
Clinical Assessment
One of the 13 participating clinicians evaluated the severity of symptoms and obtained a detailed medical history. The severity of fecal incontinence was assessed according to the grading system of Vaizey et al (15).
Imaging Tests
Because imaging modalities were not available at all of the 13 participating centers, both endoanal US and endoanal MR imaging were performed in seven centers. Consequently, certain patients underwent their examinations at a center other than the one from which they originated. All endoanal US examinations, except one, were performed by six clinicians with experience ranging from 10 to 14 years. At one center, endoanal US was performed by a technician with 10 years of experience. All endoanal MR imaging examinations were performed by technicians with experience ranging from 1 to 5 years. For logistic reasons, endoanal US was performed prior to endoanal MR imaging.
Endoanal US.—Endoanal US was performed by using a US scanner (model 3535, Bruel and Kjaer, Gentofte, Denmark; or Multiview, Aloka, Tokyo, Japan) with a radial endoscopic probe (7.5- or 10.0-MHz transducer) and a sonolucent plastic cone. For this procedure, the patients lay in the left lateral position with their knees bent at a 90° angle (2,11,16). The endoscopic probe was introduced into the anal canal, positioned at the upper aspect of the puborectalis sling, and slowly withdrawn until all levels, perpendicular to the anal canal, were scanned.
Endoanal MR imaging.—Endoanal MR imaging was performed by using a 1.0- or 1.5-T MR unit (Gyroscan ACS-NT, Philips Medical Systems, Best, the Netherlands; or Horizon Echospeed, GE Medical Systems, Milwaukee, Wis) with a dedicated endoanal coil that had a diameter of 19 mm (11,14,17). All patients were asked to fast for 4 hours prior to MR imaging to reduce artifacts from bowel peristalsis. Bowel relaxants—either 1 mL of butylscopolamine bromide (20 mg/mL, Buscopan; Boehringer, Ingelheim, Germany) or 1 mg of glucagon hydrochloride (Glucagen; Novo Nordisk, Bagsvaerd, Denmark)—were used at one of the institutions. The endoanal coil was covered with a condom and, after the application of lubricant, was inserted in the anal canal in a left lateral position. After positioning of the endoanal coil, the patients turned in the supine position, and supportive pads were used to stabilize the coil.
Scanning parameters were optimized for the MR imaging machines on the basis of extensive previous experience. The following T2-weighted fast spin-echo sequences were used according to the standardized imaging protocol that was established during joint meetings: 2500–3500/70–90 (repetition time msec/echo time msec), echo train length of 10, field of view of 10 x 10 cm (transverse) and 16 x 16 cm (coronal), imaging matrix of 256 x 512, section thickness of 3 mm, intersection gap of 0.3 mm, and two signals acquired. Transverse and coronal images with a section orientation perpendicular for the transverse images and parallel for the coronal images to the anal sphincter and endoanal coil were obtained.
Image analysis.—Images were analyzed separate from the imaging session. Endoanal US images were analyzed with a personal computer to capture the series of the endoanal US images. Endoanal MR image analysis was performed by using workstation viewing software (IMPAX SP4 SU4 DS3000, Agfa, Mortsel, Belgium, or Easy Vision Workstation, Philips Medical Systems).
An EAS defect at endoanal US was defined as a discontinuity of the muscle ring (anatomic defect) and/or was characterized by a loss of normal architecture, with an area of amorphous texture usually of low reflectiveness (functional defect or scar tissue) (4). An EAS defect at endoanal MR imaging was defined as a discontinuity of the muscle ring (anatomic defect) and/or was recognized by identifying a hypointense deformation of the normal pattern of the muscle layer owing to replacement of muscle cells by fibrous tissue (functional defect or scar tissue) (14).
The endoanal US images were assigned scores by one of six observers at the seven centers where imaging was performed. These observers included two gastroenterologists and four surgeons, all of whom were experts in the field and had a considerable amount of experience (10–14 years) in reading endoanal US images.
The endoanal MR images were assigned scores by one of three observers (including R.G.H.B. and J.S.) at the participating centers. All observers who assigned scores to the endoanal images were radiologists who had 8–12 years of experience in evaluating abdominal MR images.
Both endoanal US and endoanal MR images were evaluated separately. Observers were blinded to the findings of the other technique and to the medical history of the patients, except for age, sex, and the presence of fecal incontinence.
Anterior Anal Repair
The decision to perform surgery was made by a participating surgeon (J.F.M.S. or M.F.G.) on the basis of imaging findings (eg, the extent of the EAS lesion at endoanal imaging and/or the degree of sphincter atrophy), complementary clinical information (eg, the severity of fecal incontinence and the willingness of the patient to undergo surgery), and findings from anorectal physiologic testing.
Overlapping anterior anal sphincter repair was performed as previously described (18,19) at eight participating centers by one of eight experienced colorectal surgeons (6–25 years of experience). Surgical findings were recorded and used as the reference standard.
Statistical Analysis
Patient groups were compared with respect to their characteristics by using an analysis of variance and the 
2 test.
The depiction of EAS defects at endoanal MR imaging was compared with the depiction of EAS defects at endoanal US. To calculate the level of agreement between the two diagnostic techniques, we used the Cohen statistic with 95% confidence intervals (CIs). Agreement was classified as poor (
0.20), fair (0.21–0.40), moderate (0.41–0.60), good (0.61–0.80), or very good (>0.80) (20).
Findings of EAS defects, as determined at surgery, were compared with findings from endoanal MR imaging and endoanal US. To assess whether significant differences existed between diagnostic techniques, the McNemar test was used. Sensitivities and positive predictive values with 95% CIs were calculated for the depiction of EAS defects, with surgical results used as the reference standard and the imaging techniques used as the index tests.
For all statistical tests, P values of less than .05 were considered to indicate a significant difference. Software (SPSS for Windows, version 11.5; SPSS, Chicago, Ill) was used to perform statistical analysis.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCES IN KNOWLEDGEReferences |
|---|
|
|
= 0.24 [95% CI: 0.12, 0.36]) (Table 3). Most defects were assigned scores for location on the anterior or anterior-lateral side of the EAS by using either technique (104 [96%] of 108 patients at endoanal MR imaging and 130 [95%] of 137 patients at endoanal US).
|
|
The mean interval between imaging and surgery was 10 months (range, 7–17 months). Data were collected from 36 patients who underwent anterior anal repair, 34 (94%) of whom were women. The mean age of these 36 patients was 51 years ± 12.5. Clinical characteristics are summarized in Table 4.
|
|
|
|
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCES IN KNOWLEDGEReferences |
|---|
Our study results show that the agreement between endoanal MR imaging and endoanal US is fair for the depiction of EAS defects. Despite the fact that all observers can be classified as experienced, the variety in evaluation of EAS defects was substantial. In earlier studies, researchers demonstrated that both imaging techniques were accurate in mapping defects of the EAS (14,21–26). In one study, deSouza and coauthors (22) concluded that, in seven patients, endoanal MR imaging correctly diagnosed sphincter tears, all of which were validated at surgery. Others showed that, by confirming the defect at surgery, endoanal US accurately depicted sphincter lesions in patients with fecal incontinence (24–26). Unfortunately, those studies consisted of small patient groups and compared only one imaging technique with surgery.
There are two comparative studies published on endoanal MR imaging versus endoanal US for the depiction of EAS defects in a population of patients with fecal incontinence (3,10). A prospective study of 52 patients with fecal incontinence found complete agreement between endoanal MR imaging, endoanal US, and final diagnosis in 62% of patients (10). A retrospective study of 22 patients with fecal incontinence demonstrated fair agreement for the diagnosis of EAS damage between endoanal MR imaging and endoanal US ( = 0.38) (3). This is concordant with the findings of our prospective comparative study.
The reliability of our findings might have been influenced by the variety between various observers from different centers. However, despite the fact that the two cited comparative studies are single-center studies, the results of our multicenter study do not differ substantially. We also found fair agreement ( = 0.24) between both imaging techniques in a large cohort of patients with fecal incontinence. Furthermore, the results of our multicenter study are a better reflection of daily clinical practice and are therefore more applicable to external validity.
A number of potential limitations of our study should be addressed. A major limitation is partial verification bias, because we do not know the surgical findings in the nonsurgical patient group. This may lead to an overestimation of sensitivity and an underestimation of specificity. Therefore, potential true- or false-negative findings cannot be calculated.
Also, in our study, only endoluminal imaging of EAS lesions was evaluated. These results were compared with findings at surgical anterior anal repair. An anterior EAS defect can generally be considered as a surgically remedial tear. The EAS defects in our study population were mainly located at the anterior side of the sphincter complex. This confirmed our expectations, because the majority of our cohort consisted of women with one or more obstetric risk factors (3). We did not include internal anal sphincter findings in our comparative study. As for isolated internal anal sphincter damage, there is no surgical option available (10) except for injectable silicone biomaterial implants (a new experimental therapy which is still under investigation) (27).
In the selection of candidates for surgery, previous study results have shown that endoanal MR imaging, contrary to endoanal US, is an accurate diagnostic technique for the depiction of EAS atrophy (7,16,28–33). The accurate demonstration of EAS at endoanal MR imaging, especially of its borders and fat content, facilitates the evaluation of atrophy. EAS atrophy is characterized by generalized thinning of the muscle fibers and/or by fatty replacement (30). EAS atrophy negatively affects continence after anterior anal repair (28,34). Although anterior anal repair seems to confer substantial benefits on these patients (35), short-term results vary and are contradictory in the literature (9,36–40).
To prevent unnecessary surgery, endoanal MR imaging seems to be a useful diagnostic technique for preoperative assessment. Unfortunately, our study results show that, in 14 patients, endoanal MR imaging depicted an EAS defect accompanied by EAS atrophy that was confirmed at surgery in only two patients. This finding implies that exclusion of patients on the basis of atrophy at MR imaging may not have been reasonable. However, because surgery may allow determination of generalized sphincter thinning rather than fatty infiltration, histologic analysis is needed to confirm the latter. Thus, histologic analysis is the reference standard for EAS atrophy. In our study, histologic analysis was not performed. Consequently, patients with EAS atrophy characterized by fatty replacement could not be assessed at surgery. Therefore, we can hypothesize that, in reality, more patients than recorded are affected by EAS atrophy, as was suggested at MR imaging.
We demonstrated fair agreement between endoanal MR imaging and endoanal US in a large cohort of patients with fecal incontinence. We were able to validate our findings in only a subgroup of patients who underwent surgery. The selection of this subgroup was based on diagnostic imaging. We do not know to what extent a certain preference for one of the imaging modalities might have played a role in decision making by the clinician. Endoanal US is widely available, contrary to the limited availability of endoanal MR imaging. The use of the latter has been restricted to specialized centers, because the required endoanal coil is not yet available with every MR machine (6). Therefore, it is possible that experience with one technique influenced the selection process for surgery.
When we validated our results in a small cohort of patients and compared these results with surgical findings, we concluded that both imaging techniques can be considered useful in the selection of patients for surgery. Endoanal MR imaging is capable of depicting EAS atrophy, which is associated with a poor outcome of anterior anal repair. The technique of choice in clinical decision making may depend on the infrastructure of the center.
|
ADVANCES IN KNOWLEDGE |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCES IN KNOWLEDGEReferences |
|---|
|
FOOTNOTES |
|---|
Abbreviations: CI = confidence interval • EAS = external anal sphincter
Authors stated no financial relationship to disclose.
Author contributions: Guarantors of integrity of entire study, A.C.D., P.M.M.B., J.S.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; manuscript final version approval, all authors; literature research, A.C.D., M.P.T.; clinical studies, A.C.D., M.P.T.; statistical analysis, A.C.D., P.M.M.B.; and manuscript editing, all authors
|
References |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCES IN KNOWLEDGEReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  A. C. Dobben, R. J. F. Felt-Bersma, F. J. W. ten Kate, and J. Stoker Cross-Sectional Imaging of the Anal Sphincter in Fecal Incontinence Am. J. Roentgenol., March 1, 2008; 190(3): 671 - 682. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Berton, G. Gola, and S. R. Wilson Sonography of Benign Conditions of the Anal Canal: An Update Am. J. Roentgenol., October 1, 2007; 189(4): 765 - 773. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. C. Dobben, M. P. Terra, M. Deutekom, J. F. M. Slors, L. W. M. Janssen, P. M. M. Bossuyt, and J. Stoker The Role of Endoluminal Imaging in Clinical Outcome of Overlapping Anterior Anal Sphincter Repair in Patients with Fecal Incontinence Am. J. Roentgenol., August 1, 2007; 189(2): W70 - W77. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
