| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research |
1 Institute of Diagnostic Radiology, University Hospital Zürich,
Rämistrasse 100, CH-8091 Zürich, Switzerland.
2 Visceral Surgery, Kantonspital St. Gallen, St. Gallen, Switzerland.
Received July 20, 2007;
accepted after revision September 9, 2007.
Address correspondence to D. Weishaupt
(dominik.weishaupt{at}usz.ch).
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. Twenty patients underwent MR defecography twice on a 0.5-T open-configuration system in the sitting position. During the first imaging session, MR defecography was performed with a rectal enema consisting of potato starch mixed with gadolinium (PS group). During the second session, the enema consisted of ultrasound gel mixed with gadolinium (US group). The imaging protocol consisted of midsagittal T1-weighted gradient-recalled echo MR images obtained at rest, at maximal sphincter contraction, at straining, and during defecation. All images were analyzed quantitatively by measuring the contrast-to-noise ratio (CNR) and reviewed by three independent observers with regard to the visibility of pelvic floor abnormalities and the extent of those abnormalities.
RESULTS. The CNR values in the PS group (mean ± SD, 167.49 ± 44.4) were significantly higher than those obtained in the US group (150.2 ± 37.8) (p < 0.05). The visibility scores for anterior rectoceles and intussusceptions were higher in the PS group than in the US group (mean visibility scores: PS group, 2.8 ± 0.42 and 2.6 ± 0.56, respectively; US group, 2.3 ± 0.77 and 2.2 ± 0.74, respectively). The size and the number of incompletely emptying anterior rectoceles were higher in the PS group.
CONCLUSION. Ultrasound gel and potato starch provide good contrast and depiction of relevant pelvic floor abnormalities. However, the visibility of pelvic floor abnormalities and extent of those abnormalities depend on the composition of the rectal enema. In particular, the size and degree of anterior rectocele evacuation and intussusception size are often underestimated when ultrasound gel is used for rectal enema.
Keywords: contrast media • enema • MR defecography • MRI • potato starch enema • ultrasound gel enema
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Although MR defecography is being established as an alternative to conventional defecography (evacuation proctography) in many institutions, there is no uniform agreement about the technical details of the examination [3–6]. Consensus seems to exist that the examination should include MRI at different pelvic positions (at rest, at squeezing, at straining, and during evacuation) and that a rectal enema is necessary. However, examiners do not uniformly agree about the composition of the rectal enema. Some authorities use ultrasound gel for rectal enemas [7, 8], whereas others have proposed a more viscous agent such as potato starch for this purpose [9, 10].
Over the past few years, we have used both ultrasound gel and potato starch as agents for rectal enemas. However, when performing the examination with rectal enemas consisting of ultrasound gel, we got the impression that some abnormalities on the pelvic floor—in particular, with regard to the posterior compartment—may be underestimated. To our knowledge, no study assessing different contrast enemas has been published to date.
Hence, the purpose of this study was to compare intraindividually two rectal enema compositions for MR defecography.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Twenty consecutive patients (three men and 17 women; median age, 55 years; age range, 27–77 years) were included in this study. The patients were recruited from clinical referrals for MR defecography. The indication for MR defecography was an outlet obstruction in 11 patients (55%) and fecal incontinence in nine patients (45%). Twelve (71%) of the 17 women had undergone hysterectomy. Another 12 patients (60%) had a history of pelvic surgery including rectopexy in four patients (20%), sigmoid resection in another four patients (20%), and cystourethropexy in two patients (10%).
All patients included in the study agreed to undergo MR defecography on two separate occasions with an interval not to exceed 3 weeks (mean, 8.3 ± 3.05 days; range, 3–15 days) between imaging examinations. No change in treatment was made and no surgery was performed between the two imaging sessions.
MRI
All patients underwent MR defecography twice in a superconducting
open-configuration 0.5-T MRI system (Signa SP, GE Healthcare) while in the
sitting position, according to the technique described by Schoenenberger et
al. [11]. The same technique
for MR defecography was performed at both imaging sessions with the only
difference being the composition of the enema. MR examinations were performed
in all patients without bowel preparation and with a comfortably full
bladder.
Before the first MRI session, the rectum of 10 patients was filled with 300 mL of a synthetic stool consisting of potato starch (Stocki, Knorr) mixed with 1.5 mL of gadopentetate dimeglumine (377 mg/mL) (Magnevist, Bayer HealthCare), producing a gadolinium concentration of 2.5 mmol/L. We used gadopentetate dimeglumine in combination with the enema because we performed T1-weighted gradient-recalled echo sequences to obtain images of the different pelvic positions.
During the second MRI session, the rectum of the same 10 patients was filled with 300 mL of ultrasound gel (Kendall, Tyco Healthcare) combined with the same amount of gadopentetate dimeglumine as for the enema consisting of potato starch. In the remaining 10 patients, the rectal enema during the first imaging session consisted of ultrasound gel, and the enema for the second imaging session consisted of potato starch. The potato starch used for the study was a convenience food product.
For the purpose of the study, 125 g of potato starch powder was mixed with 200 mL of water. All enemas were administered via a rectal tube with the patient on the MR table. After the rectum had been filled, the patient was placed upright on a wooden chair that fits into the magnet rings. A flexible transmit–receive radiofrequency coil was wrapped around the patient's pelvis.
For reporting the results, MR defecography examinations performed with a rectal enema consisting of potato starch combined with gadolinium are reported as the potato starch group (PS group), whereas MR defecography performed with ultrasound gel combined with gadolinium are reported as the ultrasound gel group (US group). On the basis of axial localizing images in the axial, coronal, and sagittal planes, a multiphase fast T1-weighted spoiled gradient-recalled echo sequence was performed in the midsagittal plane of the anal canal with an image update every 2 seconds. To keep the imaging plane, patients were instructed not to move on the table. The imaging parameters for this sequence were as follows: TR range/TE range, 22.2–22.4/10.6–10.7; flip angle, 90°; section thickness, 1.5 cm with no interslice gap; bandwidth, 12.5 kHz; field of view, 29–32 cm; image matrix, 256 x 160; and 1 signal acquired. Images of the pelvis were obtained using this sequence with the patient at rest, at maximal voluntary sphincter and pelvic floor muscle contraction (squeezing), at straining, and during evacuation. The acquisition time varied from 2.5 to 4 minutes depending on evacuation ability and collaboration of the patient.
If an intussusception (internal rectal prolapse) or a lateral rectocele was suspected on the basis of findings on the midsagittal scans, additional gradient-recalled echo images using the same sequence parameters were acquired in the transaxial plane. To obtain images at the different pelvic restlike positions—that is, maximal voluntary sphincter and pelvic floor muscle contraction (squeezing), at straining, and during evacuation, the technician performing the examination coached patients through a microphone and a headset. All images acquired at the different positions were formatted into a cine loop presentation to enable assessment of the dynamics of rectal emptying and of pelvic floor movement. The overall imaging time for MR defecography including patient preparation varied between 20 and 25 minutes.
Viscosity Measurement
In vitro viscosity measurements for both rectal enemas were performed
according to standard methods using a rotational viscosimeter (VT
501/SV2 sensor, Thermo Haake). Prepared samples were maintained at a constant
temperature of 37°C with a water bath. The apparent viscosity, shear
stress, and shear rate were obtained with the software supplied with the
instrument. All reported viscosities were measured at shear rates of
0–500 s–1 in which the samples show Newtonian
behavior.
Image Analysis
Quantitative image analysis—For quantitative MR data
analysis, signal intensities were measured in regions of interest (ROIs)
positioned in the rectal lumen in stationary tissues adjacent to the rectum
and outside the body. The contrast-to-noise ratios (CNRs) were calculated
using the following equation:
 |
To avoid volume artifacts, the average CNR was calculated using three repeated measurements of signal intensity. The circular ROIs measured 50–70 mm2.
The time of the evacuation phase was measured during both MRI sessions and was defined as the interval between initial pelvic floor descent (evacuation initiation) to completion of rectal emptying in seconds.
Qualitative image analysis—All MR images were separately and independently analyzed by two radiologists with different levels of experience in reading MR defecography (1 year of experience and 10 years of experience) and by a visceral surgeon experienced in proctology with 10 years of experience reading MR defecography images. The readers (readers 1–3) were blinded to all clinical data and to which enema composition had been administered. Randomization was performed with regard to the MR defecography examinations performed with the different enema compositions.
The imaging analysis was performed on a separate workstation (Advantage Workstation, GE Healthcare Europe). MR defecography image interpretations were based on all of the source images and cine loops obtained with the patient at different positions (at rest, at sphincter contraction, at straining, and during evacuation).
Each reader was asked to analyze all MR images with regard to pelvic floor structure abnormalities, including anterior rectoceles, enteroceles, rectal prolapses, rectal descents (descent of the posterior pelvic floor compartment), cystoceles (descent of the anterior pelvic floor compartment), and vaginal vault descents (descent of the middle pelvic floor compartment) or descents of any part of the remaining cervix in patients who had undergone hysterectomy. Observers recorded standard measurements with electronic calipers that were included as part of the standard software delivered with the workstation.
After the observers performed all quantitative measurements on images obtained at each pelvic position in two separate measurements, the mean values for each observer were calculated. When reporting the results of these measurements, we use the mean of all three readers.
As in previous studies, a pubococcygeal line (PCL), defined as a line joining the inferior border of the symphysis pubis to the last coccygeal joint on a midline sagittal image, was drawn on the midline sagittal MR image [12, 13].
The largest measured distance, usually at the end of defecation, was used for further analysis [3, 4]. A cystocele represented an anterior pelvic floor compartment abnormality and was defined as a descent of the bladder base below the PCL. A vaginal vault descent represented a middle pelvic floor compartment abnormality and was defined as a descent of the vaginal vault below the PCL or of any part of the remaining cervix in patients who had undergone hysterectomy below the PCL. Similarly, a rectal descent represented a posterior pelvic floor compartment abnormality and was defined as a descent of the anorectal junction below the PCL [1, 4]. An enterocele was defined as a descent of either the peritoneum containing the small bowel or the sigmoid colon below the PCL. The extent of any cystocele, enterocele, anterior rectocele, or vaginal vault or rectal descent was measured at a 90° angle to the PCL and was graded with a three-category scoring system as small, moderate, or large, as shown in Table 1 [1, 4].
|
An anterior rectocele was defined as a protrusion of the rectal wall anterior to a line extending upward through the anal canal. The size (depth) of an anterior rectocele was identified as the depth of wall protrusion beyond the expected margin of the normal rectal wall and was classified as small, moderate, or large [1, 4] (Table 1). In addition, anterior rectoceles were classified as those with complete defecation and those with incomplete defecation on the basis of whether evacuation of the contrast material was complete or incomplete at the end of defecation.
When readers identified an anterior rectocele or an intussusception, they were asked to grade the visibility of that finding using a 3-point scale: 1, poor visibility; 2, moderate visibility; or 3, good visibility.
Finally, when an intussusception was present, the length and thickness were measured on the midsagittal image during maximal straining or evacuation, depending on the position where the intussusception was obvious. The length of the intussusception was measured as the distance between the point of inversion on the rectal wall and the distal edge of the intussusception. In addition, the thickness of the anterior component and posterior component of the intussusception was measured at the upper point of invagination between the intussuscipiens end and the intussusception contour [14].
Statistical Analysis
Statistical analysis was performed using commercially available software
(SPSS version 11.5, SPSS). The Wilcoxon's signed rank test was used to compare
paired nonparametric data, which are presented as the median and range.
We determined the interobserver agreement between the free independent readers' interpretations by calculating kappa values and 95% CIs. The kappa values were tested for a significant difference from zero. Poor agreement was considered when the kappa value was 0.00; slight agreement, 0.01–0.20; fair agreement, 0.21–0.40; moderate agreement, 0.41–0.60; good agreement, 0.61–0.80; and excellent agreement, 0.81–1.00.
For all tests, a p value of 0.05 or less was considered a statistically significant difference.
|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Viscosity Measurement
The viscosity measurements at different rotation frequencies (D)
showed that the potato starch is significantly more viscous than ultrasound
gel. The potato starch viscosity varied between 18.143 (D = 10.41
s–1) and 0.987 Pa (D = 499.4 s–1)
depending on the rotation frequency. The corresponding viscosity of ultrasound
gel obtained at the same rotation frequencies varied from 10.694 to 0.612 Pa
(p = 0.0025). The differences observed are presented in detail in
Figure 1.
Quantitative imaging analysis—The quantitative analysis of the MR defecography data sets revealed that the CNR values between the rectum and the perirectal tissue in the PS group (mean, 167.49 ± 44.4) were significantly higher than in the US group (mean, 150.2 ± 37.8 seconds).
The duration of the evacuation phase was significantly longer in the PS group than in the US group (158.9 ± 65.5 vs 109.3 ± 39.1 seconds, respectively; p = 0.002).
Qualitative imaging analysis—In both enema groups, each of the three readers identified 16 anterior rectoceles (Table 2). However, all three readers identified more moderate and more large anterior rectoceles in the PS group than in the US group: The number of moderate and large rectoceles in the PS group for readers 1, 2, and 3, respectively, was 11, 12, and 11 and the number of small anterior rectoceles in the PS group was five, four, and five (Figs. 2A, 2B, 3A, 3B, 3C, 4A, 4B, 5A, 5B). In addition, the frequency of anterior rectoceles with incomplete evacuation was greater in the PS group than in the US group; the number of incomplete rectoceles in the PS group for readers 1, 2, and 3 was 11, 11, and 12, respectively, and in the US group eight, nine, and eight, respectively. All three readers judged the visibility of the anterior rectoceles to be better in the PS group than in the US group (mean visibility score: PS group, 2.8 ± 0.42; US group, 2.3 ± 0.77) (Figs. 2A, 2B and 3A, 3B, 3C). These differences were statistically significant for all three readers (p = 0.029–0.007).
|
|
|
|
|
|
|
|
|
|
All three readers identified an intussusception in 12 patients (60%) in the PS group, whereas in the US group, intussusception was seen in only 10 (50%) and 11 (55%) patients. Similar to the visibility of the anterior rectoceles, the visibility of the intussusceptions was rated higher in the PS group than in the US group (2.6 ± 0.56 vs 2.2 ± 0.74, respectively; p = 0.027). In the PS group, the length and thickness of the intussusception were greater than in the US group (PS group: mean length, 19.8 ± 7.17 mm, thickness of anterior component, 7.9 ± 3.2 mm, thickness of posterior component, 5.8 ± 1.6 mm; US group: 16.7 ± 4.84 mm, 6.5 ± 1.92 mm, 5.6 ± 1.91 mm, respectively) (Figs. 3A, 3B, 3C and 4A, 4B). The differences were statistically significant for the length (p = 0.014) and for the anterior compartment thickness (p = 0.023), but not for the posterior compartment thickness (p = 0.594).
All three readers identified a posterior pelvic floor descent in 19 instances in both enema groups. The visibility score given by all three readers for the posterior pelvic floor descent was higher in the PS group than in the US group (mean visibility score: PS group, 2.8 ± 0.5; US group, 2.6 ± 0.65). However, these differences did not reach statistical significance (Table 2). The extent of pelvic floor descent was rated to be larger in the PS group by all three readers (mean size: PS group, 63.29 ± 8.96 mm; US group, 60.6 ± 8.58 mm; p = 0.029).
The frequency of the distributed abnormalities in the anterior and middle pelvic compartments and the frequency of enteroceles as assessed by the three readers are displayed in Table 3.
|
Except anterior pelvic floor compartment descents (
=
0.571–0.857) (Table 4),
the interobserver agreement was generally good to excellent for all evaluated
pelvic floor abnormalities visualized with both enema compositions ( =
0.613–1.0).
|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MR defecography using either potato starch or ultrasound gel enema provides valuable information regarding the dynamics of rectal evacuation [6, 10–13]. As we expected, in all three pelvic compartments no significant differences in the extent of pelvic floor descent or in the detection of enteroceles were observed between the two groups. The differences were more obvious in the evaluation of the posterior compartment.
The different enema compositions did not result in a difference with regard to the frequency of anterior rectoceles, but the size and completeness of evacuation varied. In general, the extent and degree of evacuation of anterior rectoceles tended to be underestimated when using an ultrasound gel enema in MR defecography. Although the indication for surgical treatment of an anterior rectocele is mainly based on the severity of symptoms, the size (> 3 cm) of the rectocele and the presence of incomplete evacuation are considered important criteria in selecting patients for surgery [4]. Thus, the differences we observed could be clinically relevant.
The cause of rectoanal intussusceptions is still controversial, and the diagnosis of intussusceptions is notoriously difficult using conventional MRI or MR defecography [15]. In a recent investigation, investigators found that large intussusceptions could be the reason for obstructed defecation syndrome [16]. Although there is no general agreement about the appropriate treatment of intussusceptions, surgical treatment of intussusception is being increasingly performed. If surgical treatment is intended, Boccasanta and colleagues [16] and another group [17] proposed that the size (> 10 mm) of the intussusceptions and its extent into the anal canal be used.
In our study, one of 12 intussusceptions diagnosed on MR defecography with potato starch enema was missed on MR defecography with ultrasound gel enema. In addition, the visibility and size of an intussusception scored on MR defecography using potato starch enema were significantly higher on MR defecography using ultrasound gel enema.
The influence of rectal enemas on defecography has not been well investigated in the literature. The only study available to date, which was conducted by Ikenberry et al. [18], comparing three barium contrast media with different viscosity levels in conventional defecography showed that the viscosity level did not affect the radiologic findings.
Our results are not in accordance with those of the study of Ikenberry et al. [18]. Our study revealed significant differences in the visibility and extent of posterior pelvic floor abnormalities in images obtained with different rectal enema compositions. We believe that these differences are due to the different viscosity levels of the enemas. As shown in the in vitro experiment, the viscosity of potato starch is significantly higher than that of ultrasound gel. The higher viscosity most likely results in a prolonged evacuation time with a higher degree of straining. The higher degree of straining may facilitate the formation of intussusceptions and may influence the size of the anterior rectoceles.
We acknowledge the following limitations. First, our study group was relatively small. Second, we did not show that the difference in enema compositions would alter clinical management of patients. Third, we did not measure intrarectal pressure with different rectal enema compositions. Finally, a limitation may relate to the fact that we did not assess the patients for the presence of lateral rectoceles. Because we did not acquire coronal images in all patients, we decided not to include lateral rectoceles as a study parameter.
In conclusion, our study results show that both ultrasound gel enema and potato starch enema provide good contrast and depiction of relevant pelvic floor abnormalities on MR defecography. However, the visibility and extent of pelvic floor abnormalities depend on the composition of the rectal enema. In particular, the size of anterior rectocele, degree of anterior rectocele evacuation, and size of intussusception are often underestimated when using an enema with ultrasound gel.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
