| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1
Ultrasonography Unit, Centre de Diagnòstic per
la Imatge, Hospital Clínic, Villarroel 170,
08036 Barcelona, Spain.
2
Department of Gastroenterology, Institut de Malalties Digestives, Hospital
Clínic, Institut
d'Investigació
Biomèdica August Pi i Sunyer (IDIBAPS),
University of Barcelona, Villarroel 170, 08036 Barcelona, Spain.
3
Department of Nuclear Medicine, Centre de
Diagnòstic per la Imatge, Hospital
Clínic, Villarroel 170, 08036 Barcelona,
Spain.
Received October 23, 2000;
accepted after revision January 3, 2001.
Supported by grant SAF 00/057 from Ministerio de Ciencia y
Tecnología. M. Sans is a recipient of grants
from Fundació
Universitària
Agustí Pedro Pons and Instituto Nacional de
Salud Carlos III.
Abstract
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
SUBJECTS AND METHODS. Sixty-eight consecutive patients with active inflammatory bowel disease (34 ulcerative colitis and 34 Crohn's disease), 12 with inactive inflammatory bowel disease, and 10 control subjects were prospectively studied. Patients with active disease underwent clinical assessment, hydrocolonic sonography, scintigraphy, and colonoscopy within 72 hr, whereas patients with inactive disease and control subjects underwent clinical examination and hydrocolonic sonography.
RESULTS. Involvement of a colonic segment by active inflammatory bowel disease was best defined by mucosal thickness greater than 1.5 mm, bowel wall thickness greater than 4 mm, mucosal irregularity, or the absence of haustra; and involvement of the terminal ileum by bowel wall thickness greater than 4 mm. Using these criteria, hydrocolonic sonography had 100% sensitivity for identifying patients with active inflammatory bowel disease and a greater overall accuracy (87%) than scintigraphy (77%) in the assessment of disease extension. In addition, strong correlation was shown between a hydrocolonic sonography activity index and clinical and endoscopic activity indexes.
CONCLUSION. This prospective study provides precise sonographic criteria for the definition of bowel involvement by active inflammatory bowel disease. Hydrocolonic sonography has a greater accuracy than scintigraphy for assessing disease extension and activity. Therefore, hydrocolonic sonography should be considered a first-choice technique to complete the study of inflammatory bowel disease after confirmation of the diagnosis by histology.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Hydrocolonic sonography consists of transabdominal sonography after retrograde instillation of saline solution into the colon. This technique, initially developed by Limberg [21], allows more detailed imaging of the bowel wall and lumen and a more precise measurement of mucosal and bowel wall thickness than does conventional sonography [21]. Usefulness of this technique in the diagnosis of colonic polyps and cancer has been documented [22, 23], although a recent study found that the accuracy of hydrocolonic sonography is less than that of colonoscopy in this setting [24]. Another study by Limberg and Osswald [25] showed that hydrocolonic sonography has a high accuracy for differentiating between ulcerative colitis and Crohn's disease, as well as for assessing disease extension and activity.
In spite of its potential usefulness, a major drawback to the routine use of conventional sonography or hydrocolonic sonography in the assessment of inflammatory bowel disease is the lack of precise criteria to define bowel involvement by inflammatory bowel disease, which, in addition, makes comparisons of various studies difficult. Thus, although several sonographic characteristics of the bowel have been described in patients with inflammatory bowel disease, no studies have specifically addressed the usefulness of each of these sonographic signs in the assessment of inflammatory bowel disease, and a general consensus with respect to the optimal sonographic definition of bowel involvement by inflammatory bowel disease has not been established.
Therefore, the aims of this study were to characterize the hydrocolonic sonography patterns of the colon and the terminal ileum in patients with active and inactive inflammatory bowel disease and in healthy control subjects; to define the sonographic criteria of bowel involvement by active inflammatory bowel disease; to establish the sonographic differences between patients with ulcerative colitis and those with Crohn's disease; to determine the usefulness of hydrocolonic sonography in identifying patients with active inflammatory bowel disease and in evaluating inflammatory bowel disease extension and activity; and to compare the usefulness of hydrocolonic sonography with that of another noninvasive low-risk examination such as 99mTc-HMPAO—labeled leukocyte scintigraphy in the assessment of inflammatory bowel disease extension and activity.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Study Design
All patients with known or suspected active inflammatory bowel disease
underwent complete clinical examination within 3 days, including determination
of erythrocyte sedimentation rate and C-reactive protein, scintigraphy,
colonoscopy, and hydrocolonic sonography. To avoid potential therapy-induced
alterations in disease activity, the beginning of treatment was delayed in
patients whose disease was newly diagnosed, and no changes in current
treatment (if any) were allowed in patients with previously diagnosed disease,
until completion of all examinations. Investigators who performed colonoscopy,
scintigraphy, and sonography were unaware of diagnoses and clinical records of
the patients or of the results of other examinations. Patients with inactive
inflammatory bowel disease and control patients underwent clinical examination
and hydrocolonic sonography, but not scintigraphy or colonoscopy.
Assessment of Inflammatory Bowel Disease Extension
Colonoscopy was considered the gold standard for the evaluation of
inflammatory bowel disease extension. Endoscopic examinations were performed
in all patients by two experienced endoscopists using a CF-100 endoscope
(Olympus America, Melville, NY). Patients were sedated with IV midazolam
(Rovi, Madrid, Spain) when required. Involvement by inflammatory bowel disease
was assessed in six bowel segments (the rectum; the sigmoid, descending,
transverse, and ascending colons; and the ileum), and endoscopic findings were
graded for every segment using the following scale, which was slightly
modified from a previously described score: 0, lack of lesions and normal
vascular pattern (segment not involved by inflammatory bowel disease); 1, loss
of vascular pattern or mucosal edema or erythema, without ulcers (mild
involvement); 2, superficial ulcers (moderate involvement); and 3, deep ulcers
(severe involvement) [27,
28]. Because endoscopic
assessment of the terminal ileum is often not possible with endoscopy, a
small-bowel barium study was also performed in patients with Crohn's disease,
and ileal involvement by Crohn's disease was defined using pathologic
ileoscopy or barium studies.
Assessment of Inflammatory Bowel Disease Activity
The clinical activity indexes of Van Hees et al.
[29] and Seo et al.
[30] were used for clinical
evaluation of disease activity in patients with inflammatory bowel disease.
Laboratory determinations of erythrocyte sedimentation rate and C-reactive
protein were also used as estimates of disease activity.
Hydrocolonic Sonography
Hydrocolonic sonography was performed within 72 hr before colonoscopy (on
the same day, before endoscopy, in most patients). Colonic preparation for
hydrocolonic sonography or colonoscopy consisted of the oral administration of
a polyethylene glycol-4000 electrolyte solution 10 hr before the procedure.
Sonography was performed after an overnight fast using an SSA 140A scanner
(Toshiba Medical Systems, Madrid, Spain) with color capabilities and using
3.75- (convex) and 8-MHz (linear) transducers. After complete abdominal
sonography, saline (mean, 972 ± 490 mL) was instilled into the colon,
and transabdominal sonography of the bowel was started. The volume of saline
was adjusted for each patient to avoid abdominal pain or discomfort. The
procedure was well tolerated, and hydrocolonic sonography did not induce any
side effects. The total examination time was approximately 15-20 min.
To establish inflammatory bowel disease extension, the bowel was divided into five segments (the sigmoid, descending, transverse, and ascending colons; and the ileum). As previously described by others [25], and as confirmed in our own preliminary study (data not shown), the rectum could not be examined in detail because of air accumulation and the penetration limits of transducers. Thus, assessment of rectal involvement by inflammatory bowel disease was not intended, and patients with ulcerative colitis or Crohn's disease limited to the rectum were not included. The following six items were studied in each colonic segment: mucosal thickness (mm), bowel wall thickness (mm), mucosal regularity, distensibility of the bowel wall, preservation of the bowel wall stratification, and characteristics of the haustra (normal, shortened, or absent). Finally, a hydrocolonic sonography activity index was defined for each patient as the sum of the wall thickness of all bowel segments [6].
99mTc-HMPAO—Labeled Leukocyte Scintigraphy
Leukocytes were labeled "in vitro" with 99mTc-HMPAO,
using a previously described method
[4,
31]. An average of 200 MBq of
99mTc-HMPAO—labeled cells was injected into each patient.
Scintigraphy was performed, obtaining anterior images of the abdomen and
pelvis after the injection of the labeled leukocytes using a large field of
view and a 609 gamma camera (Elscint, Haifa, Israel). The time chosen for
scintigraphic scanning, according to a previous study by our group
[32], was 3 hr after leukocyte
administration. Tracer uptake in the colon was independently graded by two
nuclear medicine physicians according to a previously described scale: 0, no
uptake; 1, mild uptake (less than or equal to that of bone marrow); 2,
moderate uptake (greater than that of bone marrow and less than that of
liver); and 3, severe uptake (greater than or equal to that of liver)
[3,
27,
33]. A scintigraphy activity
index was defined for each patient as the sum of scores of all bowel segments
[3].
Statistical Methods
Data were analyzed using analysis of variance with the
Scheffé (post hoc) test or the Kruskal-Wallis
test, when appropriate. Values are expressed as mean ± SE. Statistical
significance was set at the level of p less than 0.05. Correlation
analyses were performed using the least squares method.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Applicability of Colonoscopy, Hydrocolonic Sonography, and
Scintigraphy
Colonoscopy could examine the rectum and sigmoid colon in all patients, the
descending colon in 65 (95.6%) of 68 patients, the transverse colon in 48
(70.6%) of 68, the ascending colon in 34 (54.0%) of 63, and the terminal ileum
in eight (11.8%) of 68 patients with active inflammatory bowel disease. The
major reasons for incomplete colonoscopy were bowel stenosis and the risk of
perforation in patients with severe inflammatory bowel disease. As explained
in our methods section, the rectum could not be examined in detail using
hydrocolonic sonography because of air accumulation and the penetration
limitations of the transducer. Hydrocolonic sonography revealed the sigmoid
colon in 66 (98.5%) of 67 patients, the descending colon in 67 (98.5%) of 68,
the transverse colon in 66 (97.1%) of 68, the ascending colon in 60 (95.2%) of
63, and the terminal ileum in 59 (86.8%) of 68 patients with active
inflammatory bowel disease. All bowel segments of patients with active
inflammatory bowel disease could be assessed using scintigraphy.
Definition of Sonographic Criteria for Active Inflammatory Bowel
Disease
Involved colonic segments of patients with active inflammatory bowel
disease were characterized by a marked increase in bowel wall thickness, which
was significantly greater than that of noninvolved segments and that of
segments of patients with inactive inflammatory bowel disease or control
subjects (Fig. 1). That
increase was greater in patients with Crohn's disease (5.94 ± 0.32 mm)
than in patients with ulcerative colitis (4.86 ± 0.18 mm, p
< 0.01). A cutoff value of bowel wall thickness greater than 4 mm had a
sensitivity of 70.1% and a specificity of 92.9% for the identification of
involved segments in patients with active inflammatory bowel disease. Involved
colonic segments in patients with active inflammatory bowel disease were also
characterized by a marked increase in mucosal thickness
(Fig. 2). That increase was
similar in patients with Crohn's disease (2.38 ± 0.18 mm) and
ulcerative colitis (2.0 ± 0.14 mm, p not significant). A
cutoff value of mucosal thickness greater than 1.5 mm had a sensitivity of
55.7% and a specificity of 92.9% for the identification of involved segments
in patients with active inflammatory bowel disease.
|
|
Accuracy of hydrocolonic sonography qualitative items in assessing inflammatory bowel disease extension is summarized in Table 1. Globally, all qualitative items were characterized by a high specificity. However, mucosal irregularity and the absence of haustra had a much greater sensitivity and overall accuracy than bowel wall distensibility and stratification for identifying inflammatory bowel disease involvement of bowel segments. Taking into account the previous results, involvement of a colonic segment by active inflammatory bowel disease was defined as the presence of one or more of the following criteria: mucosal thickness greater than 1.5 mm, wall thickness greater than 4 mm, mucosal irregularity, or the absence of haustra (Figs. 3,4A,4B,5).
|
|
|
|
|
Assessment of ileal involvement by inflammatory bowel disease was easier in patients with valve incontinence because of the reflux of saline into the terminal ileum. However, the assessment of mucosal thickness and regularity as well as bowel wall distensibility and stratification was not possible in some cases, even in patients with ileocecal valve incontinence. For this reason, we found the better definition for ileal involvement by Crohn's disease to be bowel wall thickness greater than 4 mm (Fig. 6).
|
Identification of Patients with Active Inflammatory Bowel Disease
Using Hydrocolonic Sonography
Patients with inflammatory bowel disease were considered to have active
disease on hydrocolonic sonography when one or more bowel segments were
involved according to the previously mentioned criteria. Following this
criteria, hydrocolonic sonography proved to have a high accuracy in the
identification of patients with active disease, with a sensitivity of 100%,
specificity of 81.8%, positive predictive value of 94.3%, negative predictive
value of 100%, and an overall accuracy of 95.5%. In addition, a cutoff value
of 13 for the hydrocolonic sonography activity index was also useful in
identifying patients with active inflammatory bowel disease, providing a high
specificity (97.6%), sensitivity (82.3%), and overall accuracy (85.5%).
Usefulness of Hydrocolonic Sonography and
99mTc-HMPAO—Labeled Leukocyte Scintigraphy in Assessing
Inflammatory Bowel Disease Extension
Values for sensitivity, specificity, positive and negative predictive
values, and overall accuracy of hydrocolonic sonography and scintigraphy are
summarized in Table 2.
Hydrocolonic sonography proved to be more useful than scintigraphy in
evaluating inflammatory bowel disease extension. For both diagnostic
techniques, assessment of inflammatory bowel disease extension was more
accurate in patients with ulcerative colitis than in patients with Crohn's
disease. Accuracy of hydrocolonic sonography and scintigraphy for assessing
inflammatory bowel disease extension also varied according to the location of
bowel segments.
|
In addition, we analyzed whether intrinsic characteristics of the disease process (disease duration or severity) or extrinsic factors (treatment) may affect the accuracy of hydrocolonic sonography in assessing inflammatory bowel disease extension (Table 3). The accuracy of hydrocolonic sonography was somewhat less in patients with disease of long duration, those with mild activity, and those receiving steroid treatment.
|
Usefulness of Hydrocolonic Sonography and
99mTc-HMPAO—Labeled Leukocyte Scintigraphy in Assessing
Inflammatory Bowel Disease Activity
The hydrocolonic sonography activity index had a significant correlation
with the clinical activity indexes (r = 0.66, p < 0.0001;
Fig. 7) that was slightly
stronger for patients with Crohn's disease (r = 0.71, p <
0.0001) than for patients with ulcerative colitis (r = 0.61,
p = 0.0003). By contrast, the scintigraphy activity index had a poor
correlation with the clinical activity indexes (r = 0.38, p
= 0.02). Correlation of erythrocyte sedimentation rate and C-reactive protein
with the hydrocolonic sonography activity index was weak (r = 0.44,
p = 0.0006; and r = 0.41, p = 0.006, respectively),
and with the scintigraphy activity index was not significant (r =
0.14, p = 0.28; and r = 0.21 and p = 0.15,
respectively).
|
Good correlation was observed between bowel wall thickness measured with hydrocolonic sonography and the corresponding endoscopic severity of disease in bowel segments (Fig. 8).
|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
A marked increase in bowel wall thickness and mucosal thickness is observed in bowel segments with endoscopically revealed involvement by active inflammatory bowel disease, compared with noninvolved segments or with bowel segments of patients with inactive inflammatory bowel disease and control subjects. In addition, cutoff values of bowel wall thickness greater than 4 mm and mucosal thickness greater than 1.5 mm have a high specificity for identifying involvement of a bowel segment by inflammatory bowel disease. The cutoff point of bowel wall thickness greater than 4 mm is in keeping with that proposed in the study of Limberg and Osswald [25], and in other studies using sonography [7, 8, 15]. However, greater [10] as well as lesser [14] values of bowel wall thickness have also been proposed as sonographic criteria for the diagnosis of inflammatory bowel disease. In addition, our study is in keeping with previous studies that have documented greater increments in bowel wall thickness in Crohn's disease than in ulcerative colitis [25, 34].
We also assessed the usefulness of four qualitative sonographic parameters related to characteristics of the bowel wall—namely, mucosal irregularity, lack of distensibility of the bowel wall, loss of bowel wall stratification, and alterations of the haustra—that have been previously described in patients with inflammatory bowel disease. Although all these items had a high specificity, the sensitivity of mucosal irregularity and the absence of haustra was much greater than that of lack of bowel wall distensibility or loss of bowel wall stratification for identifying bowel involvement by inflammatory bowel disease.
In the study of Limberg and Osswald [25], differentiating between ulcerative colitis and Crohn's disease was possible using hydrocolonic sonography because mucosal irregularity with pseudopolyps was present mainly in patients with active ulcerative colitis, whereas loss of bowel wall stratification was predominantly found in patients with active Crohn's disease. By contrast, in our study mucosal irregularity was frequently found in involved segments of patients with both ulcerative colitis and Crohn's disease. As for the loss of normal bowel wall stratification, our findings are in keeping with previous evidence showing that this sonographic sign is only found in a minority of patients with active inflammatory bowel disease [15]. The discrepancies between the study of Limberg and Osswald and our study might be related to the skipping distribution of mucosal changes in Crohn's disease, and the fact that we considered mucosal irregularity to be present when it was detected in any area of the bowel segment explored by hydrocolonic sonography.
A novel aspect of our study is the proposal of combined sonographic criteria that brings about the greatest accuracy in the definition of involvement of a bowel segment by inflammatory bowel disease. The presence of one or more of the following sonographic signs—mucosal thickness greater than 1.5 mm, bowel wall thickness greater than 4 mm, mucosal irregularity, and the absence of haustra—accurately identified involvement of a colonic segment by active inflammatory bowel disease. These criteria proved useful in the assessment of patients with ulcerative colitis and those with Crohn's disease. In addition, the finding of a bowel wall thickness greater than 4 mm provided the best sonographic definition of involvement of the terminal ileum by active Crohn's disease.
Our study also disclosed several limitations of the use of hydrocolonic sonography in assessing inflammatory bowel disease. Our preliminary experience was in agreement with that of most previous sonography and hydrocolonic sonography studies, in which assessment of the rectum was not possible because of air accumulation and the penetration limitations of the transducers [7, 15, 25]. In addition, our study reveals some difficulties in the assessment of ileal involvement by Crohn's disease, which is influenced by ileocecal valve continence and the subsequent reflux of saline into the terminal ileum. Although in some cases of severe ileitis this segment could be easily assessed using hydrocolonic sonography without reflux of the saline solution into the ileum, in other cases of less severe ileitis the assessment was suboptimal in the absence of saline reflux.
Hydrocolonic sonography proved to be more useful than 99mTc-HMPAO—labeled leukocyte scintigraphy in the assessment of inflammatory bowel disease. Although both techniques had a high sensitivity, specificity, and overall accuracy in identifying patients with active inflammatory bowel disease, hydrocolonic sonography was more useful than scintigraphy in evaluating disease extension. In a previous study, similar sensitivities for conventional sonography (73%) and labeled leukocyte scintigraphy (69%) were shown in the assessment of inflammatory bowel disease extension [35]. Our study, confirming the relatively low sensitivity for scintigraphy (72%) but proving a greater sensitivity for hydrocolonic sonography (86%), strongly suggests that the latter might be preferable to both scintigraphy and conventional sonography for evaluating inflammatory bowel disease extension. In addition, the hydrocolonic sonography activity index had a better correlation with the clinical activity index than the scintigraphy activity index had. Finally, the usefulness of hydrocolonic sonography has also been analyzed in various subgroups of patients, showing a slightly lower accuracy for assessing inflammatory bowel disease in patients with disease of long duration, patients with mild disease, and patients receiving steroid treatment.
Collectively, the observations from our study indicate that hydrocolonic sonography has a definite role in the assessment of inflammatory bowel disease extension and activity in clinical practice. Considering its high accuracy, low risk, and cost, we believe hydrocolonic sonography should be considered a first-choice technique in the study of disease extension in patients with incomplete colonoscopy, after confirmation of the diagnosis by histology. In patients in whom the diagnosis of inflammatory bowel disease has been established, hydrocolonic sonography may be sufficient to reassess disease extension and provide an estimate of disease severity.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  G. A. J. Sempere, V. Martinez Sanjuan, E. Medina Chulia, A. Benages, A. Tome Toyosato, P. Canelles, A. Bulto, F. Quiles, I. Puchades, J. Cuquerella, et al. MRI Evaluation of Inflammatory Activity in Crohn's Disease Am. J. Roentgenol., June 1, 2005; 184(6): 1829 - 1835. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
