| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research |
1 All authors: Department of Medicine and Clinical Science, Graduate School of Medical Sciences, Kyushu University, Maidashi 3-1-1, Higashi-ku, Fukuoka 812-8582, Japan.
Received September 14, 2007;
accepted after revision January 8, 2008.
Address correspondence to T. Matsumoto
(matane{at}intmed2.med.kyushu-u.ac.jp).
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. One hundred twenty-four patients who underwent double-balloon endoscopy during the period 2004–2006 were classified into those with abnormal radiographic findings (n = 45), normal radiographic findings (n = 31), and no small-bowl radiographs (n = 48). The classification was based on the use of small-bowel radiography and the diagnosis before double-balloon endoscopy. The indications for, approaches to, and diagnostic yields of double-balloon endoscopy were compared for the three groups. The diagnostic yield of small-bowel radiography was considered positive when any sign of pathologic change in the small bowel was identified. The diagnostic yield of double-balloon endoscopy was considered positive when endoscopic or biopsy findings explained the clinical manifestations.
RESULTS. The group with abnormal findings on small-bowel radiography was younger (15–86 years) and less frequently had obscure bleeding (8.9%) than the group with normal findings on small-bowel radiography (age, 17–84 years; frequency of obscure bleeding, 45.2%) (p = 0.01) or the group without small-bowel radiographs (age, 15–91 years; frequency of obscure bleeding, 64.6%) (p < 0.0001). The positive diagnostic yield of double-balloon endoscopy was highest in the group with abnormal findings on small-bowel radiography (71.1%), followed by the group with no small-bowel radiographs (45.8%) and the group with normal findings on small-bowel radiography (35.5%) (p = 0.0002). Among patients who did undergo small-bowl radiography, the accuracy of the technique was 68.4%, the positive predictive value was 71.1%, and the negative predictive value was 64.5%. The positive diagnostic yields of small-bowel radiography and double-balloon endoscopy were not statistically different (59.2% for small-bowel radiography, 56.6% for double-balloon endoscopy; p > 0.1).
CONCLUSION. The diagnostic accuracy of double-balloon endoscopy seems to improve if the procedure is preceded by small-bowel radiography.
Keywords: diagnosis • enteroscopy • small-bowel radiography
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Video capsule endoscopy has been found superior to small-bowel radiography in the detection of pathologic conditions in the small intestine [5–7], especially Crohn's disease [8–11] and intestinal polyposis [12]. However, video capsule endoscopy is not optimal for localization of pathologic changes in the small bowel, and when preparation is poor or intestinal strictures are present, lesions can be missed. For these reasons, small-bowel radiography, especially with the double-contrast technique, has been suggested as a complementary procedure to video capsule endoscopy [13, 14]. It also has been suggested [15–18] that double-balloon endoscopy is equivalent or even superior to video capsule endoscopy for the diagnosis of small-intestine lesions in patients with gastrointestinal bleeding of obscure origin. In this investigation, we retrospectively analyzed the contribution of small-bowel radiography to the diagnostic yield of double-balloon endoscopy.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We routinely perform small-bowel radiography before double-balloon endoscopy. The patients were thus informed of both examinations. Patients who refused small-bowel radiography were not exam ined with that procedure. In the cases of patients with poor performance status or whose condition necessitated an immediate diag nostic procedure, we discussed the decision mak ing for the diagnos tic evaluation with the refer ring physicians.
Indication for Small-Bowel Examinations
The indications for small-bowel radiography and double-balloon endoscopy
were divided into three categories modified from those described by Davies et
al. [19] in 1995. The first
category was obscure bleeding, defined as a clinical manifestation that
includes all three of the following symptoms: hemoglobin concentration less
than 11 g/dL, apparent hematochezia or positive result of a fecal occult blood
test for 2 days or more, and normal findings at both
esophagogastroduodenoscopy and total colonoscopy. Twelve of our 49 patients
with obscure bleeding were further examined with scinti graphy with normal
results. The second cate gory of indications for small-bowel examination was
polyposis of the gastro intestinal tract detected with esophago gastro duo
denoscopy or total colonoscopy. The third category of indic ations was
symptoms sug gestive of tumorous, in flammatory, or miscellaneous pathologic
con ditions of the small intestine. These symptoms included abdominal pain,
vomiting, diar rhea, weight loss, fever, and malabsorption. The indi cation
was obscure bleeding for 49 patients, poly posis for 18 pa tients, and other
symptoms for 57 patients.
|
|
|
|
Double-Balloon Endoscopy
Double-balloon endoscopy was performed with a double-balloon enteroscopy
system (Fig. 3A,
3B). Antegrade double-balloon
endoscopy was perform ed through the mouth and retrograde double-balloon endo
scopy through the anus as described by Yamamoto et al.
[2,
3]. With the patient under
light sedation with 5 mg of midazolam, the scope and the overtube were
advanced to the small intestine with reciprocal inflation and deflation of the
balloons attached to the tips. During insertion of the scope, the balloon of
the overtube was inflated and attached to the small-intestine wall. As an
alter native, the balloon of the scope was inflated and attached to the
intestinal wall during insertion of the overtube. When necessary, the overtube
was advanced under fluoroscopy. At the most distal site during antegrade
double-balloon endoscopy and at the most proximal site during retrograde
double-balloon endoscopy, the small-intestine mucosa was marked with a tattoo
by means of submucosal injection of sterilized ink through an injection
catheter. The small intestine within the reach of the scope was then examined
carefully (Figs. 1B and
2B). Forceps biopsy was
performed if the enteroscopist deemed it necessary.
|
|
The patients were divided into three groups according to the small-bowel radiographic findings: those with the diagnosis of a pathologic condition of the small intestine on the basis of small-bowel radiographic findings, those who had normal small-bowel radiographic findings, and those who did not undergo small-bowel radiography. We compared the positive diagnostic yield of double-balloon endoscopy for the three patient groups. We also assessed the diagnostic values (sensitivity, specifi city, accuracy, and positive and negative predictive values) of small-bowel radiography and compared them with the findings for patients examined with both procedures.
Statistical Analysis
Continuous variables were expressed as median and range and were compared
among the groups with use of the Kruskal-Wallis test and the Mann-Whitney
test. Categoric variables were expressed as percentages and compared among the
groups with the chi-square test or Fisher's exact probability test. The
diagnostic yields of small-bowel radiography and double-balloon endoscopy were
compared with use of the McNemar test. Stepwise logistic regression analysis
was performed to evaluate significant variables, which affected the diagnostic
yield of double-balloon endoscopy. All statistical values were calculated with
a two-tailed method with statistical software (SPSS version 10.0 for Microsoft
Windows, SPSS Japan). A value of p < 0.05 was considered
significant.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Comparison of Clinical and Procedure-Related Features Among Groups
The three groups were heterogeneous in clinical features and in the
double-balloon endoscopy procedure used. As shown in
Table 1, the group with
abnormal findings on small-bowel radiography was significantly younger (median
age, 48 years; range, 15–86 years) than the group with normal
small-bowel radiographic findings (median age, 58; range, 17–84 years)
and the group who did not undergo small-bowel radiography (median age, 62;
range, 15–91 years) (p < 0.01). In addition, the indications
for double-balloon endoscopy were different for the groups (p <
0.0001). Obscure bleeding was the most frequent indication in the group who
did not undergo small-bowel radiography (65%) but was the indication for 45.2%
of the group with normal small-bowel radiographic findings and only 8.5% of
the group with abnormal findings on small-bowel radiography. Indications other
than obscure bleeding and polyposis were most frequent in the group with
abnormal findings on small-bowel radiography (71.1%).
|
The antegrade approach was used most frequently in the group with positive findings on small-bowel radiography (42.2%), and the retrograde approach was used most in the groups with negative findings on small-bowel radiography (54.8%) and who did not undergo small-bowel radiography (41.7%). Both antegrade and retrograde approaches were used in the cases of 20.0% of the patients in the group with abnormal findings on small-bowel radiography, 29.0% of the group with normal small-bowel radiographic findings, and 29.7% of the group who did not undergo small-bowel radiography. The route of double-balloon endoscopy was not different among the three groups.
Final Diagnosis
The final diagnoses established with a combination of small-bowel
radiographic, double-balloon endoscopic, surgical, and histologic findings are
summarized in Table 2.
Seventy-four (59.7%) of the patients were found to have pathologic conditions
of the small intestine. The diagnoses included localized tumors in 10
patients, gastrointestinal polyposis in 13 patients, vascular lesions in 16
patients, inflammatory disease in 30 patients, and other diseases in five
patients. The small-intestine lesion was more frequently verified in the group
with abnormal findings on small-bowel radiography (38 [84.4%] of 45 patients)
than in the group with normal small-bowel radiographic findings (13 [41.9%] of
31 patients) and the group who did not undergo small-bowel radiography (22
[45.8%] of 48 patients) (p < 0.0001). The prevalence of
small-intestine lesions, however, was not different between the group with
normal small-bowel radiographic findings and the group who did not undergo
small-bowel radiography.
|
Comparison of Diagnostic Yield of Double-Balloon Endoscopy Among Groups
Sixty-five patients had abnormal findings at double-balloon endoscopy. When
the indications were taken into consideration, the positive diagnostic yield
of double-balloon endoscopy among patients with polyposis was not different
between the two small-bowel radiography groups (77.7% vs 66.7%)
(Table 3). However, the
positive yield of double-balloon endoscopy among patients with obscure
bleeding and those with other symptoms was higher in the group with abnormal
findings on small-bowel radiography than in the group with normal small-bowel
radiographic findings. The overall positive diagnostic yield was significantly
different among the three groups (p = 0.0002), the highest value
(71.1%) occurring in the group with abnormal findings on small-bowel
radiography followed by the group who did not undergo small-bowel radiography
(45.8%) and the group with normal small-bowel radiographic findings
(35.5%).
|
Stepwise logistic regression analysis revealed that age, sex, indication, and approach to double-balloon endoscopy did not affect the diagnostic yield of double-balloon endoscopy. Small-bowel radiographic findings, how ever, were a significant variable affecting the diagnostic yield of double-balloon endoscopy. Normal findings on small-bowel radiography (odds ratio, 0.18; 95% CI, 0.06–0.54) and no small-bowel radiography (odds ratio, 0.24; 95% CI, 0.09–0.69) contributed significantly to the diagnostic yield of double-balloon endoscopy compared with abnormal small-bowel radiographic findings.
Diagnostic Results of Small-Bowel Radiography
Table 4 indicates the
sensitivity, specificity, accuracy, positive predictive value, and negative
predictive value of small-bowel radiography among 76 patients. Small-bowel
radiography had a sensitivity of 74.4% and specificity of 60.6% for positive
double-balloon endoscopic diagnosis with a prevalence of 56.6%. Both the
accuracy (68.4%) and the positive predictive value (71.1%) of a double-balloon
endoscopic diagnosis increased when small-bowel radiography was performed
first.
|
Table 5 shows the results of a comparison of the diagnostic yield of small-bowel radiography with that of double-balloon endoscopy. Small-bowel radiography depicted abnormalities in 45 (59.2%) of 76 patients, and double-balloon endoscopy did so in 43 (56.6%) of the patients. Small-bowel radiography and double-balloon endoscopy had concordant findings in 52 (68.4%) of the patients, 32 of whom had abnormal findings. The diagnosis was discordant in the other 24 patients. Among the 24 patients, the small-bowel radiographic findings were normal in 11 patients, and the double-balloon endoscopic findings were normal in 13 patients. The diagnostic yield was not statistically different between small-bowel radiography and double-balloon endoscopy. All 13 patients with abnormal small-bowel radiographic and normal double-balloon endoscopic findings had indications other than occult bleeding. In addition, five patients had undergone abdominal surgery, and eight patients had radiographic signs of diffuse inflammation. In contrast, five of 11 patients with normal findings on small-bowel radiography and abnormal double-balloon endoscopic findings were examined because of occult bleeding.
|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Since the introduction and establishment of the procedure, double-balloon endoscopy has become widely accepted, especially in Europe, for the diagnosis and management of pathologic conditions of the small intestine [21–24]. Results of single- and multiple-center analyses indicate that double-balloon endoscopy has contributed greatly to the identification of small-intestine lesions. Although the investigations have proved the safety and feasibility of double-balloon endoscopy, to our knowledge, the diagnostic yield of double-balloon endoscopy has not been compared with that of small-bowel radiography. In four prospective investigations [15–18] with small sample sizes, the diagnostic yield of video capsule endoscopy was compared with that of double-balloon endoscopy. In those investigations, abnormal findings were identified more frequently with video capsule endoscopy than with double-balloon endoscopy in patients with obscure bleeding [15–17], although the diagnostic yield of the latter procedure was greater in patients with suspected small-intestine protrusions or ulcers [15, 18]. It thus seems possible that the yield of each diagnostic procedure varies with indication in patients with suspected small-intestine lesions.
In our retrospective analysis, the prevalence of abnormal findings at double-balloon endoscopy was not affected by indications. It was determined by the result of preceding small-bowel radiography. This observation was further supported by the results of logistic regression analysis, which indicated abnormal findings on small-bowel radiography were the single independent factor associated with abnormal findings at double-balloon endoscopy. From our results and in consideration of the prolonged examination time, the risk of complications, and the invasiveness of double-balloon endoscopy, small-bowel radiography seems an inevitable procedure for the diagnosis of small-intestine lesions. The two procedures also seem complementary; one third of our patients with normal findings on small-bowel radiography had significant findings on double-balloon endoscopy.
During the initial period of use of video capsule endoscopy, the diagnostic value of the procedure was tested in comparison with radiography in patients with obscure bleeding. This issue was initially examined by Costamagna et al. [5], who found a higher diagnostic value of video capsule endoscopy (45%) than of barium follow-through examination (20%). An extremely lower yield of barium follow-through examination was found in an investigation by Hara et al. [7] in which the positive rate was only 3% for barium follow-through examination and 55% for video capsule endoscopy. Results of a meta-analysis indicated that barium follow-through examination was less sensitive in the identification of small-bowel lesions causing obscure bleeding [25]. Although the number of subjects was small, we detected abnormal findings in only four of 18 patients with obscure bleeding, even with small-bowel radiography with the double-contrast technique. These observations suggest that enteroscopy, rather than small-bowel radiography, is the first choice of procedure in the care of patients with obscure bleeding.
Comparisons have been attempted for specific categories of small-intestine lesions—Crohn's disease [9, 10, 26–28] and other forms of gastrointestinal polyposis [12, 29]. In the former category, a meta-analysis showed video capsule endoscopy superior to barium follow-through examination for initial diagnosis and for diagnosis of recurrence [28]. Three prospective studies have shown video capsule endoscopy superior to detailed enteroclysis in the detection of diminutive small-intestine lesions of Crohn's disease [10, 26, 27]. In patients with gastrointestinal polyposis, the diagnostic yield of each procedure was assessed more optimally with the incidence and number of protrusions in the small intestine. Brown et al. [29] detected greater numbers of small-bowel polyps with video capsule endoscopy than with barium follow-through in 19 patients with Peutz-Jeghers syndrome. In an investigation in which the subjects were 24 patients with Peutz-Jeghers syndrome or familial adenomatous polyposis, Mata et al. [12] identified small-intestine polyps more frequently (29%) with video capsule endoscopy than with barium follow-through examination (13%). Because of these observations, video capsule endoscopy has been suggested as an alternative to barium follow-through in the diagnosis of small-intestine lesions.
In contrast to data reported previously, our data showed small-bowel radiography equal to double-balloon endoscopy in the identification of small-intestine lesions in patients with heterogeneous indications for enteroscopy. The favorable diagnostic yield of small-bowel radiography in our patients seems partly explained by the characteristics of our procedure, namely state-of-the-art double-contrast technique with barium and a large amount of air. Because even conventional enteroclysis with barium diluted in methylcellulose solution has been shown superior to CT enteroclysis and MR enteroclysis in the depiction of diminutive mucosal alterations [30–33], our double-contrast technique may be more appropriate for mucosal lesions of the small bowel. It thus seems possible that a certain proportion of patients with abnormal findings on double-contrast small-bowel radiography who have inflammatory or miscellaneous diseases can avoid diagnostic double-balloon endoscopy.
There were several limitations in the interpretation of our results. First, we may have underestimated the diagnostic value of double-balloon endoscopy because we included patients in whom total enteroscopy could not be performed. However, because the prevalence of abnormal findings at double-balloon endoscopy among our subjects (52.4%) was similar to those of other investigations [3, 21–24, 34], false-negative findings at double-balloon endoscopy seem to have contributed little to our results. Second, we may have overestimated small-bowel radiography because the number of patients with obscure bleeding was small and the prevalence of vascular lesions was low. Whereas small-bowel radiography is theoretically inappropriate for flat, diminutive vascular lesions such as angiodysplasia and portal hypertensive enteropathy, those lesions were identified in only 22% of our patients with abnormal finding at enteroscopy. In other investigations from Western countries [21–24, 34], the incidence of vascular lesions was much higher among patients with abnormal findings at double-balloon endoscopy. Third, the retrospective nature of the study resulted in biases in the background of the study populations. There were trends toward use of double-balloon endoscopy without small-bowel radiography in the care of aged patients with bleeding and toward use of both examinations in the care of adolescent patients with presumably inflammatory or miscellaneous diseases. We thus might have underestimated the diagnostic value of double-balloon endoscopy to patients who did not undergo small-bowel radiography. A prospective randomized study with standardized inclusion criteria is warranted to determine the practical role of small-bowel radiography before double-balloon endoscopy.
The results of our retrospective analysis indicated that small-intestine lesions were found more frequently at double-balloon endoscopy of patients with abnormal small-bowel radiographic findings than in those with normal small-bowel radiographic findings or those not examined with small-bowel radiography. In addition, the positive diagnostic yield was not significantly different between double-balloon endoscopy and small-bowel radiography. These results suggest that invasive double-balloon endoscopy should not necessarily be the first choice of procedure for the evaluation of suspected small-intestine lesions. The roles of small-bowel radiography and, preferably, video capsule endoscopy in combination with double-balloon endoscopy need to be evaluated prospectively to establish a diagnostic and therapeutic algorithm for pathologic conditions of the small intestine. In addition, CT and MR enteroclyses seem likely complementary tools because the image quality of these procedures has improved dramatically [35–37]. The issue needs to be elucidated further because enteroscopy has become more routinely used than it was previously.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  D. D. T. Maglinte, M. D. Kohli, S. Romano, and J. C. Lappas Air (CO2) Double-Contrast Barium Enteroclysis Radiology, September 1, 2009; 252(3): 633 - 641. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
