| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Innovation |
1 Department of Medical Diagnostic Sciences and Special Therapies, Padua
University Hospital, Via Giustiniani 2, 35128 Padua, Italy.
2 Department of Environmental Medicine and Public Health, Padua University
Hospital, Padua, Italy.
3 Department of Oncologic and Surgical Sciences, Padua University Hospital,
Padua, Italy.
4 IRCCS-IOV, Padua, Italy.
Received September 6, 2006;
accepted after revision March 28, 2007.
Address correspondence to F. Pomerri
(fabio.pomerri{at}unipd.it).
Abstract
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CONCLUSION. With the barium trace technique, the anatomic conspicuity of colonic segments is improved, a correct segmental marker count can be obtained, and colonic inertia can be more easily distinguished from distal constipation.
Keywords: colon • constipation • gastrointestinal imaging • radiopaque markers • transit time study
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Two main radiographic techniques can be used to analyze marker transit data from the colon. With one, markers are ingested in a single dose and abdominal images are obtained every 24 hours until the markers are no longer visible [5]. This technique has been criticized because it incurs exposure to radiation [4, 6] and is time-consuming and inconvenient to the patient.
With the other technique, an identical number of markers are ingested on each of three consecutive days, and an abdominal radiograph is obtained on the fourth day [6]. Using this technique, radiologists can monitor the progress of markers in the colon only over 72 hours, thus leading to an underestimation of colonic transit time in cases of colonic transit that take more than 72 hours. Another radiograph on day 7 is required for a more accurate quantification of the degree of abnormality [6]. The main criticism of this method is that it provides inaccurate measurements because the equilibrium between the mean daily output of markers and the marker input [7] is not necessarily reached within 3 days [4]. Furthermore, the segmental colonic marker count is necessary for calculating colonic transit times using all equations described by different authors. Markers are located and counted on abdominal radiographs according to bone landmarks and clear bowel outlines [5]. These widely used criteria are flawed. The aim of the present study was therefore to test the feasibility of a more effective technique for counting radiopaque markers in the radiologic measurement of colonic transit times.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
There were no set time limits for the viewing sessions. Patient radiograph order was random. Each observer counted markers twice at an interval of 6 weeks and was blind to the results of the previous count and to other observers' findings until study completion.
Statistical calculations were made using statistics software (SAS version
8.2, SAS Institute). Assessment of agreement in marker counts for each colonic
segment, made using the repeatability coefficient and the 95% limits of
agreement [8], was based on
within observers, between observers (first count used), and between methods
(first count used) variation. Intraobserver repeatability was analyzed by
plotting the differences between replicated counts against their mean and
calculating the repeatability coefficient, which was estimated to express the
95% limits within which the differences between two marker counts by the same
observer would lie. Interobserver reproducibility and between-method
reproducibility were expressed as the difference between the counts of each
pair of observers and the difference between marker counts obtained with two
methods, respectively, plotted against their mean. Interobserver and
between-method agreements were also described by the 95% limits of agreement,
defining the range within which 95% of the differences are expected to lie.
One-way analysis of variance was used to analyze total and segmental colonic
transit times. Statistical significance was p 
0.05.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
In group 1, no clear bowel outlines were identified in 19 (36.5%) examinations, bowel outline visibility was partial in 30 (57.7%), and a clear bowel outline of the entire colon was visible in three (5.8%). The colon was inadequately conspicuous to appropriately assign all markers to the various segments in 49 (94.2%) abdominal radiographs. In group 2, barium trace was present on abdominal radiographs throughout the entire colon in 45 patients (42.1%). In the remaining, barium trace was present in two of the three colonic segments, and the markers not counted in the two bariumtraced colonic segments were, obviously, assigned to the third colonic segment. Focal barium accumulation, present in 17 (15.9%) examinations, may have hidden some markers. To assess the reliability of the marker counts, we considered the differences between counts, in relation to observer, method, and colonic segment; observers, in relation to method and colonic segment; and methods, in relation to observer and colonic segment.
On evaluating intraobserver repeatability (type 1 differences), the repeatability coefficient, evenly distributed from 0.38 to 3.52, suggested that individual differences were fewer than four markers. On evaluating interobserver reproducibility (type 2 differences) in the plain marker count, the 95% limits of agreement between observer pairs were more than four markers except for the left colon. By contrast, in zonal (groups 1 and 2) and barium-traced (group 2) marker counts, the 95% limits of agreement were never more than four markers. On evaluating between-method reproducibility (type 3 differences), the 95% limits of agreement were always more than five markers.
To graph the poor agreement in between-method reproducibility, differences between barium-traced and zonal marker counts related to the right colon and the rectosigmoid were plotted against the corresponding means for observer A (Fig. 2A, 2B). The graphs obtained showed that most differences different from zero calculated in the right colon had positive values (mean difference, 2.53 markers; Fig. 2A), whereas for the rectosigmoid most differences different from zero had negative values (mean difference, -3.57 markers; Fig. 2B). The mean colonic transit times for barium-traced versus zonal marker count were, respectively, 31.51 ± 25.03 versus 25.44 ± 21.31 hours for the right colon (p = 0.057; near to significance), 28.22 ± 26.82 versus 25.68 ± 23.16 hours for the left colon (p = 0.460; nonsignificant), and 15.21 ± 16.24 versus 23.80 ± 22.36 hours for the rectosigmoid (p = 0.001; significant) (observer A, first count used).
|
|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The main result for between-method agreement was that most differences different from zero were positive for the right colon and negative for the rectosigmoid. An important finding was that the imaginary lines drawn on bone landmarks identify projection zones of colonic segments and not anatomic colonic segments. The topography of colonic loops was identified at barium-traced imaging and 34 (31.8%) abdominal radiographs failed to show one or two anatomic colonic segments in the projection zones identified according to bone landmarks, frequently owing to pelvic cecum and transverse colon. Consequently, the negative differences in the rectosigmoid corresponded to the positive differences in the cecum and transverse colon, suggesting that colonic barium trace is an understandable anatomic count criterion.
The main finding for transit time measurements was that rectosigmoid transit time was significantly overestimated by the classic marker count, hindering the differentiation between distal and slow-transit constipation. At least three subtypes of slow colonic transit were identified by colonic transit times, measured on the basis of the distribution of markers visible on abdominal radiographs, although there may have been an overlap between these subtypes: colonic transit is slow, the major site of delay being the ascending colon (colonic inertia); the delay is predominantly found in the descending colon (hindgut dysfunction); and marker retention is visible mainly in the rectosigmoid area (outlet obstruction) [2, 4, 5].
The overlap between patterns of colonic hypomotility may be due to the dysfunction of a colonic segment or to distal fecal impaction, which can cause nonpropagation or back-propagation of the markers [2]. In our two cohorts, retention of all markers or almost all markers throughout the whole colon or in one colonic segment was never found on the day 11 abdominal radiograph. However, if no marker discharge is observed and an exact numeric value is required for total and segmental colonic transit times in patients with a total colonic transit time of more than 240 hours, marker ingestion should be continued for a few more days and the marker count applied to a later abdominal radiograph, after stools and part of markers have been passed. A limitation of our technique was occasional focal barium accumulation, which may have hidden some markers. In such cases, a graded compression radiograph or an abdominal radiograph in a projection other than anteroposterior is required to allow correct marker counting.
In conclusion, classic marker counts appear to be unreliable because it is difficult to identify the anatomy of the colonic outlines. The efficacy of colonic marker count might be improved by oral administration of a small quantity of barium. Our proposed technical modification allows the identification of the projective overlap of functionally distant colonic segments, making the localization of all markers accurate, thus representing an improvement over traditional techniques for the measurement of colonic transit time.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
