HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Slater, A. Articles by Halligan, S. Search for Related Content PubMed PubMed Citation Articles by Slater, A. Articles by Halligan, S.

Colonic Polyps: Effect of Attenuation of Tagged Fluid and Viewing Window on Conspicuity and Measurement—In Vitro Experiment with Porcine Colonic Specimen¹

¹ From the Department of Imaging, University College Hospital, University of London, 250 Euston Rd, London NW1 2BU, England (S.A.T., S.H.); and Department of Intestinal Imaging, St Mark's Hospital, London, England (A.S., D.B., L.G., J.S.). Received June 13, 2005; revision requested August 9; revision received August 22; accepted September 15; final version accepted October 17. Address correspondence to S.A.T. (e-mail: csytaylor{at}yahoo.co.uk).

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION ADVANCES IN KNOWLEDGE References

 
Purpose: To investigate effect of attenuation of tagged fluid and viewing window on polyp conspicuity and measurement with porcine colonic specimen.

Materials and Methods: Eleven (3–10-mm-diameter) polyps were created in porcine colon and the specimen submerged in saline. Four–detector row CT was performed after gas distension and after filling with six barium sulfate suspensions (attenuation, 100–1000 HU). Two readers independently measured maximal two-dimensional polyp diameter on each data set with the following four viewing windows and window levels and window widths, respectively: colon (–150 HU, 1500 HU), lung (–500 HU, 1500 HU), bone (500 HU, 2500 HU), and abdomen (40 HU, 400 HU). In consensus, polyp conspicuity (compared with air data set) was assigned a grade of 1–4 for each viewing window (grade 1, not seen or barely visible; grade 4, optimally seen). For statistical analysis, conspicuity grades were collapsed to a two-point scale. Data were analyzed with Mann-Whitney, Kruskal-Wallis, and ² tests.

Results: Accuracy of polyp measurement was independent of viewing window for attenuation of tagged fluid of 100–300 HU but differed significantly for 500–1000 HU (P < .001); that for colonic and bone viewing windows was superior (median size difference, 1.0 mm; interquartile range, 0.5–1.5). Conspicuity differed significantly according to viewing window at all attenuation values (P < .001). For 100–300 HU with abdominal viewing window, 83% (24 of 29) of observations were assigned grade 3 or 4 (best). For 500–1000 HU with bone viewing window, 94% (30 of 32) of observations were assigned grade 3 or 4 (superior). Overall conspicuity was best with bone viewing windows at 700 HU.

Conclusion: Polyp conspicuity and measurement in tagged data sets were optimized at 700 HU with bone viewing windows. At less than 300 HU, conspicuity improved with abdominal viewing windows.

© RSNA, 2006

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION ADVANCES IN KNOWLEDGE References

 
Recent discrepant results from large multicenter trials with computed tomographic (CT) colonography (1–3) have led to a refocusing of attention on optimization of technique and interpretation. In the largest and most successful trial to date, in addition to primary three-dimensional analysis, Pickhardt and colleagues used a fecal-and-fluid–tagging regimen (1), and it has been suggested that this regimen most likely contributed to their success (4). Some form of oral tagging is increasingly used in many CT colonographic protocols, although the optimal regimen is under investigation (5–8).

Two-dimensional (2D) reading strategy for nontagged data sets is well established (9–13), with most observers advocating the use of both colonic and soft-tissue viewing window review. (A nontagged data set is a data set in which residual fluid attenuation has not been increased with addition of a contrast agent.) Findings in recent work have also suggested that 2D polyp measurement is best performed by using a colonic viewing window (14). At the time of this writing, however, there is no clear consensus for optimal analysis of tagged data sets. (A tagged data set is a data set in which residual fluid attenuation has been increased with addition of a contrast agent.) The attenuation of tagged fluid may influence conspicuity of submerged polyps. It is unclear what the best viewing window is for image analysis and polyp measurement in this context and whether the best viewing window differs according to the attenuation of tagged fluid. Although subtraction software eventually may simplify interpretation, this software is not without artifacts (15–17), and its use is not widespread. Thus, the purpose of our study was to investigate the effect of attenuation of tagged fluid and viewing window on polyp conspicuity and measurement with a porcine colonic specimen.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION ADVANCES IN KNOWLEDGE References

 
Specimen Preparation
A 25-cm-long section of porcine colon was washed, trimmed of mesentery, and inverted (A.S., L.G.). An abattoir provided the specimen (from an animal previously slaughtered for human consumption). Eleven "polyps" were created (J.S.) by puckering the mucosa with surgical forceps and forming a purse-string suture around the base by using 2/0 absorbable sutures (Vicryl; Ethicon Endo-Surgery, Cincinnati, Ohio) (Fig 1). The purpose was to alter the volume of mucosa lifted for each polyp to produce sessile lesions ranging in diameter from approximately 3 mm to more than 10 mm, and the size was confirmed by measurement with hand-held calipers, as discussed later.

View larger version (45K): [in this window] [in a new window] [Download PPT slide]   Figure 1: Inverted colonic specimen demonstrates method of polyp (arrows) formation. From left to right, actual polyp sizes were 3, 6, 10, 5, 10, 3, 9, 7, 4, 5, and 7 mm.

Barium Suspensions
Six 1-L bags of barium sulfate suspensions with attenuation values approximating to 100, 200, 300, 500, 700, and 1000 HU were then created (A.S., L.G.) by mixing different volumes of barium sulfate (Readicat 2%, E-Z-Em; Baritop, Sanochemia Diagnostics UK, Bristol, England) with normal saline. The attenuation of the suspensions was checked prior to scanning: Each bag was scanned with a four–detector row CT scanner (LightSpeed Plus; GE Medical Systems, Milwaukee, Wis) and the following parameters: 1.25-mm collimation, pitch of 1.5, 120 kVp, and 50 mA. The attenuation in Hounsfield units was measured in large regions of interest (at least 10 cm²) placed within the suspension.

Imaging
The prepared colonic specimen was then placed (A.S., L.G.) in a plastic box filled with 20 L of 0.9% saline mixed with 60 mL of diatrizoate meglumine containing 370 mg of iodine per milliliter (Gastrografin; Schering Health Care, Burgess Hill, West Sussex, England). The specimen provided an average attenuation value of 36 HU, which was similar to that of abdominal tissue (18). The specimen was arranged so as to mimic the normal anatomic configuration of the human colon as closely as possible. The plastic box had a locking lid, which pushed down on a plastic spacing device, thus submerging and securing the specimen. The colonic specimen was reinflated with air by using a hand pump until it was visibly well distended. CT scanning was then performed by using the same four–detector row CT scanner as was mentioned before with the following parameters: 1.25-mm collimation, pitch of 1.5, 120 kVp, and 50 mA (Fig 2). Images were reconstructed by using 50% overlap (0.625 mm) and were viewed by using a 26-cm field of view (matrix, 512 x 512).

View larger version (137K): [in this window] [in a new window] [Download PPT slide]   Figure 2: Three-dimensional endoluminal reformatted image of the gas-filled specimen demonstrates morphologic features of created polyps (arrows).

The whole process was repeated for each of the remaining five barium suspensions, ending with the suspension with an attenuation value of 1000 HU. The colon was completely emptied of each barium solution prior to introduction of the next. Saline washing was, however, not performed between the introduction of suspensions for fear of disrupting the sutured polyps.

Polyp Measurement and Location
At the end of the experiment, the specimen was reinverted and gently washed with saline, and the maximal dimension of the base of each polyp again was measured to the nearest 0.5 mm by using hand-held calipers to ensure that there had been no change from the measurement determined before the experiment. The measurements before and after the experiment were performed by the same observer (A.S.) by using the same plane of measurement for each polyp.

The study coordinator (A.S., with 3 years of experience in CT colonography) then carefully documented the exact location of each polyp in the inverted specimen with reference to measurements to stable landmarks of the sutured end and serosal staples and with measurements from the polyp to neighboring polyps. In this fashion, a detailed polyp map of the specimen was produced. Each of the seven data sets was loaded onto a workstation (Advantage Windows 4.2; GE Medical Systems) with dedicated CT colonographic software (CT Colonography 2; GE Medical Systems), and the study coordinator, with reference to the position measurements and map drawing, then noted the section number of each polyp in each data set. If a polyp could not be visualized on any viewing window, the observer wrote down the section number of its theoretical position on the basis of the known distance to other visible polyps. In data sets obtained when the colonic specimen was filled with barium suspensions, the observer also noted if the polyp was not completely submerged in the tagged fluid and excluded it if it was not. The final attenuation value of each barium suspension was estimated by calculating the average of the attenuation values of three large regions of interest that were at least 10 cm² and were placed along the length of the colonic specimen.

Image Interpretation
Two experienced readers (S.A.T., D.B.), who had experience with more than 400 CT colonographic data sets with full endoscopic correlation, analyzed each of the seven data sets presented in random order independently by using the same viewing software. Each reader was given the exact polyp location on a data sheet that detailed the individual polyp locations by means of section numbers and a graphic representation of the specimen. Readers were not blinded to polyp location because the main purpose of the study was to investigate the effect of attenuation of tagged fluid and viewing window on polyp conspicuity and measurement rather than on polyp detection. Readers, however, were blinded to the polyp size. For each polyp, the observer then measured the maximal 2D diameter (on a transverse or multiplanar [sagittal, coronal, or oblique] reconstructed image as deemed appropriate) with electronic calipers embedded within the software by using each of four prespecified viewing windows as follows: colon (window level, –150 HU; window width, 1500 HU), lung (window level, –500 HU; window width, 1500 HU), bone (window level, 500 HU; window width, 2500 HU), and abdomen (window level, 40 HU; window width, 400 HU).

The study coordinator randomized the order of the viewing window used for each polyp. Readers' measurements were handed to the study coordinator after each data set was analyzed, and a reading took place during two sessions temporally separated by 1 week.

At the end of the second reading session, the two readers reviewed all data sets obtained when the colonic specimen was filled with barium suspensions and, in consensus, subjectively assigned a grade of 1–4 to the conspicuity of each submerged polyp: grade 1, not visible or barely visible, grade 2, poorly seen; grade 3, well seen; and grade 4, optimally seen. The readers were unaware of the attenuation of the tagged fluid when these observations were made. The data set obtained when the colonic specimen was distended with room air and viewed on colonic viewing windows was the baseline reference standard against which conspicuity was judged (Fig 3). This assessment was made at each of the four viewing windows described previously, and, again, the viewing windows were applied in a random order.

View larger version (58K): [in this window] [in a new window] [Download PPT slide]   Figure 3a: Transverse 2D images show polyps with conspicuity grades assigned by two experienced readers in consensus. (a) Air-distended specimen with 9-mm polyp (arrow) is reference standard for assignment of conspicuity grade for data sets containing intraluminal tagged fluid. (b-e) Specimen filled with tagged fluid (attenuation value, 700 HU) viewed on (b) abdominal window (window level, 40 HU; window width, 400 HU), (c) lung window (window level, –500 HU; window width, 1500 HU), (d) colonic window (window level, –150 HU; window width, 1500 HU), and (e) bone window (window level, 500 HU; window width, 2500 HU) images. Same polyp (arrow) as in a is barely visible (observer conspicuity grade 1) in b, poorly seen (observer conspicuity grade 2) in c, well seen (observer conspicuity grade 3) in d, and optimally visible (observer conspicuity grade 4) in e.

View larger version (64K): [in this window] [in a new window] [Download PPT slide]   Figure 3b: Transverse 2D images show polyps with conspicuity grades assigned by two experienced readers in consensus. (a) Air-distended specimen with 9-mm polyp (arrow) is reference standard for assignment of conspicuity grade for data sets containing intraluminal tagged fluid. (b-e) Specimen filled with tagged fluid (attenuation value, 700 HU) viewed on (b) abdominal window (window level, 40 HU; window width, 400 HU), (c) lung window (window level, –500 HU; window width, 1500 HU), (d) colonic window (window level, –150 HU; window width, 1500 HU), and (e) bone window (window level, 500 HU; window width, 2500 HU) images. Same polyp (arrow) as in a is barely visible (observer conspicuity grade 1) in b, poorly seen (observer conspicuity grade 2) in c, well seen (observer conspicuity grade 3) in d, and optimally visible (observer conspicuity grade 4) in e.

View larger version (68K): [in this window] [in a new window] [Download PPT slide]   Figure 3c: Transverse 2D images show polyps with conspicuity grades assigned by two experienced readers in consensus. (a) Air-distended specimen with 9-mm polyp (arrow) is reference standard for assignment of conspicuity grade for data sets containing intraluminal tagged fluid. (b-e) Specimen filled with tagged fluid (attenuation value, 700 HU) viewed on (b) abdominal window (window level, 40 HU; window width, 400 HU), (c) lung window (window level, –500 HU; window width, 1500 HU), (d) colonic window (window level, –150 HU; window width, 1500 HU), and (e) bone window (window level, 500 HU; window width, 2500 HU) images. Same polyp (arrow) as in a is barely visible (observer conspicuity grade 1) in b, poorly seen (observer conspicuity grade 2) in c, well seen (observer conspicuity grade 3) in d, and optimally visible (observer conspicuity grade 4) in e.

View larger version (81K): [in this window] [in a new window] [Download PPT slide]   Figure 3d: Transverse 2D images show polyps with conspicuity grades assigned by two experienced readers in consensus. (a) Air-distended specimen with 9-mm polyp (arrow) is reference standard for assignment of conspicuity grade for data sets containing intraluminal tagged fluid. (b-e) Specimen filled with tagged fluid (attenuation value, 700 HU) viewed on (b) abdominal window (window level, 40 HU; window width, 400 HU), (c) lung window (window level, –500 HU; window width, 1500 HU), (d) colonic window (window level, –150 HU; window width, 1500 HU), and (e) bone window (window level, 500 HU; window width, 2500 HU) images. Same polyp (arrow) as in a is barely visible (observer conspicuity grade 1) in b, poorly seen (observer conspicuity grade 2) in c, well seen (observer conspicuity grade 3) in d, and optimally visible (observer conspicuity grade 4) in e.

View larger version (79K): [in this window] [in a new window] [Download PPT slide]   Figure 3e: Transverse 2D images show polyps with conspicuity grades assigned by two experienced readers in consensus. (a) Air-distended specimen with 9-mm polyp (arrow) is reference standard for assignment of conspicuity grade for data sets containing intraluminal tagged fluid. (b-e) Specimen filled with tagged fluid (attenuation value, 700 HU) viewed on (b) abdominal window (window level, 40 HU; window width, 400 HU), (c) lung window (window level, –500 HU; window width, 1500 HU), (d) colonic window (window level, –150 HU; window width, 1500 HU), and (e) bone window (window level, 500 HU; window width, 2500 HU) images. Same polyp (arrow) as in a is barely visible (observer conspicuity grade 1) in b, poorly seen (observer conspicuity grade 2) in c, well seen (observer conspicuity grade 3) in d, and optimally visible (observer conspicuity grade 4) in e.

The level of agreement between observers was then assessed by using the Bland-Altman limits of agreement (mean difference ± [1.96 x standard deviation of differences]).

To determine the effect of attenuation of tagged fluid and viewing window on the accuracy of the diameter measurements, the absolute differences between the mean measurement for the two observers and the actual polyp size were calculated. These differences were not normally distributed and, thus, were compared across the four viewing windows by using the Kruskal-Wallis test for each of the six individual data sets obtained when the colonic specimen was filled with barium suspensions and then by combining the data from all six data sets.

Initial analysis of the polyp conspicuity data revealed that very few observations were assigned a grade of 4, and a minority of them were assigned a grade of 1. Therefore, the outcome measure was reduced to a two-point scale, a grade of either 1 or 2 (poorly seen) or a grade of either 3 or 4 (well seen).

The relationship between conspicuity grades and the viewing windows used for visualization was then examined by using the ² test for each attenuation value of tagged fluid. In an attempt to produce a simplified analysis, this calculation was repeated after combining the results from the 100–300-HU data sets and again after combining the results from the 500–1000-HU data sets. For the combined results, the analysis was repeated for polyps that were 5 mm or smaller and those that were larger than 5 mm in size.

To determine the optimal viewing window paradigm for tagged data, the viewing window with the highest polyp conspicuity values at each attenuation value for tagged fluid was selected and compared by using the ² test for trend.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION ADVANCES IN KNOWLEDGE References

 
Polyp Measurement and Attenuation of Barium Suspension
The maximal diameters for the 11 polyps in the specimen were 3, 3, 4, 5, 5, 6, 7, 7, 9, 10, and 10 mm, and they were unaffected by the actual experiment.

For the six bags of barium sulfate suspension (measured in the specimen), the final attenuation values were 114, 188, 277, 501, 722, and 1028 HU. For ease of presentation, the attenuation values of these suspensions will be referred to as 100, 200, 300, 500, 700, and 1000 HU, respectively. One polyp (7 mm) in the 100-HU data set, three polyps (3, 9, and 10 mm) in the 300-HU data set, and one polyp (3 mm) in the 500-HU data set were not submerged in tagged fluid and were excluded from the analysis.

Interobserver Agreement for Polyp Diameter
There was no evidence of any association between the difference in polyp diameter measured by the readers and attenuation of tagged fluid for any of the four viewing windows (absolute difference Pearson coefficient range, –0.03 to 1.9; P = .15–.83). Interobserver agreement for polyp diameter was very similar for all four viewing windows, with 95% Bland-Altman limits of agreement spanning around 5 mm (Table 1).

Viewing window was not significantly associated with the accuracy of diameter measurements when polyp diameter measurements were combined from data sets obtained when the colonic specimen contained barium sulfate suspensions of attenuation values of 100–300 HU. When diameter measurements were combined from data sets obtained when the colonic specimen contained barium sulfate suspensions of attenuation values of 500–1000 HU, however, there was a statistically significant association between the viewing window used and accuracy, again with colonic and bone viewing windows superior to lung and abdominal viewing windows.

Polyp Conspicuity
There was a highly significant association between polyp conspicuity grade and the four viewing windows used at each attenuation value for tagged fluid (Table 3). Polyps were least conspicuous when lung viewing windows were used for all attenuation values. For each attenuation value from 100 to 300 HU, the abdominal viewing window was superior. Polyps were most conspicuous when bone and colonic viewing windows were used at 500 and 700 HU, whereas polyps were most conspicuous when the bone viewing window alone was used at 1000 HU (Fig 4).

View larger version (39K): [in this window] [in a new window] [Download PPT slide]   Figure 4: Bar chart shows relationship between attenuation value of tagged fluid and viewing window on polyp conspicuity classified as well seen (grade 3) or optimally seen (grade 4).

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION ADVANCES IN KNOWLEDGE References

Tagged feces and fluid in CT colonographic data sets present new challenges to the reporting radiologist. Most workers experienced in 2D interpretation of data sets with untagged fluid routinely use a colonic viewing window (window level, approximately –150 HU, window width, 1500 HU) for analysis (12,22,23), with abdominal viewing windows recommended for flat adenomas (13). To our knowledge, however, there are no data that support this strategy for viewing of tagged data sets.

We found that the attenuation value of tagged fluid was significantly associated with both polyp conspicuity and diameter measurement. For relatively dilute suspensions (300 HU or less), conspicuity was significantly improved when abdominal viewing windows were used. This observation held true for polyps of all sizes. As the attenuation value of tagged fluid increased, however, the performance with both colonic and bone viewing windows became superior. At the highest attenuation value we tested (1000 HU), the performance with the bone viewing window was best, in terms of both polyp conspicuity and accuracy of polyp diameter measurement, as discussed later. This advantage was maintained when the results from data sets with attenuation values for tagged fluid of 500–1000 HU were combined, and this advantage also held true for both small and large polyps.

To put our findings into a clinical context, if standard colonic viewing windows alone were used when we analyzed a tagged data set, more than half of all submerged polyps would be either invisible or poorly seen if the attenuation value of tagged fluid happened to be either 300 HU or less or approximately 1000 HU or more. It would, therefore, seem advisable to change viewing windows contingent on the attenuation value of tagged fluid when one reviews submerged areas to ensure that no polyp is missed. This strategy would increase report times, but efficiency could be improved if only those areas that are observed to be submerged on both supine and prone data sets are reanalyzed; recent data suggest that approximately 22% of the colonic surface may be submerged in fluid, on average (24). Even low-volume barium-based tagging regimens produce substantial volumes of tagged fluid (25).

When polyps were surrounded by air, reader measurement accuracy was good, although we found no association between the viewing window and the diameter measurement. As might be expected, however, in air, the narrowest interquartile range between measured and actual polyp diameter was obtained by using colonic viewing windows. Although polyp conspicuity in lower-attenuation data sets was significantly improved by using abdominal viewing windows, this did not translate into more accurate measurement of diameter on the basis of the mean measurements of both observers. By way of explanation, it seems likely that any small improvement in accuracy could have been offset by the baseline interobserver variation in diameter. Furthermore, given that there is inherent measurement error even with a colon that is optimally distended with gas, it cannot be assumed that increased conspicuity will translate into improved accuracy. For example, if a reader tends to overestimate polyp size in air, reduction of conspicuity (and apparent polyp dimension) may offset this overestimation. To reproduce clinical conditions, we did not specify the plane for polyp measurement but instead allowed readers to choose their own, which also may have decreased accuracy.

Although interobserver Bland-Altman 95% limits of agreement were relatively wide at around 5 mm, it should be remembered that this value is a robust statistical measurement of precision. In simplistic terms, the data suggest that, for any given polyp (regardless of size), 95% of the first reader's measurements will be no more than 2 mm smaller or 3 mm larger than the second reader's estimate. The absolute mean difference between readers' measurements was relatively small (approximately 0.4 mm).

We found a clear link between improved conspicuity and measurement accuracy for the higher-attenuation data sets. The conspicuity with both colonic and bone viewing windows was significantly better than it was with the other viewing windows, and that with the bone viewing window alone was superior at the highest attenuation value tested (1000 HU). This result has important clinical implications. The subsequent management of polyps detected during CT colonography is essentially governed by their size, because there is a well-established relationship between increased diameter and potential for malignancy (26). Recent guidelines from the consensus group formed at the Fourth International Symposium on Virtual Colonoscopy held in Boston, Mass, in 2003 (27) include the recommendation that polyp management strategies be highly dependent on diameter measurement. As fluid tagging becomes incorporated into mainstream CT colonographic practice, it is increasingly likely that polyps will be submerged in tagged fluid. Investigators in a recent study found that 18% of polyps may be submerged (24). Our data strongly suggest that, when the attenuation value of this fluid is greater than 500 HU, use of bone viewing windows will maximize both detection and measurement accuracy.

Extrapolation of our results suggests that, when one is searching for submerged lesions, the optimal 2D viewing window paradigm may be a combination of bone viewing window with an attenuation value of tagged fluid of approximately 700 HU (all polyps had high conspicuity grades with this combination). Clearly, in the clinical environment, it is almost impossible to ensure that an exact attenuation value is achieved in any patient. There is no consensus as to the optimal tagging regimen (barium sulfate suspension, iodinated contrast medium, or a combination of the two); moreover, the final attenuation value will depend on patient factors, such as compliance with cleansing, intestinal transit time, and volume of fluid ingested. Our data, however, at least provide some guidance as to the optimal end point for any particular tagging regimen.

Ultimately, the development of robust and accurate computer subtraction software may help to solve many of the problems associated with interpretation of tagged data sets. At the time of this writing, however, such software is limited to few vendor-specific workstations, and its use is not widespread. Furthermore, artifacts may limit clinical application (16,17), although recent experimental data suggest that accuracy of diameter measurements may be maintained after electronic subtraction (28).

Our study had limitations. We used artificially created polyps, and it cannot be assumed automatically that our results are applicable in vivo in humans. We used a porcine model, however, which we believe is perhaps more likely to reflect the real clinical situation than is a plastic phantom. We did not attempt to create flat adenomas in our specimen, and clearly future work is required. Our readers were not blinded to the polyp location (the main purpose of the study was not to investigate any association of attenuation of tagged fluid with detection). Again, further research into this issue would be worthwhile. Although we examined the effect of attenuation of tagged fluid on interobserver variability, we did not specifically address intraobserver variability, and results of investigation of the latter may have been useful. The specimen was not washed between experiments. However, with introduced barium sulfate suspensions in order of increasing attenuation, review of the data sets showed no evidence of barium coating from the previous experiment. Although the readers were fully blinded to the attenuation values of the barium suspensions tested, which were presented in random order, it is possible that readers had a subjective appreciation of the differences in the attenuation values of tagged fluid between experiments. Finally, we used a subjective classification system for assigning grades, although this assignment was performed in consensus by two experienced observers who were blinded to the identity of the data sets.

In conclusion, we found that the attenuation value of surrounding tagged fluid is significantly associated with both polyp conspicuity and measurement accuracy.

Practical application: On the basis of our results, a simple strategy would be to review areas submerged in tagged fluid by using abdominal viewing windows if the fluid attenuation value is less than 500 HU or by using bone viewing windows if the fluid attenuation value is 500 HU or greater.

	ADVANCES IN KNOWLEDGE

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION ADVANCES IN KNOWLEDGE References

 

• The attenuation value of surrounding tagged fluid and the viewing window are significantly associated with the conspicuity of submerged colonic polyps.
  
• The attenuation value of surrounding tagged fluid and the viewing window are significantly associated with the size measurement of submerged colonic polyps.
  
• Abdominal viewing windows (window level, 40 HU; window width, 400 HU) optimize polyp conspicuity when the attenuation value of tagged fluid is less than 500 HU.
  
• Bone viewing windows (window level, 500 HU; window width, 2500 HU) optimize both polyp visualization and measurement when the attenuation value of tagged fluid is 500 HU or greater.
  
• A fluid attenuation value of 700 HU viewed on bone windows may be the best viewing window paradigm for tagged data sets.
  

	ACKNOWLEDGMENTS

 
The authors thank Paul Bassett, BSc, for his statistical advice.

	FOOTNOTES

 

Abbreviations: 2D = two-dimensional

S.H., S.A.T., and D.B. are remunerated consultants for Medicsight, S.H. has a research agreement with Barco, and S.A.T. received grant support from E-Z-Em.

Author contributions: Guarantor of integrity of entire study, S.A.T.; 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.S., S.A.T.; clinical studies, S.A.T., D.B.; experimental studies, A.S., S.A.T., L.G., J.S.; statistical analysis, S.A.T.; and manuscript editing, A.S., S.A.T., D.B., S.H.

	References

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION ADVANCES IN KNOWLEDGE References

 

1. Pickhardt PJ, Choi JR, Hwang I, et al. Computed tomographic virtual colonoscopy to screen for colorectal neoplasia in asymptomatic adults. N Engl J Med 2003;349:2191–2200.[Abstract/Free Full Text]
2. Cotton PB, Durkalski VL, Pineau BC, et al. Computed tomographic colonography (virtual colonoscopy): a multicenter comparison with standard colonoscopy for detection of colorectal neoplasia. JAMA 2004;291:1713–1719.[Abstract/Free Full Text]
3. Rockey DC, Paulson E, Niedzwiecki D, et al. Analysis of air contrast barium enema, computed tomographic colonography, and colonoscopy: prospective comparison. Lancet 2005;365:305–311.[Medline]
4. Ferrucci JT. Colonoscopy: virtual and optical—another look, another view. Radiology 2005;235:13–16.[Free Full Text]
5. Iannaccone R, Laghi A, Catalano C, et al. Computed tomographic colonography without cathartic preparation for the detection of colorectal polyps. Gastroenterology 2004;127:1300–1311.[CrossRef][Medline]
6. Lefere P, Gryspeerdt S, Baekelandt M, Van Holsbeeck B. Laxative-free CT colonography. AJR Am J Roentgenol 2004;183:945–948.[Free Full Text]
7. Lefere PA, Gryspeerdt SS, Dewyspelaere J, Baekelandt M, Van Holsbeeck BG. Dietary fecal tagging as a cleansing method before CT colonography: initial results—polyp detection and patient acceptance. Radiology 2002;224:393–403.[Abstract/Free Full Text]
8. McFarland EG, Zalis ME. CT colonography: progress toward colorectal evaluation without catharsis. Gastroenterology 2004;127:1623–1626.[CrossRef][Medline]
9. Bruzzi JF, Moss AC, Brennan DD, MacMathuna P, Fenlon HM. Colonic surveillance by CT colonography using axial images only. Eur Radiol 2004;14:763–767.[CrossRef][Medline]
10. Fenlon HM, Clarke PD, Ferrucci JT. Virtual colonoscopy: imaging features with colonoscopic correlation. AJR Am J Roentgenol 1998;170:1303–1309.[Free Full Text]
11. Fletcher JG, Johnson CD, MacCarty RL, Welch TJ, Reed JE, Hara AK. CT colonography: potential pitfalls and problem-solving techniques. AJR Am J Roentgenol 1999;172:1271–1278.[Free Full Text]
12. McFarland EG. Reader strategies for CT colonography. Abdom Imaging 2002;27:275–283.[Medline]
13. Fidler JL, Johnson CD, MacCarty RL, Welch TJ, Hara AK, Harmsen WS. Detection of flat lesions in the colon with CT colonography. Abdom Imaging 2002;27:292–300.[Medline]
14. Burling D, Halligan S, Taylor SA, Brennand D. Polyp size measurement by CT colonography: effect of viewing conditions on inter- and intra-observer agreement, and agreement with colonoscopy (abstr). In: Radiological Society of North America Scientific Assembly and Annual Meeting Program. Oak Brook, Ill: Radiological Society of North America, 2004; 617.
15. Zalis ME, Hahn PF. Digital subtraction bowel cleansing in CT colonography. AJR Am J Roentgenol 2001;176:646–648.[Free Full Text]
16. Zalis ME, Perumpillichira J, Del Frate C, Hahn PF. CT colonography: digital subtraction bowel cleansing with mucosal reconstruction—initial observations. Radiology 2003;226:911–917.[Abstract/Free Full Text]
17. Pickhardt PJ, Choi JH. Electronic cleansing and stool tagging in CT colonography: advantages and pitfalls with primary three-dimensional evaluation. AJR Am J Roentgenol 2003;181:799–805.[Free Full Text]
18. Dachman AH, Lieberman J, Osnis RB, et al. Small simulated polyps in pig colon: sensitivity of CT virtual colography. Radiology 1997;203:427–430.[Abstract/Free Full Text]
19. Fletcher JG, Johnson CD, Welch TJ, et al. Optimization of CT colonography technique: prospective trial in 180 patients. Radiology 2000;216:704–711.[Abstract/Free Full Text]
20. Thomeer M, Carbone I, Bosmans H, et al. Stool tagging applied in thin-slice multidetector computed tomography colonography. J Comput Assist Tomogr 2003;27:132–139.[CrossRef][Medline]
21. Callstrom MR, Johnson CD, Fletcher JG, et al. CT colonography without cathartic preparation: feasibility study. Radiology 2001;219:693–698.[Abstract/Free Full Text]
22. Taylor SA, Halligan S, Bartram CI. CT colonography: methods, pathology and pitfalls. Clin Radiol 2003;58:179–190.[CrossRef][Medline]
23. Fenlon HM, Ferrucci JT. First International Symposium on Virtual Colonoscopy. AJR Am J Roentgenol 1999;173:565–569.[Free Full Text]
24. Franaszek M, Summers RM, Pickhardt PJ, Choi JR, Schindler WR. Assessment of obscured colonic surface in CT colonography (abstr). In: Radiological Society of North America Scientific Assembly and Annual Meeting Program. Oak Brook, Ill: Radiological Society of North America, 2004; 618.
25. Lefere P, Gryspeerdt S, Marrannes J, Baekelandt M, Van Holsbeeck B. CT colonography after fecal tagging with a reduced cathartic cleansing and a reduced volume of barium. AJR Am J Roentgenol 2005;184:1836–1842.[Abstract/Free Full Text]
26. Muto T, Bussey HJ, Morson BC. The evolution of cancer of the colon and rectum. Cancer 1975;36:2251–2270.[Medline]
27. Zalis ME, Barish MA, Choi JR, et al. CT colonography reporting and data system: a consensus proposal. Radiology 2005;236:3–9.[Free Full Text]
28. Zalis ME, Perumpillichira JJ, Kim JY, Del Frate C, Magee C, Hahn PF. Polyp size at CT colonography after electronic subtraction cleansing in an anthropomorphic colon phantom. Radiology 2005;236:118–124.[Abstract/Free Full Text]

This article has been cited by other articles:

				A. H. de Vries, H. W. Venema, J. Florie, C. Y. Nio, and J. Stoker Influence of Tagged Fecal Material on Detectability of Colorectal Polyps at CT: Phantom Study Am. J. Roentgenol., October 1, 2008; 191(4): W181 - W189. [Abstract] [Full Text] [PDF]

This Article Abstract Figures Only Full Text (PDF) Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Slater, A. Articles by Halligan, S. Search for Related Content PubMed PubMed Citation Articles by Slater, A. Articles by Halligan, S.