Send letter to journal:
Re: Author's reply

Dear Editor

We read with interest the eLetter of Laghi et al. (14 August 2002).

We agree with the first comment of Laghi et al. concerning the experience of the investigators. At the beginning of this study, our experience was limited to the cases done for the study published in Gut[1] and to a couple of multidetector computed tomography colonography (MDCT- colonography) in patients with incomplete conventional colonoscopy. We were entirely aware about the limits of our clinical experience in CT- colonography.[1,2] Independent of this fact, we were able to demonstrate the advantages of MDCT colonography in the detection of colorectal lesions. We do not agree with the second comment in the eLetter of Laghi et al. As described in the methods section the image reconstructions were performed with a slice thickness of 2.5 mm with reconstruction intervals of 2 mm (overlap 0.5 mm) that means that we used an optimal technical setting. But it is true that we have been using only 5 mm slices in our first study.[1] Aware about our limited experience we included in our paper a retrospective analysis of all missed lesions. If we add lesions not detected due to perception errors to our first view results we reach sensitivities as good as those of more experienced investigators:[3,4] lesions 10 and more mm 100 % sensitivity, lesions 6 to 9 mm 80 % sensitivity.

In conclusion, we showed clearly that the moderate results in detection of small polyps is due to the limited experience and not to intrinsic limitations of MDCT colonography.

References

(1) Pescatore P, Gluecker T, Delarive J, et al. Diagnostic accuracy and interobserver agreement of CT colonography (virtual colonoscopy). Gut 2000; 47: 126-30.

(2) Gluecker T, Meuwly JY, Pescatore P, Schnyder P, Delarive J, Jornod P, Meuli R, Dorta G. Effect of investigator experience in CT colonography. Eur Radiol 2002;12:1405-9.

(3) Laghi A, Iannaccone R, Carbone I, et al. Computed tomographic colonography (virtual colonoscopy): blinded prospective comparison with conventional colonoscopy for the detection of colorectal neoplasia. Endoscopy 2002;34: 441-6.

(4) Macari M, Bini EJ, Xue X et al. Colorectal Neoplasms: Prospective Comparison of Thin-Section Low-Dose Multi-Detector Row CT Colonography and Conventional Colonoscopy for Detection. Radiology 2002;224: 383-92.

Send letter to journal:
Re: Performance of multidetector computed tomography colonography in the detection of colorectal lesions

Dear Editor

We read with interest the recent article by Gluecker et al. (Gut 2002; 51: 207-211) investigating the performance of multidetector computed tomography (MDCT) colonography in the detection of colorectal lesions in comparison with that of conventional colonoscopy. The authors concluded that MDCT colonography provides good sensitivity and specificity for the detection of colonic lesions larger than 10 mm in diameter, but certain weaknesses in the study design should be emphasised. We would like to comment further on two of these weaknesses, as they are potentially important in interpreting the results of Gluecker et al. (Gut 2002; 51: 207 -211).

First, the evaluation of MDCT colonographic data sets was performed by two observers with little (radiologist with experience of approximately 60 CT colonographic examinations) or no (gastroenterologist) experience in the procedure. We believe that this represents a major limitation to the study. It is already well known from the literature [1,2] that data interpretation of CT colonography requires a long learning process. The authors too acknowledged this key issue in the manuscript (Gluecker et al. Gut 2002; 51: 207-211). Indeed, the first reading of the MDCT colonography images identified 1/41 lesions 5 mm in diameter or smaller, 5/15 lesions between 6 and 9 mm, and 9/11 lesions 10 mm or larger. However, retrospective evaluation of the same data (i.e., a second reading) already allowed the authors to detect many of the lesions missed in the first reading session (18/40 lesions 5 mm or smaller; 7/10 lesions between 6 and 9 mm; 2/2 lesions 10 mm or larger). All lesions identified retrospectively were missed due to perceptive errors. Although perceptive errors may occur with MDCT, the high incidence of this type of errors in the study by Gluecker et al. (Gut 2002;51: 207-211) is best explained with the low experience of the observers, and should not be construed as an intrinsic limitation of MDCT colonography.

Second, it should be noted that one of the major advantages introduced by MDCT technology is represented by the ability of combining thin beam collimation (1 or 2.5 mm) with short acquisition times.[3] The use of 5 mm beam collimation considerably reduces this advantage and could explain the poor performance of MDCT in the detection of small lesions (lesions 6-9 mm, sensitivity 33 %; lesions 5 mm or smaller, sensitivity 4 %) in the hands of the authors. In particular, the use of a thin beam collimation would have significantly increased the spatial resolution of CT images with consequent improvement in image quality, while still allowing for short acquisition time and only slightly higher radiation exposure delivered to patients.[3,4]

In recent years, several studies have demonstrated that CT colonography has a performance similar to that of conventional colonoscopy in the detection of colorectal lesions 6 mm in diameter or larger.[5-7] In particular, with regard to CT colonography as peformed with thin-beam- collimation MDCT, both published[8,9] and unpublished [Laghi A, et al. Multi-slice spiral CT colonography for the detection of colorectal polyps and neoplasms, RSNA 2001, personal communication] data have shown that MDCT colonography can increase the detection rate of small lesions. In addition, the use of thin-beam-collimation MDCT colonography can reduce the number of false positive findings by enabling better assessment of the morphology and CT density of suspected lesions, thus allowing to better differentiate residual stool or hypertrophic colonic folds from colorectal lesions.[10]

We therefore believe that MDCT colonography can reliably detect lesions smaller than 10 mm (in particular those in the 6 to 9 mm range), provided that experienced readers interpret the results and that the technological potential of the current devices is fully exploited with the use of thin-beam-collimation protocols.

References

(1) Pescatore P, Gluecker T, Delarive J, et al. Diagnostic accuracy and interobserver agreement of CT colonography (virtual colonoscopy). Gut 2000; 47: 126-30.

(2) Spinzi G, Belloni G, Martegani A, et al. Computed tomographic colonography and conventional colonoscopy for colon diseases: a prospective, blinded study. Am J Gastroenterol 2001; 96: 394-400.

(3) Laghi A, Iannaccone R, Panebianco V, et al. Multislice CT colonography: technical developments. Semin Ultrasound CT MR 2001; 22: 425 -31.

(4) Laghi A, Iannaccone R, Mangiapane F, et al. Experimental colonic phantom for the evaluation of the optimal scanning technique for CT colonography using a multidetector spiral CT equipment. Eur Radiol 2002 (in press).

(5) Fenlon HM, Nunes DP, Schroy PC, et al. A comparison of virtual and conventional colonoscopy for the detection of colorectal polyps. N Engl J Med 1999; 341: 1496-503.

(6) Laghi A, Iannaccone R, Carbone I, et al. Computed tomographic colonography (virtual colonoscopy): blinded prospective comparison with conventional colonoscopy for the detection of colorectal neoplasia. Endoscopy 2002; 34: 441-6.

(7) Laghi A, Iannaccone R, Carbone I, et al. Detection of colorectal lesions with virtual computed tomographic colonography. Am J Surg 2002; 183: 124-31.

(8) Rogalla P, Meiri N, Rückert JC, et al. Colonography using multislice CT. Eur J Radiol 2000; 36: 81-5.

(9) Wessling J, Fischbach R, Domagk D, et al. Colorectal polyps: Detection with multi-slice CT colonography. Rofo Fortschr Geb Rontgenstr Neuen Bildgeb Verfahr 2001; 173: 1069-71.

(10) Macari M, Bini EJ, Xue X, et al. Colorectal neoplasms: prospective comparison of thin-section low-dose multi-detector row CT colonography and conventional colonoscopy for detection. Radiology 2002 (in press).