Reply to K.-I. Fu et al.

We read with interest the reply by Fu at al. to our study. We would like to congratulate them on their excellent work. It is interesting that their results without magnification using conventional colonoscopy with chromoendoscopy are similar to our study and thus validate our results. They state in their letter that they achieved an accuracy of approximately 90 % using chromoendoscopy and conventional colonoscopy.

Fu et al. comment on two issues concerning our study:

An unexpectedly high number of nonneoplastic lesions were classified endoscopically as nonneoplastic. Of the 192 nonneoplastic lesions in the study, 5 could not be retrieved for histological analysis. Of the 187 remaining, 153 (82 %) were correctly classified as nonneoplastic, achieving a specificity of 82 % (95 % CI 76 % - 87 %) and a negative predictive value of 78 % (95 % CI 72 % - 84 %). It is always better to overdiagnose a nonneoplastic lesion as neoplastic rather than the other way round. Interobserver variation. Two endoscopists (BJR and SAS) performed endoscopies and classified lesions. BJR is a fully trained endoscopist and SAS performed the colonoscopies under his direct supervision. Hence, we are unable to perform agreement statistics, as our methodology would be biased.

The authors conclude that the use of chromoendoscopy alone is “insufficient” and that an unacceptable number of lesions would be misdiagnosed. However, their data do not fully support this conclusion. In their larger and more recent study from 2004 [1], colonoscopy with dye spraying allowed accurate characterization of 184 of 206 colonic lesions. With the use of a dye spraying and magnification, they were able to correctly distinguish a further 13 lesions. The increased characterization of another 13 colonic polyps in a study population of nearly 1000 patients may be regarded as insignificant and of insufficient magnitude to justify the extra expense of purchasing a magnifying colonoscope.

References

• 1 Fu K I, Sano Y, Kato S. et al . Chromoendoscopy using indigo carmine dye spraying with magnifying observation is the most reliable method for differential diagnosis between non-neoplastic and neoplastic colorectal lesions: a prospective study.  Endoscopy. 2004;  36 1089-1093