Colonoscopy making or not use of chromoscopy methods on the diagnosis of flat, depressed and augmented colorectal lesions

Colorectal cancer (CRC) is one of the largest causes of death on the industrialized world. Its annual incidence of 800.000 new cases means 8,5% of all the new ones and 12% of deaths related to this disease. In Brazil, excluding the non-melanoma skin-cancers, CRC is the fourth more frequent among men and the third one among women. The risk for developping CRC is approximately of 5 to 6% on the Western population. There are epidemiological evidences for reducing CRC on 60-90% when colonoscopy with polypectomy is used preventively. Colonoscopy is still the best method both for the early dyagnosis of CRC and precursor lesions. However, there are non-contemptible failures on the detection. This paper purpose was comparing the result of colon and rectum mucous membrane detailed test through conventional colonoscopy, chromoendoscopy and NBI, on the detection of augmented, depressed and flat lesions in patients submitted to it without any personal or familiar antecedents. Between January 2007 and October 2009 181 patients were selected randomically and divided into 3 groups: A: 48 control patients; B: 29 patients, NBI; C: 104 patients, diffuse chromoscopy. It is observed that, from the 181 examined patients, 38 (21%) didn’t present lesions. The 143 patients with lesion, presented an average number of 2,65 lesions, with a minimum of 1 and a maximum of 7 lesions. On the total of the 181 patients and on the whole of the 143 patients with lesions it was not observed any statistically significant difference among the three groups A, B and C as for Age, the Rectum-Cecum Time and the Cecum-Rectum Time, while there was a statistical variation for Height, Weight and consequent bmi. The average size of the 379 lesions found on the 143 patients, assessed by its diameter was of 5,45 mm (2.14 in.) + 2,84 mm (1,11 in), without any statistical variation among the groups, among the hemicolons and among the hemicolons in the groups. The size of the lesions were gathered into three distinct intervals: up to 5 mm [1.9 in.] (76,30%), from 6 mm [2.3 in.] to 10 mm [3.9 in] (19,50%) and from 11 to 20 mm [4.3 to 7.8 in] (4,20%). From the total of 379 lesions, 203 (53,6%) revealed themselves neoplastic and 176 (46,4%) non-neoplastic. The average size of the 203 neoplastic lesions was of 5,96 mm (2.34 in.), and of the 176 non-neoplastic ones, 4,87 mm [12,36 in]. Neoplastic lesions have shown larger than the non-neoplastic ones, with a stastistical significance. On the groups there is any significant variation between neoplasia and non-neoplasia, but a significant difference between the neoplasias and non-neoplasias size. There was any statistical difference among the lesion size on both hemicolons, however, a significant difference among the sizes of neoplastic and non-neoplastic ones. The same is observed when colon segments were analyzed individually. The two segments that have presented significant lesions, specifically on what concerns the size of neoplastic and non-neoplastic ones were the Sigmoid and the Transverse. It is observed that all the subpediculated lesions and the flat-augmented ones with a central depression were neoplastics. The flat and neoplastic lesions are proportionally more visible on the right hemicolon at groups B (85,7%) and C (67,9%), without any statistical difference. The diagnostic hypotheses of the lesions grown during the colonoscopic test were compared to the histopathological results. On control group (A) it is observed a 82,7% sensibility, a 59% specificity with a concordance rate of 72,5%, considered regular. On group B it is observed a 92,3% sensibility, a 61,9% specificity, with a regular concordance rate of 78,7%. On group C it is observed a 88,8% sensibility, a 79,3% specificity, a 84,2% concordance rate, considered good. Proportionally group C has a larger number of patients with 3 or more lesions and or more neoplastic lesions, but with no statistical value. On what concerns general data, it is concluded that there wasn’t any statistical difference among the 181 patients examined and the 143 ones presenting lesions as for the relative number, the type and size of the lesions. Neoplastic lesions appear to be larger when compared to non-neoplastic ones, with a statistical significance. The concordance between the colonoscopic diagnostic hypothesis and the histology is larger on chromoscopy group.