Practical fecal calprotectin cut-off value for Japanese patients with ulcerative colitis

AIM To determine appropriate fecal calprotectin cut-off values for the prediction of endoscopic and histologic remission in Japanese patients with ulcerative colitis(UC). METHODS We performed a cross-sectional observational study of 131 Japanese patients with UC and measured fecal calprotectin levels by fluorescence enzyme immunoassay. The clinical activity of UC was assessed with the partial Mayo score(PMS). Relapse was defined as increase of PMS by 2 points or more in stool frequency or rectal bleeding subscore. The endoscopic and histologic activities of UC were evaluated in 50 patients within a 2-mo period from fecal sampling. Endoscopic activity was determined by Mayo endoscopic subscore, Rachmilewitz endoscopic index, and ulcerative colitis endoscopic index of severity. The histologic grade of inflammation was evaluated with biopsy specimens obtained from the endoscopically most severely inflamed site, according to the scheme by Matts grade and Riley's score.RESULTS Fecal calprotectin levels varied from 1-20783 μg/g. There was a significant correlation between the partial Mayo score and fecal calprotectin levels(r = 0.548, P < 0.001). In 50 patients who underwent colonoscopy with biopsy, levels were significantly correlated with the Mayo endoscopic subscore(r = 0.574, P < 0.001), Rachmilewitz endoscopic index(r = 0.628, P < 0.001), ulcerative colitis endoscopic index of severity(r = 0.613, P < 0.001), Riley's histologic score(r = 0.400, P = 0.006), and Matts grade(r = 0.586, P < 0.001). Receiver-operating characteristic analyses identified the best cut-off value for the prediction of endoscopic remission as 288 μg/g, with an area under the curve of 0.777 or 0.823, while that for histologic remission was 123 or 125 μg/g, with an AUC of 0.881 or 0918, respectively. Of the 131 study patients, 88 patients in clinical remission were followed up 6 mo. During the follow-up period, 19 patients relapsed. The best fecal calprotectin cut-off value for predicting relapse was 175 μg/g.CONCLUSION Fecal calprotectin is a predictive biomarker for endoscopic and histologic remission in Japanese patients with UC.

Association between white opaque substance under magnifying colonoscopy and lipid droplets in colorectal epithelial neoplasms

AIM To examine the association between white opaque substance(WOS) and histologically verified lipiddroplets in colorectal epithelial neoplasms.METHODS We reviewed colonoscopy records at our institution from 2014 to 2016 and identified cases of endoscopically or surgically resected colorectal epithelial neoplasms observed by magnifying narrow-band imaging(M-NBI) colonoscopy. Immunohistochemistry was used to stain tumors with a monoclonal antibody specific to adipophilin as a marker of lipids. The expression and distribution of adipophilin were compared between WOS-positive and WOS-negative lesions and among tumors classified by histologic type and depth of invasion.RESULTS Under M-NBI colonoscopy, 81 lesions were positive for WOS and 48 lesions were negative for WOS. The rate of adipophilin expression was significantly higher in WOS-positive lesions(95.1%) than in WOS-negative lesions(68.7%)(P = 0.0001). The incidence of deep adipophilin expression was higher in WOS-positive lesions(24.7%) than in WOS-negative lesions(4.2%)(P = 0.001). The incidence of deep expression was predominant among cancers with massive submucosal invasion(62.5%) compared to adenoma(7.2%) and high-grade dysplasia or cancers with slight submucosal invasion(12.7%)(P = 0.0001).CONCLUSION The distribution of lipid droplets may be closely associated with the visibility of WOS under M-NBI colonoscopy, and with histologic grade and depth of tumor invasion.