Study on the effectiveness of the proportion of inflammatory cells in the lamina propria of intestinal mucosa in predicting mucosal histological healing in patients with ulcerative colitis / 中华消化杂志

ABSTRACT

Objective:To determine a simpler and more practical scoring standard for predicting mucosal histological healing in ulcerative colitis (UC).Methods:From April 11, 2017 to February 8, 2021, 68 UC patients diagnosed with mucosal healing under endoscopy and hospitalized at Department of Gastroenterology, the Tenth People′s Hospital of Tongji University and during the same period 60 healthy individuals who underwent endoscopy for health checkup were retrospectively analyzed. Modified Mayo score and ulcerative colitis endoscopic index of severity (UCEIS), the modified Nancy index and Robarts histopathology index were determined based on the collected clinical data, endoscopic reports and histopathological evaluation. The proportions of neutrophils, eosinophils, and plasma cells in the colonic mucosal lamina propria were calculated. The proportions of activated neutrophils and T cells in the colonic mucosal lamina were calculated according to CD177 and CD40L, respectively. The new clinical and laboratory diagnostic formulas were determined by multivariate logistic regression analysis, the effectiveness of the equations was evaluated by receiver operating characteristic curve (ROC).Results:Among the 68 patients with UC, the modified Mayo score was 0.7 (0.4, 1.1), the UCEIS was 0.5 (0.3, 0.8), the Nancy index was 5.9±3.2, and the Robarts histopathology index was 2.6±1.7. According to multivariate logistic regression analysis, the formula for clinical diagnosis of histological healing was Y1=-21.09+ 355.9 X1+ 305.8 X2+ 44.91 X3 ( X1, X2 and X3 were the proportions of neutrophils, eosinophils, and plasma cells, respectively). The results of ROC analysis indicated that Y1<-0.747 was the cut-off value of diagnosis of histological healing, and the area under the curve (AUC) was 0.986 and 95% confidence interval ( CI) was 0.922 to 1.000 ( P<0.001), the sensitivity was 97.10% and the specificity was 91.20%. The formula of laboratory diagnosis of histological healing was Y2=-10.57+ 469.1 X1 + 132.7 X2 + 101.2 X3 + 18.56 X4 ( X1, X2, X3, and X4 were the proportions of CD177 + neutrophils, eosinophils, CD40L + T cells and plasma cells, respectively). The results of ROC analysis indicated that Y2<1.960 was the cut-off value of diagnosis of histological healing, and the AUC was 0.980, 95% CI was 0.913 to 0.999 ( P<0.001), the sensitivity was 84.78%, and the specificity was 100.00%. The new clinical and laboratory diagnostic criteria were positively correlated with the Nancy histological index ( r=0.411 and 0.308, P=0.001 and 0.011), and Robarts histopathology index ( r=0.311, 0.273, P=0.010 and 0.024). Conclusions:Compared with the Nancy index, the new clinical and laboratory diagnostic criteria are simpler and more practical. The new clinical diagnostic formula Y1<-0.747 and the new laboratory diagnosis formula Y2<1.960 are the independent factors for predicting histological healing in UC patients.