ABSTRACT
Background: Assessment of disease severity of ulcerative colitis (UC) is crucial for the decision of drug treatment strategy, the appraisal of response to drugs, and the prognosis evaluation. Aims: To establish and validate a discriminant model for the quantification of disease severity of UC. Methods: The clinical data of 169 active UC cases admitted at the Beijing Haidian Hospital were analyzed retrospectively. Based on Truelove-Witts criteria, seven widely accepted indices including temperature, pulse, defecation frequency, hematochezia, hemoglobin (Hb), erythrocyte sedimentation rate (ESR) and albumin (Alb) were selected and a stepwise discriminant analysis was performed to establish a discriminant model using Fisher's linear discriminant function, and then the discriminant model was validated from different aspects. Results: The discriminant model: F1 (mild)=-51.109+0.587X1+0.295X2+0.379X3+0.071X4; F2 (moderate)=-49.905+0.624X1+0.512X2+0.333X3+0.095X4; F3 (severe)=-57.294+0.709X1+0.599X2+0.296X3+0.141X4 [ X1, pulse (beats /min); X2, defecation frequency (times/d); X3, Hb (g/L); X4, ESR (mm/h)]. The result of the newly established discriminant model was moderately correlated with the Truelove-Witts criteria (r
ABSTRACT
0.1) were not significantly different.Conclusion The seropositivity of H.pylori,anti-CagA and anti-H.pylori-IgG are associated with the risk of gastric cancer.