Search

Objective To establish a risk prediction model for early identification of patients with severe bee stings, so as to provide a quantitative basis for whether patients with bee stings admitted to the emergency department should choose intensive treatment in the Intensive Care Unit (ICU).Methods Clinical data of 402 patients with bee stings hospitalized from September 15,2012 to December 31,2019 were collected as the modeling group, and clinical data of 63 patients with bee stings hospitalized from January 1,2020 to May 30,2022 were collected as the verification group. The 402 cases in the modeling group were divided into 343 cases in the general ward group and 59 cases in the ICU group according to whether they were admitted to ICU within 48 hours. Univariate analysis and logistic regression were used to conduct retrospective analysis to screen out the independent risk factors affecting the admission of patients with bee stings to ICU, and establish a risk prediction model of severe disease and conduct the risk stratification. Meanwhile, according to the established model, 63 cases in the validation group（41 cases in the general ward group and 22 cases in the ICU group） were scored and the receiver operating characteristic (ROC) curve was drawn to verify the effectiveness of the model. Results Logistic regression analysis showed that the number of stings, systemic allergic rash symptoms, soy-colored urine and white blood cell number were independent risk factors for patients with bee stings to admitted to ICU (number of bee stings: Z=-6.603,P<0.001;systemic allergic rash symptoms: χ~2=82.679,P<0.001;soy-color urine: χ~2=147.723,P<0.001;white blood cell count: Z=-8.941,P<0.001）,and the final early risk prediction model of severe bee stings was 0.093×A+2.965×B+2.647×C+0.134×D（A was the number of stings, B was one if there was systemic allergic rash, B was zero if there was no systemic allergic rash, C was one if there was soy-colored urine, and C without soy-colored urine was zero. D was white blood cell count/10~9）.The area under curve (AUC) of the ROC curve of the model was 0.960,the sensitivity was 0.915 and the specificity was 0.901.The AUC, sensitivity and specificity of the ROC curve of the validation group were 0.952,0.96 and 0.88,respectively.Conclusion The number of bee stings, systemic allergic rash symptoms, soy-colored urine and white blood cell number are independent risk factors affecting the admission of patients with bee stings to ICU. The established early risk prediction model of severe bee stings has good discriminant validity and application value, and can better predict the admission risk of patients with bee stings to ICU. The model can provide a certain reference for the early evaluation of the condition of bee stings and the guidance of stratified treatment. [ABSTRACT FROM AUTHOR]