Diagnosis Of Coronary Artery Lesions In Children With Kawasaki Disease Based On Z-Score Regression Model

Objective: To establish the most suitable formula for calculating coronary artery Zscore and to plot the Z-score curve for children in Sichuan by studying the relationship of coronary artery internal diameter with their age,height,weight,body surface area in healthy children.By comparing the detection rate of coronary artery lesions in Kawasaki disease with different echocardiographic diagnostic methods,in order to provide a basis for early diagnosis,timely treatment and optimal follow-up management plan for clinical Kawasaki disease.Methods: 1.Modeling group: We included healthy children who were examined in Sichuan Provincial People’s Hospital from January 2020 to December 2022,collecting the information of their gender,age,height and weight,calculating their body surface area by Haycock’s formula,and measured coronary artery internal diameter values by using echocardiography.Pearson correlation analysis was used to identify the highest correlation coefficient between coronary artery internal diameter values and each physiological index and regression models were constructed through various function transformations.Then,the regression model with the best coefficient of determination and the smallest mean square error was found to establish the Z-score calculation formula,and the Z-score curve was drawn.2.Validation group: We included children who were hospitalized at Sichuan Provincial People’s Hospital with a diagnosis of Kawasaki disease from January 2018 to December 2022,collecting information on age,sex,height,weight,and coronary artery internal diameter values measured by echocardiography,and compared the detection rate of coronary artery lesions by different methods in children with Kawasaki disease.Results: 1.The differences in the mean values of each coronary artery internal diameter were statistically significant among all age groups(all P < 0.01),and each coronary artery internal diameter increased with the age of children.Pearson correlation analysis revealed that the coronary artery internal diameter values were significantly correlated with age,height,weight,and body surface area(r value range 0.684 to 0.815),with the highest correlation coefficient for body surface area.Among them,the correlation coefficients of left main coronary artery,left anterior descending,left circumflex branch,right coronary artery internal diameter and body surface area were0.815,0.793,0.704,0.802,respectively.The body surface area was selected and transformed by various functions to construct the regression model,among which the power function model was designed with fewer parameters,simple form,best determination coefficient and better regression effect,so the power function model was used as the most suitable regression model to calculate the Z-score formula and plot the Z-core curve.2.The detection rate of coronary artery lesions in children with Kawasaki disease was evaluated using different echocardiographic diagnostic methods,and the results showed differences in the detection rate of different echocardiographic diagnostic methods.Conclusion: The Z-score formula standardized via body surface area and established by using a power function regression model reduces the influence of growth and development on coronary arteries in children and can better reflect the real situation of coronary artery internal diameter,which is conducive to early diagnosis,timely treatment and follow-up management of clinical Kawasaki disease.