Study On Scoring The Risk Level Of Developing Coronary Artery Lesions Secondary To Kawasaki Disease In Chinese Children

Part I:A Meta-analysis of the biomarkers associated with coronary artery lesions secondary to Kawasaki disease in Chinese childrenObjective:To determine the biomarkers associated with coronary artery lesions(CAL) secondary to Kawasaki disease (KD) in Chinese children by the way of meta-analysis.Methods:We searched documents published from January 1997 to December 2009 from medical electronic databases; According to inclusion and exclusion criteria, eligible full texts were identified. All children with Kawasaki disease were divided into CAL group and NCAL group according to weather complicated with CAL. We conducted a comprehensive quantitative analysis by Statal0.0 statistical software to assess heterogeneity among documents, calculate the summary effects and analyze publication, sensitivity.Results:A total of 92 documents and 16 biomarkers were identified. All documents were case-control studies, and included 2398 patients in CAL group and 5932 patients in NCAL group. The meta-analysis showed that platelet count, neutrophil count, platelet hematocrit, platelet distribution width, mean platelet volume, erythrocyte sedimentation rate, cardiac troponin I, endothelin-1 of CAL group were significantly higher than NCAL group's. Serum albumin and hemoglobin were significantly lower in CAL group (all P values<0.05). White blood cell count, serum sodium, matrix metalloproteinase 9, total cholesterol, CD3+T lymphocytes percentage have no statistically significant difference between the two groups.Conclusion:Our results indicated that the 10 biomarkers of platelet count, nertrophil count, platelet hematocrit, platelet distribution width, mean platelet volume, erythrocyte sedimentation rate, cardiac troponia I, endothelin-1, albumin, hemoglobin are associated with coronary artery lesions secondary to Kawasaki disease in Chinese children. Furthermore, these biomarkers may contribute to the pathological process of CAL; The levels of white blood cell count, serum sodium, matrix metalloproteinase-9, total cholesterol, platelet hematocrit, and CD3+T lymphocytes percentage do not have a relationship with the occurrence of CAL among Chinese children with Kawasaki disease. PartⅡ:Study on scoring the risk level of developing coronary artery lesions secondary to Kawasaki disease in Chinese childrenObjective:To determine the risk level of developing coronary artery lesions(CAL) secondary to Kawasaki disease in Chinese children, based on the biomarkers'test values before initial treatment.Methods:We reviewed the continuous clinical records of a total of 113 Kawasaki disease patients, which included 43 children in CAL group and 70 children in NCAL group, in Union Hospital of Tongji Medical College of HUST during January 2007 to December 2010. Receiver operating characteristics(ROC) curves were drawn to determine the best cut-off levels of 9 biomarkers including platelet count, neutrophil count, platelet hematocrit, platelet distribution width, mean platelet volume, erythrocyte sedimentation rate, serum albumin, hemoglobin and C-reaction protein, according to the test values of the 9 biomarkers before initial treatment. To identify risk factors of CAL secondary to KD in Chinese children, a multivariable logistic regression analysis was conducted to build a logistic regression model. A simple scoring system was build for determining the risk level of developing CAL secondary to KD in Chinese children by assigning integer score points to all risk factors. The best discriminatory capacity of the simple scoring models was assessed by the area under the ROC curves.Results:The cut-off values of the 9 biomarkers were 564(109/L),10.1(109/L),0.415(%), 10.05(%),7.25(fL),81.5(mm/h),31.5(g/L),100.50(g/L),14.4(mg/L), respectively. Six risk factors of developing CAL secondary to Kawasaki disease in Chinese children including gender, age, neutrophil count, platelet hematocrit, serum albumin and C-reaction protein, were selected by multivariable logistic regression analysis; the logistic regression model was established, P=ey/(1+ey),y=1.1149×gender+1.1934×age+1.6443×neutrophil count+1.3708×platelet hematocrit+1.0543×albumin+1.0036×C-reaction protein. A simple scoring system was established on the basis of the logistic regression model:male, 1.1 points; age≤1 year,1.2 points; neutrophil count>10.1×109/L,1.6 points; PCT>0.415%,1.4 points; CRP>14.4mg/L,1 point; ALB<31.5g/L,1 point. Total scores were 7.3 points,0-2.9 points were classified to low risk group,3.0-7.3 points were classified to high risk group. The areas below ROC curves of the simple scoring system was 0.85, sensitivity and specificity were 81.4% and 78.6%, consistent rate with actual situation that CAL complicated with KD was 79.6%.Conclusion:Our scoring system established in this study showed high sensitivity and specificity, and has important clinical value.