[Assessment of myocardial injury following anthracycline chemotherapy in children with leukemia by echocardiography].
AI interpretation is pending for this paper.
Open original publication →What the AI sees
Not AI summarized yet.
Research significance
Pending deeper interpretation.
Source abstract
OBJECTIVES: To evaluate the clinical value of echocardiography in assessing myocardial injury following anthracycline chemotherapy in children with leukemia. METHODS: A total of 267 children with leukemia who received anthracycline-based chemotherapy at Suzhou Wujiang District Children's Hospital from June 2020 to May 2025 were enrolled. Based on cardiac troponin I levels, the children were divided into a myocardial injury group (n=81) and a non-myocardial injury group (n=186). Clinical data, myocardial injury markers, and echocardiographic parameters were compared between groups. A random forest algorithm was used to screen for influencing factors, and multivariable logistic regression was employed to identify independent predictors. Receiver operating characteristic curve analysis was used to evaluate the predictive performance of individual and combined factors, and the DeLong test was employed to compare combined diagnostic performance across sex- and age-stratified subgroups. RESULTS: Random forest analysis indicated the lowest out-of-bag error rate when the model included left ventricular ejection fraction (LVEF), cumulative anthracycline dose, left ventricular end-diastolic volume index, left ventricular end-systolic volume index, E-wave velocity, A-wave velocity, and E-wave velocity/A-wave velocity (E/A) ratio. Multivariable logistic regression showed that cumulative anthracycline dose (OR=8.998, 95%CI: 4.759-15.013), and pre-chemotherapy LVEF (OR=7.714, 95%CI: 3.985-14.932), E/A ratio (OR=12.207, 95%CI: 6.269-23.769), E-wave velocity (OR=5.842, 95%CI: 2.959-11.532), and A-wave velocity (OR=7.114, 95%CI: 3.913-12.933) were independent predictors of myocardial injury following chemotherapy (P<0.05). Receiver operating characteristic curve analysis showed that the combined diagnostic model achieved superior predictive performance compared with any single indicator, with an area under the curve of 0.936, sensitivity of 0.820, and specificity of 0.917; predictive efficacy was maintained in both sex- and age-stratified subgroups. CONCLUSIONS: A combined model integrating echocardiographic parameters (LVEF, E-wave velocity, A-wave velocity, and E/A ratio) with cumulative anthracycline dose demonstrates high predictive efficacy for post-chemotherapy myocardial injury in children with leukemia and is applicable across sexes and age groups.