Objective To explore the risk factors for anthracycline related acute cardiotoxicity (ACT) in pediatric patients with hematological tumors, and to construct a column chart risk prediction model.
Methods 184 children with hematological tumors in a tertiary hospital from January 2016 to December 2023 were selected as the research subjects. The children were divided into a model group (129 cases) and a validation group (55 cases) in a 7:3 ratio. The model group was used to construct a risk prediction model, while the validation group was used to verify the accuracy of the model's predictions. Using multiple logistic regression analysis to explore the independent influencing factors of anthracycline related acute ACT in children with hematological tumors, and further establishing a risk column chart prediction model using R software.
Results Age, Targeted therapy, TC, and LDL-H were independent risk factors for anthracycline related acute ACT in children with hematological tumors (all P<0.05), while the combination of dexamethasone and dexamethasone was an independent protective factor for anthracycline related acute ACT in children with hematological tumors (P<0.05). A column chart prediction model was constructed using the above influencing factors as indicators. The area under the working characteristic curve (AUC) of the subjects in the model group was 0.804 [95% CI (0.725, 0.869), P<0.001]. The results of the Hosmer Lemeshow goodness of fit test were χ2=9.448, P=0.306, AUC in the validation group was 0.738 [95% CI (0.635,0.848), P<0.001], and the results of the Hosmer Lemeshow goodness of fit test were χ2=9.448, P=0.306. When the threshold probability of the clinical decision curve is between 0.01 and 0.92, the clinical net benefit is higher.
Conclusion: The risk prediction model for anthracycline related acute ACT in children with hematological tumors constructed in this study has good predictive value and clinical applicability. It can help medical staff screen high-risk populations, take timely and effective intervention measures, and achieve maximum survival benefits.