Abstract 12585: Flexible Modeling of the Association Between Cumulative Exposure to Low-Dose Ionizing Radiation From Cardiac Procedures and Risk of Hematopoietic Cancer in Children With Congenital Heart Disease

Abstract

Background: High-dose ionizing radiation is a well-established risk factor for childhood malignancies, including hematopoietic cancers (HC). However, data on the effect of low-dose ionizing radiation (LDIR) from medical imaging is conflicting and scant, especially in the pediatric population with congenital heart diseases (CHD). This study evaluated the association between cardiac LDIR exposure and hematopoietic cancers among children with CHD. Methods: A nationwide population-based cohort study was conducted using the Canadian Congenital Heart Disease (CanCHD) database. The study population included children born between 1999 and 2017 with at least one CHD diagnosis in their medical records. The cumulative dose of ionizing radiation corresponding to cardiac diagnostic and therapeutic procedures was quantified considering a 6-month exposure lag. The recency-weighted cumulative exposure (WCE) model, a flexible extension of Cox’s proportional hazards model, was used to assess the association. Results: We identified 139,975 children with CHD born between 1999 and 2017 and followed them for 1,388,681 person-years since birth. In this population, 718 hematopoietic cancer cases were observed. Children with HC were exposed to low-dose ionizing radiation earlier in life (median age at first exposure: 6 vs. 10 months; p=0.03) and had more procedures than those without cancer (mean number of procedures: 0.4 vs. 0.2; p<0.001). The cases received higher cumulative LDIR doses than their counterparts (mean dose: 2.3 vs. 1.1 mSv; p<0.001). We observed that cumulative LDIR doses within five years were associated with increased risk of hematopoietic cancer with the maximum association magnitude around 2 years. Conclusion: This is the first large population-based study documenting increased risk of HC associated with increased dose and recency of the LDIR exposure among children with CHD. Along with these findings, future studies focusing on detecting a threshold effect will help physicians decide the exposure point at which increased surveillance on LDIR exposure should be initiated.