Abstract 4364340: Survivors of Childhood Cancers Exposed to Cardiotoxic Therapies Demonstrate Diastolic Dysfunction by Cardiac MRI

Toxicity from cancer therapy is a leading cause of morbidity and mortality in childhood cancer survivors (CCS). Research in the cardiotoxic effects of cancer therapy has focused on systolic dysfunction. Less data are available regarding the effects of cancer treatments on diastolic parameters, especially diastolic cardiac magnetic resonance (CMR) parameters, including peak filling rate (PFR), time to PFR (tPFR), PFR/end diastolic volume (EDV), peak ejection rate (PER), time to PER (tPER), and PER/EDV. The objective of this study was to evaluate whether CCS exhibit changes in CMR diastolic function. We hypothesized that CCS patients have worse CMR markers of diastolic dysfunction compared to healthy pediatric patients. This is a single-center retrospective analysis. CCS who underwent CMR imaging as part of cardio-oncology follow up were identified. All CMR studies were performed on a 1.5 or 3T scanner. Standard volumetric and functional analysis was performed, including left ventricular ejection fraction (LVEF), left ventricular end diastolic volume indexed (LVEDVi), and left ventricular end systolic volume indexed (LVESVi). All phases of the short axis stack of cine imaging were contoured to calculate filling and ejection curves. CMR analysis was performed using Medis QMass. A previously collected retrospective cohort of healthy controls with filling and ejection curves was used for comparison. Cohorts were compared using a Mann Whitney U and Logistic regression to correct for baseline LVEF. A total of 94 CCS with a median age of 15 years (43% female) were compared with 96 controls with a median age of 15 years (45% female). 84 subjects (89%) received anthracyclines and 31 (33%) chest radiation. The CCS cohort had a lower LVEF and smaller LVEDVi (LVEF 56% interquartile range (IQR) (52,58) vs 62% IQR(58, 65), p<0.001; LVEDVi 75ml/m2 IQR(68,86) vs 82ml/m2 IQR(75,91), p=0.002). CCS had slower PER, PFR, and tPER as well as decreased PER and PFR indexed to EDV (Table 1). PFR and PER remained significantly different between CCS and control after correction for age, BSA, systolic blood pressure, and LVEF (p=0.001 and 0.002, respectively). CCS have worse filling and ejection compared to healthy controls, even when correcting for demographics and known risk factors. PFR and PER may serve as early markers of cardiotoxicity in CCS.

Duke Scholars

Author Michael Jay Campbell Pediatrics, Cardiology