Cardiac magnetic resonance versus echocardiography: An analysis examining risk reclassification by left ventricular ejection fraction and cost burden analysis

BACKGROUND: Heart failure (HF) is a leading cause of morbidity and mortality in the United States and is projected to increase in the next decade. Left ventricular ejection fraction (LVEF) is used to guide optimal medical therapy and is typically quantified using two-dimensional transthoracic echocardiography (TTE) due to ease of accessibility and cost. However, LVEF measurements by cardiovascular magnetic resonance (CMR) are considered the gold standard due to their accuracy and precision. Despite this, CMR is not the first imaging modality selected for LVEF evaluation due to perceptions of long study time, high cost, and inaccessibility. Our study aims to determine the cost of imaging studies (e.g., CMR, TTE) relative to the overall HF-related health care costs and associated outcomes. METHODS: A retrospective single-center cohort study of 420 participants with same-day TTE and CMR from 2009-2019, including participants >18 years of age with good image quality with or at risk for cardiovascular disease. Primary outcome was a composite outcome defined as HF admission, left ventricular assist device, cardiovascular disease-related death, heart transplantation, and implantable cardioverter defibrillator implantation. HF risk groups were determined based on clinically relevant LVEF cutoffs. All costs were calculated and adjusted to 2022 US$. RESULTS: Participants were 49 ± 17 years old, 52% (219/420) female, 50% (209/420) White, and 41% (174/420) Black. Median follow-up was 4 years. HF was the most common co-morbidity (31%). LVEF measured by CMR predicted HF outcomes better than TTE (p = 0.005). Continuous net reclassification index of CMR LVEF was 0.36 (95% confidence interval: 0.16-0.56); p = 0.001 due to predominant reclassification to lower risk groups. On an individual level, HF health care cost increased from low- to high-risk groups irrespective of modality. High-risk individuals classified by CMR had lower average per-person HF health care costs compared to TTE counterparts. Cost of CMR and TTE was <1% of the total HF health care cost. CONCLUSION: The cost of non-invasive imaging studies accounted for <1% of the cost compared to other components of HF care. Downstream cost prediction based on LVEF classification using CMR has the potential to better predict cost burden compared to TTE in patients with HF.