Purpose: To assess prechemotherapy cardiac imaging practices in relation to patients' heart failure (HF) risk.
Methods: We performed a population-based retrospective cohort study of women receiving chemotherapy for early-stage breast cancer in Ontario between 2007 and 2012. We surveyed for baseline cardiac imaging 6 months before chemotherapy or within 30 days thereafter. The proportion of patients who underwent imaging and cumulative incidence of major adverse cardiac event (MACE) rates was determined based on chemotherapy regimen and HF risk factors. Logistic regression was used to assess predictors of pretreatment cardiac imaging.
Results: We studied 18,444 women who had been treated with chemotherapy (median age, 55 years). There was near-universal imaging of women treated with trastuzumab-containing regimens, including those without additional HF risk factors. Women who received anthracyclines without trastuzumab underwent imaging more frequently if they had additional HF risk factors (73.3% v 62.6%; P < .001). The 5-year incidence of MACE was two to six times higher in patients with HF risk factors across all treatment regimens. Patients with HF risk factors who received anthracyclines without trastuzumab had a higher 5-year incidence of MACE (4.5%) than patients without HF risk factors who received trastuzumab without anthracyclines (2.6%). However, cardiac imaging was less frequent in the former group (73.3% v 93.6%; P < .001). Logistic regression indicated that most variation in baseline imaging was related to chemotherapy, followed by physician-level factors. The odds of imaging were doubled with female physicians. Patient-specific factors, including HF risk factors, made minimal contribution to variation in imaging.
Conclusion: Baseline cardiac imaging was driven by chemotherapy regimen rather than HF risk. This risk-imaging mismatch is an impetus to reconsider current cardiac imaging practices in patients who receive chemotherapy for breast cancer.