4119Predictors of progression of preclinical left ventricular diastolic dysfunction to clinical heart failure in patients with malignancy

Background Preclinical diastolic dysfunction PDD is a well-known but poorly understood risk factor for clinical heart failure CHF. Preclinical diastolic dysfunction and predictors of progression of PDD to CHF in patients with malignancy has never been studied. Purpose The purpose of our study was to determine the incidence of CHF in cancer patients who have preexisting preclinical diastolic dysfunction, with a particular focus on identifying the risk factors contributing to progression of PDD to CHF. Study design We retrospectively sought all the patients who underwent echocardiography from 2003 to 2008 and were found to have preclinical diastolic dysfunction at our institution. PDD was defined as presence of grade I diastolic dysfunction, LVEF ≥50%, and free of clinical diagnosis of heart failure. Of these, only the patients who had a confirmed histopathological diagnosis of malignancy were included in the study. The end point was incidence of clinical heart failure, HFrEF and HFpEF. Multivariable adjusted Cox proportional hazards regression were performed to study the predictors of CHF. Results Out of 1086 patients with PDD and a histopathological diagnosis of malignancy, 208 patients were diagnosed with new onset CHF during the 10 years follow-up. Incidence of CHF, HFpEF and HFrEF was found to be 32.1 per 1,000 person-year, 13.9 per 1,000 person-year and 2.93 per 1,000 person-year respectively. Patients who developed clinical heart failure were found to have more hypertension, diabetes, myocardial infarction, peripheral vascular disease, pulmonary disease, renal disease, hematological malignancies and metastatic disease when compared to patients who did not develop heart failure. In multivariable adjusted Cox regression analysis, age at index date (HR 1.02, 95% CI 1.01–1.04, p=0.001), diabetes (HR 1.74, 95% CI 1.28–2.35, p<0.001), myocardial infarction (HR 2.19, 95% CI 1.49–3.20, p<0.001), lung disease (HR 1.53, 95% CI 1.14–2.06, p=0.004) and renal disease (HR 2.03, 95% CI 1.45–2.84, p<0.001) were independent predictors of CHF in patients with PDD and malignancy. Among chemotherapy agents, Vincristine, Ifosfamide, transtuzumab and Bortezomib were found to be associated with CHF. Conclusion This is the first study ever to report the incidence of CHF, HFrEF and HFpEF and identify a distinct set of predictors of heart failure in patients with PDD and malignancy. We estimated a much higher incidence of CHF (19%) in our study population, when compared to the general population with malignancy. Our study also depicted a significantly high incidence of HFpEF compared to HFrEF (8.3% vs 1.75%). Our study shows that patients with malignancy are at a significantly higher risk of developing heart failure if they have preexisting diastolic dysfunction. This emphasizes the importance of identifying patients with diastolic dysfunction and a more rigorous cardiovascular surveillance to prevent them from potential heart failure. Acknowledgement/Funding None