Abstract 18131: Prior Chemotherapy Does Not Provide Additional Predictive Value to Other Stage A Heart Failure Risk Factors for Left Ventricular Systolic Function
Introduction: Prior chemotherapy (PC) is diagnostic of stage A Heart Failure (SAHF), but its association with LV systolic dysfunction in the community has not been compared with other SAHF risk factors (RF).
Hypothesis: Addition of PC to presence of other SAHF RF improves prediction of LV systolic function.
Methods: We performed a cross-sectional study of community-dwelling pts aged over 65 with at least 1 SAHF RF (i.e. diabetes, hypertension, obesity, family history of heart failure or PC). Exclusion criteria included valvular heart disease, coronary artery disease, atrial fibrillation and heart failure. All patients underwent clinical evaluation and comprehensive echocardiography assessment, including global longitudinal strain (GLS) and 3D-LVEF. A healthy cohort aged over 65 years with no SAHF RF was assembled from local hospital echocardiographic databases.
Results: 630 patients were recruited (mean age 70.9 ± 4.8 years, 52.5% female, 51.1% diabetic). 79 (12.5%) had PC treatment (median follow-up duration 6.9 years; cancer origin 40% breast, 24% haematological, 14% colorectal, 20% other). No significant difference was seen between PC group vs other SAHF RF (n=551) for GLS (-18.7% vs -18.5%, p=0.71) or 3D-LVEF (60.4% vs 60.9%, p=0.45). An age-and gender-matched other SAHF subgroup (n=158) was compared with PC as sole RF subgroup (n=31), with no significant difference in GLS (-18.7% vs -18.9%, p=0.99) or 3D-LVEF (61.1% vs 61.1%, p=0.89). A significant difference was seen between PC group and age-matched healthy controls (n=37) for GLS (-18.7% vs -20.4%, p=0.009), but not for other echocardiographic findings. No significant association was found for PC with GLS as a continuous variable (β -0.081, p=0.74) or when added to other SAHF RF in a sequential logistic model (χ2 change from 64 to 65, p=0.68). Receiver-operator curve analysis demonstrated no discriminative power of PC for reduced GLS<18% (AUC 0.50) or 3D-LVEF<53% (AUC 0.48)
Conclusions: Patients treated with PC have abnormal GLS in comparison with healthy age-matched controls, but no significant differences in other echocardiographic findings in comparison to pts with other SAHF RFs. Addition of PC to presence of other SAHF RF did not improve prediction of LV systolic function.
Author Disclosures: M.T. Nolan: None. H. Yang: None. F. Pathan: None. M. Saito: None. K. Negishi: None. T.H. Marwick: None.
- © 2016 by American Heart Association, Inc.