Abstract 20906: Prognosis in Heart Failure: A TNM-Like Classification for Risk Stratification

Abstract

Introduction: NYHA is the most used classification for heart failure (HF), but it does not include patient’s clinical features. We proposed a new staging system for HF, named HLM (JACC 2014;20;63(19):1959-60), analogous to the TNM classification used in Oncology. HLM refers to heart damage (H), lung involvement (L), and malfunction (M) of peripheral organs (i.e. kidney, liver, hematopoiesis and central nervous system). Hypothesis: The aim was a preliminary comparison between HLM and NYHA at 6 and 12 months to assess the most accurate prognosis of HF patients in terms of rehospitalization for major adverse cardiovascular and cerebrovascular events (MACCE) and mortality. Methods: We enrolled 1064 consecutive patients with the diagnosis of or at risk for HF. According to HLM classification, all parameters for heart, lungs and peripheral organs function were collected and each patient was classified according to NYHA and HLM. At 6 and 12 months patients were followed up for rehospitalization for MACCE and cardiac death rates were calculated. Results: At 6 and 12 months follow-up, comparing to NYHA, HLM showed a greater area under the ROC curve (AUC) for rehospitalization as well as for cardiac death (Fig. 1). MACCE and cardiac death rates have been assessed separately for each parameter (H, L and M): L and M showed the most accurate prognostic power; at 1 year follow-up, L2 was significantly correlated (p=0.004) with rehospitalization as well as L3 (p=0.000), M1 (p=0.004), M2 (p=0.008) and M3 (p=0.003). M2 was significantly correlated (p=0.02) with cardiovascular mortality as well as M3 (p=0.004 ). Conclusions: Our preliminary data suggest that, compared to NYHA, HLM better stratifies risk of rehospitalization for MACCE and of cardiac death in HF patients. Regarding HLM, lung involvement (L2-3) and two or more organs dysfunction (M2-3) presented the most accurate prognostic power. Those preliminary results need to be confirmed in a greater population with a longer follow up.