Durante o seguimento, oito doentes morreram, um foi colocado em AVM e um foi transplantado. A análise de Cox univariável revelou potenciais marcadores ecocardiográficos de prognóstico na amostra tais como a dimensão da AE em modo M (HR-1,12; IC: 0,99-1,26; p = 0,067); VAE (HR-1,02; IC: 1,00-1,04; p = 0,046); VAE ajustado à superfície corporal (HR-1,03; IC: 1,00-1,07; p = 0,049); E/A (HR-0,99; IC: 0,99-1,81; p = 0,060); E/A > 2 (HR-7,00; IC:1,48-32,43; p = 0,014) e E/E' mitral (HR-1,04; IC: 1,00-1,09; p = 0,074). Na análise multivariável a única variável que permaneceu no modelo foi o VAE com o ponto de corte de 63 ml 7, IC: 0,61, p = 0,05 Objective: To evaluate the long-term prognostic value of LA volume (LAV) in patients with DCM. Methods: We prospectively studied patients admitted from January to December 2004 with a diagnosis of DCM, in sinus rhythm. Complete echocardiographic study at rest and after pharmacological stress was performed in all patients. The composite endpoint of mechanical ventricular assistance (MVA), heart transplantation or death during follow-up was assessed by univariate and multivariate analysis using a Cox regression model. Results: The study population consisted of 35 patients (68.6% male, mean age 52.0) with DCM, 82.9% of non-ischemic etiology. Ejection fraction (EF) at rest was 31.1±9.4%.During follow-up, eight patients died, one was placed on MVA and one underwent transplantation. Univariate Cox analysis showed various potential echocardiographic markers of prognosis in our population, including LA size in M-mode (HR 1.12, CI: 0.99-1.26, p=0.067), LAV (HR 1.03, CI: 1.00-1.07, p=0.046), LAV adjusted for body surface area (HR 1.03, CI: 0.99-1.26, p=0.049), E/A ratio (HR 0.99; CI: 0.99-1.81; p=0.060); E/A >2 (HR 7.00, CI: 1.48-32.43, p=0.014) and mitral E/E' ratio (HR 1.04, CI: 1.00-1.09, p=0.074).The only variable that remained in the multivariate model was LAV, with a cut-off value of 63 ml (HR 7.7, p=0.05). Conclusions: LAV was the only echocardiographic determinant of MVA, heart transplantation or death in our population with DCM. The echocardiographic parameters commonly used for risk stratification such as EF, left ventricular end-diastolic diameter and contractile reserve did not show prognostic significance in our study.