Background and objective: Left atrium diameter (LAD) is a very simple and easy parameter to obtain by echocardiography. It is influenced by systolic and diastolic ventricular dysfunction and by the coexistence of mitral regurgitation. We evaluated LAD as a predictor of prognosis (2 year mortality) in a heart failure (HF) population admitted to an outpatient HF unit. We compared LAD (mm/m2) with other echocardiographic parameters (left ventricular ejection fraction, left ventricular end-diastolic and end-systolic diameters [mm/m2], mitral regurgitation, degree of diastolic dysfunction and pulmonary artery pressure).
Patients and method: We studied 368 patients (73% men; mean age [standard deviation]: 65.2 [11] years; 60% of ischemic etiology). The mean left ventricular ejection fraction by echocardiography was 32.3% (13.1%). The majority of patients were in NYHA (New York Heart Association) class II (48%) or III (43%).
Results: Two years mortality was 20.6%. In the univariate analysis LAD (p < 0.001), left ventricular end-diastolic diameter (p < 0.001), left ventricular end-systolic diameter (p = 0.003), the degree of mitral regurgitation (p = 0.002) and the pattern of diastolic dysfunction (p = 0.004) showed a significant relationship with 2 years mortality, but not left ventricular ejection fraction and pulmonary pressure. In the echocardiographic multivariate analysis, only LAD remained significantly associated with mortality. In the multivariate analysis including important clinical parameters such as age, sex, etiology, time lapsed since symptoms onset, NYHA functional class, and the presence of diabetes, hypertension and atrial fibrillation, LAD remained as independent predictor of 2 years mortality. Patients with LAD less than 25 mm/m2 have a 10.9% mortality, whereas those with LAD equal or greater than 25 mm/m2 have a 30.1% mortality (p < 0.001).
Conclusions: LAD was a good predictor of 2 years mortality, better than other echocardiographic parameters in patients of our outpatient HF unit and was independent of strong clinical parameters.