Optimal noninvasive assessment of diastolic heart failure in patients with atrial fibrillation: comparison of tissue doppler echocardiography, left atrium size, and brain natriuretic peptide.

OBJECTIVE: The aim of this study was to evaluate whether the ratio of early diastolic transmitral velocity to early diastolic mitral annular velocity (E/E') can be applied to identify diastolic heart failure (HF) noninvasively rather than using brain natriuretic peptide (BNP) or enlargement of left atrium (LA) in patients with atrial fibrillation (AF) by comparing the severity of HF symptoms. Moreover, we investigated the relationship between the changes in E/E' and the severity of HF or LA remodeling in the follow-up period.

METHODS: We examined 73 patients with nonvalvular AF disease and preserved left ventricular ejection fraction (>50%), ie, patients with diastolic HF accompanied with New York Heart Association (NYHA) functional class I to IV (n = 32, HF group) and those without HF (n = 41, non-HF group). No patients showed dyspnea caused by anemia, renal failure, lung disease, or other disease states except HF. We evaluated E, E', and E/E' by Doppler echocardiography, and the LA area (LAA) by 2-dimensional echocardiography. BNP levels were also examined. A follow-up study was performed in 18 of the 32 patients with HF.

RESULTS: E/E', LAA, and BNP were higher in the HF group than in the non-HF group (E/E', 15 +/- 5 vs 9 +/- 2; LAA, 24 +/- 6 vs 20 +/- 6 cm(2); and BNP, 321 +/- 200 vs 140 +/- 76 pg/mL, each P < .01). Using the receiver operating characteristic curve for identification of symptomatic diastolic HF with NYHA functional class II to IV, the areas under the curves were: E/E', 0.96 (95% confidence interval 0.91-1.0); LAA, 0.77 (95% confidence interval 0.64-0.89); and BNP, 0.85 (95% confidence interval 0.75-0.95). In the HF group, 18 patients who were re-examined 17 +/- 9 weeks after were divided into two groups, depending on the improvement in NYHA functional class, ie, improved group (n = 10) and unchanged group (n = 8). In the follow-up period, E (112 +/- 20-94 +/- 21 cm/s), E/E' (17.1 +/- 5-13.1 +/- 3), and LAA (28 +/- 5-24 +/- 4) decreased in the improved group (each P < .05), but E' and BNP did not.

CONCLUSIONS: E/E' could be useful in identifying symptomatic diastolic HF and evaluating the functional state in the process of HF in patients with AF. Moreover, E/E' is able to assess the improvement of diastolic HF in AF.