ATHENS, GREECE. It is standard practice to postpone cardioversion if a patient has gone more than 48 hours since the onset of an AF episode. However, some emergency departments do not accept the patient's estimate of the time elapsed since onset, and routinely postpone the cardioversion for 3 weeks during which warfarin is administered to ensure (not always successfully) that no clots are present in the left atrium (LA) or left atrial appendage (LAA). Some hospitals will, however, perform immediate cardioversion if a transesophageal echo cardiogram (TEE) shows no evidence of thrombi (blood clots) or spontaneous echocardiographic contrast (SEC). Now Greek researchers propose that measuring the blood level of brain natriuretic peptide (BNP) in a patient arriving for cardioversion will give an accurate measure of time since onset and, more importantly, the risk of finding thrombi or SEC in the LA and LAA.

Brain natriuretic peptide, a cousin of atrial natriuretic peptide (ANP), is a hormone released from the walls of the ventricles when stretched, such as during unusually strenuous activity. It is stored as a prohormone within secretory granules in the ventricles and is secreted as a N-terminal fragment, N-terminal pro-brain natriuretic peptide (nt-pro-BNP), and the smaller active hormone BNP. BNP has effects similar to those of ANP, that is, it decreases sodium re absorption rate, renin release, and aldosterone release; it also increases vagal (parasympathetic) tone and decreases adrenergic (sympathetic) tone. Because nt-pro-BNP is easier to measure than BNP, it is often used as a marker for BNP.

Most afibbers are well aware that ANP and BNP levels are elevated in the early stages of an AF episode. Both are strong diuretics and are responsible for "the big pee". Research has shown that lone afibbers have elevated BNP (nt-pro-BNP) levels even when in sinus rhythm. There is also evidence that BNP levels decline significantly after a successful cardioversion and that a successful pulmonary vein isolation (PVI) procedure also results in a return to normal levels. Furthermore, an elevated BNP level is associated with an enlarged left atrium and with heart failure and structural heart disease. SeeThe Role of BNP in Atrial Fibrillation.

A high BNP level is highly predictive of acute heart failure with the following age-dependent cut-off points:

• Age below 50 years – greater than 450 pg/mL
• Age between 50 and 75 years – greater than 900 pg/mL
• Age over 75 years – greater than 1800 pg/mL

The Greek study involved 86 patients (65%) men) with an average age of 62 years. Fifty percent of the study participants had hypertension, 20% had coronary heart disease or other vascular disease, and 23% had diabetes. None had heart failure or severe valvular heart disease. Average left ventricular ejection fraction (LVEF) was 60% and average LA diameter was 42 mm. Patients were separated into 2 groups and had blood samples drawn upon admission and 6 and 12 hours after for BNP determination – group A consisted of 43 patients whose highest BNP level was above the age-adjusted cut-off (average of 1231 pg/mL), while group B consisted of 43 patients with maximum BNP levels below the age-adjusted cut-off (average of 537 pg/mL). The only significant difference between the two groups was that patients in group B had a larger LA diameter (43 mm vs. 41 mm).

All patients underwent a TEE. This detected thrombi in two group A patients (4.7%) vs. 13 among group B patients (30.2%). NOTE: The two patients in group A in whom thrombi were detected had grossly enlarged LA diameters (51 mm and 48 mm) and a high stroke risk score. SEC was detected in 14% of group A vs. 37.2% of group B. The presence of thrombi was also associated with advanced age, hypertension, diabetes, higher stroke risk score (CHAD2DS2VASc) and larger LA diameter.

However, multivariate analysis showed that a BNP level below the cut-off point was by far the most accurate predictor of thrombi. A patient with a BNP level below the cut-off had a 25 times greater chance of having thrombi in the LA or LAA than did a patient with a BNP value above the cut-off. The researchers suggest that, if a patient admitted with AF of unknown time of onset has markedly elevated BNP levels in the absence of heart failure symptoms, it may mean that the episode began less than 36 to 48 hours before admission, and that immediate cardioversion would be safe.

Deftereos, S, Stefanadis, C, et al. Estimation of atrial fibrillation recency of onset and safety of cardioversion using NTproBNP levels in patients with unknown time of onset. Heart, Vol. 97, 2011, pp. 914-17

Editor's comment: In view of the fact that BNP level can not only be used to predict risk of developing AF, estimate elapsed time since onset, predict the risk of thrombi in the LA and LAA, and to predict the success of cardioversion, it is to be hoped that cardiologists and electrophysiologists will make BNP determinants standard procedure when evaluating afibbers.