SUMMARYWe previously demonstrated that the severity of arrhythmias is reflected by circulating brain natriuretic peptide (BNP) concentrations in patients without signs of congestive heart failure. In the present study, we evaluated the relationships between the severity of the arrhythmia, BNP concentration, and echocardiographic findings.The subjects consisted of 52 patients with ventricular premature contractions (VPC) but no manifestations of heart failure and no digoxin or beta-blocker therapy. Patients underwent Holter monitoring, plasma sampling for BNP measurement, and transthoracic echocardiography (TTE). We scored the motion of 16 left ventricular segments, deriving a wall-motion score index (WMSI) by totaling the scores and dividing by the number of segments scored.Twenty-three patients with Lown grade I to II arrhythmias constituted group A while group B consisted of 29 Lown III to IV patients. Group B had BNP concentrations triple those in group A (57.2 versus 18.1 pg/mL, P < 0.01). Left ventricular ejection fraction (LVEF) was similar in groups A and B (65.2% versus 62.1%, NS). Although left ventricular end-diastolic dimension (LVEDD) was normal in both groups, group B exhibited a larger LVEDD than group A (50 versus 46 mm, P < 0.005). The correlation (r) between BNP and interventricular septum thickness (IVST) was 0.27 (P = 0.013) in group A and 0.37 (P < 0.0001) in group B. Between BNP and posterior wall thickness (PWT), the correlation was 0.23 (P = 0.014) in group A versus 0.33 (P < 0.0001) in group B. The WMSI in group B was higher than in group A (1.34 versus 1.11, P < 0.05).We believe that besides the changes in echocardiographic parameters, the BNP elevation in group B could be a response to abnormal wall stress from the severe ventricular arrhythmias. (Jpn Heart J 2004; 45: 771-777)

patient's optimal dose of oral medication for longterm antiarrhythmic treatment must be determined by therapeutic monitoring, usually frequent measurements of serum drug concentration; however, such measurements are costly and time-consuming. In previous studies, clinical administration of antiarrhythmic agents was found to cause changes in signal-averaged electrocardiographic (SAECG) parameters 1 and each antiarrhythmic agent affected the electrocardiographic (ECG) variables in a different manner. 2,3 Also, it has been reported that SAECG is more sensitive than a standard ECG in detecting the changes in QRS duration resulting from administration of class I antiarrhythmic drugs. 4,11 Therefore, the purposes of the present study were to compare the ECG changes caused by administration of 3 class I antiarrhythmic agents (flecainide, pilsicainide and, pirmenol), and to identify the SAECG parameters that might be used to estimate changes in serum concentration of antiarrhythmic agents during monitoring for long-term therapy. Methods PatientsWe enrolled 36 patients (12 taking flecainide; 12 pilsi-