A comparative study of two different methods for detecting late potentials in patients with and without ventricular tachycardia.

To compare the clinical utility of 2 methods in detecting late potentials (LPs), our method (K-method) and Simson's method (S-method) were used in 96 patients with or without ventricular tachycardia(VT). Indices of LPs were LP duration (LPd) in K-method, and mean voltage of the filtered vector magnitude for the last 40 ms of the QRS complex (V40) in S-method. For patients with underlying cardiac disease (UD), LPd in patients with sustained VT (s-VT; 39.4 +/- 20.1 ms) was significantly longer than that in patients with nonsustained VT (ns-VT; 16.1 +/- 6.6 ms) or without VT (non-VT; 13.3 +/- 5.2 ms) (p less than 0.01), while V40 in patients with s-VT, ns-VT, and non-VT were 24.7 +/- 26.4 microV, 56.5 +/- 31.7 microV, and 98.1 +/- 75.3 microV, respectively (p less than 0.05). Patients without UD had less apparent LPs in both methods. There was a significant reciprocal correlation between LPd and V40; LPd = (378/V40) + 6.66 (r = 0.89, p less than 0.001). The sensitivity and specificity with the highest predictive accuracy for s-VT in K-method were 50% and 100% and those in S-method were 50% and 97% respectively. In conclusion, despite many differences in these 2 methods, sensitivity and specificity were nearly equal and they were both useful to identify patients with s-VT.