The use of impedance cardiography in automatic external defibrillators to discriminate between shockable and nonshockable ventricular tachycardia in real time

Purpose: The purpose of this study was to compare computer versus manual spatial QRS-T angle calculations in 12-lead ECGs. Methods: 12-Lead ECGs with narrow QRS complexes (b120 milliseconds) were selected from an existing database. All ECGs were printed on standard computer paper, and the QRS-T angle was manually calculated using the QRS/Tsimple formula. Manual measurements were made by a reviewer blinded to the computer scores and to all clinical data. The computergenerated calculations were obtained using automated measurements from SuperECG software (Mortara, Milwaukee WI). Mean QRS-T angle values were compared using paired t tests, and Pearson coefficient was used to test the correlation. Results: The spatial QRS-T angle was calculated from 16 ECGs (n = 16). On average, the computer-generated QRS-T angle was greater than the manual values (104 ± 17 versus 101 ± 27, P = .000). The range for the computer values was smaller than for the manual values, 78 to 132 versus 54 to 168 degrees, respectively. A moderate correlation was observed between the manual and computer measurements (r = 0.70, P b .003). Conclusions: Although computer measurements of the spatial QRS-T angle are on average only 3 degrees greater than manual measurements, the correlation between the 2 measures was not strong. Thus, computer and manual calculations of the spatial QRS-T angle should not be interchanged in research.