Precision and reproducibility of non-automatic measurement of the QRS complex in potential candidates for cardiac resynchronization therapy.

BACKGROUND: Accurate measurement of QRS complex duration (QRSd) remains crucial for the selection of patients for cardiac resynchronization therapy (CRT). However, assessment of QRSd on conventional surface electrocardiograms (ECG), especially when performed without computer assistance, may be challenging due to the limited accuracy of the human eye to discriminate differences in the range of 10 ms at 25 mm/s. The value and reproducibility of visual assessment of QRSd at 25 mm/s on conventional ECGs was compared to those obtained using an electrophysiology recording system (EPRS) with simultaneous 12 lead traces at 100 mm/s, which was considered the gold standard. METHODS: The ECGs of 102 consecutive patients with left ventricular dysfunction undergoing electrophysiological evaluation were collected. Two sets of measurements were obtained: 1) QRSd-25 measured on conventional 12-lead ECGs printed at 25 mm/s with standard amplification (10 mm/mV) by 4 different observers, and 2) QRSd-100 measured on simultaneous 12-lead traces at 100 mm/s and 40 mm/mV by 2 different observers using electronic callipers. RESULTS: Significant differences were observed between QRSd-100 and QRSd-25 measurements (19.3 ±â€¯9.9 ms, range 1.0-47.5, p < 0.001). QRSd-25 showed significant inter and intra-observer variability. When categorizing individual ECGs in three QRSd-25 subgroups (<120 ms, 120-149 ms and ≥150 ms), low concordance was observed between both techniques (kappa index 0.25, p < 0.001). The sensitivity and specificity of QRSd-25 to detect QRSd-100 ≥ 150 ms was 36.6% and 100.0% respectively. CONCLUSIONS: Visual measurement of QRSd at 25 mm/s often underestimates its magnitude and presents significant inter and intraobserver variability.