BACKGROUND:Transcutaneous cardiac pacing (TCP) is deeply entwined with the problem of assessing ventricular capture on the electrocardiogram (ECG). We sought clarification of ventricular capture during TCP. METHODS:We studied one hundred and ten patients (75 ± 12 years) with bradycardia who underwent pacemaker or implantable cardioverter-defibrillator implantation. The cohort was stratified by structural heart disease (SHD) status and presence of narrow or wide QRS during spontaneous heart rhythm. We compared 12-lead ECG data at baseline (48 ± 7 beats/min) with those of TCP as well as of transvenous pacing (TVP) at a similar increased heart rate (76 ± 9 beats/min) to ensure constant ventricular capture. The QT interval was corrected for heart rate (QTc) using Bazett's method as well as by the Hodge's and Rautaharju's formulae depending on the presence of narrow or wide QRS at baseline. Electromechanical coupling was assessed by noninvasive arterial pressure measurement. RESULTS:TCP (median 80 mA) produced a QRS pattern resembling left bundle branch block. Overall, both TCP and TVP induced significant QRS and QTc prolongations when compared with baseline measures (p < 0.001). TCP created narrower QRS than TVP in those patients with SHD and narrow QRS (p < 0.006). There was no significant QTc duration difference between TCP and TVP. Mean arterial pressure underwent similar significant decrease following either TCP or TVP over baseline (p < 0.001), without difference between the two pacing approaches in any patient group. CONCLUSION:TCP is associated with similar ECG and hemodynamic responses to those of TVP, regardless of the presence of SHD.