QRS complex and T wave planarity for the efficacy prediction of automatic implantable defibrillators

Figures

• 

  Figure 1. Examples of QRS planarity assessment. The images in each of the two columns of figure show different projections of three-dimensional QRS complex loops of ECGs of two different study subjects (ie, each column of the figure shows different views on the very same QRS complex loop), both around 70 years of age, with similar underlying heart rate and with the same QRS duration (see online supplemental material for the images of source ECGs). The images in (A) show that the loop of the QRS complex was planar, that is, that it collapses into a practically straight line when viewed from the side of the plane of the QRS vector movement. The QRS non-planarity (ie, the departures from the plane of the vector movement) was 2.17%; the patient survived the study follow-up. On the contrary, the images in (B) show that the loop of the QRS complex vector movement was twisted out of a single plane and that consequently, its three-dimensional nature was visible in all possible projections. The QRS non-planarity was 10.9% and the patient died 278 days after ICD implantation. The T wave planarity was assessed in the same way (note that small T wave loops are also visible in the images). Online…

• 

  Figure 2. Comparison of QRS complex non-planarity values (A,B) and T wave non-planarity values (C,D) in patients who did and did not survive during the study follow-up (A and C) and in patients who experienced and did not experience appropriate ICD shocks (B and D). In each panel, cumulative distributions of the non-planarity values are shown together with Kolmogorov-Smirnov statistics and their corresponding p values. The non-planarity values shown were derived from the analysis of all eight independent leads of standard 12-lead ECGs. See the online supplemental material for the same comparison of non-planarity values derived from the ECG leads of the modelled chest belt (see the text for details). ICD, implantable cardioverter-defibrillator.

• 

  Figure 3. Kaplan-Meier analysis of the probability of death despite ICD protection (A and C) and of the probability of first appropriate ICD shock (B and D). (A,B) Comparison of the subgroups stratified by QRS complex non-planarity (QRS loop twist); (C,D) comparison of the subgroups stratified by the T wave non-planarity (T wave loop twist). The characteristics used in the comparisons were derived from eight independent leads of the complete ECG recordings. χ2 statistics and corresponding p values comparing the Kaplan-Meier curves are shown in each panel. The number of patients at risk in these groups is shown below the panels in colours corresponding to the individual graphs. Note that the data on ICD shocks were not available for the complete population—see the text for details. ICD, implantable cardioverter-defibrillator.

• 

  Figure 4. Kaplan-Meier analysis of the probability of death despite ICD protection (A and C) and of the probability of first appropriate ICD shock (B and D). (A,B) Comparison of the subgroups stratified by QRS complex non-planarity (QRS loop twist); (C,D) comparison of the subgroups stratified by the T wave non-planarity (T wave loop twist). The characteristics used in the comparisons were derived from three leads modelling a chest belt (see the text for details). χ2 statistics and corresponding p values comparing the Kaplan-Meier curves are shown in each panel. The number of patients at risk in these groups is shown below the panels in colours corresponding to the individual graphs. Note that the data on ICD shocks were not available for the complete population—see the text for details. ICD, implantable cardioverter-defibrillator.