Assessment of activation delay in the right ventricular outflow tract as a potential complementary diagnostic tool for Brugada Syndrome

Abstract

Aims: In patients with Brugada syndrome (BrS), diagnosis relies primarily on the presence of the characteristic type 1 electrocardiographic (ECG) pattern. The aim of this study was to propose an alternative diagnostic method in situations where ECG alone is uncertain.

Methods and results: This study was conducted in two phases: (i) Phase 1: cut-off determination. Controls and BrS patients were analysed to develop a predictive model based on electrocardiographic imaging (ECGi) parameters for the diagnosis of BrS. Patients with right bundle branch block (RBBB) were analysed separately. All patients underwent ajmaline infusion. Concealed BrS patients were evaluated in both the absence and presence of a type 1 ECG pattern. The right and left ventricular 'epicardium' maps obtained with ECGi were divided into eight regions, and the mean activation time (ATm) was calculated for each region. The ATm for each area was normalized to QRS length (ATm%); ATm and ATm% were compared across populations. (ii) Phase 2: cut-off validations. A new cohort of control and BrS patients was used to perform a blinded validation of the proposed method. In Phase 1 (cut-off determination), 57 patients affected by BrS, and 10 controls were included. Analysis of ATm and ATm% in right ventricular outflow tract (RVOT) showed significant differences between controls and BrS patients both with either concealed or manifested Pattern 1 ECG (3 721 ± 6.23 vs. 68.33 ± 14.73 ms, P < 0.001; 37.21 ± 6.23 vs. 107.57 ± 21.16 ms, P < 0.001). The relationship between the anterior-RV and the RVOT ATm was used to develop a predictive model to identify a diagnostic threshold for BrS diagnosis. An increase of 45% in anterior-RV ATm was determined to be the optimal predictor of delayed RVOT activation in BrS patients (area under the receiver operating characteristic curve = 0.97, accuracy = 0.92, F-score = 0.95). In RBBB patients, the ATm delay cut-off was reached exclusively in cases with concomitant BrS. In Phase 2, 7 out of 7 control patients exhibited a percentage increase between the anterior-RV and RVOT of <45%. Among BrS patients with concealed pattern (pattern-concealed), 11 out of 20 showed a percentage increase >45% (accuracy 67%). In BrS patients with manifested Pattern 1 (pattern-positive), 19 out of 20 showed a percentage increase of >45% (accuracy 96%).

Conclusion: In BrS, the delay in RVOT activation can be identified using a threshold value of 45% above the mean activation time in the anterior-RV for each patient, offering a reliable diagnostic tool when standard ECG method alone falls short.

Keywords: Activation time; Brugada syndrome; Diagnostic tool; ECGi.

Graphical Abstract

Figure 1

Four patients diagnosed with BrS in Phase 1 of the study, after pharmacological induction. All ECGs were evaluated by an expert operator and all patients achieved a Shanghai score ≥ 3.5 points. The baseline ECGs are shown in Supplementary material online, Figure S1 Supplementary Materials. (A) Classic ECG presentation of BrS pattern in leads V2 and V2-third intercostal space. (B) ECG with initially uncertain interpretation. A BrS pattern is present in leads V1 and V1-third intercostal space (coved ST segment, T-wave inversion), but the leads exhibit microvoltage, and the J-point elevation does not reach 2 mm above the isoelectric line. The PVC at the start of the ECG trace (also recorded with standard lead positions) originates from the anterior RVOT. (C) ECG with initially uncertain interpretation. The ST segment in the right precordial leads appears steep and lacks the classic convexity associated with the BrS pattern. (D) ECG with initially uncertain interpretation. The patient exhibited a baseline incomplete right bundle branch block (RBBB). Only during the washout phase of the pharmacological provocation test did the ST segment in lead V2-second intercostal space exhibit a coved morphology (as shown), whereas it remained concave during the peak of drug induction. BrS, Brugada syndrome; RVOT, right ventricular outflow tract.

Figure 2

Example of a patient with baseline RBBB (A) and BrS pattern at the peak of the pharmacological provocation test (B).

Figure 3

The CT scan was used to capture images of the heart, and the epicardial surface was divided into 4 regions for each ventricle. The left anterior descendent (LAD) coronary artery served as a reference to distinguish the right ventricle (RV) and the left ventricle (LV). The regions were numbered 1–8, with regions 1–4 corresponding to the RV and regions 5–8 to the LV: (1) anterior RV, (2) RV apex, (3) RV outflow tract (RVOT), (4) posterior-lateral RV, (5) anterior LV, (6) LV apex, (7) LV outflow tract (LVOT), and (8) posterior-lateral LV.

Figure 4

Mean differences and 95% confidence interval (CI) of mean activation times (ATm), mean activation-repolarization intervals (ARIm), and mean repolarization times (RTm) between BrS pattern-positive patients and the control group, recorded via ECGi and divided by cardiac zones. The greatest difference with the narrowest dispersion (95% CI range) corresponds to ATm in zone 3 (RVOT). The 95% CI of ATm does not include zero in 6 out of 8 zones and are in general the narrowest, confirming AT as the parameter with the highest statistical precision among those available. Zone 1 (anterior-RV) was chosen as a reference, as it shows the lowest variability in ATm, ARIm, and RTm compared with all other zones and exhibits the earliest ATm both in BrS and controls, as shown in Supplementary material online, Table S2a of the supplementary materials. Zone 1: anterior-RV; 2: RV apex; 3: RVOT; 4: posterior-lateral RV; 5: anterior LV; 6: LV apex; 7: LVOT; 8: posterior-lateral LV. ARIm, activation repolarization time; ATm, activation time, mean; BrS pattern-positive, BrS patients during manifested ECG pattern 1; RTm, repolarization time, mean.

Figure 5

Post-processed AT maps generated using the MatLab tool for the Phase 1 study populations. The first epicardial breakthrough is indicated with asterisks. The bottom panels display the window of interest for the electrogram recordings. The blue line represents the -dV/dT, and the purple line indicates the annotation of the first breakthrough. BrS, Brugada syndrome; BrS pattern-concealed, BrS patients with concealed ECG; BrS pattern-positive, BrS patients during manifested ECG pattern 1; RBBB, right bundle branch block.

Figure 6

Distribution of ATm by absolute (A and B) and relative (B and C) frequency. The scatter plots (A and C) show that the anterior RV and RVOT ATm values for the control group are concentrated in the first half of the graph, while in BrS patients, values are dispersed in the right part of the graph, particularly for RVOT values in BrS pattern-positive patients. The stacked bar charts (B and D) indicate that the median ATm values are around 40 ms, with values exceeding 100 ms observed exclusively in the BrS + population. anterior-RV, right ventricle anterior; ATm, activation time mean; BrS, Brugada syndrome; BrS pattern-concealed, BrS patients with concealed ECG; BrS pattern-positive, BrS patients during manifested ECG pattern 1; RVOT, right ventricle outflow tract.

Figure 7

Distribution of ATm% by absolute (A and B) and relative (C and D) frequency. The scatter plots (A and C) demonstrate that the ATm values normalized to QRS duration are less dispersed compared with the previous figure, showing two areas of concentration: in the first part of the graph (including all anterior-RV ATm% and RVOT ATm% values from the control group) and in the second part (where only RVOT ATm% values from the BrS population are found). The stacked column graphs (B and D) illustrate that all ATm% values fall within 60% of QRS duration, except for RVOT values in BrS patients and a small percentage of RV anterior values in the BrS pattern-positive population, where values are mostly above 60% of QRS length. anterior-RV, right ventricle anterior; ATm%, activation time mean%; BrS, Brugada syndrome; BrS pattern-concealed, BrS patients with concealed ECG; BrS pattern-positive, BrS patients during manifested ECG pattern 1; RVOT, right ventricle outflow tract.

Figure 8

Diagnostic performance tests. (A) The ROC curve shows the trade-off between the true positive rate and the false positive rate at different classification thresholds, represented by the percentage increase in ATm from the anterior RV to the RVOT. A 45% cut-off corresponds to 100% specificity and >90% sensitivity (AUC = 0.97). (B) A 45% increase in ATm from the anterior RV to the RVOT yields the highest accuracy and F-score. (C) The x-axis represents the cut-off values as the percentage increase in ATm from the RV anterior to the RVOT. The two curves refer to false positives in the total BrS population and false negatives in the control population. As can be seen, the two curves intersect above 40%, and at 45%, the false positive population is equal to zero. BrS, Brugada syndrome; FN, false negative; FP, false positive; FPR, false positive rate; TPR, true positive rate.