We included 51 patients in the BrS group (7 females, 44 males, mean age 43 ± 11 years), 49 patients with localized ARVD/C and no ST-segment elevation in the right precordial leads (14 females, 35 males, mean age 39 ± 13 years), and 14 control patients (7 females, 7 males, mean age 38 ± 16 years). The main results regarding the BrS group population are summarized in Tables , . Sex ratio M/F was significantly higher in the BrS group compared to patients with localized ARVD/C without ST abnormalities and to control patients (respectively 6.3 vs. 2.5 and 1, p = 0.0001, Table ). This difference remained significant when comparing localized ARVD/C without ST abnormalities, either with a BS population pooling together BrS group I and BrS group II (i.e., BrS-ECG and no ARVD/C) or with the BrS group III (sex ratios M/F respectively of 2.5 vs. 5.8 and 9, p = 0.0001).
Results of investigations carried out in 41 patients with a BrS-ECG without angiographic arguments for ARVD/C (BrS groups I and II).
Results of investigations achieved in 10 patients belonging to BrS group III (BrS-ECG with angiographic ARVD/C).
Data from hemodynamic and angiographic quantitative study: patients with a BrS-ECG vs. localized ARVD/C without ST syndrome and vs. control patients.
Among BrS group, 19/51 patients (37%) presented with a spontaneous BrS type 1 ECG. The remaining 32 had a baseline BrS type 2 or 3 ECG, converted to type 1 under ajmaline infusion. Familial screening of BrS concerned only 5 patients of our series, whereas the remaining 46 were probands. Nineteen patients (37%) were symptomatic for aborted sudden death, agonal nocturnal respiration or syncope. Two patients presented with aborted SCD (documented nocturnal VF), 2 others with nocturnal seizures and syncope history, and 15 with at least one syncopal episode. Among patients without aborted sudden death, agonal nocturnal respiration or syncope, 2 patients presented palpitations and dizziness, 6 complained of lipothymia, and 5 others of palpitations. Remarkably, 12/34 patients (24%) had atypical chest pain, which was the cause of achieving an ECG in 10 patients. For the remaining 9 asymptomatic patients (18%), BrS-ECG diagnosis was fortuitous (systematic ECG). At the time of inclusion, 3 patients had an ICD because of type 1 ECG associated with history of aborted SCD, nocturnal agonal respiration or syncope. There was no statistical difference in terms of symptoms (syncope, SCD, seizures) between patients regarding the BrS-ECG type (1, 2, or 3) (p = 0.5). Moreover, no statistical significant difference was observed between BrS groups I, II, and III, in terms of symptoms (atypical chest pain, syncope or pre-syncope, seizures, SCD, p = 0.08).
Primary outcome (tables , )
Among BrS group, 34/51 patients (67%) fulfilled BrS HRS/EHRA 2005 criteria (Antzelevitch et al., 2005
), mainly because 19/51 (37%) were symptomatic with aborted sudden death, agonal nocturnal respiration or syncope, because of a positive ventricular stimulation in 14 patients (28%) or because of a familial history of SCD before the age of 45 years in 10 patients (20%). Among BrS group III, 8/10 patients (16% of all the BrS-ECGs) fulfilled international ESC/WHF 2000 ARVD/C criteria (McKenna et al., 1994
; Corrado et al., 2000
) and 5/10 (10% of all the BrS-ECGs) fulfilled BrS HRS/EHRA 2005 diagnostic criteria. An overlap was observed in 4 patients (8% of all the BrS-ECGs) who fulfilled both ARVD/C and BrS international criteria.
Non-invasive electrophysiology (BrS group)
Systematic ECG recording in one upper space above the usual right precordial leads (3rd intercostal space for V1H (V1 “high”) and V2H, 4th intercostal space for V3H) showed a BrS type 1 ECG in 63% of the cases (n = 26, Figure ) and allowed the diagnosis of BrS-ECG in 34% (n = 14, where a coved-type pattern was only visible in one of the classical leads V1, V2, V3, but also in one or more of the leads V1H, V2H, V3H). Reversible conduction slowing was observed in 24% of the cases (n = 10, mainly PR interval and QRS duration (>30%) prolongation). No further complication occurred.
Figure 3 Two typical BrS-ECGs (speed 25 mm/s; gain 1 mm/mV). Top: patient 33 with a baseline type 1 Brugada ECG (BrS group I). Bottom: patient 43 with a spontaneous type 1 Brugada ECG (left panel), becoming caricatural after ajmaline infusion (right panel, BrS (more ...)
Baseline ECG showed a prolonged PR interval (≥200 ms) in 7 patients (14%), a left anterior fascicular block (QRS axis beneath −30°) in 7 patients (14%) and a left posterior fascicular block in one patient (patient 28, BrS group II, with a PR interval of 200 ms). Remarkably, a left anterior fascicular block was present in 2 patients with ARVC/D (BrS group III) as compared to 5 patients (p = NS) without ARVD/C [BrS (I+II) groups]. A prolonged PR interval alone was present in 2 patients with ARVD/C (BrS group III) and in 4 patients without ARVD/C [BrS group (I+II), p = NS].
Holter-ECG or continuous monitoring showed premature ventricular contractions in 12/51 patients (24%), with a left bundle branch block pattern in 9, and associated with non-sustained RVOT tachycardia in one patient from BrS group III (n = 42).
SA-ECG was performed in 37/51 patients from BrS group (73%) at 40 and/or 25 Hz. When combining the results of SA-ECG at 40 and/or 25 Hz, late potentials were present in 27 patients out of 37 (73%). The presence of late potentials (at 40 and/or 25 Hz) was associated with a sensitivity of 100%, a specificity of 37%, a positive predictive value (PPV) of 37% and a negative predictive value (NPV) of 100% for angiographic ARVD/C diagnosis (BrS group III). Late potentials were significantly more frequent at 40 Hz and at 40 and/or 25 Hz in BrS group III compared to patients with a BrS-ECG and no ARVD/C [BrS groups (I+II), χ2 test, p = 0.04 and p = 0.02 respectively, p = 0.06 for 25 Hz alone].
Invasive electrophysiological study (BrS group)
Mean HV interval was 48 ± 7 ms (range 38–65 ms). Mean RV effective refractory period was 226 ± 20 ms. A sustained/syncopal VF or polymorphic VT was induced in 28% of the cases (n = 14), while non-sustained polymorphic VT was induced in 7 additional patients. We found no difference concerning inducibility between BrS groups I, II, and III, neither between asymptomatic and symptomatic patients, nor between patients with or without spontaneous type 1 BrS-ECG. Finally, one of the 2 patients who experienced previous SCD was not inducible (n = 25).
Non-invasive imaging (BrS group)
Cardiac MRI was performed in 29 BrS patients. Overall, MRI abnormalities (mainly localized RV wall motion abnormalities) were seen in 45% (13/29) of BrS patients, with fatty infiltration in 17% (5/29), which was often considered as non-specific by radiologists.
Radionuclide angiography was performed in 32 BrS patients. Mean RVEF was 52 ± 7% and mean LVEF was 64 ± 5% (versus contrast angiography, p = NS and p = 0.03, respectively). We noticed in 25% of the 32 explored cases (n = 8) a RVOT phase delay and a localized RV hypokinesia in 13% (n = 4). Seven patients from BrS group III underwent angioscintigraphy, which confirmed in two cases the diagnosis of ARVD/C made with contrast angiography (marked localized hypokinesia associated with a RVOT phase delay). Despite a poor sensitivity for ARVD/C diagnosis, angioscintigraphy had an excellent PPV (100%) and a good NPV (86%) as compared to contrast angiography (Table , bottom).
Hemodynamical study (table )
All groups under study were comparable in terms of right or left heart pressures, pulmonary or systemic vascular resistances (PVR, SVR), arteriovenous oxygen content difference, or hemoglobin content. The significant difference in heart rate observed between BrS group and patients with localized ARVD/C without ST segment elevation was attributed to the frequent prescription of β-blockers in the last group (76 ± 14 vs. 69 ± 14 bpm, p = 0.009). Cardiac index was significantly higher in patients from the BrS group compared to patients with localized ARVD/C without ST abnormalities (3.9 ± 0.8 vs. 3.3 ± 0.8 L/min/m2, p = 0.003). This was also the case when comparing BrS group III (ST-syndrome and ARVD/C) with localized ARVD/C without ST abnormalities (3.9 ± 0.7 vs. 3.3 ± 0.8 L/min/m2, p = 0.01). This finding could be also attributed to the lower heart-rate of this group. We observed no significant difference in terms of cardiac index between BrS groups I, II, III, and control patients.
Digitized quantitative contrast angiography
Qualitative analysis. The 26 patients from BrS group II (without overt structural abnormalities) had three different types of segmental wall motion abnormalities (Figure ): (1) diffuse anterior RV hypokinesia or akinesia (n = 15), corresponding to an extended anteroapical RV hypokinesia, sometimes associated with an inferior RV hypokinesia (n = 9); (2) localized apical RV hypokinesia or akinesia (n = 7); and (3) hypokinesia or akinesia restricted to the anterior RV infundibulo-trabecular junction (n = 4). BrS group III patients (n = 10) had RV structural abnormalities (deep fissuring) corresponding to early or localized ARVD/C, predominantly affecting the RV anteroapical wall (Figures , ).
LV anatomy was comparable between BrS groups I, II, and III. None of the patients had RV or LV microaneurysms (Frustaci et al., 2005
Quantitative assessment of rv systolic function (Table ). The RVEF and TAPSE were comparable between BrS groups I, II, and III, as well as between BrS group and control patients. However, RVEF was significantly higher in BrS group compared to ARVD/C patients without ST segment elevation (57 ± 5 vs. 54 ± 5%, p = 0.006). This difference was related to a higher RVEF among patients with a BrS-ECG and no ARVD/C (BrS groups I+II, n = 41) compared to ARVD/C patients without ST segment elevation (58 ± 5 vs. 54 ± 5%, p = 0.004). RVEF was comparable between the BrS group III and ARVD/C patients without ST segment elevation (56 ± 5 vs. 54 ± 5%, p = NS).
Statistically significant differences in terms of segmental excursion between BrS groups I, II and III were seen for the anterior, apical (BrS group I vs. II, BrS group I vs. III) and inferior (BrS group I vs. II) walls of the RV. On one hand, there was no difference in terms of RV segmental excursion between BrS groups II and III. On the other hand, only localized ARVD/C (with or without BrS-ECG) provided anteroapical dyskinesia. The excursion of the mid-lateral wall was significantly larger in patients with a BrS-ECG and no ARVD/C (BrS groups I+II), as compared to control patients and ARVD/C patients without ST segment elevation (12.6 ± 3.3 vs. 10.3 ± 3.0 and 10.6 ± 3.8 mm, p = 0.02).
Quantitative assessment of lv systolic function (Table ). The LVEF was comparable between BrS groups I, II, and III, and between BrS group and control patients.
No complication occurred during follow-up.
The results of our genetic study (n
= 45) are reported in Tables , . A missense R965C SCN5A
mutation was found in one patient (patient 41, BrS group II), resulting in a single amino-acid substitution. A Black african male (patient 33) belonging to BrS group I carried an African SCN5A
polymorphism (R481W) concerning a highly preserved gene segment among 5 species and 300 controls, which has been described in a recent publication as a special polymorphism (Tan et al., 2005
). He had also a double polymorphism of DSG2
. Patient 37, native of Cambodia, carried a TRPM4
mutation p.Q131H (c.393G>C) and had a baseline incomplete right bundle branch block, associated with a type 3 BrS-ECG and a left anterior fascicular block.
Among the 10 patients from BrS group III who underwent a more extensive genetic study, no desmosomal gene mutation was found. Patient 43 carried a rare PKP2 polymorphism (missense, A264V) and patient 44 carried a DSG2 polymorphism (p.Ser882Pro, c2644T>C). No SCN5A mutation was found in this group (Table , bottom). Another PKP2 polymorphism was found in patient 24 (BrS group II).