Risk stratification in Brugada syndrome: the challenge of the grey zone

This editorial refers to ‘Robustness and relevance of predictive score in sudden cardiac death for patients with Brugada syndrome’†, by V. Probst et al., on page 1687. Risk stratification according to the Shanghai and Sieira scores. Risk stratification according to the Shanghai and Sieira scores. Risk stratification in subjects with a Brugada ECG pattern (BrS type 1 ECG) is a debated issue.1–3 In subjects with a BrS type 1 ECG, the main problem is to identify those who are at risk of sudden death (SD) and those who are not. Current European guidelines only partially help in this difficult challenge, as they equate a BrS type 1 ECG with Brugada syndrome.5 In many cases, however, these terms are not synonymous. Indeed, according to some authors,3,4 a BrS type 1 ECG is not specific for a Brugada syndrome as it may be observed in many other conditions, different from Brugada syndrome. Some authors6,7 have suggested to identify patients with BrS type 1 ECG at risk of SD (i.e. affected by Brugada syndrome) on the basis of a multiparametric clinical and instrumental approach including history of familial SD, syncope of presumably cardiac origin, a positive electrophysiological study, etc. Their results6,7 have demonstrated that the patients at highest risk are those with multiple risk factors. Recently, two scores, the Shanghai and Sieira scores,8,9 have been proposed: the former in order to diagnose a Brugada syndrome; the latter to stratify the risk of SD in subjects with a BrS type 1 ECG. Both scores are calculated on the basis of several risk factors, with points being assigned to each factor. The sum of all points determines the final score. The paper by Probst et al. published in this issue of the European Heart Journal10 retrospectively tested the ability of these scores to predict malignant events during follow-up in subjects with a BrS type 1 ECG. In this elegant paper, the authors examined a very large population (1613 patients) over long-term follow-up (mean 6.5 years). To reach their objective, they subdivided these patients into two populations: 477 with an implantable cardioverter defibrillator (ICD) and 1136 without an ICD. In the former, the endpoint was the recording of fast ventricular arrhythmias by the device or appropriate shocks. In the latter, the endpoint was SD or aborted SD (aSD). The results of this study substantially confirm those of the original papers (by Sieira et al.8 and the Shanghai Group9). Patients with low scores had few events, while those with high scores had many events. Those with intermediate scores had an intermediate incidence of events. However, while there was a statistically significant difference in the frequency of events between subjects with low and high scores, in subjects with intermediate scores the difference was not statistically significant. The authors concluded that, in patients with intermediate scores, the clinical usefulness of these scores was limited and that their use was of little help in therapeutic decision-making. This finding is not surprising. In fact, the impact of different risk factors on the final score can in some cases inaccurately tilt the balance in the wrong direction. One example is syncope, which may have several possible causes (including benign vaso-vagal forms). As it is not always easy to distinguish the various conditions, the inclusion of this potential risk factor may change a low score into an intermediate score or, in contrast, may shift a patient from a high score category to an intermediate score category. As a consequence, intermediate scores should be considered grey areas (Graphical abstract). In the case of the inclusion of a single patient in the grey zone, further investigation may be useful. For example when using the Shangai score, an electophysiological study may be proposed (which is one the factors included in the Sieira score, but not in the Shangai score). Alternatively, using the Sieira score, other factors included in the Shangai score may be considered, e.g. familial history of Brugada syndrome or atrial fibrillation before 30 years of age (which are not included in the Sieira score). In any case, the results of this study are very interesting because they demonstrate the correctness of the clinical approach applied by the authors when deciding whether or not to implant an ICD. This approach is similar to the one commonly used by most centres worldwide. Indeed: (i) 16.5% (79/477) of ICD patients had events vs. 1% (12/1136) of patients without an ICD. In other words, patients considered at high risk correctly underwent ICD implantation. (ii) SD occurred in 1% of patients without an ICD, which means 0.15% per year. This value is similar to the incidence of SD in the general population (0.1% per year).11 Therefore, in the vast majority of these patients too, the authors correctly decided not to implant an ICD. The results obtained in patients without an ICD are of further interest. Indeed, in this group, among the few patients who suffered SD, almost all (except one) had a Sieira score of ≥5, with an annual incidence of 4.76%. In contrast, the incidence of SD in patients with scores between 2 and 4 was 0%. This latter observation is very encouraging because, in clinical practice, the decision not to implant an ICD in subjects with a type BrS type 1 ECG who do not have major risk factors is often difficult, and the fear of a future unexpected SD remains. In conclusion, this study confirms that the Sieira and Shanghai scores are useful in helping physicians to make therapeutic decisions, but they cannot substitute the clinical approach. The final lesson is that, in the clinical setting, the decision-making of physicians cannot be replaced by a calculating machine alone. Conflict of interest: none declared. The opinions expressed in this article are not necessarily those of the Editors of the European Heart Journal or of the European Society of Cardiology.