In 1992, the Brugada brothers observed that some patients with idiopathic cardiac arrest (i.e., without obvious heart disease) had a particular ECG pattern. This pattern was characterized by elevation of the J point and down-sloping ST elevation followed by a negative T wave in right precordial leads (BrS type 1 ECG). On the basis of this observation, they introduced a new syndrome, named Brugada Syndrome [1].

This observation created panic among cardiologists and (presumed) patients worldwide. Indeed, as this ECG pattern is not so rare, the first obvious question was: are all subjects with a BrS type 1 ECG at risk of sudden death (SD), or only some of them?

Similar questions have frequently arisen in the history of medicine when an instrumental sign has been associated to a dangerous disease at risk of SD. For instance, more than seventy years ago, negative T waves were defined as subepicardial ischemia and were correlated with coronary heart disease [2]. Only after several decades was it established that negative T waves were not pathognomonic of ischemia, but may be determined by many causes, including physiologic variants.

Another example is mitral valve prolapse (MVP) which was also associated with the risk of SD. In this case, too, only after many years were it established that MPV is very frequent in the general population and is almost always benign. Indeed, only recently, autopsy studies have demonstrated that the few patients who suddenly die have fibrosis of the papillary muscles and myxomatosis of the mitral leaflets, thus indicating that mitral valve prolapse is a consequence of this disease and not the cause of SD [3].

With regard to Brugada Syndrome, a similar scenario appears probable. Indeed, the BrS type 1 ECG is relatively frequent and, in most cases, is discovered by chance [4]. In an increasing number of cases, it is discovered after the administration of drugs such as 1 C antiarrhythmic drug to treat atrial fibrillation or administered as a diagnostic test to discover this ECG pattern.

The vast majority of patients with a BrS type 1 ECG have a benign prognosis, particularly those with a drug-induced BrS type 1 ECG [5-6]. We recently published cumulative data from a multicenter prospective study which collected over 1500 patients with a BrS type 1 ECG without ICD, the majority of whom were clinically judged to be at low risk [7]. In these patients, the incidence of SD during follow-up was very low. Specifically, those with a spontaneous BrS type 1 ECG had an incidence of SD which was only slightly higher than that of the general population, while those with a drug-induced type 1 BrS type 1 ECG had an incidence of SD death similar (slightly lower) to that of the general population. It is therefore evident that BrS type 1 ECG and Brugada syndrome are not synonyms, in contrast with the suggestions of current European Guidelines [8].

It is currently accepted that Brugada Syndrome is determined by defects of SCN5A gene which encodes the alpha subunit of the main sodium channel Nav1.5. [9], leading to a loss of its function.

However, a defect of this gene is found in no more than 30% of patients with a Brugada syndrome.

In addition, in families with SCN5A mutations, the BrS type 1 ECG is present in only 47% of mutation carriers, while among genotype negative subjects 5% have a BrS type 1 ECG [10].

Furthermore, similar genetic defects may be found in other diseases such as in Lenegre disease, sick sinus syndrome, conduction defects, early repolarization syndrome etc. (J wave syndromes) [11].

Finally, in families with SCN5A mutations some patients present with a BrS type 1 ECG, some present with conduction disturbances and other both ECG anomalies [12].

Although it is difficult to reach a conclusion, all these data suggest that SCN5A mutations are not directly causal to the occurrence of a BrS type 1 ECG and that other factors (genetic and non-genetic) may play a role on the occurrence of the BrS type 1 ECG. Among these factors a subepicardial fibrosis of the right ventricular outflow tract has been suggested [13].

Several prospective studies [14-26] have identified a number of risk factors which may be useful in identifying patients with Brugada syndrome (at risk of malignant arrhythmias) among the great number of subjects with a BrS type 1 ECG: familial juvenile SD, syncope of presumed cardiac origin, spontaneous BrS type 1 ECG, first degree AV block, fragmented QRS, positive EPS etc. Unfortunately these factors singly considered, generally have a low positive predictive value and, as a consequence, they present with a little clinical usefulness to decide a therapy or not [24].

Recently, an increasing number of prospective studies have suggested that patients at highest risk are those with multiple risk factors [27-30]. Of note all these studies included patients with and without ICD in whom malignant events were considered ventricular arrhythmias recorded by ICD and cardiac arrest in those without ICD.

Interestingly, also in our cumulative prospective study of patients without ICD [7], the few patients who suffered SD/aSD had multiple risk factors in addition to a BrS type 1 ECG (i.e. they should have undergone ICD implantation).

If every patient with a BrS type 1 ECG is regarded as having a Brugada syndrome, a number of discrepancies arise on comparing these patients with those affected by other channellopathies, such as LQTS [31,32]:

In patients with a Brugada type 1 ECG, there is no correlation between the magnitude of ECG anomalies and prognosis. By contrast, in LQTS, the longer the QT the worse the prognosis. Paradoxically, patients with Brugada syndrome and cardiac arrest frequently have only mild anomalies, while subjects with a benign prognosis may display marked ECG changes [Figures 1-3] [33].

Figure 1: Three patients with aborted sudden death (A and B) or sudden death (C). A=male of 42 years; B=male of 17 years; C=male of 26 years. In A and B the ECG were recorded after the cardiac arrest, in C the ECG was recorded one year before. ICS=intercostal space.

Figure 2: Three patients with BrS type 1 ECG who reived an ICD in primary prevention. All of them during follow up received an appropriate shock. A=male of 42 years; B=male of 58 years; C= male of 37 years. ICS=intercostal space.

Figure 3: Three patients with a BrS type 1 ECG at low risk who had no therapy. A=male of 35 years, no clinical risk factor, negative EPS, follow up 12 years; B=male of 60 years, no clinical risk factor, negative EPS, no arrhythmias after 5 years (loop recorded implanted); C=female of 60 years, no clinical risk factor, negative EPS. ICS=intercostal space.

Only in a few cases (less than 30%) is a genetic defect found in Brugada Syndrome. By contrast, in LQTS, a genetic defect is present in over 70% of cases.

In patients with the BrS type 1 ECG, prospective and retrospective studies do not show overlapping characteristics. Accordingly, we recently published a study [34] including over 200 published cases and 23 personal cases of patients resuscitated from cardiac arrest in which a Brugada Syndrome was retrospectively diagnosed on the basis of a BrS type 1 ECG recorded after the event. Interestingly, in the personal cases and in those published in the last 10 years, a spontaneous BrS type 1 ECG was recorded after the event in only 60% of patients, while in the remaining 40% it was observed after administration of a class 1 C drug. In addition, few patients had a family history of Brugada syndrome and/or of SD, previous syncope or other commonly recognized risk factors. In other words, most of these patients could not be classified before the event as being at risk on the basis of risk factors identified by prospective studies.

The high number of drug-induced BrS type 1 ECG found in our study [34], certainly depended on the progressive diffusion of drug testing in patients with idiopathic cardiac arrest. Consequently, the question arises as to whether the cause of cardiac arrest in these cases was a Brugada syndrome, or whether the BrS type 1 ECG might have been a by-stander sign. Indeed, even if we accept that a “Brugada defect” is the cause of the ECG anomaly, and a possibly a co-factor, it is possible that other, unrecognized, causes might contribute to SD/aSD in these patients.

Unfortunately, there is no answer to these questions, mainly because no study has ever compared the prevalence of the BrS Type 1 ECG (retrospectively diagnosed) in subjects with idiopathic aSD with that of the general population. Therefore, it is not possible to demonstrate a by-stander role of BrS type 1 ECG in these subjects. Nor is it possible, however, to demonstrate the contrary.

26 years after the description of Brugada syndrome, too many questions remain and too few answers are available. In our opinion, Brugada Syndrome is an entity which needs to be re-defined on the basis of more scientific information, much of which is, as yet, lacking. In any case, we believe that the diagnosis of Brugada Syndrome should not be made only on the basis of a BrS type 1 ECG pattern. Rather, risk stratification in subjects with a BrS type 1 ECG should be performed on the basis of data available from prospective studies, which have clearly identified several risk factors and have clearly demonstrated that subjects at highest risk are those with multiple risk factors. Consequently, aggressive treatment (ICD, ablation) in subjects with only a BrS type 1 ECG pattern, and without other risk factors, is not ethically justified.