Introduction

Electrocardiogram (ECG) is a determining tool in the assessment of patients with chest pain, since it allows establishing the diagnosis of acute myocardial infarction (AMI), especially in those with ST-segment elevation (STEMI); however, in the presence of left bundle branch block (LBBB), repolarization abnormalities modify the morphology of waves and difficult their reading¹.

In view of these considerations, several criteria have been proposed to facilitate AMI diagnosis in the presence of LBBB, but none has reached optimal diagnostic performance². Sgarbossa, et al. were the first to propose an electrocardiographic score for the diagnosis of AMI in the presence of this entity³; however, although it has high specificity, sensitivity is still very low. Subsequently, Smith, et al. proposed a modification to these criteria that, according to their results, significantly increase sensitivity⁴.

Left bundle branch block as ST equivalent?

In the context of this diagnostic uncertainty, the presence of a new or presumably new LBBB together with ischemic symptoms has been traditionally considered an electrocardiographic equivalent to STEMI, which must be brought to urgent reperfusion²; however, recent studies have shown a low incidence of AMI in patients with LBBB^5,6. Owing to these findings, treatment guidelines radically changed after 20 years.

The 1996 and 2004 American Heart Association (AHA) guidelines^7,8 recommended immediate reperfusion therapy in patients with a new or presumably new LBBB who had ischemic symptoms, based on a meta-analysis that showed that fibrinolysis was associated with a survival increase in patients with LBBB and AMI⁹; however, this recommendation was abolished in the 2013 guidelines¹⁰. This change reduces the incidence of unnecessary fibrinolytic therapy that may cause hemorrhages, mainly in high-risk population, although not performing adequate reperfusion therapy in patients with a true coronary artery occlusion can increase morbidity and mortality in this population¹.

In 2012, the European Society of Cardiology (ESC) STEMI guidelines, and in 2014, the European Society of Cardiothoracic Surgery myocardial revascularization guidelines, recommended the performance of percutaneous coronary intervention (PCI) or pharmacological reperfusion for patients with ischemic symptoms showing a new or presumably new LBBB in the ECG^11,12. In 2017, the latest ESC STEMI guidelines were published, where they do not only continue making the same recommendation, but they also suggest the use of the Sgarbossa criteria, emphasizing that concordant ST-segment elevation is one of the best indicators of AMI¹³.

In 1996, Sgarbossa et al. published their criteria based on a sub-analysis of the GUSTO trial³, where 0.6% of patients with AMI had LBBB. Sgarbossa's criteria (Fig. 1) are based on a scoring system where three or more points are required for AMI diagnosis in the presence of LBBB, typically known as weighted criteria. Criteria without this scoring system can be also used and, in this case, any of the three criteria is accepted, a modality known as unweighted, which is more sensitive but less specific^4,14.

Figure 1 Sgarbossa's criteria for STEMI diagnosis in the presence of LBBB. A score lower than 3 is not traditionally considered diagnostic for STEMI, but does not rule it out. 

These criteria largely contributed to solve multiple problems when facing an ECG with LBBB in patients with ischemic symptoms; however, the study had two important limitations,^1,4 and thus it is not surprising that the attempts to validate the electrocardiographic criteria have shown a useful specificity but poor sensitivity:

Although Sgarbossa's criteria are an important advance, they have the limitation of their low sensitivity. In a meta-analysis, a general sensitivity of 20% and specificity of 96% were observed, for a Sgarbossa score higher than or equal to three^14,15. However, the combination of the first two criteria (ST-segment concordant elevation in any lead and ST-segment concordant depression in V1-V3), contributes to sensitivity and specificity increase by 37.2 and 96.6%, respectively¹⁶. Similar findings have been encountered in multiple studies where ST-segment concordant elevation or depression have been shown to be an independent AMI predictor¹⁷, and ST-segment concordant elevation has been shown to be the most specific individual criterion for the diagnosis of AMI in the presence of LBBB¹⁸.

Approximately nine years after the emblematic study by Sgarbossa et al., a report by the Hirulog and early reperfusion or occlusion trial (HERO-2) researchers compared the three Sgarbossa criteria and analyzed their 30-day prognostic value. In the original HERO-2 study, 300 patients had ischemic symptoms (more than 30 minutes with angina) and new or presumably new LBBB; these subjects were compared with a control group of 300 patients with no LBBB in the ECG. Of the 300 patients with LBBB, 92 met the criteria established by Sgarbossa et al.¹⁹.

AMI, as determined by creatine kinase levels, occurred in 80.7% of patients with LBBB and in 88.7% of controls (p = 0.006). ST-segment changes were specific but not sensitive for AMI diagnosis. In addition, mortality was higher in patients with LBBB. The study confirmed the following¹⁹:

The low usefulness of Sgarbossa's third criterion (discordant elevation higher than 5 mm in relation to QRS in any lead) is because the relationship between ST-segment elevation and the QRS voltage that precedes it (ST/S ratio) is not considered, a concept known as proportionality¹⁴; this corresponds to the ratio obtained between the ST-segment discordant elevation amplitude measured from point J in precordial leads⁴ and anterior QRS amplitude (Fig. 2)^1,20. In other words, this rule is based on the fact that repolarization is proportional to depolarization, an essential concept for STEMI diagnosis in the scenario of morphological abnormalities present in branch blocks or ventricular hypertrophies^1,21.

Figure 2 Electrocardiographic proportionality criterion: ST/S ratio calculation (adapted from Cai et al., 2013)²³. 

Smith et al. were the first to assess Sgarbossa's algorithms using PCI in the definition of their outcome,^4,22 and in an attempt to improve sensitivity, they replaced the third criterion with a statistical relationship. In this case-control study, where patients with LBBB and acute coronary occlusion were analyzed, implementation of both concordance criteria together with an ST/S ratio ≤ -0.25 had better results, with a sensitivity of 91%, specificity of 90% and LR+ 9, LR- 0.1^4,14,16.

Smith's rules I and II correspond to Sgarbossa's criteria with and without the scoring system, respectively; with a score of three or more for rule I and two or more for rule II being necessary for STEMI diagnosis. Rule III represents a modification of rule II, where the use of an absolute measurement is replaced by a relative discordance measurement. Rule IV consists in making a relative measurement of discordant deviation, either ST-segment elevation or depression, instead of ST-segment elevation absolute discordance. Finally, rule V is based on proportional discordance, either by ST-segment elevation or depression (Table 1)⁴.

In a study carried out by Meyers HP et al., Smith's rules were found to be significantly more sensitive than Sgarbossa's criteria¹. Di Marco A et al. had similar results; 145 patients were included in their study, out of whom 37 had a clinical presentation consistent with STEMI. They demonstrated that Smith's rules III and IV showed better results than Sgarbossa's criteria and had good specificity, of 90 and 97% respectively; however, sensitivity was 67% for Smith's rule III and 54% for rule IV, which is still suboptimal²².

Even though Smith's rules significantly improved sensitivity with respect to Sgarbossa's criteria for the diagnosis of STEMI in the presence of LBBB, they have a significant limitation related to the difficulty of implementation in clinical practice. Cai et al. proposed a simplified diagnostic algorithm that facilitates its systematic use, which includes the implementation of additional diagnostic aids that favor the approach to this type of patients (Fig. 3)²³.

Figura 3 Simplified algorithm for the diagnosis of STEMI in the presence of LBBB (adapted from Cai et al., 2013)²³. PCI: Percutaneous coronary intervention. 

Conclusion

Electrocardiographic definition of STEMI in the presence of LBBB continues to be a challenge despite multiple attempts to accurately characterize this condition. Thus, it is no surprise that great international scientific societies provide different recommendations. The last 2017 ESC guidelines for the treatment of STEMI are the only ones so far in recommending the use of electrocardiographic criteria to facilitate the diagnosis of AMI in the presence of LBBB and in considering it as an ST equivalent; however, why do they recommend the use of Sgarbossa's criteria when Smith's rules have been shown to be more sensitive and specific?

The answer could be associated with the ease in their application versus Smith's rules, which despite being promising, still have poor sensitivity.

There are still many questions to be answered, which makes for appropriate treatment of patients with suspected AMI in the presence of LBB to remain controversial and challenging, which is why there is the need to continue in the search for electrocardiographic criteria that enable identifying those patients who really need urgent reperfusion and avoiding the performance of procedures that might generate more morbidity and mortality. In addition, the use of tools complementary to ECG, such as echocardiogram and biomarkers, can provide useful information in diagnostic and therapeutic dilemmas.

Ethical disclosures

Protection of people and animals. The authors declare that no experiments have been conducted on humans or animals for this research.

Confidentiality of data. The authors declare having followed the protocols of their work center on the publication of patient data.

Right to privacy and informed consent. The authors declare that no patient data appear in this article.