Background: Left bundle branch block (LBBB) complicates the diagnosis of acute myocardial infarction (AMI). The Sgarbossa criteria were developed from GUSTO I to surmount this diagnostic challenge but have not been prospectively validated in a large population with presumed AMI. We evaluated their utility in the diagnosis and risk stratification of AMI patients in ASSENT 2 & 3.
Methods: Baseline electrocardiograms (ECG) of LBBB patients were scored using Sgarbossa's criteria (0-10) by 2 readers blinded to the CK/CK-MB data and clinical outcomes; 267 (1.2%) patients had LBBB on their baseline ECG.
Results: Among 253 LBBB patients with available peak CK/CK-MB data, 158 (62.5%) had peak CK/CK-MB levels > 2x ULN, thereby qualifying for the diagnosis of AMI. A Sgarbossa score of 3 was shown in 48.7% of LBBB patients with elevated CK/CK-MB versus in 12.6% of those without a CK/CK-MB rise (P < .001). Patients with higher Sgarbossa scores, ie, 3, had a higher mortality compared with those with a score < 3, (23.5% vs 7.7% at 30 days P < .001; and 33.7% vs 20.2% at 1 year, P < .001, respectively).
Conclusions: Our findings validate the utility of Sgarbossa criteria for diagnosing AMI in the setting of LBBB. These criteria provide a simple and practical diagnostic approach to risk stratify this diagnostically challenging high-risk group and optimize risk-benefit of acute therapy.