Predictive value of a comprehensive atrial assessment with cardiac magnetic resonance in non-ischemic cardiomyopathy: keep it simple.

Cardiac magnetic resonance (CMR) can provide a multi-parametric evaluation of left atrial (LA) size and function. A complete CMR-based LA assessment might improve the risk stratification of patients with non-ischemic dilated cardiomyopathy (DCM). We performed a comprehensive CMR-based evaluation of LA size and function, in order to assess the prognostic impact of specific LA parameters in DCM. Secondary analysis of a prospective registry (UHSM-CMR study, NCT02326324) including 648 consecutive patients with DCM and CMR evaluation of LA area and LA length. Of these, 456 had complete LA assessment covering reservoir, conduit and booster pump function and including LA reservoir strain evaluated with feature tracking. The heart failure (HF) endpoint included HF hospitalizations, HF death and heart transplant. The arrhythmic endpoint included ventricular arrhythmias (VA) (sustained or treated by implantable defibrillator) and sudden death (SD). At median follow-up of 23 months, 34 patients reached the HF endpoint; in a multivariable model including NYHA class and LVEF, LA length had incremental predictive value. LA length ≥ 69 mm was the best cut-off to predict HF events (adjusted HR 2.3, p = 0.03). Among the 456 patients with comprehensive LA assessment, only LA length was independently associated with the HF endpoint after adjusting for LVEF and NYHA class. By contrast, no LA parameter independently predicted the arrhythmic risk. In DCM patients, LA length is an independent predictor of HF events, showing stronger association than other more complex parameters of LA function. No atrial parameter predicts the risk of VA and SD.

Assessment of Smith Algorithms for the Diagnosis of Acute Myocardial Infarction in the Presence of Left Bundle Branch Block.

INTRODUCTION AND OBJECTIVES: Recently, a new electrocardiography algorithm has shown promising results for the the diagnosis of acute myocardial infarction in the presence of left bundle branch block (LBBB). We aimed to assess these new electrocardiography rules in a cohort of patients referred for primary percutaneous coronary intervention (pPCI). METHODS: Retrospective observational cohort study that included all patients with suspected myocardial infarction and LBBB on the presenting electrocardiogram, referred for pPCI to 4 tertiary hospitals in Barcelona, Spain. RESULTS: A total of 145 patients were included. Fifty four (37%) had an ST-segment elevation myocardial infarction (STEMI) equivalent. Among patients with STEMI, 25 (46%) presented in Killip class III or IV, and in-hospital mortality was 15%. Smith I and II rules performed better than Sgarbossa algorithms and showed good specificity (90% and 97%, respectively) but their sensitivity was 67% and 54%, respectively. In a strategy guided by Smith I or Smith II rules, 18 (33%) or 25 (46%) patients with STEMI would have not received a pPCI, respectively. Moreover, the severity and prognosis of STEMI patients was similar regardless of the positivity of Smith rules. Cardiac biomarkers were positive in 54% of non-STEMI patients, limiting their usefulness for initial diagnostic screening. CONCLUSIONS: Diagnosis of STEMI in the presence of LBBB remains a challenge. Smith rules can be useful but are limited by suboptimal sensitivity. The search for new electrocardiography algorithms should be encouraged to avoid unnecessary aggressive treatments in the majority of patients, while providing timely reperfusion to a high-risk subgroup of patients.