ABSTRACT Background: Different electrocardiographic abnormalities have been associated with left ventricular systolic dysfunction (LVSD), although the association with standard electrocardiographic frontal plane ST-segment depression (STD) has not been established. Objective: The aim of this study was to evaluate whether lead I STD (STD-I) allows predicting the presence of LVSD. Methods: Patients with risk factors or stable chronic heart disease, and baseline electrocardiogram (ECG) and echocardiogram that provided evaluation of left ventricular ejection fraction (LVEF), left ventricular wall motility, and dichotomous evaluation of left ventricular hypertrophy (LVH), were prospectively included in the study. ST-segment morphology in leads I and V6 was evaluated, defining horizontal (≥1mm at 80 ms from the J point) or downsloping STD as abnormal STD-I and STD-6. Results: A total of 691 patients; with mean age of 69.8 ± 12 years and 61.6% men, were prospectively analyzed. STD-I and STD-6 were identified in 250 (36.2%) and 199 (28.8%) cases, respectively. Presence of STD-I and STD-6 was associated with a significantly lower LVEF compared with the absence of this finding: 44.8 ± 13.9% vs. 55.6 ± 8.9% (p <0.0001) and 45.8 ± 14.1% vs. 54.1±10.4% (p <0.0001), respectively. Both were associated with the presence of LVSD, defined as LVEF <50%, although STD-I showed better diagnostic performance than STD-6 [area under the ROC curve 0.72 (95% CI 0.69-0.76) vs. 0.64 (95% CI 0.61-0.68), p = 0.0001]. Conclusions: This study showed that STD-I predicts the presence of LVSD better than STD-6. The potential relevance of these findings should be placed in the current context of the emerging use of wearable devices that analyze electrocardiographic information through a single lead.

RESUMEN Objetivo: Distintas alteraciones del electrocardiograma (ECG) han sido asociadas a disfunción sistólica ventricular izquierda (DSVI), si bien la asociación con el infradesnivel del segmento ST (IST) del plano frontal del ECG estándar no se encuentra establecida. El objetivo del presente trabajo fue evaluar si el IST de la derivación DI (IST-1) permite predecir la presencia de DSVI. Material y métodos: Se incluyeron de forma prospectiva pacientes portadores factores de riesgo o cardiopatías crónicas estables, con ECG basal y ecocardiograma que aportara evaluación de la fracción de eyección (FEVI), motilidad ventricular izquierda y evaluación dicotómica sobre la presencia de hipertrofia ventricular izquierda (HVI). Evaluamos la morfología del segmento ST en derivaciones DI y V6, definiéndose como anormal (IST-1; IST-6) al ST infradesnivelado (≥1mm a 80mseg del punto J) o descendente. Resultados: Se analizaron en forma prospectiva 691 pacientes, edad media 69,8 ± 12 años, 61,6% hombres. Se identificó IST-1 e IST-6 en 250 (36,2%) y 199 (28,8%) casos, respectivamente. La presencia de IST-1 e IST-6 se asoció a una FEVI significativamente menor comparado con la ausencia de dicho hallazgo: 44,8 ± 13,9% vs. 55,6 ± 8,9%, (p <0,0001) y 45,8 ± 14,1% vs. 54,1 ± 10,4% (p <0,0001) respectivamente. Ambos se asociaron a la presencia de DSVI, definida como FEVI <50%, aunque el IST-1 mostró mejor rendimiento diagnóstico que el IST-6 [área bajo la curva 0,72 (IC 95% 0,69-0,76) vs. 0,64 (IC 95% 0,610,68), p = 0,0001]. Conclusiones: Este estudio mostró que la depresión del segmento ST de la derivación DI permite predecir la presencia de DSVI mejor que IST-6. La potencial relevancia de dichos hallazgos debería situarse en el contexto actual de la emergente utilización de dispositivos wearables que analizan la información electrocardiográfica mediante una única derivación.

Resumo Fundamento: Os dados de reprodutibilidade da extensão e dos padrões de realce tardio pelo gadolínio (RTG) em cardiomiopatia hipertrófica (CMH) são limitados. Objetivo: Explorar a reprodutibilidade da espessura parietal (EP) da região, extensão do RTG e padrões de RTG em pacientes com CMH avaliados com ressonância magnética cardíaca (RMC). Métodos: A extensão do RTG foi avaliada pelo número de segmentos com RTG e pela massa total do VE com RTG (%RTG) e foi definido o padrão RMC com RTG para cada segmento. Resultados: A população do estudo foi composta por um total de 42 pacientes (672 segmentos) com CMH. As medições médias de EP mostraram uma diferença média entre observadores de -0,62 ± 1,0 mm (6,1%), com limites de concordância de 1,36 mm, -2,60 mm e um coeficiente de correlação intraclasse (CCI) de 0,95 (95% IC 0,93-0,96). Medições máximas de EP mostraram uma diferença média entre observadores de -0,19 ± 0,8 mm (0,9%), com limites de concordância de 1,32 mm, -1,70 mm e CCI de 0,95 (95% IC 0,91-0,98). O % RTG mostrou uma diferença média entre observadores de -1,17 ± 1,2% (21%), com limites de concordância de 1,16%, -3,49% e CCI de 0,94 (95% IC 0,88-0,97). A diferença média entre observadores com relação ao número de segmentos com RTG foi de -0,40 ± 0,45 segmentos (11%) com limites de concordância de 0,50 segmentos, -1,31 segmentos e CCI de 0,97 (95% IC 0,94-0,99). Conclusões: O número de segmentos com RTG pode ser mais reprodutível do que o percentual da massa do VE com RTG.

Abstract Background: Reproducibility data of the extent and patterns of late gadolinium enhancement (LGE) in hypertrophic cardiomyopathy (HCM) is limited. Objective: To explore the reproducibility of regional wall thickness (WT), LGE extent, and LGE patterns in patients with HCM assessed with cardiac magnetic resonance (CMR). Methods: The extent of LGE was assessed by the number of segments with LGE, and by the total LV mass with LGE (% LGE); and the pattern of LGE-CMR was defined for each segment. Results: A total of 42 patients (672 segments) with HCM constituted the study population. The mean WT measurements showed a mean difference between observers of -0.62 ± 1.0 mm (6.1%), with limits of agreement of 1.36 mm; -2.60 mm and intraclass correlation coefficient (ICC) of 0.95 (95% CI 0.93-0.96). Maximum WT measurements showed a mean difference between observers of -0.19 ± 0.8 mm (0.9%), with limits of agreement of 1.32 mm; -1.70 mm, and an ICC of 0.95 (95% CI 0.91-0.98). The % LGE showed a mean difference between observers of -1.17 ± 1.2 % (21%), with limits of agreement of 1.16%; -3.49%, and an ICC of 0.94 (95% CI 0.88-0.97). The mean difference between observers regarding the number of segments with LGE was -0.40 ± 0.45 segments (11%), with limits of agreement of 0.50 segments; -1.31 segments, and an ICC of 0.97 (95% CI 0.94-0.99). Conclusions: The number of segments with LGE might be more reproducible than the percent of the LV mass with LGE.