Hipertrofia Ventricular Esquerda: acurácia e valor prognóstico do eletrocardiograma e do Ecocardiograma Transtorácico – estudo ELSA Brasil

Abstract

INTRODUCTION: Left ventricular hypertrophy (LVH) is defined by an increase in left ventricular mass above pre-established limits. It is an independent predictor of cardiovascular mortality, and its early diagnosis is of fundamental importance. There are three main approaches to diagnosing LVH: through electrocardiogram (ECG), transthoracic echocardiography (TTE) and cardiac magnetic resonance imaging (MRI). Data from the literature have suggested that LVH identified by ECG and imaging techniques are clinically distinct entities, with different prognostic information that, when properly understood, can be used to benefit the patient. OBJECTIVES: To analyze the accuracy of the electrocardiographic criteria for LVH compared to Echo-TT and the prognostic value of the presence of LVH by ECG and Echo-TT for the composite outcome of death from acute myocardial infarction (AMI), heart failure and stroke, here called major adverse cardiovascular events (MACE). METHODS: prospective longitudinal study and a cross-sectional study of the accuracy of the diagnostic method. The study population consisted of individuals from the first wave of ELSA Brazil, with valid ECG and ECHO, excluding those with cardiac pacemakers, Wolff-Parkinson-White syndrome and complete intraventricular blocks on ECG. The electrocardiographic criteria for LVH analyzed were Minnesota codes 3.1; 3.3 and 3.1 + 3.3. To analyze the accuracy of the cross-sectional diagnostic method, the performance of each electrocardiographic criterion for LVH was compared to the available gold standard, Echo-TT. For the prognostic analysis, the cumulative incidence of the primary outcome in participants with LVH defined by ECG and Echo-TT was assessed. RESULTS: 2849 individuals were evaluated, 43.6% male. We found that 12.5% had LVH by EcoTT; 4.1% had LVH by Minnesota code 3.1; 1.9% had LVH by Minnesota code 3.3; 5.9% had LVH by Minnesota code 3.1 +3.3 and 17% had LVH by EcoTT in addition to code 3.1 +3.3. For the diagnosis of LVH, Minnesota code 3.1 had a sensitivity (S) of 7%, specificity (E) of 96.3%, positive predictive value (PPV) of 21.6% and negative predictive value (NPV) of 87. 9%. Code 3.3 presented S of 5%, E of 98.6%, PPV of 32.1% and NPV of 87.8%. The sum 3.1 + 3.3 presented S of 12%, E of 94.9%, PPV of 24.9% and NPV of 88.2%. In the prognostic analysis, the presence of LVH by EcoTT resulted in a hazard ratio (HR) for MACE of 3.2 (IC 95% 1.7-5.9). The presence of LVH by Minnesota code 3.1 had an HR for MACE of 3.6 (IC 95% 1.5 – 8.4) and the presence of LVH by codes 3.1 + 3.3, an HR for MACE of 3.4 (IC 95% 1.6 – 7.3). CONCLUSIONS: The low accuracy of the electrocardiographic criteria for LVH is consistent with that documented in the literature. The prognostic value of the electrocardiographic criteria for death from MACE was like that of Echo-TT and was maintained after adjusting for other variables, providing exploratory evidence that LVH by ECG offers independent prognostic information. It is likely that more accurate estimates will be obtained in studies with larger populations, more clearly delineating the significance of these ECG alterations in LVH.