The overestimation of concentric hypertrophy in patients with HFpEF as determined by 2D-echocardiography

Background: Heart failure with preserved ejection fraction continues to pose multiple challenges in terms of accurate diagnosis, treatment, and associated morbidity. Accurate left ventricular (LV) mass calculation yields essential prognostic information relating to structural heart disease. Two-dimensional (2D) echocardiography-based calculations are solely limited to LV geometric assumptions of symmetry, whereas three-dimensional (3D) echocardiography could overcome these limitations. This study aims to compare the performance of 2D and 3D LV mass calculations. Methods: A prospective review of echocardiography findings at the University of Louisville, Kentucky, was conducted and assessed. Normal ejection fraction (EF) was defined as >=52% in males and >=54% in females. The following calculations were performed: relative wall thickness (RWT) = 2x posterior wall thickness/LV internal diastolic dimension (LVIDd) and 2D LV mass = 0.8{1.04([LVIDd + IVSd +PWd]³ - LVIDd³)} + 0.6. Concentric hypertrophy was RWT >0.42 and LV mass >95 kg/m² in females or >115 kg/m² in males. The same cut-offs were used for 2D and 3D echocardiography. Results: Echocardiographic findings for a total number of 154 patients in the study were investigated. There was a weak positive correlation between 2D and 3D LV mass indices (R = 0.534, r2 = 0.286, p = 0.001). Seventy patients had 3D EF >=45% with clinical heart failure (HFpEF). Among HFpEF patients, LV hypertrophy (LVH) was present in 74% of patients by 2D echocardiography and 30% by 3D echocardiography (McNemar test p = 0.001). Using 3D echocardiography as the reference, 68% of normal patients were misdiagnosed with LV hypertrophy by 2D echocardiography. Two-thirds of the patients with concentric remodeling by 3D echocardiography were misclassified as having concentric hypertrophy by 2D echocardiography (p = 0.001). Conclusion: Adapting necropsy-proven LV mass index cutoffs, 2D over-diagnosed LV hypertrophy through overestimation of the mass, compared to 3D echocardiography. In turn, the majority of HFpEF patients showed no structural hypertrophy of the LV on 3D imaging. This suggests that the majority of patients with HFpEF may qualify for pharmacological prevention to prevent further progression to LV remodeling or LVH.