Left ventricular hypertrophy identified by cardiac computed tomography and ECG in hypertensive individuals: a population-based study

Objectives: Screening of left ventricular hypertrophy (LVH) is a biomarker of organ damage in hypertensive individuals and associated with increased mortality. Cardiac computed tomography (CT) is widely expanding worldwide; however, the value of CT assessment of LVH is unknown. We aimed to identify individuals with LVH using both cardiac CT and electrocardiograms (ECG) and to explore potential differences between these phenotypical distinct diagnostic modalities. Methods: Participants in the Copenhagen General Population Study underwent 12-lead ECG and cardiac CT and were evaluated for the presence of LVH. Multiple ECG signs of LVH were compared with LVH by CT. Results: Out of 4942 participants, 1347 had untreated hypertension and in this group, 13% presented with anatomical LVH by CT and 10% by ECG with an overlap of 4%. ECG signs of LVH had negative predictive values between 87 and 89% compared with CT. Using a combination of the Sokolow-Lyon index, the Cornell voltage duration product and/or a Romhilt-Estes score at least 4, lead to an increased C-statistics (P < 0.001) compared with the use of any single ECG sign of LVH. Individuals with solely CT but not ECG signs of LVH had higher SBPs (152 vs. 144 mmHg, P < 0.001) and larger left atria (49 vs. 45 ml/m(2), P < 0.001) compared with individuals with solely ECG LVH. Conclusion: CT and ECG identifies LVH in 19% of hypertensive individuals with only a small diagnostic overlap. Commonly used ECG criteria for LVH cannot safely rule out the presence of anatomical LV organ damage.