Predictive value of electrocardiographic left ventricular hypertrophy in the general population: A meta-analysis

Abstract Background Conflicting results have been reported on the predictive value of the electrocardiographic left ventricular hypertrophy (LVH) in the general population. This meta-analysis sought to compare the predictive value of different electrocardiographic criteria of LVH in the general population. Methods We comprehensively searched PubMed and Embase databases until May 9, 2020 to identify observational studies investigating the predictive value of different electrocardiographic criteria for LVH (Sokolow-Lyon voltage, Cornell voltage or Cornell product) in the general population. Outcome measures were major adverse cardiovascular events (MACEs), cardiovascular or all-cause mortality. Results Ten studies enrolling 58,400 individuals were included. Comparison with and without electrocardiographic LVH, the pooled risk ratio (RR) of MACEs was 1.62 (95% confidence interval [CI] 1.40–1.89) for the Sokolow-Lyon voltage criteria, 1.70 (95% CI 1.27–2.29) for the Cornell voltage criteria, and 1.56 (95% CI 1.17–2) for the Cornell product criteria. The pooled RR of all-cause mortality was 1.47 (95% CI 1.10–1.97) for the Sokolow-Lyon voltage criteria and 1.87 (95% CI 1.29–2.71) for the Cornell voltage criteria. Furthermore, the pooled RR of cardiovascular mortality was 1.38 (95% CI 1.19–1.60) for the Sokolow-Lyon criteria, 1.66 (95% CI 1.24–2.33) for the Cornell voltage criteria, and 1.82 (95% CI 0.65–5.09) for the Cornell product criteria. Conclusions Different electrocardiographic criteria for evaluating LVH had a similar value in predicting MACEs among the general population. LVH detected by the Cornell voltage appeared to have a stronger predictive value in prediction of cardiovascular or all-cause mortality.