Performance of Sokolow-Lyon index in detection of echocardiographically diagnosed left ventricular hypertrophy in a normal Eastern German population - results of the CARLA study

Background: Arterial hypertension is a common disease with high prevalence in the general population. Left ventricular hypertrophy (LVH) is an independent risk factor in arterial hypertension. Electrocardiographic indices like the Sokolow-Lyon index (SLI) are recommended as diagnostic screening methods for LVH. We assessed the diagnostic performance of the SLI in a cohort of a large general population. Methods: We used electrocardiographic and echocardiographic data from the prospective, population-based cohort study CARdio-vascular Disease, Living and Ageing in Halle (CARLA). Linear and logistic regression models were used to assess the association of SLI with LVH. To assess the impact of the body-mass-index (BMI), we performed interaction analyses. Results: AUC of SLI to predict LVH was 55.3 %, sensitivity of the SLI was 5 %, specificity 97 %. We found a significant association of SLI after covariate-adjustment with echocardiographically detected LVH (increase of left-ventricular mass index, LVMI 7.0 g/m(2) per 1 mV increase of SLI, p < 0.0001). However, this association was mainly caused by an association of SLI with the left-ventricular internal diameter (LVIDd, increase of 0.06 cm/m(2) per 1 mV increase of SLI, p < 0.0001). In obese (BMI > 30 kg/m(2)) we found the strongest association with an increase of 9.2 g/m(2) per 1 mV. Conclusions: Although statistically significant, relations of SLI and echocardiographic parameters of LVH were weak and mainly driven by the increase in LVIDd, implicating a more eccentric type of LVH in the collective. The relations were strongest when obese subjects were taken into account. Our data do not favour the SLI as a diagnostic screening test to identify patients at risk for LVH, especially in non-obese subjects without eccentric LVH.