BACKGROUND: The relationship between LV mass and systolic blood pressure at baseline was studied in the pooled database of the APTH (n = 408) and THOP (n = 376) trials. METHODS: Partial correlation coefficients were calculated between electrocardiographic and echocardiographic LV mass at baseline and different types of blood pressure, that is, office blood pressure (OBP), 24-h ambulatory blood pressure (ABP), daytime ABP and night-time ABP. RESULTS: Electrocardiographic LV mass correlated equally well (P < or = 0.001) with all types of systolic blood pressure (partial r = 0.19 to 0.22 for the Sokolow-Lyon index, 0.20 to 0.22 for the sum of the Sokolow-Lyon index plus the R-wave in aVl, and 0.12 to 0.19 for the Cornell voltage). The findings were similar for echocardiographic LV mass index (partial r = 0.15 to 0.22). However, for echocardiographic mean wall thickness, there was a stronger association with ABP than with OBP (partial r = 0.30 for 24-h ABP, 0.29 for daytime ABP, 0.26 for night-time ABP versus 0.15 for OBP (P < or = 0.05)).CONCLUSIONS In the combined dataset of the APTH and THOP trials, electrocardiographic and echocardiographic LV mass at baseline correlated significantly with all types of systolic blood pressure. For mean wall thickness, the relationship was stronger for ABP than OBP.