Global left ventricular wall thickness by coronary computed tomography angiography – derivation, validation, and normal reference values

Aim Left ventricular hypertrophy (LVH) is a prognostic marker in the assessment of cardiovascular disease. Left ventricular mean global wall thickness (GT) can be measured using cardiovascular magnetic resonance, and has been shown to improve diagnostic and prognostic performance beyond left ventricular mass (LVM). The aim of this study was to use coronary computed tomography angiography (CTA) to derive and validate a formula to estimate GT from coronary CTA, and to determine normal reference values. Methods and results This was a retrospective cohort study among subjects aged 50-64 years randomly selected from the general population as part of the Swedish CArdioPulmonary bioImage Study (SCAPIS). GT, LVM, and left ventricular mid-diastolic volume (LVMDV) were measured by coronary CTA. A formula for estimating GT using only LVM and LVMDV was derived and validated. The study included 414 subjects (age 59±4 years, 67% male). Calculated GT was derived in a derivation subset (n=207) as GT[mm]=1.55xLVM[g]^0.77xLVMDV[ml]^-0.41, and agreed with measured GT in a separate validation subset (n=207, R 2 =0.97, p<0.001, bias 0.01±0.23 mm). Among a subset of healthy subjects, calculated GT for females (n=27) was 7.0±0.9 mm (normal range 5.2-8.8 mm), and for males (n=20) was 8.9±1.3 mm (normal range 6.4-11.4 mm). Systolic blood pressure, age and coronary artery calcium score were all weakly correlated with GT (R 2 =0.03–0.11, p<0.05 for all). Conclusion GT from coronary CTA can be calculated using only LVM and LVMDV with excellent accuracy and precision, and sex-specific normal reference values in healthy subjects are presented for use in the clinical evaluation of LVH.