Cardiac MRI: accuracy of simultaneous measurement of left and right ventricular parameters using three different sequences.

Summary Background and aim:  We evaluated a value of Qp/Qs (left-to-right shunt measurement) using volumetric cardiac magnetic resonance (CMR) cardiac output (CO) measurements. We defined intraobserver, interobserver variability and reproducibility of left and right ventricular parameters by CMR. Furthermore, we studied whether shortened acquisition time has an effect on the accuracy of left and right ventricular parameters both in healthy volunteers and in patients with cardiovascular disease. Methods:  Sixteen subjects were enrolled in this study. Group A (n = 8, five males) consisted of healthy volunteers with a mean age of 25·9 years (range 24–30). Group B (n = 8, four males) was heterogenic consisting of patients with left ventricular (LV) hypertrophy, hypertension or coronary artery disease with their mean age of 56·3 years (range 38–70). Results:  The measured Qp/Qs as calculated from the right and LV CO was 0·87 ± 0·13. Overall variability [as presented with the lowest coefficient of variation (CV)%– the highest CV % of intraobserver, interobserver variability or reproducibility] of LV parameters were for ejection fraction (EF) 1·5–2·8%, stroke volume (SV) 1·3–3·2%, CO 1·4–3·2%, end-diastolic volume 0·5–3·0%, end-systolic volume 1·8–6·2% and LV mass 1·1–2·6%. Corresponding values for right ventricular parameters were for EF 1·1–4·2%, SV 1·9–8·2%, CO 1·9–7·6%, end-diastolic volume 2·1–7·6%, end-systolic volume 2·8–10·2% and right ventricle mass 2·9–8·3%. There was no statistically significant difference between the results of different sequences. Conclusions:  The CMR allows accurate Qp/Qs observation but the absolute value is at slightly different level compared with reference methods. Both left and right ventricular parameters are highly reproducible and even small clinically relevant changes can be measured with CMR. The shortened acquisition does not affect significantly to the accuracy of CMR-derived parameters.