Assessment of the Accuracy and Reproducibility of RV Volume Measurements by CMR in Congenital Heart Disease
Are right ventricular (RV) volumes more accurately and reproducibly measured by cardiac magnetic resonance (CMR) in an axial orientation or in a short-axis orientation in patients with congenital heart disease?
Patients undergoing CMR were retrospectively identified. Underlying congenital heart disease consisted of tetralogy of Fallot, valvular pulmonary stenosis, and repaired truncus arteriosus in a majority of patients. Multiple measurements of RV volumes using both imaging planes were made by two observers. Intra- and inter-observer variability was assessed. RV stroke volumes were compared with forward flow in the pulmonary trunk, as calculated by phase contrast imaging.
There were no differences in the two contouring methods in the measurement of RV end-diastolic volume (EDV), end-systolic volume (ESV), ejection fraction, or stroke volumes. There was increased interobserver variability in RVESV measurements using short-axis images, although the difference was small (-2.5 ml/m2, 95% confidence interval, -5.6 to 0.0). In patients with large ventricles (RVEDV ≥150 ml/m2), RV stroke volume measured using axial contours correlated better with forward flow measured in the pulmonary trunk (concordance correlation coefficient [CCC] = 0.63), as compared with measurements made using short-axis contours (CCC = 0.56, p = 0.007).
The authors concluded that trends favoring the axial orientation for measurement of RV parameters were not clinically significant. In patients with large ventricles, volumes measured using axial views showed better agreement with flow measured in the pulmonary trunk, as compared with volumes measured using short-axis views.
This study compares two methods of calculation of RV volumes and function, based on the imaging planes used for measurements. CMR has become a standard tool in the assessment of RV size and function in patients with congenital heart disease, and it is essential that clinicians can count on the measurements reported by their colleagues in the CMR reading room. While this study showed a relatively small and likely clinically unimportant difference between measures derived from short-axis as compared with axial views, the inter-observer variability using the axial views seemed to be more of an issue at 9.5 ml/m2 for the EDV measurements and 12.7 ml/m2 for the ESV measurements. There is also the challenge of the lack of an alternative gold standard for comparison of the RV volume measurements. Flow measurements using phase contrast in the pulmonary trunk were used in this study, although there are limitations to this method, particularly in patients with tricuspid insufficiency.
Keywords: Reproducibility of Results, Tricuspid Valve Insufficiency, Cardiology, Stroke Volume, Risk Factors, Pulmonary Artery, Magnetic Resonance Spectroscopy, Heart Ventricles, Cerebral Ventricles
< Back to Listings