Abstract
OBJECTIVE To evaluate accuracy of quantification of right ventricle volume (RVV) by use of 3-D echocardiography (3DE) and ECG-gated multidetector CT (MDCT).
ANIMALS 6 healthy hound-cross dogs.
PROCEDURES ECG-gated MDCT and complete 3DE examinations were performed on each dog. Right ventricular end-diastolic volumes (EDVs), end-systolic volumes (ESVs), stroke volume (SV), and ejection fraction (EF) were measured for 3DE and MDCT data sets by use of software specific for RVV quantification. Correlation and level of agreement between methods were determined. Intraobserver and interobserver variability were assessed for 3DE.
RESULTS No significant differences were detected between SV and EF obtained with MDCT and 3DE. Significant differences were detected between right ventricular EDV and ESV obtained with MDCT and 3DE. No significant difference in heart rate was detected between methods. The correlation between MDCT and 3DE was very good (r = 0.87) for EDV and ESV, moderate (r = 0.60) for EF, and poor (r = 0.31) for SV. Bland-Altman analysis revealed a systematic underestimation of RVV derived by use of 3DE, compared with the RVV derived by use of MDCT (mean bias, 15 and 10.3 mL for EDV and ESV, respectively). Intraobserver (EDV, 12%; ESV, 18%) and interobserver (EDV, 14%; ESV, 11%) variability were acceptable for 3DE.
CONCLUSIONS AND CLINICAL RELEVANCE There was substantial variance for RVV measured by use of 3DE in healthy dogs and a significant underestimation of volumes, compared with results for MDCT, despite the fact there were no significant differences in SV and EF.
