Atrial size is a predictor of cardiovascular mortality. Non-ECG-gated computed tomography pulmonary angiography (CTPA) is a common test for cardiopulmonary evaluation but normative values for biatrial volumes are lacking. We derived normal CT biatrial volumes using manual and semiautomated segmentation with contemporaneous transthoracic echocardiography (TTE) to confirm normal diastology. Thirty-five consecutive cases in sinus rhythm with no history of cardio-vascular, renal, or pulmonary disease and normal diastolic function were selected. Planimetric CTPA measurements were compared to TTE volumes measured using area length method. TTE and CTPA derived normal LAVi and RAVi were 27 + 5 and 20 + 6 mL/m2, and 30 + 8 and 29 + 9 mL/m2, respectively. Bland-Altman analysis revealed an underestimation of biatrial volumes by TTE. TTE-CT mean biases for LAV and RAV were -5.7 + 12.0 mL and -16.2 + 14.8 mL, respectively. The CT intraclass correlation coefficients (ICC 95% CI) for LA and RA volumes were 0.99 (0.96-1.00) and 0.96 (0.76-0.99), respectively. There was excellent correlation (p < 0.001) between the semiautomated and manual measurements for LA (r 0.99, 95% CI 0.98-0.99) and RA (r 0.99, 95% CI 0.99-1.00). Atrial volumetric assessment on CTPA is easy and reproducible and can provide additional metric in cardiopulmonary assessment.
Keywords: CT atrial size; CT atrial volumes; CT biatrial size; CT biatrial volumes; CT indexed atrial volumes; CT planimetry.