Non-ECG Gated Computed Tomographic Left Atrium-to-Aortic Diameter Ratio is not a Sensitive Discriminator for...

Echocardiography and radiography are frequently used to evaluate left atrial size, which serves as an important prognostic marker for canine cardiac conditions. However, there are no published criteria in veterinary medicine that allow for non-ECG gated CT to detect the presence or absence of left atrial enlargement (LAE). The availability of a validated measurement of left atrial size would have the potential to improve the detection of cardiac disease in dogs having thoracic exams for non-cardiovascular reasons. The objective of this study was to evaluate a non-ECG gated CT left atrium-to-aortic diameter ratio (LA:Ao) for detecting LAE, while using a two-dimensional (2D) echocardiogram as the gold standard. One hundred twenty-three dogs that underwent thoracic non-ECG gated CT and echocardiogram within 1 month of one another were evaluated retrospectively. Thirty-three dogs were diagnosed with LAE on 2D echocardiogram (short axis LA:Ao ≥ 1.6) while 90 dogs had a normal left atrial size. The CT LA:Ao was not a sensitive discriminator for LAE in dogs; however, larger ratios were highly specific. Dogs with a CT LA:Ao ≥ 3.4 (97% specificity, 15% sensitivity, 71% PPV, and 76% NPV) may benefit from further evaluation with an echocardiogram at the discretion of the clinical care team. Overall, the method evaluated in this study was not able to reliably detect dogs that have echocardiographically confirmed LAE and, as such, appears to have low clinical utility. The use of multiplanar reconstructed images to emulate the right parasternal short-axis plane is recommended for future studies.