Comparison of Two-Dimensional Crescentic Ellipsoidal Models and Three-Dimensional Echocardiography in Assessment of Right Ventricular Volume and Function in Dogs
Fellow: Robert Ciardullo
Mentor: Weihow Hsue
Co-Mentor: Romain Pariaut
DESCRIPTION (provided by applicant):
Right ventricular (RV) volume and function play pivotal roles in determining prognosis and treatment in both right- and left-sided heart diseases. Two-dimensional echocardiography (2DE) is the standard diagnostic modality for assessing heart disease in dogs. However, accurate RV volume quantification methods are lacking because of the RV’s complex geometric shape. The crescentic ellipsoidal model (CEM) is a simple geometric formula based on the two-dimensional area of the RV that has been shown to mimic three-dimensional RV shape in humans. There is also a linear alternative that utilizes solely one-dimensional measurements. Both methods can be obtained from either the right parasternal long-axis or left apical views. The objective of this study is to ascertain the utility of variations of the CEM in dogs by determining their agreement with three-dimensional echocardiography (3DE) and their reliability across the same and different operators. MATERIALS AND METHODS: Fifty canine patients, half with and half without right-sided heart disease, will be recruited for this study. Each dog will undergo an echocardiogram. The end-diastolic volume, end-systolic volume, and ejection fraction will be calculated from 2DE methods (CEM, linear-CEM, and Simpson’s method of disk, each applied in two different views) and 3DE. All 2DE and 3DE measurements will be performed separately and in a blinded fashion. The limits of agreement and presence of systematic bias between each 2DE value and 3DE will be evaluated with Bland-Altman analysis, and their correlation will be determined using linear regression. In addition, a subset of 15 dogs will receive four total echocardiographic exams over a morning session and an afternoon session in the same day, with each session consisting of two different operators. Intraoperator and interoperator reliabilities for each method will be determined using intraclass correlation coefficients, coefficients of variation, and reproducibility coefficients. EXPECTED OUTCOME: We expect that the CEM from the left apical view will generate the smallest limits of agreement and the highest intraoperator and interoperator reliabilities. SIGNIFICANCE: The findings of this study will establish a clinically practical method that can be immediately and widely applied to every routine 2DE study in dogs.