Cardiac computed tomography in dogs: insights into structural imaging and therapeutics

Abstract

Visualization of intracardiac structures with medical imaging has always been a formidable task due to the inherently complex structure of mammalian hearts. Traditional two-dimensional (2D) imaging such as fluoroscopy and 2D echocardiography present challenges in reconstructing the three-dimensional (3D) morphology of cardiac structures and alterations in intracardiac architecture caused by cardiovascular disease. Challenges in accurately visualizing complex cardiac geometry can lead to incomplete or inaccurate assessments during surgical and transcatheter interventions. To overcome this issue, cross-sectional imaging such as cardiac computed tomography (CCT) may be utilized to allow a comprehensive, dynamic, and 3D evaluation of cardiac valves, chambers and major vascular structures. One potential application of 3D imaging is preprocedural prediction of optimal fluoroscopic projections that can be used for intraprocedural guidance of cardiac interventions. For any given cardiac structure, there is an infinite number of 2D projections orthogonal to the straight-on (en face) projection forming a 360-degree circle of perpendicularity. The circle of perpendicularity becomes an arch of perpendicularity when considering projections above the fluoroscopic table only. In Chapter 1, this concept was explored by identifying the en face projections of various cardiac structures on CCT. The arch of perpendicularity (also known as the S curve) can be derived by using a trigonometric formula to solve the spatial relationship between the en face and perpendicular projections. Once the S curves are established, the optimal fluoroscopic projections (OFPs) can then be characterized and used to guide cardiac interventions. Cardiac computed tomography studies from eighteen healthy dogs were used to generate S-curves and OFPs for selected cardiac structures. The projections were defined by the cranial-caudal angulation and left-right rotation angles of the C-arm assuming the dog is in dorsal recumbency. Mean OFP S-curves and 95% confidence curve areas were determined for the aortic, mitral, and pulmonary valve along with interatrial septum. The impact of thoracic conformation on OFPs was found to be non-significant. By analyzing the CCTs from 18 dogs with severe pulmonary stenosis, the S-curves and OFPs of the pulmonary valve were described in Chapter 2. The coordinates of those optimized projections were found to be different from those of healthy dogs. Cardiac computed tomography from dogs with pulmonary stenosis were also used as a reference to examine the accuracy of pulmonary annular diameters measured on transthoracic echocardiography and angiocardiography. It was found that transthoracic echocardiography underestimates pulmonary annular diameter. Angiographic pulmonary annular diameters on standard lateral projection were also shown to be less precise when compared to the measurements made on CCT-derived OFPs. The inodilator pimobendan has been shown to delay the onset of congestive heart failure and decrease cardiac mortality in dogs with degenerative mitral valve disease. The benefit of pimobendan may go beyond its effect on cardiac contractility and vasodilation. In Chapter 3, CCTs from 20 dogs with subclinical degenerative mitral valve disease were used to investigate the effects of pimobendan on mitral annular dynamics. By comparing the mitral annular dynamics before and after the administration of pimobendan, it was shown that pimobendan augmented the systolic contraction of the mitral annulus which likely reduced the regurgitant orifice area and severity of mitral regurgitation. In conclusion, integration of CCT into clinical practice enhances our ability to evaluate and treat structural heart disease. Cardiac computed tomography can also be used as a reference to refine other imaging techniques and investigate the mechanism of action underlying various cardiovascular therapies.