Deformation imaging

Abstract

Recent developments in the field of clinical echocardiography allow the ultrasonographer to objectively quantify both regional and global myocardial function. Regional deformation (Strain) and deformation rate (Strain Rate) are novel indices of ventricular function that can be estimated non-invasively by ultrasonographic interrogation of the heart. Deformation Imaging (DI) represents a family of echocardiographic techniques that can be employed to detect, quantify and display the characteristics of physical deformation of the myocardium. The objective of the studies described in this thesis was to investigate Doppler-derived deformation imaging in dogs. This Master's Degree thesis is structured in two separate studies. The initial study compared Doppler-derived DI measures of ventricular function with more traditional invasive indices of cardiovascular function in healthy adult anesthetized dogs over a range of hemodynamic conditions created by serial pharmacologic manipulations. Five adult healthy dogs underwent simultaneous cardiac catheterization and transthoracic echocardiography under general anesthesia. The following invasive indices were monitored during the study: cardiac output (CO), femoral arterial (FA) and left ventricular (LV) pressures, +dP/dt[subscript]max, -dP/dt[subscript]max, right atrial (RA) and pulmonary arterial (PA) pressures, and pulmonary capillary wedge pressure (PCWP). Sequential manipulations of systolic function, afterload and preload were performed, respectively, by means of dobutamine, nitroprusside, and hetastarch infusions. Significant changes were induced in the following invasive hemodynamic parameters: cardiac output (p<0.0001, range 2.600 - 9.340 L/min), +dP/dt[subscript]max (p=0.0152, range 953.7 - 3822 mmHg/s), left ventricular end-diastolic pressure (LVEDP) (p<0.0001, range 0.210 -16.43 mmHg), mean right atrial pressure (p<0.0004, range -2.490 - 9.920 mmHg), systemic vascular resistance (SVR) (p<0.0001, range 510.0 - 2652 dyne*sec/cm⁵), and pulmonary vascular resistance (PVR) (p<0.05, range 256.4 -1635 dyne*sec/cm⁵). All dogs underwent a complete echocardiographic exam before anesthesia, after induction of general anesthesia and after each hemodynamic manipulation. A total of 750 regions of interest (ROIs) were included in the final analysis. Measurable plots were obtained in 741 (98.67%) out of 750 ROIs. Peak systolic strain rate (SSR) values obtained at the 6 ROIs did not differ significantly from each other and the avSSR (average of the 6 ROIs) negatively correlated with +dP/dt[subscript]max (r=0.9792, p=0.0208). Individual values of SSR from the different ROIs also negatively correlated with +dP/dt[subscript]max, with the exception of ROI 4 (basal portion of the interventricular septum). The lack of correlation was attributed to high variability in the measurements obtained from this specific ROI. The avSSR was significantly reduced, i.e. systolic function was increased, by the dobutamine infusion, but was not significantly changed during nitroprusside and hetastarch infusions. The LVEDP was significantly reduced during infusion of nitroprusside, apparently not influencing avSSR, and suggesting the possible preload independence of this parameter in the hemodynamic range generated in the study. Nitroprusside infusion effected a significant reduction of systemic vascular resistance without alteration of avSSR, suggesting a possible afterload independence of this parameter, at least in the hemodynamic range obtained in this study. Study 2 was conducted on healthy adult non-sedated dogs divided into four groups (group 1: Body weight (BW) < 15 kg, group 2: BW 15 - 30 kg, group 3: BW > 30 kg, and group 4: Doberman pinscher dogs). The aim of the study was to collect information on the feasibility, repeatability and reproducibility of Doppler-derived DI performed in a clinical setting. Of the initial 56 dogs from which echocardiographic data were collected, 7 dogs (12.5%) were excluded because of poor image quality, and data from 49 (87.5%) dogs were included in the final analysis. Of a total of 1470 ROIs analyzed, 4.65% yielded good quality curves, 88.3% were characterized by a significant amount of noise but retained a discernible pattern, and 7.14% were considered non-interpretable. Some pair-wise comparisons of the SSR values obtained from the 6 ROIs reached statistically significant difference. DI values in the Doberman pinscher group did not differ significantly compared to values obtained from a BWmatched group. DI values in group 1 dogs, but not the traditional indices FS and EF, were significantly higher compared to groups 2 and 3. Results from a Bland-Altman analysis revealed overall poor clinical repeatability and reproducibility, as a result of the wide variability of the data. Based on the results of study 1 we conclude that Doppler-derived SSR represents a useful load-independent index of global systolic function, as demonstrated in anesthetized dogs. However study 2 suggests that, in a population of non-sedated dogs, with a relatively wide range of breeds and BW, Doppler-derived DI measurements are characterized by less than optimal intra-operator and inter-operator variability which may limit its value as a tool to evaluate myocardial function in a clinical setting. Comparison to other newly developed non-invasive techniques to estimate regional myocardial deformation is needed to determine the eventual clinical utility of the method.