This protocol provides a technical guide for the image acquisition of the thoracic and abdominal aorta in mice, using a high-frequency ultrasound system. Ultrasound aortic imaging has potential confounders, such as probe position and cardiac cycle, that may compromise the accuracy of the aortic measurements, particularly in the proximal thoracic aorta. This protocol describes detailed instructions and strategies for image acquisition, measurement, and data analysis, in order to accurately measure aortic dimensions.
For imaging the proximal thoracic aorta, there are several approaches to probe placement. The right parasternal long axis view shown in Figure 2A was used for ultrasound imaging in this protocol. This view facilitates the acquisition of high-quality images from the aortic sinus to the aortic arch portion. It is not optimal for the descending aorta because of interference of the ultrasonic waves. This protocol is applicable to most mouse models of thoracic aortic aneurysms because they exhibit luminal dilation predominantly in the aortic root to the ascending aorta. This includes chronic angiotensin II infusion that causes aneurysm formation in the ascending aorta of mice18,19,20,21,22,23. Mouse models of Marfan syndrome (fibrillin 1C1041G/+ and fibrillin 1mgR/mgR mice) display both aortic root and ascending aortic dilation23,24,25. Loeys-Dietz syndrome mouse models (postnatal deletion of TGF-β receptor 1 or 2 in smooth muscle cells) also develop aneurysm in the aortic root and ascending aorta18,26,27,28. Therefore, the right parasternal long axis view is appropriate for aortic imaging in these mouse models of thoracic aortic aneurysms. On the other hand, the right parasternal short axis view has the potential to capture aortic images diagonally because aneurysms are often complicated by aortic tortuosity, which may cause an overestimation of diameters. Unlike the thoracic aorta, the short axis view was used for the imaging of the abdominal aorta in this protocol. Since aortic curvature and tortuosity are modest in the abdominal aorta compared to the thoracic aorta, the acquisition of images in the short axis view ameliorates underestimations of the aortic diameter. It is important to note that different probe positions provide different viewing angles, and the aortic diameter may be different in each view angle. Therefore, reliable aortic diameter measurements are enhanced by applying the same probe position for all images within a study. Interestingly, three-dimensional (3D) ultrasound images of the heart and aorta have been reported recently29,30,31,32. In addition, current ultrasound systems can obtain 3D images over time as four-dimensional images33. Thus, these 3D imaging technologies have the potential to demonstrate the aortic structure more precisely, which may solve the problem of probe positioning.
Ultrasound images can be captured in either 2D brightness mode (B-mode) or one-dimensional motion mode (M-mode). Although some articles have used M-mode for the measurement of aortic diameter, B-mode is preferable15,34,35,36. M-mode has the capacity to image in two dimensions to increase temporal and spatial resolution. However, this mode relies on the assumption that the aorta is a concentric cylinder being imaged perpendicularly to the ultrasonic waves. This assumption may not hold true in an aneurysmal state and the curvature of the ascending aorta makes this difficult, even in nonaneurysmal states. In addition, the aorta does not remain in a fixed position throughout the cardiac cycle37. Therefore, M-mode may cause measurement errors, including over- and underestimations.
It is also important to note that the cardiac cycle affects the luminal diameter in the aorta. As expected, the aortic diameter in systole is greater than in diastole (Figure 4A-H), which is associated with aortic wall elasticity and strain. Aortic wall elasticity and strain can be calculated from the difference of aortic diameters between systole and diastole. Elasticity and strain are decreased in aneurysmal aortas compared to normal aortas31,34,35,38,39,40. Aortic stiffness cannot be measured directly by ultrasound. Measuring pulse wave velocity (PWV) can evaluate its stiffness as a proxy, which is reported to be increased in aneurysmal aortas31,35,41,42. PWV is calculated by the transit time between two arterial sites, using pulse wave Doppler images and their corresponding distance. For comparing aortic diameters, unlike clinical examination, there is no rigorous standardization in terms of cardiac phase for aortic measurements in mice. Therefore, it is still unclear which cardiac phase is appropriate for aortic measurements. However, to ensure reliable and reproducible comparisons, aortic diameters should be measured in a defined phase of the cardiac cycle.
This protocol provides detailed instructions for aortic imaging and data analysis in order to measure aortic dimensions accurately. The aortic measurement, using this protocol, was consistent with the actual ex vivo aortic diameter (Figure 5A). We also confirmed consistencies of inter- and intraobserver reproducibility (Figure 5B,C). All steps in this protocol, especially probe position and cardiac cycle, are necessary for accurate measurements. However, even when using appropriate procedures, artifacts during ultrasound imaging are unavoidable. The location of ribs and lung, as well as respiration and cardiac pulsation, can affect the image quality of the thoracic aorta. Intestinal gas can also cause artifacts in abdominal imaging. Thus, we suggest defining exclusion criteria when following this protocol in case of poor aortic images.
With the advent of high-resolution ultrasound systems, the aortic structure of mice can be examined in exquisite detail, both serially and conventionally, thereby greatly contributing to the understanding of aortic aneurysms. Ultrasound imaging, with the protocol as described above, is a reliable and reproducible noninvasive approach for quantifying aortic aneurysms in mice.