Fig 1.
Visualization of the abdominal aorta in a control mouse by MR-angiography on a clinical MR system in comparison to high-frequency ultrasound.
Images demonstrate the visualization of the abdominal aorta in a sham-operated ApoE-/- mouse by in vivo MR-angiography on a clinical MR system (A) and dedicated high-frequency ultrasound (C) in comparison to histology (B). The reconstructed TOF angiogram (A1, maximum intensity projection (MIP)) of the suprarenal part of the nondilated abdominal aorta is shown. Red lines indicate the orientation of subsequently performed transverse MRI sequences (A2, A3, A4). Corresponding ex vivo histological sections (B1- B6), Elastica van Gieson (EvG) stain (B1, B2, B3), hematoxylin eosin (HE) stain (B4, B5, B6) demonstrate a nondilated abdominal aorta at different levels (red lines). High-frequency ultrasound images (C) of an abdominal aorta using a dedicated imaging system (Vevo 770). Corresponding longitudinal (C4) and transversal imaging planes (C1, C2, C3) are shown.
Fig 2.
Evaluation of the abdominal aorta in an ApoE-/- mouse by MR-angiography on a clinical MR system and high-frequency ultrasound.
Evaluation of an abdominal aortic aneurysm in an ApoE-/- mouse by in vivo MRI and ultrasound 4 week after continuous infusion of angiotensin II (4-week group). The maximum intensity projection (MIP) of the time-of-flight (TOF) angiogram (A1) demonstrates a significantly dilated aortic lumen. The location of transverse slices (A2, A3) are depicted by the red lines in A1. Corresponding ex vivo histological sections (Elastica van Gieson (EvG) stain (B1, B2), hematoxylin eosin (HE) stain (B3, B4) confirm the dilation of the aortic lumen. Magnifications of B2 and B4 highlight the site of rupture of the elastic laminae in the tunica media of the aorta in EvG stain and HE stain. Corresponding ultrasound images of abdominal aorta using the dedicated high-frequency US imaging system (Vevo 770) in sagittal (C3) and transversal orientation (C1, C2).
Fig 3.
Time course of aortic dilatation assessed by MR-angiography on a clinical MR system.
In vivo assessment of the dimensions of abdominal aortic aneurysm measured in the TOF angiography in an ApoE-/- mouse model of aortic aneurysms. The luminal aortic areas were measured in vivo after one, two, three and four weeks of angiotensin II infusion. In the control group (sham group) an average aortic area of 1.16 ± 0.12 mm2 was measured. After one week of angiotensin II infusion an average area of 2.3 ± 0.7 mm2 was measured. A further dilation was observed after three and four weeks with an average aortic area of 2.94 ± 0.8 mm2 and 3.79 ± 1.12 mm2. ToF: Time of flight.
Fig 4.
Signal to noise measurements of the aortic lumen on a clinical MR system and on a dedicated high-frequency ultrasound system.
This figure shows that magnetic resonance angiography (MRA, black bar) demonstrated a significantly (p < 0.001) higher signal to noise ratio (SNR) compared to ultrasound (US, grey bar). MRI and ultrasound measurements were performed at comparable locations of the aorta. The time-of-flight technique in MR and the B-mode in ultrasound are techniques which are also frequently used in a clinical setting.
Fig 5.
In vivo area measurements on the TOF MR-angiography and dedicated high-frequency ultrasound compared to ex vivo histopathology.
The closest correlation (A) between in vivo measurements and ex vivo histology was found for the TOF angiogram (R2 = 0.98; p < 0.001). This can be explained by the higher signal to noise ratios of MRA compared to ultrasound. A high, however slightly lower however significant correlation (B, C) was measured for ultrasound in systole (R2 = 0.92; p < 0.001) and diastole (R2 = 0.93; p < 0.001). A slight overestimation of luminal areas was measured for both MR-angiography and ultrasound. This can be explained with the shrinkage of the histological specimens following the processing of the tissue samples. MRA: Magnetic resonance angiography.
Table 1.
Summary of results from in vivo magnetic resonance imaging, ultrasound and ex vivo histology.
Fig 6.
Correlation of area measurements on the TOF MR-angiography with a dedicated high-frequency ultrasound.
Area measurements of aortic aneurysms showed a close correlation between time-of-flight MR angiography and high-frequency ultrasound measurements in both cardiac phases. Measurements in diastole (R2 = 0.94; p < 0.001) showed a closer correlation compared to measurements in systole (R2 = 0.91; p < 0.001). This can be explained by the acquisition technique of the time-of-flight angiography. The time-of-flight angiography is acquired continuously through both cardiac cycles. The resulting images therefore reflect the larger diameter in diastole.
Fig 7.
Reproducibility of MR-angiography and ultrasound measurements.
MR-angiography (A, B) measurements showed the highest reproducibility of measurements between both readers with the smallest 95% confidence intervals (-0.73 to 0.46, R2 = 0.96; p < 0.001). Intraobserver reproducibility for high-frequency ultrasound was slightly lower with slightly wider 95% confidence intervals (-0.7 to 0.88, R2 = 0.87; p < 0.001). This can be explained by the overall higher signal to noise ratio of the magnetic resonance angiography and therefore the improved delineation of the vascular boundaries.