5299 A Comparison Study of Imaging CBF Change in Transient MCAO Rat Brain with the SRT 1 Method and the CASL Technique

Target Audience: Researchers interested in perfusion imaging and stroke model imaging. Purpose: Cerebral blood flow (CBF) and its dynamic change are extremely important to brain function, metabolism, and tissue viability associated with many diseases. The saturation-recovery T1 (SR-T1) method provides a noninvasive and effective tool for monitoring rat brain CBF change and other useful information including blood-oxygen-level dependent (BOLD) signal and basal CBF simultaneously. However, this method was only applied to either the normal rat brain or global fourvessel occlusion in the previous studies, and its efficacy, sensitivity and reliability in the preclinical diseased model has not been tested. Stroke is one of the leading cerebral vascular diseases of death and long-term disability in the United States, and its animal model and clinical stroke trials are an active area of research. In this study, we aim to imaging CBF changes in response to the mild hypercapnia in varied lesion regions of the middle cerebral artery occlusion (MCAO) rat brain with the SR-T1 method and compare it with that measured using the continuous arterial spin labeling (CASL) technique. Materials and Methods: Ten MCAO rats (384±45g) under 1.8% isoflurane anesthesia were scanned on day 1 after a 1-hour MCA occlusion. MRI measurements were performed using a 9.4T/31cm magnet interfaced with VNMRJ consoles (Varian) and a H surface coil (2.8cm×2cm). The T2-weighted images were acquired with a fast spin echo sequence (TE=10ms; TR=4sec; FOV=3.2×3.2cm; matrix=256×256; thickness=1 mm; 8 echo train length). Gradient echo EPI (TE=17ms; FOV=3.2×3.2cm; image matrix=64×64; 1 mm thickness) combined with the saturation-recovery preparation was used for imaging T1 (or R1=1/T1) with 32 varied saturation recovery time (TSR) values range from 0.011 to 12s. The CBF change (ΔCBF) can be calculated with λ×ΔR1 (Eq. 1) , where λ (=0.9ml/g) is the blood-tissue water partition coefficient; ΔR1 is the longitudinal relaxation rate difference between the R1 measured under hypercapnia and normocapnia conditions. A modified TurboFLASH sequence (TE=30ms; TR=3sec; FOV=3.2×3.2cm; image matrix=64×64; 1 mm thickness) was used for the CASL experiment. The duration of the RF labeling pulse was 2.2 second. The CBF was computed following CBF=[λ×R1×(SC-SL)]/[SL+(2×α-1) ×SC] (Eq. 2), where SC and SL are signal intensity of the image without and with the RF spin labeling respectively, α is the effective efficiency of the arterial spin labeling and it is determined by: α=α1×TL/TR. α1 is the initial degree of spin labeling measured at the labeling plane, which is equal to 1 for inversion and 0.5 for saturation; the duty cycle is defined as TL/TR, where TL is the length of the labeling RF pulse, TR is the repetition time. Mild hypercapnia was induced with 6% CO2 inhalation. The CBF change equals to the CBF values obtained at hypercapnia condition minus the CBF values derived at normocapnia condition. Paired t-test was carried out to: i) compare the ΔCBF measured with the SR-T1 method and with the CASL technique in a certain ROI; ii) compare ΔCBF values in the varied lesion ROIs with that in the corresponding ROIs at the control side. Results: Figure 1 shows three continuous brain slices of the T2-weighted images, ΔCBF images induced by hypercapnia generated with the SR-T1 method as well as with the CASL technique in one representative rat scanned on day 1 of a 1-hour occlusion. The lesion areas show hyper-intensity in T2-weighted images, mostly localized in the subcortex region of the brain. Impaired vascular response to CO2 was observed in the lesion (right) side of the rat brain in ΔCBF images created with both techniques. Interestingly, the areas with compromised vascular reactivity toward CO2 (demarcated with white line shown in Fig. 1) spread more than the sub-cortex lesion regions shown in the corresponding T2weighted images, indicating the extensive involvement of the vascular reaction to the ischemic attack induced by the MCA occlusion. Table 1 summarizes the mean and standard deviation of ΔCBF induced by hypercapnia in different ROIs calculated with the SR-T1 method and the CASL technique in 10 MCAO rats on day 1 of post-occlusion. It shows the consistent ΔCBF results measured with these two image techniques in different ROIs at both the lesion side and the control side. In addition, the ΔCBF values at the varied lesion ROIs are significantly smaller than that in their individual homologous ROIs at the control side (p<0.01). Discussion: There is an excellent agreement of CBF change induced by the mild hypercapnia measured with the SR-T1 method and with the CASL technique in varied lesion regions at lesion side of the rat brain (Table 1). This is also true for the CBF change in the homologous ROIs at the control side. Paired t-test is used to compare the CBF change measured with the SR-T1 method and with the CASL technique in a given ROI, no statistical difference was found between them and the p value ranges from 0.25 to 0.74. Moreover, the CBF change in varied lesion ROIs is distinct from the values in the corresponding ROIs at the control side, indicating the sensitivity of perfusion change computed with these two MRI methods. For example, the CBF increase measured with the SR-T1 method and the CASL technique in the somatosensory cortex during hypercapnia at the lesion side is about 64% of that at the control side on day 1 after the occlusion. In contrast, there is almost no CBF change in the peripheral and core area at the lesion side when hypercapnia is induced on day 1 after the occlusion, suggesting the impairment of the vascular response to CO2 after the MCA occlusion. In addition, the spatial pattern and magnitude of the ΔCBF images created with the SR-T1 method is consistent with those generated with the CASL technique (Fig. 1), although the former images appear less uniform in the deep brain region, possibly due to the nature of EPI sensitive to inhomogeneous B0 and B1 fields. Conclusion: In summary, we have shown the consistent results of CBF change measured with the SR-T1 method and with the CASL technique in a preclinical MCAO rat model, and for validating the efficacy of the SR-T1 method of imaging CBF change under the diseased condition. Therefore, the SR-T1 method is able to noninvasively and reliably image the CBF change in an absolute scale induced by physiological and pathological conditions associated with cerebrovascular diseases. Acknowledgments: NIH grants NS057560, NS041262, NS070839, P41 RR08079 & EB015894, P30 NS057091 & NS076408 and WM Keck Foundation. References: 1. Wang et. al., ISMRM Proceedings, 1481, 2009; 2. Wang et.al ISMRM Proceedings, 1213, 2010; 3. Wang et. al., ISMRM Proceedings, 4008, 2011; 4. Divani et.al XXVIth ISCBFM, 2013. ROI (pixels) lesion side control side