Synopsis

7T TOF-MRA has demonstrated an exquisite capacity for imaging the lenticulostriate artery (LSA) due to its high spatial resolution and in-flow effect. However, due to the morphological variability of the LSA, a clinical application is needed to identify abnormalities of this vessel. In this study, we analyzed the LSA using 7T TOF-MRA on patients with unilateral lacunar stroke. Comparing the results with other morphological parameters, we found that a reduced contrast-to-noise ratio (CNR) was a more sensitive parameter for reflecting impairment of the LSA on the ipsilateral side of the lacunae.

Introduction

Method

Four patients with unilateral infarcts in the BG and LV regions were enrolled in this study, with a confirmed clinical diagnosis collected prospectively. The local IRB approved the study, and informed consent was obtained from all the participants. All the images were acquired on a 7T research system (Siemens Healthcare, Erlangen, Germany) equipped with a Nova 32-channel head coil. The subjects underwent TOF-MRA, diffusion-weighted imaging (DWI), T1w, and FLAIR imaging.

The TOF-MRA was positioned to cover the middle cerebral artery (MCA) and LSA. The following parameters were used: FA = 20°, TR = 15 ms, TE = 4.30 ms, voxel = 0.23x0.23x0.36 mm³, FOV = 180x135x47 mm³, GRAPPA = 2, and TA = 7:34 min.

The DWI, T1w, and FLAIR images were inspected in OsiriX to identify the lacunar infarction. Vascular tracing of the MCA and LSA originating from M1 segments was manually conducted in SimVascular. The length and distance between the root and the extremity of the vascular branch were measured, and the tortuosity was calculated as the ratio of length to distance. Three branches with longest lengths in every hemisphere were selected to measure the contrast-to-noise ratio (CNR). The CNR was calculated at the sections of the LSA with vertical distances of 5, 10, and 15 mm from the MCA. The CNR was defined as [ max(Signal_LSA) - mean(Signal_tissue) ] / std_Noise.

Results

The extracted vascular trees were visually in good agreement with the coronal maximal intensity projection (MIP) of the TOF-MRA (Fig. 1). The numbers of stems and branches of the LSA are listed in Table 1. Independent sample t-tests showed no significant difference in the number of stems (p=1.000) or the number of branches (p=0.374) between the ipsilateral side (IPS) and the contralateral side (CON) of the lacunar infarction. There was also no significant difference between the IPS and CON sides in the length (p=0.365), distance (p=0.407), or tortuosity (p=0.705) of the three longest branches, as displayed in the box-scatter plots in Fig 2.

The box-scatter plot of the CNR was shown in Fig. 3, and it demonstrated a significantly lower CNR on the IPS side than on the CON side at all locations of measurement (p= 0.039 at 5 mm, p= 0.049 at 10 mm, and p= 0.004 at 15 mm). The CNR of the LSA branches decreased with increasing distance from the parent MCA, on both the IPS and CON sides.

Discussion

As an angiographic imaging technique, the contrast of TOF-MRA derives from the in-flow effect of blood flow. Therefore, the impairment of vessels will initially affect the signal intensity in TOF-MRA. Our results showed that the CNR on the IPS side was significantly lower than that on the CON side, indicating that the lacuna was caused by the lesions in the ipsilateral LSA. Meanwhile, other quantitative parameters failed to reveal the difference in the LSA between IPS and CON sides.

Since 7T TOF-MRA was initially developed for imaging the LSA, many parameters have been defined to quantify the LSA, including the number of stems and branches, as well as its length, distance, and tortuosity. However, these parameters can barely reveal the difference between an impaired LSA and a normal one because of the great morphological differences of the LSA among individuals. Previous research has suggested that the number of branches was slightly lower in patients with chronic stroke compared with healthy volunteers, and this conclusion was made after studying a relatively large sample (n=10)³. Our study demonstrates that the CNR is a sensitive parameter reflecting lesion of LSA in patients with unilateral lacunar stroke. The CNR could therefore be used as an indicator to evaluate the blood supply of the LSA. Further studies using a larger sample size are needed to evaluate the CNR of the LSA in healthy controls.

Conclusion

Acknowledgements

References

1. Gold G, Kovari E, Hof PR, Bouras C, Giannakopoulos P. Sorting out the clinical consequences of ischemic lesions in brain aging: a clinicopathological approach. J Neurol Sci 2007; 257:17–22.

2. Cho ZH, Kang CK, Han JY et al. Observation of the lenticulostriate arteries in the human brain in-vivo using 7.0T MR Angiography. Stroke 2008; 39:1604–6.

3. C-K Kang, C-A Park, C-W Park, Y-B Lee, Z-H Cho, Y-B Kim Lenticulostriate arteries in chronic stroke patients visualised by 7 T magnetic resonance angiography. Int J Stroke 2010; 5:374-380.