Synopsis

We present a novel pulse sequence and modeling algorithm to quantify the water exchange rate (kw) across the BBB without contrast, and to evaluate its clinical utility in a cohort of elderly subjects at risk of cerebral small vessel disease (SVD). A diffusion preparation module was integrated with pseudo-continuous ASL (pCASL) with background suppressed 3D GRASE readout. The results showed good reproducibility of kw measurements (ICC=0.75) in elderly subjects with repeated scans ~2 weeks apart. Average kw was increased in subjects with diabetes and hypercholesterolemia, and was correlated with vascular risk factors, cognitive function and white matter hyperintensities.

Background

Methods

Figure 1 (a) shows the sequence diagram. Diffusion preparation was implemented by non-selective pulses and bipolar gradients along z-direction with their timing optimized to minimize the eddy current.² The non-CPMG component of DW pCASL signal was eliminated by applying a de-phasing gradient as proposed by Alsop³ (Figure 1(b,c)). Imaging parameters were: FOV=224mm, matrix size=64×64, 12 slices (10% oversampling), resolution=3.5×3.5×8 mm³, turbo factor=14, EPI factor=64, bandwidth=3125Hz/pixel, TE=36.5ms, TR=4000ms, label/control duration=1500ms. A 2-stage approach was employed⁶: 1) PLD=900ms and b=0,14 s/mm² for FEAST scan (15 measurements, TA=4min); and 2) PLD=1800ms and b=0,50 s/mm² for kw measurement (20 measurements, TA=5min20sec).

The single-pass approximation (SPA) was applied to fit the water exchange rate (kw) by estimating pCASL signals in the capillary/brain tissue compartments. To minimize spuriously high kw due to noise, a novel total generalized variation (TGV)⁴ regularized SPA modeling algorithm was applied for estimating ATT and kw, based on DW pCASL data acquired at two PLDs with respective b values:

$$argmin_{\tiny{ATT,ATT'}}\large[\small\frac{1}{2\lambda}||ATT-g(\frac{\tiny\triangle \small M_{b_{ATT}}^{900}}{\tiny\triangle \small M_{0}^{900}})||_2^2+\alpha_1|\triangledown ATT-\triangledown ATT^{'}|_1+\frac{\alpha_0}{2}|\triangledown ATT^{'}-\triangledown ATT^{'T}|_1\large]\small...[1]$$

$$argmin_{\tiny{kw,kw'}}\large[\small\frac{1}{2\lambda}||kw-f(1-\frac{\tiny\triangle \small M_{b_{DW}}^{1800}}{\tiny\triangle \small M_{0}^{1800}},ATT)||_2^2+\alpha_1|\triangledown kw-\triangledown kw^{'}|_1+\frac{\alpha_0}{2}|\triangledown kw^{'}-\triangledown kw^{'T}|_1\large]\small...[2]$$

where $$$\tiny\triangle \small M_{b-value}^{PLD}$$$ is ASL signal with PLD and b-value indicated by superscript and subscript. Vascular signal was suppressed with b_ATT=14s/mm² (VENC=7.5mm/s) at PLD=900ms,⁵ and tissue/capillary fraction of the ASL signal at PLD=1800ms was separated by b_DW=50s/mm².⁶ Function g( ) calculates ATT based on the FEAST technique⁵ and f( ) describes the monotonic relationship between kw and the capillary fraction at a given ATT.⁶ λ= 0.05 is the weighting factor balancing data fidelity and TGV penalty function, while α₁=1 and α₀=2, balances between the first and second derivative of ATT and kw maps.⁴ Nineteen aged subjects (7 male, age=68.8±7.6 yrs) underwent two MRIs approximately 2 weeks apart on a Siemens 3T Prisma system (Erlangen, Germany) using a 20-channel head coil. Subjects underwent a physical exam, medical history evaluation and blood draw before the first MRI scan. Descriptions of clinical assessments are summarized in Table 1. Correlations between average kw from both test and retest scans and clinical/behavioral assessments were evaluated using mixed effects linear regression model incorporating age and gender as covariates and time (test/retest) as the random variable.

Results and discussion

Figure 2 shows the comparison of ATT and kw maps estimated using direct SPA modeling and the proposed SPA modeling with TGV regularization respectively. TGV regularized SPA modeling preserved the original image resolution. The local bright regions (indicated by red arrows, kw>200 min^-1) with spuriously high kw values in direct SPA modeling were suppressed by TGV regularized SPA modeling.

Figure 3 (a,b) shows six slices of kw maps from test-retest scans (global kw = 95.3 and 96.5 min^-1) of one representative subject (Female,64yrs). Good test-retest reproducibility (ICC=0.75) was achieved for the proposed sequence across nineteen subjects. Average kw values from repeated scans were shown in figure 3 (c).

Increased kw was found in subjects with type 2 diabetes (β=25.7,P<0.001) (figure 4(a)) and hypercholesterolemia (β=17.8,P=0.04) (figure 4(b)), which is consistent with DCE-MRI⁷ and biochemical studies.⁸ Increased kw was found in subjects with higher vascular risk factors (β=9.4, P=0.02) (figure 4(c)). Both CDR scores were significant predictors of kw (CDR-GS: β=44.6,P=0.002; CDR-SB: β=21.0,P=0.001) (figure 4(e, f)), which indicates increased BBB permeability is associated with a greater severity of functional impairment. NIH toolbox measurements: DCCs (β=-1.10,P=0.02), PSMTa (β=-0.98,P=0.03) and PSMTb (β=-1.19,P=0.001) were significant correlated with kw, and a trend of negative correlation was found between Flanker and kw (β=-0.58,P=0.08) (figure 4(g-j)). kw was also significantly correlated with the Fazekas scale of WMH (β=10.61,P=0.04) (figure 4(d)), which indicates kw is associated with severity of WMH.

Conclusion

Acknowledgements

References

[1] Paulson OB. Blood-brain barrier, brain metabolism and cerebral blood flow. Eur Neuropsychopharmacol 2002;12(6):495-501.

[2] Reese T G, Heid O, Weisskoff R M, et al. Reduction of eddy‐current‐induced distortion in diffusion MRI using a twice‐refocused spin echo[J]. Magnetic Resonance in Medicine: An Official Journal of the International Society for Magnetic Resonance in Medicine, 2003, 49(1): 177-182.

[3] Alsop D C. Phase insensitive preparation of single‐shot RARE: Application to diffusion imaging in humans[J]. Magnetic resonance in medicine, 1997, 38(4): 527-533.

[4] Knoll F, Bredies K, Pock T, et al. Second order total generalized variation (TGV) for MRI[J]. Magnetic resonance in medicine, 2011, 65(2): 480-491.

[5] Wang J, Alsop D C, Song H K, et al. Arterial transit time imaging with flow encoding arterial spin tagging (FEAST)[J]. Magnetic resonance in medicine, 2003, 50(3): 599-607.

[6] St Lawrence K S, Owen D, Wang D J J. A two-stage approach for measuring vascular water exchange and arterial transit time by diffusion‐weighted perfusion MRI[J]. Magnetic resonance in medicine, 2012, 67(5): 1275-1284.

[7] Starr JM, Wardlaw J, Ferguson K, MacLullich A, Deary IJ, Marshall I. Increased blood-brain barrier permeability in type II diabetes demonstrated by gadolinium magnetic resonance imaging. J Neurol Neurosurg Psychiatry 2003;74(1):70-76.

[8] Acharya NK, Levin EC, Clifford PM, Han M, Tourtellotte R, Chamberlain D, Pollaro M, Coretti NJ, Kosciuk MC, Nagele EP, Demarshall C, Freeman T, Shi Y, Guan C, Macphee CH, Wilensky RL, Nagele RG. Diabetes and hypercholesterolemia increase blood-brain barrier permeability and brain amyloid deposition: beneficial effects of the LpPLA2 inhibitor darapladib. J Alzheimers Dis 2013;35(1):179-198.