Quantification of blood-brain barrier water exchange and permeability with multidelay diffusion-weighted pseudo-continuous arterial spin labeling - PubMed (original) (raw)

Quantification of blood-brain barrier water exchange and permeability with multidelay diffusion-weighted pseudo-continuous arterial spin labeling

Xingfeng Shao et al. Magn Reson Med. 2023 May.

Abstract

Purpose: To present a pulse sequence and mathematical models for quantification of blood-brain barrier water exchange and permeability.

Methods: Motion-compensated diffusion-weighted (MCDW) gradient-and-spin echo (GRASE) pseudo-continuous arterial spin labeling (pCASL) sequence was proposed to acquire intravascular/extravascular perfusion signals from five postlabeling delays (PLDs, 1590-2790 ms). Experiments were performed on 11 healthy subjects at 3 T. A comprehensive set of perfusion and permeability parameters including cerebral blood flow (CBF), capillary transit time (τc ), and water exchange rate (kw ) were quantified, and permeability surface area product (PSw ), total extraction fraction (Ew ), and capillary volume (Vc ) were derived simultaneously by a three-compartment single-pass approximation (SPA) model on group-averaged data. With information (i.e., Vc and τc ) obtained from three-compartment SPA modeling, a simplified linear regression of logarithm (LRL) approach was proposed for individual kw quantification, and Ew and PSw can be estimated from long PLD (2490/2790 ms) signals. MCDW-pCASL was compared with a previously developed diffusion-prepared (DP) pCASL sequence, which calculates kw by a two-compartment SPA model from PLD = 1800 ms signals, to evaluate the improvements.

Results: Using three-compartment SPA modeling, group-averaged CBF = 51.5/36.8 ml/100 g/min, kw = 126.3/106.7 min-1 , PSw = 151.6/93.8 ml/100 g/min, Ew = 94.7/92.2%, τc = 1409.2/1431.8 ms, and Vc = 1.2/0.9 ml/100 g in gray/white matter, respectively. Temporal SNR of MCDW-pCASL perfusion signals increased 3-fold, and individual kw maps calculated by the LRL method achieved higher spatial resolution (3.5 mm3 isotropic) as compared with DP pCASL (3.5 × 3.5 × 8 mm3 ).

Conclusion: MCDW-pCASL allows visualization of intravascular/extravascular ASL signals across multiple PLDs. The three-compartment SPA model provides a comprehensive measurement of blood-brain barrier water dynamics from group-averaged data, and a simplified LRL method was proposed for individual kw quantification.

Keywords: blood-brain barrier (BBB); diffusion-weighted arterial spin labeling (DW-ASL); water exchange rate (kw); water permeability (PSw).

PubMed Disclaimer

Figures

Figure 1.

A. Simulated residue functions of the tissue and vascular compartments according to the SPA model. Vascular contribution decays fast before _τ_c due to rapid exchange into tissue compartment. B. ASL signals were calculated as the convolution between AIF (Eq. [2]) and residue functions. t indicates the time from the beginning of ASL labeling. ASL signal is zero before bolus arrival, builds up from t = ATT, and then decays after the duration of labeling (t = ATT + δ). Decay of the vascular compartment is faster than the tissue compartment due to water exchange. An inset (dashed box) shows the tissue and combined ASL signals within the acquisition window (PLD ~ 1.5 sec to 3 sec). Black, red, and blue traces indicate combined, tissue and vascular compartment, respectively. Parameters used for this simulation are: R1a = 0.6 sec−1, R1t = 0.77 sec−1, ATT = 1.3 sec, δ = 1.87 sec, CBF = 60 ml/100g/min, kw = 133.3 min-1, Vc = 1.5 ml/100g, _τ_c = 1.5 sec and PSw = 200 ml/100g/min.

Figure 2.

Simulated ΔMc(t) (A) and log(ΔMc(t)) (B) with three kw values (50, 100 and 150 min−1) according to Eq. [10]. ΔMc(t) decays faster and slope (S) of log(ΔMc(t)) is greater with higher kw values. C shows the plot of kw and S, and kw can be determined by estimating the slope of log(ΔMc(t)). R1a was assumed to be 0.6 sec−1. Dashed lines show the kw versus S plot with ±20% variation in R1a. Parameters used for this simulation: ATT = 1.3 sec, CBF = 60 ml/100g/min.

Figure 3.

A. Sequence diagram of pCASL. A pre-saturation pulse at the beginning of labeling and two non-selective HS pulses within PLD were used for background suppression. B. Implementation of M1 compensated and diffusion weighted (MCDW) GRASE. The original slice crusher pairs (yellow lobes) of the first refocusing pulse were replaced by three-lobe M1-motion compensated gradients (blue lobes) with maximal amplitude and prolonged duration to achieve sufficient diffusion weighting (blue lobes). Ramp time = 0.74 msec, gradient duration = 2.85/5.7 msec for A and 2A lobes, δt = 7.68 msec and effective b-value = 40.4 sec/mm2. Gradient amplitude was reduced to 4.6 mT/m for the remaining crusher pairs (b-value = 0.3 sec/mm2) (orange lobes). 180° sinc-shaped pulses were used for GRASE refocusing while a BIR4 non-selective refocusing pulse was used as the first refocusing pulse to improve the robustness to B1+/B0 field inhomogeneities.

Figure 4.

A. Magnitude and phase of optimized BIR4 non-selective refocusing pulse. Amplitude and frequency modulated hyperbolic tangent functions were optimized for a range of B1+ (80% to 130%) and B0 (−500 Hz to 500 Hz) with constraints of pulse duration (7.68 msec) and maximal scanner transmit voltage (550 V). B. Bloch simulation of magnetization (Mx, My, Mz) being inverted from +Y (0, 1, 0) to −Y (0, −1, 0) axis by BIR4 refocusing pulse. C. Heatmap of 1+My reflecting robustness of refocusing in existence of B1+/B0 field inhomogeneities, while zero indicates perfect refocusing. D. Diffusion weighted (b=40.4 sec/mm2) perfusion images acquired at PLD = 1890 msec. Imaging parameters were kept identical between two rows except different first refocusing pulses. The proposed BIR4 refocusing pulse significantly improved motion robustness and reduced artifacts (white arrows) as compared to a conventional sinc-shaped 180° refocusing pulse.

Figure 5.

Group-averaged perfusion images at 5 PLDs (columns) in MNI space. The first and second row show perfusion maps without (vascular+tissue) and with (tissue) diffusion weighting. Perfusion signal intensities decreased at longer PLDs. ΔMb=0 is overall higher than ΔMb=40.4. The difference (vascular) image is shown in the third row. Vascular contribution was relatively higher in gray matter at PLD = 1590 msec (gray arrow) and became relatively higher in WM at PLD = 2790 msec (white arrow). Color scale indicates the intensity of the perfusion signal in relation to M0.

Figure 6.

A. Plot of perfusion signals and fitting results in GM (top) and WM (bottom). ATT values were obtained from DP-pCASL and T1 values were obtained from 5-PLD control signals. CBF, _τ_c and kw were simultaneously fitted by 3-compartment SPA model. PSw and Vc were derived according to Eqs [5,6]. Black-square and red-triangle marks indicate subject-averaged perfusion signal without (tissue+vascular) and with (tissue) diffusion weighting, and black and red traces indicate simulated perfusion signals with fitted parameters. Black arrows indicate the under-estimated perfusion signals at long PLDs especially in white matter, which is suspected to be caused by underestimated tissue T1 values. B. Results of 3-compartment SPA fitting including tissue T1 as an unknown parameter. Fitted T1 values were 1438.6 msec and 1250.3 msec in GM and WM, which were longer than T1 values obtained from DP-pCASL (1304.3 msec and 935.4 msec). RSMEs were significantly reduced indicating improved accuracy of model fitting. C. Plots of vascular fraction in GM (top) and WM (bottom). Black-square marks indicate subject-averaged vascular fraction measured at 5 PLDs, and black traces indicate simulated vascular fractions with fitted parameters as shown in B. Measured vascular fractions were 19.1%, 15.2%, 11.5%, 7.4%, 5.4% in GM and 23.9%, 20.9%, 12.4%, 8.0%, 11.2% in WM at 5 PLDs. Simulated vascular fractions were 21.7%, 12.7%, 8.2%, 6.4%, 6.0% in GM and 25.6%, 17.1%, 12.6%, 10.7%, 10.8% in WM at 5 PLDs. Tissue fraction can be calculated as 1-vascular fraction, which were summarized in Table 1. Error bars indicate standard deviation across 11 subjects.

Figure 7.

Results of 3-compartment SPA modeling. A. Parameters obtained prior to model fitting. T1 values were estimated from background suppressed control images acquired at 5 PLDs. ATT map was obtained from DP-pCASL scan and normalized to MNI space. B. CBF, _τ_c and kw maps were simultaneously computed by 3-compartment SPA fitting of the 5-PLD ΔMb=0 and ΔMb>0 signals. C. Two derived parameters Vc and PSw calculated according to Eqs. [5,6].

Figure 8.

Five representative slices of reconstruction results from one representative subject. (Female, 25 years old). A and B show CBF and kw maps obtained from the MCDW-pCASL LRL (top) and DP-pCASL (bottom) techniques. C and D show Ew and PSw maps obtained from the proposed technique. E shows the ATT map obtained from DP-pCASL. 36 and 12 slices were acquired by the proposed MCDW-pCASL and DP-pCASL, respectively. And reconstruction results with full slices can be found in Supporting Figure S1.

Figure 9.

Comparison of whole brain average CBF (A) and kw (B) acquired by the MCDW-pCASL (X-axis) and DP-pCASL (Y-axis).

References

1. Sweeney MD, Zhao Z, Montagne A, Nelson AR, Zlokovic BV. Blood-brain barrier: from physiology to disease and back. Physiological reviews 2019;99(1):21–78. -PMC -PubMed
1. Farrall AJ, Wardlaw JM. Blood-brain barrier: ageing and microvascular disease--systematic review and meta-analysis. Neurobiol Aging 2009;30(3):337–352. -PubMed
1. Ingrisch M, Sourbron S, Morhard D, Ertl-Wagner B, Kümpfel T, Hohlfeld R, Reiser M, Glaser C. Quantification of perfusion and permeability in multiple sclerosis: dynamic contrast-enhanced MRI in 3D at 3T. Investigative radiology 2012;47(4):252–258. -PubMed
1. Heye AK, Culling RD, Valdes Hernandez Mdel C, Thrippleton MJ, Wardlaw JM. Assessment of blood-brain barrier disruption using dynamic contrast-enhanced MRI. A systematic review. Neuroimage Clin 2014;6:262–274. -PMC -PubMed
1. Montagne A, Toga AW, Zlokovic BV. Blood-Brain Barrier Permeability and Gadolinium: Benefits and Potential Pitfalls in Research. JAMA Neurol 2016;73(1):13–14. -PMC -PubMed

Quantification of blood-brain barrier water exchange and permeability with multidelay diffusion-weighted pseudo-continuous arterial spin labeling - PubMed (original) (raw)