Cerebrospinal fluid (CSF) T2 mapping can potentially be used to investigate CSF composition. A previously proposed CSF T2–mapping method reported a T2 difference between peripheral and ventricular CSF, and suggested that this reflected different CSF compositions. We studied the performance of this method at 7 T and evaluated the influence of partial volume and B1 and B0 inhomogeneity.
Materials and methods
T2-preparation-based CSF T2-mapping was performed in seven healthy volunteers at 7 and 3 T, and was compared with a single echo spin-echo sequence with various echo times. The influence of partial volume was assessed by our analyzing the longest echo times only. B1 and B0 maps were acquired. B1 and B0 dependency of the sequences was tested with a phantom.
Results
T2,CSF was shorter at 7 T compared with 3 T. At 3 T, but not at 7 T, peripheral T2,CSF was significantly shorter than ventricular T2,CSF. Partial volume contributed to this T2 difference, but could not fully explain it. B1 and B0 inhomogeneity had only a very limited effect. T2,CSF did not depend on the voxel size, probably because of the used method to select of the regions of interest.
Conclusion
CSF T2 mapping is feasible at 7 T. The shorter peripheral T2,CSF is likely a combined effect of partial volume and CSF composition.
The online version of this article (doi:10.1007/s10334-017-0659-3) contains supplementary material, which is available to authorized users.
Introduction
Qin [1] proposed a fast 3-T MRI method to map the volume and T2 of cerebrospinal fluid (CSF) in the brain. A striking finding with this method was the observation of a shorter T2 in the peripheral CSF compared with the T2 of the CSF in the lateral ventricles. Qin suggested that this T2 difference is caused by differences in CSF composition between both areas, implying that CSF T2 (T2,CSF) can be used as a noninvasive biomarker for CSF composition. This would be highly relevant in the light of the recent attention to the clearance of brain waste products, in which CSF is involved [2‐5]. A method that can noninvasively assess CSF composition would provide a noninvasive window on the brain clearance system, with great potential for applications in studying diseases related to dementia such as Alzheimer’s disease and cerebral small vessel disease. If T2,CSF is indeed useful as a functional marker of the brain clearance system, it could be studied next to other advanced imaging markers of early brain damage such as microbleeds, microinfarcts, and hippocampus subfield volumes and atrophy. As many of these advanced markers are acquired at 7 T [6‐8], it is desirable to implement and evaluate CSF T2 mapping at 7 T as well.
At 7 T, B1 inhomogeneity is considerable and may influence the T2 mapping results, despite the relative B1 insensitivity of the used CSF T2 mapping method. Even at 3 T, considerable B1 inhomogeneity in the brain can be observed [9]. Also, when T2 is measured in peripheral CSF, partial volume effects with tissue cannot be avoided. So, we hypothesized that these partial volume effects and B1 imperfections can explain the previously observed T2 differences. De Vis et al. [10] obtained a rough estimation of the influence of partial volume effects on the estimated T2,CSF by scanning with two different resolutions. The higher resolution resulted in longer T2 times, suggesting a role for partial volume effects. Because the influence of B1 inhomogeneity and partial volume effects is not clear yet, it remains uncertain to what extent T2,CSF can be used to assess the composition of CSF.
Anzeige
In this work we studied the performance of Qin’s CSF T2 mapping method at 7 T. The specific goals were to investigate the influence of B1 and B0 imperfections on the estimated T2,CSF, to assess the influence of partial volume effects, and to evaluate to what extent the previously observed difference in T2,CSF between periphery and ventricles can be explained by B1, B0, and partial volume effects. B1 and B0 sensitivity was investigated with phantom measurements, and by comparison of the method for 7 and 3 T in healthy humans. Partial volume effects were estimated by removal of the influence of partial volume with tissue, through selection of only the last (longest) echo times (TEs). Also, scanning was done at different resolutions.
Materials and methods
Sequence
The CSF T2 mapping sequence used in this research is based on T2 preparation, and has been described elsewhere [1, 10]. For this study the method was further extended to improve the fit reliability of the long T2 times by implementation of longer refocusing pulse trains, yielding longer TEs. Briefly, the sequence consists of four parts (Fig. 1). First, a set of four nonselective water suppression enhanced through T1 effects (WET) pulses are applied for saturation to prevent slice history effects. These pulses are optimized for saturation of free water (applicable for T1 between 3 and 6 s); the pulse angles are 156°, 71°, 109°, and 90° [11]. Second, a delay time (Tdelay) follows, where T1 relaxation occurs, followed by crusher gradients. Third, T2 preparation is applied, consisting of a nonselective 90° pulse, a set of 4, 8, 16, or 32 nonselective refocusing pulses (R) according to the Malcolm Levitt (MLEV) phase cycling scheme, and a nonselective −90° pulse with a crusher gradient to crush any remaining transverse magnetization [12]. Each refocusing pulse R is a composite pulse, consisting of \(90_{\text{x}}^{ \circ } , \, 180_{\text{y}}^{ \circ } , \, 90_{\text{x}}^{ \circ }\) rectangular pulses (or the inverse \({\bar{\text{R}}}:90_{\text{ -x}}^{ \circ } , \, 180_{\text{ -y}}^{ \circ } , \, 90_{\text{ -x}}^{ \circ }\)). The duration of a single refocusing pulse was 2.6 ms. T2 relaxation occurs during TET2-prep, which is determined by the number of refocusing pulses and the spacing between the centers of the refocusing pulses (τ). To achieve long TET2-prep times, τ was chosen as 150 ms. This resulted in TET2-prep durations of 600, 1200, 2400, and 4800 ms. Also, one scan was acquired without any refocusing pulses; this scan was not used in data analysis. The fourth part of the sequence is a single-shot 2D spin echo (SE) echo planar imaging (EPI) readout. During the EPI train, T2 decay also occurs; this can, however, be regarded as a constant factor, and was therefore disregarded in the analysis.
Fig. 1
Cerebrospinal fluid T2 mapping pulse sequence. The sequence consists of four parts: water suppression enhanced through T1 effects (WET) presaturation, a fixed delay with duration Tdelay, a T2 preparation module with Malcolm Levitt (MLEV) phase cycling, and a spin echo echo planar imaging (SE-EPI) image acquisition. T2 relaxation occurs during TET2-prep. To perform T2 mapping the sequence was repeated, while the number of refocusing pulses (and therefore TET2-prep) was increased for a fixed interpulse delay τ. The applied RF pulses are shown on the RF axis, and the applied gradients are shown on the frequency (F), phase (P), and slice (S) encoding axes
×
Although the refocusing train in the T2 preparation is relatively insensitive to B1 inhomogeneities because of the MLEV phase cycling scheme, the 90° rectangular pulses before and after the train may fail in the case of B1 deviations. Consequently, we hypothesized that a fraction of the magnetization may be unaffected by the T2-preparation module, which could have a relatively large impact on the measured signal in the case of partial volume effects. Also, in the case of imperfect B1, T1-weighted stimulated echoes may influence the T2 measurements, although this effect is expected to be relatively small because of the long T1 of CSF (4.4 s [13]). Therefore, T2 mapping with a single echo SE-EPI sequence with various TEs was used as a truly B1-insensitive reference (shown in the electronic supplemental material).
Nonselective pulses were used where possible to minimize motion sensitivity. Consequently, only the excitation pulse of the SE-EPI readout was selective.
Anzeige
Phantom measurements
Phantom measurements were performed to test the B1 and B0 sensitivity. The experiments were performed with a 7-T Philips Achieva scanner (Philips Medical Systems, Best, Netherlands) with a 32-channel head coil (Nova Medical, Wilmington, MA, USA), with a tap water phantom at room temperature. The phantom size was 200 × 95 × 20 mm3. The phantom was scanned with the same TEs as used for the in vivo experiments. A single slice was acquired with 3 × 3 × 6 mm3 resolution, field of view (FOV) of 240 × 96 mm2, Tdelay of 15 s, [which is more than three times the T1 of CSF (4.4 s [13])], and sensitivity encoding (SENSE) [14] (with SENSE factor 1, meaning that the coil sensitivities of the receive coils were used for optimal coil combination without imaging acceleration). A series of T2 maps with increasing through-plane B0 gradients were acquired to study the effect of diffusion for B1 of 100%. The following through-plane B0 gradient strengths were applied by addition of this strength to the linear shim term in the user interface: 0, 0.05, 0.1, 0.2, 0.3, and 0.5 mT/m. The phantom appeared sensitive to free induction decay (FID) artifacts, because of the relatively large volume of water and more pronounced B1 inhomogeneity [15]. The acquisition with the highest B0 gradient yielded images free of FID artifacts, which allowed us to study the B1 dependency of the CSF T2 mapping sequence. B0 and B1 maps were acquired, with use of identical resolution and FOV as for the T2 mapping acquisitions. The B0 map was obtained with a gradient echo sequence with two different TEs (1.64 and 2.64 ms). The B1 mapping sequence was based on the actual flip angle method [16], with first repetition time (TR) of 40 ms, second TR of 160 ms, TE of 0.96 ms, and a flip angle of 50°.
In vivo measurements
In vivo experiments were performed at both 7 and 3 T to test the feasibility of CSF T2 mapping at 7 T, to further assess the sensitivity to B1 inhomogeneities, and to explore the influence of partial volume effects. Seven healthy volunteers (three men, mean age 34 ± 11 years, age range 21–54 years) participated in this study. Informed consent was given by all volunteers in accordance with the requirements of the Institutional Review Board of the University Medical Center Utrecht (Utrecht, Netherlands). All volunteers were scanned with both a 3-T Philips Achieva scanner with an eight-channel head coil (Philips Healthcare, Best, Netherlands) and the 7-T scanner that was also used for the phantom study. The CSF T2 mapping scans were acquired in a single coronal slice, planned through both the lateral ventricles and the fourth ventricle (Fig. 2a). The scanning parameters are summarized in Table 1. The fixed Tdelay was 15 s, and TR ranged between 20 and 25 s depending on TET2-prep. Other parameters were as follows: SENSE factor 2.3 in the left–right direction and FOV of 240 × 240 mm2. Because of the long TR, the specific absorption rate (SAR) remained well within the specific absorption rate limits, also at 7 T. No additional methods were used to correct the scans for eddy currents. The low bandwidth of the scan may cause distortions in areas with poor shimming, such as the nasal cavities. However, shimming was good in the selected coronal slice. Also B0 and B1 maps were acquired.
Fig. 2
a Planning of the cerebrospinal fluid (CSF) T2 mapping scans, through the lateral ventricles and the fourth ventricle. b CSF T2 mapping scans at 7 T for increasing echo times (TET2-prep) (0.6, 1.2, 2.4, and 4.8 s), shown with equal intensity scaling. c The region of interest masks used: the periphery (PER; white), the lateral ventricles (LAT; yellow), and the fourth ventricle (FOU; red)
Table 1
Scan parameters used for the in vivo experiments for the cerebrospinal fluid T2 mapping sequence (based on T2 preparation)
Resolution (mm3)
TEreadout (ms)
TET2-prep (ms)
EPI factor
Bandwidth (phase/frequency) (Hz/voxel)
Scan duration (min)
3 T
1 × 1 × 4
133
0–4800a
105
8.1/961
2:59
3 × 3 × 6
42
0–4800a
67
28.2/2308
2:59
7 T
1 × 1 × 2
127
0–4800a
107
8.4/1082
3:04
1 × 1 × 4
126
0–4800a
107
8.4/1082
3:04
3 × 3 × 6
23
0–4800a
37
55.1/2642
3:04
EPI echo planar imaging, TE echo time, TET2-prep echo time of T2-preparation
aThe TET2-prep values used were 0, 600, 1200, 2400, and 4800 ms.
×
Data analysis
Phantom
The resulting T2 estimates were analyzed as a function of B1 and B0. B1 sensitivity was assessed with use of the scans with the highest B0 gradient strength (0.5 mT/m), where no FID artifacts were present. The B1 range present in the scan was used. On the basis of B1 in each voxel, the voxels were sorted over eight bins of 5% B1, leading to B1 bins ranging from (85 ± 2.5)% to (120 ± 2.5)%. The signal was averaged over each B1 bin, and T2 values were fitted over this averaged signal. B0 sensitivity was assessed with the various applied B0 gradient strengths (0, 0.05, 0.1, 0.2, 0.3, and 0.5 mT/m). In each scan, only voxels with B1 between 97.5% and 102.5% were included. An additional intensity threshold was applied on the scan with the longest TET2-prep to minimize the influence of artifacts in the lower B0 gradient scans. This threshold was set at 75% of the maximum intensity for the longest TE. The signal of all voxels included was averaged over each scan, and T2 values were fitted over this averaged signal.
In vivo
Three regions of interest (ROIs) were defined on the acquired in vivo scans: the lateral ventricles, the fourth ventricle, and peripheral CSF. The ROI masks were made by our applying an intensity threshold to the first TE (TET2-prep = 0.6 s). The intensity threshold was set at 25% of the maximum intensity in the image. Figure 2b and c shows the acquired CSF T2 mapping scan at 7 T with a resolution of 1 × 1 × 4 mm3 at all TEs for one volunteer, and the ROIs used. Conservative ROIs were used in the ventricles by our eroding the intensity-based ROIs with one voxel to minimize both partial volume and motion sensitivity. Erosion of the peripheral ROIs was not feasible.
The signal was averaged over each ROI, and T2 values were fitted over this averaged signal, with use of a single exponential decay model. Also mean B0 and B1 values were determined for each ROI. To minimize the influence of, for example, motion or partial volume effects on the data analysis, only fit results with R2 of 0.99 or higher were considered.
Partial volume assessment
In the peripheral CSF an additional assessment of the influence of partial volume was made by our performing a partial volume correction. Only TET2-prep values of at least 1200 ms (excluding the shortest TET2-prep of 600 ms) were taken into account in the analysis. Thereby, maximal nulling of, for example, tissue signal was achieved, since the T2 values of tissue are below 100 ms [17, 18], about ten times shorter than the minimal TET2-prep used. The analysis with only the last TET2-prep times was also performed on the phantom scans to check for any systematic errors for all B0 gradient strengths applied and B1 between 97.5% and 102.5%.
All data analysis was performed in MATLAB (version 2015B, The MathWorks, Natick, MA, USA). IBM SPSS Statistics (version 21.0) was used for statistical analysis. Median T2,CSF values and full ranges are reported. Wilcoxon signed-rank tests (significance level p < 0.05) were used to compare CSF T2 values in the lateral and fourth ventricles with those in the periphery to explore the observed T2 differences.
Results
Phantom measurements
Figure 3 shows the results of the phantom measurements for the B1 dependency (Fig. 3a) and B0 gradient dependency (Fig. 3b). The CSF T2 mapping sequence measured a T2 of 1.71 s (95% confidence interval 1.66–1.76 s) for B1 of (100 ± 2.5)% and B0 gradient strength of 0 mT/m. The sequence showed only minor B1 sensitivity (assessed in the scans with B0 gradient strength of 0.5 mT/m), with T2 ranging from 1.41 s (95% confidence interval 1.38–1.43 s) at B1 of (85 ± 2.5)% to 1.49 s (95% confidence interval 1.40–1.57 s) at B1 of (105 ± 2.5)%. Also minor B0 gradient dependency was observed.
Fig. 3
Results of the phantom measurements for the B1 (a) and B0 gradient dependency (b) showing the fitted T2 for different B1 values and through-plane B0 gradient strengths, respectively. The error bars show the 95% confidence interval of the fitted T2. The cerebrospinal fluid T2 mapping sequence shows only minor sensitivity to B1 and to the through-plane B0 gradient (and thus to diffusion). The effect of the B0 gradient was most apparent for the highest B0 gradient. The B1 dependency was determined with a B0 gradient strength of 0.5 mT/m to avoid free induction decay (FID) artifacts in regions with B1 deviating from 100%. For a B0 gradient of 0.2 mT/m, the confidence interval was greater because of FID artifacts
×
In vivo measurements
Thirty-five CSF T2 mapping scans were acquired, for both field strengths and the different resolutions. Per scan, three fits were made, one per ROI, resulting in a total of 105 fits (42 at 3 T, 63 at 7 T). On the basis of the strict requirement on minimum R2, 14 fits were excluded (four at 3 T, ten at 7 T), which corresponds to 13% of the total number of fits (10% at 3 T, 16% at 7 T); see Table 2 for a detailed overview.
Table 2
Number of scans acquired and T2 fits performed, and the number of excluded T2 fits per region of interest
Resolution (mm3)
Scans
Fits
Excluded fits
Lateral ventricles
Fourth ventricle
Periphery
3 T
1 × 1 × 4
7
21
0
1
0
3 × 3 × 6
7
21
2
1
0
Total
14
42
2
2
0
7 T
1 × 1 × 2
7
21
0
1
0
1 × 1 × 4
7
21
1
3
0
3 × 3 × 6
7
21
1
4
0
Total
21
63
2
8
0
The in vivo results for the scans with a resolution of 1 × 1 × 4 mm3 are summarized in Fig. 4. The results for the other resolutions were not significantly different from the data shown here (all data are shown in Tables S3, S4, S5). Although T2 differences between the resolutions were not significant, in most cases the shortest T2 times were observed for the largest voxel sizes.
Fig. 4
In vivo results: T2 (a), B1 (b), and B0 gradient (c) for the three different regions of interest. Outliers are represented by a square. Significant differences in measured T2 were found between the periphery and the lateral and fourth ventricles at 3 T (indicated by an asterisk)
×
At 7 T significantly shorter T2 times were found compared with at 3 T. At 3 T the T2 times measured in the periphery were significantly shorter than those measured in the lateral and fourth ventricles. The T2 times measured at 7 T were not significantly different between the three ROIs. At 3 T the median B1 in the periphery was 85% (range 79–90%), while in the lateral and fourth ventricles it was 109% (ranges 106–112% and 103–114%). The median B0 gradient in the periphery was 0.13 mT/m (range 0.07–0.38 mT/m), while in the lateral and fourth ventricles it was 0.02 and 0.03 mT/m, respectively (range 0.02–0.06 mT/m and 0.01–0.03 mT/m, respectively). At 7 T, lower B1 values were observed in the periphery and the fourth ventricle [median 86% (range 75–94%) and 93% (range 62–101%), respectively], and higher B1 values were observed in the lateral ventricles [median 111% (range 109–116)]. Similar B0 gradients were observed in the three ROIs [median 0.07 mT/m (range 0.03–0.10 mT/m), 0.06 mT/m (range 0.04–0.08 mT/m), and 0.06 mT/m (range 0.05–0.09 mT/m), for the lateral ventricles, the fourth ventricle, and the periphery, respectively].
Partial volume assessment
Figure 5 shows the results for the additional analysis of peripheral CSF to assess the influence of partial volume. The partial volume correction resulted in longer T2 times, with a significant increase of 118 ms at both 3 and 7 T. At 7 T the corrected peripheral CSF T2 was quite similar to the ventricular T2 (1.01 s vs 1.05 s), while at 3 T the mean peripheral T2 was still approximately 200 ms shorter than the ventricular T2 times [1.79 s (range 1.49–1.82 s) vs 2.03 s (range 1.73–2.16 s), p = 0.02]. The results for this analysis of the phantom data are shown in Fig. 6. Both analysis methods (including all TEs or only the longest TEs) resulted in similar T2 values, indicating no systematic errors in the additional analysis with only the longest TEs.
Fig. 5
T2 values of peripheral cerebrospinal fluid resulting from the use of only the longest echo times (TEs) compared with the original analysis. Outliers are represented by a square. At both 3 and 7 T an increase in T2 can be observed. The asterisk indicates a significant difference with the original analysis (including all TEs)
Fig. 6
T2 values of the phantom, resulting from the use of only the longest echo times (TEs; orange) compared with the original analysis (blue). Both analyses result in similar T2 values
×
×
Discussion
In this research we have shown the feasibility of CSF T2 mapping at 7 T with a dedicated CSF T2 mapping sequence based on T2 preparation, which was initially developed at 3 T. We investigated the sensitivity of this sequence for the influence of B1, diffusion (B0 gradient), and partial volume effects. The sequence appeared to be relatively insensitive to B1 and B0 inhomogeneity. Partial volume effects tend to lower the observed T2 values at the periphery. T2,CSF was considerably shorter at 7 T than at 3 T in all three ROIs. The peripheral T2,CSF was significantly shorter than the ventricular T2,CSF at 3 T (but not at 7 T).
Anzeige
The peripheral T2,CSF increased considerably on partial volume correction, as obtained from analysis of long TEs (more than ten times the tissue T2). The partial volume correction for the SE-EPI sequence, which was used as a relatively B1-insensitive reference (data shown in the electronic supplementary material), did not significantly increase T2 values, although the SE-EPI sequence showed an even larger T2 difference between the periphery and ventricles. The ventricular T2 values measured with the CSF T2 mapping sequence at 3 T match with T2 values found in literature [1, 10, 19, 20]. Given the results from our measurements and analysis, we believe that the observed T2 difference between the ventricular and peripheral CSF could be partly due to physiological differences. However, the different results for different sequences and field strengths and the confounding influence of partial volume effects will make it challenging to accurately isolate and quantify any true physiological effect from confounders. This will hamper applications in research focusing on in vivo evaluation of the (regional) composition of CSF.
B1 and B0 dependency
In the phantom measurements only minor B1 dependency was found for the CSF T2 mapping sequence, as shown in Fig. 3a. Also, the measurements with an increasing through-plane B0 gradient showed only limited B0 gradient dependency, except for the highest B0 gradient (0.5 mT/m), as shown in Fig. 3b. In the in vivo measurements, the B0 gradient was similar between the periphery and the ventricles at 7 T, and differed by a maximum of 0.38 mT/m (median B0 gradient was 0.13 mT/m) at 3 T (Fig. 4c). This difference in B0 homogeneity between 3 and 7 T is probably due to different shimming techniques: image-based third-order shimming was used at 7 T, and linear shimming was used at 3 T. The low sensitivity to B0 gradient shows that the T2 mapping sequence is relatively insensitive to diffusion. It is not likely that B0 gradients due to imperfect shimming contributed considerably to the observed difference in T2 between periphery and ventricles.
Partial volume effects
The different resolutions used at both field strengths did not yield considerably different T2 values (Tables S3, S4, S5), although there is a trend of longer measured ventricular T2 times for smaller voxel sizes at 3 T, similarly to what was found by to De Vis et al. [10]. As the ventricular ROIs were eroded, the voxels at the edges, where more partial volume is expected, were discarded. For the periphery, however, erosion was not feasible because of the thin shape of the ROI. Moreover, the ROI definition was based on an intensity threshold, which depends on the CSF fraction in each voxel. Since the total subarachnoidal CSF volume is quite small, and distributed over a relatively large area [21], partial volume is probably present in all peripheral ROIs, independently of the voxel sizes used in this work.
The role of partial volume effects regarding the measured peripheral T2,CSF was investigated by use of the longest TEs only (Fig. 5) to maximally remove the influence of partial volume. It could seem unexpected that the use of the late TEs reveals a considerable partial volume effect, since the first TET2-prep is already relatively long compared with the tissue T2. The T2 of gray matter is approximately 90 ms at 3 T [18, 22] and 55 ms at 7 T [18, 23], while the first TE was 600 ms. However, it is possible that partial volume occurs with a compartment with a relatively long T2 in the cerebral cortex, like arterial blood (T2 around 150 ms at 3 T [24, 25]) or the outer rim of the cortex (unknown but long T2, greater than 100 ms, at 7 T [26]). At the shortest TET2prep (600 ms), the signal of arterial blood has decayed to 2%. However, in the case of small partial volume fractions of CSF in the periphery, this could still have a considerable influence on the measured T2. The outer layer of the cerebral cortex (layer I) may have a long T2 because it contains almost no neuronal cell bodies, and many glial cells instead, similarly to gliotic lesions, which also have a long T2 [26].
Anzeige
Peripheral versus ventricular CSF T2 and field strength dependence
De Vis et al. [10] found a T2 difference of 609 ± 133 ms between the periphery and the ventricles at 3 T, and Qin found a T2 difference of 420 ± 155 ms at 3 T. Also in this work a shorter T2,CSF was measured in the periphery compared with the ventricles, as shown in Fig. 4a. This T2 difference is larger at 3 T than at 7 T: the T2 difference is 365 ms at 3 T and 161 ms at 7 T, which corresponds to differences of 18% and 15% relative to the T2 in the lateral ventricles for 3 and 7 T, respectively. Partial volume correction, which led to a peripheral CSF T2 increase of 118 ms at both field strengths (Fig. 5), resulted in remaining T2 differences of 247 ms and only 43 ms for 3 and 7 T, respectively. These correspond to a T2 difference of 12% and 4% relative to the ventricular CSF T2 for 3 and 7 T, respectively. A relatively larger T2 difference was found when a SE-EPI sequence was used, and remained largely unchanged after partial volume correction (data shown in the electronic supplementary material).
These results indicate a true T2 difference between peripheral and ventricular CSF. A potential physiological explanation for this observed T2 difference could be sought in differences in, for example, in the levels of O2, protein, and/or glucose, since these substrates are known to decrease T2 [20, 27‐29]. However, relatively large concentration differences are necessary to bridge the difference between peripheral and ventricular T2,CSF. So although differences in CSF composition may partly cause the observed T2 difference, it seems unlikely that these are the only contributor.
The shorter in vivo CSF T2 at 7 T than at 3 T (Fig. 4a) is in line with published in vivo measurements by Daoust et al. [20]. However, Daoust et al. suggested that the T2 of CSF is not field strength dependent, but that residual field gradients cause errors in in vivo measurements at higher field strengths. If the T2 measurements are strongly dependent on residual gradients, one might expect that the T2 difference between periphery and ventricles observed at 3 T is also largely due to residual field gradients, such as B0 gradients. However, the CSF T2 mapping sequence used in our study showed negligible B0 gradient dependency for the measured T2 up to 0.3 mT/m, while the observed B0 gradients in the brain were between 0.07 and 0.38 mT/m, and on average well below 0.20 mT/m. The limited diffusion sensitivity of the CSF T2 mapping sequence is also visible from the results of the long TE analysis on the phantom measurements. The measured T2 remained unchanged when only long TEs (with stronger diffusion weighting) were used (see Fig. 6).
Implications
Before CSF T2 mapping can be used as a parameter to study diseases such as cerebral small vessel disease, several uncertainties need to be resolved. It is not yet clear to what extent the T2 difference between ventricular and peripheral CSF reflects physiological differences in CSF composition. The CSF T2 mapping sequence shows a much smaller T2 difference compared with SE-EPI, while the difference also varies with field strength. Overall, the T2 difference between peripheral and ventricular CSF could (partly) be explained by (a combination of) physiological differences. The possibility that the shorter peripheral T2 is entirely caused by an artifact, like B0 gradients caused by imperfect shimming and/or partial volume effects between tissue, blood, and CSF, seems unlikely.
Anzeige
Care should be taken when one is interpreting T2 measurements of CSF, and more work is necessary to find the true explanations for the T2 differences between 3 and 7 T and between the peripheral and ventricular CSF at 3 T.
Limitations
The major limitation of this work is that it is an observational study, which limits the extent to which underlying mechanisms causing the observations can be identified. Despite our efforts to separate the effects of partial volume and true physiological differences, it remains uncertain to what extent the observed shorter peripheral T2,CSF is due to different CSF compositions.
Furthermore, the statistical power of this study was limited by the low number of volunteers combined with the stringent R2 criterion, which resulted in a relatively large dropout of ROIs.
Moreover, only macroscopic B0 gradients could be determined in the in vivo scans, and the magnitude of microscopic, subvoxel B0 gradients remains unknown.
Finally, no in vitro CSF sample was used to validate the in vivo measurements. In vitro CSF is prone to changes in, for example, O2 content, compared with in vivo CSF, which may induce T2 differences between in vitro and in vivo CSF.
Conclusion
CSF T2 mapping with a dedicated sequence is feasible at both 3 and 7 T, and yields shorter CSF T2 times at 7 T compared with 3 T. At 3 T, shorter T2 times were found for peripheral CSF compared with ventricular CSF; at 7 T this effect was much smaller. Partial volume effects can partly explain this T2 difference, but a physiological contribution to the difference in T2 between ventricular and peripheral CSF is possible. The different results for different sequences and field strengths, and the confounding influence of partial volume, will make it challenging to accurately isolate and quantify any true physiological effect for applications in research focusing on in vivo evaluation of the (regional) composition of CSF.
Funding
The research leading to these results received funding from the European Research Council (ERC) under the European Union’s Seventh Framework Programme (2007-2013)/ERC grant agreement no. 337333 (SmallVesselMRI), and the European Union’s Horizon 2020 program/ERC grant agreement no. 637024 (HEARTOFSTROKE) and under grant agreement no. 666881 (SVDs@target).
Authors’ contribution
JMS: Protocol/project development, Data collection, Data analysis. ETP: Protocol/project development, Data analysis. JH: Protocol/project development, Data analysis. PL: Protocol/project development. JJMZ: Protocol/project development, Data collection, Data analysis.
Compliance with ethical standards
Conflict of interest
The authors declare that they have no competing interests.
Ethical approval
All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.
Informed consent
Informed consent was obtained from all individual participants included in the study.
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.
Mit e.Med Radiologie erhalten Sie Zugang zu CME-Fortbildungen des Fachgebietes Radiologie, den Premium-Inhalten der radiologischen Fachzeitschriften, inklusive einer gedruckten Radiologie-Zeitschrift Ihrer Wahl.
