Optimal sequences and sequence parameters for GBCA-enhanced MRI of the glymphatic system: a systematic literature review

Abstract

Background

The glymphatic system (GS) is a recently discovered waste clearance system in the brain.

Purpose

To evaluate the most promising magnetic resonance imaging (MRI) sequence(s) and the most optimal sequence parameters for glymphatic MRI (gMRI) 4–24 h after administration of gadolinium-based contrast agent (GBCA).

Material and Methods

Multiple literature databases were systematically searched for articles regarding gMRI or MRI of the perilymph in the inner ear until 11 May 2020. All relevant MRI sequence parameters were tabulated for qualitative analysis. Their potential was assessed based on detection of low dose GBCA, primarily measured as signal intensity (SI) ratio.

Results

Thirty articles were included in the analysis. Three-dimensional fluid attenuated inversion recovery (3D-FLAIR), 3D Real Inversion Recovery (3D-Real IR), and multiple 3D T1-weighted gradient echo sequences were used. In perilymph, 3D-FLAIR with a TE of at least 400 ms yielded the highest SIRs. In the qualitative analysis of inner ear studies using 3D-FLAIR, TR was in the range of 4400–10,000 ms, TI 1500–2600 ms, refocusing flip angle (rFA) (range 120°–180°), and echo train length (ETL) 23–173. In the gMRI studies, quantitative analysis was not possible. In the qualitative analysis, 3D-FLAIR was used in the majority (8/12) of the studies, usually with TR 4800–9000 ms, TI 1650–2500 ms, TE 311–561 ms, rFA 90°–120°, and ETL 167–278.

Conclusion

Long TE 3D-FLAIR is the most promising sequence for detection of low-dose GBCA in the GS. Clinical and/or phantom studies on other MRI parameters are needed for further optimization of gMRI.

Keywords

Glymphatic system magnetic resonance imaging gadolinium cerebrospinal fluid systematic review

Introduction

The glymphatic system (GS) of the brain is a recently discovered waste clearance system of the brain (1). The cerebrospinal fluid (CSF) in the subarachnoid space can enter the brain parenchyma via perivascular spaces around perforating arteries. The outflow pathway is still under debate. Whereas some groups describe the outflow pathway to follow similar perivascular spaces around veins (1,2), other evidence points towards outflow through intramural periarterial drainage pathways (3 –5).

The GS is more active during sleep and it is suggested that the restorative function of sleep may be due to enhanced waste clearance (6), while sleep deprivation results in impairment of the GS. Impairment of the GS is associated with ageing (7), but also plays a key role in Alzheimer’s disease and other dementias, idiopathic normal pressure hydrocephalus, stroke, and brain trauma (1,2,5,8,9).

Currently, only macrocyclic gadolinium-based contrast agents (GBCA) are approved by the European Medicines Agency and these are usually administered intravenously in a dose of 0.1 mmol/kg (10). After intravenous (IV) injection of GBCAs, plasma concentrations in the first 90 min follow three phases: distribution; rapid elimination; and long-lasting residual elimination from a third compartment (11). In this third phase, small amounts of GBCA cross the blood–brain barrier into the CSF and GS. This is a slow process, peaking at 4–24 h after the injection. Therefore, GBCA may function as a tracer of GS function in vivo (12). Long-term retention of GBCA in the brain has been shown for less stable, linear GBCA but not for macrocyclic GBCA (13,14). An impairment of the GS could be a contributing factor to the formation of GBCA depositions with lower thermodynamic and kinetic stabilities. An off-label use of GBCA is through direct intrathecal (IT) administration via cervical or lumbar puncture. As GBCA is injected directly into the CSF, this may yield higher GBCA concentration in CSF.

Contrast-enhanced magnetic resonance imaging (MRI) could therefore be a valuable tool for in vivo research of the anatomy, function, and role of GS in several pathologies. The protocol for glymphatic MRI (gMRI) should link high-resolution morphologic (T2-weighted [T2W]) imaging with functional information (e.g. diffusion-weighted imaging, susceptibility-weighted imaging) and sequences that can accurately detect low doses of GBCA in perivascular spaces. Inner ear MRI is a model to test MRI sequences for detection of low doses of GBCA. The perilymph (PL) in the scala tympani of the cochlea is similar to CSF in composition and the GBCA concentration in PL after IV administration is similarly low (15,16). Therefore, parameters of sequences that are sensitive to low doses of GBCA in PL may be suitable for detection of GBCA in the CSF elsewhere in the brain. The research on imaging of the PL (e.g. endolymphatic hydrops in Ménière’s disease) has advanced more than on gMRI and may therefore yield important insights into possibilities for advancing gMRI.

The aim of the present systematic literature review was to identify the most used and reported sequences and parameter settings for detection of low doses of GBCA to be used in future gMRI protocols.

Material and Methods

Search strategy

This systematic review was performed in accordance with the PRISMA guidelines (17). The PubMed, Embase, and Web of Science databases were searched until 11 May 2020 for literature on gMRI with IV or IT administration of GBCA and MRI of PL with IV administration of GBCA as a model for gMRI. Phantom studies on these topics were also included. All study types, except case reports, were included. The specific search strings are presented in Appendix A.

Study selection and data extraction

Articles had to be available through the medical libraries of the Erasmus University Medical Center Rotterdam or the Leiden University Medical Center. Language was limited to English. Studies were considered eligible if they reported primary or secondary outcomes regarding image quality or signal intensity (SI) in GBCA-enhanced MRI of intracranial systems (or a simulation of such a system). These results also had to be obtained without the use of image processing. GBCA had to be administered intravenously or intrathecally. The used sequence(s) and parameter settings had to be specified. Articles were initially assessed independently by two of the researchers on title and abstract; in a second round, final inclusion was done based on full-text analysis. Disagreements were resolved by a third researcher.

From the selected studies, data on title, name of the first author, study design, relevant outcome measures, and patient characteristics were extracted. Primary outcome measures were SI ratios between PL or CSF and (a part of) the brain stem or thalamus, signal-to-noise ratios (SNR), and contrast-to-noise ratios (CNR). Secondary outcomes were measures describing image quality or SI.

Quality assessment

In order to assess the quality of selected articles a quality assessment tool was made, based on questions from QUADAS-2 (Appendix B) (18). Articles that consisted of a combination of human, phantom, and/or animal test subjects were assessed for each of those parts individually. Articles were excluded if they scored poorly (human studies score < 7, phantom studies score < 5, animal studies score < 4).

Data reviewing

Articles were grouped based on subject type and outcomes were reported. Scanner manufacturer, field strength, and key parameters such as repetition time (TR), echo time (TE), inversion time (TI), echo train length (ETL), and refocusing pulse flip angle (rFA) were listed for all glymphatic, inner ear, and phantom MRI studies. Through this, qualitative analysis was possible to provide a practical range for clinical practice or for phantom optimization.

Results

Study selection and quality assessment

In the initial search and citation tracking, a total of 624 articles were identified, after removing duplicates. After assessing the abstract and title, full-text evaluation, and quality assessment (results in appendix C), 30 eligible articles were included in the review (Fig. 1) (19 –48).

Fig. 1.

The PRISMA flowchart for study selection.

Study characteristics

Study characteristics are shown in Table 1. Twelve studies focused on gMRI, of which five used IV administration and five used IT administration of GBCA. The remaining two studies were animal studies in which IV administration was used (19 –30). There were 15 studies on PL of the inner ear and three phantom studies (31 –48).

Table 1.

Study characteristics, alphabetically ordered per subject type and administration route.

	First author	Year	Study type	FS	Brand	Sequence(s)	Voxel size (mm3)	No. subjects	Imaged subjects	Administration and dose	GBCA type
1	Deike-Hofmann (19)	2019	D	1.5	Siemens	3D-FLAIR	0.38	40	B	IV 0.1	MC
2	Naganawa (20)	2018	CS	3	Siemens	3D-FLAIR	0.25	21	B	IV 0.1	MC
3	Naganawa (21)	2016	D	3	Siemens	3D-FLAIR	0.25	27	B	IV 0.1	MC/L
4	Naganawa (22)	2011	D	3	Siemens	3D-FLAIR	0.39	10	B, IE	IV 0.1	L
5	Ohashi (23)	2019	CS	3	Siemens	3D-FLAIR	0.25	25	B	IV 0.1	MC
6	Eide (24)	2018	PC	3	Philips	3D-T1-GRE	1.00	38	B	IT 0.5 mmol	MC
7	Jacobsen (25)	2019	PC	3	Philips	3D-T1-GRE	1.00	10	B	IT 0.5 mmol	MC
8	Ringstad (26)	2020	CS	3	Philips	3D-FLAIR; 3D-T1-BB	1.00 1.00	18	B	IT 0.5 mmol	MC
9	Ringstad (27)	2017	PC	3	Philips	3D-T1-GRE	1.00	23	B	IT 0.5 mmol	MC
10	Zhou (28)	2020	PC	3	GE	2D-FLAIR; 3D-T1-GRE	0.95 1.00	35	B	IT 0.5 mmol	L
11	Jost (29)	2017	A	1.5	Siemens	3D-FLAIR	0.05	84	B,A	IV 1.8	MC/L
12	Taoka (30)	2018	A	4.7	BioSpec	T1-TSE	0.05	6	B,A	IV 1.0	L
13	Eliezer (31)	2018	D	3	GE	3D-FLAIR	0.51	20	IE	IV 0.1/0.2	MC
14	Lee (32)	2019	D	3	Philips	3D-FLAIR	0.39	123	IE	IV 0.1	MC
15	Li (33)	2019	D	3	Siemens	3D-real IR	0.02	79	IE	IV 0.2	L
16	Naganawa (34)	2016	D	3	Siemens	3D-FLAIR	0.22	17	IE	IV 0.1	MC
17	Naganawa (35)	2010	CS	3	Siemens	3D-FLAIR	0.39	4	IE	IV 0.2	MC
18	Nahmani (36)	2020	D	3	Siemens	3D-FLAIR	0.24	16	IE	IV 0.1	MC
19	Pakdaman (37)	2016	CS	3	Siemens	3D-FLAIR	0.32	32	IE	IV 0.2	L
20	Sano (38)	2011	PC	3	Siemens	3D-FLAIR	0.35	10	IE	IV 0.1	L
21	Shi (39)	2019	CS	3	Siemens	3D-real IR	0.02	23	IE	IV 0.2	MC
22	Shi (40)	2018	CS	3	Siemens	3D-real IR	0.02	154	IE	IV 0.2	MC
23	Suzuki (41)	2011	D	3	Siemens	3D-FLAIR	0.19	32	IE	IV 0.1/0.2	MC/L
24	Tagaya (42)	2011	CS	3	Siemens	3D-FLAIR	0.35	12	IE	IV 0.2	MC
25	Tagaya (43)	2010	D	3	Siemens	3D-FLAIR	0.35	10	IE	IV 0.2	MC
26	Yamazaki (44)	2012	D	3	Siemens	3D-FLAIR	0.39	17	IE	IV 0.2	MC
27	Yamazaki (45)	2011	D	3	Siemens	3D-FLAIR	0.39	22	IE	IV 0.2	MC
28	Freeze (46)	2019	P	3	Philips	3D-FLAIR	1.00	–	P	0.004, 0.008, 0.016, 0.031, 0.063, 0.125, 0.250, 0.500, 1.0 mM	MC
29	Kato (47)	2019	P	3	Siemens	3D-FLAIR	0.24	–	P	0.008, 0.016, 0.031, 0.063, 0.125, 0.250 mM	L
30	Naganawa (48)	2010	P	3	Siemens	3D-FLAIR	0.25	–	P	0.015, 0.021, 0.031, 0.063, 0.125 mM	L

3D-FLAIR, 3D fluid-attenuated inversion recovery; 3D-real IR, 3D real inversion recovery; 3D-T1-BB, 3D T1 black blood; 3D-T1-GRE, 3D T1 gradient echo; A, animal; B, brain; CS, case series; D, diagnostic; FS, field strength; GBCA, gadolinium-based contrast agent; IE, inner ear; IT, intrathecal (in mmol/kg unless stated otherwise); IV, intravenous; L, linear; MC, macrocyclic; P, phantom; PC, prospective cohort; T1-TSE, T1 turbo spin echo.

gMRI studies

In all articles, the used sequences and parameter settings resulted in enhancement of CSF. In total, eight studies included a T2W fluid-attenuated inversion recovery (FLAIR) sequence, of which six studies employed a TE > 400 ms, which is commonly termed heavily T2W (hT2W) (19 –23,26,28,29). Because hT2W-FLAIR and regular FLAIR are essentially the same sequence, the term “FLAIR” will be used for both sequences throughout this review. The FLAIR studies reported enhancement of the CSF in several spaces surrounding the cortex, perineural spaces, and spaces in and surrounding the eye segments. One study specifically reported that hT2W-3D-FLAIR, with a TE of 561 ms and ETL of 278, enabled visualization of the GBCA pathway into and through the GS (19).

Six studies included a T1-weighted (T1W) sequence (24 –28,30). Three of these studies adopted a 3D-T1W gradient-echo (GRE) sequence with equal parameter settings and were also able to visualize several spaces surrounding the cortex (24,25,27). Two of these studies specifically reported the ability to visualize clearance of GBCA from the CSF. One study used a 3D T1-GRE sequence in which the ventricular system, the frontal horn, and the inferior frontal gyrus were visualized (28). In the present review, visualization of clearance through the GS was reported. One study used a T1W turbo spin echo sequence to image the fourth ventricle and the subarachnoid space of the prepontine cistern (30).

gMRI sequence parameters

An overview of the sequence parameters is depicted in Table 2. In the gMRI studies, ranges of parameter settings in 3D-FLAIR sequences were: TR = 4800–9000 ms; TE = 311–561 ms; TI = 1650–2500 ms; ETL = 167–278; rFA = 90°–120°. In one study that used 2D-FLAIR, a TE of 152 ms was used. Parameter settings of 3D-T1-GRE sequences after IT administration were TR 5.1–7.3 ms, TE 2.3–3.0 ms, and excitation pulse flip angle (eFA) = 8°.

Table 2.

Overview of MRI protocols in gMRI studies.

First author	No. subjects	IV(A)/ IT	GBCA dose	Sequence	FS (T)	TR (ms)	TE (ms)	TI (ms)	ETL	rFA (°)	eFA
Deike-Hofmann (19)	40	IV	0.1 mmol/kg	3D-FLAIR	1.5	9000	561	2500	278	120	NA
Naganawa (21)	27	IV	0.1 mmol/kg	3D-FLAIR	3	9000	544	2250	173	120*	NA
Naganawa (20)	21	IV	0.1 mmol/kg	3D-FLAIR	3	9000	544	2250	173	120*	NA
Naganawa (22)	10	IV	0.1 mmol/kg	3D-FLAIR	3	9000	540	2350	107	120*	NA
Ohashi (23)	25	IV	0.1 mmol/kg	3D-FLAIR	3	9000	544	2250	173	120*	NA
Eide (24)	38	IT	0.5 mmol	3D-T1-GRE	3	∼5.1^†	∼2.3^†	NA	NA	NA	8
Jacobsen (25)	10	IT	0.5 mmol	3D-T1-GRE	3	∼5.1^†	∼2.3^†	NA	NA	NA	8
Ringstad (26)	18 18	IT IT	0.5 mmol 0.5 mmol	3D-FLAIR 3D-T1-BB	3 3	4800 700	311 35	1650 NA	167 55	90 80	NA NA
Ringstad (27)	23	IT	0.5 mmol	3D-T1-GRE	3	∼5.1^†	∼2.3^†	NA	NA	NA	8
Zhou (28)	35 35	IT IT	0.5 mmol 0.5 mmol	2D-FLAIR 3D-T1-GRE	3 3	8400 7.3	152 3.0	– NA	– NA	90 NA	NA 8
Jost (29)	84	IV, A	1.8 mmol/kg	3D-FLAIR	3	9000	554	2250	79	120*	NA
Taoka (30)	6	IV, A	1.0 mmol/kg	T1-TSE	4.7	1015	18.2	NA	6	90	NA

*Rapidly decreased from 180° to 120°.

^†Shortest time was chosen.

3D-T1-BB, 3D T1 black blood; 3D-T1-GRE, 3D T1 gradient echo; A, animal; eFA, excitation pulse flip angle; ETL, echo train length; FLAIR, fluid-attenuated inversion recovery; FS, field strength; GBCA, gadolinium-based contrast agent; gMRI, glymphatic magnetic resonance imaging; IT, intrathecal; IV, intravenous; MRI, magnetic resonance imaging; NA, not applicable; rFA, refocusing pulse flip angle; T1-TSE, T1 turbo spin echo; TE, echo time; TI, inversion time; TR, repetition time.

Inner ear studies

Table 3 shows the results of all articles that reported the SI ratio between regions of interest (ROIs) on the scala tympani of the basal turn of the cochlea and the cerebellum white matter (CC-SIR). Apparent from this overview is that 3D-real IR resulted in lower CC-SIRs than the 3D-FLAIR sequence with long TE (34,38 –41). Additionally, the studies that used the 3D-real IR sequence for the detection of GBCA, used double contrast agent doses (0.2 mmol/kg) for IV administration in their subjects. One study described the effect of administering a double dose of GBCA compared to a single dose (0.1 mmol/kg), while using the 3D-FLAIR sequence (41). The CC-SIR increased from mean 12.5 ± 5.0 in the single dose group to 18.3 ± 5.2 in the double dose group. This illustrates an even larger efficiency of this sequence compared to the 3D-real IR sequence. Several of the seven articles additionally reported preference for the use of a long TE (i.e. longer than 400 ms) over conventional 3D-FLAIR (38,41). All articles with a long TE of ∼540 ms reported considerably higher CC-SIRs than those with shorter TE of 128 ms or 181 ms (Table 3).

Table 3.

SI ratios between the basal turn of cochlea and cerebellum (CC-SIR) per inner ear study and corresponding MRI protocols.

First author	Pt. (ears)	GBCA dose (mmol/kg)	Sequence	TR (ms)	TE (ms)	TI (ms)	ETL	rFA (degrees of arc)	CC-SIR (diseased group) (mean ± SD)	CC-SIR (healthy group) (mean ± SD)
Shi (39)	23 (46)	0.2	3D-real IR	6000	181	1850	NA	NA	1.44 ± 0.46	1.15 ± 0.33
Shi (40)	154 (308)	0.2	3D-real IR	6000	181	1850	NA	NA	1.39 ± 0.37	1.18 ± 0.29
Naganawa (34)	17 (26)	0.1	3D-FLAIR	9000	544	2250	173	120*	11.6 ± 3.4	7.9 ± 1.5
Sano (38)	10 (20)	0.1	3D-FLAIR	9000	540	2250	173	120*	12.2 ± 2.4	9.2 ± 1.8
Suzuki (41)	32 (64)	0.1 0.2	3D-FLAIR	9000	540	2250	173	120*	12.5 ± 5.0^† 18.3 ± 5.2^†
Tagaya (42)	12 (24)	0.2	3D-FLAIR	9000	128	2500	23	180	1.12 ± 0.36	0.82 ± 0.15
Tagaya (43)	10 (20)	0.1 0.2	3D-FLAIR	9000	128	2500	23	180	0.95 ± 0.64 1.17 ± 0.40	0.72 ± 0.38 0.84 ± 0.05

*Rapidly decreased from 180° to 120°.

^†Diseased and healthy groups not reported separately.

A, animal; CC-SIR, cochlea–cerebellum signal intensity ratio; eFA, excitation pulse flip angle; ETL, echo train length; FLAIR, fluid-attenuated inversion recovery; GBCA, gadolinium-based contrast agent; MRI, magnetic resonance imaging; NA, not applicable; rFA, refocusing pulse flip angle; T1-TSE, T1 turbo spin echo; TE, echo time; TI, inversion time; TR, repetition time.

Moreover, in four articles that compared CC-SIRs in inner ears affected by endolymphatic hydrops (EH) to healthy volunteers or healthy contralateral ears, three articles reported significantly higher CC-SIR in affected ears (38 –40).

Table 4 shows the results of all articles that reported the SI ratio between ROIs in the basal turn of the cochlea and the medulla oblongata (CM-SIR). One study found no significant difference in CM-SIR between single and double GBCA doses in affected ears, but the single dose resulted in higher CM-SIR than double dose when using 3D-FLAIR with a low TE (P = 0.009) (31). One study specifically studied rFA and reported significant benefit in using a constant rFA of 140° compared to a variable rFA in both affected and unaffected ears (P < 0.001) (36). Additionally, three articles that used variable rFA reported lower CM-SIRs than did the article with constant rFA of 120° (31,37,44,45). Moreover, articles with high CM-SIR generally used longer TE.

Table 4.

SI ratios between basal turn of cochlea and medulla oblongata (CM-SIR) per inner ear study and corresponding MRI protocols.

First author	Pt. (ears)	GBCA dose	Sequence	TR (ms)	TE (ms)	TI (ms)	ETL	rFA (degrees of arc)	CM-SIR (diseased group) (mean ± SD)	CM-SIR (healthy group) (mean ± SD)
Eliezer (31)	20 (40)	0.1 0.2	3D-FLAIR	8000	130	2059	–	Var	1.58 (1.46–1.97)* 1.3 (1.15–1.84)*	1.62 (1.49–1.70)* 1.21 (1.17–1.42)*
Nahmani (36)	16 (32)	0.1	3D-FLAIR	10,000	640	2600	–	140 Var	7.16 (5.65–8.46)* 1.54 (1.44–1.84)*	7.00 (5.85–7.60)* 1.45 (1.30–1.68)*
Pakdaman (37)	32 (64)	0.2	3D-FLAIR	9000	534	2350	144	120	12.6 ± 7.4	8.0 ± 3.1
Yamazaki (44)	17 (28)	0.2	3D-FLAIR	9000	458	2500	119	Var^†	1.46 ± 0.45	–
Yamazaki (45)	22 (44)	0.2	3D-FLAIR	9000	458	2500	119	Var^†	1.86 ± 0.74	1.29 ± 0.31

*Median (interquartile range).

^†Average of 120°.

CM-SIR, cochlea–medulla signal intensity ratio; ETL, echo train length; FLAIR, fluid-attenuated inversion recovery; GBCA, gadolinium-based contrast agent; MRI, magnetic resonance imaging; rFA, refocusing pulse flip angle; TE, echo time; TI, inversion time; TR, repetition time.

Out of five articles that evaluated CM-SIR, two reported no significant difference between affected and unaffected ears, whereas two articles reported significantly higher CM-SIR in the affected ears (31,36,37,45).

Four articles were not suitable for quantitative analysis and will therefore be evaluated separately. Three articles evaluated 3D-FLAIR, of which two studies reported no relevant quantitative outcomes, but were able to visualize PL (22,32,35). The third study reported a significant increase in SIR of the PL, with respect to the pontine parenchyma (22). Lastly, one study described SNRs of PL (42.81 ± 22.66) and brainstem (12.55 ± 5.58) when scanning with 3D-real IR. The resulting CNR between PL and brainstem was 38.87 ± 26.09 (33).

Inner ear MRI sequence parameters

An overview of the sequence parameters is depicted in Tables 3 and 4. In the inner ear studies, parameter settings in 3D-real IR sequences were as follows: TR = 6000 ms; TE = 181 ms; and TI = 1850 ms. ETL was not listed for all three 3D-real IR studies, while rFA was listed in one study as 180° (33). For 3D-FLAIR parameters ranges were as follows: TR = 4400–10,000 ms; TE = 128–640 ms; TI = 1500–2600 ms; ETL = 23–173; and rFA = 90°–180°.

Phantom studies

Of the three included phantom studies for optimization, two imaged GBCA dilutions using 3D-FLAIR with a long TE (46,47). The final phantom study compared TEs for 3D-FLAIR sequences when imaging the inner ear (48). Within these phantom studies, dilutions of GBCA were either made using saline or demineralized water as dissolvant. The GBCA concentrations were made to be comparable to conditions in vivo. Temperature adjustments were made in two phantom studies, either heating the scanning samples to the level of body temperature, or modelling outcomes with temperature adjusted T1 and T2 values (47,48).

TR was in the range of 9000–16,000 ms. At lower concentrations of GBCA (0.008 mM), choosing a longer TR (>12,000 ms) becomes progressively more important, because normalized SI (compared to SI at TR = 9000 ms) increases when GBCA concentrations decrease (47). Longer TR has also shown to be effective against SI loss due to saturation effects of the high T1 value of CSF. However, lengthening the TR increases the SI of GBCA in CSF, but also the surrounding brain, with a decrease in the signal difference between CSF and the brain (46). Scan time will increase proportionally. Similar positive effects of long TR in 3D-FLAIR with a long TE have been shown in an inner ear phantom (48).

Reported TE for hT2W-3D-FLAIR was in the range of 500–544 ms, whereas the conventional 3D-FLAIR has a TE of around 140 ms (46 –48). Increasing TE resulted in the suppression of brain tissue, facilitating detection of subtle changes in GBCA concentration in CSF (46).

Discussion

In this first systematic literature review on glymphatic MRI, hT2W-3D-FLAIR is the most widely used and promising sequence for the visualization of GBCA in CSF after IV administration of GBCA. These results are supported by the inner ear model for the detection of GBCA in the PL in the cochlea.

The range of outcome measures used in the included studies was wide, which complicated the comparison between studies. Though several studies reported CC- or CM-SIR, the studies were too heterogeneous with regards to, for example, imaging equipment, parameter settings, and type of GBCA to perform a meta-analysis. The studies on the GS were too heterogeneous in outcome measures to be able to quantitatively compare the results. Future research is therefore required to systematically quantify the sensitivity of MRI techniques for low concentrations of GBCA in CSF.

Based on the outcomes of the inner ear studies, it can be stated that 3D-FLAIR outperformed 3D-real IR regarding contrast between the GBCA and the brain. In addition, 3D-real IR showed good results on Siemens scanners, while results were unstable in hands of others, who have discontinued its development (49). 3D-T1-GRE techniques are more sensitive to higher GBCA dose levels (50), for instance when GBCA is injected intrathecally (as was done in the included 3D-T1-GRE studies). This sequence could be of use as a secondary, supporting sequence in a gMRI protocol.

Even though one of the inner ear studies indicated that a constant rFA was to be preferred over a variable rFA, most studies employed 3D-FLAIR techniques with variable rFA, which is inherent in SPACE (Siemens Healthineers), CUBE (GE Healthcare), or VISTA (Philips Medical Systems) sequences. Therefore, more evidence has to be gathered on the optimal rFA settings.

The current literature provides insufficient guidance on parameter settings such as TR/TI and ETL. There is a wide range of parameters used in which TI for CSF suppression depends directly on the chosen TR for total brain coverage. In essence, to avoid blurring, a low echo spacing is advisable. Other parameters, such as matrix size, slice thickness, and field of view, should also be optimized, but were not included in the current review.

A sequence which was not found in the current literature, but which could yield promising results, is double inversion recovery (DIR). DIR could be used to minimize SI from CSF as well as SI from brain matter surrounding the GS. This could provide a clearer distinction of the GS.

The present systematic review has some limitations. The first (and largest) limitation is the fact that a significant portion of the available literature involved research on MRI of the PL in the inner ear, which can only be seen as a model for gMRI. Since several of the inner ear studies concluded that enhancement measures of the PL are dependent on, for example, pathology, there is no linear relationship between this model and the GS. Thus, further targeted investigation into gMRI is needed through clinical studies or phantom studies.

Second, the included studies often lacked a detailed discussion of the main outcome measures of this review. The main outcome and conclusion were often with regards to the value of the used technique for diagnosis or evaluation of a specific pathology. Thus, the interpretation of the results relevant to this review was not straightforward.

Third, MRI equipment used in the included studies was heterogeneous. In the majority of studies, especially in inner ear MRI, Siemens MRI scanners have been used for imaging. In addition, field strengths were 1.5–4.7 T using 8–32-channel head coils.

Fourth, only a few studies included the acquisition time per scan as a determining factor for their recommendations. Long acquisition times are difficult for the patient and may result in unwanted artifacts. For example, a longer TR also increases the SI of tissue surrounding the CSF spaces; the clinically optimal TR still remains to be investigated (19,46).

Fifth, many articles in this systematic review were from a limited number of research groups, which could potentially introduce bias.

In conclusion, hT2W-3D-FLAIR should be further explored for imaging the GS. Currently, there are still uncertainties regarding optimal parameter settings, which require more systematic investigation on multiple scanners.

Supplemental Material

sj-pdf-1-acr-10.1177_0284185120969950 - Supplemental material for Optimal sequences and sequence parameters for GBCA-enhanced MRI of the glymphatic system: a systematic literature review

Supplemental material, sj-pdf-1-acr-10.1177_0284185120969950 for Optimal sequences and sequence parameters for GBCA-enhanced MRI of the glymphatic system: a systematic literature review by Liesje SP Mijnders, Feline WR Steup, Mette Lindhout, Paul A van der Kleij, Wyger M Brink and Aart J van der Molen in Acta Radiologica

Footnotes

Declaration of conflicting interests

The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: AvdM has received funding for presentations on contrast media safety by Bayer Healthcare.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Liesje SP Mijnders

Aart J van der Molen

Supplemental material

Supplemental material for this article is available online.

References

Iliff

Wang

Liao

, et al. A paravascular pathway facilitates CSF flow through the brain parenchyma and the clearance of interstitial solutes, including amyloid β. Sci Transl Med 2012; 4:ra11.

Mestre

Mori

Nedergaard

The Brain’s Glymphatic System: Current Controversies. Trends Neurosci 2020;43:458–466.

Albargothy

Johnston

MacGregor-Sharp

, et al. Convective influx/glymphatic system: tracers injected into the CSF enter and leave the brain along separate periarterial basement membrane pathways. Acta Neuropathol 2018; 136:139–152.

Bakker

Bacskai

Arbel-Ornath

, et al. Lymphatic clearance of the brain: perivascular, paravascular and significance for neurodegenerative diseases. Cell Mol Neurobiol 2016; 36:181–194.

Rasschaert

Weller

Schroeder

, et al. Retention of gadolinium in brain parenchyma: pathways for speciation, access, and distribution. A critical review. J Magn Reson Imaging 2020;52:1293–1305.

Xie

Kang

, et al. Sleep drives metabolite clearance from the adult brain. Science 2013; 342:373–377.

Benveniste

Liu

Koundal

, et al. The glymphatic system and waste clearance with brain aging: a review. Gerontology 2019; 65:106–119.

Rasmussen

Mestre

Nedergaard

The glymphatic pathway in neurological disorders. Lancet Neurol 2018; 17:1016–1024.

Taoka

Naganawa

Glymphatic imaging using MRI. J Magn Reson Imaging 2020; 51:11–24.

10.

European Medicines Agency. EMA's final opinion confirms restrictions on use of linear gadolinium agents in body scans, 457616 (2017). https://www.ema.europa.eu/en/documents/press-release/emas-final-opinion-confirms-restrictions-use-linear-gadolinium-agents-body-scans_en.pdf.

11.

Lancelot

Revisiting the pharmacokinetic profiles of gadolinium-based contrast agents: differences in long-term biodistribution and excretion. Invest Radiol 2016; 51:691–700.

12.

Berger

Kubik-Huch

Niemann

, et al. Gadolinium distribution in cerebrospinal fluid after administration of a gadolinium-based MR contrast agent in humans. Radiology 2018; 288:703–709.

13.

Jost

Frenzel

Boyken

, et al. Long-term excretion of gadolinium-based contrast agents: linear versus macrocyclic agents in an experimental rat model. Radiology 2019; 290:340–348.

14.

Robert

Fingerhut

Factor

, et al. One-year retention of gadolinium in the brain: comparison of gadodiamide and gadoterate meglumine in a rodent model. Radiology 2018; 288:424–433.

15.

Bernaerts

De Foer

Imaging of Meniere Disease. Neuroimaging Clin N Am 2019; 29:19–28.

16.

Wangemann

Supporting sensory transduction: cochlear fluid homeostasis and the endocochlear potential. J Physiol 2006; 576:11–21.

17.

Moher

Liberati

Tetzlaff

, et al. Preferred Reporting Items for Systematic Reviews and Meta-Analyses: The PRISMA Statement. PLoS Med 2009; 6:e1000097.

18.

Bristol Uo. QUADAS-2 Bristol 2019. Available at: https://www.bristol.ac.uk/population-health-sciences/projects/quadas/quadas-2/

19.

Deike-Hofmann

Reuter

Haase

, et al. Glymphatic pathway of gadolinium-based contrast agents through the brain: overlooked and misinterpreted. Invest Radiol 2019; 54:229–237.

20.

Naganawa

Nakane

Kawai

, et al. Differences in signal intensity and enhancement on mr images of the perivascular spaces in the basal ganglia versus those in white matter. Magn Reson Med Sci 2018; 17:301–307.

21.

Naganawa

Nakane

Kawai

, et al. Gd-based contrast enhancement of the perivascular spaces in the basal ganglia. Magn Reson Med Sci 2016; 16:61–65.

22.

Naganawa

Yamazaki

Kawai

, et al. Contrast enhancement of the anterior eye segment and subarachnoid space: detection in the normal state by heavily T2-weighted 3D FLAIR. Magn Reson Med Sci 2011; 10:193–199.

23.

Ohashi

Naganawa

Ogawa

, et al. Signal intensity of the cerebrospinal fluid after intravenous administration of gadolinium-based contrast agents: strong contrast enhancement around the vein of Labbe. Magn Reson Med Sci 2019; 18:194–199.

24.

Eide

Ringstad

Delayed clearance of cerebrospinal fluid tracer from entorhinal cortex in idiopathic normal pressure hydrocephalus: A glymphatic magnetic resonance imaging study. J Cereb Blood Flow Metab 2018; 39:1355–1368.

25.

Jacobsen

Ringstad

Jorstad

, et al. The human visual pathway communicates directly with the subarachnoid space. Invest Ophthalmol Vis Sci 2019; 60:2773–2780.

26.

Ringstad

Eide

PK.

Cerebrospinal fluid tracer efflux to parasagittal dura in humans. Nat Commun 2020; 11:354.

27.

Ringstad

Vatnehol

SAS

Eide

PK.

Glymphatic MRI in idiopathic normal pressure hydrocephalus. Brain 2017; 140:2691–2705.

28.

Zhou

Cai

Zhang

, et al. Impairment of the glymphatic pathway and putative meningeal lymphatic vessels in the aging human. Ann Neurol 2020; 87:357–369.

29.

Jost

Frenzel

Lohrke

, et al. Penetration and distribution of gadolinium-based contrast agents into the cerebrospinal fluid in healthy rats: a potential pathway of entry into the brain tissue. Eur Radiol 2017; 27:2877–2885.

30.

Taoka

Jost

Frenzel

, et al. Impact of the glymphatic system on the kinetic and distribution of gadodiamide in the rat brain: observations by dynamic MRI and effect of circadian rhythm on tissue gadolinium concentrations. Invest Radiol 2018; 53:529–534.

31.

Eliezer

Poillon

Gillibert

, et al. Comparison of enhancement of the vestibular perilymph between gadoterate meglumine and gadobutrol at 3-Tesla in Meniere’s disease. Diagn Interv Imaging 2018; 99:271–277.

32.

Lee

Kim

Lee

, et al. Quantitative analysis of cochlear signal intensity on three-dimensional and contrast-enhanced fluid-attenuated inversion recovery images in patients with Meniere’s disease: Correlation with the pure tone audiometry test. J Neuroradiol 2019; 46:307–311.

33.

Sha

Wang

, et al. Comprehensive comparison of MR image quality between intratympanic and intravenous gadolinium injection using 3D real IR sequences. Acta Otolaryngol 2019; 139:659–664.

34.

Naganawa

Kawai

Taoka

, et al. Cochlear lymph fluid signal increase in patients with otosclerosis after intravenous administration of gadodiamide. Magn Reson Med Sci 2016; 15:308–315.

35.

Naganawa

Ishihara

Iwano

, et al. Estimation of gadolinium-induced T1-shortening with measurement of simple signal intensity ratio between the cochlea and brain parenchyma on 3D-FLAIR: correlation with T1 measurement by TI scout sequence. Magn Reson Med Sci 2010; 9:17–22.

36.

Nahmani

Vaussy

Hautefort

, et al. Comparison of enhancement of the vestibular perilymph between variable and constant flip angle-delayed 3D-FLAIR sequences in Meniere disease. AJNR Am J Neuroradiol 2020; 41:706–711.

37.

Pakdaman

Ishiyama

, et al. Blood-labyrinth barrier permeability in meniere disease and idiopathic sudden sensorineural hearing loss: findings on delayed postcontrast 3D-FLAIR MRI. AJNR Am J Neuroradiol 2016; 37:1903–1908.

38.

Sano

Teranishi

Yamazaki

, et al. Contrast enhancement of the inner ear in magnetic resonance images taken at 10 minutes or 4 hours after intravenous gadolinium injection. Acta Otolaryngol 2011; 132:241–246.

39.

Shi

Zhou

Wang

3D-real IR MRI of Meniere’s disease with partial endolymphatic hydrops. Am J Otolaryngol 2019; 40:589–593.

40.

Shi

Guo

Wang

Magnetic resonance imaging of Meniere’s disease after intravenous administration of gadolinium. Ann Otol Rhinol Laryngol 2018; 127:777–782.

41.

Suzuki

Teranishi

Sone

, et al. Contrast enhancement of the inner ear after intravenous administration of a standard or double dose of gadolinium contrast agents. Acta Otolaryngol 2011; 131:1025–1031.

42.

Tagaya

Yamazaki

Teranishi

, et al. Endolymphatic hydrops and blood-labyrinth barrier in Meniere’s disease. Acta Otolaryngol 2011; 131:474–479.

43.

Tagaya

Teranishi

Naganawa

, et al. 3 Tesla magnetic resonance imaging obtained 4 hours after intravenous gadolinium injection in patients with sudden deafness. Acta Otolaryngol 2010; 130:665–669.

44.

Yamazaki

Naganawa

Kawai

, et al. Gadolinium distribution in cochlear perilymph: differences between intratympanic and intravenous gadolinium injection. Neuroradiology 2012; 54:1161–1169.

45.

Yamazaki

Naganawa

Tagaya

, et al. Comparison of contrast effect on the cochlear perilymph after intratympanic and intravenous gadolinium injection. AJNR Am J Neuroradiol 2011; 33:773–778.

46.

Freeze

Ter Weele

Palm

, et al. Optimal detection of subtle gadolinium leakage in CSF with heavily T2-weighted fluid-attenuated inversion recovery imaging. AJNR Am J Neuroradiol 2019; 40:1481–1483.

47.

Kato

Bokura

Taoka

, et al. Increased signal intensity of low-concentration gadolinium contrast agent by longer repetition time in heavily T2-weighted-3D-FLAIR. Jpn J Radiol 2019; 37:431–435.

48.

Naganawa

Kawai

Sone

, et al. Increased sensitivity to low concentration gadolinium contrast by optimized heavily T2-weighted 3D-FLAIR to visualize endolymphatic space. Magn Reson Med Sci 2010; 9:73–80.

49.

Attye A. Personal communication. Discontinuing acquisition 3D-real IR. to: Brink W, 2020.

50.

Naganawa

Gd deposition: How does Gd enter the brain, when BBB is intact? ISMRM presentation. Nagoya: Nagoya University Graduate School of Medicine, Department of Radiology, 2017.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.27 MB