White matter hyperintensities (WMHs), was called Leukoaraiosi (LA) initially (Alexander, 1987). It presents as a symmetrical, diffuse speckled or patchy irregular hypodense area in the paraventricular and centrum semiovale white matter on CT images and hyperintensity on T2-weighted (T2WI) or/and fluid-attenuated inversion recovery (FLAIR) MRI. WMHs was classified into two subtypes based on possible different pathogenic mechanisms: periventricular white matter hyperintensities (PVWMHs) and subcortical white matter hyperintensities (SCWMHs) (Kim et al., 2008). WMHs was one of the earliest and most established markers of cerebral small vessel disease (SVD), with a pathophysiological substrate in demyelination and small artery atherosclerosis. And a recent study shows that increased WHM volume is associated with amyloid accumulation (Shirzadi et al., 2023). WMHs frequently raise the risk of stroke, dementia, and mortality in the elderly although they are considered healthy people clinically (Debette and Markus, 2010; Debette et al., 2019), jeopardizing their health and quality of daily life to a certain extent. Investigation of the WMHs is necessary to clarify the progression of SVD and may provide new insights into the mechanism of demyelination produced by ischemic.

Proton magnetic resonance spectroscopy (¹H-MRS) is a non-invasive imaging technique that quantitatively detects metabolite levels in vivo. Conventional MRS at 3T can quantify various brain metabolites including N-acetylaspartate (NAA), choline (Cho), creatine (Cr), and glutamate + glutamine (Glx) etc. Previous studies have detected decreased metabolic ratios of NAA/Cho in patients with SVD (Hund-Georgiadis et al., 2001; Wang et al., 2012), but the ratio does not explicitly elaborate whether it is the numerator or the denominator that has altered. Recent ¹H-MRS research revealed that absolute quantification outperforms relative ratio in distinguishing patients from healthy controls (Watanabe et al., 2008).

Gamma-aminobutyric acid (GABA) is the main inhibitory neurotransmitter in the human central nervous system. Quantitative measurement of intra-cerebral GABA using conventional ¹H-MRS is restricted because the low content of GABA (Govindaraju et al., 2000) and the signal of GABA greatly overlap with other main metabolites such as glutamate (Glu), NAA, Cr and macromolecules (MM). While GABA concentration can be reliably measured in vivo by J-edited MRS (Mescher et al., 1998). Mescher-Garwood point resolved spectroscopy (MEGA-PRESS) is the most common used J-edited MRS approach for quantifying GABA in human brain, which made the signal of GABA largely free from overlap with more abundant molecules, and it has been applied successfully to measure GABA in several neurologic and psychiatric disorders (Marenco et al., 2016; Chen et al., 2019) and healthy control subjects (Gao et al., 2013; Frank et al., 2022).

However, to date, there has been a paucity of studies exploring changes in GABA concentration in WMHs patients. The primary purpose of the present study was to determine whether there were abnormalities in the level of GABA, and alterations in the levels of NAA, Cr, Cho, and Glx in WMHs patients.

In this study, the brain GABA+ levels of 24 WMHs patients and 20 healthy controls were assessed with J-edited MR spectroscopy. Further, the levels of other major metabolites (including Glx, tNAA, tCho, and tCr) have also been analyzed. This study demonstrates significant reductions of GABA+ levels in centrum semiovale in WMHs patients when compared to age and gender-matched healthy controls. In addition, we found that tCho levels were elevated in moderate-severe WMHs subgroup patients compared to healthy controls.

The spectroscopic volumes that were chosen for edited MR spectroscopy in this study was centrum semiovale. This is because centrum semiovale is the area of the so-called boundary zone positioned distal to the deep and superficial branches of the middle cerebral artery. This territory is particularly susceptible to inadequate blood perfusion or ischemia secondary to risk elements associated with the deep vascular region, and this region of interest is frequently chosen in patients with cerebral small vessel disease (Nitkunan et al., 2006, 2009).

The primary finding of this study is that the GABA+ level estimated from J-edited MRS using MEGA-PRESS was substantially lower in WHMs patients when compared to healthy controls. GABA is the primary inhibitory neurotransmitter in the central nervous system and plays an important role in effective information processing in the brain (Bartos et al., 2007) as well as normal behavioral performance or cognitive function (Boy et al., 2011). Therefore, reduced GABA+ concentrations identified in this study may be underlying pathological mechanisms for the concealed cognitive deficits in WMHs patients, as reported in individuals with cerebral small vessel disease (Zanon Zotin et al., 2021). In the subgroup analysis, lower levels of GABA+ were also detected both in moderate-severe and mild WMHs subgroup patients compared to HCs. This may imply that abnormalities in GABA+ levels develop early in the disease, and variations in GABA+ levels have a high sensitivity for studying pathological processes in WMHs. Moreover, the gray matter tissue fraction showed a significant difference between moderate-severe WMHs subgroup and HCs but not between mild WMHs subgroup and HCs, which may suggest that metabolic changes precede structural brain changes. Further exploration of the reason for this phenomenon is needed.

Our findings also revealed that the level of tCho in the centrum semiovale was considerably higher in patients with moderate-severe WMHs than healthy controls. tCho is involved in the composition of cell membranes, reflecting cell metabolism and myelin formation. Increased tCho levels may reflect membrane breakdown or demyelination (De Stefano et al., 2005; Lin et al., 2005), and the elevated tCho levels in our study represent ischemic brain injury in patients with moderate-severe WMHs. Similarly, a previous study found patients with transient ischemic attack within 3 days of onset show an increase in choline (Bisschops et al., 2002). Ischemia or hypometabolism are also potential pathogenic mechanisms for LA (Sun et al., 2022). In the subgroup analysis, the level of tCho showed no significant increase in mild WMHs subgroup patients compared to healthy controls, which suggests that mild cerebral white matter hyperintensities do not cause changes in the level of tCho until the disease has progressed to a certain level.

Interestingly, we did not find a significant difference in the level of tNAA between patients with WMHs patients and healthy controls, which contract to the results of the previous study (Wang et al., 2012), which may be a result of the small sample size, the severity of white matter lesions, or the clinical heterogeneity. In addition, given the varying severity of WMHs, part of moderate-severe patients with frontal-parietal as well as paraventricular demyelination contains white matter lesion within the MRS voxels, and the inclusion of demyelinating lesions within the voxels may affect metabolite quantification outcomes, which needs further investigation. It is equally worthy of attention that a significant difference in hypertension was found between the WMHs patients and healthy controls, this may indicate that WMH patients are more likely to have concomitant hypertension than healthy individuals, in other words, hypertension is one of the underlying pathogenic mechanisms in WMHs (Markus and de Leeuw, 2023). Furthermore, it has been shown that intensive blood pressure lowering slows the progression of WHMs (Nasrallah et al., 2019). Our findings to some extent reveal potential metabolic markers of WMHs; in future studies, we will further explore which high-risk factors are associated with metabolic changes in WMHs patients to better prevent the occurrence of WHMs in individuals.

Our study has some limitations that need to be discussed. Firstly, previous studies have shown that GABA concentrations vary with location (Evans et al., 2010; Gao et al., 2013), thus further studies should be performed to explore whether there are also differences in GABA concentration in other different brain regions (e.g., the right centrum semiovale, the frontal lobe, the temporal lobe or hippocampus). Secondly, the edited GABA signal contains a significant contribution from co-edited macromolecules and homocarnosine. A previous study has developed a new method for macromolecule suppression to extract pure GABA (Edden et al., 2012), but the technique has practical limitations: MM-only spectra is time-consuming and the signal-to-noise ratio of GABA may be degraded; acquiring MM-suppressed edited experiments makes the acquisition substantially more sensitive to field changes. Additionally, for the majority of applications, variations in macromolecules between subjects are slightly compared with the variation in GABA levels (Kreis et al., 2005). Homocarnosine concentration is much less than GABA, and its impact on the results can be ignored. Comprehensively considered, GABA+ measurement is more appropriate. Finally, our results show the moderate-severe WMH group has a decreased mean GABA+ levels but increased median GABA+ levels than mild WMH group, which may be due to our modest sample size. Thus, future studies with a larger cohort of participants will boost the stability of our study results.

In conclusion, we found reduced GABA+ levels in WMHs patients and increased tCho levels in moderate-severe WMHs subgroup patients compared to healthy controls. These findings support the potential for MRS to establish GABA+ and tCho measurements as clinically relevant metabolic markers for WMHs, although further research efforts are needed to better understand the findings reported here.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving humans were approved by The First Affiliated Hospital of Zhengzhou University. The studies were conducted in accordance with the local legislation and institutional requirements. The participants provided their written informed consent to participate in this study.

XW: Writing – original draft, Writing – review and editing. CW: Supervision, Writing – original draft, Writing – review and editing. PM: Resources, Writing – original draft. YW: Resources, Writing – original draft. LL: Conceptualization, Software, Writing – review and editing. ZL: Supervision, Writing – review and editing. YZ: Supervision, Validation, Writing – review and editing. JC: Funding acquisition, Writing – review and editing. CR: Supervision, Writing – review and editing.