γ-aminobutyric acid spectroscopy of the thalamus in diabetic neuropathy

Iain D Wilkinson¹, Pillai Shillo¹, Marni Greig¹, Solomon Tesfaye¹, Richard A Edden², and Dinesh Selvarajah¹

¹Academic Radiology, University of Sheffield, Sheffield, United Kingdom, ²Kennedy Krieger Institute, Baltimore, MA, United States

Synopsis

The sensory system is affected in diabetic neuropathy (DN), a common ailment associated with diabetes mellitus. The thalamus is part of the brain’s sensory pathway. This study applies MEGA-PRESS to assess thalamic GABA in-vivo in patients with and without DN. Differences in GABA/H20 ratios were identified between those with and without DN, demonstrating potential differences in the neuronal inhibitory status of the thalamus.

Purpose

To ascertain the thalamic γ-aminobutyric acid (GABA) features on proton spectroscopy in patients with and without diabetic neuropathy and in matched subjects without diabetes.

Introduction

The thalamus forms a major component of the ascending sensory pathway to the brain, the function of which may play an important role in patients with diabetic neuropathy (DN). Previous Magnetic Resonance (MR) studies have demonstrated thalamic metabolite involvement (particularly N-Acetyl Aspartate) on proton MR Spectroscopy (1,2). γ-Aminobutyric acid (GABA) is an inhibitory neurotransmitter that can be assessed using proton spectral editing techniques (3). This study assessed thalamic GABA in-vivo using single-voxel MEGA-PRESS in well phenotyped patients with and without DN.

Methods

Forty-four type 2 diabetes (T2DM) Caucasian subjects (14 Painful-DPN, 15 Painless-DPN and 15 No-DPN) and 15 healthy volunteers without diabetes (HV) underwent detailed clinical and neurophysiological assessments. Type-2 diabetic subjects were divided into three groups based on a neuropathy composite score [NIS(LL)+7 and Douleur Neuropathique 4 score(DN4) assessed using recognized clinical and physiological measures. All groups were age-matched. Subjects underwent localised H-MRS at 3T (Ingenia, Philips Healthcare, Best, NL) to assess GABA relative to unsuppressed water and creatine using a single-voxel, spin-echo, spectral editing technique (MEGAPRESS; echo time=68ms) centred over thalami. Post-acquisition spectral analysis was performed using the Gannet software toolkit (4). The fitted GABA resonance area was expressed relative to that of parenchymal water.

Results

Analysis of Variance revealed differences in group mean GABA/H20 ratios [ANOVA p<0.01; Painless-DPN 1.47(sd=0.23), Painful-DPN 1.61(0.33), HV 1.75(0.25) and T2DM with No-DPN 1.84(0.38)]. Post-hoc comparisons indicated significantly lower mean GABA/H20 in Painless-DPN compared to No-DPN (p< 0.005) and significantly lower mean GABA/H20 in Painless-DPN compared with HV (p<0.05).

Discussion

This study containing a well-sized, well-characterized cohort demonstrates lower relative levels of the ‘H-MRS-visible’ inhibitory neurotransmitter GABA in the thalamus of patients with DPN. Previous published pilot data from 7 diabetics with neuropathy and 7 non-diabetics reported lower GABA in the posterior insula with no significant difference in the thalami in the context of DN (5). Our current larger sample size indicates significantly lower GABA within the thalamus often considered the sensory gateway to the brain. A further understanding of the cerebral neuronal excitatory/inhibitory balance inferred from this HMRS technique may help determine the mechanistic basis of central nervous system involvement in pain perception associated with DN.

Acknowledgements

Many thanks to the volunteers and technicians.

References

1. Sorensen L et al; Differences in metabolites in pain-processing brain regions in patients with diabetes and painful neuropathy. Diabetes Care. 2008; 31(5):980-1.

2. Selvarajah D et al; Thalamic neuronal dysfunction and chronic sensorimotor distal symmetrical polyneuropathy in patients with type 1 diabetes mellitus , Diabetologia, 2008; 51(11):2088-92.

3. Mullins PG et al; Current practice in the use of MEGA-PRESS spectroscopy for the detection of GABA, Neuroimage 2012; 86:43-52.

4. Eden RA et al; Gannet: A batch-processing tool for the quantitative analysis of gamma-aminobutyric acid–edited MR spectroscopy spectra. JMRI 2014; 40(6):1145-52.

5. Petrou Met al; Aletered excitation inhibition balance in the brain of patients with diabetic neuropathy. Acad Radiology 2012; 19(5):607-612.

Figures

Fig. 1: Example subtracted spectrum using the MEGA_PRESS technique. GABA(+) resonance is shown at 3ppm.

Fig. 2: Mean GABA/water ratio for the 4 study groups. The differences between the painful-neuropathy group and both the non-neuropathy diabetic group and the non-neuropathy non-diabetic group show statistical significance (ANCOVA, p<0.05).

Proc. Intl. Soc. Mag. Reson. Med. 24 (2016)

1231