hide
Free keywords:
-
Abstract:
Introduction:
Spinal cord injury (SCI) leads to immediate sensorimotor and autonomic dysfunction below the level of injury and can result in neuropathic pain (NP). Along with neural structural changes [1], neurodegeneration within the sensorimotor system is reflected by biochemical changes in the neural tissue [2]. Magnetic resonance spectroscopy (MRS) non-invasively detects and quantifies several metabolites in vivo which entails molecular information of the tissue. Alterations of cellular processes after SCI have been associated with neurodegeneration in the thalamus [3] and anterior cingulate cortex [4] of SCI patients with NP. However, little is known about how metabolites change after SCI in the cervical cord above the lesion in NP patients. This study therefore aimed to investigate metabolic changes at cervical level in chronic SCI patients with or without NP and assessed the relationships of these biomarkers to clinical measures of NP.
Methods:
We recruited 14 chronic SCI patients with NP (12 men, age=52.2±10.5 [y], years since injury=11.3±9.2), 10 patients without NP (10 men, age=50.0±10.3 [y], years since injury=18.4±10.5), and 21 healthy controls (HC, 18 men, age=45.0±11.9 [y]). All participants underwent MRS measurement on a 3T scanner (Philips, Netherlands) with a 16 channel SENSE neurovascular coil. T2-weighted images (0.5x0.5x3.2 mm3) were used to place the spectroscopic voxel (6x9x35 mm3) at spinal level C2 and the metabolite cycling (MC) technique [5] was applied. Each MRS measurement contained 512 signal averages and the data were fitted using LCModel [6]. MRS allowed us to reliably detect total N-Acetyl-Aspartate (tNAA), choline containing compound (tCho), and myo-Inositol (mI) at cervical level (CRLB<25%). Groups were compared regarding their tNAA/mI and tCho/mI ratios. All patients were clinically assessed using the International Standards for Neurological Classification of Spinal Cord Injury (ISNCSCI) protocol for light-touch and pin-prick scores [7]. Statistical analyses were performed using R (R Core Team, 2016, Version 3.4.3). Group differences were assessed with the Kruskal-Wallis test and Spearman's rank correlations were used to investigate associations between metabolic ratios and clinical outcome. Results with an uncorrected p-value ≤ 0.05 were regarded as significant.
Results:
We found lower tNAA/mI (P=0.005) and tCho/mI (P=0.002) ratios in chronic SCI patients without NP compared to HC (Fig. 1). Additionally, tCho/mI was lower in SCI patients without NP in comparison to patients with NP (P=0.024). However, in patients with NP, tCho/mI and tNAA/mI were at a similar level as in HC. As shown in Fig. 2, the tCho/mI ratio was positively associated with pin-prick (P=0.006, R2=0.298) and light-touch (P=0.001, R2=0.381) scores.
Conclusions:
This study shows that chronic SCI patients without NP had decreased tNAA/mI and tCho/mI ratios compared to HC and lower tCho/mI ratios compared to patients with NP. Reductions of these ratios can either result from an increase of mI (marker for reactive gliosis and neuro-inflammation) or a decrease in tNAA (neuronal cell integrity marker) and tCho (marker for cell membrane turnover), or both. Thus, our findings could point towards either ongoing neuro-inflammation or neurodegeneration in patients without NP, or both. Patients with NP, on the other hand, are likely to have more spinothalamic tract-specific viable neurons mediating NP and therefore more closely resemble HC. Worse clinical outcome measures were associated with lower tCho/mI ratios, likely originating from reactive astrocytes counteracting axonal regeneration and mediating the clinical impairment [8]. In conclusion, sensitive MRS at the cervical cord can provide metabolic biomarkers underlying neurodegenerative changes and neuro-inflammation after SCI which have the potential to be used in clinical trials for patient stratification and therapy monitoring.
