MR properties of excised neural tissue following experimentally induced inflammation

Abstract

Changes in the MR parameters of inflamed neural tissue were measured in vitro. Tumor necrosis factor-alpha (TNF-␣) was injected into rat sciatic nerves to induce inflammation with negligible axonal loss and demyelination. The MR parameters, such as T 1 /T 2 relaxation and magnetization transfer (MT), were measured 2 days after TNF-␣ injection and were found to be substantially different from those of normal nerves. The average T 1 /T 2 relaxation times increased, whereas the MT ratio (MTR) and the quantitative MT parameter M 0B (which describes the semisolid pool of protons) decreased. The MR parameters correlated very well with the extracellular volume fraction (EM) of neural tissue evaluated by quantitative histopathology. The multicomponent T 2 relaxation was shown to provide the best quantitative assessment of changes in neural tissue microstructure, and allowed us to distinguish between the processes of inflammation and demyelination. In comparison, the MT measurements were less successful due to competing contributions of demyelination and pH-sensitive changes in the MT effect. Inflammation commonly occurs in a wide spectrum of nervous system diseases, including multiple sclerosis (MS) (1), stroke (2), dementia (3), and traumatic brain injury (4). It is difficult to distinguish the processes of inflammation from those of neural tissue degeneration, such as axonal loss and demyelination, because similar qualitative changes occur in the MR signal intensities of T 1 -, T 2 -, or magnetization transfer (MT)-weighted images. MRI evaluations of tissue pathology are further complicated by the fact that inflammation, axonal loss, and demyelination often occur concurrently. The precise identification of the pathophysiological processes that occur in nervous system disorders is of considerable importance in our ability to accurately diagnose disease, understand the mechanisms and dynamics of neural tissue degeneration, and evaluate treatment efficacy. There is, therefore, a great need to establish MRI protocols that are capable of differentiating and quantitatively assessing these pathologies.Quantitative MR measures are thought to provide more specific information about changes in tissue microstructure. For example, the multicomponent T 2 relaxation is sensitive to white matter (WM) demyelination in multiple sclerosis (MS). The observed short T 2 component, which occurs at approximately 10 -20 ms, has been associated with water in the myelin sheath (5), and thus provides a measure of the processes of demyelination and remyelination. We recently demonstrated that the magnitude of the short T 2 component correlates very well with the amount of myelin in nerves undergoing degeneration and spontaneous regeneration following traumatic injury of the peripheral nervous system (PNS) (6). Conversely, the MT effect is thought to be mainly mediated by the processes of MT exchange between water and lipid protons within the myelin sheath (7,8). It has been shown (9) that the observed changes in the MT ratio (MTR) and the decrea...

Myelination and long diffusion times alter diffusion-tensor-imaging contrast in myelin-deficient shiverer mice

“…Although the sensitivity of T 2 to detect changes in myelin has been an issue of debate, myelin content had been reported to correlate with T 2 (MacKay et al, 1994;Stanisz et al, 2004). We have no definitive explanation for this apparent discrepancy.…”

Characterizing healthy and diseased white matter using quantitative magnetization transfer and multicomponent T2 relaxometry: A unified view via a four‐pool model

“…Real T 2 increases likely to be present in the case of true edema clearly affected the observed ͗T 2 ͘ and ͗T 2ie ͘; moreover, we observed that the position of the ͗T 2ie ͘ peak in the distribution does not strictly follow the model value of T 2ie but is underestimated by the QT2 analysis. The different effects of demyelination and edema on the T 2 distribution may allow the distinction of the two effects using in vivo imaging in human WM, as previously suggested by Stanisz and colleagues (33,39). A linear relationship was assumed between increases in the IE water content and T 2ie in these simulations as the functional form of this relationship is not distinguishable from published data (33): better knowledge regarding this relationship would help determine the potential of QT2 to distinguish between edema and demyelination.…”

“…Specifically, we performed simulations reducing both compartments by the same relative amount of 25 to 100%. Finally, edema simulations were also performed with increases of up to 25% in the IE water compartment size, in 5% steps, increases larger than reported in MS (32) but selected to cover of the potential range of experimentally induced inflammation (33). In addition, these simulations were performed with and without an increase of T 2ie of up to 100%, linearly scaled with the increase of IE water content.…”

Differentiation and quantification of inflammation, demyelination and axon injury or loss in multiple sclerosis

“…For specifically identifying axon injury and loss, reduced NAA content measured by magnetic resonance spectroscopy (MRS) has been used (De Stefano et al, 1999;Aboul-Enein et al, 2010;Wood et al, 2012). However, inflammation can interfere with measuring myelin and axons because inflammation-induced tissue swelling falsely reduces the measured myelin water and NAA content (oedema water contains no NAA and dilutes myelin water fraction), MTR (oedema water exhibits no transfer effect while contributing to the total signal), and PET marker intensity (diluting marker density) (Schmierer et al, 2004;Stanisz et al, 2004;McCreary et al, 2009;Vavasour et al, 2011). Despite that DTI-derived axial diffusivity and radial diffusivity can concurrently quantify coexisting axonal injury and demyelination in some settings (Song et al, 2002(Song et al, , 2003(Song et al, , 2005, DTI becomes inaccurate in the presence of inflammation (Sun et al, 2006b;Lodygensky et al, 2010;Xie et al, 2010), tissue loss (Kim et al, 2007), crossing fibres (Wheeler-Kingshott and Cercignani, 2009), and CSF contamination (Karampinos et al, 2008;Cheng et al, 2011) and cannot measure the inflammatory components.…”