Elsevier

NeuroImage

Volume 41, Issue 3, 1 July 2008, Pages 706-717
NeuroImage

Quantitative cerebral water content mapping in hepatic encephalopathy

https://doi.org/10.1016/j.neuroimage.2008.02.057Get rights and content

Abstract

There is increasing evidence that the pathophysiology of hepatic encephalopathy is tightly associated with low-grade cerebral oedema; however, no method has yet specifically and unambiguously confirmed this hypothesis in vivo. The current study describes the quantitative measurement of localised water content using MRI in a cohort of 38 patients suffering from hepatic encephalopathy. A significant global increase in cerebral water content was observed in white matter whereas water content in grey matter was globally unaffected. However, significant spatial variations in the water content distribution, especially in grey matter, were observed and were correlated with disease grade and critical flicker frequency. In addition, regions-of-interest were defined and a significant change in water content with disease grade was found in the frontal and occipital white matter, the globus pallidus, the anterior limb of the internal capsule and the putamen. No association of water content and HE grade was established for the occipital visual and frontal cortices, the thalamus, the posterior limb of the internal capsule, the caudate nucleus and the coronal white matter. In conclusion, the measurements presented here are the first direct and quantitative demonstration of the presence of low-grade cerebral oedema in patients with hepatic encephalopathy. Further, absolute changes in tissue water content were quantified for various brain regions.

Introduction

Hepatic encephalopathy is a multi-factorial disease and a common neuropsychiatric complication in patients with cirrhosis. Neurological manifestations of this disease are thought to result from the presence of a low-grade cerebral oedema (Häussinger et al., 1994, Häussinger et al., 2000). Hepatic encephalopathy is, at least in part, a consequence of astrocyte swelling mediated by cytokines, hyponatremia, benzodiazepines and neurotoxins such as NH3 and it has been proposed that the amount of swelling and the resulting brain oedema correlate with disease grade (Häussinger et al., 2000, Spahr et al., 2000, Rovira et al., 2001, Cordoba et al., 2001, Spahr et al., 2002, Rovira et al., 2002).

The direct assessment of cerebral water changes in HE is largely hindered by the non-quantitative measurement of water content changes associated with low-grade oedema. Magnetic resonance imaging (MRI), despite being a very sensitive tool for the detection of functional or structural changes in the human brain, has only provided indirect evidence thus far to support the hypothesis for an association between HE and cerebral oedema. Using the signal intensity of T2-weighted FLAIR images, Rovira et al. (2002) concluded that the observed changes were characteristic for increased water content in cirrhotic patients. Increased white matter signal intensity in or around the corticospinal tract in cirrhotic patients was observed by these authors and was interpreted as a manifestation of low-grade cerebral oedema. Results from other measures of H2O content, including magnetisation transfer contrast enhanced imaging, which is an indirect measure, and MR spectroscopy, a more direct measure, also suggest the presence of a low-grade cerebral oedema in hepatic encephalopathy (Häussinger et al., 1994, Häussinger et al., 2000, Spahr et al., 2000, Rovira et al., 2001, Cordoba et al., 2001, Spahr et al., 2002, Rovira et al., 2002, Shah et al., 2003, Lodi et al., 2004, Miese et al., 2006, Kale et al., 2006, Laubenberger et al., 1997). However, even though these conventional MR techniques are relatively sensitive to pathological alterations, they lack specificity. Magnetisation transfer contrast, e.g., is not only determined by the total water content, but is sensitive to the exchange properties between bound protons (e.g. in macromolecules) and free protons (e.g. in water). A similar argument applies to the transverse relaxation time, T2, which is, above all, sensitive to changes in the molecular environment. Therefore, none of these conventional methods is able to quantitatively and unambiguously measure the presence of brain oedema in vivo, its true spatial extent or the brain regions affected in patients with hepatic encephalopathy.

MRI, however, is intrinsically a quantitative procedure enabling the direct and quantitative measurement of parameters such as tissue relaxation times or water content. In the work described here, we have used a recently published and validated method for the quantitative mapping of water content in vivo to study the putative changes in cerebral hydration associated with minimal and manifest hepatic encephalopathy (Neeb et al., 2004, Neeb et al., 2006a, Neeb and Shah, 2006, Neeb et al., 2006b). The method enables the absolute quantification of water content with high precision within clinically relevant measurement times (approx. 20 min). Furthermore, due to the quantitative nature of the data acquired, automated image analysis with the possibility to extract derived quantitative parameters becomes feasible. We have recently developed a method to extract parameters from water content maps which enable an automated assessment of global water content changes in grey and white matter as well as changes in the spatial distribution of water in the brain (Neeb et al., 2006b). It was shown, e.g., that these derived quantities enable a precise prediction of the phenotypes age and gender in normal healthy subjects.

Thus, a two-fold approach has been taken in the current study. First, regions-of-interest were interactively defined in the putamen, the white matter of the corona radiata, the frontal and occipital white matter, the occipital visual and frontal cortices, the globus pallidus, the thalamus, the caudate nucleus and the posterior and anterior limb of the internal capsule; from these individual regions average water content was recorded. Correlations between the water content in the brain regions investigated and the different severities of HE as well as the critical flicker frequency (CFF) were investigated. It is to be noted that CFF is a sensitive marker for disease-associated changes (Kircheis et al., 2002). Second, white and grey matter were segmented and the parameters described in Neeb et al. (2006b) were determined. Furthermore, it was investigated whether the current classification of minimal HE (MHE), i.e. the separation between the grades HE-0 and MHE, is also reflected in the multivariate combination of a typical parameter used for the clinical assessment such as CFF and the information acquired with quantitative MRI. The results obtained in this study highlight the relevance of quantitative imaging using MRI as a valuable tool in clinical research and diagnosis of hepatic encephalopathy.

Section snippets

Patient population

Fifty-four patients with different grades of HE were studied from whom reliable water content maps were reconstructed for thirty-eight patients only (13 HE-0, 12 MHE, 10 HE-I and 3 HE-II). Strong artefacts caused by movement during the measurement were observed for the remaining patients who were therefore removed from further data analysis. Due to the small number of patients in the HE-II group from whom artefact-free images could be acquired, data from HE-I and HE-II were pooled ("overt HE").

Results

A quantitative water content map from a randomly chosen patient with grade HE-0 is shown in Fig. 1a. The contrast between white and grey matter as well as CSF is clearly visible as a result of the differences in water content. The water content distribution at a similar anatomical location is remarkably different in a patient classified as HE-II as exemplified in Fig. 1b. The water content in this patient is higher; this increase is best visible within the basal ganglia and in white matter.

Discussion

The evidence for a close association between the pathophysiology of hepatic encephalopathy and low-grade cerebral oedema has significantly increased since the first report on changes observed in the myo-inositol and the glutamine/glutamate signal in HE patients using magnetic resonance spectroscopy (Häussinger et al., 1994). Magnetic resonance imaging and/or spectroscopy have been used as the main tool to investigate structural and functional changes which are characteristic for increased

References (46)

  • NeebH. et al.

    A new method for fast quantitative mapping of absolute water content in vivo

    NeuroImage

    (2006)
  • NeebH. et al.

    Fully automated detection of quantitative water content changes: application to the study of age- and gender related cerebral H2O patterns with quantitative MRI

    NeuroImage

    (2006)
  • O' CarrollR.E. et al.

    Regional cerebral blood flow and cognitive function in patients with chronic liver disease

    Lancet

    (1991)
  • ShahN.J. et al.

    A new method for fast multi-slice T1 mapping

    NeuroImage

    (2001)
  • ShahN.J. et al.

    Quantitative T1 mapping of hepatic encephalopathy using magnetic resonance imaging

    Hepatology

    (2003)
  • SpahrL. et al.

    Magnetic resonance imaging and proton spectroscopic alterations correlate with parkinsonian signs in patients with cirrhosis

    Gastroenterology

    (2000)
  • ZafirisO. et al.

    Neural mechanism underlying impaired visual judgement in the dysmetabolic brain: an fMRI study

    NeuroImage

    (2004)
  • BüchelC. et al.

    Modulation of connectivity in visual pathways by attention: cortical interactions evaluated with structural equation modelling and fMRI

    Cereb. Cortex

    (1997)
  • BüchelC. et al.

    The functional anatomy of attention to visual motion. A functional MRI study

    Brain

    (1998)
  • ConnH.O.

    Quantifying the severity of hepatic encephalopathy

  • FristonK.J. et al.

    Attentional modulation of effective connectivity from V2 to V5/MT in humans

    PNAS

    (2000)
  • HaberS.M. et al.

    The place of the thalamus in frontal cortical–basal ganglia circuits

    Neuroscientist

    (2001)
  • KaleR.A. et al.

    Demonstration of interstitial cerebral edema with diffusion tensor MR imaging in type C hepatic encephalopathy

    Hepatology.

    (2006)
  • Cited by (112)

    • Gut-liver-brain axis in chronic liver disease with a focus on hepatic encephalopathy

      2021, The Complex Interplay Between Gut-Brain, Gut-Liver, and Liver-Brain Axes
    • Longitudinal neurometabolic changes in the hippocampus of a rat model of chronic hepatic encephalopathy

      2019, Journal of Hepatology
      Citation Excerpt :

      These changes have since been thought to underlie the low grade cerebral oedema observed in chronic HE in humans. Later, other magnetic resonance imaging (MRI) techniques (i.e. diffusion weighted/tensor imaging, water mapping) further expanded our understanding of chronic HE by demonstrating the presence of low grade cerebral oedema.13,25–28 Although these studies were instrumental in demonstrating neurometabolic disturbances, their contribution was limited by comparatively low magnetic field strengths used and the ensuing limitation in the number of metabolites quantified (i.e. Glx, total creatine (tCr), total choline (tCho) and Ins5,28–31).

    View all citing articles on Scopus
    View full text