Research ArticleMagnetic resonance quantification of water and metabolites in the brain of cirrhotics following induced hyperammonaemia
Introduction
Hepatic encephalopathy (HE) is an important complication of both acute and chronic liver disease [1]. About 50% of patients with cirrhosis, being evaluated for liver transplantation in our unit, exhibit minimal hepatic encephalopathy [2], while clinically overt HE occurs in approximately 35% of patients with end-stage liver disease, especially in those requiring therapeutic transjugular intrahepatic shunts [3]. Grade 2 or greater HE in hospitalised patients is associated with a 3.9-fold increase in risk of death [4].
HE in cirrhosis is now thought to be related to the development of low grade cerebral oedema [5]. However, current treatments for this condition are controversial [6], [7], [8]. In a previous study, we have shown reversal of minimal encephalopathy by liver transplantation [2] but simpler more effective treatments are required. The shortage of treatments for HE may be explained by the lack of consensus regarding the pathogenesis of cerebral oedema and its association with the development of the condition. This is due to the fact that the processes leading to cerebral oedema are complex and multifactorial. One possible driving mechanism for oedema development in cirrhosis is an elevation in blood ammonia levels in these patients compared to control subjects, ammonia being a known neurotoxin. A recent study has suggested that venous ammonia is associated with cerebral oedema severity in patients with cirrhosis [9].
Diffusion tensor MRI (DTI) is a sensitive technique to examine water distribution in the brain. Diffusion of water within brain tissues is restricted by collisions with cellular and sub-cellular structures, most notably the cell membrane [10], [11], leading to a measured apparent diffusion coefficient (ADC), which is significantly less than free water diffusion and reflects the compartmentalisation of water within the tissue. DTI has been used extensively in the clinical diagnosis of acute brain ischaemia where the ADC of brain-water falls by up to 50% within seconds of loss of cellular homeostasis [10]. These changes are often interpreted as reflecting shifts in compartmentalisation of water within the brain parenchyma, and the formation of cytotoxic oedema. DTI, therefore, provides a sensitive tool to monitor brain water homeostasis in vivo, and this is supported by in vitro data in brain slice preparations and cultured cells in the context of detection of osmotic changes. In isolated turtle cerebellum exposed to a hypotonic environment, ADC was reported to fall, while under hypertonic conditions, ADC rose significantly [12], supporting the hypothesis that changes in ADC can reflect cellular swelling. A similar finding has also been reported in rat brain hippocampal slices [13].
Recently, DTI has been applied to a cohort of patients with end-stage liver disease and HE. In these patients, ADC was found to be increased by up to 15% compared with control subjects, reflecting the extent of brain oedema, suggesting a potential role for DTI in diagnosing or monitoring the progression of HE [9]. ADC was also found to be correlated with HE severity in non-alcoholic liver disease [14]. These previous studies show that ADC is useful for detecting changes in brain water in HE, but the cross-sectional nature of the data prevents these studies from directly addressing the mechanisms driving oedema development. Here, we aimed to evaluate dynamic changes in brain water and metabolites in cirrhotic patients, before and after hyperammonaemia induced by amino acid challenge, to determine the pathogenesis of cerebral oedema development. This was assessed non-invasively by diffusion tensor imaging (DTI) and also by proton magnetic resonance spectroscopy (MRS), an associated technique which measures key brain metabolites in HE [15]. We hypothesised that hyperammonaemia would dynamically alter brain water diffusivity, reflecting shifts in brain water concentration and that this would be related to alterations in cerebral metabolites associated with ammonia detoxification and cell volume regulation.
Section snippets
Subjects
Thirteen out-patients with well characterised stable cirrhosis, evidence of portal hypertension, and no evidence of infection were recruited from those attending for assessment of their suitability for liver transplantation. The severity of liver disease was determined according to the Child-Pugh [16] and Meld [17] score. Full demographic and psychometric data is provided in Table 1. The study was approved by the Joint Ethics Committee of Newcastle and North Tyneside Health Authority,
1H MR spectroscopy
During each scanning session, all patients also underwent metabolic measurements using proton MRS to measure myo-inositol, glutamate, glutamine, creatine, choline, and N-acetyl-aspartate (NAA). A localised 8 cm3 voxel located in frontal white matter was sampled and a fully relaxed MR spectrum was acquired with a short TE PRESS sequence using a TR of 3 s, TE of 36 ms, 128 averages, 1024 samples, and a spectral bandwidth of 2000 Hz [20]. A series of non-water suppressed spectra were also collected at
Blood ammonia and plasma amino acid concentrations
A maximum increase in venous blood ammonia concentration of 58 + 41 μmol/L (p <0.0001) was observed 1–3 h after amino acid administration (Fig. 1A). Basal and 120-min plasma amino acid profiles are shown in Table 2. As expected, there was a large (8–9-fold) increase in plasma levels of the administered amino acids threonine, serine, and glycine. This resulted in up to 2-fold increases in plasma taurine, aspartate, asparagine, glutamine, proline, alanine, and citruline while gamma amino butyric acid
1H MRS
Myo-inositol concentration varied widely amongst patients (ranging from 2.2 mMol/L down to undetectable in one patient, attributed with zero concentration) consistent with previous reports in HE [27]. The mean pre and post challenge myo-inositol concentration was 0.8 ± 0.7 mMol/L and 0.1 ± 0.1 mMol/L, respectively, showing a significant reduction in myo-inositol concentration after the challenge (p <0.005, Fig. 1C). There were no significant differences between pre (21.6 ± 6.4 mMol/L) and post challenge
Discussion
In a previous study in patients with cirrhosis, we demonstrated that hyperammonaemia from an oral glutamine challenge is associated with a rise in blood ammonia and deterioration in psychometric scores, suggesting that induced hyperammonaemia is a suitable model for investigating the pathophysiology and treatment of HE [28]. Since gastrointestinal bleeding is a known precipitant of HE, we have also used an amino acid mixture resembling the composition of haemoglobin and have shown that EEG
Conclusions
This study shows that ammonia can directly drive changes in brain water distribution as a mechanism for cerebral oedema development in patients with HE. Since cerebral astrocytes contain glutamine synthetase, our MRS data suggests intracerebral formation of glutamine from ammonia. The rapid decrease in myo-inositol suggests that this organic osmolyte may play a protective role in HE and is released from astrocytes in order to maintain cell volume. Low levels of myo-inositol concentration may
Conflict of interest
The authors who have taken part in this study declared that they do not have anything to disclose regarding funding or conflict of interest with respect to this manuscript.
Financial support
The study received financial support from the Newcastle Healthcare Charity.
References (46)
- et al.
Computerized psychometric testing in minimal encephalopathy and modulation by nitrogen challenge and liver transplant
Gastroenterology
(2008) - et al.
Dispelling myths in the treatment of hepatic encephalopathy
Lancet
(2005) Treatment of hepatic encephalopathy
Lancet
(2005)- et al.
Diagnosis of hepatic encephalopathy by proton magnetic resonance spectroscopy
Lancet
(1990) - et al.
Detection of minimal hepatic encephalopathy: normalization and optimization of the Psychometric Hepatic Encephalopathy Score. A neuropsychological and quantified EEG study
J Hepatol
(2008) - et al.
Comparison of the effect of various amino acids upon the blood ammonia concentration of patients with liver disease
Am J Clin Nutr
(1973) - et al.
Absolute quantification of water and metabolites in the human brain I. Compartments and water
J Magn Reson Ser B
(1993) - et al.
A global optimisation method for robust affine registration of brain images
Medical Image Analysis
(2001) - et al.
SVD-based quantification of magnetic-resonance signals
J Magn Reson
(1992) - et al.
Oral glutamine challenge in cirrhotics awaiting liver transplantation: a psychometric and analysed EEG study
Hepatology
(1997)
Amino acid challenge in patients with cirrhosis: a model for the assessment of treatments for hepatic encephalopathy
J Hepatol
Induced hyperammonemia alters neuropsychology, brain MR spectroscopy and magnetization transfer in cirrhosis
Hepatology
Quantitative cerebral water content mapping in hepatic encephalopathy
Neuroimage
Proton magnetic-resonance spectroscopy studies on human brain myo-inositol in hypo-osmolarity and hepatic-encephalopathy
Gastroenterology
Hepatic encephalopathy after transjugular intrahepatic portosystemic shunt insertion: a decade of experience
Q J Med
Hepatic encephalopathy as a predictor of survival in patients with end stage liver disease
Liver Transplant
Pathogenic mechanisms of hepatic encephalopathy
Gut
Treatment of hepatic encephalopathy
Am J Gastroenterol
Diffusion MRI shows increased water apparent diffusion coefficient in the brains of cirrhotics
Neurology
Longitudinal magnetic resonance imaging study of perfusion and diffusion in stroke: evolution of lesion volume and correlation with clinical outcome
Ann Neurol
The basis of anisotropic water diffusion in the nervous system – a technical review
NMR Biomed
Apparent diffusion coefficient and MR relaxation during osmotic manipulation in isolated turtle cerebellum
Magn Reson Med
Cited by (0)
- †
Authors H.M. and F.E.S. are co-first authors on this manuscript.