Submitted on March 9, 2004
Accepted on May 5, 2004
Low myo-inositol and high glutamine levels in the brain are associated with neuropsychological deterioration following induced hyperammonemia
D. L Shawcross1, S. Balata2, S. W. M Olde Damink3, P. C Hayes2, J. Wardlaw4, I. Marshall5, N. E. P Deutz3, R. Williams1, and R. Jalan1*
1 Liver Failure Group, Institute of Hepatology, University College London, London, United Kingdom
2 The Scottish Liver Transplantation Unit, Royal Infirmary of Edinburgh, Edinburgh, Scotland, United Kingdom
3 Department of Surgery, Academic Hospital, Maastricht University, Maastricht, The Netherlands
4 Division of Clinical Sciences, Western General Hospital, University of Edinburgh, Edinburgh, Scotland, United Kingdom
5 Department of Medical Physics, University of Edinburgh, Edinburgh, Scotland, United Kingdom
* To whom correspondence should be addressed. E-mail: r.jalan{at}ucl.ac.uk.
Background: The neuropsychological effect of hyperammonemia is variable. This
study tests the hypothesis that the effect of ammonia on the neuropsychological
function in patients with cirrhosis is determined by the ability of the brain to 'buffer'
ammonia-induced increase in glutamine within the astrocyte by losing osmolytes like
myo-inositol (mI) and not by the magnitude of the 'induced hyperammonemia'.
Methods: Fourteen cirrhotic patients with no evidence of overt HE were given a 75
gram amino acid (aa) solution mimicking the hemoglobin molecule to induce
hyperammonemia. Measurement of a battery of neuropsychological function tests
including immediate memory, ammonia, aa and short-echo time proton magnetic
resonance spectroscopy were performed prior to and 4 hours after administration of
the aa solution.
Results: Eight patients showed deterioration in the Immediate Memory Test at 4
hours. Demographic factors, severity of liver disease, change in plasma ammonia and
aa profiles following the aa solution were similar in those that showed a deterioration
compared with those who did not. In the patients who showed deterioration in the
memory test, the mI/creatine ratio was significantly lower at baseline than those that
did not deteriorate. In contrast, the Glx/creatine ratio was significantly greater in the
patients that deteriorated.
Conclusions: The observation that the deterioration in the memory test was greater in
those with lower mI/creatine ratio supports the hypothesis that the neuropsychological
effects of induced hyperammonemia is determined by the capacity of the brain to
handle the ammonia-induced increase in glutamine.
