brain quinolinic acid levels correlate with the severity of hepatic encephalopathy. Methods: Postmortem samples of selected brain regions and plasma samples.
GASTROENTEROLOGY1995;108:818-823
The Relationship Between Plasma and Brain Quinolinic Acid Levels and the Severity of Hepatic Encephalopathy ANTHONY S. BASILE,*
KUNIAKI SAITO, t HANAAN A L - M A R D I N I , § CHRISTOPHER O. RECORD, §
ROBIN D. HUGHES, II PHIL HARRISON, H ROGER WILLIAMS, II YONG L I , * and MELVYN P. HEYES* *Laboratory of Neuroscience, National Institute of Diabetes and Digestiveand Kidney Diseases, National Institutes of Health, Bethesda, Maryland; tSection on Analytical Biochemistry, Laboratoryof Clinical Science, National Institute of Mental Health, National Institutes of Health, Bethesda, Maryland; ~GastroenterologyUnit, The Royal Victoria Infirmary, Newcastle-Upon-Tyne,England; and illnstitute of Liver Studies, King's College School of Medicine and Dentistry, London, England
Background/Aims: Quinolinic acid is an endogenous
neuroexcitant derived from tryptophan. Brain quinolinic acid concentrations are reportedly elevated in chronic liver failure. The aim of this study was to determine if brain quinolinic acid levels correlate with the severity of hepatic encephalopathy. Methods: Postmortem samples of selected brain regions and plasma samples taken at several stages of encephalopathy were obtained from patients with acute and chronic liver failure. Quinolinic acid levels were measured by mass spectroscopy using [180]quinolinic acid. Results: Plasma quinolinic acid levels were significantly increased by stage I encephalopathy in patients with acute liver failure and by stages II and III in patients with chronic liver failure. Brain quinolinic acid levels were elevated only in patients with acute liver failure and were uniformly distributed at concentrations below those observed in plasma. Conclusions: The uniform distribution of quinolinic acid at subplasma concentrations in the brains of patients with acute liver failure suggests that it is synthesized peripherally and enters the brain across a permeabilized blood-brain barrier. Whereas the elevation of brain quinolinic acid levels in patients who died of acute but not chronic liver failure suggests that the involvement of quinolinic acid in the pathogenesis of hepatic encephalopathy is minimal, it could predispose these patients to seizures.
he tryptophan metabolite quinolinic acid is an endogenous neuronal excitant acting at the N-methylD-aspartate receptor that can cause convulsions and neuronal degeneration. 1 Because the plasma and brain concentrations of tryptophan are elevated in acute and chronic liver failure, 2'~ increased rates of quinolinic acid synthesis in the central nervous system (CNS) and periphery might result from the increased availability of tryptophan in these regions. Indeed, increased quinolinic acid levels have been reported in the brains of hyperammonemic and portacaval shunted rats 4 and in the cerebrospinal fluid and frontal cortex of humans who died of
chronic liver failure) This led to the proposal that quinolinic acid may play a role in the pathogenesis of hepatic encephalopathy (HE), a neuropsychiatric disorder that often accompanies acute or chronic hepatic failure. A role for quinolinic acid in the pathogenesis of HE is particularly intriguing because the neuroexcitatory, convulsant, and neurotoxic characteristics of this agent contrast with the general manifestations of HE. Specifically, patients with HE show evidence of depressed neuronal activity manifested as lethargy, decreased cognitive function, and electroencephalographic slowing. 6.s Seizures are relatively rare, and neuronal degeneration is not a significant characteristic of this reversible, metabolic encephalopathy. 8'9 Moreover, because the mechanisms underlying the pathogenesis of HE caused by acute or chronic liver failure are not entirely the same, the involvement of quinolinic acid in patients with HE associated with chronic liver failure may not be relevant to patients with HE caused by acute or fulminant hepatic failure. In view of these incongruities, we have measured quinolinic acid levels in the plasma and selected brain regions of patients with either acute or chronic liver failure at several stages of HE. This may provide an insight into the role that quinolinic acid may play in the pathogenesis of acute episodes of HE associated with either acute or chronic liver failure.
Materials and Methods
T
Patient Characteristics The clinical characteristics of the 28 patients with acute liver failure, 30 patients with chronic liver failure, and 18 normal subjects are presented in Table 1. The manifestations of HE appeared acutely in the patients with chronic liver failure as well as the acute liver failure group. One of the patients with an acute presentation of Wilson's disease was included Abbreviations used in this paper: HE, hepatic encephalopathy. This is a U.S. governmentwork, There are no restrictionson its
use, 0016-5085/95/$0,00
March 1995
QUINOLINIC ACID AND HEPATIC ENCEPHALOPATHY 819
Table 1. Clinical Data on Patients With HE and Control Subjects Sex (M/F)
Age (yr)
Acute liver failure
16/12
29 _+ 2
Chronic liver failure
20/12
53 + 2
9/9
46 + 4
Group
Control subjects
Cause of death Acetaminophen overdose (19) Other drug overdose (2) Hepatitis (5) Wilson's disease (1) Alcoholic cirrhosis (11) Primary biliary cirrhosis (10) Cryptogenic cirrhosis (4) Hepatic carcinomatosis (2) Hepatitis (3) Wilson's disease (1) Ischemic heart failure (14) Motoneuron disease (1) Rheumatic heart disease (1) Lung cancer (2)
Prothrombin time
(seconds)
85 _+ 9
41 + 7
NOTE. The data represent the mean _+ SE.
in the acute liver failure group because the clinical course of the condition was identical to that of patients with acute liver failure rather than that of a patient with chronic liver failure. The primary cause of death for patients in the acute liver failure category was acetaminophen overdose (68%), whereas most patients with chronic liver failure had alcoholic cirrhosis (37%) or primary biliary cirrhosis (33%). The principal cause of death among the controls was ischemic heart failure (78%). Of the patients with hepatitis, 3 of 5 patients with acute liver failure had non-A, non-B hepatitis; 1 of 5 had ischemic hepatitis; and 1 of 5 had drug-induced hepatitis, whereas all 3 of the patients with chronic liver failure had hepatitis B. Some of the patients received ranitidine, cefuroxime, vitamin K, and parentrovite (containing 500 mg ascorbic acid, 900 mg glucose, 160 mg nicotinamide, 50 mg pyridoxine, 4 mg riboflavine, and 250 mg thiamine HC1 per 10 mL) as routine therapy. Furthermore, 9 patients received noradrenaline or adrenaline for hypotension. The severity of HE in each of these patients was graded according to the general criteria previously described, lo Patient plasma samples from only one stage of HE were included in subsequent analyses. There was no significant difference between the groups in the numbers of men and women studied. The age of the patients in the acute liver failure group was significantly less than either the control or chronic liver failure groups (P < 0.01; analysis of variance). However, the ages of the patients in the control and chronic liver failure groups were not significantly different from each other. Not all patients were sampled for both brain and plasma.
Quantification of Quinolinic Acid Plasma samples ( 3 - 1 5 mL) were taken from patients at various times during their hospital care in the United Kingdom with the first sample taken within 24 hours after admission. Brain samples were taken an average of 20 _+ 2 hours after death according to previously described techniques. 11 The samples were stored at - 8 0 ° C for 3 - 4 months and then shipped on dry ice for analysis in the United States, where they were analyzed within 1 month.
Aliquots of the plasma samples (20 btL) were diluted with 1 mL of 1 mol/L HC1, whereas brain samples (50 mg) were homogenized by sonication with 1 mL of 1 mol/L HC1. Both preparations were then centrifuged at 12,000g for 20 minutes at 4°C, and aliquots of the supernatant (300 }.tL of plasma and 800 btL of brain) were retained. Quinolinic acid standards were dissolved in equivalent volumes of deionized water. [iSO]Quinolinic acid (2 ng) was added as an internal standard to each tube. All tissue extracts and corresponding quinolinic acid standards were washed with 3 mL of chloroform, and equal volumes of the aqueous layer were collected. The samples and standards were then freeze-dried overnight. Quinolinic acid and [lSO]quinolinic acid were derivatized to their dihexafluoroisopropanol esters with complete retention of the isotopes, washed with 250 [.tL of water, extracted into 3 0 0 - 1 0 0 0 btL of heptane, and then quantified using a Hewlett-Packard 5988 mass spectrometer (San Fernando, CA) operated in the electron capture negative chemical ionization mode using methane as the reagent gas. Aliquots of each sample were injected into a 1 m X 0.53 m m (inside diameter) fused silica precolumn (on-column injection at 80°C) that had been sealed to a 15 m X 0.25 m m (inner diameter) DB5 analytical column (J & W Scientific, Folsum, CA). Column temperature was 112°C, and helium was the carrier gas. The molecular anions of quinolinic acid (m/z 467) and the [180]quinolinic acid internal standard (mass/charge = 471) were monitored and each peak area quantified at the appropriate retention time. All quantification was performed at a signal/noise ratio of >--20:1. The minimum detectability was 300 fg at a signal/noise ratio of -----5:1.
Statistics Intergroup comparisons were made using analysis of variance followed by post hoc multiple comparison analysis with Fisher's least significant difference test (Systat, Evanston, IL). Correlations were determined using the Pearson correlation matrix with Bonferroni-adjusted probabilities.
820
0 4J
BASILE ET AL.
GASTROENTEROLOGY Vol. 108, No. 3
tients with chronic liver failure were not significantly increased above control levels until stages II and III of HE, where the maximum concentrations reached approximately 5 btmol/L (Figure 1 and Table 1). Overall, plasma samples from 13 of 23 patients with chronic liver failure had quinolinic acid levels within the range of normal values. Analyzed by themselves, a strong correlation was found between the plasma concentration of quinolinic acid and the stage of HE (Pearson correlation coefficient = 0.75; P < 0.01) in patients with chronic liver failure. When the plasma quinolinic acid concentrations from both patients with acute and chronic liver failure were pooled, there was a moderate but significant correlation between the stage of HE and the quinolinic acid concentration (Pearson correlation coefficient = 0.33; P
18-
L.
4-1 c(J to Or', U--I
-O,,J < (J co ¢-
16 14
12lo-
8 6-
o
o A
0
O
AA
:
2
B
Ct:rl
O
oo , , , t a ~
o
i Stage
O
~o
o
A A
0
AA
O
4
o_
CY
8
A
^o
z'z of
ziz
ooo
i~,
HE
Figure 1. Distribution of plasma quinolinic acid levels in controls (F1)
and patients with acute (©) or chronic (A) liver failure. Patients were sampled only once at any stage of encephalopathy. Approximately 35% of the patients with acute liver failure and 57% of the patients with chronic liver failure had quinolinic acid levels that fell within the control range, regardless of the stage of HE.
=
0.01).
The quinolinic acid concentrations in samples of cerebral cortex, caudate nucleus, or cerebellum from patients who died from chronic liver failure in stage IV HE were not significantly different from controls (Figure 2 and Table 3). However, mean quinolinic acid concentrations in patients who died from acute liver failure were elevated by approximately 1 0 0 % - 4 0 0 % above control levels in the cerebral cortex, caudate nucleus, and cerebellum (F(8,795 = 3.736; P < 0.01, P < 0.05, and P < 0.05, respectively). There was no significant difference in quinolinic acid concentrations between the brain regions from patients with acute liver failure (P = 0.06, caudate nucleus vs. cortex; P = 0.19 caudate nucleus vs. cerebellum). Approximately 35% of the patients with acute liver failure had brain quinolinic acid concentrations that fell within normal levels.
Results
Approximately 35% of the patients with acute liver failure had plasma quinolinic acid levels within the control range (
