- Email: [email protected]
Brain γ-aminobutyric acid and benzodiazepine receptor binding in dialysis encephalopathy
Neuroscience Letters, 58 (1985) 241 244
241
Elsevier Scientific Publishers Ireland Ltd.
NSL 03419
BRAIN 7-AMINOBUTYRIC ACID AND BENZODIAZEPINE RECEPTOR BINDING IN DIALYSIS ENCEPHALOPATHY
S T E P H E N J. KISH l'*, T H O M A S L. P E R R Y 2, V I N C E N T P. S W E E N E Y ~ and OLEH H O R N Y K I E W I C Z t,4
*Human Brain Laboratory, Clarke Institute of Psychiatry, Toronto, Ont.. Departments of 2pharmaeology and Therapeutics, and iMedieine, University of British Columbia, Vancouver, B.C. (Canada); and 4UniversiO' ~[" Vienna, Vienna (Austria) (Received March 29th, 1985; Accepted April 23rd, 1985)
Key words." benzodiazepine - benzodiazepine receptor - dialysis encephalopathy - diazepam - y-aminobutyric acid
We measured y-aminobutyric acid (GABA) and benzodiazepine binding in autopsied frontal cortex of 8 patients dying with dialysis encephalopathy (DE). No alteration in [3H]GABA binding was observed. However, a mild reduction (-23~o, P < 0.05) of [3H]flunitrazepam-binding density was found in DE cortex. The magnitude of this reduction was similar to that observed in frontal cortex of amygdala-kindled rats [10]. We suggest that a reduction in benzodiazepine receptor number, in combination with markedly reduced GABA concentration in DE cerebral cortex may contribute to some of the clinical features (especially seizures) characteristically observed in this syndrome.
Dialysis encephalopathy (DE) is a progressive and fatal disorder occurring in patients receiving chronic hemodialysis. The clinical symptoms of this syndrome typically include a speech abnormality, electroencephalographic changes with spikes and sharp waves, seizures (usually generalized), progressive global dementia and psychotic episodes [1, 6, 9]. Recently, we observed a marked deficiency of ~,-aminobutyric acid (GABA) concentration in cerebral cortex, caudate and thalamus in 5 cases [14]. We have now extended our original investigation to include measurement by radioligand binding technique of two proposed subunits of the GABA receptor complex, namely the GABA receptor recognition site and the benzodiazepine receptor (cf. ref. 11). This investigation was prompted by several lines of evidence: (1) changes in GABA receptor number or affinity might be expected as a consequence of the markedly reduced GABA concentration in DE brain (implying altered GABAergic neuronal activity) observed in our original study; (2) much evidence implicates alterations in GABA/benzodiazepine receptor function in seizure disorders (for extensive review see ref. 12). Thus, it is conceivable that the seizures and electroencephalographic changes (suggestive of epileptic activity [9]) which are present in the majority of DE patients [6] could be related to abnormal GABA receptor sensitivity. Finally, *Author for correspondence at: H u m a n Brain Laboratory, Clarke Institute of Psychiatry, 250 College Street, Toronto, Ont. M 5 T IR8, Canada. 0304-3940/85/$ 03.30 © 1985 Elsevier Scientific Publishers Ireland Ltd.
242 (3) altered GABA-benzodiazepine receptor mechanisms might be involved in view of the observations that the benzodiazepine diazepam reverses the language disorder of DE when administered early in the course of the disorder [9, 14]. In our study we chose to examine biochemically the frontal cortex, since in DE the electroencephalographic changes show dominant presentation in the frontal lobes [2]. In addition, altered (reduced) benzodiazepine binding has been reported in this brain region in an experimental animal model of epilepsy (kindling model [10]). Autopsied brain was obtained from 8 adults (51 + 5 years, m e a n + S . E . M . ) who died with the classical symptoms of DE (including characteristic speech impairment, electroencephalogram (EEG) changes, and generalized seizures) following chronic hemodialysis treatment for kidney failure, and from 10 adults (54_+6 years) dying without evidence of neurological disease. Drug information could be obtained on 7 of the 8 DE patients. Of these 7 patients, 6 had been treated with benzodiazepines (diazepam or triazolam) for varying periods (days to months) of time. Brains were bisected sagittally and one-half was frozen at 7 0 C for neurochemical analysis, The mean intervals between death and freezing of the brain tissue did not differ significantly (P>0.05: DE, 11 + 6 h; controls, 15_+3 h). G A B A receptor binding in extensively washed brain homogenates was determined at a single concentration of [3H]GABA (25 nM) as described previously [4]. Centraltype benzodiazepine binding in unwashed brain homogenates was performed using varying concentrations of [3H]flunitrazepam (0.3 12 riM) and 1 /zM clonazepam to determine non-specific binding [5]. A single concentration of [SH]Ro 5-4864 (2 nM) was employed to determine 'peripheral'-type benzodiazepine binding to unwashed brain homogenates [5]. Protein content was measured using the procedure of Lowry et al. [8]. Statistical analysis of the data was made using the Student's two-tailed ttest. TABLE I (;ABA AND BENZODIAZEPINEBINDING IN HUMAN FRONTAL CORTEX Values represent mean_+S.E.M. GABA binding in extensively washed tissue was measured at a concentration of 25 nM. Ro 5-4864binding in unwashed tissue was measured at 2 nM. [3H]flunitrazepam-binding density (Bin,0 and affinity (Kj) (in unwashed tissue) was determined by Scatchard analysis. *P < 0.05, Student's two-tailed t-test.
[3H]GABAbinding (fmol/mgprotein) [~H]Ro 5-4864binding (fmol/mgprotein) [3H]flunitrazepam-bindingdensity (B..... fmol/mgprotein) [~H]flunitrazepam-bindingaffinity (Ka, nM)
Controls
Dialysis encephalopathy
525_+34 0o) 19.4+_1.I (8) 771 ± 29 (8) 1.67-+0.17 (8)
526_+54 (8) 21.6_+3.4 (8) 596 _+77* (8) 2.72+0.77 (8)
243 Mean [3H]GABA and [3H]Ro 5-4864 binding were not significantly different in DE frontal cortex as compared with the controls (Table I, P>0.05). However, the [3H]flunitrazepam-binding density was slightly but significantly reduced by 23~o (P < 0.05). In the DE group, [3H]flunitrazepam-binding affinity (Ka) was (nonsignificantly) elevated by 63~. An analysis of the individual Kd values revealed that the elevation was due to abnormally high Kd values in 3 patients (3.76, 6.18 and 5.13), two of which were known to have recently received benzodiazepines within 2 days of death. The results of our study demonstrate no significant alteration in either [3H]GABA or [3H]Ro 5-4864 binding in frontal cortex of patients dying with DE. In contrast, in the DE group we observed a mild but statistically significant decrease in cortical [3H]flunitrazepam binding density. The lack of change in [3H]GABA binding in DE cerebral cortex is in contradistinction to observations in another seizure disorder, namely temporal lobe epilsepsy, in which reduced binding has been reported in biopsied (mainly temporal lobe) epileptogenic foci [7]. Our finding of normal [3H]GABA binding in the presence of markedly reduced GABA concentration in DE frontal cortex [14] additionally suggests that there exist no marked compensatory changes in cortical GABA receptor number in this disorder. Benzodiazepine binding sites in brain can be classified into two types: the central neuronal site (preferentially labelled by [3H]flunitrazepam), which appears to be relevant to the pharmacological actions of the benzodiazepines, and the peripheral-type binding site (preferentially labelled by [3H]Ro 5-4864), which appears to be mainly localized, in the central nervous system, to glial cells (cf. ref. 13). The results of our study indicate a reduction in number of the central type of benzodiazepine receptor in DE. Our finding of a (non-significant) elevation in receptor affinity (Kd) is most likely explained by the presence, in brain of 3 of the DE patients, of residual amounts of a benzodiazepine compound (or metabolites). The mechanism involved in reduced cortical benzodiazepine binding in DE is uncertain, Prolonged administration of benzodiazepines (which most of the DE patients had been receiving) does not seem to be involved, as indicated by normal benzodiazepine binding density observed in cerebral cortex of Lesch-Nyhan patients who had received chronic benzodiazepine treatment [3]. On the other hand, our present demonstration of reduced benzodiazepine binding, taken together with our earlier finding of markedly reduced GABA concentration in DE cerebral cortex, could be explained by a loss of neurons on which benzodiazepine receptors were localized. Although neuropathological analysis revealed no significant neuronal loss in frontal cortex of our DE patients [14], it is conceivable that these receptors could have been localized on endings of neurons originating outside of the cortex. Finally, the possible relationship between the reduced benzodiazepine receptor density and the seizures present in the DE patients has to be considered. In this regard, reduced benzodiazepine receptor density (in unwashed but not washed brain homogenates) of similar magnitude ( - 25~o) as in the DE material ( - 23~o) has been observed in rat frontal cortex in the kindling model of epilepsy [10]. With respect
244 to functional considerations, much circumstantial evidence has linked altered (reduced) G A B A - b e n z o d i a z e p i n e brain m e c h a n i s m s with seizure c o n d i t i o n s (cf. ref. 12 for comprehensive review). O u r data are consistent with these experimental observations a n d suggest that an i m p a i r m e n t of G A B A - m e d i a t e d n e u r o t r a n s m i s s i o m as evidenced by reduced G A B A c o n c e n t r a t i o n and benzodiazepine binding, m a y be involved in the generation of seizures observed in DE. In addition, our finding of reduced cortical benzodiazepine b i n d i n g density might help to explain the clinical observation that the efficacy of benzodiazepines in D E (especially with respect to the language a b n o r m a l i t y ; V. Sweeney, u n p u b l i s h e d observations) markedly declines with the progression of the disorder. This study was supported by the Clarke Institute of Psychiatry and by the Medical Research Council of C a n a d a (T.L.P.). S.J.K. is a Career Scientist of the O n t a r i o Ministry of Health. Some control b r a i n material was received from Drs. E m s o n and Nyssen, D e p a r t m e n t of Pathology, University Hospital, Saskatoon, a n d the C a n a dian Brain Tissue Bank. 1 Alfrey, A.C., Dialysis encephalopathy syndrome, Annu. Rev. Med., 29 (1978) 93-98. 2 ttughes, J.R. and Schreeder, M.T., EEG in dialysis encephalopathy, Neurology, 30 (1980) 1148 1154. 3 Kish, S.J., Fox, I.H, Kapur, B.M., Lloyd, K. and Hornykiewicz, O., Brain benzodiazepine receptor binding and purine concentration in Lesch-Nyhan syndrome, Brain Res., in press. 4 Kish, S.J., Perry, T.L. and Hornykiewicz,O., Increased GABA receptor binding in dominantly inherited cerebellar ataxias, Brain Res., 269 (1983) 370-373. 5 Kish, S.J., Sperk, G. and Hornykiewicz, O., Alterations in benzodiazepine and GABA receptor binding in rat brain followingsystemic injection of kainic acid, Neuropharmacology, 22 (1983) 1303-1309. 6 Lederman, R.J. and Henry, C.E., Progressivedialysisencephalopathy, Ann. Neurol., 4 (1978) 199 204. 7 Lloyd, K.G., Munari, C., Bossi, L. and Morselli, P.L., Status of GABAergic neurones in human epileptic foci, as defined by neurochemistry, J. Neurochem., 41, Suppl. (1983) SI01C. 8 Lowry, O.H., Rosebrough, N.J., Farr, A.L. and Randall, R.J., Protein measurement with the Folin phenol reagent, J. Biol. Chem., 193 (1951) 265-275. 9 Nadel, A.M. and Wilson, W.P., Dialysis encephalopathy: a possible seizure disorder, Neurology, 26 (1976) 1130 1134. 10 Niznik, H.B., Burnham, W.M. and Kish, S.J., Benzodiazepinereceptor binding following amygdalakindled convulsions: differing results in washed and unwashed brain membranes, J. Neurochem., 43 {1984) 1732 1736. 11 Otsen, R.W., GABA. benzodiazepine-barbiturate receptor interactions, J. Neurochem., 37 (1981) 1 13. 12 Olsen, R.W., Snowman, A.M., Lee, R., Lomax, P. and Wamsley, J.K., Role of the 7-aminobutyric acid receptor-ionophore complex in seizure disorders, Ann. Neurol., 16, Suppl. (1984) $90-$97. 13 Richards, J.G. and Mohler, H., Benzodiazepinereceptors, Neuropharmacology, 23 (1984) 233-242. 14 Sweeney,V.P., Perry, T.L., Price, J.D.E., Reeve, C.E., Godolphin, W.J. and Kish, S.J., Brain ?-aminobutyric acid deficiencyin dialysis encephalopathy, Neurology, 35 (1985) 180-184.
241
Elsevier Scientific Publishers Ireland Ltd.
NSL 03419
BRAIN 7-AMINOBUTYRIC ACID AND BENZODIAZEPINE RECEPTOR BINDING IN DIALYSIS ENCEPHALOPATHY
S T E P H E N J. KISH l'*, T H O M A S L. P E R R Y 2, V I N C E N T P. S W E E N E Y ~ and OLEH H O R N Y K I E W I C Z t,4
*Human Brain Laboratory, Clarke Institute of Psychiatry, Toronto, Ont.. Departments of 2pharmaeology and Therapeutics, and iMedieine, University of British Columbia, Vancouver, B.C. (Canada); and 4UniversiO' ~[" Vienna, Vienna (Austria) (Received March 29th, 1985; Accepted April 23rd, 1985)
Key words." benzodiazepine - benzodiazepine receptor - dialysis encephalopathy - diazepam - y-aminobutyric acid
We measured y-aminobutyric acid (GABA) and benzodiazepine binding in autopsied frontal cortex of 8 patients dying with dialysis encephalopathy (DE). No alteration in [3H]GABA binding was observed. However, a mild reduction (-23~o, P < 0.05) of [3H]flunitrazepam-binding density was found in DE cortex. The magnitude of this reduction was similar to that observed in frontal cortex of amygdala-kindled rats [10]. We suggest that a reduction in benzodiazepine receptor number, in combination with markedly reduced GABA concentration in DE cerebral cortex may contribute to some of the clinical features (especially seizures) characteristically observed in this syndrome.
Dialysis encephalopathy (DE) is a progressive and fatal disorder occurring in patients receiving chronic hemodialysis. The clinical symptoms of this syndrome typically include a speech abnormality, electroencephalographic changes with spikes and sharp waves, seizures (usually generalized), progressive global dementia and psychotic episodes [1, 6, 9]. Recently, we observed a marked deficiency of ~,-aminobutyric acid (GABA) concentration in cerebral cortex, caudate and thalamus in 5 cases [14]. We have now extended our original investigation to include measurement by radioligand binding technique of two proposed subunits of the GABA receptor complex, namely the GABA receptor recognition site and the benzodiazepine receptor (cf. ref. 11). This investigation was prompted by several lines of evidence: (1) changes in GABA receptor number or affinity might be expected as a consequence of the markedly reduced GABA concentration in DE brain (implying altered GABAergic neuronal activity) observed in our original study; (2) much evidence implicates alterations in GABA/benzodiazepine receptor function in seizure disorders (for extensive review see ref. 12). Thus, it is conceivable that the seizures and electroencephalographic changes (suggestive of epileptic activity [9]) which are present in the majority of DE patients [6] could be related to abnormal GABA receptor sensitivity. Finally, *Author for correspondence at: H u m a n Brain Laboratory, Clarke Institute of Psychiatry, 250 College Street, Toronto, Ont. M 5 T IR8, Canada. 0304-3940/85/$ 03.30 © 1985 Elsevier Scientific Publishers Ireland Ltd.
242 (3) altered GABA-benzodiazepine receptor mechanisms might be involved in view of the observations that the benzodiazepine diazepam reverses the language disorder of DE when administered early in the course of the disorder [9, 14]. In our study we chose to examine biochemically the frontal cortex, since in DE the electroencephalographic changes show dominant presentation in the frontal lobes [2]. In addition, altered (reduced) benzodiazepine binding has been reported in this brain region in an experimental animal model of epilepsy (kindling model [10]). Autopsied brain was obtained from 8 adults (51 + 5 years, m e a n + S . E . M . ) who died with the classical symptoms of DE (including characteristic speech impairment, electroencephalogram (EEG) changes, and generalized seizures) following chronic hemodialysis treatment for kidney failure, and from 10 adults (54_+6 years) dying without evidence of neurological disease. Drug information could be obtained on 7 of the 8 DE patients. Of these 7 patients, 6 had been treated with benzodiazepines (diazepam or triazolam) for varying periods (days to months) of time. Brains were bisected sagittally and one-half was frozen at 7 0 C for neurochemical analysis, The mean intervals between death and freezing of the brain tissue did not differ significantly (P>0.05: DE, 11 + 6 h; controls, 15_+3 h). G A B A receptor binding in extensively washed brain homogenates was determined at a single concentration of [3H]GABA (25 nM) as described previously [4]. Centraltype benzodiazepine binding in unwashed brain homogenates was performed using varying concentrations of [3H]flunitrazepam (0.3 12 riM) and 1 /zM clonazepam to determine non-specific binding [5]. A single concentration of [SH]Ro 5-4864 (2 nM) was employed to determine 'peripheral'-type benzodiazepine binding to unwashed brain homogenates [5]. Protein content was measured using the procedure of Lowry et al. [8]. Statistical analysis of the data was made using the Student's two-tailed ttest. TABLE I (;ABA AND BENZODIAZEPINEBINDING IN HUMAN FRONTAL CORTEX Values represent mean_+S.E.M. GABA binding in extensively washed tissue was measured at a concentration of 25 nM. Ro 5-4864binding in unwashed tissue was measured at 2 nM. [3H]flunitrazepam-binding density (Bin,0 and affinity (Kj) (in unwashed tissue) was determined by Scatchard analysis. *P < 0.05, Student's two-tailed t-test.
[3H]GABAbinding (fmol/mgprotein) [~H]Ro 5-4864binding (fmol/mgprotein) [3H]flunitrazepam-bindingdensity (B..... fmol/mgprotein) [~H]flunitrazepam-bindingaffinity (Ka, nM)
Controls
Dialysis encephalopathy
525_+34 0o) 19.4+_1.I (8) 771 ± 29 (8) 1.67-+0.17 (8)
526_+54 (8) 21.6_+3.4 (8) 596 _+77* (8) 2.72+0.77 (8)
243 Mean [3H]GABA and [3H]Ro 5-4864 binding were not significantly different in DE frontal cortex as compared with the controls (Table I, P>0.05). However, the [3H]flunitrazepam-binding density was slightly but significantly reduced by 23~o (P < 0.05). In the DE group, [3H]flunitrazepam-binding affinity (Ka) was (nonsignificantly) elevated by 63~. An analysis of the individual Kd values revealed that the elevation was due to abnormally high Kd values in 3 patients (3.76, 6.18 and 5.13), two of which were known to have recently received benzodiazepines within 2 days of death. The results of our study demonstrate no significant alteration in either [3H]GABA or [3H]Ro 5-4864 binding in frontal cortex of patients dying with DE. In contrast, in the DE group we observed a mild but statistically significant decrease in cortical [3H]flunitrazepam binding density. The lack of change in [3H]GABA binding in DE cerebral cortex is in contradistinction to observations in another seizure disorder, namely temporal lobe epilsepsy, in which reduced binding has been reported in biopsied (mainly temporal lobe) epileptogenic foci [7]. Our finding of normal [3H]GABA binding in the presence of markedly reduced GABA concentration in DE frontal cortex [14] additionally suggests that there exist no marked compensatory changes in cortical GABA receptor number in this disorder. Benzodiazepine binding sites in brain can be classified into two types: the central neuronal site (preferentially labelled by [3H]flunitrazepam), which appears to be relevant to the pharmacological actions of the benzodiazepines, and the peripheral-type binding site (preferentially labelled by [3H]Ro 5-4864), which appears to be mainly localized, in the central nervous system, to glial cells (cf. ref. 13). The results of our study indicate a reduction in number of the central type of benzodiazepine receptor in DE. Our finding of a (non-significant) elevation in receptor affinity (Kd) is most likely explained by the presence, in brain of 3 of the DE patients, of residual amounts of a benzodiazepine compound (or metabolites). The mechanism involved in reduced cortical benzodiazepine binding in DE is uncertain, Prolonged administration of benzodiazepines (which most of the DE patients had been receiving) does not seem to be involved, as indicated by normal benzodiazepine binding density observed in cerebral cortex of Lesch-Nyhan patients who had received chronic benzodiazepine treatment [3]. On the other hand, our present demonstration of reduced benzodiazepine binding, taken together with our earlier finding of markedly reduced GABA concentration in DE cerebral cortex, could be explained by a loss of neurons on which benzodiazepine receptors were localized. Although neuropathological analysis revealed no significant neuronal loss in frontal cortex of our DE patients [14], it is conceivable that these receptors could have been localized on endings of neurons originating outside of the cortex. Finally, the possible relationship between the reduced benzodiazepine receptor density and the seizures present in the DE patients has to be considered. In this regard, reduced benzodiazepine receptor density (in unwashed but not washed brain homogenates) of similar magnitude ( - 25~o) as in the DE material ( - 23~o) has been observed in rat frontal cortex in the kindling model of epilepsy [10]. With respect
244 to functional considerations, much circumstantial evidence has linked altered (reduced) G A B A - b e n z o d i a z e p i n e brain m e c h a n i s m s with seizure c o n d i t i o n s (cf. ref. 12 for comprehensive review). O u r data are consistent with these experimental observations a n d suggest that an i m p a i r m e n t of G A B A - m e d i a t e d n e u r o t r a n s m i s s i o m as evidenced by reduced G A B A c o n c e n t r a t i o n and benzodiazepine binding, m a y be involved in the generation of seizures observed in DE. In addition, our finding of reduced cortical benzodiazepine b i n d i n g density might help to explain the clinical observation that the efficacy of benzodiazepines in D E (especially with respect to the language a b n o r m a l i t y ; V. Sweeney, u n p u b l i s h e d observations) markedly declines with the progression of the disorder. This study was supported by the Clarke Institute of Psychiatry and by the Medical Research Council of C a n a d a (T.L.P.). S.J.K. is a Career Scientist of the O n t a r i o Ministry of Health. Some control b r a i n material was received from Drs. E m s o n and Nyssen, D e p a r t m e n t of Pathology, University Hospital, Saskatoon, a n d the C a n a dian Brain Tissue Bank. 1 Alfrey, A.C., Dialysis encephalopathy syndrome, Annu. Rev. Med., 29 (1978) 93-98. 2 ttughes, J.R. and Schreeder, M.T., EEG in dialysis encephalopathy, Neurology, 30 (1980) 1148 1154. 3 Kish, S.J., Fox, I.H, Kapur, B.M., Lloyd, K. and Hornykiewicz, O., Brain benzodiazepine receptor binding and purine concentration in Lesch-Nyhan syndrome, Brain Res., in press. 4 Kish, S.J., Perry, T.L. and Hornykiewicz,O., Increased GABA receptor binding in dominantly inherited cerebellar ataxias, Brain Res., 269 (1983) 370-373. 5 Kish, S.J., Sperk, G. and Hornykiewicz, O., Alterations in benzodiazepine and GABA receptor binding in rat brain followingsystemic injection of kainic acid, Neuropharmacology, 22 (1983) 1303-1309. 6 Lederman, R.J. and Henry, C.E., Progressivedialysisencephalopathy, Ann. Neurol., 4 (1978) 199 204. 7 Lloyd, K.G., Munari, C., Bossi, L. and Morselli, P.L., Status of GABAergic neurones in human epileptic foci, as defined by neurochemistry, J. Neurochem., 41, Suppl. (1983) SI01C. 8 Lowry, O.H., Rosebrough, N.J., Farr, A.L. and Randall, R.J., Protein measurement with the Folin phenol reagent, J. Biol. Chem., 193 (1951) 265-275. 9 Nadel, A.M. and Wilson, W.P., Dialysis encephalopathy: a possible seizure disorder, Neurology, 26 (1976) 1130 1134. 10 Niznik, H.B., Burnham, W.M. and Kish, S.J., Benzodiazepinereceptor binding following amygdalakindled convulsions: differing results in washed and unwashed brain membranes, J. Neurochem., 43 {1984) 1732 1736. 11 Otsen, R.W., GABA. benzodiazepine-barbiturate receptor interactions, J. Neurochem., 37 (1981) 1 13. 12 Olsen, R.W., Snowman, A.M., Lee, R., Lomax, P. and Wamsley, J.K., Role of the 7-aminobutyric acid receptor-ionophore complex in seizure disorders, Ann. Neurol., 16, Suppl. (1984) $90-$97. 13 Richards, J.G. and Mohler, H., Benzodiazepinereceptors, Neuropharmacology, 23 (1984) 233-242. 14 Sweeney,V.P., Perry, T.L., Price, J.D.E., Reeve, C.E., Godolphin, W.J. and Kish, S.J., Brain ?-aminobutyric acid deficiencyin dialysis encephalopathy, Neurology, 35 (1985) 180-184.