Redox state, antioxidative activity and lipid peroxidation in erythrocytes and plasma of chronic ambulatory peritoneal dialysis patients

ELSEVIER

Clinica Chimica Acta 234 (1995) 127-136

Redox state, antioxidative activity and lipid peroxidation in erythrocytes and plasma of chronic ambulatory peritoneal dialysis patients F. Canestrari *a, U. Buoncristiani b, F. Galli a, A. Giorgini a, M.C. Albertini a, C. Carobi b, M. Pascucci c, M. Bossl~ a alnstitute of Biochemistry 'G. Fornaini', University of Urbino, Urbino, Italy bNephrology and Dialysis Unit, Hospital 'R. Silvestrini', Perugia, Italy CDialysis Unit, Urbino Hospital, Urbino, Italy Received 9 March 1994; revision received 29 September 1994; accepted 31 October 1994

Abstract

Red blood ceils and plasma reduced and oxidized glutathione levels, glutathione peroxidase (GSH-Px) activity, thiobarbituric acid reactants (TBAR) of both chronic ambulatory peritoneal dialysis (CAPD) patients and a matched control group were investigated in this study. Oxidized and reduced pyridinic nucleotides in red blood cells (RBC), in which NADPH is a direct expression of hexose monophosphate shunt function, were also studied. The results obtained indicate that RBC and plasma are exposed to oxidative stress in CAPD. This condition is characterized by a decreased GSH/GSSG ratio, particularly evident in RBC as a consequence of the GSSG accumulation. Lipid peroxidation is increased, as indicated by raised TBAR levels, and reduced pyridinic nucleotides are decreased. Increased GSH-Px levels and unmodified or slightly increased GSH content were observed in the RBC but not in plasma, which showed decreased GSH and unmodified peroxidase activity. Peroxidase correlated positively with TBAR levels in the RBC lysates. In a subgroup of patients treated with erythropoietin (vs. untreated patients and controls) no differences were observed in the glutathione-related parameters studied. These data suggest that a mechanism for adaptation to oxidative conditions may be present in CAPD and its effects on RBC integrity are discussed in comparison with the hemodialysis conditions previously studied. Keywords: Uremic anemia; Chronic ambulatory peritoneal dialysis (CAPD); Reduced glutathione (GSH); Oxidized glutathione (GSSG); Lipid peroxidation; Glutathione peroxidase; Hexose monophosphate shunt * Corresponding author, Institute of Biological Chemistry 'G. Fornaini', University of Urbino, Via Saffi 2, 61029 Urbino, Italy. Tel: +39-722-305260; Fax: +39-722-320188. 0009-8981/95/$09.50 © 1995 Elsevier Science B.V. All rights reserved SSDI 0009-8981(94)05990-A

128

F Canestrari et al. / Clinica Chimica Acta 234 (1995) 127-136

1. Introduction

Oxygen free radicals (OFR) are a critical factor in the anemia caused by chronic renal failure (CRF) [1,21. In this context, reduced red blood cell (RBC) life-span has been observed to be a consequence of oxidative stress, particularly in hemodialysis (HD) patients [2,3]. In fact, HD has been shown to induce acute OFR production by neutrophil activation [4,51 and increased RBC lipid peroxidation and redox modifications have been reported [6,7]. In order to better understand the consequences that dialysis can have on uremic blood components, extensive studies have been carried out on OFR activity under different dialysis conditions [81. Chronic ambulatorial peritoneal dialysis (CAPD) appears to be more physiological than HD, but its effects on RBC and plasma redox are not well understood [9]. A comparison of the two dialysis techniques is therefore of great interest in distinguishing the role played by each dialysis procedure, in RBC damage and in modification of the plasma environment. At the moment complete information is not available on plasma and the RBC redox state in CAPD. In particular, glutathione metabolism has not been studied in this context and hexose monophosphate shunt (HMPS) function has been studied only through indirect methods [10,11]. However, it is well known that glutathione and its related enzymes are crucial in the regulation of OFR activity [12]. In RBC, this compound plays a key role in the maintenance of the reduced form of hemoglobin, lipid unsaturation and protein thiols [13]. In this context, we investigated reduced and oxidized glutathione (GSH and GSSG) levels, glutathione peroxidase (GSH-Px) activity and thiobarbituric acid reactant (TBAR) production in RBC and compared the plasma of CAPD subjects and matched controls. Oxidized and reduced pyridinic nucleotides in RBC were also studied and the question of whether NADPH is a direct expression of HMPS function addressed. 2. Materials and methods

2.1. Subjects A group of 18 CAPD patients (ages 62.9 ± 13.9 years) and a control group (n = 15, ages 59.2 ± 8.3 years) were studied (see Table 1). In the CAPD subjects, the primary renal diseases present were chronic glomerulonephritis (n = 5), chronic pyelonephritis (n = 5), polycystic kidneys (n = 2), nephroangiosclerosis (n = 1) and unknown (n = 5). Dialysis patients were monitored over a period of 26.56 ± 23.27 months (range 6-76 months) using a standard procedure (8 l/day in four exchanges). Patients received neither blood nor plasma infusions. Four of them received recombinant erythropoietin (6000-10 000 units/week by two infusions). Treatment with other drugs capable of interfering with the determinations was suspended for a sufficient period preceding the study. RBC count, hematocrit and hemoglobin concentration in the CAPD group were, respectively: 3.072 ± 0.647 x 106/ram 3, 27.3 ± 8.6% and 9.84 ± 1.81 g/dl. Erythropoietin-treated patients showed Hb concentrations of 8.77 -4- 1.55 g/dl. Con-

129

F. Canestrari et al./ Clinica Chimica Acta 234 (1995) 127-136

Table 1 Hematological and clinical parameters in CAPD patients Parameter

Mean

S.D.

Control range

Age (years) Sex (M/F) Dialysis duration (months) BUN (mg/dl) Creatinine (mg/dl) Ht (%) Hb (g/dl) RBC number ( x 106) Serum folic acid (ng/ml) Total iron (/~g/dl) Erythropoietin (mU/ml) a Vit.Bl2 (pg/ml) Transferrin (#g/dl)

62.9 11/7 26.5 165.1 10.9 27.3 9.8 3.07 7.57 94.3 12.5 522.0 334.5

13.9 -23.2 32.5 1.9 8.7 1.8 0.64 5.46 20.8 9.3 212.8 52. I

27-70 --11-50 0.6-1.2 40-54 13-16 4.4-5.8 3-14 50-150 8-t5 100-800 250-450

aUntreated patients. The four CAPD subjects treated with recombinant erythropoietin showed a value of 21.7 (mU/ml).

trol group Hb was 14.90 ± 2.50 g/dl. All subjects in the CAPD group showed reticulocyte counts of 15%o or less. All subjects gave informed consent before participating in this study. 2.2. Blood Immediately before dialysis sessions, 5 ml of blood from the antecubital vein of each patient were drawn by Vacutainer technique into green-top heparinized tubes. Samples were immediately centrifuged (3000 rev./min for 10 min at 4°C) and plasma and RBC carefully removed, discarding the buffy-coat by aspiration. The RBC were then washed three times in buffered saline (0.50 mmol/1 KH~PO4, 3.45 mmol/l K2HPO4 and 150 mmol/l NaCI at pH 7.4) and filtered by the method of Beutler and co-workers [141 before utilization. 2.3. RBC and plasma GSH and GSSG determination GSH determinations on RBC lysates or plasma were made by a spectrophotometric method described by Beutler [15]. GSH formation was measured using 5,5'-dithio-bis (2-nitrobenzoic acid) (DTNB), which reacts specifically with GSH to form a highly-coloured yellow anion measured in solution at 412 nm. GSSG levels were estimated by the method of Srivastava and Beutler [16], which monitors the NADPH enzymatic oxidation carried out by glutathione reductase (GSSG-Rd) at 340 nm. Briefly, packed RBC or plasma were mixed with 0.25 mol/1 N-ethylmaleimide (NEM) at a ratio of 5:1 (v/v). This alkylating agent prevents nonenzymatic GSH oxidation. After tricloroacetic acid (TCA; 30% w/v) precipitation, excess NEM was removed by ether extraction and the pH of the sample neutralized. Final extract (900/zl) was mixed with a GSSG assay buffered mixture containing 80

130

F. Canestrari et al. /Clinica Chimica Acta 234 (1995) 127-136

#1 of I mol/l K2HPO4/KH2PO4 plus 4 mmol/l EDTA at pH 7.4 and 10 #1 12 mmol/l NADPH. After equilibration at 37°C and baseline absorbance recording, GSSG-Rd (90 IU/ml) was added and a further absorbance measurement was performed immediately. The difference in the absorbance values of different samples is directly related to differences in their GSSG concentration. GSH and GSSG concentrations were expressed in/~mol/g Hb in the case of the RBC and in #mol/g proteins in the case of the plasma and were reported as the means ± S.D. of two experiments in separate dialysis sessions. 2.4. Glutathione peroxidase assay The RBC glutathione peroxidase assay was carried out on RBC lysate obtained by adding 19 vol of ice-cold bidistilled water to 1 vol of packed RBC or, in intact plasma, using t-butyl hydroperoxide (TBHP) as substrate [17]. GSH-Px activity at 37°C was expressed for the RBC in IU/g Hb and for plasma in IU/g proteins. 2.5. RBC oxidized and reduced pyridine coenzyme assay Concentrations of RBC pyridine nucleotides, nicotinamide adenine dinucleotide (NAD+ and NADP+) and their reduced forms (NADH and NADPH) were determined by a reverse-phase HPLC method previously developed in our laboratory [181. 2.6. TBAR determination (TBA test) TBA tests were carried out on plasma and RBC lysates (1 vol. of packed RBC in 39 vol. of ice-cold bidistilled water) according to the method of Buege and Aust [19]. Appropriate standard curves were determined using malonaldehyde bis-diethylacetal (1,1,3,3-tetraethoxy-propane) as source of malonaldehyde. TBAR values are expressed in nmol/g Hb or in nmol/g proteins in the RBC and plasma, respectively. 2. 7. RBC hemoglobin and plasma protein determination Hemoglobin concentrations were determined on RBC lysates by a ferricyanidecyanide reagent method as described in Ref. [15]. Plasma protein determination was performed according to Lowry et al. [21] with bovine serum albumin as a standard. 2.8. Chemicals Enzymes, coenzymes and nucleotide standards were purchased from Sigma Chemical Co. (St. Louis, MO, USA). The standard coenzymes (NAD+, NADH, NADP+ and NADPH) were obtained from Boehringer-Mannheim (Mannheim, Germany). All other reagents were from Merck (Darmstadt, Germany), Fluka (Buchs, Switzerland) or Carlo Erba (Milano, Italy). 2.9. Statistics Statistical analyses were carried out using Student's t-tests. Values were the means ± S.D. of duplicate experiments. Correlations between variables were determined by linear regression analyses and regression line comparisons were made by t-test (P > 0.05 was not accepted).

F. Canestrari et al. / Clinica Chimica Acta 234 (1995) 127-136

131

3. Results Fig. 1 shows erythrocyte and plasma GSH, GSSG and TBAR levels in patients and controls. Erythrocyte G S H was unmodified or slightly increased (in/zmol/g Hb: 6.40 4- 1.36 in the patients vs. 4.97 4- 0.78 in the control group) while a nearly 3fold increase in GSSG was observed in the C A P D group GSSG (in/zmol/g Hb: 0.039 4- 0.014 vs. 0.012 4- 0.011; P < 0.05). Plasma GSH was significantly decreased (in/zmol/g proteins: 1.53 4- 0.69 in the C A P D group vs. 2.07 4- 0.20 in the control group; P < 0.05) and GSSG levels were slightly increased in patients (in/~mol/g proteins: 0.011 4- 0.006 vs. 0.009 4- 0.005). Increased RBC and plasma TBAR levels were found in all patients (in nmol/g Hb or in nmol/g proteins: 10.8 4- 3.8 vs. 4.8 4- 2.6with P < 0.01 in t h e R B C a n d 74.5 4- 20.5 vs. 46.6 4- 16.1 w i t h P < 0.05 in the plasma). The levels of RBC pyridine coenzymes in the C A P D group and in controls are shown in Fig. 2. Decreased levels of reduced nucleotides can be observed especially in the case of N A D P H (in nmol/ml of cells: 14 4- 5 vs. 24 4- 5 with P < 0.01 for the N A D P H and 16 4- 13 vs. 32 4- 9 with P < 0.05 for the NADH). Oxidized nucleotides were not significantly modified. Table 2 shows RBC and plasma glutathione peroxidase activity levels in the two groups. Higher erythrocyte GSH-Px levels were observed in the C A P D group (in

400

LU :::) .-I >>. 300 ._J 0 n," IZ 0 Z

200

u.I I I...-

100

RBC

PLASMA

Fig. 1. Reduced and oxidized glutathione and TBAR levels in RBC and plasma of CAPD patients. Values (means and S.D.) are expressed as % of the mean control value (dotted line). *P < 0.05 and **P < 0.01.

132

F. Canestrari et al. / Clinica Chimica A cta 234 (1995) 127-136

80

60.

en

n,"

E

40.

O

E ¢20.

NAD+

NADH

NADP+

NADPH

Fig. 2. RBC reduced and oxidized pyridine coenzymes (means and S.D.) in CAPD group and controls. *P < 0.05 and **P < 0.01.

units/g Hb: 48.7 ± 11.5 in CAPD subjects and 36.1 ± 8.5 in controls; P < 0.05). Plasma GSH-Px was similar in CAPD and controls (in units/g proteins: 5.02 ± 1.36 vs. 5.50 ± 0.79). In the RBC, a positive correlation was found between GSH-Px levels and TBARs levels (r = 0.6748 and P < 0.005; Fig. 3). A positive but statistically insignificant correlation was observed between RBC oxidized glutathione and RBC peroxidase levels (data not shown). No differences were observed in the glutathione-related parameters in the erythropoietin-treated patients with respect to untreated patients and controls. 4. Discussion

The heterogeneity of the factors which affect the RBC number in uremic patients does not permit us definitively to characterize uremic anemia. However, RBC oxidaTable 2 RBC and plasma glutathione peroxidase activity in CAPD patients and controls

RBC (IU/g Hb) Plasma (IU/g proteins) *P < 0.05 and **P < 0.01.

CAPD

Controls

48.7 ± 11.5" 5.02 ± 1.36

36.1 ± 8.5 5.50 ± 0.79

F. Canestrari et al. I Clinica Chimica Acta 234 (1995) 127-136

133

70 "

60

1-

50

n

v

_>

40.

o < 30. •

R = 0.67475 S D = 8.74336, N = 17 P = 0.00296

20

I

I

4

8

I

12

[TBAR] nmol/g Hb Fig. 3. Linear regression analysis between RBC peroxidase activity and TBAR levels in the CAPD group.

tive injury seems to play a crucial role even if erythropoietin deficiency is the main cause of this condition [1,2]. Since RBC destruction by intravascular hemolysis or by other mechanisms such as phagocytosis can be a consequence of the oxidative stress conditions both in vitro and in vivo [22], uremic RBC have been investigated in this light by several methods. Some authors have concentrated their studies on the relationship between dialysis effects and the severity of the anemia in these patients [231. Several reports have described increased RBC and plasma lipid peroxidation in HD [24,27]. In these patients, a severe involvement of RBC redox, only in part linked to an HMPS defect, was also observed [28-31]. Furthermore, we recently observed an inductive response of antioxidant glutathione-related enzymes [31]. All these data agree with the hypothesis of increased endogenous production of OFR during HD [71. The CAPD technique appears to be more physiological than HD [32] but its effects on the RBC and plasma redox are still unclear. Anemia is better controlled in CAPD [9] than in HD (possibly due to reduced OFR production). Recently, Girelli et al. [32] confirmed the observation of Taccone-Gallucci and co-

134

F. Canestrari et al. /Clinica Chimica Acta 234 (1995) 127-136

workers [11] that the RBC lipid peroxidation profile is not significantly modified in CAPD. Taylor et al. [8] have reported, on the contrary, a severe plasma peroxidative condition in these patients. Although these conflicting results can be explained by the different approaches used, they nevertheless indicate that heterogeneity in the CAPD population and in laboratory protocols can produce apparently contrasting results. It was for this reason that we decided to investigate the redox state and the degree of lipid peroxidation in the RBC and plasma of a CAPD group. Furthermore, no conclusive data on glutathione and glutathione-related enzymes in CAPD subjects have been reported in the literature. These enzymes are crucial in protecting both the cell and biological fluids from OFR insult. In particular, RBC depend on GSH for the maintenance of their redox state [12], and the HMPS pathway restores GSH levels by producing NADPH, the substrate of glutathione reductase. This pyridinic coenzyme, together with NADH, is the main source of reducing power in the RBC. In accordance with Yawata and Jacob [10], we have recently observed decreased levels of NADPH in HD patients [31]. Plasma glutathione metabolism is mainly regulated by the liver and kidney metabolism [12] and its characterization is an important parameter in blood OFR activity determination. The present study demonstrates that RBC and plasma GSH/GSSG ratios are decreased in CAPD subjects by about 50% of the control values. In RBC this redox defect seems to be a consequence of the 3-fold increase in GSSG concentrations accompanied by unmodified or slightly increased GSH and a 3-fold increase in TBAR levels. Our results also show that the peroxidative pathway is modified mainly at the cellular level, in agreement with Taccone-Gallucci et al. [33]. In accordance with results obtained by Costagliola et al. [23] we had observed a similar oxidative pattern in hemodialyzed uremic patients, explaining the key role of the GSSG levels in causing cellular damage [30,31]. For this reason we hypothesize that, concomitant with decreased HMPS function, a defective efflux of GSSG may explain its strong increase in the RBC. Chronic oxidative exposure can result in an adaptative response of the RBC precursors (in agreement with the observation of increased antioxidant enzyme levels such as GSH-S-T and GSH-Px) in CAPD as well as in HD, although to a lesser degree in the latter [31]. This cellular response fits with the results reported by several authors on uremic anemia in the presence of unmodified cytosolic GSH levels [26,29,34]. GSH-Px is not only indicative of oxidative stress, but also appears to be a reliable index of this adaptation mechanism. This observation confirms the evidence found in our in vivo oxidative studies in a rabbit-RBC oxidative stress model (unpublished data). ,. In the case of plasma, GSH levels were decreased and GSSG slightly increased. In this context we also observed unmodified GSH-Px activity and increased TBAR levels. The absence of a specific defect of this enzyme in CAPD, postulated in HD, could be explained by a better selenium metabolism regulation or by a minor plasma oxidative injury and consequent enzyme damage [31,34,35]. Other studies are in progress to better understand the role of GSH-Px in the increased plasma OFR activity in uremic subjects.

F. Canestrari et aL / Clinica Chimica Acta 234 (1995) 127-136

135

In conclusion, we found that RBC and plasma are exposed to oxidative stress in CAPD. This condition is characterized by a decreased GSH/GSSG ratio and is particularly evident in RBC, in which TBAR values are increased up to three times as compared with normal levels. This appears to be a consequence of GSSG accumulation which can, therefore, be considered a good index of qualitative and quantitative RBC damage. However, the RBC and plasma redox defect observed in CAPD appeared smaller than in HD patients previously studied [30,31] and RBC GSH peroxidase induction appeared more effective. From this point of view, the increased RBC survival and reduced need for hemotransfusion and erythropoietin support in CAPD, as compared with HD, can be explained by the better preservation of the redox state and lesser oxidative stress in the RBC and plasma. Given these results, it is still impossible to identify with certainty the primary event causing oxidative damage and shortened RBC lifespan. We can, however, speculate that metabolic changes related to the physiology of dialysis are important factors involved in uremic anemia.

References [1] Eschbach JW, Adamson JW. Anemia of end-stage renal disease (ESDR). Kidney Int 1985;28:1-5. [2] Jacobs HS, Eaton JW, Yawata Y. Shortened blood cell survival in uremic patients; beneficial and deleterious effects of dialysis. Kidney Int 1975;7:139-143. [3] Naets JP. Hematological disorders in renal failure. Nephron 1975;14:181-194. [4] Paul JL, Roch-Arveiller M, Man NK, Luong N, Moatti N, Raichvarg D. Influence of uremia on polymorphonuclear leukocytes oxidative metabolism in end-stage renal disease and dialyzed patients. Nephron 1991;57:428-432. [5] Horl WH, Riegel W, Schollauer P. Granulocyte activation during hemodialysis. Clin Nephrol 1986;26:30-34. [6] Taccone-Gallucci M, Giardini O, Lubrano R. Red blood cell lipid peroxidation in predialysis chronic renal failure. Clin Nephrol 1987;27:238-241. [7] Suha-Yalcin A, Yurtkuran M, Dilek K, Kilinc A, Taga Y, Emerk K. The effect of vitamin E therapy on plasma and erythrocyte lipid peroxidation in chronic hemodialysis patients. Clin Chim Acta 1989;185:109-112. [8] Taylor JE, Scott N, Bridges A, Henderson IS, Stewart WK, Belch JJF. Lipid peroxidation and antioxidants in continuous ambulatory dialysis patients. Perit Dial lnt 1992;12:252-256. [9] De Paepe MBJ, Scheistraete KHG, Ringoir SMG, Lameire NH. Influence of continuous ambulatory peritoneal dialysis on anemia of end stage renal disease. Kidney lnt 1983;23:744-748. [10] Yawata Y, Jacob HS. Abnormal red cell metabolism in patients with chronic uremia: nature of the defect and its persistence despite adequate hemodialysis. Blood 1975;45:231-239. [11] Taccone-Gallucci M, Giardini O, Lubrano R et al. Red blood cell lipid peroxidation in continuous ambulatory peritoneal dialysis patients. Am J Nephrol 1986;6:92-95. [12] Meister A, Anderson ME. Glutathione. Annu Rev Biochem 1983;52:711-760. [13] Bates DA, Winterbourn CC. Haemoglobin denaturation, lipid peroxidation and haemolysis in phenylhydrazine-induced anaemia. Biochim Biophys Acta 1984;798:84-87. [14] Beutler E, West C, Blume KG. The removal of leukocytes and platelets from whole blood. J Lab Clin Med 1976;88:329-333. [15] Beutler E. A manual of biochemical methods (3rd ed). New York: Grune and Stratton, 1984. [16] Srivastava SK, Beutler E. Accurate measurement of oxidized glutathione content of human, rabbit, and rat red blood cells. Anal Biochem 1968;25:70-76. [17] Ceballos-Picot I, Trivier JM, Nicole A, Sinet PM, Thevenin M. Age-correlated modifications of

136

[18]

[19] [20] [21] [22] [23] [24] [25] [26]

[27] [28] [29]

[30]

[31]

[32]

[33] [34]

[36]

F. Canestrari et al./ Clinica Chimica Acta 234 (1995) 127-136

copper-zinc superoxide dismutase and glutathione-related enzyme activities in human erythrocytes. Clin Chem 1991;38:66-70. Stocchi V, Cucchiarini L, Canestrari F, Piacentini MP, Fornaini G. A very fast ion-pair reversedphase HPLC method for the separation of the most significant nucleotidcs and their degradation products in human red blood cells. Anal Biochem 1987;167:181-190. Buege JA, Aust SD. Microsomial lipid peroxidation. Methods Enzymol 1978;LII:302-310. Eeterbauer H, Checseman KH. Determination of aidehydic lipid peroxidation products: malonaldehyde and 4-hydroxynonenal. Methods Enzymol 1990;186:407-421. Lowry OH, Rosembrough N J, Farr AL, Randall RJ. Protein determination with the folin reagent. J Biol Chem 1951;193:265-275. Magnani M, Stocchi V, Cucchiarini L, Chiarantini L, Fornaini G. Red blood cell phagocytosis and lysis following oxidative damage by phenylhydrazine. Cell Biochem Funct 1986;4:263-269. Costagliola C, Romano L, Sorice P. Anemia and chronic renal failure: the possible role of the oxidative state of glutathione. Nephron 1989;52:11-14. Giardini O, Taccone-Gallucci M, Lubrano R. Evidence of red blood cell membrane lipid peroxidation in haemodialysis patients. Nephron 1984;36:235-237. Dasgupta A, Hussain S, Ahmad S. Increased lipid peroxidation in patients on maintenance hemodialysis. Nephron 1992;60:56-59. Toborek M, Wasik T, Drozdz M, Klin M, Magner-Wrobel K, Kopieczna-Grzebieniak E. Effect of hemodialysis on lipid peroxidation and antioxidant system in patients with chronic renal failure. Metabolism 1992;41:1229-1232. Paul J-L, Sail N-D, Soni T et ai. Lipid peroxidation abnormalities in haemodialysis patients. Nephron 1993;64:106-109. Vaneila A, Geremia E, Pinturo R et al. Superoxide dismutase activity and reduced glutathione content in erythrocytes of uremic patients on chronic dialysis. Acta Haematol 1983;70:312-315. Chauhan DP, Gupta PH, Nampoothiri MRN. Determination of erythrocyte superoxide dismutase, catalase, glucose-6-phosphate dehydrogenase, reduced glutathione and maiondialdehyde in uremia. Clin Chim Acta 1982;123:153-159. Galli F, Canestrari F, Pascucci M et al. Metabolismo del glutatione e condizione eritrocitaria nel soggetto in dialisi extracorporea: evidenze per uno specifico meccanismo di danno ossidativo cellulare. In: Timio M, Wizemann V, eds. Cardionephrology. Vol. 2. Milano: Wichtig, 1993;180-183. Canestrari F, Galli F, Giorgini A, Galiotta P, Pascucci M, Boss6 M. Erythrocyte (RBC) redox state in uremic anemia: hemodialysis (HD) effects and glutathione metabolism relevance. Acta Haematol 1994;91:187-193. Girelli D, Lupo A, Trevisan MT et al. Red blood cell susceptibility to lipid peroxidation, membrane lipid composition, and antioxidant enzymes in continuous ambulatory peritoneal dialysis patients. Petit Dial Int 1992;12:205-210. Taccone-Gallucci M, Lubrano R, Belli A. Lack of oxidative damage in serum polyunsaturated fatty acids before and after dialysis in chronic uremic patients. Int J Artif Organs 1989;12/8:515-518. Mimic-Oka J, Djukanovic L, Simic T, Stefanovski J, Ramie Z. Glutathione and its associated enzymes in chronic renal failure. Amino Acids 1991;1:169-170. Saint-G-eorges MD, Bonnefont D J, Bourely BA et al. Correction of selenium deficiency in hemodialyzed patients. Kidney Int 1989;36:$274-$277.