Barry E. Walker

Decreased antioxidant status and increased lipid peroxidation in patients with septic shock and secondary organ dysfunction

OBJECTIVE To determine antioxidant vitamin concentrations, lipid peroxidation, and an index of nitric oxide production in patients in the intensive care unit (ICU) with septic shock and relate the findings to the presence of secondary organ failure. DESIGN A prospective, observational study. SETTING A nine-bed ICU in a University teaching hospital. PATIENTS Sixteen consecutive patients with septic shock, defined as: a) clinical evidence of acute infection; b) hypo- or hyperthermia (< 35.6 degrees C or > 38.3 degrees C); c) tachypnea (> 20 breaths/min or being mechanically ventilated); d) tachycardia (> 90 beats/min); e) shock (systolic pressure < 90 mm Hg) or receiving inotropes. Fourteen patients also had secondary organ dysfunction. INTERVENTIONS None. MEASUREMENTS AND MAIN RESULTS Antioxidant vitamin concentrations were significantly lower in the patients than the reference range obtained from a comparable group of healthy controls. The mean plasma retinol (vitamin A) concentration was 26.5 +/- 19.3 micrograms/dL compared with 73.5 +/- 18.3 micrograms/dL in healthy subjects (p < .01). Additionally, 13 (81%) patients had retinol values below the lower limit of our reference range (< 37.0 micrograms/dL). Tocopherol (vitamin E) plasma concentrations were below the reference range in all patients (< 9.0 mg/L), with a mean value of 3.6 +/- 2.0 mg/L compared with 11.5 +/- 1.3 mg/L in healthy subjects (p < .001). Plasma beta carotene and lycopene concentrations were undetectable (< 15 micrograms/L) in eight (50%) patients, and below our reference range (< 101 micrograms/L and < 154 micrograms/L, respectively) in the remaining patients. In the five patients with three or more dysfunctional secondary organs, plasma thiobarbituric acid-reactive substances were significantly increased (p < .05), suggesting increased lipid peroxidation. Concentrations of thiobarbituric acid-reactive substances correlated negatively with both plasma retinol and plasma tocopherol (r2 = .42, p < .01 and r2 = .48, p < .005, respectively). In the five patients from whom we were able to collect urine, nitrite excretion was increased approximately 400-fold (p < .001). CONCLUSIONS These data indicate decreased antioxidant status in the face of enhanced free radical activity, and suggest potential therapeutic strategies involving antioxidant repletion.

THE EFFECTS OF INTRAVENOUS ANTIOXIDANTS IN PATIENTS WITH SEPTIC SHOCK

Oxidative stress is implicated in septic shock. We investigated the effect of intravenous antioxidant therapy on antioxidant status, lipid peroxidation, hemodynamics and nitrite in patients with septic shock. Thirty patients randomly received either antioxidants (n-acetylcysteine 150 mg/kg for 30 min then 20 mg/kg/h plus bolus doses of 1 g ascorbic acid and 400 mg alpha-tocopherol) or 5% dextrose. Basal vitamin C was low and redox-reactive iron was elevated in all patients. In the 16 patients receiving antioxidants, vitamin C increased (p = .0002) but total antioxidant capacity was unaffected. Lipid peroxides were elevated in all patients but did not increase further in the patients receiving antioxidants. Plasma total nitrite also increased (p = .007) in the antioxidant group. Heart rate increased in patients receiving antioxidants at 60 min (p = .018) and 120 min (p = .004). Cardiac index also increased at 60 min (p = .007) and 120 min (p = .05). Systemic vascular resistance index decreased at 120 min in the antioxidant treated patients (p = .003). The effect of antioxidants on hemodynamic variables has not previously been reported. Antioxidant administration may be a useful adjunct to conventional approaches in the management of septic shock.

Regulation of nitric oxide synthase activity in cultured human endothelial cells: Effect of antioxidants

Nitric oxide release is induced in many cells, including vascular endothelium, as part of the host response to inflammation. Nitric oxide synthase activity is increased in patients with sepsis, associated with increased oxidant demands and decreased antioxidant protection. We used a human vascular endothelial cell line to investigate the influence of antioxidants on nitric oxide synthase activity. Cells were cultured to confluence and incubated with interferon gamma, tumor necrosis factor, and lipopolysaccharide in the combined presence of the antioxidants ascorbic acid, Trolox, catalase, or superoxide dismutase, singly and in combination, for 48 h. Additionally, some cells were incubated with hypoxanthine-xanthine oxidase or a nitric oxide donor. Nitric oxide synthase activity was upregulated by cytokine exposure (p < .0005). Ascorbic acid and superoxide dismutase/ catalase resulted in decreased enzyme activity (p < .05). Superoxide anion release from xanthine oxidase caused increased activity (p < .05) and exogenous nitric oxide tended to suppress synthase activity. We suggest that antioxidants scavenge superoxide anion, enabling feedback inhibition of nitric oxide synthase activity by nitric oxide, and thus reducing enzyme activity. Exogenous nitric oxide also has a similar effect. Superoxide generation suppresses this feedback inhibition. This study has important implications in patients with sepsis in whom nitric oxide synthase inhibitor therapy is currently under investigation.

Zinc concentrations in pure populations of peripheral blood neutrophils, lymphocytes and monocytes

It is doubtful if the measurement of plasma or serum zinc is of value in assessing zinc status. Leucocyte zinc has been suggested as an alternative since it may be representative of tissue zinc stores; but in many studies poorly defined cell populations make interpretation difficult. This paper describes detailed techniques for the isolation and analysis of pure populations of neutrophils, lymphocytes and monocytes. Zinc concentrations (± 1SD) in normal subjects were 1·;26 ± 0·;27 nmol/mg protein, 1·;85 ± 0·;32 nmol/mg protein and 2·;58 ±0·;65 nmol/mg protein in neutrophils, lymphocytes and monocytes respectively. Fasting caused a significant decrease in neutrophil and lymphocyte zinc, and an increase in monocyte zinc. Supplementation of zinc-replete subjects with 135 mg zinc/day for 3 weeks had no significant effect on cellular zinc concentrations.

The effect of anticoagulant choice on apparent total antioxidant capacity using three different methods

We assessed total antioxidant capacity using three different methods, in plasma samples treated with either EDTA or heparin as anticoagulant, from 26 healthy subjects. Total antioxidant capacity was determined using an oxygen electrode (as the total peroxyl radical-trapping antioxidant parameter), by enhanced chemiluminescence, and by measurement of the antioxidant-mediated quenching of the absorbance of a radical cation. The choice of anticoagulant had a profound effect on antioxidant capacity with heparinized plasma giving consistently higher values than plasma anticoagulated with EDTA. Using the oxygen electrode the mean value was 786·5 ± 171·5 μmol/L (heparin) compared to 681·4 ± 160·4 μmol/L (EDTA, P < 0·01). The chemiluminescence technique gave a mean antioxidant capacity of 915·6 ± 214·1 μmol/L in heparin samples and 714·4 ± 195·4 μmol/L in EDTA samples (P < 0·0001). The absorbance quenching technique gave a mean value of 867·0 ± 199·2 μmol/L (heparin) and 675·5 ± 245·4 μmol/L (EDTA, P < 0·001). All methods tested showed comparable results for EDTA plasma, but the chemiluminescence technique gave higher apparent antioxidant capacity than either of the other two techniques when heparin plasma was used. We suggest that either heparin is interacting to enhance antioxidant protection perhaps through release of superoxide dismutase, or the chelation of metal ions by EDTA is limiting the activity of antioxidant metalloenzymes. Consistency in the choice of anticoagulant is clearly extremely important.

Monocyte zinc andIn vitro prostaglandin E2 and interleukin-1β production by cultured peripheral blood monocytes in patients with Crohn's disease

This study investigated the relationship between zinc status and prostaglandin E2 and interleukin-1β production by cultured monocytes in patients with Crohns disease. Monocyte zinc was significantly decreased in both 12 inpatients and 22 outpatients compared with controls (P<0.001) but lymphocyte and polymorphonuclear cell zinc were normal. When cultured monocytes from 10 outpatients with Crohns disease were stimulated with lipopolysaccharide, prostaglandin E2 production increased markedly, coupled with a fall in monocyte zinc. In matched controls, prostaglandin E2 production was significantly less and monocyte zinc remained stable. No difference in interleukin-1 release was noted between patients and controls. The addition of prednisolone to cell cultures suppressed prostaglandin E2, interleukin-1 synthesis, and monocyte zinc did not change. Zinc chloride augmented prostaglandin E2 production in patients, but not controls, and interleukin-1 remained stable. These results demonstrate a link between low monocyte zinc concentration and excessive prostaglandin production in patients with Crohns disease.

The effects of acute infection on indices of zinc status

Various indices of zinc status were assessed in 12 patients with acute urinary tract or chest infections on Day 1 and Day 7 of the infection. Leucocyte counts were raised on Day 1 but had returned to near normal by Day 7. Plasma zinc was decreased on Day 1 in conjunction with depressed plasma albumin concentrations (r = 0.71, p < 0.001) but both had returned to normal by Day 7. Mononuclear cell zinc was raised in all patients on Day 1 compared to Day 7 and control values, but polymorphonuclear cell zinc remained unchanged. However, polymorphonuclear cell alkaline phosphatase activity was grossly increased on Day 1 and correlated with leucocyte count (r = 0.61, p < 0.01). Plasma alkaline phosphatase activity was variable. These results indicate that in patients with infections measurement of plasma mononuclear cell zinc concentration and alkaline phosphatase activity are misleading indicators of zinc status. Polymorphonuclear cell zinc is unaffected by leucocytosis, inflammation and stress and may therefore provide a more reliable index of zinc status in such patients.

Effect of Fasting, Self-Selected and Isocaloric Glucose and Fat Meals and Intravenous Feeding on Plasma Zinc Concentrations

This study investigated the effect of fasting, self-selected meals and isocaloric oral glucose and fat meals and intravenous (i.v.) feeding on plasma zinc concentrations in men. Plasma zinc remained stable when volunteers fasted all day, but self-selected meals and 600 kCal of dextrose or fat emulsion caused significantly reduced plasma zinc concentrations [mean(SD) 12 · 1 (1 · 4), 12 · 3 (0 · 6) and 12 · 2 (0 · 7) μmol/L at 1400 h, respectively, compared with a fasting level of 13 · 9 (1 · 6) μmol/L at 0800 h, P < 0 · 05]. In patients undergoing intravenous hyperalimentation, plasma zinc decreased from 12 · 0(1 · 4) μmol/L at 0800 h to 10 · 0 (1 · 1) μmol/L at 1400 h [mean(SD), P < 0 · 01]. These data show that both enteral and i.v. feeding cause a decline in plasma zinc and that glucose alone is not responsible for this post-prandial fall since ingestion of isocaloric amounts of glucose or fat have a similar effect.

Leucocyte zinc in non-Hodgkin's lymphoma and Hodgkin's disease.

Zinc status and the effect of zinc supplementation were assessed in groups of patients with non-Hodgkins lymphoma and Hodgkins disease; patients were either untreated or in remission. In the patients in remission, plasma zinc was normal; and whereas 30% of untreated patients had low plasma zinc, the group as a whole did not differ from normal. For mononuclear cell zinc, the range of values in the disease group was far wider than in controls, but there was no significant difference between the means of the groups. Granulocyte zinc was significantly lower in both the groups of patients in remission from non-Hodgkins lymphoma and Hodgkins disease compared with the control group. Significant increases were found in the plasma copper, ceruloplasmin, and the copper-to-zinc ratio in several of the patient groups. Plasma zinc increased by 23% with zinc supplementation (50 mg elemental Zn/day), but there was no effect on mononuclear cell or granulocyte zinc. Apart from granulocyte zinc, there is little evidence of zinc deficiency in non-Hodgkins lymphoma or Hodgkins disease. However, the presence of depleted granulocyte zinc levels could modify the immune function of this cell population.

Correction of cellular zinc depletion by oral zinc supplementation in elderly subjects

Immune function declines with age, and has been implicated in the increased incidence of cancer and infections in the elderly. In this hospital, many elderly patients have evidence of zinc depletion. In the present study, we supplemented those elderly patients who had depressed polymorphonuclear cell (PMNC) zinc levels with 135 mg oral zinc sulphate for 4 weeks. Plasma and PMNC zinc increased markedly but the percentage of peripheral blood T-lymphocytes expressing the surface markers CD3, CD4 and CD8 were unchanged. Plasma concentrations of vitamins A and E also remained constant. This study confirms the 25-30% incidence of cellular zinc depletion in this patient population, and demonstrates that zinc concentrations can be brought back to within normal limits by oral zinc supplements, but with no effect on T-cell phenotypes.