Antioxidant properties of drugs used in Type 2 diabetes management: could they contribute to, confound or conceal effects of antioxidant therapy?

ABSTRACT Objectives: This is a narrative review, investigating the antioxidant properties of drugs used in the management of diabetes, and discusses whether these antioxidant effects contribute to, confound, or conceal the effects of antioxidant therapy. Methods: A systematic search for articles reporting trials, or observational studies on the antioxidant effect of drugs used in the treatment of diabetes in humans or animals was performed using Web of Science, PubMed, and Ovid. Data were extracted, including data on a number of subjects, type of treatment (and duration) received, and primary and secondary outcomes. The primary outcomes were reporting on changes in biomarkers of antioxidants concentrations and secondary outcomes were reporting on changes in biomarkers of oxidative stress. Results: Diabetes Mellitus is a disease characterized by increased oxidative stress. It is often accompanied by a spectrum of other metabolic disturbances, including elevated plasma lipids, elevated uric acid, hypertension, endothelial dysfunction, and central obesity. This review shows evidence that some of the drugs in diabetes management have both in vivo and in vitro antioxidant properties through mechanisms such as scavenging free radicals and upregulating antioxidant gene expression. Conclusion: Pharmaceutical agents used in the treatment of type 2 diabetes has been shown to exert an antioxidant effect..


Introduction
The number of people worldwide with diabetes mellitus (DM) has increased from 30 million to 180 million since 1985 [1][2][3]. This increase is anticipated to continue, with the fastest increases seen in Asia [1,2,4]. Over 90% of diabetes is Type 2 DM, which is also associated with dyslipidemia, hypertension, and elevated plasma uric acid [3,5], which can lead to long term micro-and macro-vascular complications.
Owing to their various metabolic problems, most Type 2 DM subjects are commonly treated by a 'polypharmacy' of oral hypoglycemic agents, statins, fibrates, and anti-hypertensive drugs of various types [6]. Some of these have been reported to have pleiotropic effects and antioxidant properties [7][8][9]. Oxidative stress is increased in DM, and there is often a depletion of body stores of ascorbic acid (vitamin C), which is an important dietary-derived antioxidant. Antioxidant supplementation has been suggested as a potentially beneficial adjunct therapy [10,11]. In this article, we review the literature in relation to antioxidant effects of drugs used in the management of diabetes, and discuss results of antioxidant supplementation studies in DM patients from the perspective of the possible confounding or concealing influence of drug-induced antioxidant effects.

Type 2 DM and its management: a brief overview
Type 2 DM is a state of continuous or intermittent hyperglycemia caused by a relative deficiency of insulin. Tissues are insulin resistant and there may also be β-cell dysfunction or failure [12]. Hypoglycemic agents are aimed at improving insulin secretion and action, and limiting or slowing absorption of glucose in the gastrointestinal tract [13]. These drugs are taken orally, and may be combined with insulin injections for those patients who fail to respond acceptably to any of the oral hypoglycemic agents alone (Table 1).
Type 2 DM is often accompanied by a spectrum of other metabolic disturbances, including elevated plasma total and low density lipoprotein cholesterol (TC and LDL-C), low high density lipoprotein cholesterol (HDL-C), elevated triglycerides (Tg), elevated uric acid, hypertension, endothelial dysfunction, central obesity, and elevated inflammatory biomarkers (high sensitivity C-reactive protein (hsCRP) and inflammatory cytokines such as interleukin 6 (IL-6) [24]. The drugs most commonly used to treat dyslipidemia and hypertension in Type 2 DM patients are presented in Table 2.
Antioxidant effects of drugs used in the management of Type 2 DM As noted, there is a polypharmacy armory for the management of patients with DM, and although many patients may be managed using dietary strategies alone, many patients with diabetes will be on a combination of several of these drugs. Many of these drugs exhibit pleiotropic effects, that is, beneficial effects separate from their primary action. As summarized in Table 3, in vitro studies, using the Ferric 1-butyl-3-(4-methylphenyl)sulfonylurea C 12 H 18 N 2 O 3 S Same as glipizide. 500 mg. C max = 63 ± 11 μg/ ml AUC = 721 ± 94 μg.h/ml [17] Nateglinide (Starlix) (2S)-3-phenyl-2-[(4-propan-2-ylcyclohexanecarbonyl)amino]propanoic acid C 19 H 27 NO 3 Stimulates pancreas to produce insulin (similar to the sulfonylureas). Appears to have a faster onset and a shorter duration of action than sulfonylureas. May prevent the rapid, transient rise in blood glucose that occurs immediately following a meal.
60 or 120 mg three times daily. C max = 3.09 ± 1.64 μg/ml AUC = 6.93 ± 1.99 μg.h/ml [18] Repaglinide ( Stimulates insulin production. Rapid onset and short duration of action. Immediately before a meal. C max = 30.96 ± 9.06 μg/l AUC = 36.03 ± 6.00 μg.h/l [19] Rosiglitazone maleate (Avandia)  [21] (Continued )    Reducing Antioxidant Power (FRAP) Assay, have shown that some drugs do show antioxidant properties (our unpublished data). While the FRAP assay was originally developed to investigate antioxidant power in biological samples, such as plasma and urine, it has since been applied to foods and health products, and since it is a test for the intrinsic chemical antioxidant properties of an agent, it was applied to drugs in this instance [42,43]. Studies by other investigators have shown that these drugs used in the management of DM can improve antioxidant status and ameliorate oxidative stress in cell culture, animal, and human trials (Table 4) [44,45].
The antioxidant effects of statins have come under intense research [90][91][92]. Atorvastatin has decreased superoxide production in human endothelial cells exposed to high glucose, and also protected these cells from hydrogen peroxide mediated damage [93,94]. In human studies, statins have been shown to protect lymphocytic DNA from oxidative damage, decrease concentrations of oxidized-LDL, plasma concentrations of protein-bound tyrosines, urinary F 2 - Competes for binding at beta (1)-adrenergic receptors in heart and vascular smooth muscle, inhibiting sympathetic stimulation, resulting in lowered systolic and diastolic blood pressure.
C max = 709 ± 221 ng/ ml AUC = 6120 ± 2171 ng.h/ml [41] isoprostanes as well as to increase concentrations of an erythrocyte antioxidant enzyme, superoxide dismutase (SOD) [59,62,66,69,81,87,93,95]. Although both in vitro and in vivo studies provide evidence of the antioxidant properties of statins, conflict exists with regard to the effect of statins on plasma tocopherols (vitamin E). Cangemi et al. [69] found in a retrospective study that subjects with metabolic syndrome on statin therapy (simvastatin or atorvastatin) for 6 months or more had significantly higher concentrations of plasma vitamin E (p = 0.02) and lower concentrations of plasma 8-hydroxy-2'deoxyguanosine (8-OHdG) (p < 0.01) compared to those subjects with metabolic syndrome not treated with statins [81]. The statin-treated group had similar plasma vitamin E and 8-OHdG concentrations as the healthy control group. While in Jula et al.'s (2002) study of hyper-cholesterolemic subjects, 12-week treatment with simvastatin had no effect on serum ascorbic acid, but decreased serum α-tocopherol by 16.2% and β-carotene by 19.5% [60]. Plasma α-tocopherol was also decreased in normocholesterolemic subjects given atorvastatin for 3 months [60,65]. It is noted here though that Jula's and Oranje's work did not lipid-standardize the measurements of the lipid-soluble vitamins, and therefore, the decreases seen may simply be due to the lipid-lowering effect of the statins [60,65]. Plasma α-tocopherol was not significantly changed after 2 months of simvastatin treatment in hyper-cholesterolemic subjects in De Caterina's [59] study, and additional supplementation with vitamin E did not enhance the antioxidant effect seen when simvastatin was taken alone.
The biguanide metformin is one of the most widely prescribed oral hypoglycemic medications [96,97]. In vivo studies have shown that metformin was able to scavenge hydroxyl radicals, and to reduce the production of ROS in bovine aortic endothelial cells, though the results in these studies are mixed [98,99]. Metformin significantly decreased urinary F 2 -isoprostanes and increased plasma concentrations of vitamins A and E in Type 2 DM subjects, although no effects were seen in serum malondialdehyde (MDA) and total antioxidant status (TAS) in subjects with polycystic ovarian syndrome after 12-week treatment [49,51]. In the same study, treatment with rosiglitazone was able to increase plasma TAS from 0.95 to 1.21 mmol/l significantly (p < 0.005), and decrease plasma MDA from 7.46 to 4.02 nmol/l (p < 0.001). In an animal study of Type 2 DM mice, rosiglitazone treatment for 7 days was able to significantly decrease serum F 2 -isoprostanes and vascular superoxide production, and increase vascular catalase concentrations (all p < 0.05) [100].

What is the source of increased oxidative stress in DM?
There is no generally agreed source of the increased oxidative stress found in DM [120]. It has been proposed, but disputed, that acute elevation of plasma glucose is the trigger [120][121][122]. Others suggest that it is glycemic variability that is the root cause [123,124]. However, conceptually at least, if antioxidant status is high, then oxidative damage to key biomolecules in Type 2 DM might be avoided, improving outcome regardless of the exact relationship between glycemic control and generation of reactive oxygen species. This has led to the suggestion that antioxidant supplementation should be investigated as an adjunct therapy in DM, the hypothesis being that increasing antioxidant defense will lower oxidative stress and help slow or prevent vascular changes that lead to complications [125,126].

Can antioxidant supplementation decrease oxidative stress in DM?
This has been explored in human supplementation trials with various types and combinations of antioxidants, and using diverse biomarkers of oxidative stress. Some researchers have also investigated the effects of antioxidant supplementation on inflammation, lipids, blood pressure, and glycemic control. These studies (summarized in Table 5) have not revealed clear evidence of benefit. However, none of these studies considered the possible confounding effect of therapy with drugs with antioxidant (or other) effects.

Evidence of decreased oxidative stress in DM after antioxidant supplementation
In Ward et al.'s [127] study, the only oxidative stress biomarker measured was plasma and 24-hour urine F 2 -isoprostanes. It was found that plasma F 2 -isoprostanes were significantly reduced in both the tocopherol treatment groups when compared to the placebo group, but no difference was seen in   Serum α-tocopherol, β-carotene, and ascorbic acid.
Serum LDL-C fraction used for the determination of diene conjugation and total peroxyl radical trapping antioxidant potential. Antioxidant potential of isolated LDL-C samples was determined.
Simvastatin treatment did not affect serum ascorbic acid, but decreased serum α-tocopherol by 16.2% and βcarotene by 19.5% and total peroxyl radical trapping potential of serum LDL-C by 16.9%. Relative antioxidant power of LDL-C preparations (LDL-C TRAP/mmol of LDL-C) increased by 17.4 Note: The crude value of αtocopherol was not corrected for the cholesterol levels.

Combination
Type 2 DM subjects with marked hypertriglyceridemia. All diabetic patients were treated with either metformin or a combination of metformin and a sulfonylurea to achieve HbA1c of < 9% before the study. 6-week washout/run-in period for lipid-lowering medications before subjects were randomly assigned to a treatment group [82] (  urinary F 2 -isoprostanes in any of the groups. This could be because urinary F 2 -isoprostane excretion may be confounded by local production in the kidney and is more variable [128]. In another study, vitamin C supplementation had no effect on the vitamin E concentration of HDL-C in patients with diabetes, although an increase in intracellular glutathione (GSH) was seen (p < 0.01) when compared to the placebo group [129]. The investigators state that these subjects were on either glibenclamide, metformin, or insulin to control their diabetes, but it is likely that the subjects would also be on a statin or anti-hypertensive drugs [129]. No effect on conjugated dienes, thiobarbituric acid reactive substances (TBARS), or susceptibility of LDL-C or HDL-C to oxidation was seen post supplementation [129]. The increase seen in erythrocyte GSH, a naturally occurring antioxidant, is probably due to the low GSH found in this group before supplementation (0.5 ± 0.7 nmol/mg protein), as the normal range for erythrocyte GSH is between 745 and 1473 μmol/l in healthy individuals aged 12-69 years [130].
Nuttall et al.'s study of nine Type 2 DM subjects included the most advanced age cohort and one of two cross-over studies under review here, with 500 mg vitamin C and 400IU vitamin E taken during the first supplementation period and 1000 mg vitamin C and 800IU vitamin E taken during the second period [131]. The study was not placebo controlled [131]. With the exception of plasma antioxidants, lipid peroxides were the only marker of oxidative stress measured [131]. It is noted here that all subjects had advanced complications, including neuropathy, cardiovascular complications, and cerebrovascular disease, and the subjects were taking sulphonylureas, and/or metformin, and insulin, although no subjects were taking lipid-lowering therapy [131]. A bigger decrease (p < 0.01) in lipid peroxides was seen after the first (lower dose) supplementation than the second (p < 0.05) [131]. The investigators note that the higher doses of vitamins C and E taken during the second supplementation period may have reached the pro-oxidant threshold [131].
Gazis et al. [132] supplemented 48 subjects with Type 2 DM and no vascular complications with 1600IU α-tocopherol for 8 weeks. This study was the only one to include individual information on the drugs each subject was using to manage diabetes, but no separate statistical analysis was performed on supplementation outcome taking into consideration the medication history of the subjects. No direct biomarkers of oxidative stress were measured and no significant changes in blood flow or vasodilation were seen. In Tessier et al.'s study, 36 subjects with Type 2 DM were randomized into one of three groups, a placebo group, a group taking 0.5 g and a group taking 1 g daily of vitamin C for 12 weeks. DM status was stable in all subjects with HbA1c ≤ 9% [129]. Conjugated dienes, TBARS, vitamin E content of LDL and HDL lipoproteins and granulocyte levels of vitamin C, GSH and glutathione disulfide (GSSG) were measured [129]. After supplementation (both the 0.5 g per day and 1.0 g per day groups), there was a significant increase (p < 0.05) in cellular GSH when compared to baseline, although no changes in GSSG were observed [129]. No changes in markers of lipid peroxidation (TBARS and conjugated dienes) were seen [129].
In a supplementation study using mixed tocopherols, 55 Type 2 DM subjects were randomized to take (1) 500 mg αtocopherol per day, or (2) a combination of 75 mg α-  Multiple DNA base oxidation products, but not DNA base de-amination or chlorination products were found to be elevated in white blood cell DNA from DM patients compared with age-matched controls. No significant difference seen in 5chlorouracil and concentrations of the base de-amination products hypoxanthine and xanthine. 8-hydroxyguanine, 5-OH uracil, 5-OH methyluracil, thymine lycol, 5-OH methyldantoin, 5-OH hydantoin, 5-OH cytosine, 2-OH adenine, 8-OH adenine, and FAPy-adenine concentrations were elevated, whereas FAPy-guanine remained unchanged. Total DNA base oxidative damage products in DM subjects was double than that found in healthy controls, p < 0.01. Type 2 DM [n = 24] Age-, sex-, and BMI-matched subjects (obese group) [n = 19] Unmatched control group [n = 34] Exclusion criteriapresence of Prader-Willi Syndrome, hypothyroidism, known alcohol or drug abuse, congenital CVD, history of malignancy, use of glucocorticoids, and chronic renal failure or known primary renal disease. None of the participants were taking lipid-lowering drugs [115] Plasma IL-6 concentrations were 29.3% higher in the obese group compared with the control group, DM group had a plasma IL-6 concentration 44.5% higher than the control group and 21.5% higher than the obese group, p < 0.001 for all. Ultra-sensitive CRP was not different among the groups. OxLDL was elevated 1.6 fold in the DM group compared with the control (p = 0.002). Plasma TAS was not different between the three groups. Type 2 DM subjects (mean duration of diabetes: 8.5 years HbA1c ∼7.1%) [n = 26] 52 age-matched healthy control subjects [n = 52] All Type 2 DM patients were free of microangiopathy. Exclusion criteriasmoking, inflammatory diseases, known myocardial infarction, stroke. *8 DM subjects were on with diet and exercise, 18 subjects were on oral hypoglycemic agents [116] Urinary 8-OHdG (11.1 ng/mg. creatinine) and isoprostane (0.87 ng/g. creatinine) were also significantly higher among diabetic subjects (p = 0.019, p = 0.009 respectively). Urinary 8OHdG level in control was 8.8 ng/mg. creatinine; urinary isoprostane 0.32 ng.g creatinine.
Type 2 DM patients (33 without diabetic complications and 27 with diabetic retinopathy, of which 21 were classified as non-proliferative diabetic retinopathy, 6 were proliferative diabetic retinopathy) [n = 60 in total] Healthy, age-matched control subjects [n = 32] Exclusion criteriaacute and chronic infections, fever, malignancy, acute, and chronic nephritis, cirrhosis, and congestive heart failure. All patients were treated with insulin only [117] Significant higher serum MDA found in DM subjects (6.72 vs 3.12 nmol/ml for controls, p < 0.01 tocopherol, 315 mg γ-tocopherol, and 110 mg of δ-tocopherol, or (3) placebo for a duration of 6 weeks [133]. Plasma and urinary F 2 -isoprostanes and erythrocyte SOD and glutathione peroxidase (GPx) were measured at baseline and at the completion of the supplementation period [133]. Treatment with either α-tocopherol or the mixed tocopherols significantly decreased plasma F 2 -isoprostanes when compared with the placebo group, (p < 0.001), although neither treatment affected urinary F 2 -isoprostanes [133]. There were no changes in SOD and GPx post supplementation of either α-tocopherol or mixed tocopherols [133]. Diabetes status in this population was well controlled, with mean HbA1c at around 6.6% [133]. Although the investigators documented the percentage of subjects taking oral hypoglycemics, anti-hypertensives, lipid-lowering drugs, aspirin, and statins, analysis of the data did not take account any influence these drugs may have had on the results. In a study by Ble-Castillo et al., Type 2 DM subjects were randomized to receive 800IU α-tocopherol per day (n = 13) or placebo (n = 21) for 6 weeks [134]. Subjects on insulin, hormonal therapy, antioxidant supplements, smokers, and with hypertensive were excluded [134]. The investigators reported a 46% reduction in erythrocyte MDA and a significant increase (p < 0.001) in serum total antioxidant in the α-tocopherol group post supplementation when compared to baseline [134]. These results are in conflict with other studies investigating the effect of antioxidant supplementation on oxidative stress markers in subjects with Type 2 DM and may be unreliable. The actual concentrations of serum TAS and the erythrocyte MDA were not given, and the results are simply shown in graphical form. In addition, the serum-TAS concentrations reported are lower than the range given by the manufacturers of the kit (Randox) by a factor of a thousand. In summary, vitamin C and vitamin E are the most popular antioxidants used in supplementation studies in subjects with Type 2 DM. The supplementation period of studies reviewed here is between 3 and 12 weeks. Ascorbic acid depletionrepletion studies have shown that both plasma and intracellular ascorbic acid will reach saturation at doses of 500 mg per day within about 40 days of supplementation, which is around 6 weeks [135]. This could mean plasma and tissue concentrations of ascorbic acid would not have reached optimal levels for that dose in studies with a supplementation period of less than 6 weeks, although the doses used are many fold higher than the recommended daily amount (RDA) of up to 90 mg [136]. All studies using vitamin E also supplemented subjects with much more than the RDA of 22.5IU (15 mg) [137].
Although all studies reviewed in this section (and summarized in Table 6) are antioxidant supplementation studies, the study population are diverse. Ages ranged from 40 to 77 years, and while some used subjects who were comparatively complication-free, subjects used in other studies had suffered serious micro-and macro-vascular complications [6,131].
With all these differences in age, methods of DM management and inclusion/exclusion criteria in the study populations and outcomes assessed to determine benefit, it is very difficult to compare results across studies. Most investigators excluded subjects who were already taking the supplement in question, while some asked subjects to stop taking any type of antioxidant supplements weeks prior to the start of the study [129,134]. Gaede et al. [6] asked their subjects to stop taking vitamin C and E and ACE-inhibitors 8 weeks prior to the start of the supplementation period, although hypoglycemic drugs continued to be used. However, Levine et al. did not specify such exclusion criteria [136]. It is noted here that the subjects with Type 2 DM in Darko et al.'s study were not deficient in vitamin C, with mean plasma concentrations of 58 ± 6 µmol/l, which is higher than the normal range of 23-50 µmoll [140,143]. The fact that the subjects were not deficient in vitamin C may have contributed to the lack of benefit seen in the only biomarker of oxidative stress measured, plasma F 2 -isoprostanes.
Although some studies collected information on the polypharmacy used in the management of Type 2 DM in the subjects studied, none of the investigators analyzes data taking into account the medication of the subjects.
With the exception of one study, it is clear that supplementing Type 2 DM subjects with oral antioxidants will not further decrease oxidative stress, as seen by the oxidative stress markers measured. It is noted though that none of the above studies looked at subjects who were on oral hypoglycemics, insulin, or statins separately from those subjects who were not taking any of these medications.
In the only antioxidant supplementation study involving patients with Type 1 DM, Beckman et al. gave, vitamin C (1 g) and vitamin E (800IU), per day, or placebo to Type 1 DM subjects (n = 26), Type 2 DM subjects (n = 23), and healthy matched controls (n = 45) for 6 months [144]. In those with Type 1 DM, antioxidant supplementation increased endothelium-dependent vasodilation (p = 0.023) when compared to baseline, whereas no such effect was seen in those with Type 2 DM [144]. Although no mention was given as to the number of Type 2 DM subjects who were on oral hypoglycemics, it is surmised that the majority of them would be on polypharmacy, whereas the Type 1 DM subjects would most likely be solely on injected insulin therapy. A search of the literature did not result in any studies which have investigated any antioxidant effect of insulin. Beckman et al. did note, however, that at baseline, the Type 1 DM subjects were younger, had lower total cholesterol (TC), glucose, and mean arterial pressure than the Type 2 DM subjects [144].
It cannot be discounted, therefore, that differences in age, cholesterol, and glucose concentrations contributed to the difference seen in the result, but it is proposed here that it is possible that the Type 2 DM subjects who are likely to be on oral hypoglycemics and statins would have reached antioxidant-effect-saturation. This does not mean that their tissues are saturated with antioxidants, but it might be feasible that, even if the tissue antioxidant concentration was increased, no further benefits will be seen. This is shown clearly in two controlled studies, De Caterina et al. [59] and Pereira et al. [61], summarized in Table 4, where no further oxidative stress-lowering effects were seen when vitamin E was added to the regimen of subjects taking statins.
It should be taken into consideration also that in health, the balance between reactive species and antioxidant defenses lies slightly in favor of the reactive species so that they are able to fulfill their biological roles and artificially increasing the antioxidant concentration may possibly result in deleterious effects [145].
Drugs used in the management of Type 2 diabetes and their mechanisms of antioxidant action Sulfonylureas stimulate insulin release by binding to the sulfonylurea receptor, a subunit of the ATP-dependent K + (K ATP ) channel complex on the cell membrane of pancreatic β-cells, causing an increase in intracellular calcium, which in turn increases the secretion of pro-insulin. Gliclazide is known to be a free-radical scavenger, and an in vivo study of 44 Type 2 DM subjects taking gliclazide for 10 months resulted in a decrease in 8-isoprostanes, a marker of lipid oxidation, and an increase in the total antioxidant capacity and SOD [146,147]. Other sulfonylureas, including glipizide tolazamide, and glibenclamide do not exhibit antioxidant activity [148]. Thiazolidinediones (TZD) work to decrease insulin resistance by binding to peroxisome proliferator-activated receptors gamma (PPARγ). These receptor molecules are found inside the cell nucleus, and when activated migrate to DNA and activate the transcription of a number of genes, including SOD and catalase [149][150][151]. In addition to being able to indirectly upregulate antioxidant genes, certain TZD, for example, troglitazone, may be able to exert direct antioxidant effects through a side chain which resembles α-tocopherol [152,153]. All TZDs exhibit intracellular antioxidant properties as they are able to reduce NO production through the transrepression of iNOS [153,154]. Significant increase in cellular GSH was observed (0.60 vs. 0.33 nmol/mg protein in the placebo group, p < 0.01) in patients on 0.5 g vitamin C/ day. Significant increases in cellular GSH (0.93 vs. 0.33 nmol/mg proteins in the placebo, p < 0.01), seen in those on 1.0 g vitamin C/day. Vitamin C had no effect on the vitamin E content of HDL in any group. At week 12, no significant difference was seen in lipid peroxidation/oxidative stress markers (conjugated dienes, TBARS) and susceptibility of LDL and HDL to oxidation in the two treatment groups compared to placebo.
Individual medication details are not given, but subjects on glibenclamide, metformin or insulin Type 2 DM patients with diabetes for >= 1 year and no clinical evidence of overt vascular diseases, acute or chronic inflammatory diseases [141] Subjects randomized to (1) Vitamin C (1 g/day) [n = 14] (2) Acetate cellulose placebo 1000 mg/day, [n = 13] Treatment period, 6 weeks MDA significantly decreased at fasting (p = 0.006) and postprandial states (p < 0.001) in vitamin C group, compared to placebo group. No significant differences were seen in fasting and postprandial lipid profile.
Subjects were on standard oral hypoglycemic agents, none on lipid-lowering drugs, hormone replacement therapy and diuretics, β-blockers, and aspirin.
Type 2 DM patients with no history of other chronic diseases, kidney stones, hyperparathyroidism, pregnancy and lactation, current insulin treatment and treatment for weight reduction [142] (1) Vitamin E (800 mg/day) [n = 37] (2) Placebo[n = 34] Treatment duration, 2 months Tg decreased significantly in vitamin E group (212 ± 85 mg/dl), compared to placebo group (254 ± 170 mg/dl). No significant differences were seen in serum fasting glucose, HbA1c and β-cell function after vitamin E supplementation, compared to placebo group.

No information given
Biguanides decrease hepatic glucose output and increase the uptake of glucose by skeletal muscle. Metformin exerts its antioxidant effect through mechanisms other than direct radical scavenging [155]. In vitro experiments have shown that metformin is able to prevent the formation of advanced glycation end-products when albumin was incubated in the presence of dicarbonyl compounds [156]. In vitro experiments with endothelial cells grown in hyperglycemic (10 mmol) conditions have shown that co-incubating the cells with metformin (20 mmol) can inhibit the formation of NAD(P)H oxidase, and therefore, decrease production of hydrogen peroxide [157]. Co-incubation with metformin also leads to an increase in the activity of catalase in both euglycemic and hyperglycemic conditions [157].
Alpha-glucosidase inhibitors slow the digestion of starch in the small intestine so that glucose can enter the bloodstream more slowly. Alpha-glucosidase inhibitors in use include acarbose and miglitol. To date, no antioxidant effects have been reported in these drugs.
Meglitinides, including repaglinide and nateglinide, bind to a K ATP channel on the cell membrane of pancreatic βcells in a similar manner to sulfonylureas but at a separate binding site. Meglitinides have not been shown to possess any antioxidant properties.
Statins act by competitively inhibiting HMG-CoA reductase and decreasing cholesterol synthesis in the liver. By inhibiting Rac isoprenylation, statins can lead to a reduction in NAD(P)H oxidase and generation of reactive oxygen species [158,159]. In an animal study, simvastatin was able to decrease the formation of superoxide by inhibiting Rac1, a signaling protein involved in cell growth, cell cycle, cell-cell adhesion, and the activation of protein kinases [160].
Pleiotropic effects of diabetic polypharmacy and their possible confounding effects on observational and supplementation trials of antioxidants In vitro, animal and human studies have been conducted on the pleiotropic, and in particular, the antioxidant effects of statins and hypoglycemic drugs. Tables 1 and 2 summarize the data on statins and hypoglycemic drugs, concentrating on studies which have looked at changes in antioxidant/oxidative stress balance. From the controlled studies of statin ingestion with antioxidant vitamins, some insight into whether these pharmaceuticals are co-operators, confounders, or concealers of oral antioxidants taken in supplement form is offered.

Conclusions
Evidence that Type 2 DM subjects are in a state of oxidative stress is unequivocal. Short-term supplementation with antioxidant vitamins C and/or E has not further decreased oxidative stress markers in Type 2 DM subjects. That is not to say that antioxidant vitamin supplementation has no role in Type 2 DM, but the effects of these antioxidants may be concealed or confounded by the cocktail of medications already being taken by the patient, and pharmacological agents themselves may have strong antioxidant effects. Further study is required to ascertain if oral antioxidant vitamin supplementation can be a cheap and safe method which can benefit those Type 2 DM subjects whose diabetes, hypertension, and hypercholesterolemia is being controlled by diet alone.