The Impact of Glycemic Status and

Original Article
Metformin Administration on Red Blood
Cell Indices and Oxidative Stress in Type 2
Diabetic Patients
Adel Abdel-Moneim1, Eman Salah Abdel-Reheim1,
Margit Semmler2, Wessam Addaleel1

Submitted: 6 Dec 2018 1

Physiology Division, Zoology Department, Faculty of Science, Beni-Suef
Accepted: 3 May 2019 University, Egypt
Online:  29 Aug 2019
2
Diabetes Research Institute, Düsseldorf University, Düsseldorf, Germany

To cite this article: Abdel-Moneim A, Abdel-Reheim ES, Semmler M, Addaleel W. The impact of glycemic
status and metformin administration on red blood cell indices and oxidative stress in type 2 diabetic patients.
Malays J Med Sci. 2019;26(4):47–60. https://doi.org/10.21315/mjms2019.26.4.6

To link to this article: https://doi.org/10.21315/mjms2019.26.4.6

Abstract
Background: Most guidelines all over the world recommended metformin as the first-line
treatment for in type 2 diabetic patients. Therefore, the present study was suggested to assess the
outcome of metformin administration and glycemic status on alterations in red blood cell (RBCs)
indices as well as the oxidative stress in type 2 diabetic patients.
Methods: Between December 2016 and October of 2017, a total of 158 eligible individuals
were classified as 50 healthy subjects and 108 diabetic patients who were subdivided into six
groups according to the type of anti-diabetic treatments.
Results: Overall, the results elucidated that hemoglobin concentration was markedly
diminished, while red cell distribution width (RDW) value was significantly (P < 0.001) elevated
in all diabetic groups as compared to control. Moreover, in all diabetic groups, malondialdehyde
(MDA) concentration was elevated noticeably (P < 0.001), while reduced glutathione (GSH)
revealed a lower concentration (P < 0.001) than that of control.
Conclusion: The present study exhibited the amelioration effect of metformin
administration on oxidative stress and glycemic status which reflected on some RBCs indices.
However, hemoglobin concentration showed a noticeable diminution in all metformin-treated
groups in spite of the improvement in glycemic and oxidative stress status which indicated that the
metformin-induced anemia is independently from diabetic complications.

Keywords: red blood cell indices, glycemic status, metformin, oxidative stress

Introduction Currently, the orally employed anti-

diabetic agents: biguanides (e.g., metformin),
Diabetes is a worldwide health problem sulfonylureas (e.g., glimepiride), α-glycosidase
with a close link exists between diabetes and inhibitors (e.g., acarbose) and thiazolidinediones
cardiovascular disease (CVD), whereas CVD (e.g., pioglitazone), act to regulate a specific
is the most prevalent cause of mortality and pathological pathway (3). Metformin has
morbidity among the diabetic populations been utilised for 50 years and approved by the
(1). The prevalence of type 2 diabetes mellitus US Food and Drug Administration (FDA) in
(T2DM) has been globally growing rapidly. 1994. Moreover, most guidelines all over the
Approximately 592 million people worldwide world have recommended metformin as the
will become diabetic by the year 2035, with a first-line treatment for T2DM patients (4).
universal prevalence of 10.1% (2). Glimepiride is a second-generation sulfonylurea

Malays J Med Sci. Jul–Aug 2019; 26(4): 47–60

www.mjms.usm.my © Penerbit Universiti Sains Malaysia, 2019
This work is licensed under the terms of the Creative Commons Attribution (CC BY) 47
(http://creativecommons.org/licenses/by/4.0/).
Malays J Med Sci. Jul–Aug 2019; 26(4): 47–60

that stimulates pancreatic β-cells to release its hypoglycemic effect can reduce the oxidation
insulin and it has been shown to act through status and diabetic complications. Consequently,
additional extrapancreatic mechanisms. In the goal of our study was to evaluate the impact
addition, metformin is preferentially selected for of metformin administration (mono- and
combination therapy with sulfonylurea or insulin dual therapy) and glycemic control on RBC
to achieve the glycemic target in patients who indices alterations and oxidative stress status in
are not satisfactorily controlled by monotherapy Egyptian patients with T2DM.
alone (5).
Oxidative stress and inflammation Subjects and Methods
contribute to the development of diabetic
complications. Intracellular hyperglycemia
Patients Inclusion and Exclusion Criteria
promotes the production of mitochondrial
reactive oxygen species (ROS). ROS is directly The study was conducted on 158 individuals
increase the expression of inflammatory and of both sexes (83 males and 75 females), aged
adhesion factors, the formation of oxidised-low between 30 and 75 years. These individuals
density lipoprotein, and insulin resistance (6). enrolled in this study were classified into 50
Depletion of blood glutathione (GSH) has been normal healthy subjects and 108 T2DM patients
recorded in many clinical issues including T2DM who were fol­ lowed up at the Diabetic Section
(7). Kumawat et al. revealed that hyperglycemia of General Institution of Healthy Insurance,
induces oxidative stress through increasing Egypt, between December 2016 to October
the malondialdehyde (MDA) levels in diabetic 2017. Written agreements were obtained from
patients (8). Moreover, oxidative stress produced all patients before participation in the study.
from an imbalance between free radicals and the The present study was performed in accordance
body’s antioxidant defense systems leads to red with the declaration of Helsinki and good clinical
blood cell dysfunction and tissue injury (9). practice guidelines and also approved by the
Recently, a great attention re-given to the committee of General Institutions of Health
hematological indices as predictors of endothelial Insurance.
dysfunction and inflammation status (10). Enrolled patients were allocated to normal
Anemia is twice in diabetic patients, and mild healthy subjects (control) who had no previous
anemia also has been recorded in diabetics history of chronic diseases and free of type I or
with normal kidney function (11). In addition, 2 diabetes, and 108 patients diagnosed as T2DM
anemia is associated with duration of disease according to WHO 1999 criteria. Pregnant and
and microvascular complications; diabetic lactating women, patients receiving immuno-
neuropathy, nephropathy, retinopathy and CVD modulatory drugs and patients with medical
(12). Red blood cells (RBCs) count is a pivotal conditions such as infections, cerebrovascular
marker for the ability to recognise diabetic diseases, ischemic heart disease, malignancies,
patients at risk of microvascular complications. autoimmune disorders, eczema, respiratory
Lower RBCs counts are, therefore, considered disorder, thyroid dysfunction, kidney failure,
as an independent predictor biomarker of the liver dysfunction and alcohol abuse were
risk of microvascular complications in T2DM excluded from the study. In addition, diabetic
patients (13). Additionally, erythrocytes of patients who underwent medication changes
diabetic patients aggregate more readily that is during the 2 months preceding participation
obviously enhance whole-blood viscosity (WBV), were also excluded.
and adversely influence the microcirculation and
Study Design
finely leading to microangiopathy (14).
Among RBCs indices, the red blood cell The diabetic patients were subdivided
distribution width (RDW) is a simple and into six groups according to the treatment
non-expensive parameter, which can reflect administration.
the level of erythrocyte volume heterogeneity 1. Group 1: normal healthy subjects (control)
(anisocytosis) and is commonly used in the (50 subjects)
laboratory for differential analysis of anemia
(15). Furthermore, RDW values showed a 2. Group 2: diabetics (recent diagnoses) non-
significant increase in T2DM patients than in treated (20 patients)
control subjects (16). Thus, the ideal therapy 3. Group 3: diabetics treated with metformin
for diabetes would be a medication that besides only (15 patients)

48 www.mjms.usm.my
 Original Article | Effect of metformin on erythrocytes in diabetes

4. Group 4: diabetics treated with glimepiride hemoglobin (Hb), hematocrit (PCV), mean
only (19 patients) corpuscular volume (MCV), mean corpuscular
5. Group 5: diabetics treated with metformin + hemoglobin (MCH), mean corpuscular
glimepiride (15 patients) hemoglobin concentration (MCHC), RDW
values and RBCs count were determined using
6. Group 6: diabetics treated with insulin only a MICROS ABX auto-analyser according to the
(20 patients) manufacturer's protocol.
7. Group 7: diabetics treated with metformin +
insulin (19 patients) Statistical Analysis
The data were analysed by one-way
The demographic data regarding analysis of variance (ANOVA). To compare
anthropometric variables such as height, weight, the difference between the groups, post hoc
body mass index (BMI), gender, duration of testing was performed by the Duncan test with
disease and blood pressure (BP) were collected. least significant difference (LSD). Pearson’s
Blood samples were taken from participants after correlation coefficient analysis was used to
overnight fasting in EDTA and plain tubes (4 mL determine the correlations between different
each). Following an incubation period of 30 min studied parameters. Statistical analysis was
at room temperature, blood in plain tubes was performed using the Statistical Package for the
centrifuged at 4,000x g for serum isolation. Sera Social Science (SPSS) for Windows (version 22.0,
were rapidly separated, aliquoted and stored at Chicago, IL, USA). Data are expressed as mean
−40 °C until the biochemical measurements. (SD). Values with P < 0.05 were considered
The second tube blood sample was taken on statistically significant.
potassium fluoride for immediate glucose
estimation. The third part of the blood sample Results
was taken on EDTA for determination of HbA1c
levels and complete blood picture at the same The current results revealed that family
time, then, the blood was centrifuged, the plasma history, BMI, total cholesterol, LDL-c, and
and leucocytes layer were then removed and HbA1c levels were significant elevated in recently
the packed erythrocyte sediments were washed diagnosed diabetic patients as well as in all
three times with normal saline and hemolyzed treated diabetic groups compared to the healthy
by adding approximately 1.5 volumes of ice-cold control (Table 1). In addition, MCV, MCH and
distilled water. The stock hemolysate stored at MCHC values showed non-significant changes in
−40  °C for estimation of MDA and GSH levels. all studied groups compared to healthy control.
Fresh morning urine samples were obtained A non-significant lowering in RBCs count was
for the measurement of microalbuminuria found in all diabetic groups except patients
concentration. treated with metformin plus insulin [4.89(0.7)
vis 4.30(0.5)]. In addition, HCT% revealed
Laboratory Assays
a significant decrease in diabetic patients
The concentrations of glucose, creatinine, treated with glimepiride therapies compared
total cholesterol, triglyceride and HDL-c were to the healthy control group. Moreover, Hb
estimated using commercially available assay concentration was reduced markedly in different
kits obtained from Reactivos Spinrect, Spain. diabetic groups, while RDW values had elevated
low density lipoprotein (LDL)-cholesterol level significantly (P < 0.001) in all diabetic groups
was calculated according to Friendewald formula as compared to the healthy control [Figure
(17). The anti-atherogenic factor indices were 1(A)]. While, urinary microalbuminuria was
calculated using Ross formula (18). HbA1c % significantly elevated (P < 0.001) in the treated
was determined using reagent kits purchased diabetic groups when compared to the healthy
from Spectrum. Microalbuminuria in the urine and recent-diabetic groups [Figure 1(B)].
was estimated by reagent kits purchased from However, lipid peroxidation biomarker (MDA)
Bio-Diagnostic. Moreover, reduced glutathione was elevated markedly (P < 0.001) in all diabetic
(GSH) and lipid peroxidation (determined groups, while the antioxidant GSH level was
indirectly by measuring MDA) were assayed in lowered noticeably (P < 0.001) in different
the blood hemolysate using colorimetric kits diabetic groups when compared to the healthy
(BioVision, Milpitas, CA, USA) according to the control ones [Figures 1(C) and 1(D)].
kit instructions provided. Hematology profile;

www.mjms.usm.my 49
Malays J Med Sci. Jul–Aug 2019; 26(4): 47–60

Table 1. Demographic, biochemical and RBC indices in control, recent diabetic and treated diabetic
groups
Control R Diabetic Metformin Glimepiride Metf +Sulf Insulin Metf+Ins
N= 50 20 15 19 15 20 19
Gender
N(%)
M 25 (50)a 10 (50)c 8 (53) c 9 (47)c 8 (53)c 10(52)c 13(65)b
F 25 (50)a 10 (50)b 7(47)c 10(53)b 7(47)c 9(47)b 7(35)c
F. hist. (0)a
4(20) b
6 (40) bc
5(26) b
4(27) b
11(58) d
8(40)c
Duration (y) 0.0(0) a
0.0(0) a
4.4(3) b
5.8(4) b
6.2(5) b
16(8) c
14(7) c
M(SD)
Age (y) 42(16) a 49(14) a 57(8) bc 63(11) c 55(9) bc 57(8) bc 62(7) c
BMI 27(7) a
36(5) b
33(7) ab
34(4) ab
32(8) ab
33(10) ab
29(4) a
HbA1c% 5.11(0.51)a 9.78(1.46)d 7.31(1.32)b 7.55(1.19)bc 9.80(2.59)d 7.58(1.94)bc 8.51(2.18)c
Creatinine (mg/dL) 1.03(0.23) abc
0.96(0.25) abc
0.83(0.18) a
1.15(0.45) bc
1.04(0.25) abc
1.22(0.89) c
0.90(0.24) ab
TC (mg/dL) 168(28) a
218(48) bcd
191(51) ab
197(55) bc
227(41) d
194(24) abc
220(42)cd
LDL (mg/dL) 99(38) a 151(45) bc 124(49) ab 136(45) bc 165(39) d 128(23) bc 154(39) cd
Hb (g/dL) 14.30(1.0) c
12.27(1.9) a
13.22(0.76) b
12.65(1.3) a
12.71(0.5) a
12.97(1.3) a
12.14(1.1) a
RBCs (x10 /mm )
6 3
4.89(0.7) bc
4.56(0.7) ab
4.73(0.3) bc
4.50(0.4) ab
4.55(0.4) ab
4.74(0.5) bc
4.30(0.5) a
HCT (%) 40.53(3) b 37.36(6) ab 39.18(2) ab 36.65(2) a 38.49(2) ab 39.20(5) ab 37.13(2) ab
MCV (fl) 83.27(2) ab
81.98(7) a
82.88(5) ab
82.14(8) ab
85.05(4) b
82.65(6) ab
87.15(7) b
MCH (Pg) 28.15(2) 27.08(2) 27.25(2) 27.28(3) 28.18(2) 27.48(2) 28.25(3)
MCHC (g/dL) 33.85(2) b 32.90(2) ab 31.81(3) a 33.16(1) b 33.07(1) b 32.86(1) ab 32.72(1) ab

Data are expressed as mean = SD. Values which share the same superscript symbol are not significantly different. R = recent, Metf = metformin,
Sulf = sulfonylurea (glimepiride), Ins = insulin, F hist = family history, BMI = body mass index, HbA1c = glycosylated hemoglobin,
TC = total cholesterol, LDL = low density lipoprotein, MCV = mean corpuscular volume, MCH = mean corpuscular hemoglobin, MCHC = mean
corpuscular hemoglobin concentration

Overall, Hb showed a negative significant Moreover, HCT value observed a negative

correlation with HbA1c, microalbuminuria, correlation between HbA1c, microalbuminuria
and MDA in patients treated with metformin and MDA with metformin monotherapy and
monotherapy [r; -0.778 (95%CI: -0.937, combined therapies of metformin. On the other
-0.612), r; -0.709 (95% CI: -0.925, -0.475), r; hand, HCT value had a positive significant
-0.906 (95% CI: -0.971, -0.818), respectively] correlation between GSH with mono-and dual
and dual therapy groups, however, HB observed metformin therapies with a sulphonylurea and
a positive significant correlation with GSH insulin [0.925 (95% CI: 0.831, 0.975), 0.988
in all metformin therapy groups [Table 2, (95% CI: 0.974, 0.997) 0.918 (95% CI: 0.735,
Figures 2(A), 2(B), 3(A), 3(B), 5(B) and 5(D)]. 0.991), respectively] [Table 2, Figures 2(C), 2(D),
Moreover, RBCs count revealed a significant 4(C) and 4(D)]. RDW level showed a significant
negative correlation with HbA1c, MDA and positive correlation between HbA1c, MDA and
microalbuminuria in metformin therapies microalbuminuria with metformin monotherapy
groups. While, RBCs count induced a positive (r; 0.536 (95% CI: 0.082, 0.885), 0.885
significant correlation with GSH in metformin (95%CI: 0.596, 0.973), 0.880 (95%CI: 0.563,
monotherapy as well as combination therapy 0.971), respectively) and dual therapies, while,
groups with a sulphonylurea (glimepiride) it correlated negatively with GSH in all tested
and insulin (r; 0.960 (95% CI: 0.935, 0.982, groups [Table 2, Figures 2(C), 2(D), 3(C), 3(D),
0.824(95% CI: 0.661, 0.936), respectively) 5(A) and 5(C)].
[Table 2, Figures 2(A), 2(B), 4(A) and 4(B)].

50 www.mjms.usm.my
 Original Article | Effect of metformin on erythrocytes in diabetes

(A) RDW (B) Microalbuminuria

(C) MDA (D) GSH

Figure 1. Values of RDW% (A), microalbuminuria (B), MDA (C) and GSH (D) of control, recent
diabetic and different metformin treated groups
*
Significance compared to normal control, *P < 0.05, **P < 0.01, ***P < 0.001, #significance compared to the recent
diabetic. #P < 0.05, ##P < 0.01, ###P < 0.001. Metf = metformin, R diabetic = recent diabetic, Sulf = sulfonylurea
(glimepiride), Ins = insulin,

Table 2.  Pearson correlation of HbA1c, MDA, GSH and microalbuminuria with RBCs indices in
different treated groups
HbA1c Microalbuminuria MDA GSH

r 95% CI r 95% CI r 95% CI r 95% CI

Metf
Hb −0.778** (−0.937, −0.612) −0.709** (−0.925, −0.475) −0.906** (−0.971, −0.818) 0.884*** (0.802, 0.957)
RBCs −0.621* (−0.905, −0.291) −0.798** (−0.937, −0.537) −0.867** (−0.951, −0.640) 0.846** (0.579, 0.932)
HCT −0.713** (−0.953, −0.496) −0.803** (−0.968, −0.580) −0.957** (−0.987, −0.914) 0.925** (0.831, 0.975)
RDW 0.536* (0.082, 0.885) 0.885** (0.596, 0.973) 0.880** (0.563, 0.971) −0.882** (−0.962, −0.532)
Metf + Sulf
Hb −0.838** (−0.934, −0.562) −0.899** (−0.959, −0.754) −0.961** (−0.985, −0.917) 0.961** (0.906, 0.990)
RBCs −0.774** (−0.909, −0.563) −0.882** (−0.954, −0.799) −0.885** (−0.973, −0.795) 0.960** (0.935, 0.982)
HCT −0.790** (−0.926, −0.547) −0.904** (−0.970, −0.812) −0.917** (−0.971, −0.862) 0.988** (0.974, 0.997)
RDW 0.888** (0.630, 0.965) 0.955** (0.867, 0.986) 0.972** (0.937, 0.991) −0.978** (−0.991, −0.961)
Metf + Ins
Hb −0.704** (−0.192, −0.894) −0.681** (−0.431, −0.953) −0.831** (−0.394, −0.989) 0.862** (0.973, 0.568)
RBCs −0.687** (−0.855, −0.460) −0.578** (−0.795, −0.398) −0.721** (−0.885, −0.382) 0.824** (0.661, 0.936)
HCT −0.688** (−0.875, −0.170) −0.665** (−0.898, −0.364) −0.791** (−0.972, −0.325) 0.918** (0.735, 0.991)
RDW 0.853** (0.092, 0.950) 0.895** (0.313, 0.968) 0.865** (0.469, 0.957) −0.862** (−0.949, −0.549)

*Correlation is significant at the 0.05 level. **Correlation is significant at the 0.01 level. ***Correlation is significant at the 0.001 level.
Metf = metformin, Sulf = sulfonylurea (glimepiride), Ins = insulin, HbA1c = glycosylated hemoglobin

www.mjms.usm.my 51
Malays J Med Sci. Jul–Aug 2019; 26(4): 47–60

(A) HbA1c% (B) Microalbuminuria

(C) MDA (D) GSH

Figure 2. Correlation coefficient of Hb, RBCs, HCT and RDW values with HbA1c% (A),
microalbuminuria (B), MDA (C) and GSH (D)
*Correlation is significant at the 0.05 level. **Correlation is significant at the 0.01 level. ***Correlation is significant
at the 0.001 level. R = recent, Metf = metformin, R. diabetic = recent diabetic, Sulf = sulfonylurea (glimepiride),
Ins = insulin, HbA1c = glycosylated hemoglobin, r = correlation coefficient

52 www.mjms.usm.my
 Original Article | Effect of metformin on erythrocytes in diabetes

(A) (B)

15 r = −0.778, 15 r = −0.709,
P < 0.001 P < 0.010
14.5 14.5

14 14

13.5 13.5

Hb
Hb

13 13

12.5 12.5

12 12

11.5 11.5
5 7 9 11 0 200 400 600
HbA1c% Microalbuminuria

(C) (D)

15 r = 0.880, 15 r = −0.882,
P < 0.001 P < 0.001

14.5 14.5

14 14
RDW

RDW

13.5 13.5

13 13

12.5 12.5
10 30 50 70 30 35 40 45 50
MDA GSH

Figure 3. Correlations of Hg% with HbA1c% (A) and microalbuminuria (B), and RDW with MDA (C)
and GSH (D) among metformin monotherapy group
HbA1c = glycosylated hemoglobin, r = correlation coefficient

www.mjms.usm.my 53
Malays J Med Sci. Jul–Aug 2019; 26(4): 47–60

(A) (B)

r = −0.885, r = 0.960,
5.5 P < 0.010 5.5 P < 0.010

5 5
RBCs

RBCs
4.5 4.5

4 4
15 25 35 45 55 25 35 45 55 65
MDA GSH

(C) (D)

44 44 r = −0.904,
r = −0.790,
P < 0.010
P < 0.010

41 41
HCT

HCT

38 38

35 35
5 10 15 0 50 100 150 200
HbA1c Microalbuminuria

Figure 4. Correlations between RBCs with MDA (A) and GSH (B), and HCT with HbA1c (C) and
microalbuminuria (D) among metformin plus glimepiride group
HbA1c = glycosylated hemoglobin, r = correlation coefficient

54 www.mjms.usm.my
 Original Article | Effect of metformin on erythrocytes in diabetes

(A) (B)

16 r = 0.853, r = −0.681,
P < 0.001 P < 0.001
15.5 14

15
13
14.5
RDW

Hb
14 12

13.5
11
13

12.5 10
4 7 10 55 0 100 200 300
HbA1c% Microalbuminuria

(C) (D)

16 r = 0.865, r = 0.862,
P < 0.010 P < 0.001
15.5 14

15
13
14.5
RDW

Hb

14 12

13.5
11
13

12.5 10
10 20 30 40 50 60 30 40 50 60
MDA Microalbuminuria

Figure 5. Correlations of RDW with HbA1c% (A) and MDA (C), and Hb% with microalbuminuria (B),
and GSH (D) among metformin plus insulin patients
HbA1c = glycosylated hemoglobin, r = correlation coefficient

www.mjms.usm.my 55
Malays J Med Sci. Jul–Aug 2019; 26(4): 47–60

Discussion and anemia in patients with T2DM (27, 28). It

is reported that long-term use of metformin was
Overall, our results confirmed that Hb associated with biochemical B12 deficiency and
concentration was lowered in all diabetic groups, anemia (28). Since vitamin B12 is essential to
while RDW was elevated significantly in diabetic nutrition, the metformin-induced B12 reduction
patients. Also, the erythrocytes (RBCs) count may have detrimental effects in patients with
revealed a significant decrease in patients treated T2DM (29). Moreover, Kang et al. recommended
with metformin plus insulin. The decrease in the need for regular vitamin B12 monitoring
RBCs counts in diabetic patients may be due to in patients with T2DM, particularly patients
RBC membrane protein alterations, a decrease in receiving a higher daily dosage of sulfonylurea
hemoglobin levels and erythropoietin deficiency plus metformin treatment for a long time period
(13). Hemoglobin is the major component (27).
of erythrocytes; thus, when the HbA1c level The main pathophysiological consequences
is elevated, hyperglycemia may increase the of free radical-induced lipid peroxidation are
β-sheet structure content of Hb causing it disturbing the assembly of cell membranes,
to aggregate which subsequently increases which inevitably will impact membrane
WBV (19). In fact, the excessive aggregation fluidity, and lipid-protein interaction dynamics,
of RBC is one of the most prominent features membrane permeability, and physicochemical
in patients with diabetes with poor glycemic properties (30). Moreover, in this study, MDA
control. The erythrocyte aggregation is an concentration was elevated markedly in all
important hemorheological parameter because diabetic groups, while, the GSH level was
it directly affects WBV (20). Also, some recent lowered significantly in diabetic subjects. In
epidemiological studies have reported that addition, RBCs count, Hb and HCT values
T2DM is characterised by increased erythrocyte showed a negative correlation between HbA1c,
osmotic fragility (21). In diabetes, erythrocyte MDA and microalbuminuria with metformin
membranes are affected by the chronic exposure therapies. However, a positive correlation
to glucose, and several biochemical modifications with GSH in all diabetic groups was observed.
are triggered, with subsequent structural and Hyperlipidemia and increased lipid peroxidation
functional disruption of erythrocytes and were strongly associated with increased systemic
decreased the lifespan of RBCs (22) which may inflammation, and as well as the increased
be related to RBCs count dimension. Meanwhile, MDA production has also been recorded in the
RBC properties are critically affected by erythrocyte membrane of diabetic patients (8).
hyperglycemia and decreased deformability (23). Additionally, in parallel with our findings, De
Anemia is a highly prevalent condition Souza Bastos et al. suggested that diabetes was
in people with T2DM. The causes of associated with dyslipidemia and increased
diabetic anemia are multifactorial including lipid peroxidation (31). Moreover, Lee et al.
inflammation, concomitant autoimmune concluded that the high level of lipids may alter
diseases, antidiabetic medication, hormonal the morphology and flow behaviour of RBCs,
changes and kidney diseases (12). The which could contribute to the impairment of
manifestation of anemia in diabetics has been microcirculatory-related disorders (32).
attributed to the increase of non-enzymatic In addition, in accordance with the current
glycosylation of RBC membrane proteins, which study, several findings showed that patients with
was correlated with hyperglycemia (24). The T2DM had lower GSH content in erythrocytes
protein oxidation and hyperglycemia in diabetics than that in the control group. Also, a decrease
induced an elevation in the production of lipid in intracellular glutathione level in patients with
peroxides that may lead to RBCs hemolysis T2DM was reported especially in the presence
(25) and subsequently decrease in RBCs of microvascular complications (33), which
count and hemoglobin levels. Moreover, the was depend on the degree of hyperglycemia.
diabetic and anemic patients had high levels of Furthermore, patients with T2DM have a
C-reactive protein and ferritin ultra-sensible lower concentration of intracellular GSH,
with low iron contents which might refer to which increase the susceptibility of blood cells
association between ferritin increase and chronic to the damaging effects of ROS. Waggiallah
inflammatory process in diabetics (26). However, and Alzohairy found a significant impact of
long-term treatment with metformin is known oxidative stress (low glutathione) on glutathione
to be associated with vitamin B12 deficiency peroxidase which could reduce Hb concentration

56 www.mjms.usm.my
 Original Article | Effect of metformin on erythrocytes in diabetes

in diabetic patients, which means that oxidative associated with microalbuminuria in

stress of diabetes is one of the causes of anemia patients with T2DM (42). Some investigators
in diabetics independently from diabetic attributed the association between RDW and
nephropathy (34). microalbuminuria to the presence of chronic
In accordance with our results, the inflammation in diabetics (43).
RDW values were significantly higher in When fructose-fed rats were treated with
diabetic patients than that of healthy subjects metformin, the imbalance between peroxidation
(35). Furthermore, the RDW level showed a and antioxidants defense system was mitigated
positive correlation with HbA1c, MDA and (44). However, treatment with glimepiride and
microalbuminuria in all metformin administered vildagliptin improved erythrocyte deformability
therapies. The exact pathophysiological in patients with T2DM, the improvement seemed
mechanism underlying the association between to be correlated with improved glycemic control
RDW and diabetes still unknown, however, it (45). Moreover, it was observed that patients of
was established that inflammation and oxidative good glycemic control have lower RDW than the
stress can alter erythrocyte homeostasis and patients of poor control (35). On the other hand,
increase RDW values. Although the specific metformin or glibenclamide might ameliorate
mechanisms between the RDW and adverse oxidative stress in the kidneys of diabetic rats to
health outcomes were fully unclear, it was a certain extent with regard to SOD and MDA
suggested that it could be related to increased (46). Tessier et al. reported that the improved
oxidative stress and inflammatory cytokines glycemic control with gliclazide and metformin
(36). Inflammation inhibits bone marrow therapy was associated with improvement in
function and iron metabolism, and pro- the antioxidant/lipid peroxidation status (47).
inflammatory cytokines have been proven to In addition, Chukwunonso-Obi et al. found that
inhibit erythropoietin-induced maturation and administration of metformin, glibenclamide,
proliferation of erythrocytes. All these factors and repaglinide exhibited a significant reduction
may contribute to the increase in RDW% (37). in MDA concentration and considerable
Additionally, high RDW indicated impairment of improvement in the altered activities of
erythropoiesis, reflecting chronic inflammation antioxidant enzymes (48).
and increased level of oxidative stress, both of
which were significant signs of T2DM (38). As Conclusion
we all know, diabetes mellitus is considered
as a chronic inflammatory disease (39). Thus,
The data indicated that metformin
the achievement of the glycemic remission
administration can induce amelioration in
might result from a lower level of inflammation
hyperglycemia and oxidative stress as well as
and oxidative stress which was indicated by
inflammation status regarded to MDA, GSH and
lowering the RDW level. Increased RDW value
RDW levels. However, Hb concentration showed
was therefore reflect the significant deregulation
a reduction in metformin-treated groups in
of erythrocyte homeostasis. This deregulation
spite of improvements in glycemic and oxidative
induced chronic inflammation, the elevation of
status. This mean that the metformin-induced
oxidative stress, erythrocyte fragmentation, poor
anemia is independent of diabetes.
nutritional status, hypertension, dyslipidemia,
impairment of erythropoiesis, and erythropoietin
dysfunction (15). Acknowledgements
Microalbuminuria, the best sign for
diabetic nephropathy development in None.
T2DM, was also, accepted as an indicator
of diabetic microangiopathy (40). Urinary Conflict of Interest
microalbuminuria, in this study, was significantly
increased in all treated diabetic groups. The None.
erythropoietin dysregulation, caused by early
damage to renal tubules, has been suggested
as one of the contributors to anemia in patients Funds
with diabetes (41). Magri and Fava found that
RDW was strongly associated with diabetic None.
nephropathy and may be independently

www.mjms.usm.my 57
Malays J Med Sci. Jul–Aug 2019; 26(4): 47–60

Authors’ Contributions 6. Yuan T, Yang T, Chen H, Fu D, Hu Y, Wang

J, et al. New insights into oxidative stress
Conception and design: AAM, ESA-R, MS and inflammation during diabetes mellitus-
Analysis and interpretation of the data: WA accelerated atherosclerosis. Redox Biol.
Drafting of the article: AAM, ESA-R 2018;20:247–260. https://doi.org/10.1016/j.
Critical revision of the article for important intellectual redox.2018.09.025
content: AAM
Final approval of the article: AAM, ESA-R, MS, WA 7. Gawlik K, Naskalski JW, Fedak D, Pawlica-
Provision of study materials or patients: WA Gosiewska D, Grudzień U, Dumnicka P,
Statistical expertise: WA et al., Markers of antioxidant defense in
Administrative, technical, or logistic support: ESA-R patients with type 2 diabetes. Oxid Med Cell
Collection and assembly of data: WA Longev. 2015;2016:2352361. https://doi.
org/10.1155/2016/2352361
Correspondence
8. Kumawat M, Sharma TK, Singh I, Singh N,
Professor Dr Adel Abdel Moneim Ghalaut VS, Vardey SK, et al. Antioxidant
PhD (University of Cairo), MSc (University of Cairo) enzymes and lipid peroxidation in type 2 diabetes
Molecular Physiology Division, mellitus patients with and without nephropathy.
Faculty of Science, Beni-Suef University, N Am J Med Sci. 2013;5(3):213–219. https://doi.
Salah Salem St, 62511, Beni Suef, Egypt. org/10.4103/1947-2714.109193
Tel: +2 01006 5451764
Fax: +2 082 2356845 9. Comazzi S, Spagnolo V, Bonfanti U. Erythrocyte
E-mail: adel_men2020@yahoo.com & adel.hassan@ changes in canine diabetes mellitus: in vitro
science.bsu.edu.eg effects of hyperglycaemia and ketoacidosis. Comp
Clin Pathol. 2004;12(4):199–205. https://doi.
References org/10.1007/s00580-004-0502-x

10. Demirtas L, Degirmenci H, Akbas EM, Ozcicek

1. Matheus AS, Tannus LR, Cobas RA, Palma CC,
A, Timuroglu A, Gurel A, et al. Association
Negrato CA, Gomes MB. Impact of diabetes
of hematological indices with diabetes,
on cardiovascular disease: an update. Int J
impaired glucose regulation and microvascular
Hypertens. 2013;2013:653789. https://doi.
complications of diabetes. Int J Clin Exp Med.
org/10.1155/2013/653789
2015;8(7):11420–11427.
2. Guariguata L, Whiting DR, Hambleton I,
11. Barbieri J, Fontela PC, Winkelmann ER,
Beagley J, Linnenkamp U, Shaw JE. Global
Zimmermann CE, Sandri YP, Mallet EK, et al.
estimates of diabetes prevalence for 2013 and
anemia in patients with type 2 diabetes mellitus.
projections for 2035. Diabetes Res Clin Pract.
Anemia. 2015;2015:354737. https://doi.
2014;103(2):137–149. https://doi.org/10.1016/j.
org/10.1155/2015/354737
diabres.2013.11.002
12. Hosseini MS, Rostami Z, Saadat A, Saadatmand
3. American diabetic association (ADA).
SM, Naeimi E. Anemia and microvascular
Standards of medical care in diabetes—2012,
complications in patients with type 2 diabetes
American Diabetes Association. Diabetes
mellitus.  Nephrourol Mon. 2014;6(4):e19976.
Care. 2012;35(Suppl 1):S11–S63. https://doi.
https://doi.org/10.5812/numonthly.19976
org/10.2337/dc12-s011
13. Wang Z-S, Song Z-C, Bai J-H, Li F, Wu T, Qi
4. Bouchoucha M, Uzzan B, Cohen R. Metformin
J, et al. Red blood cell count as an indicator of
and digestive disorders. Diabetes Metab.
microvascular complications in Chinese patients
2011;37(2):90–96. https://doi.org/10.1016/j.
with type 2 diabetes mellitus. Vasc Health
diabet.2010.11.002
Risk Manag. 2013;9:237–243. https://doi.
5. Basit A, Riaz M, Fawwad A. Glimepiride: org/10.2147/VHRM.S43211
evidence-based facts, trends, and observations
14. Cho YI, Mooney MP, Cho DJ. Hemorheological
(GIFTS). [corrected]. Vasc Health Risk Manag.
disorders in diabetes mellitus. J Diabetes Sci
2012;2012:463–472. https://doi.org/10.2147/
Tech. 2008;2(6):1130–1138. https://doi.
VHRM.S33194
org/10.1177/193229680800200622

58 www.mjms.usm.my
 Original Article | Effect of metformin on erythrocytes in diabetes

15. Salvagno GL, Sanchis-Gomar F, Picanza A, Lippi 25. Shenoy AG, Goyal RK. Improvement of insulin
G. Red blood cell distribution width: a simple sensitivity by perindopril in spontaneously
parameter with multiple clinical applications. Crit hypertensive and streptozotocin-diabetic rats.
Rev Clin Lab Sci.2015;52(2):86–105. https://doi. Indian J Pharmacol. 2002;34:156–164.
org/10.3109/10408363.2014.992064
26. Andrews M, Arredondo M. Ferritin levels
16. Cakir L, Aktas G, Enginyurt O, Cakir SA. Mean and hepcidin mRNA expression in peripheral
platelet volume increases in type 2 diabetes mononuclear cells from anemic type 2 diabetic
mellitus independent of HbA1c level. Acta Medica patients. Biological Trace Element Research.
Mediterranea. 2014;30:425–428. 2012;149(1):1–4. https://doi.org/10.1007/
s12011-012-9389-6
17. Friedewald WT, Levy RI, Fredrickson DS.
Estimation of the concentration of low-density 27. Kang D, Yun J-S, Ko S-H, Lim T-S, Ahn Y-B,
lipoprotein cholesterol in plasma, without use of Park Y-M, et al. Higher prevalence of metformin-
the preparative ultracentrifuge. Clinical Chemist. induced vitamin B12 deficiency in sulfonylurea
1972;18(6):499–502. combination compared with insulin combination
in patients with type 2 diabetes: a cross-
18. Ross R. The pathogenesis of atherosclerosis: sectional study. PLoS One. 2014;9(10):e109878.
a perspective for the 1990s. Nature. https://doi.org/10.1371/journal.pone.0109878
1993;362(6423):801–809. https://doi.
org/10.1038/362801a0 28. Aroda VR, Edelstein SL, Goldberg RB, Knowler
WC, Marcovina SM, Orchard TJ, et al. Long-term
19. Chalew SA, McCarter RJ, Hempe JM. Biological Metformin Use and Vitamin B12 Deficiency in the
variation and hemoglobin A1c: relevance to Diabetes Prevention Program Outcomes Study.
diabetes management and complications. Pediatr J Clin Endocrinol Metab. 2016;101(4):1754–
Diabetes. 2013;14(6):391–398. https://doi. 1761. https://doi.org/10.1210/jc.2015-3754
org/10.1111/pedi.12055
29. Liu Q, Li S, Quan H, Li J. Vitamin B12 status in
20. Kim S, Popel AS, Intaglietta M, Johnson PC. metformin treated patients: systematic review.
Effect of erythrocyte aggregation at normal PLoS One. 2014;9(6):e100379. https://doi.
human levels on functional capillary density in rat org/10.1371/journal.pone.0100379
spinotrapezius muscle. Am J Physiol Heart Circ
Physiol. 2006;290(3):H941–H947. https://doi. 30. Volinsky R, Kinnunen PK. Oxidized
org/10.1152/ajpheart.00645.2005 phosphatidylcholines in membrane-
level cellular signaling from biophysics to
21. Lippi G, Mercadanti M, Aloe R, Targher G. physiology and molecular pathology. FEBS
Erythrocyte mechanical fragility is increased in J. 2013;280(12):2806–2816. https://doi.
patients with type 2 diabetes. Eur J Intern Med. org/10.1111/febs.12247
2012;23(2):150–153. https://doi.org/10.1016/j.
ejim.2011.11.004 31. de Souza Bastos A, Graves DT, de Melo Loureiro
AP, Júnior CR, Corbi SCT, Frizzera F, et al.
22. Gabreanu GR, Angelescu S. Erythrocyte Diabetes and increased lipid peroxidation
membrane in type 2 diabetes mellitus. are associated with systemic inflammation
Discoveries. 2016;4(2):e60. https://doi. even in well-controlled patients. J Diabetes
org/10.15190/d.2016.7 Complications. 2016;30(2):1593–1599. https://
doi.org/10.1016/j.jdiacomp.2016.07.011
23. Agrawal R, Smart T, Cardoso J, Richards C,
Bhatnagar R, Tufail A, et al. Assessment of 32. Lee C-YJ, Kim K-C, Park H-W, Song J-H, Lee
red blood cell deformability in type 2 diabetes C-H. Rheological properties of erythrocytes from
mellitus and diabetic retinopathy by dual male hypercholesterolemia. Microvasc Res.
optical tweezers stretching technique. Sci Rep. 2004;67(2):133–138. https://doi.org/10.1016/j.
2016;6:15873. https://doi.org/10.1038/srep15873 mvr.2003.12.006
24. Oyedemi S, Yakubu M, Afolayan A. Antidiabetic
activities of aqueous leaves extract of Leonotis
leonurus in streptozotocin induced diabetic rats.
J Med Plants Res. 2011;5:119–125.

www.mjms.usm.my 59
Malays J Med Sci. Jul–Aug 2019; 26(4): 47–60

33. Lutchmansingh FK, Hsu JW, Bennett FI, Badaloo 42. Magri CJ, Fava S. Red blood cell distribution
AV, McFarlane-Anderson N, Gordon-Strachan width and diabetes-associated complications.
GM, et al. Glutathione metabolism in type 2 Diabetes Metab Syndr: Clinl Res Rev.
diabetes and its relationship with microvascular 2014;8(1):13–17. https://doi.org/10.1016/j.
complications and glycemia. PLoS One. dsx.2013.10.012
2018;13(6):e0198626. https://doi.org/10.1371/
journal.pone.0198626 43. Zhang M, Zhang Y, Li C, He L. Association
between red blood cell distribution and renal
34. Waggiallah H, Alzohairy M. The effect of oxidative function in patients with untreated type 2
stress on human red cells glutathione peroxidase, diabetes mellitus. Ren Fail. 2015;37(4):659–
glutathione reductase level, and prevalence 663. https://doi.org/10.3109/088602
of anemia among diabetics. N Am J Med Sci. 2X.2015.1010938
2011;3(7):344–347. https://doi.org/10.4297/
najms.2011.3344 44. Srividhya S, Ravichandran M, Anuradha C.
Metformin attenuates blood lipid peroxidation
35. Nada AM. Red cell distribution width in type 2 and potentiates antioxidant defense in
diabetic patients. Diabetes Metab Syndr Obes. high fructose-fed rats. J Biochem Mol Biol
2015;2015:525–533. https://doi.org/10.2147/ Biophys. 2002;6:379–385. https://doi.
DMSO.S85318 org/10.1080/1025814021000024280

36. Lippi G, Targher G, Montagnana M, Salvagno 45. Berndt-Zipfel C, Michelson G, Dworak M, Mitry
GL, Zoppini G, Guidi GC. Relation between M, Löffler A, Pfützner A, et al. Vildagliptin in
red blood cell distribution width and addition to metformin improves retinal blood
inflammatory biomarkers in a large cohort of flow and erythrocyte deformability in patients
unselected outpatients. Arch Pathol Lab Med. with type 2 diabetes mellitus–results from
2009;133:628–632. an exploratory study. Cardiovasc Diabetol.
2013;12:59. https://doi.org/10.1186/1475-2840-
37. Osterholm EA, Georgieff MK. Chronic 12-59
inflammation and iron metabolism. J
Pediatr. 2015;166(6):1351–1357. https://doi. 46. Erejuwa OO, Sulaiman SA, Wahab MSA, Salam
org/10.1016/j.jpeds.2015.01.017 SKN, Salleh MSM, Gurtu S. Comparison of
antioxidant effects of honey, glibenclamide,
38. Dada OA, Uche E, Akinbami A, Odesanya metformin, and their combinations in the
M, John-Olabode S, Adediran A, et al. The kidneys of streptozotocin-induced diabetic rats.
relationship between red blood cell distribution Int J mol Sci. 2011;12(1):829–843. https://doi.
width and blood pressure in patients with type 2 org/10.3390/ijms12010829
diabetes mellitus in Lagos, Nigeria. J Blood Med.
2014;5:185–189. https://doi.org/10.2147/JBM. 47. Tessier D, Maheux P, Khalil A, Fülöp T. Effects of
S67989 gliclazide versus metformin on the clinical profile
and lipid peroxidation markers in type 2 diabetes.
39. Donath MY, Shoelson SE. Type 2 diabetes as Metabolism. 1999;48(7):897–903. https://doi.
an inflammatory disease. Nat Rev Immunol. org/10.1016/S0026-0495(99)90226-3
2011;11:98–107. https://doi.org/10.1038/nri2925
48. Chukwunonso Obi B, Chinwuba Okoye T,
40. Ünübol M, Ayhan M, Güney E. The relationship Okpashi VE, Nonye Igwe C, Olisah Alumanah
between mean platelet volume with E. Comparative study of the antioxidant
microalbuminuria and glycemic control in effects of metformin, glibenclamide, and
patients with type II diabetes mellitus. Platelets. repaglinide in alloxan-induced diabetic rats. J
2012;23(6):475–480. https://doi.org/10.3109/0 Diabetes Res. 2015;2016:1635361. http://doi.
9537104.2011.634934 org/10.1155/2016/1635361
41. Dikow R, Schwenger V, Schömig M, Ritz E.
How should we manage anaemia in patients
with diabetes? Nephrol Dial Transplant.
2002;17(Supl 1):67–72. https://doi.
org/10.1093/ndt/17.suppl_1.67

60 www.mjms.usm.my