Keywords noninvasive - saliva - fructosamine - diabetes mellitus - glycated hemoglobin
Introduction
Diabetes is the most common endocrine disorder across the world. According to the
International Diabetes Federation (IDF), in the year 2014 an estimated 387 million
people suffered from diabetes. There has been an alarming rise in the prevalence of
all types of diabetes, and it has been projected that the number will increase by
55% by the year 2035.[1 ]
India has the world's largest population living with diabetes after China.[2 ] The IDF reported that in the year 2014 about 66.8 million Indians suffered from
diabetes and it may rise to 109 million by the year 2025.[1 ]
Diabetes has an asymptomatic stage that may be present for up to 7 years before it
is diagnosed.[3 ] Type-II diabetes mellitus frequently remains undetected for a long time, about half
of the patients with diabetes present with one or more irreversible complications
at the time of diagnosis.[4 ]
[5 ]
[6 ] Hence early detection and periodic monitoring of blood glucose improve microvascular
outcomes, and help both patients and their physicians assess the responses to therapy.[1 ]
In recent times the performance of the diagnostic test for diabetes has been reviewed
extensively. Although the tests recommended by the American Diabetes Association (ADA)
are widely used, each of them has their pros and cons, hence screening programs include
a combination of these tests.[6 ]
[7 ]
All the conventional diagnostic tools apart from requiring a venous puncture, which
is a major disadvantage, have other limitations. The limitations of fasting plasma
glucose (FPG) testing are overnight fast and a 12% to 15% day-to-day variance in FPG.[3 ]
[5 ] The oral glucose tolerance test is considered a first-line diagnostic test; however,
it has limitations such as poor patient compliance and reproducibility.[8 ]
[9 ]
Maillard reaction is the process by which carbonyl group of reducing sugars such as
glucose react nonenzymatically with amino groups in proteins to form aldimines or
early glycation products (Schiff's base). These early glycation products consequently
undergo Amadori rearrangement and form intermediate glycation products which include
plasma glycated hemoglobin (HbA1C) and plasma fructosamine.[10 ]
The advanced glycated end products (AGE) are formed by classic rearrangement of the
intermediate glycation products. Pentosidine and pyralline are a few AGE products.[10 ] Unlike Amadori products, AGE products affect long-living proteins. Thus, advanced
glycation reflects glycemic control over a long period. However, at present there
is no universally accepted method to detect AGE and no internationally recognized
standard unit of measurement, making the comparison of results between different laboratories
difficult. Hence plasma HbA1C and plasma fructosamine values are widely used to estimate
the glycemic control throughout 1 to 8 weeks.[10 ]
[11 ]
Plasma glycated hemoglobin (HbA1C) is formed by the glycation of valine and lysine
residues of hemoglobin. Similarly, plasma fructosamine is formed by the glycation
of plasma protein albumin. The half-life of hemoglobin is 60 days, and that of albumin
is 14 to 20 days. Thus, plasma HbA1C levels indicate the glycemic control for 6 to
8 weeks, and plasma fructosamine levels indicate the mean blood glucose concentration
over 2 to 3 weeks.[11 ]
Plasma HbA1C measurement has been recently included by the ADA as a diagnostic and
screening tool for diabetes. The chief advantages of using HbA1C measurement is the
ease of testing because it does not require fasting and acute perturbations do not
affect HbA1C levels; further micro-angiopathic complications strongly associate with
HbA1C levels.[12 ] The limitations of HbA1C testing include variation among different races. Also conditions
such as anemia and some medications may influence the results.[7 ]
Currently, plasma fructosamine levels are used to monitor the glycemic control throughout
1 to 3 weeks, as it is a more sensitive marker than plasma HbA1C.[11 ] Plasma fructosamine is unaffected by disorders of red blood cells and medications.[13 ] The limitation of plasma fructosamine test is that the test results must be interpreted
with caution in case of protein-losing disorders such as nephropathy and liver disorder
and it requires venous puncture.[11 ]
[13 ]
Saliva, a valuable biofluid, contains abundant biomarkers, produced by the major and
minor salivary glands, gingival crevicular fluid (GCF), mucosal transudations, serum
and blood sheds from oral wounds, desquamated epithelial cells, acquired pellicles,
bacterial products, viruses and fungi, other cellular components, and food debris.[14 ] Saliva is often referred to as plasma ultrafiltrate, as about 1,000 different proteins
and 19,000 unique peptide sequences have been detected in saliva.[15 ] The mechanism by which the biomarkers enter the saliva includes passive diffusion,
active transport, or extracellular ultrafiltration. As it provides a good reflection
of the physiological functions of the body, human saliva is known as the “mirror of
the body.”[14 ]
[15 ]
In recent times, substantial body of evidence, describes the use of saliva as a liquid
biopsy, for the diagnosis of oral diseases like dental caries, gingivitis, periodontitis
(chronic/aggressive), Bechet disease, oral squamous cell carcinoma, cleft palate and
lips, salivary gland diseases, oral leukoplakia, chronic graft-versus-host disease
(cGVHD), and systematic diseases such as breast cancer, diabetes, human immune deficiency
virus (AIDS).[15 ] Apart from diagnosis of diseases, saliva is also used to assess drug levels, toxicity,
infectious, immunological, and hormonal levels, and in forensic investigations. The
proteomic or genomic macromolecules present in saliva makes it a predictive, diagnostic,
and prognostic tool for carcinomatous, inflammatory, and genetic disorders.[14 ]
[15 ]
[16 ]
Studies have reported that the glycated salivary proteins correlate with plasma glucose
levels and plasma glycated proteins. Thus salivary fructosamine levels have been reported
to correlate with plasma glucose and plasma HbA1C levels.[17 ]
[18 ]
Salivary fructosamine can be used like plasma HbA1C levels to monitor the glycemic
control over a period of 1 to 3 weeks. Since the collection of saliva is simple and
painless, it can be used as a noninvasive adjuvant tool for monitoring the glycemic
control in diabetic patients, thus reducing the necessity of repeated venous punctures
to draw blood.[19 ]
The five different types of oral fluid that can be utilized in the field of salivary
diagnostics are (1) unstimulated whole saliva (UWS), (2) stimulated whole saliva (SWS),
(3) saliva from parotid gland, (4) saliva from submandibular/sublingual gland, and
(5) saliva from minor glands.[20 ] Each of them, in turn, have a specific method of collection and type of collection
device. Salivette device (Sarstedt, Numbrecht, Germany), Quantisal (Immunalysis, Pomona,
California), and SCS (Greiner-BioOne, Kremsmunster, Austria) are the most commonly
used collection devices.[16 ]
[20 ] Groschl et al have suggested that immediate processing of the sample is essential
to avoid errors in the study due to sample instability issues, and added to this the
adsorption behavior of saliva on various devices could negatively affect the sample
integrity.[21 ] Various studies have observed a significant difference in the analyte levels, depending
on the collection method used. This has led to the invention of novel devices such
as SuperSAL and VersiSAL technologies (Oasis Diagnostics Corporation, Vancouver, Washington).
These devices provide a standardized specimen, resulting in a more consistent sample
uniformity. This is made possible by the use an inert absorbent material, as the collection
media, that effectively releases the analyte from the pad, in contrast to the cotton
bases that cause interference with the test results.[16 ]
[20 ]
The various methods used to analyze biomarkers present in saliva are high-performance
liquid chromatography (HPLC), enzyme-linked immunosorbent assay (ELISA), radioimmunoassay,
two-dimensional electrophoresis (2DE), mass spectrometry (MS), 2DE and reverse-phase
liquid chromatography (LC), LC-tandem MS, matrix-assisted laser desorption/ionization
time-off light mass spectrometry (MALDI-TOF MS), and 2DE followed by MALDI-TOF MS.
The method used for analysis depends on the biomarker that is to be analyzed in saliva.[22 ] At present, novel discoveries such as the lab on a chip system and other biosensor
systems have opened new vistas to research in the field of personalized medicine and
dentistry.[22 ]
[23 ]
A substantial body of evidence suggests that Asian Indians are genetically more susceptible
to type II diabetes mellitus, mainly due to the Pro l 2 Ala polymorphism which is
not protective against diabetes or insulin resistance among Asian Indians. Coupled
with the above factors, lack of physical activity and high fat sugar–rich diet has
escalated the prevalence of type II diabetes to alarming levels in India.[24 ]
[25 ]
To the best of our knowledge, there are only a few studies conducted in India on the
levels of salivary fructosamine amongst diabetes mellitus patients; also, by extensive
literature review we could identify numerous research papers on the role of salivary
biomarkers in cancer diagnosis. However, there is a huge lacuna in the volume of research
on the potential use of saliva in the diagnosis and posttreatment monitoring of diabetes
mellitus. It has been projected that with the growing population, India will soon
become the diabetic capital of the world; hence we found a pressing need to conduct
this study.[1 ]
[2 ]
The etiopathology of the two types of diabetes mellitus are different and type II
diabetes mellitus being the most common form among the two, the present study was
contemplated to determine the association between salivary fructosamine and plasma
glycated hemoglobin, plasma fasting and postprandial glucose levels among type-II
diabetes mellitus patients and nondiabetic individuals in Chennai.
Materials and Methods
This cross-sectional study was designed to determine the association between salivary
fructosamine and plasma glycated hemoglobin, plasma fasting, and postprandial glucose
levels among type II diabetes mellitus patients. The study participants were divided
into two groups: group I (type II diabetes mellitus patients) and group II (nondiabetic
individuals). The study was conducted for a period of 5 months.
Ethical clearance was obtained from the institution review board. Permission to collect
blood and saliva was obtained from the hospital superintendent.
The inclusion and exclusion criteria of the study are as follows.
Inclusion Criteria
Patients with type II diabetes mellitus having plasma fasting glucose levels greater
than or equal to 126 mg/dL were included in group I (ADA—standards of medical care).[1 ]
Nondiabetic healthy individuals (with no other systemic disease) having plasma fasting
glucose levels less than 126 mg/dL were included in group II (ADA—Standards of medical
care).[1 ]
Patients who were willing to participate in the study.
Exclusion Criteria
Type II diabetes mellitus patients with diabetes-associated complications such as
protein-losing nephropathy and liver failure (confirmed by their hospital reports).
Patients with hemolytic anemia or hemoglobinopathy.
Patients who underwent a recent blood transfusion.
The sample size was calculated using the G power statistical software The Pearson's
correlation coefficient (r = 0.675), from a study conducted by Nakamoto et al was considered.[19 ] On considering alpha error as 0.05 and power as 90%, a sample size of 42 for each
group was obtained. Hence we included 50 individuals in each group, with a total of
100 participants.
Recruitment of Patients with Type II Diabetes Mellitus (Group I)
Patients with type II diabetes who visited the Department of Diabetology at Aringar
Anna Government Hospital Chennai were approached for the study.
Recruitment of Nondiabetic Healthy Individuals (Group II)
Individuals who enrolled in the routine master health checkup conducted once a month
by the general outpatient department (OPD)were approached for the study.
Informed consent (bilingual) was obtained from participants who met the inclusion
criteria and were willing to participate in the study.
A prevalidated questionnaire (validated by Pan American Health Organization and World
Health Organization) was administered to collect details about sociodemographic factors,
family history of noncommunicable disease, tobacco usage, and alcohol consumption.
(The demographic details relevant to the present study and questions related to diabetes
mellitus have been included in Results section.)
Collection of Saliva and Blood
Blood and saliva samples were obtained only as a part of their routine periodic checkup
(group I) and master health checkup (group II).
Blood samples were collected after an overnight fast. Blood was drawn from the antecubital
vein, and it was collected in vacutainers by a trained phlebotomist. The blood samples
were analyzed at the hospital laboratory.
After the participants rinsed their mouth with water six times, 3 mL of unstimulated
saliva samples were collected in a sterile container.[19 ] The saliva thus obtained were transferred to a sample collection box containing
gel packs and transported to the laboratory, to estimate the salivary fructosamine
levels. The samples were stored under two to 8°C at the laboratory before carrying
out the laboratory procedures.
Laboratory Procedure
The automated laboratory procedure was carried out by trained laboratory technicians,
using the following types of equipment.
Hexokinase-mediated reaction Roche/Hitachi modular P chemistry analyzer
Tosoh A1C 2.2 plus glycohemoglobin analyzer
NBT kinetic—Spectrophotometer, measuring at 520 nm (salivary fructosamine)
Statistical analysis was performed using SPSS version 20. (IBM Corp. Released 2013.
IBM SPSS Statistics for Windows, Version 22.0. Armonk, NY: IBM Corp.) Shapiro–Wilk
Normality test was run. Depending on the nature of the data, appropriate parametric
and nonparametric tests were chosen, a p value of 0.05 was considered to be significant. Simple linear regression was carried
out to determine the association between salivary fructosamine and plasma glycated
hemoglobin, plasma fasting, and postprandial glucose levels.
Results
The present study was conducted to determine the association between salivary fructosamine
and plasma glycated hemoglobin, plasma fasting, and postprandial glucose levels among
patients with type II diabetes mellitus and nondiabetic individuals.
[Table 1 ] shows the distribution based on the demographic details of the study population.
Among the demographic information that was collected, there were no significant differences
in age, sex, race, and tobacco and alcohol usage between the two groups. The family
history of type II Diabetes mellitus was found to be significantly higher among individuals
in group I when compared to group II.
Table 1
Distribution based on the demographic detail and knowledge about plasma HbA1C test
of the study population
Variables
Group I
Group II
p –Value
*Pearson chi-square test
**Fisher's exact test
# T-test.
Age
N
50
Mean (±SD)
54.38 (±9.924)
N
50
Mean (±SD)
50.90 (±9.990)
0.084
#
Sex
Male
Female
N
30
20
%
60
40
N
21
29
%
42
58
0.072*
Race
Aryo–Dravidian
Dravidian
Indo-Aryan
14
32
4
28
64
8
13
36
1
26
72
2
0.417**
Tobacco Usage
Yes
No
17
33
34
66
14
36
28
72
0.517*
Alcohol Usage
Yes
No
12
38
24
76
9
41
18
82
0.461*
Family History
Yes
No
34
16
68
32
10
40
20
80
<0.001*
Knowledge about Plasma HbA1C Test
Yes
No
32
18
64
36
24
26
48
52
0.107*
[Table 2 ] shows the distribution based on the mean BMI (body mass index), plasma fasting,
and postprandial glucose, plasma glycated hemoglobin, and salivary fructosamine levels
of the study population. A significant difference in BMI, plasma fasting, and postprandial
glucose, plasma glycated hemoglobin, and salivary fructosamine levels were observed
among the two groups. In group I female patients had significantly higher levels of
plasma fasting and postprandial glucose, plasma glycated hemoglobin, and salivary
fructosamine when compared to male participants.
Table 2
Distribution based on the mean BMI, plasma fasting and postprandial glucose, plasma
glycated hemoglobin, and salivary fructosamine levels of the study population
Variables
Group I
Group II
p –Value
* Mann–Whitney U test.
BMI
Male
N 30
Mean (±SD)
24.83 (±1.147)
Male
N 21
Mean (±SD)
22.71 (±1.521)
<0.001*
Female
20
24.90 (±1.021)
Female
29
22.55 (±1.804)
p –Value
0.918*
p –Value
0.787*
Plasma Fasting Glucose (mg/dL)
Male
30
128.20 (±46.643)
Male
21
84.14 (±6.792)
<0.001*
Female
20
167.60 (±68.775)
Female
29
84.55 (±5.968)
p –Value
0.014
*
p –Value
0.738*
Plasma Postprandial Glucose (mg/dL)
Male
30
182.77 (±96.524)
Male
21
110.19 (±14.713)
<0.001*
Female
20
259.35 (±105.068)
Female
29
115.21 (±15.486)
p –Value
0.011
*
p –Value
0.288*
Plasma HbA1c (%)
Male
30
6.53 (±2.097)
Male
21
4.90 (±0.539)
<0.001*
Female
20
8.35 (±2.700)
Female
29
4.66 (±0.484)
p –Value
0.019
**
p –Value
0.111**
Salivary Fructosamine (µmol/L)
Male
30
158.80 (±97.736)
Male
21
80.90 (±29.348)
<0.001**
Female
20
208.10 (±72.480)
Female
29
85.07 (±25.209)
p –Value
0.036
*
p –Value
0.582*
[Table 3 ] shows the correlation between salivary fructosamine, plasma fasting, and postprandial
glucose, and plasma glycated hemoglobin levels in the two groups. In group I individuals
high correlation was observed between salivary fructosamine, plasma fasting, postprandial
glucose levels, and plasma glycated hemoglobin. In group II high correlation was observed
between salivary fructosamine and plasma fasting glucose levels. A moderate correlation
was observed between salivary fructosamine and plasma postprandial glucose and plasma
glycated hemoglobin levels. [Fig. 1 ]
[Fig. 2 ] shows the correlation between salivary fructosamine and plasma glycated hemoglobin
levels in group I and II individuals, respectively.
Table 3
Correlation between salivary fructosamine, plasma fasting and postprandial glucose,
and plasma glycated hemoglobin levels in the two groups
Salivary fructosamine (umol/L)
Groups
Variable
r –value (Spearman's rho)
p –Value
Group I
Plasma fasting glucose (mg/dL)
0.934
<0.001
Plasma postprandial glucose (mg/dL)
0.910
<0.001
Plasma HbA1C (%)
0.893
<0.001
Group II
Plasma fasting glucose (mg/dL)
0.794
<0.001
Plasma postprandial glucose (mg/dL)
0.530
<0.001
Plasma HbA1C (%)
0.307
0.030
Fig. 1 Correlation between salivary fructosamine and plasma glycated hemoglobin levels in
group I individuals.
Fig. 2 Correlation between salivary fructosamine and plasma glycated hemoglobin levels in
group II individuals.
Discussion
To improve treatment outcomes and financial burden, there is the need for early diagnosis
of diabetes, by periodic monitoring of glycemic control.[26 ]
A few studies have been conducted to estimate the levels of salivary glucose, and
to determine the association between salivary glucose and blood glucose levels. However,
due to the presence of bacteria and enzymes in the oral cavity, salivary glucose is
rapidly decomposed, or it may be decomposed during the transfer from blood to saliva.[19 ]
Campbell MJA observed that the salivary glucose levels were significantly higher in
patients with diabetes when compared to nondiabetic healthy individuals. However,
no significant correlation was observed between blood glucose and salivary glucose
levels.[27 ] Similarly, a study conducted by Forbat et al reported no significant correlation
between blood and salivary glucose levels.[28 ]
Nakamoto et al compared the levels of salivary glucose among healthy controls, patients
with impaired glucose tolerance test (IGTT) and patients with diabetes. The study
revealed no correlation between salivary glucose and plasma glucose levels.[19 ]
Manjrekar et al compared the levels of salivary glucose among healthy controls, nondiabetic
first-degree relatives of patients with type II diabetes, and patients with type II
diabetes. Salivary glucose levels did not correlate with plasma glucose levels in
all the groups.[29 ]
In the present study, we did not assess salivary glucose levels as extensive literature
review shows there is no correlation between salivary and plasma glucose levels.[19 ]
Salivary fructosamine is a glycated protein formed by the nonenzymatic reaction between
glucose and plasma proteins (mostly albumin).[11 ] Since the protein combines with glucose, it is relatively stable against bacteria
when compared to salivary glucose. A substantial body of evidence shows that plasma
fructosamine is correlated with plasma glucose and HbA1C levels; however, very few
studies have been conducted to estimate the levels of salivary fructosamine in diabetic
patients.[19 ] Hence the present study was contemplated to determine the association between salivary
fructosamine and plasma glycated hemoglobin (HbA1C), plasma fasting, and postprandial
glucose levels among patients with type II diabetes mellitus and nondiabetic controls.
Gen et al conducted a study among diabetic patients and healthy controls. The study
results reported that salivary fructosamine levels were significantly higher in individuals
with diabetes (0.61 ± 0.19 mmoL DMF/L) when compared to healthy controls (0.38 ± 0.07
mmol DMF/L). Salivary fructosamine was positively correlated with plasma fasting glucose
(r = 0.449), plasma fructosamine (r = 0.526), and plasma glycated HbA1C (r = 0.411).[30 ]
Nakamoto et al reported the mean salivary fructosamine levels of the study population
as 25.2 ± 11.6 µmol/g protein. A positive correlation was observed between salivary
fructosamine, plasma HbA1C, and plasma blood glucose levels (r = 0.449, and r = 0.445, respectively). No association was observed between salivary fructosamine
and plasma fructosamine (r = 0.260). This was attributed to the protein compensation that occurs in the oral
cavity (by the combination of salivary proteins and glucose); however, such protein
compensation was not observed in plasma fructosamine. The study suggests that since
no correlation was observed between salivary glucose and plasma glucose, the correlation
between salivary fructosamine, plasma glucose, and HbA1C reflects that the salivary
fructosamine was produced in the blood or in saliva over a short period of time.[19 ]
Morenkova reported that there was an increase in salivary fructosamine concentration
with an increase in plasma glucose levels. The levels of salivary fructosamine in
healthy controls, patients with type I diabetes, and patients with type II diabetes
were 3.7 ± 0.7 mmol/L, 6.2 ± 1.7 mmol/L, and 7.3 ± 2.0 mmol/L, respectively.[31 ]
Manjrekar et al, reported that diabetic patients had significantly higher levels of
salivary fructosamine (202.1 ± 103.4 mg/dL) when compared to first-degree relatives
(130 ± 71.6 mg/dL) and control group (99.8 ± 50.1 mg/dL). However, no significant
correlation was observed between salivary fructosamine and plasma fasting glucose
levels in all the groups (group I r value = 0.061, group II r value = 0.171, group III r value = −0.078).[29 ]
In the present study, patients with type II diabetes had significantly higher levels
of salivary fructosamine when compared to nondiabetic healthy individuals (p value < 0.001). The mean levels of salivary fructosamine among group I male and female
participants were observed to be 158.80 (±97.736) and 208.10 (±72.480) µmol/L, respectively.
Among group II male and female participants the mean levels of salivary fructosamine
were observed to be 80.90 (±29.348) and 85.07 (±25.209) µmol/L, respectively. It was
also observed that female participants in group I had significantly higher levels
of salivary fructosamine when compared to male participants (p -value = 0.036). The higher levels of salivary fructosamine among group I female participants
can be attributed to the significantly higher levels of plasma fasting and postprandial
glucose levels in them, which in turn leads to increased glycation of plasma proteins.
Participants in group I had significantly higher levels of plasma fasting glucose
when compared to group II (p -value < 0.001); also in Group I female participants had significantly higher levels
of plasma fasting glucose when compared to male participants (p -value = 0.014). This may be attributed to the difference in insulin sensitivity,
insulin secretion, and hepatic glucose production, among the two genders.[32 ]
[33 ]
A high correlation was observed between salivary fructosamine and plasma fasting glucose
levels in group I (r = 0.934) and group II (r = 0.794) individuals.
Participants in group I had significantly higher levels of plasma postprandial glucose
when compared to group II (p value < 0.001); also in group I female participants had significantly higher levels
of plasma postprandial glucose when compared to male participants (p value = 0.011). This may be attributed to the prolonged gut absorption of glucose
in women.[33 ]
A high correlation was observed between salivary fructosamine and plasma postprandial
glucose levels in group I (r = 0.910). In group II, a moderate correlation was observed between salivary fructosamine
and plasma postprandial glucose levels (r = 0.530) individuals.
Participants in group I had significantly higher levels of plasma HbA1C when compared
to group II (p value < 0.001); also in group I female participants had significantly higher levels
of plasma HbA1C when compared to male participants (p value = 0.019). Glycemic and metabolic control in women are affected by postprandial
glucose levels.[33 ] Hence the higher HbA1C levels among female participants may be attributed to the
high levels of fasting and postprandial glucose levels, which in turn leads to increased
glycation of hemoglobin.[33 ]
A high correlation was observed between salivary fructosamine and plasma HbA1C levels
in group I (r = 0.893), and in group II, a moderate correlation was observed (r = 0.307).
Saliva collection apart from being simple and noninvasive can be utilized in instances
where the collection of blood is not ideal due to the patient's age (especially in
pediatric and older adults), attitude toward venous puncture, and hygiene issues in
rural areas.[34 ] Collection of saliva does not require any special training or equipment. Hence salivary
biomarkers can be used in large-scale population assessments and screening programs.[34 ]
The limitations of the study include the use of convenience sampling technique with
cross-sectional design; hence further longitudinal studies are required to validate
the use of salivary fructosamine for the diagnosis and monitoring of glycemic control
in patients with diabetes.
Conclusion
The present study throws light on the potential use of salivary fructosamine for the
diagnosis and post-treatment monitoring of glycemic control among patients with diabetes.
The overall study results showed that salivary fructosamine levels were significantly
higher in patients with type II diabetes mellitus when compared to nondiabetic individuals.
Also, a positive correlation was observed between salivary fructosamine, plasma fasting,
plasma postprandial, and plasma glycated hemoglobin.
