The Role of Hbalc as a Diagnostic Test for Type 2 Diabetes Mellitus in Bangladesh

Background zType2diabetes mellitus is a serious chronic disease with micro vascular complications such as retinopathy, nephropathy and neuropathy and macro vascular complications such as cardiac, peripheral arterial and cerebrovascular disease. Objective: The aim of the study was to investigate the value of tlbAlc as a diagnostic test for type 2 diabetes mellitus in Bangladeshi individuals. Methods: This cross sectional study was conducted in the Departrnent of Biochemistry, Bangabandhu Sheikh Mujib Medical University, Dhaka, Bangladesh. A total657 patients, who were attended in the one point sample collection cenffe of Bangabandhu Sheikh Mujib Medical University for oral glucose tolerance test (OGTT) from 1st Apitl2Ol4 to 30th Jtne2014, were purposively enrolled in this study. According to WHO criteria and based on OGTT findings study subjects were categorized into Normoglycemic (257),IFG (82), IGT (174), and DM (347). Fasting plasma glucose, tlbAlc and plasma glucose at 2 hour after glucose load on OGTT was done from all the study subjects. Results: With a cut-off value of 6.l%o,IIbAlc had a maximal sensitivity and specificity of 97.0%o and 49.OVo respectively with a positive predictive valte 65.5Vo and a negative predictive valae 94.OVo.IIbAlc had a sensitivity of 93 .l%o and a specificity of 63 .0Vo was calculated with a cut-off valu e of 6 .57o with positive predictive vahte 77 .5Vo and negative predictive vahte 90.07o. Both fasting plasma glucose levels and 2 hour plasma glucose levels were showed significantpositive correlation with HbAlc (r= 0.788, P = 0.000 andr= 0.800, P = 0.000respectively). Conclusion: The study suggests that measurement of HbAlc could be used to make diagnosis of T2DM in the Bangladeshi population.

20t4 Jul Based, orr cuffent recommendations, diagnosis of DM requires the presence of a fasting plasma glucose concen- tration of >7.1mmollL, or a 2hour plasma glucose level of > 1 1.1 mmol/L on an oral glucose tolerance test (OGTT).On the other hand, international committee members selected by the American Diabetes Association (ADA) and the Alliance for European Diabetes Research (EURADIA) recently suggested that glycosylated hemo- globin (HbAlc) could be used as an alternative for making diagnosis of DM8.The committee concluded that an HbAlc level of >6.5Vo was diagnostic for DM, without requiring a determination of plasma glucose levels.However, the use of standard glucose measure- ments is still recommended for individuals when HbAlc assays are deemed unreliablee.
HbAlc is formed as a result of the addition of a stable glucose molecule to the N-terminal group of an HbA0 molecule via a non enzymatic glycation processl0, and is considered a reliable indicator of the glycemic status of the previous 3 monthsll.
Despite the cloud of controversy regarding the limitations of HbAlc for making diagnosis of DM, many experts believe that HbAlc may be superior to the OGTT in daily clinical practice as there was no need of preparation of patient, no fasting is necessary, sample collection and test procedure is simple, there was no variation of test result like plasma glucosel2.
The aim of this study was to investi gate the value of HbAlc as a diagnostic test for T2DM in Bangladeshi individuals.

Methods :
This cross sectional study was undertaken in the Depart- ment of Biochemistry, Bangabandhu Sheikh Mujib Medical University, Dhaka, Bangladesh.Patients, who were attended in the one point sample collection centre of Bangabandhu Sheikh Mujib Medical University for oral glucose tolerance test (OGTT) from 1st AprLl2014 to 30th From each patient, blood samples were obtained at 0800 hours and onward following 10 hours fast, from the antecubital vein in a sitting position, for the determination of HbAlc as well as fasting plasma glucose level.A11 patients were then subjected to a 7 5-gm OGTT on the same day, and second blood samples were obtained 2 hour after glucose loading for determination of plasma glucose level.For determination of HbAlc whole blood samples were collected in vacuum tube contains EDTA and for plasma glucose measurement blood samples were collected in fluoride tube.
An NcsP-approved for the percent determination of HbAlc in human whole blood using ion-exchange high- performance liquid chromatography (HPLC) method on a D-10 Hemoglobin A1c program (Bio-Rad, Japan)   analyzer.Blood samples were automatically diluted on the D-10 and injected in the analytical cartridge.The D-10 delivers a programmed buffer gradient of increasing ionic strength into the analytical cartridge, where the hemoglobin is separated base on their ionic interactions with the cartridge material.The separated hemoglobin then passes through the flow cell of the photometer, where changes in absorbance at 4I5 nm were measured.
Glucose measurements were made on the same day as HbAlc measurements, using the glucose oxidation method on Dead Behring (Bio-Rad, Japan) auto analyzer.
Subjects were categonzed into 4 groups based on their OGTT results, according to the crtterra put forth by the WHO: noffnoglycemic (NG), impaired fasting glucose (IFG), impaired glucose tolerance (IGT), and diabetes mellitus (DM).
Statistical analyses were performed using SPSS windows version 20.Yalues for HbAlc and plasma glucose levels were provided as mean + standard deviation.The sensitivity, specifi city, positive predictive values, and negative predictive values for both tests were calculated by plotting a receiver operating characteristic (ROC) curve.Cor-rela- The role of HbAlc as a diasnostic test for Tvoe 2 diabetes mellitus in Baneladesh Md Matiur Rahman et al   tion analyses between fasting plasma glucose, 2 hour plasma glucose, and HbAlc levels were performed using Spearman's correlation test.A p-value of less than 0.05 was considered indicative of statistical significance.

Results:
A total of 651 consenting participants (360 male and 297 female) were included in this study.Based on OGTT results and according to criteria put forth by the WHO, 257 individuals were categofized as normoglycemic, 82   were IFG, 114 were IGT and 347 were DM.The demo- graphic characteristics and laboratory findings of the study population have been summarized in the  The OGTT was considered the gold-standard test for the diagnosis of DM.The area under the ROC curve for the diagnosis of DM by HbAlc was 0.90 (P < 0.001) (Figure 1).With a cut-off value of 6.17o, HbAlc had a maximal sensitivity and specificity of 9l .07oand 49.07o respec- tively with a positive predictive value 65.57o and negative predictive value 94.07o.With a cut-off value of 6.57o, HbAlc had a sensitivity of 93 .07oand a specificity of A significant positive coffelation was observed between HbAlc, fasting plasma glucose (r = 0.788, P = 0.000), and 2hplasma glucose (r = S.800, P -0.000) levels in patients with DM (Figures 2 & 3).
Several studies have investigated the value of HbAlc for the diagnosis of DM with a cut-off level of 6.17o.In such a study by  HbAl c (r -0.800, p -0.000). Discussion: In this study fasting plasma glucose and 2 hour plasma glucose on OGTT showed significant positive correlation with HbAlc (r = 0.788, p = 0.000 and r = 0.800, p = 0.000).Riet et al.'3 reported that weak positive correla- tions between HbAlc and fasting plasma glucose levels (r -0.46), as well as 2 hour plasma glucose (r = 0.33) levels, were observed in individuals from the general population.
Ginis et al.'o reported that both fasting plasma glucose and 2 h plasma glucose levels were found to correlate moder- ately with HbAlc levels (r -0.47, P -0.001 and r = 0.52, P = 0.000, respectively) and were consisted with our Fig-L z Receiver operating characteristic curve for HbAl c in patient with diabetes mellitus.

Table - I
Demographic and Laboratory findings of study population Tavintharan et al.ls a sensitivity of 8l7o wasreported with a specificity of 847o.Simil arly,Ko et a1.16reported on a sensitivity and specificity of 17 .5%oand78.8%o,raspectively.With a cut-off value for the diagnosis of DM of 6.17o, HbAlc had a sensitivity of 81.87o and a specificity of 807o, with positive and negative predictive values of 80.2%o and 81.05%o, respectively showed by Ginis et aL.|a.Comparable results were observed in our study for diagnosis of DM by HbAlc with a cut-off value et al.r1 reported on a sensitivity of 65%o and a specificity of 88%o with positive and negative predictive values of 75.27o and 96.5%o respectively for diagnosis of DM by HbAlc with a cut-off value of 6.5Vo.Similar study done by Ginis el al.ta also reported that a sensitivity of 56.87o and a specificity of 89.27o for diagnosis DM with a cut-off value of 6.5 To.Inour study with a cut-off value of 6.57o, HbAlc had a sensitivity of 93 .0%oand a specificity of 63.07o with positive predictive value 77 .57oandnegativepredictive value 90.0Vo.HbAlc reflects averuge plasma glucose over the previous eight to 12 weeks".It canbe performed at any time of the day and does not require any special preparation such as fasting.These properties have made it the preferred test for assessing glycemic control in people with diabetes.More recently, there has been substantial interest in using it as a diagnostic test for diabetes and as a screening test for persons at high risk of diabetesle.Owing in large part to levels or performing an OGTT, and day-to-day variability in glucose, an alternative to glucose measure- ments for the diagnosis of diabetes has long been sought. Kumar