Abstract
BACKGROUND
Hemoglobin A1c (HbA1c) is used to monitor long-term glycemic control in individuals with diabetes, guide therapy, predict the risk of microvascular-complications, and more recently to diagnose diabetes. An automated liquid-flow-capillary-electrophoresis method was recently developed to measure HbA1c using the Capillarys 2 Flex Piercing instrument.
METHODS
Analytical evaluation was performed at 2 clinical centers. A clinical analysis was conducted in 109 African-born individuals, 24% of whom have variant hemoglobin (HbAS or HbAC). Abnormal glucose tolerance (which includes both diabetes and prediabetes) was defined as 2 h glucose ≥ 140 mg/dl (7.8 mmol/l) during an oral glucose tolerance test.
RESULTS
Interlaboratory CVs were ≤ 2.1%. The method showed satisfactory correlation with 2 other analyzers that measure HbA1c by high-performance liquid chromatography. Neither labile HbA1c, carbamylated hemoglobin, uremia, bilirubin nor common hemoglobin variants (HbC/HbS/HbE) interfered. Forty-five individuals (41%) had abnormal glucose tolerance. The sensitivity of HbA1c for diagnosing abnormal glucose tolerance was 38%, 36% and 42% for total, normal and variant hemoglobin groups, respectively.
CONCLUSIONS
The analytical performance of HbA1c on the Capillarys 2 is suitable for clinical application. Variant hemoglobin in Africans did not interfere with the detection of abnormal glucose tolerance by HbA1c measured on the Capillarys 2.
Keywords: Hemoglobin A1c, Capillarys 2, Diabetes, Abnormal Glucose Tolerance, Capillary Electrophoresis, Sickle Cell, Africans
Introduction
Hemoglobin A1c (HbA1c) is the reaction product of glucose with the NH2-terminal valine of the β-chain of hemoglobin. In individuals with diabetes, HbA1c is used to monitor long-term glycemic control, guide therapy, and predict the risk of microvascular complications. More recently, HbA1c has been advocated by several influential diabetes organizations for the diagnosis of diabetes [1,2]. It is therefore vital that methods used to measure HbA1c are accurate, precise, reproducible and subject to minimal interference.
Initial assays to measure glycated hemoglobin were limited by a lack of comparability, resulting in very wide variability in results among laboratories [3]. The implementation of standardization activities by the National Glycohemoglobin Standardization Program (NGSP) and the International Federation of Clinical Chemistry and Laboratory Medicine (IFCC) has considerably improved HbA1c accuracy and precision [4]. Notwithstanding this progress, further improvement is required given the integral importance of HbA1c measurement for the diagnosis and management of diabetes.
Recently, Sebia (Lisses, France) developed an automated liquid-flow capillary electrophoresis (CE) method to measure HbA1c on the Capillarys 2 Flex Piercing instrument. This is the first CE technology approved by the U. S. Food and Drug Administration for HbA1c measurement. The instrument provides complete automation for fast separation and quantification of HbA1c using the principle of liquid-flow capillary electrophoresis in free solution. With this technique, charged molecules are separated by their electrophoretic mobility in an alkaline buffer with a specific pH, as well as by electrolyte pH and electroosmotic flow. Hemoglobins are detected by absorption spectroscopy at the cathodic end of the capillary [5].
Sickle cell trait (HbAS) occurs in about 0.08% of US newborns [6]. The prevalence is 6–8% in African-Americans [6,7] and as high as 40% in certain areas of Africa [8]. Hemoglobin C trait (HbAC) is the second most common variant in African Americans and prevalence reaches 15% in areas of West Africa [9]. Importantly, populations of African descent are also at particularly high risk for diabetes (prevalence of 13.3% to 14.9%), which is twice that of non-Hispanic Whites [10]. Although the diagnostic utility of HbA1c for diabetes has been investigated in several ethnic groups, data are lacking for populations of African descent with high prevalence of both diabetes and hemoglobin variants. The accuracy of HbA1c methods can be adversely affected by the presence of hemoglobin variants [11, 12]. Therefore, to assess Africans an accurate HbA1c method that is not affected by hemoglobin variants, such as HbAS or HbAC, is necessary.
Materials and Methods
Participating centers
Technical evaluation was performed at the Diabetes Diagnostic Laboratory using NGSP Secondary Reference Laboratory methods (NGSP SRL9 and SRL3), University of Missouri (Center 1), and Department of Laboratory Medicine, National Institutes of Health (NIH, Center 2). Clinical evaluation was conducted at the NIH.
Characteristics of the CAPILLARYS 2 Flex Piercing analyzer
Sample analyses were carried out following the manufacturer’s instructions (4).
Capillarys 2 Flex Piercing is a multipurpose instrument using 8 capillary tubes for multiple and simultaneous hands-free electrophoretic separations at a high throughput. The separation of the different hemoglobin fractions takes place in 8 silica capillary tubes of 25 µm internal diameter, and migration is performed at a high voltage of 9.4 kV under tight temperature control of 34°C using a Peltier device. Hemoglobins are directly detected at a specific absorption wavelength of 415 nm at the cathodic end of the capillary.
Analytical evaluation
Blood samples used in this portion of the study were left over from samples routinely submitted for HbA1c measurement. Imprecision, trueness and linearity studies were performed following Clinical and Laboratory Standards Institute (CLSI) EP-5, EP-9 and EP-6 protocols, respectively. Tosoh G8 and Bio-Rad D10 (both NGSP-certified; G8 is also an NGSP Secondary Reference Laboratory method) were the comparison methods for trueness. Sample pools of high and low HbA1c concentrations were run alternately to assess carry over.
For all interference studies (namely labile HbA1c, carbamylated hemoglobin, uremia, bilirubin, and total hemoglobin), a difference of ±0.3% (3.3 mmol/mol) HbA1c from the target was used as an acceptable limit [13]. Methodological details of the analytical evaluation are in the supplement.
Hemoglobin variants for analytical evaluation
The possible interference of variant hemoglobins was assessed. Samples with HbAS (n=5), HbAC (n=5) or HbAE (n=1) were analyzed from frozen whole blood aliquots on both the Trinity ultra2 (NGSP SRL3, Trinity Biotech, Kansas City, MO) and Capillarys 2; these methods have previously been shown to compare well in samples without variants (30). The Trinity ultra2 boronate affinity method was selected as a comparison method because these hemoglobin variants do not interfere with HbA1c measurement by this method [14]. In addition, HbSS (n=4), HbEE (n=2), HbCC (n=2), HbDD (n=1), and HbSC (n=6) samples were analyzed by the Capillarys 2.
Clinical study
This is a substudy of the Africans in America Study [15–17], which is an ongoing, prospective study of adults who were born in equatorial Africa, but resided in the US at the time of the study. The group evaluated consisted of 109 consecutively enrolled adults who came to the NIH Clinical Center between January 2012 and August 2014. Participants stated at enrollment they were healthy and denied any history or symptoms of diabetes. Informed written consent was obtained from all participants. The study protocol was approved by the NIDDK Institutional Review Board.
At the first outpatient visit a history and physical examination were conducted. Anemia, iron deficiency and renal, hepatic and thyroid dysfunction were ruled out by routine blood tests at the screening visit. Glucose homeostasis was evaluated at the second and third visits.. At Visit 2, following a 12 h fast, fasting plasma glucose (FPG), HbA1c, and a 2h plasma glucose during a standard oral glucose tolerance test (OGTT) using 75 g dextrose were obtained. At Visit 3, an insulin modified-frequently sampled intravenous glucose tolerance test (IM-FSIGT) was performed in 80 participants as described previously [18]. Twenty-nine participants did not undergo an IM-FSIG; 11 participants had an increased 2h plasma glucose during their OGTT consistent with the diagnosis of diabetes and per protocol did not proceed to the IM-FSIGT, while 18 participants declined the IM-FSIGT. The insulin sensitivity index (SI) (a measure of insulin resistance) and the acute insulin response to glucose (AIRg) (a measure of β-cell secretion) were determined as a previously described [19].
Hemoglobin, hematocrit, mean corpuscular volume (MCV), and reticulocyte count were analyzed in EDTA-anticoagulated whole blood using a Sysmex XE-5000 analyzer. Glucose, total bilirubin, direct bilirubin, AST, ALT, blood urea nitrogen, creatinine, iron, transferrin, ferritin, vitamin B12, folate and insulin were analyzed in serum or plasma using Siemens Dimension Vista analyz or Roche Cobas 6000 analyzer. HbA1c was measured using whole blood aliquots frozen at −80°C on Sebia Capillarys 2 Flex Piercing system at Center 2. Hemoglobin electrophoresis was performed on Helena Zip-Zone electrophoresis instrument to identify variant hemoglobin in 88 consecutively enrolled participants. Identities of hemoglobin types were confirmed by comparison to known samples of HbA, HbS and HbC. Other than hemoglobin electrophoresis and HbA1c, all laboratory measurements were performed on the day of the visit. Abnormal glucose tolerance (which comprises both diabetes and prediabetes) was defined according to ADA criteria as 2h glucose ≥ 140 mg/dl (7.8 mmol/l) during an OGTT [1].
Statistical analysis
Student’s t-test and the χ2 test were applied to compare the metabolic and demographic characteristics of individuals with normal or variant hemoglobins. Diagnostic sensitivity and specificity were calculated based on ADA criteria for pre-diabetes and diabetes [1], namely HbA1c ≥ 5.7% (39 mmol/mol) or FPG ≥ 100 mg/dl (5.6 mmol/l). Sensitivity and specificity values were compared via chi-square test when applicable. All statistics were performed with STATA 13.0 (StataCorp), Prism GraphPad 5.0 (GraphPad Software) or EP Evaluator Release 10 software (David G. Rhoads Associates). Statistical significance was defined as a P value ≤0.05.
Results
Analytical performance
Within-run coefficients of variation (CVs) were 1.1–2.2% for NGSP units/ 1.3%–3.7% for SI (IFCC) units (center 1) and 0.9%–2.0% for NGSP units/ 1.0%–3.4% for SI units (center 2), while between-run CVs were 0.0%–0.7% for NGSP units/ 0.0%–0.9% for SI units (Table 1). Total intra-laboratory CVs were ≤ 2% for NGSP units/ ≤ 4% for SI units for all samples in both centers. The inter-laboratory CVs were 1.1–2.1% for NGSP units/ 1.3%–3.7% for SI units.
Table 1.
Imprecision of Capillarys 2 Flex Piercing HbA1c assay.
| NGSP units | |||||||
|---|---|---|---|---|---|---|---|
| % HbA1c Mean |
Imprecision (CV, %) | ||||||
| Center 1 | Center 2 | Inter- laboratory |
|||||
| Within- run |
Between-run | Total | Within- run |
Between- run |
Total | Total | |
| 4.7 | 2.2 | 0.0 | 2.3 | 2.0 | 0.0 | 2.0 | 2.1 |
| 6.3 | 1.2 | 0.6 | 1.4 | 1.4 | 0.4 | 1.5 | 1.4 |
| 7.4 | 1.2 | 0.7 | 1.4 | 1.3 | 0.0 | 1.3 | 1.3 |
| 11.1 | 1.1 | 0.5 | 1.2 | 0.9 | 0.4 | 1.0 | 1.1 |
| SI units | |||||||
|---|---|---|---|---|---|---|---|
| HbA1c Mean (mmol/mol) |
Imprecision (CV, %) | ||||||
| Center 1 | Center 2 | Inter- laboratory |
|||||
| Within- run |
Between-run | Total | Within- run |
Between- run |
Total | Total | |
| 28 | 3.7 | 0.0 | 3.8 | 3.4 | 0.9 | 3.5 | 3.7 |
| 45 | 1.9 | 0.7 | 2.0 | 2.1 | 0.7 | 2.2 | 2.1 |
| 57 | 1.7 | 0.9 | 2.0 | 1.9 | 0.0 | 1.9 | 1.9 |
| 98 | 1.3 | 0.5 | 1.4 | 1.0 | 0.5 | 1.2 | 1.3 |
HbA1c values correlated well between the Capillarys 2 Flex Piercing instruments at the 2 centers and also with those of the comparison methods (Tosoh G8 and Bio-Rad D10), with minimal bias (Fig. 1 and supplemental Fig. 1). Linear regression analysis yielded the following: y (HbA1c Capillarys 2 center 1) = 1.033 × (HbA1c Tosoh G8) −0.34, r=0.997, Syx=0.16; y (HbA1c Capillarys 2 center 2) = 1.049 × (HbA1c Tosoh G8) − 0.38, r=0.996, Syx=0.19; and y (HbA1c Capillarys 2 center 2) = 1.084 × (HbA1c Bio-Rad D10)−0.67, r=0.995, Syx=0.22. Results of the Capillarys 2 were comparable between centers with mean bias = 0.09% (1.0 mmol/mol) HbA1c (y (HbA1c Capillarys 2 center 1) = 0.982 × (HbA1c Capillarys 2 center 2) ± 0.05, r=0.996, Syx=0.18). The HbA1c values obtained with the Capillarys 2 Flex Piercing at each center also correlated well with the NGSP secondary reference laboratory method (r ≥ 0.995 for both centers) and mean bias was −0.09% (−1.0 mmol/mol) HbA1c (Capillarys 2 center 1 vs. Tosoh G8, NGSP SRL9) and 0.00% (0 mmol/mol) HbA1c (Capillarys 2 center 2 vs. Tosoh G8, NGSP SRL9). Ninety-five % (center 1) and 97 % (center 2) of single Capillarys 2 results were within ±6% of the SRL9 mean (NGSP manufacturer certification criteria require that 37/40 or 92.5%, results be within ±6% of the SRL mean) [20, 21].
Figure 1.
Scatter and bias plots showing HbA1c results obtained in 2 centers (A, center 1; B, Center 2) compared to those obtained in an NGSP SRL (Tosoh G8) at Center 1. C, comparison of HbA1c values obtained with Capillarys 2 Flex Piercing instruments at the 2 centers. In the scatter plots, the lines represent x = y (dotted line) and the Deming regression (dashed line). In the bias plots, the dashed line represents the mean bias.
The method was linear for HbA1c results from 4.2% to 17.6% (22 to 169 mmol/mol, center 1) and 4.3% to 14.1% (23 to 131 mmol/mol, center 2), with correlation coefficients of 0.998 and 1.00, respectively, and the differences between measured and expected values were within ±6% (Supplemental Table 1, Supplemental Fig. 2).
The difference between the means of the high-low and the low-low sequence samples was 0.06% (0.3 mmol/mol), which is < 3 SD of the low-low values (1 SD of the low-low values= 0.07% (0.7 mmol/mol), indicating no significant carryover.
Evaluation of interference
There was no interference from labile HbA1c (up to 11%, 97 mmol/mol), carbamylated hemoglobin, BUN (97 mg/dl, 34.6 mmol/l) or bilirubin (40 mg/dl, 684 µmol/l). No additional peaks (i.e. for labile HbA1c or carbamylated hemoglobin) were observed on the Capillarys 2 electropherograms. Varying hemoglobin concentrations from 6.9 to 18.3 g/dl had no effect on HbA1c measurement (Supplemental Table 2, Supplemental Fig. 3).
Hemoglobin variants
All the Capillarys 2 results with HbAS, HbAC and HbAE samples were within ±7% of the Ultra2 results (Supplemental Table 3). Analysis of HbSS, HbEE, HbCC, HbDD, and HbSC samples with Capillarys 2 gave no results (data not shown). This was expected as these samples have no HbA and therefore no HbA1c.
Clinical analysis of HbA1c on Capillarys 2 for the detection of abnormal glucose tolerance
Since the Capillarys 2 method accurately measured HbA1c in individuals with HbAS or HbAC, we evaluated its ability to identify individuals with abnormal glucose tolerance in a population with a high prevalence of variant hemoglobin.
The characteristics of the study population are shown in Table 2. There was no significant difference in age or BMI between subgroups with normal and variant hemoglobin. Reticulocytes, iron, vitamin B12, folate, liver function, and renal function did not vary by hemoglobin status. Thus, factors reported to interfere with HbA1c analysis or levels (i.e., uremia, anemia or iron deficiency) [22] were absent. The variant hemoglobin subgroup had significantly lower MCV (P<0.001). Among the 109 participants, 83 had normal hemoglobin. The Capillarys 2 detected variant hemoglobin in 26 (23 HbAS and 3 HbAC) (24%) of the participants. Hemoglobin electrophoresis was performed in 88 subjects and verified that the Capillarys 2 correctly identified all variant (19 HbAS and 2 HbAC) and normal hemoglobin. Importantly, hemoglobin status did not alter glucose homeostasis; FPG, 2h glucose, SI (insulin resistance) or AIRg (β-cell function) were not significantly different (Table 2). Mean HbA1c concentrations were essentially identical among the groups (Table 2).
Table 2.
Metabolic and demographic characteristics†
| Parameter | Total Group n=109 |
Normal Hb n=83 |
Variant Hb n=26 |
P value* |
|---|---|---|---|---|
| Male (%) | 72 | 72 | 69 | NS |
| Age (y) | 40±10 | 40±10 | 41±10 | NS |
| BMI (kg/m2) | 28.1±4.5 | 28.1±4.7 | 28.0±3.7 | NS |
| WC (cm) | 92±12 | 92±12 | 92±11 | NS |
| Hemoglobin, g/dl (g/l) | 14.1±1.3 (141±13) | 14.1±1.3 (141±13) | 14.3±1.1 (143±11) | NS |
| Hematocrit (%) | 41.9±3.3 | 42.1±3.3 | 41.2±3.2 | NS |
| MCV (fl) | 84.8±5.7 | 85.2±5.5 | 80.7±4.9 | <0.001 |
| Reticulocytes (%) | 1.27±1.18 | 1.27±0.52 | 1.27±0.37 | NS |
| Bilirubin Total, mg/dl (µmol/l) | 0.53±0.22 (9.0±3.8) | 0.51±0.22 (8.7±3.8) | 0.57±0.23 (9.7±3.9) | NS |
| Bilirubin Direct, mg/dl (µmol/l) | 0.15±0.06 (2.6±1.0) | 0.14±0.06 (2.4±1.0) | 0.16±0.06 (2.7±1.0) | NS |
| ALT (U/l) | 34±17 | 33±18 | 35±16 | NS |
| AST (U/l) | 24±11 | 25±12 | 23±6 | NS |
| BUN, mg/dl (mmol/l) | 12±3 (4.3±1.1) | 12±3 (4.3±1.1) | 12±2 (4.3±0.7) | NS |
| Creatinine, mg/dl (µmol/l) | 0.89±0.17 (79±15) | 0.89±0.17 (79±15) | 0.92±0.16 (81±14) | NS |
| eGFR (mL/min/1.73 m2)‡ | 110±20 | 111±21 | 104±15 | NS |
| Iron, µg/dl (µmol/l) | 83.5±27.1 (14±5) | 82.5±28.5 (14±5) | 86.8±22.1 (15±4) | NS |
| Transferrin, mg/dl (g/l) | 246±35 (2.46±0.35) | 244±32 (2.44±0.32) | 244±32 (2.44±0.32) | NS |
| Transferrin saturation (%) | 24.8±8.7 | 24.5±9.0 | 25.9±7.7 | NS |
| Ferritin µg/l (pmol/l) | 113±90 (253±202) | 110±92 (247±206) | 121±86 (271±193) | NS |
| Vitamin B12, pg/mL (pmol/l) | 613±335 (452±247) | 601±347 (443±256) | 652±297 (481±219) | NS |
| Folate, ng/mL (nmol/l) | 12.6±4.7 (28.6±10.7) | 12.4±4.7 (28.1±10.7) | 13.5±4.8 (30.6±10.9) | NS |
| HbA1C, % (mmol/mol) | 5.4±0.8 (35±8) | 5.4±0.8 (35±9) | 5.4±0.3 (36±4) | NS |
| Abnormal glucose tolerance# (%) | 41% (45/109) | 40% (33/83) | 46% (12/26) | NS |
| Diabetes& (%) | 10% (11/109) | 10% (8/83) | 12% (3/26) | NS |
| Pre-Diabetes‖ (%) | 31% (34/109) | 30% (25/83) | 35% (9/26) | NS |
| FPG, mg/dl (mmol/l) | 92±17 (5.1±0.9) | 93±18 (5.2±1.0) | 90±10 (5.0±0.6) | NS |
| 2h glucose, mg/dl (mmol/l) | 142±45 (7.9±2.5) | 142±48 (7.9±2.7) | 140±45 (7.8±2.5) | NS |
| SI ((mU/l)−1 min−1) | 3.27±1.79 | 3.41±1.80 | 2.89±1.76 | NS |
| AIRg (mUL−1 min) | 747±526 | 711±390 | 851±801 | NS |
Data represent mean ± SD
To compare normal and variant hemoglobin subgroups, the student t-test and chi-square test were applied for continuous variables and categorical variables, respectively.
eGFR calculated based on the Modification of Diet in Renal Disease (MDRD) Equation
Defined as 2h glucose≥7.8 mmol/l
Defined as 2h glucose≥11.1 mmol/l
Defined as 2h glucose≥7.8 mmol/l and <11.1 mmol/l
Abbreviations: BMI, body mass index; MCV, mean corpuscular volume; eGFR, estimated glomerular filtration rate; HbA1c, hemoglobin A1C; FPG, fasting plasma glucose; visceral adipose tissue; WC, waist circumference; SI, insulin sensitivity index; AIRg, acute insulin response to glucose.
Forty-five participants (41%) had abnormal glucose tolerance based on 2h glucose ≥ 140 mg/dl (7.8 mmol/l) (Table 2). Diabetes was detected in 11 (10%) participants, while 34 (31%) had pre-diabetes. The sensitivity of HbA1c to detect abnormal glucose tolerance in the variant hemoglobin subgroup (42%) was comparable to the whole cohort (38%) (P=0.81) and was slightly higher than in the normal hemoglobin subgroup (36%) (P=NS) (Table 3). The sensitivity of HbA1c was slightly, but not significantly, higher than that of FPG in the whole cohort (38% vs. 31%, P=NS), the subgroup with normal hemoglobin (36% vs. 33%, P=NS) and the subgroup with hemoglobin variants (42% vs. 25%, P=NS).
Table 3.
Detection of abnormal glucose tolerance*
| Sensitivity# | Specificity# | |||||
|---|---|---|---|---|---|---|
| Group | FPG | HbA1c | Combined | FPG | HbA1c | Combined |
| Whole Cohort (n=109) | 31% (14/45) | 38% (17/45) | 51% (23/45) | 97% (62/64) | 92% (59/64) | 91% (58/64) |
| Normal Hemoglobin (n=83) | 33% (11/33) | 36% (12/33) | 52% (17/33) | 96% (48/50) | 94% (47/50) | 92% (46/50) |
| Hemoglobin Variant (n=26) | 25% (3/12) | 42% (5/12) | 50% (6/12) | 100% (14/14) | 86% (12/14) | 86% (12/14) |
Abnormal glucose tolerance defined by 2h glucose ≥140 mg/dl (7.8 mmol/l)
Sensitivity and specificity to detect abnormal glucose tolerance (diabetes and prediabetes) were calculated based on the ADA criteria [1] i.e., FPG (FPG ≥ 100 mg/dl (5.6 mmol/l)), HbA1c (HbA1c ≥ 5.7% (39 mmol/mol)) or combined (either FPG ≥ 100 mg/dl (5.6 mmol/l) or HbA1c (HbA1c ≥ 5.7% (39 mmol/mol)).
Fig. 2 shows the number of true positive, false positive and false negative cases of abnormal glucose tolerance by FPG or HbA1c. Among these 45 participants, eight were identified by both FPG and HbA1c, whereas 22 were not identified by either assay. Nine participants identified by HbA1c were not detected by FPG and 6 identified by FPG were not detected by HbA1c. HbA1c and FPG falsely identified 5 and 2 cases, respectively, with 1 falsely detected by both.
Figure 2.
The number of true positive, false positive and false negative cases of abnormal glucose tolerance by FPG or HbA1c in 109 African immigrants. Dashed line, abnormal glucose tolerance defined by 2h glucose ≥ 140 mg/dl (7.8 mmol/l). True positive and false positive identified by HbA1c ≥ 5.7% (39 mmol/mol) (double lines) and FPG ≥ 100 mg/dl (5.6 mmol/l) (solid line). The numbers (percentage) of participants are indicated for each category.
Discussion
Considerable improvement has been made in the analytical performance of HbA1c measurement since the NGSP and IFCC standardization schemes were implemented [4]. Sebia recently developed a new CE-based HbA1c method on the Capillarys 2 Flex Piercing system. Initial reports of its technical performance in Europe have appeared in the last 2 y [23–27]. CE is an established technique routinely used for protein analysis in clinical laboratories [28] and has also been used for high throughput analysis of hemoglobinopathies and thalassemia [29]. In addition, a CE method to measure HbA1c was described previously [30]. The main advantages of the Capillarys 2 Flex Piercing system for HbA1c measurement include the ability to detect hemoglobin variants, high throughput, full automation, small sample volume, and ability to use primary tubes. To the best of our knowledge, this is the first detailed analysis of the Sebia Capillarys 2 Flex Piercing system performed in the US.
Measurement trueness and precision for HbA1c are essential as large, prospective clinical studies have demonstrated that a relatively small change in HbA1c (e.g., from 9% to 8%) significantly reduces the risk of diabetes microvascular complications [31, 32]. The National Institute for Health and Care Excellence (NICE) recommends that treatment regimens for individuals with diabetes should be assessed on the basis of a minimal change in HbA1c ≥ 0.5% [33]. With the recent recommendation to use HbA1c to diagnose diabetes [1,2], there is an even greater need for accurate and precise HbA1c measurements.
One hundred samples were analyzed on the Capillarys 2 at the Diabetes Diagnostic Laboratory and the NIH Clinical Center for the evaluation of trueness. Then the data from the 2 centers were directly compared both to each other and to an NGSP secondary reference laboratory method. Analysis of the same samples in 2 independent laboratories provides information important for patient care as the results from a single sample should not differ substantially between laboratories. Comparing a single result to the NGSP mean result (as would be done for NGSP certification) showed that > 92.5% of Capillarys 2 results were within 6% of the NGSP mean at each center (95% at center 1 and 97% at center 2), indicating that these methods pass the NGSP criteria for manufacturer and both Level I and Level II Laboratory certifications [21] at each site. The precision met the NACB/ADA criteria of intra-laboratory CV<2% and inter-laboratory CV<3.5% [34] and was comparable to or slightly better than CVs shown in other studies [23–25]. These data reveal that the Capillarys 2 Flex Piercing system demonstrated excellent accuracy and precision to meet clinical requirements.
Many of the methods routinely used for HbA1c measurement are subject to interference by hemoglobin variants [35]. For example, the accuracy of some HbA1c methods can be adversely affected by the presence of hemoglobin C or S trait [11,12]. The current study detected no analytical interference from three common hemoglobin variants (HbAC, HbAS, or HbAE) on the Capillarys 2 Flex Piercing assay, validating recent results [36]. By using alkaline pH buffer, normal and abnormal (or variant) hemoglobins are detected in the following order, from cathode to anode: A2/C, E, S, D, F, A0, other Hb (including minor HbA1) and then A1c. Common hemoglobin variants can be detected, but do not interfere with HbA1c results. The instrument is able to generate automatic flags from samples with hemoglobin variants and/or beta-thalassemia. This feature enables easy classification and interpretation of HbA1c results. Note that individuals with homozygous or compound heterozygous (e.g., HbSC) hemoglobin variants have no HbA and therefore no HbA1c. Thus, Capillarys 2 does not provide a value in these individuals. By contrast, boronate affinity methods (e.g., Ultra2), measure total glycated hemoglobin and calculate HbA1c, and would give an HbA1c result for these variant hemoglobins.
While several publications have evaluated analytical interference by hemoglobin variants (reviewed in Reference [35]), very little is known about their potential impact on the performance of HbA1c for the diagnosis of diabetes or prediabetes. After documenting that the Capillarys 2 accurately measures HbA1c in individuals with HbAS or HbAC, we evaluated its diagnostic performance in 109 asymptomatic African immigrants to the US. Laboratory analysis (see Table 2) revealed that the individuals did not have uremia, anemia or iron deficiency, conditions reported to interfere with HbA1c [22].The sensitivity of HbA1c on the Capillarys 2 for abnormal glucose tolerance in individuals with variant hemoglobin was slightly (but not significantly) higher than in those with normal hemoglobin. Analysis of additional individuals is required to ascertain if this initial observation is valid. Nevertheless, these data reveal that HbAS and HbAC do not impair the ability of HbA1c measured on the Capillarys 2 to detect abnormal glucose tolerance.
Although not significant, there was a trend towards higher sensitivity of HbA1c than FPG for abnormal glucose tolerance. These observations contrast with those of several prior studies. For example, the National Health and Nutrition Examination Survey (NHANES, 1988–2006) indicated that HbA1c detects a much lower prevalence of diabetes than FPG in the US population [10]. In the Epidemiological Study on the Insulin Resistance Syndrome (DESIR) cohort HbA1c identified significantly (P<0.0003) less diabetes in men than FPG: 5.3% and 7.5%, respectively [37]. Similarly, the number of individuals with diabetes identified by HbA1c was 70% of that identified by FPG in the Insulin Resistance Atherosclerosis Study (IRAS) [38]. For detecting individuals at increased risk of diabetes the superiority of FPG (over HbA1c) was less marked in African Americans than in whites [39]. Our data reveal that in immigrants to the US from equatorial African countries with a high prevalence of HbAS, HbA1c is slightly better than FPG at identifying abnormal glucose tolerance. It seems reasonable to postulate that the performance of HbA1c in Africa is likely to be similar to our observations.
A limitation of our study is the relatively small number of participants with both abnormal glucose tolerance and variant hemoglobin. A second limitation is the use of single measurements of FPG, 2h glucose and HbA1c; 2 measurements are preferable [1]. However, the detailed characterization of insulin secretion and insulin resistance provides reassurance that glucose homeostasis was not different between the normal and variant hemoglobin groups. Finally, abnormal glucose tolerance was defined by a single 2 h glucose. OGTTs are limited by a lack of reproducibility [22]. Nevertheless, a single OGTT is frequently used in clinical studies to identify diabetes and pre-diabetes [40, 41].
Overall, the analytical performance of Capillarys 2 HbA1c is within NACB/ADA [34] and NGSP [21] recommendations, has minimal interference, and is suitable for clinical application. To our knowledge, this is the first clinical investigation of HbA1c on the Capillarys 2 Flex Piercing instrument. Our cross sectional, population-based data derived from African immigrants with a high prevalence of variant hemoglobin reveal that sickle cell trait did not interfere with the ability of HbA1c (measured using the Capillarys 2) to detect abnormal glucose tolerance.
Supplementary Material
Highlights.
Analytical performance of Capillarys 2 is within NACB/ADA and NGSP recommendations
HbAS did not interfere with Capillarys 2 HbA1c to detect abnormal glucose tolerance
Capillarys 2 HbA1c is suitable for Africans with high prevalence of hemoglobinopathy
Acknowledgement
This research was supported by the Intramural Research Programs of the Clinical Center (ZZ, JB and DBS) and the National Institute of Diabetes and Digestive and Kidney Diseases (AES) of the National Institutes of Health. The authors are grateful to Sebia for providing equipment, reagents and technical support. We also thank Ms. Khanh Nghiem for analyzing the samples on the Bio-Rad D10 analyzer.
Abbreviations
- AIRg
Acute Insulin Response to Glucose
- FPG
Fasting Plasma Glucose
- NGSP
National Glycohemoglobin Standardization Program
- OGTT
Oral Glucose Tolerance Test
- SI
Insulin Sensitivity Index
- SRL
Secondary Reference Laboratory
- WC
Waist Circumference
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- 1.Standards of medical care in Diabetes--2014. Diab Care. 2014;37(Suppl 1):S14–S80. doi: 10.2337/dc14-S014. [DOI] [PubMed] [Google Scholar]
- 2.Mbanya JC, Henry RR, Smith U. Presidents' statement on WHO recommendation on HbA1c for Diabetes diagnosis. Diab Res Clin Pract. 2011;93:310–311. doi: 10.1016/j.diabres.2011.06.026. [DOI] [PubMed] [Google Scholar]
- 3.Sacks DB. Measurement of hemoglobin A(1c): a new twist on the path to harmony. Diab Care. 2012;35:2674–2680. doi: 10.2337/dc12-1348. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Little RR, Rohlfing CL, Sacks DB. Status of hemoglobin A1c measurement and goals for improvement: from chaos to order for improving Diab care. Clin Chem. 2011;57:205–214. doi: 10.1373/clinchem.2010.148841. [DOI] [PubMed] [Google Scholar]
- 5.Sebia, CAPILLARYS Hb A1c using the CAPILLARYS 2 Flex Piercing instrument. Package insert. 2013 [Google Scholar]
- 6.Michlitsch J, Azimi M, Hoppe C, et al. Newborn screening for hemoglobinopathies in California. Pediatr Blood Cancer. 2009;52:486–490. doi: 10.1002/pbc.21883. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Motulsky AG. Frequency of sickling disorders in U.S. blacks. N Engl J Med. 1973;288:31–33. doi: 10.1056/NEJM197301042880108. [DOI] [PubMed] [Google Scholar]
- 8.Allison AC. The distribution of the sickle-cell trait in East Africa and elsewhere, and its apparent relationship to the incidence of subtertian malaria. Trans R Soc Trop Med Hyg. 1954;48:312–318. doi: 10.1016/0035-9203(54)90101-7. [DOI] [PubMed] [Google Scholar]
- 9.Piel FB, Howes RE, Patil AP, et al. The distribution of haemoglobin C and its prevalence in newborns in Africa. Sci Rep. 2013;3:1671. doi: 10.1038/srep01671. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Cowie CC, Rust KF, Byrd-Holt DD, et al. Prevalence of Diabetes and high risk for Diabetes using A1C criteria in the U.S. population in 1988–2006. Diab Care. 2010;33:562–568. doi: 10.2337/dc09-1524. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Mongia SK, Little RR, Rohlfing CL, et al. Effects of hemoglobin C and S traits on the results of 14 commercial glycated hemoglobin assays. Am J Clin Pathol. 2008;130:136–140. doi: 10.1309/1YU0D34VJKNUCGT1. [DOI] [PubMed] [Google Scholar]
- 12.Roberts WL, Safar-Pour S, De BK, Rohlfing CL, Weykamp CW, Little RR. Effects of hemoglobin C and S traits on glycohemoglobin measurements by eleven methods. Clin Chem. 2005;51:776–778. doi: 10.1373/clinchem.2004.047142. [DOI] [PubMed] [Google Scholar]
- 13.Goodall I, Colman PG, Schneider HG, McLean M, Barker G. Desirable performance standards for HbA(1c) analysis - precision, accuracy and standardisation: consensus statement of the Australasian Association of Clinical Biochemists (AACB), the Australian Diabetes Society (ADS), the Royal College of Pathologists of Australasia (RCPA), Endocrine Society of Australia (ESA), and the Australian Diabetes Educators Association (ADEA) Clin Chem Lab Med. 2007;45:1083–1097. doi: 10.1515/CCLM.2007.158. [DOI] [PubMed] [Google Scholar]
- 14.Little RR, Vesper H, Rohlfing CL, Ospina M, Safar-Pour S, Roberts WL. Validation by a mass spectrometric reference method of use of boronate affinity chromatography to measure glycohemoglobin in the presence of hemoglobin S and C traits. Clin Chem. 2005;51:264–265. doi: 10.1373/clinchem.2004.043141. [DOI] [PubMed] [Google Scholar]
- 15.Sumner AE, Thoreson CK, O'Connor MY, et al. Detection of abnormal glucose tolerance in Africans is improved by combining A1C with fasting glucose: The Africans in America Study. Diab Care. 2015;38:213–219. doi: 10.2337/dc14-1179. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.O'Connor MY, Thoreson CK, Ricks M, et al. Worse cardiometabolic health in African immigrant men than African American men: reconsideration of the healthy immigrant effect. Metab Syndr Relat Disord. 2014;12:347–353. doi: 10.1089/met.2014.0026. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Ukegbu UJ, Castillo DC, Knight MG, et al. Metabolic syndrome does not detect metabolic risk in African men living in the U.S. Diab Care. 2011;34:2297–2299. doi: 10.2337/dc11-1055. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Sumner AE, Luercio MF, Frempong BA, et al. Validity of the reduced-sample insulin modified frequently-sampled intravenous glucose tolerance test using the nonlinear regression approach. Metabolism. 2009;58:220–225. doi: 10.1016/j.metabol.2008.09.017. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Boston RC, Stefanovski D, Moate PJ, Sumner AE, Watanabe RM, Bergman RN. MINMOD Millennium: a computer program to calculate glucose effectiveness and insulin sensitivity from the frequently sampled intravenous glucose tolerance test. Diab Technol Ther. 2003;5:1003–1015. doi: 10.1089/152091503322641060. [DOI] [PubMed] [Google Scholar]
- 20.Rohlfing CL, Parvin CA, Sacks DB, Little RR. Comparing analytic performance criteria: evaluation of HbA1c certification criteria as an example. Clin Chim Acta. 2014;433:259–263. doi: 10.1016/j.cca.2014.03.034. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. [Accessed March 2015]; http://www.ngsp.org/ [Google Scholar]
- 22.Sacks DB. A1C versus glucose testing: a comparison. Diab Care. 2011;34:518–523. doi: 10.2337/dc10-1546. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Marinova M, Altinier S, Caldini A, et al. Multicenter evaluation of hemoglobin A1c assay on capillary electrophoresis. Clin Chim Acta. 2013;424:207–211. doi: 10.1016/j.cca.2013.06.014. [DOI] [PubMed] [Google Scholar]
- 24.Urrechaga E. High-resolution HbA(1c) separation and hemoglobinopathy detection with capillary electrophoresis. Am J Clin Pathol. 2012;138:448–456. doi: 10.1309/AJCPVYW9QZ9EVFXI. [DOI] [PubMed] [Google Scholar]
- 25.Jaisson S, Leroy N, Meurice J, Guillard E, Gillery P. First evaluation of Capillarys 2 Flex Piercing(R) (Sebia) as a new analyzer for HbA1c assay by capillary electrophoresis. Clin Chem Lab Med. 2012;50:1769–1775. doi: 10.1515/cclm-2012-0017. [DOI] [PubMed] [Google Scholar]
- 26.Weykamp C, Waenink-Wieggers H, Kemna E, Siebelder C. HbA1c: performance of the Sebia Capillarys 2 Flex Piercing. Clin Chem Lab Med. 2013;51:e129–e131. doi: 10.1515/cclm-2012-0560. [DOI] [PubMed] [Google Scholar]
- 27.Heylen O, Van Neyghem S, Exterbille S, Wehlou C, Gorus F, Weets I. Evaluation of the Sebia CAPILLARYS 2 flex piercing for the measurement of HbA(1c) on venous and capillary blood samples. Am J Clin Pathol. 2014;141:867–877. doi: 10.1309/AJCPRU5QC2JBANSV. [DOI] [PubMed] [Google Scholar]
- 28.Petersen JR, Okorodudu AO, Mohammad A, Payne DA. Capillary electrophoresis and its application in the clinical laboratory. Clin Chim Acta. 2003;330:1–30. doi: 10.1016/s0009-8981(03)00006-8. [DOI] [PubMed] [Google Scholar]
- 29.Greene DN, Pyle AL, Chang JS, Hoke C, Lorey T. Comparison of Sebia Capillarys Flex capillary electrophoresis with the BioRad Variant II high pressure liquid chromatography in the evaluation of hemoglobinopathies. Clin Chim Acta. 2012;413:1232–1238. doi: 10.1016/j.cca.2012.03.027. [DOI] [PubMed] [Google Scholar]
- 30.Doelman CJ, Siebelder CW, Nijhof WA, Weykamp CW, Janssens J, Penders TJ. Capillary electrophoresis system for hemoglobin A1c determinations evaluated. Clin Chem. 1997;43:644–648. [PubMed] [Google Scholar]
- 31.The effect of intensive treatment of Diabetes on the development and progression of long-term complications in insulin-dependent Diabetes mellitus. The Diab Control and Complications Trial Research Group. N Engl J Med. 1993;329:977–986. doi: 10.1056/NEJM199309303291401. [DOI] [PubMed] [Google Scholar]
- 32.Intensive blood-glucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 Diabetes (UKPDS 33). UK Prospective Diabetes Study (UKPDS) Group. Lancet. 1998;352:837–853. [PubMed] [Google Scholar]
- 33.Type 2 Diabetes: newer agents. National Institute for Health and Clinical Excellence. [PubMed] [Google Scholar]
- 34.Sacks DB, Arnold M, Bakris GL, et al. Guidelines and recommendations for laboratory analysis in the diagnosis and management of Diabetes mellitus. Clin Chem. 2011;57:e1–e47. doi: 10.1373/clinchem.2010.161596. [DOI] [PubMed] [Google Scholar]
- 35.Bry L, Chen PC, Sacks DB. Effects of hemoglobin variants and chemically modified derivatives on assays for glycohemoglobin. Clin Chem. 2001;47:153–163. [PubMed] [Google Scholar]
- 36.Lin CN, Emery TJ, Little RR, et al. Effects of hemoglobin C, D, E, and S traits on measurements of HbA1c by six methods. Clin Chim Acta. 2012;413:819–821. doi: 10.1016/j.cca.2011.12.019. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Balkau B, Soulimane S, Lange C, Gautier A, Tichet J, Vol S. Are the same clinical risk factors relevant for incident Diabetes defined by treatment, fasting plasma glucose, and HbA1c? Diab Care. 2011;34:957–959. doi: 10.2337/dc10-1581. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Lorenzo C, Haffner SM. Performance characteristics of the new definition of Diabetes: the insulin resistance atherosclerosis study. Diab Care. 2010;33:335–337. doi: 10.2337/dc09-1357. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Lorenzo C, Wagenknecht LE, Hanley AJ, Rewers MJ, Karter AJ, Haffner SM. A1C between 5.7 and 6.4% as a marker for identifying pre-Diabetes, insulin sensitivity and secretion, and cardiovascular risk factors: the Insulin Resistance Atherosclerosis Study (IRAS) Diab Care. 2010;33:2104–2109. doi: 10.2337/dc10-0679. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Kramer CK, Araneta MR, Barrett-Connor E. A1C and Diabetes diagnosis: The Rancho Bernardo Study. Diab Care. 2010;33:101–103. doi: 10.2337/dc09-1366. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Olson DE, Rhee MK, Herrick K, Ziemer DC, Twombly JG, Phillips LS. Screening for Diabetes and pre-Diab with proposed A1C-based diagnostic criteria. Diab Care. 2010;33:2184–2189. doi: 10.2337/dc10-0433. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.