Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Comparison of Enzymatic Assay for HBA1C Measurement (Abbott Architect) With Capillary Electrophoresis (Sebia Minicap Flex Piercing Analyser)

Comparison of Enzymatic Assay for HBA1C Measurement (Abbott Architect) With Capillary... Abstract Objective To compare the analytical performances of the enzymatic method (EM) and capillary electrophoresis (CE) for hemoglobin A1c (HbA1c) measurement. Methods Imprecision, carryover, stability, linearity, method comparison, and interferences were evaluated for HbA1c via EM (Abbott Laboratories, Inc) and CE (Sebia). Results Both methods have shown overall within-laboratory imprecision of less than 3% for International Federation of Clinical Chemistry and Laboratory Medicine (IFCC) units (<2% National Glycohemoglobin Standardization Program [NGSP] units). Carryover effects were within acceptable criteria. The linearity of both methods has proven to be excellent (R2 = 0.999). Significant proportional and constant difference were found for EM, compared with CE, but were not clinically relevant (<5 mmol/mol; NGSP <0.5%). At the clinically relevant HbA1c concentration, stability observed with both methods was acceptable (bias, <3%). Triglyceride levels of 8.11 mmol per L or greater showed to interfere with EM and fetal hemoglobin (HbF) of 10.6% or greater with CE. Conclusion The enzymatic method proved to be comparable to the CE method in analytical performances; however, certain interferences can influence the measurements of each method. glycated hemoglobin, electrophoresis, enzymatic methods, imprecison, lipemia, fetal hemoglobin Glycated hemoglobin is composed of hemoglobin (Hb)A1a, HbA1b, and HbA1c. The major fraction of glycated hemoglobin, HbA1c, is formed by nonenzymatic process of binding glucose to the N-terminal valine of the β-chain of hemoglobin.1 The American Diabetes Association (ADA) and World Health Organization (WHO) have confirmed the use of HbA1c as a biomarker for diagnosing and monitoring of diabetes mellitus.2 Two major clinical studies, the Diabetes Control and Complication Trial (DCCT) and the United Kingdom Prospective Diabetes Study (UKPDS), have demonstrated that intensive therapy and monitoring of diabetes mellitus type 1 (T1DM) and type 2 diabetes mellitus (T2DM) can decrease complications of diabetes.3,4 In July 2002, the International Federation of Clinical Chemistry and Laboratory Medicine (IFCC) approved reference methods for HbA1c. Two combinations of methods were used for separation and quantification of HbA1c, namely, high-performance liquid chromatography (HPLC) and electrospray ionization mass spectrometry and 2-dimensional approach using HPLC and capillary electrophoresis (CE) with ultraviolet (UV) detection.5 Studies have shown that other Hb variants (HbS, HbE, HbC, and HbD) do not interfere with HbA1c measurement by CE and enzymatic methods (EMs), but HbF might be a significant source of interference error.6,7 CE and EMs correlate well with the HPLC method,8,9 but there are not many studies comparing EMs with CE. The aim of this study was to perform verification of analytical specifications (precision, carryover, linearity, interferences, and stability) and method comparison for 2 different assays for HbA1c measurement, namely, direct whole blood enzymatic assay (Abbott Laboratories, Inc) and CE (Sebia). Materials and Methods Study Design Blood specimens used for this study were leftover patient specimens referred to the laboratory for routine HbA1c testing. All blood specimens were collected in ethylenediaminetetraacetic acid (EDTA) tubes (Vacuette tripotassium (K3 )EDTA Tubes, 4 mL; Greiner Bio-One International GmbH). The study was conducted from April 2015 through December 2016. First, we performed a comparison of 2 different assays for HbA1c measurement. Precision, carryover, stability, linearity, and interferences were tested during the specified period for both methods. The precision study was performed following the Clinical Laboratory Standards Institute (CLSI) EP15-A2 guideline.10 Methods CE: Minicap Flex Piercing Analyzer The Minicap Flex Piercing (Sebia) is a 2-silica-capillary automated analyzer that uses whole blood specimens from primary capped tubes for HbA1c testing. Using CE, hemoglobin variants are separated by high voltage (9400 V) in an alkaline buffer solution (pH, 9.4) and detected at 415 nm. Phoresis software (version 8.6.3) integrates HbA1c and HbA0 peaks, and the HbA1c concentration is calculated from the ratio HbA1c / (HbA1c + HbA0). EM: Abbott ARCHITECT c8000 Analyzer The enzymatic Hemoglobin A1c assay (Abbott Laboratories, Inc) measures HbA1c and total hemoglobin (THb) from the whole blood specimen. This method specifically measures N-terminal fructosyl dipeptides of the β-chain of HbA1c cleaved by a fructosyl peptide oxidase. The HbA1c concentration is calculated from the ratio of HbA1c and THb: HbA1c (mmol/mol) = (HbA1c/THb) × 1000.11 The results are expressed in IFCC units (mmol/mol) and DCCT/National Glycohemoglobin Standardization Program (NGSP) units (%) after conversion using the IFCC-NGSP master equation HbA1c (%) = (0.09148 × IFCC) + 2.152.8,12,13 Assay specifications are featured in Table S1 (available online-only). Verification of Analytical Specifications of the Methods and Comparison Studies Precision For this purpose, patient blood specimens have been used for both methods and commercial quality controls for EM only. All specimens used for precision testing were analysed in triplicate, during a 5-day period. Precision study of Minicap Flex Piercing analyzer. Three patient specimens with low (L; 31 mmol/mol [5.0%]), medium (M; 43 mmol/mol [6.1%]), and high (H; 98 mmol/mol [11.1%]) concentration of HbA1c were chosen for precision studies. Each specimen was divided into 2 parts to test both capillaries simultaneously. Precision study of EM. For precision study of EM, 1 patient specimen (S1 = 53 mmol/mol [7.0%]) and 2 commercial control specimens (Lyphocheck Diabetes Level 1 = 28 mmol/mol [4.7%] and Level 2 = 79 mmol/mol [9.4%]; Bio-Rad Laboratories, Inc) were chosen. Acceptance criteria for intralaboratory percentage coefficient of variation (%CV) was set at less than 3% (for results expressed in IFCC units) and less than 2% (for results expressed in DCCT/NGSP units).14 Carryover Carryover study of Minicap Flex Piercing Analyzer. We tested carryover for the Minicap Flex Piercing analyzer during a 3-day period, using 2 specimens, 1 with normal and the other with high HbA1c concentration. Specimens were divided into 2 tubes to test the carryover for both capillaries at the same time. Specimens with normal HbA1c concentration were analyzed before and after testing of the specimen with high HbA1c concentration. Specimens were analyzed once per day during the 3-day period. Carryover study for EM. We tested carryover for EM using 2 protocols, specifically evaluating analyte carryover between 2 whole blood specimens and whole blood and serum specimens. In the first protocol, 1 whole blood specimen with normal HbA1c concentration was divided into 2 aliquots and analyzed before and after the specimen with the high HbA1c concentration. The second protocol was designed because serum and whole blood specimens are tested on the same analyzers (Abbott Architect c8000) during routine laboratory work. In this protocol, 1 serum specimen was divided into 2 aliquots and analyzed before and after the whole blood specimen. To evaluate potential contamination with erythrocytes from the whole blood, in serum specimen potassium, magnesium, calcium, phosphorus, lactate dehydrogenase, aspartate aminotransferase (AST), total proteins, and glucose were measured on the Architect c8000 (with the original manufacturers reagents). Measurements were performed in triplicate, and average values were used for calculation. For all methods, carryover bias was calculated as follows: Bias=([parameter valueafter−parameter valuebefore]/parameter valuebefore)×100% Acceptance criterion for HbA1c carryover was set at 3% and for all other analytes, desirable specification for bias was taken from Ricos et al15; these data are listed in Table S2 (available online-only). Stability Stability was evaluated by measuring HbA1c concentration in the same specimen during 12 consecutive days for EM and 14 days for CE. Specimens were stored in the refrigerator (2°–8°C) between measurements. For each day, concentration of HbA1c was compared with the concentration obtained on the first day of measurement by calculating bias (%). Three specimens with different HbA1c concentrations were used for both methods. A criterion for stability bias was set at 3% or less.15 Method Comparison Comparison of these 2 methods was conducted during an 8-day period using 93 patient specimens. These specimens were analyzed by EM, stored in the refrigerator (2°–8°C) for a maximum of 3 days, and then analyzed by CE. Linearity The linearity was evaluated by measuring HbA1c in 5 specimens obtained with serial dilutions of the whole blood specimen with high HbA1c concentration (156 mmol/mol [16.4%] for CE and 101 mmol/mol [11.4%] for EM) with whole blood specimens with low HbA1c concentration (32 mmol/mol [5.1%] for CE and 31 mmol/mol [5.0%] for EM). Each specimen has been assayed in duplicate and its mean value plotted against the expected value for regression analysis. Interferences Both methods were tested for the interference of lipemia using Intralipid 20% emulsion for infusion (Fresenius Kabi AG). Intralipid was added in increased concentration into the whole blood specimen, and the bias of measured HbA1c was calculated for each one.16,17 We evaluated the interference of HbF according to a protocol in which whole blood specimens were spiked with an increasing volume of pooled fresh whole blood from a neonate (left over from routine complete blood count [CBC] testing), obtaining specimens with different HbF fractions. HbA1c was measured and bias calculated for each specimen by comparison with the same specimen without HbF added. The level of HbF was estimated by using the percentage of HbF, as determined by CE of hemoglobin (CAPILLARYS 2, Sebia). For lipemia and HbF interferences measurement, HbA1c was determined in duplicate by EM; for CE, we performed a single measurement. Interference of labile HbA1c was evaluated by incubating washed erythrocytes from pooled blood in 3 levels (L = 29 mmol/mol [4.8%]; M = 45 mmol/mol [6.3%]; and H = 110 mmol/mol [12.2%], of HbA1c measured by CE) with various glucose solutions (15 mmol/L, 30 mmol/L, and 60 mmol/L). Washed erythrocytes were incubated at 37°C for 3 hours and mixed every 30 minutes.18,19 We examined the interference of carbamylated hemoglobin (cHb) by in vitro carbamylation of hemoglobin. Washed erythrocytes of pooled blood in 3 levels (L = 27 mmol/mol [4.6%]; M = 45 mmol/mol [6.3%]; H = 96 mmol/mol [10.9%], of HbA1c measured by CE) were incubated for 3 hours at 37°C with 2.5 and 5.0 mmol per L of KOCN.19,20 HbA1c in all 3 pools for labile HbA1c and cHb interference was measured via EM and CE, before and after incubation, and interferents and biases were calculated. All measurements were made in duplicate. For all interferences studies, a criterion for acceptable bias was set at 6% or less.21 Statistical Analysis For comparison of 2 methods, results were analyzed by the Bland-Altman plot.22 The difference between methods is presented on the y axis for evaluation of constant difference; the average of both methods is presented on the x axis. Statistical analysis was performed using MedCalc software, version 11.5.1.0 (MedCalc Software). Results Precision Study Obtained repeatability and intralaboratory precision data are summarized in Table S3 (available online-only). All coefficients of variation were within acceptance criteria for both methods (<3% for IFCC units, <2% for NGSP units). Carryover Study Carryover testing between high and low HbA1c concentrations in whole blood specimens obtained acceptable mean biases for EM (0.27%) and CE (−1.0%). Also, in testing of carryover of whole blood to serum, all bias values were within acceptable criteria (Table S2, available online-only). Stability Study For EM, biases were within the acceptance criteria for all specimens (Figure S1, available online-only). In the case of CE, biases were acceptable for the specimens with medium and high HbA1c concentration but not for specimen with low HbA1c concentration (32 mmol/mol [5.1%]), with obtained bias as high as 9.38% on the ninth day (Figure S2, available online-only). Figure 1 View largeDownload slide Bland-Altman plot for constant difference between hemoglobin (Hb)A1c measurement (National Glycohemoglobin Standardization Program [NGSP] units) via the Abbott ARCHITECT enzymatic method and capillary electrophoresis method (Minicap Flex Piercing analyzer); n = 93. Figure 1 View largeDownload slide Bland-Altman plot for constant difference between hemoglobin (Hb)A1c measurement (National Glycohemoglobin Standardization Program [NGSP] units) via the Abbott ARCHITECT enzymatic method and capillary electrophoresis method (Minicap Flex Piercing analyzer); n = 93. Method Comparison Study Bland-Altman analysis for comparison of EM and CE revealed significant constant (mean, 1.1 mmol/mol [0.11%]) (Figure 1 and Figure S3 [available online-only]) and proportional (mean, 2.3% [for NGSP units, 1.6%]) differences (data not shown). Limits of agreement (±1.96 SD of differences) ranged from −1.9 to 4.0 mmol/mol (−0.2% to 0.4%) for constant difference (Figure 1 and Figure S3 [available online-only]) and −4.0 to 8.6% (NGSP, −2.6% to 5.8%) for proportional difference (data not shown), respectively. EM underestimates the HbA1c value, compared with CE. Linearity of the Methods The linearity of both methods has proven to be excellent within the examined ranges of HbA1c. Regression equation for the CE method (range, 32–156 mmol/mol, 5.1%–16.4%) was y = −0.15 (95% confidence interval [CI], −6.13 to 5.83) + 1.00 (0.94 to 1.06) x. For the EM (range, 31–101 mmol/mol; 5.0%–11.4%), y = 0.64 (95% CI, −2.08 to 3.36) + 1.00 (0.96 to 1.04) x, with R2 values of 0.999 for both methods. Interferences The results of a study of lipemia interference showed that triglyceride levels as high as 42.37 mmol per L have no impact on the HbA1c determination by CE (bias, <3.0%). For EM, the acceptable bias for triglyceride levels as high as 8.11 mmol per L has been obtained; however, increasing triglyceride concentration afterward causes a steep decrease of HbA1c (Figure 2). Figure 2 View largeDownload slide Interference of lipemia on hemoglobin (Hb)A1c measurement via the enzymatic method and capillary electrophoresis method. Interference is presented with mean bias (%) for EM and bias (%) from single measurement for CE. Biases for both IFCC and NGSP units are presented. Figure 2 View largeDownload slide Interference of lipemia on hemoglobin (Hb)A1c measurement via the enzymatic method and capillary electrophoresis method. Interference is presented with mean bias (%) for EM and bias (%) from single measurement for CE. Biases for both IFCC and NGSP units are presented. The presence of HbF already at the level of 10.6% influences the HbA1c measurement by CE, resulting in underestimation of the HbA1c concentration (bias, 7.7%). At the same time, no significant impact on HbA1c measurement by EM has been detected up to HbF of 39.5% (Figure 3). Figure 3 View largeDownload slide Interference of HbF (%) on hemoglobin (Hb)A1c measurement via the enzymatic method and capillary electrophoresis method. Interference is presented with mean bias (%) for EM and bias (%) from single measurement for CE. Biases for both IFCC and NGSP units are presented. Figure 3 View largeDownload slide Interference of HbF (%) on hemoglobin (Hb)A1c measurement via the enzymatic method and capillary electrophoresis method. Interference is presented with mean bias (%) for EM and bias (%) from single measurement for CE. Biases for both IFCC and NGSP units are presented. No significant impact of labile HbA1c was found for either method, with biases as high as −4.4% for EM and as high as −2.3% for CE. For cHb interference, no significant bias was detected for either method, with 2.5 mM potassium cyanate. However, with 5 mM potassium cyanate, both methods showed significant interference but with opposite trends regarding HbA1c concentration. For EM, bias was acceptable (−5.6%) only at low HbA1c levels and increased with HbA1c concentration (−7.1% and −8.1%, respectively). With CE, bias was acceptable (3.1%) only at the high HbA1c level and increased as high as 20.0% at normal and unmeasurable points at the low HbA1c concentration. Discussion The results of this study confirmed differences between the Abbott EM assay and the Sebia Minicap CE for HbA1c concentration measurement. All tested analytical performances were in accordance with the previously defined criteria, with the exception of the lipemia interference on EM and HbF interference on the capillary method of HbA1c measurement. Herein, we compared the analytical performances of 2 different methods for HbA1c measurement. Due to the recommended role of HbA1c as a biomarker for the diagnosis of diabetes (criterion, ≥48 mmol/mol [6.5%]) and management of diabetes complications (therapeutic target, <53 mmol/mol [7%]), as well as an indicator for change of therapy (criterion, ≥64 mmol/mol [8%]), assays with high analytical precision are necessary to differentiate between the upper limit of the nondiabetic range (42 mmol/mol [6.0%) and aforementioned targets.2 Therefore, highly stringent precision requirements for HbA1c assays are necessary. Based on intra- and interindividual biologic variation, intralaboratory CV of less than 3% (<2% for NGSP units) is recommended.15,23,24 This recommendation was supported by the mathematical minimal ability to differentiate the upper normal concentration of 42 mmol/mol (6.0%) and diagnostic criterion of 48 mmol/mol (6.5%).23 In our study, both methods achieved the criterion for intralaboratory imprecision, with an even more stringent goal of less than 2% (recommended for NGSP units) achieved for medium and high HbA1c concentrations expressed in IFCC units. This finding is important because it encompasses diagnostic and therapeutic targets. In comparison with the findings of the first published evaluation of the Sebia machine as a new analyzer for HbA1c assay by CE, the results of our study showed lower coefficients of variation. The reason for this finding may be the difference between the analyzers used. Whereas the CAPILLARYS 2 Flex piercing analyzer, which was used in the initial study, uses 8 capillaries, the Minicap-Flex piercing analyzer uses only 2 capillaries, thus lowering the impact of each capillary on overall imprecision.8 Performing analysis of serum/plasma and whole blood specimens on the same analyzers might represent a risk due to potential contamination and/or carryover. Although analyzers include additional washing cycles in their protocols, one might be concerned that serum specimens analyzed after full-blood specimens might be falsely hemolyzed, due to the carryover of erythrocytes. Therefore, one of the strengths of our study is the comprehensive investigation of carryover effect on biochemistry analyzer, including analytes that are most vulnerable to the hemolysis effect. Both methods proved to be reliable, considering carryover effect between the whole blood specimens. Also, carryover effect obtained between whole blood and serum specimens was far below the desirable specifications for bias used as acceptance criteria.15 Stability studies for CE and EM at the clinically relevant concentration of HbA1c, showed overall bias lower than 2% during the entire tested period (12 days and 14 days, respectively). EM showed better overall stability; however, we were surprised to find high deviation (9.8%) for CE at a low HbA1c concentration (32 mmol/mol [5.1%]). However, even such high deviation has no impact on clinical interpretation. A combination of HPLC and CE has been approved by the IFCC as the reference method for HbA1c.5 Recent studies8,18,25 have shown that results obtained by the HPLC method and the newly developed method that uses CE for the separation and quantification of HbA1c correlate well. As a result, in our study, we used CE as a reference method against which we compared HbA1c results measured via EM. According to Bland-Altman analysis of difference between methods, significant proportional and constant differences were found for EM compared with CE. However, a difference of 1.1 mmol/mol (0.11%) is not clinically relevant with total allowable error (TAE) buffer set at 5 mmol/mol (0.5%).26 Our results showed minimal interference from lipemia for the Abbott EM at triglyceride concentration of 8.11 mmol per L or greater. However, in our laboratory, this finding impacts only 0.3% to 0.4% of specimens. Data obtained using synthetic lipid emulsion might differ in comparison to the pathophysiologically induced lipemia experienced by patients with diabetes. Nowadays, most analytical platforms use automatic detection and assessment of the degree of lipemia as an L-index, enabling the laboratory professional to avoid missing the potential interferent. Increase of circulating HbF, defined as HbF greater than >2%, occurs relatively commonly and is associated with pathologic conditions such as leukemia, thalassemia, and hereditary persistence of Hbf. Individuals with the most common form of hereditary persistence of Hbf can have HbF levels as high as 30%.27 Because these patients generally experience no symptoms, it is crucial to know whether your method of HbA1c measurement is influenced by the presence of HbF and if so, at what proportion of HbF. The results of our study have shown that HbA1c measurement by CE is influenced by the presence of HbF at a lower proportion (10.6%) than that declared by the manufacturer (15%). Our data are in concordance with those reported in a recent study in which significant interference was observed for HbA1c measurement via the CAPILLARYS 2 FP analyzer for specimens with 10% to 15% HbF (relative bias, >±7%).18 Hemoglobin F has a lower glycation rate than HbA because approximately 60% of glycation of HbA occurs at the amino terminus of the β chain, and HbF has γ in place of β chains.28 Because HbF migrates close to HbA0, separation of fractions is not efficient enough if one overestimates the HbA0 proportion, which artificially lowers the HbA1c level. However, the presence of an increased level of HbF is obvious from an interfering peak in the chromatogram and therefore, misinterpretation of HbA1c concentration can be safely avoided. The results of our study indicate an insignificant influence of HbF levels as high as 55% on HbA1c measurement with EM, which is much higher than the result of the recent study in which the bias for the same method is within 6% for HbF less than 13%.29 This discrepancy possibly could be attributed to the protocol applied for interference investigation. In this study, bias was obtained by comparing the HbA1c concentrations obtained using EM with the IFCC reference method for specimens with abnormal hemoglobins or β0-thalassemia with HbF concentrations between 1.4% and 90.1%. Our results showed that labile HbA1c has no significant impact on HbA1c measurement via EM and CE. Regarding cHb interference, significant underestimation of HbA1c concentration was observed in the EM, with overestimation in CE but only at 5 mM potassium cyanate. Discrepancies in the trend of interference can be attributed to differences in calculation of HbA1c between methods. However, because such a high concentration of interferent is not physiologic, this bias can be neglected.30,31 In our study, the interference of hemoglobin variants on HbA1c measurement was limited to HbF only because based on the only available data on the frequency of hemoglobinopathic manifestations in Croatia, other variants in hemoglobin are rare (0.8%).32 Therefore, we believe that the results of our study are applicable to our population. A limitation to this study might be that we did not investigate the influence of acetylated HbA1c. Also, other interferences listed in the manufacturer declarations for the CE method have not been investigated (bilirubin, urea, ascorbic acid, rheumatoid factor, glibenclamide). However, we believe that our study results still provide valuable data on potential interferences in HbA1c testing. In conclusion, the analytical performances of HbA1c, as tested via EM and CE, are comparable, and both methods satisfy highly stringent analytical performance requirements; however, certain interferences can influence the measurements of each method. EM has the advantage of increased capacity and suitability for general chemistry analyzers. In contrast, CE has the important advantage of its ability to detect biological interference from hemoglobin variants that would affect the erythrocyte life span and lead to clinically misinterpreted HbA1c results. Abbreviations Hb hemoglobin ADA American Diabetes Association WHO World Health Organization DCCT Diabetes Control and Complication Trial UKPDS United Kingdom Prospective Diabetes Study T1DM diabetes mellitus type 1 T2DM type 2 diabetes mellitus IFCC International Federation of Clinical Chemistry and Laboratory Medicine HPLC high-performance liquid chromatography CE capillary electrophoresis UV ultraviolet EM enzymatic methods EDTA ethylenediaminetetraacetic acid K3 tripotassium CLSI Clinical Laboratory Standards Institute THb total hemoglobin NGSP National Glycohemoglobin Standardization Program %CV percentage coefficient of variation TAE total allowable error C1 first capillary tested C2 second capillary tested DSB desired specification for bias Acknowledgments This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors. References 1. Bunn HF , Haney DN , Kamin S , Gabbay KH , Gallop PM . The biosynthesis of human hemoglobin A1c. Slow glycosylation of hemoglobin in vivo . J Clin Invest . 1976 ; 57 ( 6 ): 1652 – 1659 . 2. American Diabetes Association . Diagnosis and classification of diabetes mellitus . Diabetes Care . 2014 ; 37 ( Suppl 1 ): S81 – S90 . 3. The Diabetes Control and Complications Trial Research Group . The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus . N Engl J Med . 1993 ; 329 : 977 – 986 . 4. UK Prospective Diabetes Study (UKPDS) Group . Intensive blood-glucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes (UKPDS 33) . Lancet . 1998 ; 352 ( 9131 ): 837 – 853 . 5. Jeppsson JO , Kobold U , Barr J , et al. ; International Federation of Clinical Chemistry and Laboratory Medicine (IFCC) . Approved IFCC reference method for the measurement of HbA1c in human blood . Clin Chem Lab Med . 2002 ; 40 ( 1 ): 78 – 89 . 6. National Glycohemoglobin Standardization Program (NGSP) . Factors that Interfere with HbA1c Test Results . NGSP website. http://www.ngsp.org/factors.asp. Accessed January 5, 2018 . 7. Teodoro-Morrison T , Janssen MJ , Mols J , et al. Evaluation of a next generation direct whole blood enzymatic assay for hemoglobin A1c on the ARCHITECT c8000 chemistry system . Clin Chem Lab Med . 2015 ; 53 ( 1 ): 125 – 132 . 8. Jaisson S , Leroy N , Meurice J , Guillard E , Gillery P . First evaluation of Capillarys 2 Flex Piercing® (Sebia) as a new analyzer for HbA1c assay by capillary electrophoresis . Clin Chem Lab Med . 2012 ; 50 ( 10 ): 1769 – 1775 . 9. Ajubi NE , Kross-Brion J , Newton CJ . Performance characteristics of the enzymatic Abbott Architect HbA1c whole blood assay . Clin Chem Lab Med . 2014 ; 52 ( 12 ): e283 – e285 . 10. Clinical and Laboratory Standards Institute . User Verification of Precision and Estimation of Bias: Approved Guideline . 2nd edition. CLSI document EP15-A2. Wayne, PA : Clinical and Laboratory Standards Institute , 2005 . 11. ARCHITECT System [package insert]. Abbott Park, IL: Abbott Laboratories, Inc; 2013. 12. Hoelzel W , Weykamp C , Jeppsson JO , et al. ; IFCC Working Group on HbA1c Standardization . IFCC reference system for measurement of hemoglobin A1c in human blood and the national standardization schemes in the United States, Japan, and Sweden: a method-comparison study . Clin Chem . 2004 ; 50 ( 1 ): 166 – 174 . 13. CAPILLARYS HbA1c . Using the Capillarys 2 Flex-Piercing Instrument . Sebia website. http://www.ilexmedical.com/files/Sebia%20inserts/CAPILLARYS_Hb_A1c.pdf. Accessed January 5, 2018 . 14. Braga F , Panteghini M . Standardization and analytical goals for glycated hemoglobin measurement . Clin Chem Lab Med . 2013 ; 51 ( 9 ): 1719 – 1726 . 15. Desirable Biological Variation Database specifications . Available at: http://www.westgard.com/biodatabase1.htm. Accessed December 6, 2015. 16. Nikolac N . Lipemia: causes, interference mechanisms, detection and management . Biochem Med (Zagreb) . 2014 ; 24 ( 1 ): 57 – 67 . 17. Clinical Laboratory Standards Institute . Hemolysis, Icterus, and Lipemia/Turbidity Indices as Indicators of Interference in Clinical Laboratory Analysis; Approved Guideline. CLSI C56-A document . Wayne, PA : Clinical Laboratory Standards Institute ; 2012 . 18. Wu X , Chao Y , Wan Z , et al. A comparative evaluation of the analytical performances of Capillarys 2 Flex Piercing, Tosoh HLC-723 G8, Premier Hb9210, and Roche Cobas c501 Tina-quant Gen 2 analyzers for HbA1c determination . Biochem Med (Zagreb) . 2016 ; 26 ( 3 ): 353 – 364 . 19. 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 ( 6 ): 867 – 877 . 20. Chachou A , Randoux C , Millart H , Chanard J , Gillery P . Influence of in vivo hemoglobin carbamylation on HbA1c measurements by various methods . Clin Chem Lab Med . 2000 ; 38 ( 4 ): 321 – 326 . 21. Little RR . Performance of hemoglobin A1c assay methods: good enough ? Clin Chem . 2014 ; 60 ( 8 ): 1031 – 1033 . 22. Giavarina D . Understanding Bland Altman analysis . Biochem Med (Zagreb) . 2015 ; 25 ( 2 ): 141 – 151 . 23. 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 ( 8 ): 1083 – 1097 . 24. Sacks DB , Arnold M , Bakris GL , et al. Executive summary: guidelines and recommendations for laboratory analysis in the diagnosis and management of diabetes mellitus . Clin Chem . 2011 ; 57 ( 6 ): 793 – 798 . 25. Woodworth A , Korpi-Steiner N , Miller JJ , et al. Utilization of assay performance characteristics to estimate hemoglobin A1c result reliability . Clin Chem . 2014 ; 60 ( 8 ): 1073 – 1079 . 26. Weykamp C , John G , Gillery P , et al. ; IFCC Task Force on Implementation of HbA1c Standardization . Investigation of 2 models to set and evaluate quality targets for HB A1c: biological variation and sigma-metrics . Clin Chem . 2015 ; 61 ( 5 ): 752 – 759 . 27. Little RR , Roberts WL . A review of variant hemoglobins interfering with hemoglobin A1c measurement . J Diabetes Sci Technol . 2009 ; 3 ( 3 ): 446 – 451 . 28. Rohlfing CL , Connolly SM , England JD , et al. The effect of elevated fetal hemoglobin on hemoglobin A1c results: five common hemoglobin A1c methods compared with the IFCC reference method . Am J Clin Pathol . 2008 ; 129 ( 5 ): 811 – 814 . 29. Nitta T , Yamashiro Y , Hattori Y , Ezumi T , Nishioka M , Nakamura J . The interference by HbF on HbA1c (BM Test HbA1c) measurement in enzymatic method . Ann Clin Biochem . 2015 ; 52 ( Pt 5 ): 569 – 575 . 30. Herpol M , Lanckmans K , Van Neyghem S , et al. Evaluation of the Sebia Capillarys 3 Tera and the Bio-Rad D-100 Systems for the Measurement of Hemoglobin A1c . Am J Clin Pathol . 2016 ; 146 ( 1 ): 67 – 77 . 31. Jaisson S , Desmons A , Renard B , Chevelle B , Leroy N , Gillery P . Analytical performances of a new enzymatic assay for hemoglobin A1c . Clin Chim Acta . 2014 ; 434 : 48 – 52 . 32. Efremov GD . Hemoglobinopathies in Yugoslavia: an update . Hemoglobin . 1992 ; 16 ( 6 ): 531 – 544 . © American Society for Clinical Pathology 2018. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Laboratory Medicine Oxford University Press

Comparison of Enzymatic Assay for HBA1C Measurement (Abbott Architect) With Capillary Electrophoresis (Sebia Minicap Flex Piercing Analyser)

Download PDF
8 pages

Loading next page...
 
/lp/ou_press/comparison-of-enzymatic-assay-for-hba1c-measurement-abbott-architect-O6Rt4BqzNd
Publisher
Oxford University Press
Copyright
© American Society for Clinical Pathology 2018. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com
ISSN
0007-5027
eISSN
1943-7730
DOI
10.1093/labmed/lmx090
pmid
29528429
Publisher site
See Article on Publisher Site

Abstract

Abstract Objective To compare the analytical performances of the enzymatic method (EM) and capillary electrophoresis (CE) for hemoglobin A1c (HbA1c) measurement. Methods Imprecision, carryover, stability, linearity, method comparison, and interferences were evaluated for HbA1c via EM (Abbott Laboratories, Inc) and CE (Sebia). Results Both methods have shown overall within-laboratory imprecision of less than 3% for International Federation of Clinical Chemistry and Laboratory Medicine (IFCC) units (<2% National Glycohemoglobin Standardization Program [NGSP] units). Carryover effects were within acceptable criteria. The linearity of both methods has proven to be excellent (R2 = 0.999). Significant proportional and constant difference were found for EM, compared with CE, but were not clinically relevant (<5 mmol/mol; NGSP <0.5%). At the clinically relevant HbA1c concentration, stability observed with both methods was acceptable (bias, <3%). Triglyceride levels of 8.11 mmol per L or greater showed to interfere with EM and fetal hemoglobin (HbF) of 10.6% or greater with CE. Conclusion The enzymatic method proved to be comparable to the CE method in analytical performances; however, certain interferences can influence the measurements of each method. glycated hemoglobin, electrophoresis, enzymatic methods, imprecison, lipemia, fetal hemoglobin Glycated hemoglobin is composed of hemoglobin (Hb)A1a, HbA1b, and HbA1c. The major fraction of glycated hemoglobin, HbA1c, is formed by nonenzymatic process of binding glucose to the N-terminal valine of the β-chain of hemoglobin.1 The American Diabetes Association (ADA) and World Health Organization (WHO) have confirmed the use of HbA1c as a biomarker for diagnosing and monitoring of diabetes mellitus.2 Two major clinical studies, the Diabetes Control and Complication Trial (DCCT) and the United Kingdom Prospective Diabetes Study (UKPDS), have demonstrated that intensive therapy and monitoring of diabetes mellitus type 1 (T1DM) and type 2 diabetes mellitus (T2DM) can decrease complications of diabetes.3,4 In July 2002, the International Federation of Clinical Chemistry and Laboratory Medicine (IFCC) approved reference methods for HbA1c. Two combinations of methods were used for separation and quantification of HbA1c, namely, high-performance liquid chromatography (HPLC) and electrospray ionization mass spectrometry and 2-dimensional approach using HPLC and capillary electrophoresis (CE) with ultraviolet (UV) detection.5 Studies have shown that other Hb variants (HbS, HbE, HbC, and HbD) do not interfere with HbA1c measurement by CE and enzymatic methods (EMs), but HbF might be a significant source of interference error.6,7 CE and EMs correlate well with the HPLC method,8,9 but there are not many studies comparing EMs with CE. The aim of this study was to perform verification of analytical specifications (precision, carryover, linearity, interferences, and stability) and method comparison for 2 different assays for HbA1c measurement, namely, direct whole blood enzymatic assay (Abbott Laboratories, Inc) and CE (Sebia). Materials and Methods Study Design Blood specimens used for this study were leftover patient specimens referred to the laboratory for routine HbA1c testing. All blood specimens were collected in ethylenediaminetetraacetic acid (EDTA) tubes (Vacuette tripotassium (K3 )EDTA Tubes, 4 mL; Greiner Bio-One International GmbH). The study was conducted from April 2015 through December 2016. First, we performed a comparison of 2 different assays for HbA1c measurement. Precision, carryover, stability, linearity, and interferences were tested during the specified period for both methods. The precision study was performed following the Clinical Laboratory Standards Institute (CLSI) EP15-A2 guideline.10 Methods CE: Minicap Flex Piercing Analyzer The Minicap Flex Piercing (Sebia) is a 2-silica-capillary automated analyzer that uses whole blood specimens from primary capped tubes for HbA1c testing. Using CE, hemoglobin variants are separated by high voltage (9400 V) in an alkaline buffer solution (pH, 9.4) and detected at 415 nm. Phoresis software (version 8.6.3) integrates HbA1c and HbA0 peaks, and the HbA1c concentration is calculated from the ratio HbA1c / (HbA1c + HbA0). EM: Abbott ARCHITECT c8000 Analyzer The enzymatic Hemoglobin A1c assay (Abbott Laboratories, Inc) measures HbA1c and total hemoglobin (THb) from the whole blood specimen. This method specifically measures N-terminal fructosyl dipeptides of the β-chain of HbA1c cleaved by a fructosyl peptide oxidase. The HbA1c concentration is calculated from the ratio of HbA1c and THb: HbA1c (mmol/mol) = (HbA1c/THb) × 1000.11 The results are expressed in IFCC units (mmol/mol) and DCCT/National Glycohemoglobin Standardization Program (NGSP) units (%) after conversion using the IFCC-NGSP master equation HbA1c (%) = (0.09148 × IFCC) + 2.152.8,12,13 Assay specifications are featured in Table S1 (available online-only). Verification of Analytical Specifications of the Methods and Comparison Studies Precision For this purpose, patient blood specimens have been used for both methods and commercial quality controls for EM only. All specimens used for precision testing were analysed in triplicate, during a 5-day period. Precision study of Minicap Flex Piercing analyzer. Three patient specimens with low (L; 31 mmol/mol [5.0%]), medium (M; 43 mmol/mol [6.1%]), and high (H; 98 mmol/mol [11.1%]) concentration of HbA1c were chosen for precision studies. Each specimen was divided into 2 parts to test both capillaries simultaneously. Precision study of EM. For precision study of EM, 1 patient specimen (S1 = 53 mmol/mol [7.0%]) and 2 commercial control specimens (Lyphocheck Diabetes Level 1 = 28 mmol/mol [4.7%] and Level 2 = 79 mmol/mol [9.4%]; Bio-Rad Laboratories, Inc) were chosen. Acceptance criteria for intralaboratory percentage coefficient of variation (%CV) was set at less than 3% (for results expressed in IFCC units) and less than 2% (for results expressed in DCCT/NGSP units).14 Carryover Carryover study of Minicap Flex Piercing Analyzer. We tested carryover for the Minicap Flex Piercing analyzer during a 3-day period, using 2 specimens, 1 with normal and the other with high HbA1c concentration. Specimens were divided into 2 tubes to test the carryover for both capillaries at the same time. Specimens with normal HbA1c concentration were analyzed before and after testing of the specimen with high HbA1c concentration. Specimens were analyzed once per day during the 3-day period. Carryover study for EM. We tested carryover for EM using 2 protocols, specifically evaluating analyte carryover between 2 whole blood specimens and whole blood and serum specimens. In the first protocol, 1 whole blood specimen with normal HbA1c concentration was divided into 2 aliquots and analyzed before and after the specimen with the high HbA1c concentration. The second protocol was designed because serum and whole blood specimens are tested on the same analyzers (Abbott Architect c8000) during routine laboratory work. In this protocol, 1 serum specimen was divided into 2 aliquots and analyzed before and after the whole blood specimen. To evaluate potential contamination with erythrocytes from the whole blood, in serum specimen potassium, magnesium, calcium, phosphorus, lactate dehydrogenase, aspartate aminotransferase (AST), total proteins, and glucose were measured on the Architect c8000 (with the original manufacturers reagents). Measurements were performed in triplicate, and average values were used for calculation. For all methods, carryover bias was calculated as follows: Bias=([parameter valueafter−parameter valuebefore]/parameter valuebefore)×100% Acceptance criterion for HbA1c carryover was set at 3% and for all other analytes, desirable specification for bias was taken from Ricos et al15; these data are listed in Table S2 (available online-only). Stability Stability was evaluated by measuring HbA1c concentration in the same specimen during 12 consecutive days for EM and 14 days for CE. Specimens were stored in the refrigerator (2°–8°C) between measurements. For each day, concentration of HbA1c was compared with the concentration obtained on the first day of measurement by calculating bias (%). Three specimens with different HbA1c concentrations were used for both methods. A criterion for stability bias was set at 3% or less.15 Method Comparison Comparison of these 2 methods was conducted during an 8-day period using 93 patient specimens. These specimens were analyzed by EM, stored in the refrigerator (2°–8°C) for a maximum of 3 days, and then analyzed by CE. Linearity The linearity was evaluated by measuring HbA1c in 5 specimens obtained with serial dilutions of the whole blood specimen with high HbA1c concentration (156 mmol/mol [16.4%] for CE and 101 mmol/mol [11.4%] for EM) with whole blood specimens with low HbA1c concentration (32 mmol/mol [5.1%] for CE and 31 mmol/mol [5.0%] for EM). Each specimen has been assayed in duplicate and its mean value plotted against the expected value for regression analysis. Interferences Both methods were tested for the interference of lipemia using Intralipid 20% emulsion for infusion (Fresenius Kabi AG). Intralipid was added in increased concentration into the whole blood specimen, and the bias of measured HbA1c was calculated for each one.16,17 We evaluated the interference of HbF according to a protocol in which whole blood specimens were spiked with an increasing volume of pooled fresh whole blood from a neonate (left over from routine complete blood count [CBC] testing), obtaining specimens with different HbF fractions. HbA1c was measured and bias calculated for each specimen by comparison with the same specimen without HbF added. The level of HbF was estimated by using the percentage of HbF, as determined by CE of hemoglobin (CAPILLARYS 2, Sebia). For lipemia and HbF interferences measurement, HbA1c was determined in duplicate by EM; for CE, we performed a single measurement. Interference of labile HbA1c was evaluated by incubating washed erythrocytes from pooled blood in 3 levels (L = 29 mmol/mol [4.8%]; M = 45 mmol/mol [6.3%]; and H = 110 mmol/mol [12.2%], of HbA1c measured by CE) with various glucose solutions (15 mmol/L, 30 mmol/L, and 60 mmol/L). Washed erythrocytes were incubated at 37°C for 3 hours and mixed every 30 minutes.18,19 We examined the interference of carbamylated hemoglobin (cHb) by in vitro carbamylation of hemoglobin. Washed erythrocytes of pooled blood in 3 levels (L = 27 mmol/mol [4.6%]; M = 45 mmol/mol [6.3%]; H = 96 mmol/mol [10.9%], of HbA1c measured by CE) were incubated for 3 hours at 37°C with 2.5 and 5.0 mmol per L of KOCN.19,20 HbA1c in all 3 pools for labile HbA1c and cHb interference was measured via EM and CE, before and after incubation, and interferents and biases were calculated. All measurements were made in duplicate. For all interferences studies, a criterion for acceptable bias was set at 6% or less.21 Statistical Analysis For comparison of 2 methods, results were analyzed by the Bland-Altman plot.22 The difference between methods is presented on the y axis for evaluation of constant difference; the average of both methods is presented on the x axis. Statistical analysis was performed using MedCalc software, version 11.5.1.0 (MedCalc Software). Results Precision Study Obtained repeatability and intralaboratory precision data are summarized in Table S3 (available online-only). All coefficients of variation were within acceptance criteria for both methods (<3% for IFCC units, <2% for NGSP units). Carryover Study Carryover testing between high and low HbA1c concentrations in whole blood specimens obtained acceptable mean biases for EM (0.27%) and CE (−1.0%). Also, in testing of carryover of whole blood to serum, all bias values were within acceptable criteria (Table S2, available online-only). Stability Study For EM, biases were within the acceptance criteria for all specimens (Figure S1, available online-only). In the case of CE, biases were acceptable for the specimens with medium and high HbA1c concentration but not for specimen with low HbA1c concentration (32 mmol/mol [5.1%]), with obtained bias as high as 9.38% on the ninth day (Figure S2, available online-only). Figure 1 View largeDownload slide Bland-Altman plot for constant difference between hemoglobin (Hb)A1c measurement (National Glycohemoglobin Standardization Program [NGSP] units) via the Abbott ARCHITECT enzymatic method and capillary electrophoresis method (Minicap Flex Piercing analyzer); n = 93. Figure 1 View largeDownload slide Bland-Altman plot for constant difference between hemoglobin (Hb)A1c measurement (National Glycohemoglobin Standardization Program [NGSP] units) via the Abbott ARCHITECT enzymatic method and capillary electrophoresis method (Minicap Flex Piercing analyzer); n = 93. Method Comparison Study Bland-Altman analysis for comparison of EM and CE revealed significant constant (mean, 1.1 mmol/mol [0.11%]) (Figure 1 and Figure S3 [available online-only]) and proportional (mean, 2.3% [for NGSP units, 1.6%]) differences (data not shown). Limits of agreement (±1.96 SD of differences) ranged from −1.9 to 4.0 mmol/mol (−0.2% to 0.4%) for constant difference (Figure 1 and Figure S3 [available online-only]) and −4.0 to 8.6% (NGSP, −2.6% to 5.8%) for proportional difference (data not shown), respectively. EM underestimates the HbA1c value, compared with CE. Linearity of the Methods The linearity of both methods has proven to be excellent within the examined ranges of HbA1c. Regression equation for the CE method (range, 32–156 mmol/mol, 5.1%–16.4%) was y = −0.15 (95% confidence interval [CI], −6.13 to 5.83) + 1.00 (0.94 to 1.06) x. For the EM (range, 31–101 mmol/mol; 5.0%–11.4%), y = 0.64 (95% CI, −2.08 to 3.36) + 1.00 (0.96 to 1.04) x, with R2 values of 0.999 for both methods. Interferences The results of a study of lipemia interference showed that triglyceride levels as high as 42.37 mmol per L have no impact on the HbA1c determination by CE (bias, <3.0%). For EM, the acceptable bias for triglyceride levels as high as 8.11 mmol per L has been obtained; however, increasing triglyceride concentration afterward causes a steep decrease of HbA1c (Figure 2). Figure 2 View largeDownload slide Interference of lipemia on hemoglobin (Hb)A1c measurement via the enzymatic method and capillary electrophoresis method. Interference is presented with mean bias (%) for EM and bias (%) from single measurement for CE. Biases for both IFCC and NGSP units are presented. Figure 2 View largeDownload slide Interference of lipemia on hemoglobin (Hb)A1c measurement via the enzymatic method and capillary electrophoresis method. Interference is presented with mean bias (%) for EM and bias (%) from single measurement for CE. Biases for both IFCC and NGSP units are presented. The presence of HbF already at the level of 10.6% influences the HbA1c measurement by CE, resulting in underestimation of the HbA1c concentration (bias, 7.7%). At the same time, no significant impact on HbA1c measurement by EM has been detected up to HbF of 39.5% (Figure 3). Figure 3 View largeDownload slide Interference of HbF (%) on hemoglobin (Hb)A1c measurement via the enzymatic method and capillary electrophoresis method. Interference is presented with mean bias (%) for EM and bias (%) from single measurement for CE. Biases for both IFCC and NGSP units are presented. Figure 3 View largeDownload slide Interference of HbF (%) on hemoglobin (Hb)A1c measurement via the enzymatic method and capillary electrophoresis method. Interference is presented with mean bias (%) for EM and bias (%) from single measurement for CE. Biases for both IFCC and NGSP units are presented. No significant impact of labile HbA1c was found for either method, with biases as high as −4.4% for EM and as high as −2.3% for CE. For cHb interference, no significant bias was detected for either method, with 2.5 mM potassium cyanate. However, with 5 mM potassium cyanate, both methods showed significant interference but with opposite trends regarding HbA1c concentration. For EM, bias was acceptable (−5.6%) only at low HbA1c levels and increased with HbA1c concentration (−7.1% and −8.1%, respectively). With CE, bias was acceptable (3.1%) only at the high HbA1c level and increased as high as 20.0% at normal and unmeasurable points at the low HbA1c concentration. Discussion The results of this study confirmed differences between the Abbott EM assay and the Sebia Minicap CE for HbA1c concentration measurement. All tested analytical performances were in accordance with the previously defined criteria, with the exception of the lipemia interference on EM and HbF interference on the capillary method of HbA1c measurement. Herein, we compared the analytical performances of 2 different methods for HbA1c measurement. Due to the recommended role of HbA1c as a biomarker for the diagnosis of diabetes (criterion, ≥48 mmol/mol [6.5%]) and management of diabetes complications (therapeutic target, <53 mmol/mol [7%]), as well as an indicator for change of therapy (criterion, ≥64 mmol/mol [8%]), assays with high analytical precision are necessary to differentiate between the upper limit of the nondiabetic range (42 mmol/mol [6.0%) and aforementioned targets.2 Therefore, highly stringent precision requirements for HbA1c assays are necessary. Based on intra- and interindividual biologic variation, intralaboratory CV of less than 3% (<2% for NGSP units) is recommended.15,23,24 This recommendation was supported by the mathematical minimal ability to differentiate the upper normal concentration of 42 mmol/mol (6.0%) and diagnostic criterion of 48 mmol/mol (6.5%).23 In our study, both methods achieved the criterion for intralaboratory imprecision, with an even more stringent goal of less than 2% (recommended for NGSP units) achieved for medium and high HbA1c concentrations expressed in IFCC units. This finding is important because it encompasses diagnostic and therapeutic targets. In comparison with the findings of the first published evaluation of the Sebia machine as a new analyzer for HbA1c assay by CE, the results of our study showed lower coefficients of variation. The reason for this finding may be the difference between the analyzers used. Whereas the CAPILLARYS 2 Flex piercing analyzer, which was used in the initial study, uses 8 capillaries, the Minicap-Flex piercing analyzer uses only 2 capillaries, thus lowering the impact of each capillary on overall imprecision.8 Performing analysis of serum/plasma and whole blood specimens on the same analyzers might represent a risk due to potential contamination and/or carryover. Although analyzers include additional washing cycles in their protocols, one might be concerned that serum specimens analyzed after full-blood specimens might be falsely hemolyzed, due to the carryover of erythrocytes. Therefore, one of the strengths of our study is the comprehensive investigation of carryover effect on biochemistry analyzer, including analytes that are most vulnerable to the hemolysis effect. Both methods proved to be reliable, considering carryover effect between the whole blood specimens. Also, carryover effect obtained between whole blood and serum specimens was far below the desirable specifications for bias used as acceptance criteria.15 Stability studies for CE and EM at the clinically relevant concentration of HbA1c, showed overall bias lower than 2% during the entire tested period (12 days and 14 days, respectively). EM showed better overall stability; however, we were surprised to find high deviation (9.8%) for CE at a low HbA1c concentration (32 mmol/mol [5.1%]). However, even such high deviation has no impact on clinical interpretation. A combination of HPLC and CE has been approved by the IFCC as the reference method for HbA1c.5 Recent studies8,18,25 have shown that results obtained by the HPLC method and the newly developed method that uses CE for the separation and quantification of HbA1c correlate well. As a result, in our study, we used CE as a reference method against which we compared HbA1c results measured via EM. According to Bland-Altman analysis of difference between methods, significant proportional and constant differences were found for EM compared with CE. However, a difference of 1.1 mmol/mol (0.11%) is not clinically relevant with total allowable error (TAE) buffer set at 5 mmol/mol (0.5%).26 Our results showed minimal interference from lipemia for the Abbott EM at triglyceride concentration of 8.11 mmol per L or greater. However, in our laboratory, this finding impacts only 0.3% to 0.4% of specimens. Data obtained using synthetic lipid emulsion might differ in comparison to the pathophysiologically induced lipemia experienced by patients with diabetes. Nowadays, most analytical platforms use automatic detection and assessment of the degree of lipemia as an L-index, enabling the laboratory professional to avoid missing the potential interferent. Increase of circulating HbF, defined as HbF greater than >2%, occurs relatively commonly and is associated with pathologic conditions such as leukemia, thalassemia, and hereditary persistence of Hbf. Individuals with the most common form of hereditary persistence of Hbf can have HbF levels as high as 30%.27 Because these patients generally experience no symptoms, it is crucial to know whether your method of HbA1c measurement is influenced by the presence of HbF and if so, at what proportion of HbF. The results of our study have shown that HbA1c measurement by CE is influenced by the presence of HbF at a lower proportion (10.6%) than that declared by the manufacturer (15%). Our data are in concordance with those reported in a recent study in which significant interference was observed for HbA1c measurement via the CAPILLARYS 2 FP analyzer for specimens with 10% to 15% HbF (relative bias, >±7%).18 Hemoglobin F has a lower glycation rate than HbA because approximately 60% of glycation of HbA occurs at the amino terminus of the β chain, and HbF has γ in place of β chains.28 Because HbF migrates close to HbA0, separation of fractions is not efficient enough if one overestimates the HbA0 proportion, which artificially lowers the HbA1c level. However, the presence of an increased level of HbF is obvious from an interfering peak in the chromatogram and therefore, misinterpretation of HbA1c concentration can be safely avoided. The results of our study indicate an insignificant influence of HbF levels as high as 55% on HbA1c measurement with EM, which is much higher than the result of the recent study in which the bias for the same method is within 6% for HbF less than 13%.29 This discrepancy possibly could be attributed to the protocol applied for interference investigation. In this study, bias was obtained by comparing the HbA1c concentrations obtained using EM with the IFCC reference method for specimens with abnormal hemoglobins or β0-thalassemia with HbF concentrations between 1.4% and 90.1%. Our results showed that labile HbA1c has no significant impact on HbA1c measurement via EM and CE. Regarding cHb interference, significant underestimation of HbA1c concentration was observed in the EM, with overestimation in CE but only at 5 mM potassium cyanate. Discrepancies in the trend of interference can be attributed to differences in calculation of HbA1c between methods. However, because such a high concentration of interferent is not physiologic, this bias can be neglected.30,31 In our study, the interference of hemoglobin variants on HbA1c measurement was limited to HbF only because based on the only available data on the frequency of hemoglobinopathic manifestations in Croatia, other variants in hemoglobin are rare (0.8%).32 Therefore, we believe that the results of our study are applicable to our population. A limitation to this study might be that we did not investigate the influence of acetylated HbA1c. Also, other interferences listed in the manufacturer declarations for the CE method have not been investigated (bilirubin, urea, ascorbic acid, rheumatoid factor, glibenclamide). However, we believe that our study results still provide valuable data on potential interferences in HbA1c testing. In conclusion, the analytical performances of HbA1c, as tested via EM and CE, are comparable, and both methods satisfy highly stringent analytical performance requirements; however, certain interferences can influence the measurements of each method. EM has the advantage of increased capacity and suitability for general chemistry analyzers. In contrast, CE has the important advantage of its ability to detect biological interference from hemoglobin variants that would affect the erythrocyte life span and lead to clinically misinterpreted HbA1c results. Abbreviations Hb hemoglobin ADA American Diabetes Association WHO World Health Organization DCCT Diabetes Control and Complication Trial UKPDS United Kingdom Prospective Diabetes Study T1DM diabetes mellitus type 1 T2DM type 2 diabetes mellitus IFCC International Federation of Clinical Chemistry and Laboratory Medicine HPLC high-performance liquid chromatography CE capillary electrophoresis UV ultraviolet EM enzymatic methods EDTA ethylenediaminetetraacetic acid K3 tripotassium CLSI Clinical Laboratory Standards Institute THb total hemoglobin NGSP National Glycohemoglobin Standardization Program %CV percentage coefficient of variation TAE total allowable error C1 first capillary tested C2 second capillary tested DSB desired specification for bias Acknowledgments This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors. References 1. Bunn HF , Haney DN , Kamin S , Gabbay KH , Gallop PM . The biosynthesis of human hemoglobin A1c. Slow glycosylation of hemoglobin in vivo . J Clin Invest . 1976 ; 57 ( 6 ): 1652 – 1659 . 2. American Diabetes Association . Diagnosis and classification of diabetes mellitus . Diabetes Care . 2014 ; 37 ( Suppl 1 ): S81 – S90 . 3. The Diabetes Control and Complications Trial Research Group . The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus . N Engl J Med . 1993 ; 329 : 977 – 986 . 4. UK Prospective Diabetes Study (UKPDS) Group . Intensive blood-glucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes (UKPDS 33) . Lancet . 1998 ; 352 ( 9131 ): 837 – 853 . 5. Jeppsson JO , Kobold U , Barr J , et al. ; International Federation of Clinical Chemistry and Laboratory Medicine (IFCC) . Approved IFCC reference method for the measurement of HbA1c in human blood . Clin Chem Lab Med . 2002 ; 40 ( 1 ): 78 – 89 . 6. National Glycohemoglobin Standardization Program (NGSP) . Factors that Interfere with HbA1c Test Results . NGSP website. http://www.ngsp.org/factors.asp. Accessed January 5, 2018 . 7. Teodoro-Morrison T , Janssen MJ , Mols J , et al. Evaluation of a next generation direct whole blood enzymatic assay for hemoglobin A1c on the ARCHITECT c8000 chemistry system . Clin Chem Lab Med . 2015 ; 53 ( 1 ): 125 – 132 . 8. Jaisson S , Leroy N , Meurice J , Guillard E , Gillery P . First evaluation of Capillarys 2 Flex Piercing® (Sebia) as a new analyzer for HbA1c assay by capillary electrophoresis . Clin Chem Lab Med . 2012 ; 50 ( 10 ): 1769 – 1775 . 9. Ajubi NE , Kross-Brion J , Newton CJ . Performance characteristics of the enzymatic Abbott Architect HbA1c whole blood assay . Clin Chem Lab Med . 2014 ; 52 ( 12 ): e283 – e285 . 10. Clinical and Laboratory Standards Institute . User Verification of Precision and Estimation of Bias: Approved Guideline . 2nd edition. CLSI document EP15-A2. Wayne, PA : Clinical and Laboratory Standards Institute , 2005 . 11. ARCHITECT System [package insert]. Abbott Park, IL: Abbott Laboratories, Inc; 2013. 12. Hoelzel W , Weykamp C , Jeppsson JO , et al. ; IFCC Working Group on HbA1c Standardization . IFCC reference system for measurement of hemoglobin A1c in human blood and the national standardization schemes in the United States, Japan, and Sweden: a method-comparison study . Clin Chem . 2004 ; 50 ( 1 ): 166 – 174 . 13. CAPILLARYS HbA1c . Using the Capillarys 2 Flex-Piercing Instrument . Sebia website. http://www.ilexmedical.com/files/Sebia%20inserts/CAPILLARYS_Hb_A1c.pdf. Accessed January 5, 2018 . 14. Braga F , Panteghini M . Standardization and analytical goals for glycated hemoglobin measurement . Clin Chem Lab Med . 2013 ; 51 ( 9 ): 1719 – 1726 . 15. Desirable Biological Variation Database specifications . Available at: http://www.westgard.com/biodatabase1.htm. Accessed December 6, 2015. 16. Nikolac N . Lipemia: causes, interference mechanisms, detection and management . Biochem Med (Zagreb) . 2014 ; 24 ( 1 ): 57 – 67 . 17. Clinical Laboratory Standards Institute . Hemolysis, Icterus, and Lipemia/Turbidity Indices as Indicators of Interference in Clinical Laboratory Analysis; Approved Guideline. CLSI C56-A document . Wayne, PA : Clinical Laboratory Standards Institute ; 2012 . 18. Wu X , Chao Y , Wan Z , et al. A comparative evaluation of the analytical performances of Capillarys 2 Flex Piercing, Tosoh HLC-723 G8, Premier Hb9210, and Roche Cobas c501 Tina-quant Gen 2 analyzers for HbA1c determination . Biochem Med (Zagreb) . 2016 ; 26 ( 3 ): 353 – 364 . 19. 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 ( 6 ): 867 – 877 . 20. Chachou A , Randoux C , Millart H , Chanard J , Gillery P . Influence of in vivo hemoglobin carbamylation on HbA1c measurements by various methods . Clin Chem Lab Med . 2000 ; 38 ( 4 ): 321 – 326 . 21. Little RR . Performance of hemoglobin A1c assay methods: good enough ? Clin Chem . 2014 ; 60 ( 8 ): 1031 – 1033 . 22. Giavarina D . Understanding Bland Altman analysis . Biochem Med (Zagreb) . 2015 ; 25 ( 2 ): 141 – 151 . 23. 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 ( 8 ): 1083 – 1097 . 24. Sacks DB , Arnold M , Bakris GL , et al. Executive summary: guidelines and recommendations for laboratory analysis in the diagnosis and management of diabetes mellitus . Clin Chem . 2011 ; 57 ( 6 ): 793 – 798 . 25. Woodworth A , Korpi-Steiner N , Miller JJ , et al. Utilization of assay performance characteristics to estimate hemoglobin A1c result reliability . Clin Chem . 2014 ; 60 ( 8 ): 1073 – 1079 . 26. Weykamp C , John G , Gillery P , et al. ; IFCC Task Force on Implementation of HbA1c Standardization . Investigation of 2 models to set and evaluate quality targets for HB A1c: biological variation and sigma-metrics . Clin Chem . 2015 ; 61 ( 5 ): 752 – 759 . 27. Little RR , Roberts WL . A review of variant hemoglobins interfering with hemoglobin A1c measurement . J Diabetes Sci Technol . 2009 ; 3 ( 3 ): 446 – 451 . 28. Rohlfing CL , Connolly SM , England JD , et al. The effect of elevated fetal hemoglobin on hemoglobin A1c results: five common hemoglobin A1c methods compared with the IFCC reference method . Am J Clin Pathol . 2008 ; 129 ( 5 ): 811 – 814 . 29. Nitta T , Yamashiro Y , Hattori Y , Ezumi T , Nishioka M , Nakamura J . The interference by HbF on HbA1c (BM Test HbA1c) measurement in enzymatic method . Ann Clin Biochem . 2015 ; 52 ( Pt 5 ): 569 – 575 . 30. Herpol M , Lanckmans K , Van Neyghem S , et al. Evaluation of the Sebia Capillarys 3 Tera and the Bio-Rad D-100 Systems for the Measurement of Hemoglobin A1c . Am J Clin Pathol . 2016 ; 146 ( 1 ): 67 – 77 . 31. Jaisson S , Desmons A , Renard B , Chevelle B , Leroy N , Gillery P . Analytical performances of a new enzymatic assay for hemoglobin A1c . Clin Chim Acta . 2014 ; 434 : 48 – 52 . 32. Efremov GD . Hemoglobinopathies in Yugoslavia: an update . Hemoglobin . 1992 ; 16 ( 6 ): 531 – 544 . © American Society for Clinical Pathology 2018. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)

Journal

Laboratory Medicine – Oxford University Press

Published: Mar 8, 2018

References

  • The biosynthesis of human hemoglobin A1c. Slow glycosylation of hemoglobin in vivo
    Bunn
  • Diagnosis and classification of diabetes mellitus
    American Diabetes Association
  • The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus
    The Diabetes Control and Complications Trial Research Group
  • 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
  • Approved IFCC reference method for the measurement of HbA1c in human blood
    Jeppsson