Diabetes mellitus: the long way of standardization of HbA(1c) to the level of highest metrological order.

Glycated haemoglobin (HbA(1c)) measurements are used in clinical studies and for the management of diabetic patients. Various efforts were made to standardize the HbA(1c) measurements with consensus standards and standards based on a reference measurement procedure with external calibration. According to ISO 17511 a standard should meet highest accuracy possible, have a defined uncertainty of measurement and the calibration should be traceable to SI units. For HbA(1c) this has been realized using a LC-ID-MS procedure based on the existing reference measurement procedure.

Diabetes mellitus: The long way of standardization of HbA1c to the highest metrological order

HbA1c is a retrospective analyte of the carbohydrate metabolism reflecting the mean blood glucose levels during the last 6 to 8 weeks. Clinical studies revealed a close relation between late diabetic complications and the concentration of glycated proteins in blood. Therefore HbA1c determination is widely used to assess the metabolic status and to monitor the medical treatment of diabetic patients. The Diabetes Control and Complications Trial (DCCT) and the United Kingdom Prospective Diabetes Study (UKPDS) used HbA1c as one of the main indicators for the quality of diabetes management [1], [2]. HbA1c is meanwhile recommended for screening of diabetes, since HbA1c concentrations in blood are not affected by acute metabolic alterations in the patients.

Standardization of HbA1c determinations affords the establishing and implementation of a reference measurement procedure with the highest accuracy possible. At the end of the 80s a certified reference material of HbA1c was not available and an high performance liquid chromatography (HPLC) peak of HbA1c analysis developed by Goldstein and co-workers [3] became the reference for the calibration of HbA1c measurement.

The HPLC procedure [3] was accepted as a common „voluntary consensus standard” for the HbA1c determination in the DCCT- and UKPDS-studies. Moreover, the procedure was applied as standard protocol for analytical devices for HbA1c measurements. The National Glycohemoglobin Standardization Program (NGSP) became the basis for the calibration of HbA1c measurements in the USA. According to this calibration HbA1c values are reported in % of blood haemoglobin concentration. Other national standardization activities were performed in Japan and Sweden. Further investigations showed that the separation of HbA1c from other proteins by HPLC was incomplete and the specificity was not appropriate for international standardization [4].

At the end of the 90s, the International Federation of Clinical Chemistry (IFCC) had established a Working Group to develop an international reference measurement procedure for HbA1c traceable to a standard of higher metrological order. Chromatographically purified HbA0 and HbA1c preparations from human blood were used as calibrators [5]. A reference measurement procedure was developed using these calibrators.

The principle of the IFCC reference measurement procedure was the proteolytic treatment of the blood sample with endoproteinase Glu-C cleaving glycated or non-glycated hexapeptides from the N-terminal β-chains of the haemoglobin. The hexapeptides were analyzed by HPLC-MS or by HPLC-capillary electrophoresis. The ratio of the glycated and non-glycated N-terminal hexapeptides of the haemoglobin β-chain expressed in percent was defined as the IFCC unit. The decision created considerable confusion, since from now on there were two different calibrations with the same unit on the market.

The results between the laboratories of the Working Group were regularly monitored by intercomparison surveys. The quality of measurements, expressed in terms of inaccuracy and imprecision, was improved during the measurement campaigns. From the collected data of the IFCC Working Group a “master equation” was developed to convert values from IFCC units into NGSP units and vice versa.

Since 2007, laboratories in Europe, Australia, and Japan calibrated their analytical systems for HbA1c according to the IFCC reference measurement procedure and started to report their results in mmol/mol to avoid confusions [6]; US laboratories remained with NGSP calibration and express HbA1c values in NGSP % .

The effort of the IFCC Working Group resulted in a remarkable improvement with respect to the NGSP consensus calibration. However, the key problem of the IFCC reference measurement procedure revealed to be the quality of HbA1c and HbA0 calibrator preparations and their traceability to the SI standard. During the inter-comparison studies a more robust LC-MS reference measurement procedure improving the reproducibility of measurement was established [7], [8].

In a recent study the traceability of HbA1c, measurement results to SI units has been achieved by isotope-dilution mass spectrometry, using glycated and non-glycated β-N-terminal hexapeptides of defined purity and stated uncertainty of measurement as calibrators and deuterated β-N-terminal hexapeptides as internal standards [9]. The traceability is accomplished by hydrolysis of the unlabelled hexapeptide standards and the determination of the amino acid concentration by LC-ID-MS calibrated with certified amino acid standards [10] (Figure 1 (Fig. 1)).

Figure 1

Traceability chain of HbA1c measurements to SI units. The calibration of the reference peptides is based on certified amino acid concentrations.

In a recent article Little et al. [11] claim the development of HbA1c standardization “from chaos to order” during the last 15 years with the NGSP and IFCC standardization efforts and discuss problems arising from the implementation of the IFCC Reference System. They state that “there are different interpretations of ‘traceability’”. According to ISO Standard 17511 the traceability of measurement is realized through an “unbroken chain of comparison” using a reference measurement procedure of highest metrological order and, if possible, a reference material of highest metrological level which is traceable to SI units [12]. With this ISO standard the “traceability of measurement” is obviously defined (Table 1 (Tab. 1)).

Table 1

Overview of HbA1c standardization efforts

The new reference measurement procedure is appropriate to set target values in external quality assessment schemes.

In Figure 2 (Fig. 2) the dispersion of routine values around the different target values is demonstrated in a recent external quality assessment scheme (EQAS) for HbA1c. The three possible target values (consensus value, IFCC reference measurement procedure value, and IDMS value) are given in this scheme. The new calibration using the LC-ID-MS procedure leads to slightly higher HbA1c levels than the recent calibration according to the IFCC reference measurement procedure.

Figure 2

Results of an external quality assessment scheme run by INSTAND e.V. for HbA1c

The Youden-plot demonstrates the measurement results of two samples (sample 1 and sample 2) from 325 routine laboratories (red cross), their overall mean (black cross, 40.4 and 82.5 mmol/mol, respectively), and the results obtained by the IFCC reference measurement procedure (IFCC RMP value, green cross, 36.6 and 78.4 mmol/mol, respectively), and by the IDMS procedure (IDMS value, blue cross, 37.9 and 81.2 mmol/mol, respectively).

In summary, a new standardization of HbA1c measurement has been established at the highest level of accuracy, with stated uncertainty of measurement and with traceability to SI units following the IFCC reference measurement protocol using LC-ID-MS as procedure of highest metrological order.

Notes

Competing interests

The authors declare that they have no competing interests.