Trueness verification and traceability assessment of results from commercial systems for measurement of six enzyme activities in serum

BACKGROUND

The in vitro diagnostics directive of the European Union requires traceability to higher order reference measurement procedures and materials for analytes in assuring the result trueness and comparability of laboratory measurements. Manufacturers must ensure that the systems they market are calibrated against available reference systems. Validation of metrologically traceable calibrations is, however, required.

METHODS

A commutable serum-based material was analyzed in three reference laboratories and target values for six enzymes (ALT, AST, CK, GGT, LD, amylase) were assigned using IFCC reference measurement procedures. 70 laboratories in Germany, Italy, and The Netherlands measured the same enzymes in the material using procedures from six commercial companies. A system for maximum allowable error was developed from the biological variation model and the results of the various procedures were tested on their compliance to trueness and between-laboratory and within-laboratory variations relative to the maximum allowable.

RESULTS

For ALT results were relatively good. >95% of laboratories using systems from Dade, Olympus, Ortho and Roche are expected to comply traceability within the biologically derived limits, and 94% respectively 89% from Abbott and Beckman. For AST and GGT only Dade respectively Olympus fully complied. For CK all companies showed significant bias. Nevertheless >95% of laboratories applying Abbott, Beckman and Roche systems will comply. Finally, LD and amylase measurements require significant improvement. Some manufacturers continue to sell on the European market assays giving results which are not traceable to the internationally accepted reference systems.

CONCLUSIONS

The traceability of enzyme measurements obtained with routine procedures to internationally accepted IFCC reference systems is not yet satisfactorily accomplished in clinical practice.

External Quality Assessment in The Netherlands: time to introduce commutable survey specimens. Lessons from the Dutch “Calibration 2000” project

The performance of suitable secondary reference material for the use of trueness control of six routinely measured clinical enzymes in the Dutch External Quality Assessment (EQA) scheme is described. The reference material of choice was selected using the split-patient-sample between-field method (twin study) design as described in an earlier study of the Calibration 2000 project in The Netherlands. This material, which was proven to be commutable for all wet chemistry systems, was implemented as the national enzyme calibrator. It consisted of a cryo-protected lyophilised serum with additions of recombinant human enzymes. Various batches of the frozen version of this material without cryo-protection additive, called native EQA samples, were used in the general EQA scheme for performance evaluation. The results of Calibration 2000 calibrated and non-Calibration 2000 calibrated laboratories were compared for both the regular (spiked with non-human enzymes) and native EQA samples in terms of precision and bias with established reference method values for the native samples. The regular samples showed mean between-laboratory CV ranges for all six enzymes involved (low–high) of 5.5–10.3% for the non-calibrated users vs. 4.6–10.8% for the calibrated users. For the native samples these respective ranges were 5.2–9.9% vs. 2.2–4.9%. Without exception, the group of Calibration 2000 calibrated users showed the lowest bias against the reference method values. Regular EQA samples (spiked with non-human enzymes) showed poorer performance than native samples and are not suitable for accuracy assessment purposes, the main aim of EQA schemes. Native samples that are commutable should be used for trueness control in current EQA schemes.

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

Background: The national programs for the harmonization of hemoglobin (Hb)A1c measurements in the US [National Glycohemoglobin Standardization Program (NGSP)], Japan [Japanese Diabetes Society (JDS)/Japanese Society of Clinical Chemistry (JSCC)], and Sweden are based on different designated comparison methods (DCMs). The future basis for international standardization will be the reference system developed by the IFCC Working Group on HbA1c Standardization. The aim of the present study was to determine the relationships between the IFCC Reference Method (RM) and the DCMs.

Methods: Four method-comparison studies were performed in 2001–2003. In each study five to eight pooled blood samples were measured by 11 reference laboratories of the IFCC Network of Reference Laboratories, 9 Secondary Reference Laboratories of the NGSP, 3 reference laboratories of the JDS/JSCC program, and a Swedish reference laboratory. Regression equations were determined for the relationship between the IFCC RM and each of the DCMs.

Results: Significant differences were observed between the HbA1c results of the IFCC RM and those of the DCMs. Significant differences were also demonstrated between the three DCMs. However, in all cases the relationship of the DCMs with the RM were linear. There were no statistically significant differences between the regression equations calculated for each of the four studies; therefore, the results could be combined. The relationship is described by the following regression equations: NGSP-HbA1c = 0.915(IFCC-HbA1c) + 2.15% (r2 = 0.998); JDS/JSCC-HbA1c = 0.927(IFCC-HbA1c) + 1.73% (r2 = 0.997); Swedish-HbA1c = 0.989(IFCC-HbA1c) + 0.88% (r2 = 0.996).

Conclusion: There is a firm and reproducible link between the IFCC RM and DCM HbA1c values.

Comparison of Procedures for Evaluating Laboratory Performance in External Quality Assessment Schemes for Lead in Blood and Aluminum in Serum Demonstrates the Need for Common Quality Specifications

Background: The different scoring methods used by eight European External Quality Assessment Schemes (EQASs) for occupational and environmental laboratory medicine were compared to develop suitable quality specifications as a step toward harmonization.

Methods: Real results for blood lead and serum aluminum assays, reported by participants in Italian and United Kingdom EQASs, were evaluated according to individual scheme scoring criteria. The same results were then used to produce z scores using scheme-based between-laboratory SDs as the estimate of variability to determine whether simple performance-derived quality specifications produced better agreement among schemes.

Results: The schemes gave conflicting assessments of participants’ performance, and participants judged to be successful by one scheme could be defined as performing inadequately by another. An approach proposed by Kenny et al. (Scand J Clin Lab Invest 1999;59:585), which uses clinical inputs to set targets for analytical imprecision, bias, and total error allowable, was then used to elaborate quality specifications.

Conclusions: We suggest that the CLIA ′88 recommendations for blood lead (± 40 μg/L or ± 10% of the target concentration, whichever is the greater) could be used as a quality specification, although a revision to ± 30 μg/L or ± 10% is recommended. For serum aluminum, a suitable quality specification of ± 5 μg/L or ± 20% of the target concentration, whichever is the greater, is suggested. These specifications may be used to compare laboratory performance across schemes.

Comparison of procedures for evaluating laboratory performance in external quality assessment schemes for lead in blood and aluminum in serum demonstrates the need for common quality specifications

BACKGROUND

The different scoring methods used by eight European External Quality Assessment Schemes (EQASs) for occupational and environmental laboratory medicine were compared to develop suitable quality specifications as a step toward harmonization.

METHODS

Real results for blood lead and serum aluminum assays, reported by participants in Italian and United Kingdom EQASs, were evaluated according to individual scheme scoring criteria. The same results were then used to produce z scores using scheme-based between-laboratory SDs as the estimate of variability to determine whether simple performance-derived quality specifications produced better agreement among schemes.

RESULTS

The schemes gave conflicting assessments of participants' performance, and participants judged to be successful by one scheme could be defined as performing inadequately by another. An approach proposed by Kenny et al. (Scand J Clin Lab Invest 1999;59:585), which uses clinical inputs to set targets for analytical imprecision, bias, and total error allowable, was then used to elaborate quality specifications.

CONCLUSIONS

We suggest that the CLIA '88 recommendations for blood lead (+/- 40 micro g/L or +/- 10% of the target concentration, whichever is the greater) could be used as a quality specification, although a revision to +/- 30 micro g/L or +/- 10% is recommended. For serum aluminum, a suitable quality specification of +/- 5 micro g/L or +/- 20% of the target concentration, whichever is the greater, is suggested. These specifications may be used to compare laboratory performance across schemes.

Commutability assessment of potential reference materials using a multicenter split-patient-sample between-field-methods (twin-study) design: study within the framework of the Dutch project "Calibration 2000"

BACKGROUND

The Dutch project "Calibration 2000" aims at harmonization of laboratory results via calibration by development of commutable, matrix-based, secondary reference materials. An alternative approach to the NCCLS EP14 protocol for studying commutability of reference materials is presented, the "twin-study design", which in essence is a multicenter, split-patient-sample, between-field-methods protocol.

METHODS

The study consisted of the simultaneous analysis of fresh patient sera and potential reference materials (PRMs) for HDL-cholesterol (HDL-C) by 86 laboratories forming 43 laboratory couples. Six subgroups of method combinations were formed. The patient sera were selected and interchanged by each laboratory couple. The PRMs consisted of three types: C37, prepared according to the NCCLS C37 protocol; Fro, frozen selectively pooled human serum; and Lyo, which was the same serum pool as Fro but lyophilized in the presence of sucrose. All PRMs were provided in three HDL-C concentrations. The regression line residuals for the PRMs were normalized by expressing them as multiples of the state-of-the-art within laboratory SD (SD(SA)). In addition, the extra contribution of each PRM to the total measurement uncertainty, CV(Netto), was calculated.

RESULTS

Averaged over the three PRM concentrations, 1.6% of the C37 residuals were outside the 3 SD(SA) limit. For the Fro and Lyo PRMs, these values were 2.4% and 11.1%. CV(Netto) values for C37, Fro, and Lyo were 2.9%, 4.3%, and 5.3%, respectively.

CONCLUSIONS

The present twin-study design, as a practical alternative to the NCCLS EP14 protocol, is a viable way of studying commutability characteristics of PRMs. The study suggests that the C37 PRMs are the best candidates for a future reference material.

Commutability Assessment of Potential Reference Materials Using a Multicenter Split-Patient-Sample Between-Field-Methods (Twin-Study) Design: Study within the Framework of the Dutch Project “Calibration 2000”

Background: The Dutch project “Calibration 2000” aims at harmonization of laboratory results via calibration by development of commutable, matrix-based, secondary reference materials. An alternative approach to the NCCLS EP14 protocol for studying commutability of reference materials is presented, the “twin-study design”, which in essence is a multicenter, split-patient-sample, between-field-methods protocol.

Methods: The study consisted of the simultaneous analysis of fresh patient sera and potential reference materials (PRMs) for HDL-cholesterol (HDL-C) by 86 laboratories forming 43 laboratory couples. Six subgroups of method combinations were formed. The patient sera were selected and interchanged by each laboratory couple. The PRMs consisted of three types: C37, prepared according to the NCCLS C37 protocol; Fro, frozen selectively pooled human serum; and Lyo, which was the same serum pool as Fro but lyophilized in the presence of sucrose. All PRMs were provided in three HDL-C concentrations. The regression line residuals for the PRMs were normalized by expressing them as multiples of the state-of-the-art within laboratory SD (SDSA). In addition, the extra contribution of each PRM to the total measurement uncertainty, CVNetto, was calculated.

Results: Averaged over the three PRM concentrations, 1.6% of the C37 residuals were outside the 3 SDSA limit. For the Fro and Lyo PRMs, these values were 2.4% and 11.1%. CVNetto values for C37, Fro, and Lyo were 2.9%, 4.3%, and 5.3%, respectively.

Conclusions: The present twin-study design, as a practical alternative to the NCCLS EP14 protocol, is a viable way of studying commutability characteristics of PRMs. The study suggests that the C37 PRMs are the best candidates for a future reference material.

Selection, Preparation, and Characterization of Commutable Frozen Human Serum Pools as Potential Secondary Reference Materials for Lipid and Apolipoprotein Measurements: Study within the Framework of the Dutch Project “Calibration 2000”

Background: The Dutch project “Calibration 2000” aims at harmonization of laboratory results via calibration by development of matrix-based secondary reference materials. We considered the selection, preparation, and characterization of 34 potential reference materials (PRMs).

Methods: Sixteen PRMs were prepared either strictly according to the NCCLS C37-A protocol or in a less stringent and more convenient way. In addition, 18 commercial, so-called human serum-based calibrators or controls were purchased and tested. Lipoprotein integrity was evaluated by examining the physicochemical characteristics of the materials. Commutability of the PRMs was assessed in 86 Dutch clinical laboratories, using a multicenter split-patient-sample between-field-method (twin-study) design. Normalized residuals of the PRMs with respect to the patient regression lines were calculated; in addition, the extra contribution of each PRM to the total measurement uncertainty (CVNetto) was calculated. On the basis of these results, the most native PRM was selected to investigate its potential to reduce interlaboratory variation and to improve lipid and apolipoprotein standardization.

Results: In general, only the NCCLS C37-A-type materials displayed normalized residuals below the decision limit for commutability and had small CVNetto values ranging between 0 and 3.8%. This contrasts with the findings in regularly pooled frozen sera and lyophilized cryoprotected PRMs. In two subsequent external quality assessment surveys, the NCCLS type C37-A materials contributed to reducing the intermethod lipid and (apo)lipoprotein variation to ∼2–4%.

Conclusions: NCCLS C37-A materials have a strong potential as secondary reference materials, not only for cholesterol but also for HDL-cholesterol, LDL-cholesterol, triglyceride, and apolipoprotein measurements.

Selection, preparation, and characterization of commutable frozen human serum pools as potential secondary reference materials for lipid and apolipoprotein measurements: study within the framework of the Dutch project "Calibration 2000"

BACKGROUND

The Dutch project "Calibration 2000" aims at harmonization of laboratory results via calibration by development of matrix-based secondary reference materials. We considered the selection, preparation, and characterization of 34 potential reference materials (PRMs).

METHODS

Sixteen PRMs were prepared either strictly according to the NCCLS C37-A protocol or in a less stringent and more convenient way. In addition, 18 commercial, so-called human serum-based calibrators or controls were purchased and tested. Lipoprotein integrity was evaluated by examining the physicochemical characteristics of the materials. Commutability of the PRMs was assessed in 86 Dutch clinical laboratories, using a multicenter split-patient-sample between-field-method (twin-study) design. Normalized residuals of the PRMs with respect to the patient regression lines were calculated; in addition, the extra contribution of each PRM to the total measurement uncertainty (CV(Netto)) was calculated. On the basis of these results, the most native PRM was selected to investigate its potential to reduce interlaboratory variation and to improve lipid and apolipoprotein standardization.

RESULTS

In general, only the NCCLS C37-A-type materials displayed normalized residuals below the decision limit for commutability and had small CV(Netto) values ranging between 0 and 3.8%. This contrasts with the findings in regularly pooled frozen sera and lyophilized cryoprotected PRMs. In two subsequent external quality assessment surveys, the NCCLS type C37-A materials contributed to reducing the intermethod lipid and (apo)lipoprotein variation to approximately 2-4%.

CONCLUSIONS

NCCLS C37-A materials have a strong potential as secondary reference materials, not only for cholesterol but also for HDL-cholesterol, LDL-cholesterol, triglyceride, and apolipoprotein measurements.

Neurotransmitter metabolites in CSF: an external quality control scheme

We report an international external quality control scheme on neurotransmitter metabolites in cerebrospinal fluid (CSF). The neurotransmitter metabolites homovanillic acid (HVA), 5-hydroxyindoleacetic acid (5-HIAA) and 3-methoxy-4-hydroxyphenylglycol (MHPG) are analysed to diagnose inborn errors of neurotransmitter metabolism. HVA is the catabolite of dopamine; 5-HIAA is the catabolite of serotonin; and MHPG is the catabolite of noradrenaline. In the first phase, 12 laboratories from six countries participated in this special quality control scheme to define the present state of the art and achieve harmonization in analytical outcome and interpretation. In the second part, recoveries, dilutions and methods for sample preparation were compared. The results of 3 of 12 laboratories were excluded because of unacceptable intralaboratory coefficients of variations (CV) for HVA and/or 5-HIAA. The inter- and intralaboratory CVs, the linearity and the recovery were acceptable for the other laboratories for both parameters. Unacceptable differences in the reference ranges between laboratories, leading to differences in interpretation of the results, became obvious. There was a significant improvement of the interlaboratory CV for HVA after standardization with a calibrator. The reproducibility of MHPG measurement appeared to be adequately established in only two laboratories and recovery was low in all five measuring this metabolite. The quality control scheme is an invaluable tool for controlling the analytical outcome and providing support to laboratories to improve their quality.

Approved IFCC reference method for the measurement of HbA1c in human blood

HbA1C is the stable glucose adduct to the N-terminal group of the beta-chain of HbA0. The measurement of HbA1c in human blood is most important for the long-term control of the glycaemic state in diabetic patients. Because there was no internationally agreed reference method the IFCC Working Group on HbA1c Standardization developed a reference method which is here described. In a first step haemoglobin is cleaved into peptides by the enzyme endoproteinase Glu-C, and in a second step the glycated and non-glycated N-terminal hexapeptides of the beta-chain obtained are separated and quantified by HPLC and electrospray ionisation mass spectrometry or in a two-dimensional approach using HPLC and capillary electrophoresis with UV-detection. Both principles give identical results. HbA1c is measured as ratio between the glycated and non-glycated hexapeptides. Calibrators consisting of mixtures of highly purified HbA1c and HbA0 are used. The analytical performance of the reference method has been evaluated by an international network of reference laboratories comprising laboratories from Europe, Japan and the USA. The intercomparison studies of the network showed excellent results with intra-laboratory CVs of 0.5 to 2% and inter-laboratory CVs of 1.4 to 2.3%. Possible interferences have been carefully investigated. Due to the higher specificity of the reference method the results are lower than those generated with most of the present commercial methods which currently are calibrated with unspecific designated comparison methods. The new reference method has been approved by the member societies of the International Federation of Clinical Chemistry and Laboratory Medicine and will be the basis for the future uniform standardization of HbA1c routine assays worldwide.

Preparation of a candidate primary reference material for the international standardisation of HbA1c determinations

We prepared a candidate primary reference material for the forthcoming international standardisation of beta-N-terminal glycated hemoglobin A measurements. It consists of well-defined mixtures of purified beta-N-terminal glycated hemoglobin A and non-glycated hemoglobin A. First, beta-N-terminal glycated hemoglobin A and non-glycated hemoglobin A were isolated, purified to homogeneity, and characterised. The techniques used were cation exchange and affinity chromatography for the purification, and high performance liquid chromatography, capillary isoelectric focusing, electrospray ionisation mass spectrometry, and peptide mapping for the characterisation. Hemoglobins from blood of healthy, non-diabetic volunteers were obtained with a purity of > 99.5% for non-glycated hemoglobin A and of > 98.5% for beta-N-terminal glycated hemoglobin A. However, results from peptide mapping indicate that the beta-N-terminal glycated hemoglobin A preparations still contain some non-beta-N-terminal glycated hemoglobins, co-eluting with beta-N-terminal glycated hemoglobin A. The exact content of beta-N-terminal glycated hemoglobin A in these preparations could be determined by a procedure consisting of standard addition, enzymatic cleavage and quantification of the resulting beta-N-terminal peptides to be in the range from 95-97.5%. Since the beta-N-terminal glycated hemoglobin A and non-glycated hemoglobin A content could be exactly determined in the materials prepared, mixtures of both components could be successfully used to calibrate the candidate reference methods.

Testing the accuracy of total cholesterol assays in an external quality-control program. Effect of adding sucrose to lyophilized control sera compared with use of fresh or frozen sera

We studied the suitability of various types of human serum preparations to test the accuracy of total cholesterol measurements in the External Quality Assessment scheme in The Netherlands, in which approximately 180 laboratories participate. Checked against the certified Abell/Kendall Reference Method, large reagent-dependent negative biases were observed with lyophilized serum that was insufficiently cryoprotected. The biases for the reagents of Du Pont, Roche, and Beckman averaged -16.7%, -9.2%, and -7.6% respectively; the least bias, -0.4%, was obtained with reagent from Boehringer Mannheim. The beneficial effect of cryoprotection with sucrose was demonstrated by the decrease in interreagent variation from 5.4% to 1.9%, the latter value being comparable with the values for fresh and once-frozen pooled serum (1.3% and 1.7%, respectively). We conclude that the detrimental effect of lyophilization on serum matrix can be minimized by suitable cryoprotection with 200 g/L sucrose.

Testing the accuracy of total cholesterol assays in an external quality-control program. Effect of adding sucrose to lyophilized control sera compared with use of fresh or frozen sera

We studied the suitability of various types of human serum preparations to test the accuracy of total cholesterol measurements in the External Quality Assessment scheme in The Netherlands, in which approximately 180 laboratories participate. Checked against the certified Abell/Kendall Reference Method, large reagent-dependent negative biases were observed with lyophilized serum that was insufficiently cryoprotected. The biases for the reagents of Du Pont, Roche, and Beckman averaged -16.7%, -9.2%, and -7.6% respectively; the least bias, -0.4%, was obtained with reagent from Boehringer Mannheim. The beneficial effect of cryoprotection with sucrose was demonstrated by the decrease in interreagent variation from 5.4% to 1.9%, the latter value being comparable with the values for fresh and once-frozen pooled serum (1.3% and 1.7%, respectively). We conclude that the detrimental effect of lyophilization on serum matrix can be minimized by suitable cryoprotection with 200 g/L sucrose.