Multicenter harmonization of common enzyme results by fresh patient-pool sera

¹ Department of Clinical Chemistry, Leyenburg Hospital, P.O. Box 40551, 2504 LN, The Hague, The Netherlands.

² Department of Clinical Chemistry, Rijnland Hospital, 2350 CC, Leiderdorp, The Netherlands.

³ Department of Clinical Chemistry, Leiden University Hospital, 2300 RC, Leiden, The Netherlands.

⁴ Department of Clinical Chemistry, Bronovo Hospital, 2597 AX, The Hague, The Netherlands.
^a Author for correspondence. Fax (31) 703592158.

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion References

 
A region consisting of 19 clinical laboratories harmonized their calibration of seven common enzymes by using fresh patient-pool sera. One of the laboratories was chosen to act as Regional Reference Laboratory (RRL). This laboratory used internationally accepted (mostly IFCC) methods at 37 °C, with an intralaboratory CV 2.5%. First, the reference ranges of the RRL were verified by analysis of a reference population and calculation of the results by a parametric method. Next, all laboratories, including the RRL, received six patient-pool sera and analyzed them at the same time on the same date. Enzyme calibration factors at each laboratory were converted on the basis of the slope, and occasionally the intercept, of regression analysis with the RRL and the individual laboratory. Before harmonization, the interlaboratory CVs varied from 16.9% to 61.6%. After harmonization, CVs decreased to between 5.0% and 9.5%. These results proved to be reproducible over a period of more than a year. Using internationally accepted inaccuracy and imprecision criteria, the achieved interlaboratory CVs permit the use of one set of reference ranges by all participating laboratories. Certified Reference Materials were analyzed, resulting in interlaboratory CVs as low as those achieved with patient-pool sera. These materials can act as commutable reference preparations, except for creatine kinase.

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

 
The catalytic activity of enzymes in the same patient's serum sample can differ substantially among clinical laboratories. Appropriate interpretation of the results is possible only by comparing the corresponding reference ranges. This inconsistency of results is method-dependent (1)(2)(3)(4)(5)(6)(7)(8)(9). Although most manufacturers produce reagents according to IFCC recommendations or methods closely resembling them, the results still can differ strongly. One of the major reasons for these discrepancies is that the enzyme calibration factor in an instrument often is based on correlation with historically diverse methods. Most clinical laboratories hesitate to change their history-based reference ranges for reasons of continuity in longitudinal laboratory data and clinical interpretation.

The solution to this lack of uniformity will come from the availability of appropriate, commutable reference preparations comparable in performance with fresh human serum (1)(2)(3)(10)(11)(12)(13). However, reference materials may differ in their properties because of so-called matrix effects, which are the result of the lyophilization process, the addition of preservatives or stabilizers, and supplementation with different isoenzymes from nonhuman sources (1)(2)(12)(13). For pragmatic reasons, we conclude that fresh human serum samples should be used for harmonization purposes.

As an official region of the Clinical Chemistry Society of The Netherlands (NVKC)¹ , we decided to harmonize the catalytic activities of seven common enzymes in 19 laboratories. The enzymes concerned were alkaline phosphatase (ALP; EC 3.1.3.1), alanine aminotransferase (ALAT; EC 2.6.1.2), aspartate aminotransferase (ASAT; EC 2.6.1.1), -glutamyltransferase (GT; EC 2.3.2.2), creatine kinase (CK; EC 2.7.3.2), lactate dehydrogenase (LD; EC 1.1.1.27), and -amylase (AMYL; EC 3.2.1.1).

After investigating all methods used in our region, we concluded that most laboratories used IFCC-recommended methods (14)(15)(16)(17)(18) or methods closely resembling them. Most laboratories, however, used 37 °C as the incubation temperature and converted their results to report at 30 °C. Although IFCC advocated an incubation temperature of 30 °C, we chose to harmonize at 37 °C because most modern analyzers are available with only this fixed, calibrated temperature. Moreover, most European countries endorse a reaction temperature of 37 °C. The regional reference methods of choice were the Société Française de Biologie Clinique (SFBC)/NVKC-recommended method for LD and the method with 4,6-ethylidene-p-nitrophenyl--maltoheptaoside as a substrate for AMYL, because in our region IFCC-recommended methods are not in use as everyday, routine methods (19)(20)(21)(22).

One of the laboratories using these methods was chosen to act as the Regional Reference Laboratory (RRL). The RRL established its reference ranges for these methods from a healthy reference population. Harmonization then took place through the analysis of six fresh patient-pool sera. All laboratories, including the RRL, measured these samples at the same time on the same date. After critical evaluation, the enzyme calibration factors of each analyzer were adjusted under the guidance of regression analysis. The harmonized results have been verified by analyzing new sets of six fresh patient-pool sera on three separate occasions thus far.

To investigate the characteristics of Certified Reference Materials (CRM) for enzyme determinations, prepared by the Community Bureau of Reference (BCR) of the European Community, we examined the activity of these lyophilized, nonhuman specimens and their suitability as commutable reference materials.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

 
general investigation of methods used
The analyzers in use were manufactured by Hitachi/Boehringer Mannheim (n = 6), Beckman (n = 4), Johnson & Johnson Clinical Diagnostics (J&JCD, n = 3), Bayer (n = 3), Vitatron (n = 2), and Roche (n = 1). The temperature setting was fixed at 37 °C for almost all analyzers (n = 15). However, before harmonization at 37 °C, only 4 of 19 laboratories reported at this temperature. The others recalculated their results to report at 30 °C (n = 13) or 25 °C (n = 2).

All reagents were produced by the manufacturers according to the recommendations of the IFCC or methods closely resembling those (14)(15)(16)(17)(18), except for LD and AMYL. LD was assayed by all participants according to the recommendations of the Clinical Chemistry Societies of France and The Netherlands (SFBC/NVKC) with pyruvate as a substrate (19)(20). None of the laboratories used the IFCC-recommended method with lactate as a substrate (21). AMYL was determined by a variety of methods. However, the one with p-nitrophenyl--maltoheptaoside (22) as a substrate with or without ethylidene or chloride as a (regulating) group was used in most cases (n = 13). Dry chemistry methods (J&JCD) are based on a refractory measuring principle, but are all calibrated by the manufacturer on the same reference methods as used by the RRL, except for AMYL (2).

regional reference laboratory
One of the participants (Rijnland Hospital, Leiderdorp) of this project was chosen to act as the RRL. The common routine methods of this laboratory were the regional reference methods. The analyzer of the RRL was a Hitachi 911 (Boehringer Mannheim). Reagents were also obtained from Boehringer Mannheim. ALP (cat. no. 816396), ALAT (cat. no. 851132), ASAT (cat. no. 851124), CK (cat. no. 763870), and GT (cat. no. 1489224) were analyzed according to the methods of the IFCC, or those closely resembling them (14)(15)(16)(17)(18), but carried out at 37 °C. The method according to SFBC/NVKC (cat. no. 1442643) was the regional reference method of choice for LD (19)(20). The method (cat. no. 1489399) using 4,6-ethylidene-p-nitrophenyl--maltoheptaoside as a substrate was the regional reference method for AMYL (22). The RRL used Bio-Rad Liquichek (Unassayed Chemistry Control, Human; Level 1, cat. no. 691, lot no. 75601; and Level 2, cat. no. 692, lot no.75602) for internal quality assessment.

regional reference ranges at 37 °C
The reference ranges of the RRL methods at 37 °C were verified before they were accepted as regional reference ranges. The reference population consisted of 330 apparently healthy individuals, one-half of whom were men. The population comprised six age groups (n = 55 ± 3 each) between 20 and 80 years old, with each group spanning one decade. Blood was collected from bloodbank donors and hospital staff, as well as from patients from several areas of our region who had a physically normal condition and who were visiting the outpatient departments of orthopedic surgery, neurosurgery (hernia nuclei pulposi), and ophthalmology for presurgery screening. Samples were analyzed within 48 h and were not hemolyzed.

The reference ranges were calculated by a parametric method, described by an expert panel on theory of reference values of the IFCC. The method is implemented in the REFVAL computer program (Ver. 3.2), which was applied in this study (23). The program appropriately transforms data to fulfill gaussian distribution after outlier correction.

harmonization of the individual laboratories by patient-pool sera
Patient sera were collected within 1 week and stored at 4 °C until use. These sera were mixed into six pool sera, each in appropriate amounts, to form a set of samples covering broad ranges of catalytic activity of all enzymes (0.5–10 times the upper limit of the reference range). Each laboratory (n = 19) received these samples to calculate conversion factors for conversion to RRL (37 °C) methods. All laboratories, including the RRL, analyzed the samples in duplicate at a fixed date and time. Results of the RRL (x) and the individual participants (y) were compared by regression analysis (y = ax b) according to Passing and Bablok (24)(25). Factors to convert enzyme calibration were calculated on the basis of slope (a) and intercept (b) of these regression analysis results.

verification and reproducibility of the harmonization program
After the conversion procedure, each laboratory received a new set of six patient-pool serum samples on three separate occasions at 3, 9, and 13 months after harmonization. These samples were again analyzed (singletons) at a fixed time and date, but now standardized at 37 °C. Interlaboratory variation before and after harmonization was calculated. This was defined as the percentage of variation (interlab CV) due to differences in the results (U/L) reported by all participating laboratories, and calculated as SD.

catalytic characteristics of crm after harmonization
To further investigate inaccuracy and imprecision of the harmonized methods in our region, we analyzed CRM. These materials for ALP (CRM 371), ALAT (CRM 426), GT (CRM 319), CK (CRM 299), LD (CRM 404), and AMYL (CRM 476) were obtained from the BCR of the European Community, Brussels, Belgium. No reference material for ASAT is prepared by BCR. The BCR-certified values are a result of IFCC-recommended methods with incubation at the official IFCC temperature of 30 °C. For LD, the certified value holds for the SFBC/NVKC-recommended method, which is the same as the RRL method, but again incubated at 30 °C. To investigate inaccuracy, it is possible to recalculate the BCR-certified results by temperature conversion factors from 30 °C to 37 °C, known from the literature (6)(7)(8). Recalculation of the BCR-certified 30 °C values by means of these temperature conversion factors has been done.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

 
regional reference ranges
Part of the harmonization procedure was the verification of the reference ranges of the RRL methods at 37 °C before they were accepted as regional reference ranges. We recognized a bimodal distribution of GT results in both men and women (data not shown). The parametric REFVAL software program with outlier correction was not able to correct for this. Probably this small but distinct second population is due to the consumption of alcohol by part of the reference population, as has been observed by others (26). Therefore, we decided to maintain the existing reference ranges of the RRL, as calculated in the past by the same IFCC parametric method. These results, as well as those of the other enzymes (Table 1 ), closely resemble the reference ranges (37 °C) used in The Netherlands (5).

harmonization of participating laboratories
Before laboratories were harmonized, the catalytic activities of fresh patient-pool sera showed a great variety among the participants of the harmonization program (Fig. 1 and Table 2 , column 1). Results varied from an interlaboratory CV of 16.9% for ALP to 61.6% for AMYL (Fig. 2 and Table 2 , column 1). The conversion factors were calculated by means of regression analyses on the basis of the duplicate results of six patient-pool sera. Comparison showed that almost all enzyme calibration factors had to be adjusted by slope (a) correction (y = ax b, x = RRL and y = individual participants; Fig. 1 ). In some cases, an additional intercept correction (b) had to be introduced. Dry chemistry methods (J&JCD) showed an intercept of 67 ± 9 U/L for LD for all three laboratories.

View larger version (34K): [in this window] [in a new window]   Figure 1. Typical examples of comparisons of ALAT and GT. Catalytic enzyme activity (U/L) of six patient-pool sera, measured by the RRL (x) and the individual participating laboratories (y) of the regional harmonization program before (left panels) and after (right panels) harmonization.

View larger version (22K): [in this window] [in a new window]   Figure 2. Reproducibility within the regional harmonization program. Mean catalytic activity of enzymes ( - ) of 19 regional clinical laboratories in comparison with the RRL results (zero line). Interlaboratory imprecision is represented by bars before () and at 3, 9, and 13 months after harmonization (). Bars indicate 4x interlaboratory CV of the mean.

All methods were linear for the patient-pool sera, correlated well (r 0.99) with the RRL methods, and exhibited a small variance of the slope of the regression line (CV <2.5%), except for LD dry chemistry (J&JCD). For all three laboratories, this method showed the same small patient-related discrepancies in the RRL method, with an average r = 0.98 and an average CV = 5.5% of the slope of the regression line. However, they did not cause miscalculations of the conversion factor. In general, one should be aware of patient-related "matrix effects" introduced by differences in isoenzyme specificity and of interferences of biological substances or medication within methods, even within those resembling the recommendations of the IFCC (1)(27).

reproducibility of the harmonization program
During the harmonization period, intralaboratory quality assessment of the RRL showed low CVs for all methods under investigation (Table 2 , column 6) for both normal and abnormal levels (Bio-Rad Liquichek Levels 1 and 2). They all met quality goals formulated by a working group of the European Group for the Evaluation of Reagents and Analytical Systems in Laboratory Medicine (EGE-Lab) as being "state of the art". This is defined as the intralaboratory CV of methods, achieved by the best 0.2 fractile of laboratories participating in international quality assessment schemes (Table 2 , column 5) (28)(29). This constant and low intralaboratory variance indicates a good calibration stability of the RRL methods, an indisputable condition for the reproducibility of the harmonization program.

After converting the participants' enzyme calibration factors to the RRL results, the reproducibility of the harmonization process was verified. On three different occasions, new sets of six patient-pool sera were analyzed. The results of these three independent observations are shown in Fig. 2 . After harmonization, interlaboratory CV dropped dramatically, varying from a minimum of 5.0% for ALAT to a maximum of 9.5% for AMYL (Figs. 1 and 2 ; Table 2 , column 2). These results were reproducible 3, 9, and 13 months after harmonization (Fig. 2 ).

wet vs dry chemistry after harmonization
After harmonization, no significant differences between the results of "dry chemistry" methods of J&JCD (n = 3) and "wet chemistry" methods (n = 16) could be observed when the average catalytic enzyme activity of 18 patient-pool sera were analyzed and the average interlaboratory CVs were calculated (Fig. 3 ).

View larger version (28K): [in this window] [in a new window]   Figure 3. Wet and dry chemistry methods analyzing 18 patient-pool sera and CRM reference materials after harmonization. Mean catalytic enzyme activity ( - ) of wet chemistry methods (; n = 16) and dry chemistry methods (; n = 3) in comparison with the RRL results (zero line). Bars indicate 4x interlaboratory CV of the mean.

catalytic characteristics of crm after harmonization
Interlaboratory CVs of the CRM preparations varied from 3.3% to 11.4% (Table 2 , column 7). These imprecision results are of the same magnitude as the interlaboratory CVs of fresh patient-pool sera, except for CK (Table 2 , columns 2 and 7; Fig. 3 ). A significant difference in CK activities is observed between the method used by J&JCD ("dry chemistry") and the other methods (all IFCC-recommended "wet chemistry"); the activities were 609 ± 61 U/L and 483 ± 12 U/L, respectively.

To investigate accuracy, we recalculated the 30 °C-certified BCR values for CRM preparations by temperature conversion factors from 30 °C to 37 °C, which are known from the literature (6)(7)(8). The 37 °C-converted BCR values differed by <10% from the 37 °C results found by the RRL, except for the CK dry chemistry method (data not shown).

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

 
In this study, we show that 19 independent clinical laboratories were able to harmonize their calibration of common enzymes. Indeed, it took some time and discussion before consensus was attained with respect to the harmonization procedure. The RRL should use methods according to IFCC recommendations, but these methods should also be implemented in the routine setting for the production of clinical results. This approach makes verification of clinical laboratory methods in a regional setting inexpensive, easy, and workable. However, no routine clinical chemistry analysis system meets the exact specifications of the IFCC-recommended methods. Moreover, modern analyzers are available with only the fixed calibrated temperature of 37 °C, instead of the IFCC-recommended 30 °C. We all agreed that the (mostly IFCC) methods in use by the RRL were the best methods to act as regional reference methods for this moment. They all have good calibration stability for the enzymes under investigation, with an intralaboratory CV 2.5%, meeting state-of-the-art specifications (Table 2 , columns 5 and 6).

The alternative can be compliance with the official IFCC-recommended methodologies on a specially dedicated instrument, fully meeting the specifications of the IFCC (9)(14)(15)(16)(17)(18)(21). Apart from instrument specifications, well-defined, high-quality reagents are an absolute necessity (1). However, for more than two decades, this approach to solving the "accuracy" problem of enzyme determinations by rigorously defined methods has not led to harmonization of results.

Essential for the long-term success of this harmonization program is starting a regional quality assessment program to periodically check the reproducibility of the harmonized results. Twice a year the participants analyze a new set of six patient-pool samples. Thus far, the results show a constant reproducible quality with small interlaboratory CVs (Fig. 2 and Table 2 , column 2).

However, do these results meet international quality specifications? If we take into account that individual patient results should be exchangeable between hospitals, one could argue that clinical imprecision specifications as proposed by Fraser et al. (29)(30) and Harris (31) should be met. The desirable analytical precision (CV_a) should be smaller than one-half of the within-subject biological variation (CV_I), which is formulated as CV_a 1/2 CV_I. These quality specifications for individual patient testing are supported by EGE-Lab (29). The regional interlaboratory CVs for ALAT, ASAT, GT, and CK indeed meet these criteria, as can be observed in Table 2 (columns 2 and 3).

Unfortunately, the interlaboratory CVs of ALP, LD, and AMYL are larger than one-half of the within-subject CV_I (Table 2 , columns 2 and 3). However, intralaboratory CVs of the participants are much smaller, comparable with those of the RRL and meeting the state-of-the-art criteria (Table 2 , columns 5 and 6). For this reason, we concluded that these larger interlaboratory CVs after harmonization are (still) due to inaccuracy.

Gowans et al. (32) based their analytical quality goals for laboratories, using the same set of reference ranges, on criteria based on inaccuracy. In addition to the within-subject (CV_I) component, they also incorporated the between-subject (CV_G) component of biological variation. Inaccuracy or bias (B_a) should be <1/4 {(CV_G) (CV_I)}1/2. Comparison of our results with these specifications, also supported by EGE-Lab, shows that, for AMYL, the interlaboratory CV (9.5%) is still too high (Table 2 , columns 2 and 4). However, when we use the less stringent criterion of one-sixteenth of the reference range (Table 1 ; B_a 10.3%), the results are acceptable (29). The results for LD meet the interim inaccuracy specification of EGE-Lab (Table 2 , column 4, in parentheses) but still are slightly too high compared with the proposed inaccuracy specifications (Table 2 , column 4). For all enzymes under investigation, the results meet the US CLIA Total Error criteria for proficiency testing, the acceptable value being CV_a <20% (33)(34).

The most convenient and best solution to the lack of uniformity and accuracy of enzyme determinations will remain the availability of appropriate, commutable reference materials comparable in performance with fresh human sera. Specimens used in clinical chemistry for quality control or calibration can differ markedly in their properties because of so-called matrix effects (1)(2)(3)(4)(5)(6)(7)(8)(9)(10)(11)(12)(13). CRM materials from BCR are often prepared from pork tissue or may have nonhuman serum matrices and are lyophilizates, all of which are causes for matrix effects. However, to our surprise, except for CK, no gross differences in catalytic activities were observed for these CRM preparations after harmonization. The interlaboratory CVs found for these preparations after harmonization were small and comparable with those found for patient-pool sera (Table 2 , columns 2 and 7). No definite conclusion concerning accuracy could be drawn, because the BCR-certified values are for 30 °C IFCC or SFBC/NVKC methods rather than for 37 °C. However, the differences between 37 °C-converted BCR values and those found by the RRL were <10%. In general, we think that the results obtained with the CRM preparations are commutable, with no significant matrix effects for the methods under investigation, except for CK. However, calibration on CRM preparations is expensive and time consuming because each enzyme has its own preparation at a price of ~$150/mL. We think they can be used as verifier or primary reference material, which is the principal reason for BCR to manufacture them. Hopefully, a value for IFCC methods incubated at 37 °C will be certified by BCR in the near future.

At the start of the harmonization program, our approach demanded one reference laboratory to be the RRL. Like others (35), we felt that a consensus value could be stable and accurate as a target value. Indeed, up to now no gross discrepancies between the RRL and the consensus results have been observed (Fig. 2 ). After harmonization, the RRL has been regarded as a regular participant of the regional enzyme quality program and the consensus value as a reference. However, this concept is susceptible to drift because traceability to the original harmonization results can be lost in the long run. To check longitudinal calibration stability, we will analyze the CRM preparations as a verifier on a regular basis, in addition to the six patient-pool samples. As is documented by BCR, the CRM preparations have great stability at -20 °C, with a deterioration rate between 0.01% and 0.04% a year, depending on the enzyme. The "CRM consensus value" will function as the reference point to guarantee accuracy and stability in time.

We showed that, in contrast to patient results, the CRM results of the J&JCD dry chemistry method for CK were not identical to IFCC-recommended wet chemistry results. The CRM wet chemistry consensus can function as a regional target value for CK. Laboratories using a dry chemistry method for CK must compare their patient results with those of wet chemistry methods to check their calibration stability.

Finally, we conclude that the described procedure has shown that the calibration of enzymes can be unified in a regional setting by making use of patient-pool sera. Even the analysis of AMYL, carried out with a great variety of methods, could be harmonized. The same holds true for dry chemistry methodology, which is particularly sensitive to matrix effects. The results confirm our opinion that the lack of good calibration standards rather than differences in methodology is the major reason for discrepancies in the measurements of enzymes.

	Footnotes

 
¹ Nonstandard abbreviations: NVKC, Clinical Chemistry Society of The Netherlands; ALP, alkaline phosphatase; ALAT, alanine aminotransferase; ASAT, aspartate aminotransferase; GT, -glutamyltransferase; CK, creatine kinase; LD, lactate dehydrogenase; AMYL, -amylase; RRL, Regional Reference Laboratory; CRM, Certified Reference Materials; SFBC, Société Française de Biologie Clinique; BCR, Community Bureau of Reference; J&JCD, Johnson & Johnson Clinical Diagnostics; and EGE-Lab, European Group for the Evaluation of Reagents and Analytical Systems in Laboratory Medicine.

	References

Top Abstract Introduction Materials and Methods Results Discussion References

 

1. Rej R. Accurate enzyme activity measurements: two decades of developments in commutability of enzyme quality control materials. Arch Pathol Lab Med 1993;117:352-364. [ISI][Medline] [Order article via Infotrieve]
2. Lasky FD. Achieving accuracy for routine clinical chemistry methods by using patient specimen correlations to assign calibrator values: a means of managing matrix effects. Arch Pathol Lab Med 1993;117:412-419. [ISI][Medline] [Order article via Infotrieve]
3. Jansen RTP, Jansen AP. Standard versus standardised methods in enzyme assay. Ann Clin Biochem 1983;20:52-59.
4. Miller WG, Crane PD, Cryer C. Interlaboratory standardization of enzyme results: the Richmond project. Clin Chem 1986;32:1525-1531. [Abstract/Free Full Text]
5. Baadenhuijsen H, van Benthem E. Status incubation conditions enzyme activities in The Netherlands. Ned Tijdschr Klin Chem 1993;18:260-265.
6. Heerspink W, Hafkenscheid JCM, Siepel H. Temperature conversion factors for enzymes: comparison of methods. Enzyme 1980;25:333-341. [ISI][Medline] [Order article via Infotrieve]
7. Hafkenscheid JCM, Kohler BEM. Effects of temperature on measurement of aspartate aminotransferase and alanine aminotransferase in commercial control sera. Clin Chem 1986;32:184-185. [Abstract/Free Full Text]
8. Hafkenscheid JCM, Kohler BEM. Temperature conversion factors for four enzymes in commercial control sera [Letter]. Clin Chem 1986;32:1616.[ISI][Medline] [Order article via Infotrieve]
9. . ECCLS. Standards for enzyme determination: creatine kinase, aspartate aminotransferase, alanine aminotransferase, gamma-glutamyltransferase. Mitt Dtsch Ges Klin Chem 1989;4:186-217.
10. Lessinger JM, Férard G, Grafmeyer D, Labbé D, Maire I, Schiele F, Vassault A. Usefulness of reference materials in calibration of enzyme activity. Eur J Clin Chem Clin Biochem 1995;33:859-864. [ISI][Medline] [Order article via Infotrieve]
11. Tanishima K, Gao SX, Yammamoto R, Yoshida H. Biochemical and enzymological study of lactate dehydrogenase isoenzymes from commercial quality control sera and several animal tissue sources. Eur J Clin Chem Clin Biochem 1995;33:865-868. [ISI][Medline] [Order article via Infotrieve]
12. Powers DM. Overcoming the limitations of artificial clinical specimens: linkage to performance of patients' specimens [Editorial]. Clin Chem 1995;41:1689-1692. [ISI][Medline] [Order article via Infotrieve]
13. Henderson AR, Webb S, Richardson H, Wood DE. A comparison of pooled, fresh frozen, and lyophilised sera as a matrix for enzyme proficiency testing: the experience of Ontario laboratory proficiency testing program. Clin Biochem 1996;29:183-185. [ISI][Medline] [Order article via Infotrieve]
14. Shaw LM, Strømme JH, London JL, Theodorsen L. IFCC methods for the measurement of catalytic concentration of enzymes. Part 4. IFCC method for -glutamyltransferase [(-glutamyl)-peptide: amino acid -glutamyltransferase, EC 2.3.2.2]. J Clin Chem Clin Biochem 1983;21:633-646. [ISI][Medline] [Order article via Infotrieve]
15. Tietz N, Rinker AD, Shaw LM. IFCC methods for the measurement of catalytic concentration of enzymes. Part 5. IFCC method for alkaline phosphatase (orthophosphoric-monoester phosphohydrolase, alkaline optimum, EC 3.1.3.1). J Clin Chem Clin Biochem 1983;21:731-748. [ISI][Medline] [Order article via Infotrieve]
16. Bergmeyer HU, Hørder M, Rej R. Approved recommendation (1985) on IFCC methods for the measurement of catalytic concentration of enzymes. Part 3. IFCC method for alanine aminotransferase (L-alanine:2-oxoglutarate aminotransferase, EC 2.6.1.2). J Clin Chem Clin Biochem 1986;24:481-495. [ISI][Medline] [Order article via Infotrieve]
17. Bergmeyer HU, Hørder M, Rej R. Approved recommendation (1985) on IFCC methods for the measurement of catalytic concentration of enzymes. Part 2. IFCC method for aspartate aminotransferase (L-aspartate:2-oxoglutarate aminotransferase, EC 2.6.1.1). J Clin Chem Clin Biochem 1986;24:497-510. [ISI][Medline] [Order article via Infotrieve]
18. Hørder M, Elser RC, Gerhardt W, Mathieu M, Sampson EJ. Approved recommendation on IFCC methods for the measurement of catalytic concentration of enzymes. Part 7. IFCC method for creatine kinase (ATP:creatine N-phosphotransferase, EC 2.7.3.2). Eur J Clin Chem Clin Biochem 1991;29:435-456. [ISI][Medline] [Order article via Infotrieve]
19. . Enzymology Commission: Société Française de Biologie Clinique. Recommendations for determining the catalytic concentration of lactate dehydrogenase in human serum at 30 °C (document F). Ann Biol Clin 1982;40:160-164.
20. van der Heiden C, Bootsma J, Cornelissen PJHC, Hafkenscheid JCM, Oosterom R, Smit EM. Adaptations of the recommendations (NVKC) for the measurements of catalytic activity concentrations of enzymes in serum and plasma. Ned Tijdschr Klin Chem 1987;12:231-236.
21. Bais R, Philcox M. Approved recommendation on IFCC methods for the measurement of catalytic concentration of enzymes. Part 8. IFCC method for lactate dehydrogenase (L-lactate:NAD⁺ oxidoreductase, EC 1.1.1.27). Eur J Clin Chem Clin Biochem 1994;32:639-655. [ISI][Medline] [Order article via Infotrieve]
22. Kruse-Jarres JD, Kaiser C, Hafkenscheid JCM, Hakenwallner W, Stein W, Bohner J, et al. Evaluation of new -amylase assay using 4,6-ethylidene-(G₇)-1–4-nitrophenyl-(G₁) -D-maltoheptaoside as substrate. J Clin Chem Clin Biochem 1989;27:103-113. [ISI][Medline] [Order article via Infotrieve]
23. Solberg HE. Approved recommendation (1987) on theory of reference values. Part 5. Statistical treatment of collected reference values. Determination of reference limits. Clin Chim Acta 1987;170:S13-S32.
24. Passing H, Bablok W. A new biometrical procedure for testing the equality of measurements from two different analytical methods. Part 1. J Clin Chem Clin Biochem 1983;21:709-720. [ISI][Medline] [Order article via Infotrieve]
25. Passing H, Bablok W. A new biometrical procedure for testing the equality of measurements from two different analytical methods. Part 2. J Clin Chem Clin Biochem 1984;22:431-445. [ISI][Medline] [Order article via Infotrieve]
26. Leino A, Impivaara O, Irjala K, Mäki J, Pelola O, Järvisalo J. Health-based reference intervals for ALAT, ASAT and GT in serum, measured according to the recommendations of European Committee for Clinical Laboratory Standards (ECCLS). Scand J Clin Lab Invest 1995;55:243-250. [ISI][Medline] [Order article via Infotrieve]
27. Kroll MH, Elin RJ. Interference with clinical laboratory analysis [Review]. Clin Chem 1994;40:1996-2005. [Abstract]
28. Steigstra H, Jansen RTP, Baadenhuijsen H. Combi scheme: new combined internal/external quality-assessment scheme in The Netherlands. Clin Chem 1991;37:1196-1204. [Abstract/Free Full Text]
29. Fraser CG, Hyltoft Petersen P, Ricos C, Haeckel R. Proposed quality specifications for imprecision and accuracy of analytical systems for clinical chemistry. Eur J Clin Chem Clin Biochem 1992;30:311-317. [ISI][Medline] [Order article via Infotrieve]
30. Fraser CG. Biological variation in clinical chemistry. An update: collated data 1988–1991. Arch Pathol Lab Med 1992;116:916–23..
31. Harris EK. Statistical principles underlying analytic goal-setting in clinical chemistry. Am J Clin Pathol 1979;72:374-382. [ISI][Medline] [Order article via Infotrieve]
32. Gowans EMS, Hyltoft Petersen P, Blaabjerg O, Hørder M. Analytical goals for the acceptance of common reference intervals for laboratories throughout a geographical area. Scand J Clin Lab Invest 1988;48:757-794. [ISI][Medline] [Order article via Infotrieve]
33. Westgard JO, Seehafer JJ, Barry P. European specifications for imprecision and inaccuracy compared with operating specifications that assure the quality required by US CLIA proficiency-testing criteria. Clin Chem 1994;40:228-232.
34. US Dept. of Health and Human Services. Medicare, Medicaid, and CLIA programs; regulations implementing the Clinical Laboratory Improvement Amendments of 1988 (CLIA). Final rule. Fed Regist 1992;57:7002–186..
35. Hartmann AE, Naito HK, Burnett RW, Welch MJ. Accuracy of participant results utilized as target values in the CAP chemistry survey program. Arch Pathol Lab Med 1985;109:894-903. [ISI][Medline] [Order article via Infotrieve]