Interlaboratory variation of biochemical markers of bone turnover.

BACKGROUND: Biochemical markers of bone metabolism are used to assess skeletal turnover, but the variability of marker assays is still an issue of practical concern. We describe the results of an international proficiency testing program for biochemical bone markers among clinical laboratories. METHODS: Two serum and two urine pools (normal and increased marker concentrations) were sent on dry ice to 79 laboratories for analysis within 2 weeks of receipt. RESULTS: Data were submitted by 73 laboratories. The within-method interlaboratory CVs (CV(IL)s) were as follows: serum bone-specific alkaline phosphatase (n = 47 laboratories), 16-48%; serum osteocalcin (n = 31), 16-42%; urinary free deoxypyridinoline (n = 30), 6.4-12%; urinary total deoxypyridinoline and pyridinoline (n = 29), 27-28%; urinary N-terminal cross-linked telopeptide of type I collagen (n = 10), 39%; serum C-terminal cross-linked telopeptide of type I collagen (ICTP; n = 8), 22-27%; urinary hydroxyproline (n = 13), 12%. Analytical results showed both systematic and nonsystematic deviations. In identical samples, results obtained for the same marker by the same method differed up to 7.3-fold. In urine-based assays, correction for urinary creatinine slightly increased CVs. CONCLUSION: Even with identical assays and methods, results for most biochemical markers of bone turnover differ markedly among laboratories.

External quality assessment of molecular biology-based methods used in laboratories of clinical chemistry and human genetics.

The Reference Institute of Bioanalysis of the German Society of Clinical Chemistry has performed the first external assessment of molecular genetics methods used in medical diagnosis. The following procedures were tested: (I) DNA preparation from whole blood, (II) PCR amplification using "standard" primers, and (III) submarine agarose gel electrophoresis. Out of 50 participants, 45 returned samples for evaluation.

Overview: quality control of tumor marker assays.

Data from interlaboratory surveys for the determination of CA 125 (1987 to 1998) show large differences depending on the commercial kit used. Samples of the same material produced different analytical results in different interlaboratory surveys, thus showing that some kits did not have a satisfactory rate of reproducibility over a longer period of time.

Concentration-dependent profiles for describing the scatter of results of interlaboratory surveys.

In an interlaboratory survey for the quantitative determination of a clinical chemical quantity, samples of the same specimen are analysed in different laboratories. If the number of participating laboratories is sufficiently large, then the differences between the 50th percentile (median) and e. g. the 25th and 75th percentiles of the results give a very reliable impression of the range of interlaboratory scatter for the particular analytical technique. Results from a relatively large number of interlaboratory surveys, in which specimens containing different concentrations of the analyte are investigated, can be handled in the same way. If the resulting differences between the chosen percentiles are plotted against the median, and the corresponding two regression lines (upper and lower) are drawn, the results are asymmetric scatter profiles covering the concentration range of the specimen collective. Numerous options are available. Thus, a profile's power of characterizing the scatter correctly can be improved by weighing of the results. Moreover, scatter profiles can be based on different variables of the survey, such as the analytical method, or the observation period, etc. They may be based on the total collective of all results for a given quantity, or they can be constructed for subcollectives of results obtained with a single analytical method. Further, it is possible to present the results of all subcollectives in a single pair of scatter profiles. This latter type of analysis provides profiles of the average scatter for a collective of different analytical methods, which are unaffected by any systematic differences that may exist between the methods.

Calibrators and control samples for bilirubinometers.

The different matrix properties of neonatal serum and commercial control samples can lead to considerable errors in the calibration and control of bilirubinometers. These difficulties can be avoided by calibration with serum from healthy adults which is supplemented with unconjugated bilirubin. But this procedure is impracticable for most routine laboratories. Under certain preconditions, control samples, with bilirubin concentrations determined with correctly calibrated bilirubinometers or spectrophotometers, are also suitable as calibrators. This was established by determination of the bilirubin concentration of 16 different control samples, using both the reference method and correctly calibrated bilirubinometers or spectrophotometers in three or four specialist laboratories. This was also confirmed in several interlaboratory surveys, some involving up to 72 laboratories. The results of these investigations show that a control sample should be used for the calibration of a bilirubinometer only if it meets the following preconditions: 1. There should be no significant difference between the bilirubin values determined with the reference method and with a correctly calibrated spectrophotometer or bilirubinometer. 2. The bilirubin concentration should lie in the range 230-300 mumol/l. The photometric response of bilirubinometers has a limited linear range, so that analytical results greater than 300 mumol/l must be rated as basically unreliable.

Interlaboratory surveys of the determination of tumour markers scatter and repeatability of the results.

Data collected between 1983 and 1991 in interlaboratory surveys of the determination of tumour markers are used to show the magnitude of the scatter of results from different laboratories for the analysis of a single quantity in a given matrix. These data also show that the varying specificity of different reagent combinations appears to make a considerable contribution to this scatter, and that the used reagent combinations were not of uniform quality over a relatively extended period. The results for the following tumour markers were studied: alpha-fetoprotein (AFP), carcinoembryonic antigen (CEA), human chorionic gonadotropin (hCG), human chorionic gonadotropin+beta-subunit (hCG+beta-hCG), tissue polypeptide antigen (TPA), carbohydrate antigen 19-9 (CA 19-9), cancer antigen 15-3 (CA 15-3), cancer antigen 125 (CA 125), prostatic acid phosphatase (PAP) and prostate-specific antigen (PSA).

Our experience with quality control in current growth hormone assays.

For many years now the German Society for Clinical Chemistry has organized interlaboratory surveys by order of the Federal Medical Association. Human growth hormone (GH) is one of 20 endocrine parameters included in a set of two control specimens and offered to survey participants at least 4 times a year. Since no reference method exists for GH, the 'true value' remains unknown. Thus, evaluating survey results is limited to a presentation of how far one participant's results agree with those of other participants using the same or different methods. Medians and percentiles are calculated for all participants and, when possible, for subgroups using the same kit. Over the last years, 60-80 laboratories have participated in each GH survey with 12-16 different kits. The participants' results have shown considerable scatter with differences in GH levels of several hundred percent. The largest discrepancies have occurred with specimens of low GH concentrations, those especially interesting for pediatricians. An evaluation of subgroups using the same kit has shown a relatively good group agreement but large method-dependent differences between these groups. As regards the median, the results of different kits varied by as much as 250%. Accordingly, different reference values have to be applied to different GH kits. For the last 3 years not only have several companies changed their methods, but also one third of the laboratories participating in the survey have switched to other GH kits. This leads us to wonder if all the clinicians served by these laboratories are being informed of this and of any resulting discrepancies in their GH parameters.

External quality assessment of absorbance measurements on spectral and spectral-line photometers.

From 1984-1987, 12 quality control surveys on photometric measurements were carried out in 600-800 laboratories. The participants measured the photometric absorbance of the control samples at 4 wavelengths of the mercury spectrum: 334.1 nm, 365.4 nm, 404.7 nm and 546.1 nm. The medians of the results were without exception lower than the target values, but only very few of them deviated more than 1%. The dispersion of the values did not follow a normal distribution. Two thirds of the values were concentrated within a very small range, while about 10% lay outside the 2- to 3-fold range. It was found that longer wavelengths resulted in a smaller dispersion of readings than shorter ones. Furthermore, precision showed a significant dependency on the absorbance readings of the samples, on the one hand, and on the different photometers, on the other.

External quality control in the determination of neonatal bilirubin. An approach to the improvement of results.

The reliability of bilirubin analyses is especially important in cases of neonatal hyperbilirubinaemia. However, when the means of the results of external quality control surveys and the method-dependent stated values for control sera were compared with reference method values, differences of up to 10% were found. Further inaccuracy arose from interlaboratory imprecision, which showed coefficients of variation of at least 7%, and from greater or lesser interference from contamination of samples with haemoglobin. The present work investigates whether the current situation can be improved by available means.

Report on a joint European quality control survey for neonatal thyrotropin determinations (1986).

This paper presents the results of a joint European external quality control survey for thyrotropin determinations in blood dried on filter paper, carried out in 1986 in cooperation with several national quality control organizations. For the evaluation, 124 participants presented their individual diagnostic classifications in addition to their analytical results. Although, in relation to earlier studies of this kind, there was a significant improvement in interlaboratory precision the results still showed variance which depended on the analytical method and, possibly on the country in which it was performed. Regional differences were also evident in the diagnostic classifications.