Assay Bias Research Articles

Can we interpret serum thyroglobulin results?

Thyroglobulin (Tg) is a tumour marker for differentiated thyroid cancer. Interpretation requires a knowledge of the current thyrotropin (TSH) concentration as secretion is TSH-dependent. While a raised serum Tg may be indicative of residual or recurrent thyroid cancer, trauma to the thyroid (e.g. surgical, biopsy or due to radioiodine treatment) also causes an increase. Tg may be measured when TSH is suppressed and also following recombinant TSH (rhTSH) stimulation. Interpretation of results in pregnancy and in children is discussed. Assay bias and interference by endogenous Tg antibodies (Abs) are the main confounders in the interpretation of results. Although there is an international standard for Tg, there are large differences in results and yet there are few assay-specific clinical decision limits. Patients should therefore be monitored with the same assay. Endogenous TgAbs may cause false-negative interference in immunometric assays and may cause false-positive results in radioimmunoassay. Although the measurement of TgAbs has been advocated for predicting interference, it is now clear that interference can still occur when TgAbs have not been detected, the effect being TgAb-assay-specific. Approaches to identifying those samples where there may be interference are discussed. The laboratory should have a protocol for the investigation of possible interferences and data on the bias of the Tg assay that they use. An appreciation of the clinical uses of the service is required as an understanding by endocrinologists, oncologists and endocrine surgeons of the analytical limitations of the service.

Comparison of serum cortisol measurement by immunoassay and liquid chromatography-tandem mass spectrometry in patients receiving the 11β-hydroxylase inhibitor metyrapone

Monaghan et al. present convincing data that serum cortisol results derived from a commonly used immunoassay must be interpreted with great caution in metyrapone-treated patients, given the findings of their extensive method comparison study using liquid chromatography-tandem mass spectrometry (LC-MS/MS). These data support conclusions from previous publications which invoke cross-reactivity with cortisol precursors generated by the metyrapone blockade of the cortisol biosynthetic pathway. 11-Deoxycortisol (11DOC) is a likely candidate, and Monaghan et al. show a modest correlation between 11DOC and the difference between the immunoassay and the LC-MS/MS (Pearson’s correlation coefficient of 0.47). The authors state that variation in the data is so great that the assay bias cannot be corrected for by a knowledge of either the metyrapone dose or the serum 11DOC concentration. Based on previous data from a series of metyrapone-treated patients, we support this finding and further concluded that interference from 11DOC alone could not account for the difference between the serum cortisol immunoassay and LC-MS/MS results. We also showed that this is a widespread problem as three commonly used cortisol immunoassays are affected. Initially, we calculated the degree of 11DOC cross-reactivity for each of the cortisol assays by performing spiking studies with 11DOC. We then demonstrated in two metyraponetreated patients that the difference between the immunoassay and the LC-MS/MS serum cortisol results could not be entirely accounted for by the 11DOC concentration given the established cross-reactivity. Data presented in this abstract and unpublished data from additional local metyrapone-treated patients support the hypothesis of Monaghan et al. that this effect is most likely to be seen in adrenocorticotropic hormone (ACTH)-dependent Cushing’s syndrome, as the degree of accumulation of steroid precursors is likely to be a function of both the magnitude of the ACTH drive and the extent of metyrapone blockade. In this way it is not dissimilar to congenital adrenal hyperplasia, where interference in serum cortisol immunoassays is also well described. If the data are available, it would be enlightening to see a correlation between the plasma ACTH concentration and the difference between the serum cortisol immunoassay and LC-MS/MS result in the cohort described in this paper, as well as cross-reactivity data for exogenous 11DOC. While the nature of the interference caused by metyrapone blockade in serum cortisol immunoassays remains elusive, we wholeheartedly support the conclusion of Monaghan et al. that current immunoassay methods for serum cortisol lack the specificity for use in metyrapone-treated patients, and this interference is likely to be more marked in patients with a high ACTH drive.

The Need for Standardization of Tacrolimus Assays

Owing to the lack of an internationally recognized tacrolimus reference material and reference method, current LC-MS and immunoassay test methods used to monitor tacrolimus concentrations in whole blood are not standardized. The aim of this study was to assess the need for tacrolimus assay standardization. We sent a blinded 40-member whole-blood tacrolimus proficiency panel (0-30 μg/L) to 22 clinical laboratories in 14 countries to be tested by the following assays: Abbott ARCHITECT (n = 17), LC-MS (n = 9), and Siemens Dade Dimension (n = 5). Selected LC-MS laboratories (n = 4) also received a common calibrator set. We compared test results to a validated LC-MS method. Four samples from the proficiency panel were assigned reference values by using exact-matching isotope-dilution mass spectrometr at LGC. The range of CVs observed with the tacrolimus proficiency panel was as follows: LC-MS 11.4%-18.7%, ARCHITECT 3.9%-9.5%, and Siemens Dade 5.0%-48.1%. The range of historical within-site QC CVs obtained with the use of 3 control concentrations were as follows: LC-MS low 3.8%-10.7%, medium 2.0%-9.3%, high 2.3%-9.0%; ARCHITECT low 2.5%-9.5%, medium 2.5%-8.6%, high 2.9%-18.6%; and Siemens/Dade Dimension low 8.7%-23.0%, medium 7.6%-13.2%, high 4.4%-10.4%. Assay bias observed between the 4 LC-MS sites was not corrected by implementation of a common calibrator set. Tacrolimus assay standardization will be necessary to compare patient results between clinical laboratories. Improved assay accuracy is required to provide optimized drug dosing and consistent care across transplant centers globally.

The Effect of Hematocrit and Punch Location on Assay Bias During Quantitative Bioanalysis of Dried Blood Spot Samples

Dried blood spot (DBS) sampling, the collection of whole blood samples on paper, is being evaluated for its use in quantitative analysis. The aim of this investigation was to evaluate the impact of sample hematocrit (HCT) and punch location on assay bias in the measurement of compounds with varying physicochemical properties, when a portion of a DBS sample is analyzed. In addition, the statistical significance of factors such as compound, card type, HCT level, spot position on a card and the location of the punch on the spot were evaluated. The results of this work indicate that for quantitative analysis using a portion of the DBS, notable bias (>15%) exists due to the effect of HCT and nonhomogeneous distribution of compound across the spot. The statistical analysis of results from a split-split-plot designed study showed that the factors of compound, card type and HCT were statistically significant (p < 0.05) both individually and in complex interactions, which affect the accuracy of quantitative concentrations obtained from DBS samples for each compound on each card type. To achieve the acceptance criteria (±15%) for accuracy when removing a portion of the DBS spot for quantitative analysis, HCT and punch location for each selected card type will need to be assessed during validation and sample analysis.

The ABC of non‐inferiority margin setting from indirect comparisons

In a non-inferiority trial to assess a new investigative treatment, there may need to be consideration of an indirect comparison with placebo using the active control in the current trial. We can, therefore, use the fact that there is a common active control in the comparisons of the investigative treatment and placebo. In analysing a non-inferiority trial, the ABC of: Assay sensitivity, Bias minimisation and Constancy assumption needs to be considered. It is highlighted how the ABC assumptions can potentially fail when there is placebo creep or a patient population shift. In this situation, the belief about the placebo response expressed in terms of a prior probability in Bayesian formulation could be used with the observed treatment effects to set the non-inferiority limit.

Simultaneous measurement of cyclosporin A and tacrolimus from dried blood spots by ultra high performance liquid chromatography tandem mass spectrometry

Cyclosporin A (CsA) and tacrolimus are immunosuppressant drugs principally used in solid organ transplant recipients. Therapeutic drug monitoring (TDM) of both drugs is essential to avoid toxicity related to overdosage, and transplant rejection from underdosage. This necessitates frequent hospital visits to phlebotomy services. Capillary blood sampling onto dried blood spots (DBS) provides numerous advantages to venous whole blood sampling, including the ability for patients to send DBS to the laboratory by post, significantly reducing the number of unnecessary hospital visits. We have developed a novel, simple and rapid method for the extraction and simultaneous UPLC–MS/MS measurement of both CsA and tacrolimus from DBS. The extraction method involved a simple 30min hot solvent extraction with ultrasonication. Extract (10μL) was injected onto a Waters Acquity UPLC column filter unit security frit, coupled to a Waters Acquity BEH C18 UPLC column, with methanolic mobile phase gradient elution. Eluant was connected to a Waters Quattro Premier XE tandem mass spectrometer operating in ES+ mode. We detected multiple reaction monitoring (MRM) transitions of m/z 1220>1203 and 1231.9>1215.1 for CsA and d12 CsA respectively which co-eluted at 1.30min, and 821.6>768.5 and 809.6>756.5 for tacrolimus and ascomycin respectively which co-eluted at 1.17min. Ion suppression was negligible. Mean recovery was 95.5% for CsA and 92.8% for tacrolimus. Limit of detection and limit of quantitation were both 8.5μg/L for CsA, and 0.5 and 2.3μg/L respectively for tacrolimus. The assay was linear up to 1500μg/L for CsA (r2=0.9999), and up to 50μg/L for tacrolimus (r2=0.9994). Mean intra assay imprecision, inter assay imprecision and bias were all <10% for both CsA and tacrolimus. DBS were stable for at least 14 days at room temperature. Comparison of the DBS UPLC–MS/MS method and the routine venous whole blood LC–MS/MS assay demonstrated good agreement between the two methods for both drugs. We have developed a simple and robust method for the extraction and simultaneous measurement of CsA and tacrolimus from DBS. The method will allow TDM of transplant recipients to proceed at home using capillary blood sampling.

External Quality Assurance of Antithrombin, Protein C, and Protein S Assays: Results of the College of American Pathologists Proficiency Testing Program in Thrombophilia

Hereditary and acquired deficiencies of antithrombin (AT), protein C (PC), and protein S (PS) are risk factors for venous thromboembolism. Proper diagnosis requires high-quality assays for these proteins. To determine the accuracy and interlaboratory precision of AT, PC, and PS assays used by laboratories participating in the United States College of American Pathologists proficiency testing program in thrombophilia and to grade the performance of laboratories. Standardized normal plasma with assigned analyte values was sent in 2 separate challenges to participating laboratories. Participants measured AT, PC, and PS levels using local methods. When compared with the assigned values for the international standard, the order of assay accuracy from highest to lowest was AT activity, PC antigen, AT antigen, total PS antigen, PC activity, PS activity, and free PS antigen (range of assay bias, 2.6%-8.8%). The order of assay precision from highest to lowest was PC activity, AT activity, AT antigen, total PS antigen, PS activity, free PS antigen, and PC antigen (range of assay coefficient of variation, 6.1%-20.0%). Most testing events (87.8%) could be graded as pass or fail using a target range of ±3 standard deviations from the method-specific mean. The pass rate was 98.2% for all AT, PC, and PS testing events combined. Accuracy and precision were higher for AT assays and lower for PC and PS assays. It was feasible to grade individual laboratory performance.

Serum globulins contribute to the discrepancies observed between the bromocresol green and bromocresol purple assays of serum albumin concentration

The bromocresol green (BCG) and bromocresol purple (BCP) assays often yield discordant serum albumin results. This study seeks to test the hypothesis that bias in albumin results are influenced by the concentration of serum globulin subtypes. The concentrations of serum albumin, α1-globulin, α2-globulin, β-globulin and γ-globulin are determined in 197 human serum specimens by total serum protein quantification and protein electrophoresis, and by the BCG and BCP assays. The influence of globulins on albumin measurement is validated with protein mixtures of albumin and globulins. The BCG assay bias was directly proportional to the concentrations of α1-globulin and α2-globulin, and inversely correlated with the concentrations of β-globulin and γ-globulin (r2=0.793). The BCP assay bias was inversely proportional to the concentration of α1-globulin and α2-globulin (r2=0.464) but not related to the concentrations of β-globulin or γ-globulin. Among the 197 study participants, those with nephrotic syndrome had a significantly higher level of α2-globulin compared to those in other categories. Thus, the authors conclude that serum globulins contribute to the bias seen in the BCG and BCP assays, with the greatest effects observed for α-globulin on the BCG assay where higher concentrations contributed to a higher bias.

Multilaboratory Comparison of Streptococcus pneumoniae Opsonophagocytic Killing Assays and Their Level of Agreement for the Determination of Functional Antibody Activity in Human Reference Sera

Antibody-mediated killing of Streptococcus pneumoniae (pneumococcus) by phagocytes is an important mechanism of protection of the human host against pneumococcal infections. Measurement of opsonophagocytic antibodies by use of a standardized opsonophagocytic assay (OPA) is important for the evaluation of candidate vaccines and required for the licensure of new pneumococcal conjugate vaccine formulations. We assessed agreement among six laboratories that used their own optimized OPAs on a panel of 16 human reference sera for 13 pneumococcal serotypes. Consensus titers, estimated using an analysis-of-variance (ANOVA) mixed-effects model, provided a common reference for assessing agreement among these laboratories. Agreement was evaluated in terms of assay accuracy, reproducibility, repeatability, precision, and bias. We also reviewed four acceptance criterion intervals for assessing the comparability of protocols when assaying the same reference sera. The precision, accuracy, and concordance results among laboratories and the consensus titers revealed acceptable agreement. The results of this study indicate that the bioassays evaluated in this study are robust, and the resultant OPA values are reproducible for the determination of functional antibody titers specific to 13 pneumococcal serotypes when performed by laboratories using highly standardized but not identical assays. The statistical methodologies employed in this study may serve as a template for evaluating future multilaboratory studies.

The Effect of Hematocrit on Assay Bias When Using DBS Samples for The Quantitative Bioanalysis of Drugs

As hematocrit levels are known to vary between individuals and with disease state, its effect on the physical characteristics of dried blood spot (DBS) samples and on the accurate quantification of analytes within these samples is examined. The area of DBS samples decreases with increasing hematocrit levels in a linear manner on the three cellulose paper substrates tested. Furthermore, a bias was observed in the concentrations of two analytes determined in DBS samples at different hematocrits, which in some cases exceeded acceptable values, particularly for hematocrits outside normal values. If it is expected that the hematocrit of study samples will vary from values considered normal, then its effect on the quantitative determination of an analyte in DBS samples should be investigated as part of the method development and validation. If an unacceptable effect is observed, then this will need to be addressed, by modification of the analytical method, or the inclusion of quality control samples at different hematocrit levels to show control of the assay.

Optimized Quantification of Fragmented, Free Circulating DNA in Human Blood Plasma Using a Calibrated Duplex Real-Time PCR

BackgroundDuplex real-time PCR assays have been widely used to determine amounts and concentrations of free circulating DNA in human blood plasma samples. Circulatory plasma DNA is highly fragmented and hence a PCR-based determination of DNA concentration may be affected by the limited availability of full-length targets in the DNA sample. This leads to inaccuracies when counting PCR target copy numbers as whole genome equivalents.Methodology/Principal FindingsA model system was designed allowing for assessment of bias in a duplex real-time PCR research assay. We collected blood plasma samples from male donors in pools of 6 to 8 individuals. Circulatory plasma DNA was extracted and separated by agarose gel electrophoresis. Separated DNA was recovered from the gel in discrete size fractions and analyzed with different duplex real-time PCR Taqman assays detecting a Y chromosome-specific target and an autosomal target. The real-time PCR research assays used differed significantly in their ability to determine the correct copy number ratio of 0.5 between Y chromosome and autosome targets in DNA of male origin. Longer PCR targets did not amplify quantitatively in circulatory DNA, due to limited presence of full-length target sequence in the sample.ConclusionsPCR targets of the same small size are preferred over longer targets when comparing fractional circulatory DNA concentrations by real-time PCR. As an example, a DYS14/18S duplex real-time PCR research assay is presented that correctly measures the fractional concentration of male DNA in a male/female mixture of circulatory, fragmented DNA.

Laboratory services for thyroglobulin and implications for monitoring of differentiated thyroid cancer

Serum thyroglobulin is used as a marker for monitoring differentiated thyroid cancer. There are a number of analytical challenges: the need for stability of the assay over decades, differences in...

Opportunities to Optimize Tacrolimus Therapy in Solid Organ Transplantation: Report of the European Consensus Conference

In 2007, a consortium of European experts on tacrolimus (TAC) met to discuss the most recent advances in the drug/dose optimization of TAC taking into account specific clinical situations and the analytical methods currently available and drew some recommendations and guidelines to help clinicians with the practical use of the drug. Pharmacokinetic, pharmacodynamic, and more recently pharmacogenetic approaches aid physicians to individualize long-term therapies as TAC demonstrates a high degree of both between- and within-individual variability, which may result in an increased risk of therapeutic failure if all patients are administered a uniform dose. TAC has undoubtedly benefited from therapeutic drug monitoring, but interpretation of the blood concentration is confounded by the relative differences between the assays. Single time points, limited sampling strategies, and area under concentration-time curve have all been considered to determine the most appropriate sampling procedure that correlates with efficacy. Therapeutic trough TAC concentration ranges have changed since the initial introduction of the drug, while still maintaining adequate immunosuppression and avoiding drug-related adverse effects. Pharmacodynamic markers have also been considered advantageous to the clinician, which may better reflect efficacy and safety, taking into account the between-individual variability rather than whole blood concentrations. The choice of method, differences between methods, and potential pitfalls of the method should all be considered when determining TAC concentrations. The recommendations of this consensus meeting regarding the analytical methods include the following: encourage the development and promote the use of analytical methods displaying a lower limit of quantification (1 ng/mL), perform careful validation when implementing a new analytical assay, participate in external proficiency testing programs, promote the use of certified material as calibrators in high-performance liquid chromatography with mass spectrometric detection methods, and take account of the assay and intermethod bias when comparing clinical trial outcomes. It is also important to consider that TAC concentrations may also be influenced by other factors such as specific pharmacokinetic characteristics associated with the population, drug interactions, pharmacogenetics, adverse events that may alter TAC concentrations, and any change in the oral formulation that may result in pharmacokinetic changes. This meeting emphasized the importance of obtaining multicenter prospective trials to assess the efficacy of alternative strategies to TAC trough concentrations whether it is other single time points or area under the concentration-time curve Bayesian estimation using limited sampling strategies and to select, standardize, and validate routine biomarkers of TAC pharmacodynamics.

Multi-Site PCR-Based CMV Viral Load Assessment-Assays Demonstrate Linearity and Precision, but Lack Numeric Standardization: A Report of the Association for Molecular Pathology

Viralload (VL) assessment of cytomegalovirus (CMV) by real-time PCR is an important tool for diagnosing and monitoring CMV viremia in patients with compromised immune systems. We report results from a sample exchange organized by members of the Association for Molecular Pathology that compared PCR results from 23 laboratories; 22 such laboratories used a laboratory-developed real-time PCR assay and one laboratory used a competitive PCR assay. The samples sent to each laboratory were comprised of a dilution panel of CMV virion-derived reference materials that ranged from 0 to 500,000 copies/ml. Accuracy, linearity, and intralaboratory precision were established for the different laboratory-developed assays. Overall, PCR results were linear for each laboratory (R(2) > 0.97 in all but two). While 13 laboratories showed no significant quantitative assay bias, 10 laboratories reported VLs that were significantly different compared with expected values (bias range, -0.82 to 1.4 logs). The intralaboratory precision [mean coefficient of variance of 2% to 5% (log-scale)] suggested that changes in VLs of less than 3- to fivefold may not be significantly different. There was no significant association between laboratory-specific technical variables (PCR platform, calibrator, extraction method) and assay linearity or accuracy. These data suggested that, within each laboratory, relative VL values were linear, but additional method standardization and a CMV DNA reference standard are needed to allow laboratories to achieve comparable numeric results.

Assay bias may invalidate decision limits and affect comparability of serum thyroglobulin assay methods: An approach to reduce interpretation differences

Background Thyroglobulin (Tg) assays are used to monitor patients after ablative therapy for differentiated thyroid carcinoma (DTC). Detectable Tg suggests persistence or recurrence of disease. However, even with very sensitive assays, false positive or false negative results are likely to occur depending on the presence of positive or negative assay bias. Methods We compared 6 different Tg immunometric assays in 53 treated DTC patients in whom serum Tg during TSH-suppression therapy was below an earlier established clinical decision limit of 3 pmol/l (2 ng/ml) by our routine Tg assay, which was one of these 6. Serum Tg was measured before and after administration of recombinant human TSH (r-TSH). Assay bias and confidence intervals for each assay were calculated from the pre-r-TSH Tg values in the patient subgroup in which no significant rise after r-TSH was observed by any method. Results Upper reference limits based on confidence intervals instead of functional sensitivity resulted in significant improvement of concordance between assays and prediction of Tg-rise after r-TSH. Conclusion Confidence intervals in which assay bias are taken into account, form a better guideline for detection of possible persistence or recurrence of DTC than functional sensitivities.

Comparison of 11 Human Insulin Assays: Implications for Clinical Investigation and Research

The American Diabetes Association task force on standardization of insulin assays in 1996 showed wide variation in assay bias. Newer assays are specific for insulin, with several now available on automated immunoassay analyzers. In 2004, we compared 11 commercially available insulin assays by analyzing 150 serum samples (99 fasting/51 postprandial) from study participants with various degrees of glucose intolerance (exclusions being type 1 diabetes, insulin treatment, or presence of insulin antibodies). All assays were calibrated against International Reference Preparation 66/304. One assay was not specific for insulin and another was an RIA; 10 assays used enzyme/chemiluminescent labels. Bland-Altman difference plots were modified to use the mean insulin from all assays on the x-axis as a common comparator. As in the 1996 study, insulin values from the different assays varied by a factor of 2, with the nonspecific assay ranking in the middle of the distribution. Spearman rank correlation coefficients, for ranking samples vs the mean, were 0.983-0.997. Both offsets and concentration-dependent differences were seen in the modified difference plots. Imprecision (mean CV) for automated assays (3%) was not significantly different from manual assays (5%). Similar values were obtained when one automated assay was run in laboratories in both the UK and the US. Results of 1 assay showed lower insulin concentrations in heparinized plasma than in serum. Assay performance must be considered before comparing insulin results. The 2-fold variation in insulin results may be related to specificity, manufacturers' calibration procedures or conversion factors.

Thyroid guidelines - are thyroid-stimulating hormone assays fit for purpose?

Most thyroid-stimulating hormone (TSH) assays now have the sensitivity required by thyroid guidelines and allow the reliable identification of patients with both overt and subclinical hyperthyroidism. Clinical guidelines usually quote decision limits for TSH, but often ignore the issue of whether variability in bias between assays should be considered when such decision limits are implemented. Clinicians and laboratories should appreciate that these decision limits arise largely from historical data that used TSH assays with poorly defined bias. It is thus unlikely that laboratories will be able to apply an appropriate method-related bias adjustment to these TSH cut-offs. Clinicians should appreciate that TSH decision limits should thus be regarded as typical target figures rather than an absolute cut-off and thus can be applied with some degree of flexibility. There is currently insufficient evidence to justify a significant lowering of the upper reference limit for TSH, but fine-tuning of current reference ranges is required since there appears to be no association between the ranking of the assay bias in the UK National External Quality Assessment Service scheme and the manufacturers' quoted reference ranges. There is room for further improvement in TSH assays and this can best be achieved if manufacturers, laboratories and clinicians work together to produce TSH assays and reference ranges that show closer agreement between methods. Until this is achieved, future studies that examine the relationship of TSH with symptoms and treatment should ensure that sufficient information is included in the publication to allow the method related bias of the TSH assay to be clearly described.

Errors in centering of array data can induce biases in correlation estimates

For a general class of microarray data sets, the sample specific centering errors can constitute correlated measurement errors. These errors are perfectly correlated within samples in the sense that all spot assays within a given sample are centered with the same error. Under certain biological conditions, the centering errors can also be correlated to a specific subset of spot assays. In contrast to the within-sample correlated errors, the correlation of centering errors to select spot assays can be thought of as an across-sample correlation (e.g., for the cases presented in this manuscript, the spot assay by centering error correlation only exists if the skew of the distribution varies across samples with the variation dependent on a subset of spot assays). We demonstrate that the perfect within sample correlation of the centering errors induces a positive bias in spot assay by spot assay correlation estimates while negative across-sample correlations can induce a negative bias in said estimates. We provide analyses of a real and simulated data set and demonstrate that, under realistic conditions, the centering errors can induce biases in spot assay by spot assay correlation estimates while not significantly affecting permutation-based thresholds for significance; a combination which can lead to a breakdown in error control.

Effect of variability in assay bias and imprecision on external quality assessment bias and imprecision measures

External quality assessment/proficiency testing programmes that report consensus means and coefficients of variations (CVs) are a potential source of information on assay bias and imprecision. This study examined the effect of variability in assay bias and imprecision on consensus means and CVs, using a computerized spreadsheet model. A model with varying assay bias (mean and standard deviation) and assay imprecision (mean and standard deviation) was developed using a MS Excel 2003 spreadsheet with a macro to generate pseudo-random numbers from a Gaussian distribution. The means, standard deviations and CVs of these data points were considered to simulate the consensus measures reported in external quality assessment testing programmes. The simulated consensus CV was very sensitive to increases in assay bias variability, with greater effects at low levels of mean assay imprecision. Changes in assay imprecision variability had a much smaller effect on the simulated consensus CV, while the simulated consensus mean was relatively insensitive to any change in the other variables. Consensus imprecision measures from external quality assessment programmes are unreliable measures of assay imprecision. Consensus means are relatively unaffected by variability in assay bias and imprecision, and can be used as reliable measures of assay bias from a statistical standpoint.

Recommendations for Improving Serum Creatinine Measurement: A Report from the Laboratory Working Group of the National Kidney Disease Education Program

Reliable serum creatinine measurements in glomerular filtration rate (GFR) estimation are critical to ongoing global public health efforts to increase the diagnosis and treatment of chronic kidney disease (CKD). We present an overview of the commonly used methods for the determination of serum creatinine, method limitations, and method performance in conjunction with the development of analytical performance criteria. Available resources for standardization of serum creatinine measurement are discussed, and recommendations for measurement improvement are given. The National Kidney Disease Education Program (NKDEP) Laboratory Working Group reviewed problems related to serum creatinine measurement for estimating GFR and prepared recommendations to standardize and improve creatinine measurement. The NKDEP Laboratory Working Group, in collaboration with international professional organizations, has developed a plan that enables standardization and improved accuracy (trueness) of serum creatinine measurements in clinical laboratories worldwide that includes the use of the estimating equation for GFR based on serum creatinine concentration that was developed from the Modification of Diet in Renal Disease (MDRD) study. The current variability in serum creatinine measurements renders all estimating equations for GFR, including the MDRD Study equation, less accurate in the normal and slightly increased range of serum creatinine concentrations [<133 micromol/L (1.5 mg/dL)], which is the relevant range for detecting CKD [<60 mL.min(-1).(1.73 m2)(-1)]. Many automated routine methods for serum creatinine measurement meet or exceed the required precision; therefore, reduction of analytical bias in creatinine assays is needed. Standardization of calibration does not correct for analytical interferences (nonspecificity bias). The bias and nonspecificity problems associated with some of the routine methods must be addressed.