Repeatability imprecision from analysis of duplicates of patient samples and control materials.

Measurement imprecision is usually calculated from measurement results of the same stabilized control material(s) obtained over time, and is therefore, principally, only valid at the concentration(s) of the selected control material(s). The resulting uncertainty has been obtained under reproducibility conditions and corresponds to the conventional analytical goals. Furthermore, the commutability of the control materials used determines whether the imprecision calculated from the control materials reflects the imprecision of measuring patient samples. Imprecision estimated by measurements of patient samples uses fully commutable samples, freely available in the laboratories. It is commonly performed by calculating the results of routine patient samples measured twice each. Since the duplicates are usually analysed throughout the entire concentration interval of the patient samples processed in the laboratory, the result will be a weighted average of the repeatability imprecision measured in the chosen measurement intervals or throughout the entire interval of concentrations encountered in patient care. In contrast, the uncertainty derived from many measurements of control materials over periods of weeks is usually made under reproducibility conditions. Consequently, the repeatability and reproducibility imprecision play different roles in the inference of results in clinical medicine. The purpose of the present review is to detail the properties of the imprecision calculated by duplicates of natural samples, to explain how it differs from imprecision calculated from single concentrations of control materials, and to elucidate what precautions need to be taken in case of bias, e.g. due to carry-over effects.

An experimental study of methods for the analysis of variance components in the inference of laboratory information.

Measurement uncertainty (MU) can be estimated and calculated by different procedures, representing different aspects and intended use. It is appropriate to distinguish between uncertainty determined under repeatability and reproducibility conditions, and to distinguish causes of variation using analysis of variance components. The intra-laboratory MU is frequently determined by repeated measurements of control material(s) of one or several concentrations during a prolonged period of time. We demonstrate, based on experimental results, how such results can be used to identify the repeatability, 'pure' reproducibility and intra-laboratory variance as the sum of the two. Native patient material was used to establish repeatability using the Dahlberg formula for random differences between measurements and an expanded Dahlberg formula if a non-random difference, e.g. bias, was expected. Repeatability and reproducibility have different clinical relevance in intensive care compared to monitoring treatment of chronic diseases, comparison with reference intervals or screening.

Measurement repeatability profiles of eight frequently requested measurands in clinical chemistry determined by duplicate measurements of patient samples.

Measurement uncertainties in clinical chemistry are commonly regarded as heteroscedastic - having a constant relative standard deviation irrespective of the concentration of the measurand. The uncertainty is usually determined at two concentrations using stabilized control materials and assumed to represent the analytical goal. The purpose of the present study was to use duplicates of unselected patient samples to calculate the absolute and relative repeatability component of the intra-laboratory measurement uncertainty from duplicates, using the Dahlberg formula and analysis of variance components. Estimates were made at five different concentration intervals of ALT, AST, Calcium, Cholesterol, Creatinine, CRP, Triglycerides and TSH covering the entire concentration interval of the patient cohort. This partioning allows detailing their repeatability profiles. The calculations of the profiles were based on randomly selected results from sets of duplicates ranging from 12,000 to 65,000 pairs. The repeatability of the measurands showed substantial variability within the measuring interval. Therefore, characterizing imprecision profiles as purely homo- or heteroscedastic or by a single number may not be optimal for the intended use. The present data make a case for nuancing the evaluation of analytical goals and minimal differences of measurement results by establishing uncertainty profiles under repeatability conditions, using natural patient samples.

The use of error and uncertainty methods in the medical laboratory.

Error methods - compared with uncertainty methods - offer simpler, more intuitive and practical procedures for calculating measurement uncertainty and conducting quality assurance in laboratory medicine. However, uncertainty methods are preferred in other fields of science as reflected by the guide to the expression of uncertainty in measurement. When laboratory results are used for supporting medical diagnoses, the total uncertainty consists only partially of analytical variation. Biological variation, pre- and postanalytical variation all need to be included. Furthermore, all components of the measuring procedure need to be taken into account. Performance specifications for diagnostic tests should include the diagnostic uncertainty of the entire testing process. Uncertainty methods may be particularly useful for this purpose but have yet to show their strength in laboratory medicine. The purpose of this paper is to elucidate the pros and cons of error and uncertainty methods as groundwork for future consensus on their use in practical performance specifications. Error and uncertainty methods are complementary when evaluating measurement data.

Extending laboratory automation to the wards: effect of an innovative pneumatic tube system on diagnostic samples and transport time.

BACKGROUND: The innovative pneumatic tube system (iPTS) transports one sample at a time without the use of cartridges and allows rapid sending of samples directly into the bulk loader of a laboratory automation system (LAS). We investigated effects of the iPTS on samples and turn-around time (TAT). METHODS: During transport, a mini data logger recorded the accelerations in three dimensions and reported them in arbitrary area under the curve (AUC) units. In addition representative quantities of clinical chemistry, hematology and coagulation were measured and compared in 20 blood sample pairs transported by iPTS and courier. RESULTS: Samples transported by iPTS were brought to the laboratory (300 m) within 30 s without adverse effects on the samples. The information retrieved from the data logger showed a median AUC of 7 and 310 arbitrary units for courier and iPTS transport, respectively. This is considerably below the reported limit for noticeable hemolysis of 500 arbitrary units. CONCLUSIONS: iPTS reduces TAT by reducing the hands-on time and a fast transport. No differences in the measurement results were found for any of the investigated 36 analytes between courier and iPTS transport. Based on these findings the iPTS was cleared for clinical use in our hospital.

Is the combination of trueness and precision in one expression meaningful? On the use of total error and uncertainty in clinical chemistry.

The performance of all measurement procedures used in routine clinical laboratories shall be verified; a minimum is to verify the precision and trueness of the results. This is well established and adequate recommendations and procedures are available. Conveying this information in a form that is adequate and understandable for the practical end-user in the health care sector is still a much debated issue. By tradition, since several decades, the "total error" (TE) is presented, a quantity that is the linear sum of an imprecision and bias. Since any combination of the two can yield the same TE it may not be very helpful in finding and correcting a root-cause for an unacceptable value. Also, an acceptable TE may hide an unacceptable level of its constituents. An alternative is the measurement uncertainty (MU), which is recommended by accreditation and standardizing bodies The MU separates the imprecision and bias and expresses an interval around a best estimate within which the true value is expected with a certain probability. We describe the reporting the best estimate of a measurement result and describe how the uncertainty of the result, can be calculated, using simple custom-made software.

Discrepant post filter ionized calcium concentrations by common blood gas analyzers in CRRT using regional citrate anticoagulation.

INTRODUCTION: Ionized calcium (iCa) concentration is often used in critical care and measured using blood gas analyzers at the point of care. Controlling and adjusting regional citrate anticoagulation (RCA) for continuous renal replacement therapy (CRRT) involves measuring the iCa concentration in two samples: systemic with physiological iCa concentrations and post filter samples with very low iCa concentrations. However, modern blood gas analyzers are optimized for physiological iCa concentrations which might make them less suitable for measuring low iCa in blood with a high concentration of citrate. We present results of iCa measurements from six different blood gas analyzers and the impact on clinical decisions based on the recommendations of the dialysis' device manufacturer. METHOD: The iCa concentrations of systemic and post filter samples were measured using six distinct, frequently used blood gas analyzers. We obtained iCa results of 74 systemic and 84 post filter samples from patients undergoing RCA for CRRT at the University Medicine of Greifswald. RESULTS: The systemic samples showed concordant results on all analyzers with median iCa concentrations ranging from 1.07 to 1.16 mmol/L. The medians of iCa concentrations for post filter samples ranged from 0.21 to 0.50 mmol/L. Results of >70% of the post filter samples would lead to major differences in decisions regarding citrate flow depending on the instrument used. CONCLUSION: Measurements of iCa in post filter samples may give misleading information in monitoring the RCA. Recommendations of the dialysis manufacturer need to be revised. Meanwhile, little weight should be given to post filter iCa. Reference methods for low iCa in whole blood containing citrate should be established.

Optimizing centrifugation of coagulation samples in laboratory automation.

BACKGROUND: High acceleration centrifugation conditions are used in laboratory automation systems to reduce the turnaround time (TAT) of clinical chemistry samples, but not of coagulation samples. This often requires separate sample flows. The CLSI guideline and manufacturers recommendations for coagulation assays aim at reducing platelet counts. For measurement of prothrombin time (PT) and activated partial thromboplastin time (APTT) platelet counts (Plt) below 200×10(9)/L are recommended. Other coagulation assays may require even lower platelet counts, e.g., less than 10 × 10(9)/L. Unifying centrifugation conditions can facilitate the integration of coagulation samples in the overall workflow of a laboratory automation system. METHODS: We evaluated centrifugation conditions of coagulation samples by using high acceleration centrifugation conditions (5 min; 3280×g) in a single and two consecutive runs. RESULTS of coagulation assays [PT, APTT, coagulation factor VIII (F. VIII) and protein S] and platelet counts were compared after the first and second centrifugation. RESULTS: Platelet counts below 200×10(9)/L were obtained in all samples after the first centrifugation and less than 10 × 10(9)/L was obtained in 73% of the samples after a second centrifugation. Passing-Bablok regression analyses showed an equal performance of PT, APTT and F. VIII after first and second centrifugation whereas protein S measurements require a second centrifugation. CONCLUSIONS: Coagulation samples can be integrated into the workflow of a laboratory automation system using high acceleration centrifugation. A single centrifugation was sufficient for PT, APTT and F. VIII whereas two successive centrifugations appear to be sufficient for protein S activity.

Diagnosis of late presenting subarachnoid hemorrhage: comparison of methods for cerebrospinal fluid ferritin.

BACKGROUND: The majority of subarachnoid hemorrhage (SAH) is diagnosed using imaging techniques. The sensitivity of computed tomography scans decreases with increasing time after the bleeding event which can lead to false negative CT scans. Spectrophotometry and microscopic investigations of the cerebrospinal fluid (Csf) can provide additional diagnostic support, but may not be available for emergency diagnoses. Csf-Ferritin has been suggested as an alternative additional marker for SAH that present late and has a potency to be measured in a routine laboratory. METHODS: A routine Ferritin chemiluminescent assay (Dimension Vista) was compared with a branded and CE-marked Csf-Ferritin nephelometric assay (BN ProSpec) using surplus routine patient samples. We calculated imprecision at pertinent concentrations, compared patient samples, and established reference intervals. RESULTS: The standard deviation was about a third for the Dimension Vista assay compared to that of the BN ProSpec assay at the three tested concentrations. The correlation showed a systematic difference between the methods but the correlation was high (r = 0.955). Accordingly, the reference intervals were higher for the BN ProSPec (2.7-16.8 µg/L) than for the Dimension Vista (2.0-12.6 µg/L). CONCLUSION: The precision of the Dimension Vista measurements was considerably better than that of the BN ProSpec. The Dimension Vista results correlated well with those of the comparative method, yielding slightly lower values. This is reflected in the reference intervals. These findings permit the use of the routinely available Ferritin assay of the Dimension Vista for measuring Csf-Ferritin and complementing the late diagnosis of SAH outside office hours of specialized Csf laboratories.

Comparison of the 99th percentiles of three troponin I assays in a large reference population.

BACKGROUND: The IFCC Task Force on Clinical Applications of Cardiac Biomarkers suggests comparing several contemporary sensitive troponin assays in the same, sufficiently large reference population. METHODS: Three contemporary sensitive assays were used to measure troponin I concentration in samples from a uniquely large healthy population (2404 individuals) and in a sub-group with tighter inclusion criteria of 908 individuals. The 99th percentiles were calculated using quantile regression which takes the entire population into account. RESULTS: The 99th percentile for the ARCHITECT STAT Troponin I assay was 21 ng/L, 31 ng/L for the ADVIA Centaur Troponin I-Ultra assay and 28 ng/L for the Dimension Vista cTnI assay. Significantly higher values were found in males than in women only in the Dimension Vista cTnI assay and in the subgroup for the ARCHITECT STAT Troponin I assay. CONCLUSIONS: Quantile regression provides a tool to accurately estimate the 99th percentile and establish a continuous function of the relation between the 99th percentile and the age and gender. There was no age dependency demonstrated. A gender difference was found in one assay but inconclusive in another and not demonstrated in a third.

Effects of Consuming Fermented Fish (Surströmming) on the Fecal Microflora in Healthy Individuals.

Surströmming, a Swedish fermented fish, loved by some and avoided by others, occurs in many reports on improved or cured gastrointestinal problems even by a single meal. We tested the hypothesis that the microbes of the fermented food might have a potency to modify the gut microbiome. Two groups of voluntary participants (11 male, 8 female; aged 20-80 years) were exposed to a single meal containing the fish. A 7-day dietary intervention was carried out comprising the fish as the main source of protein in a single adult. The microbiome was characterized using 16S rRNA and metagenomic sequencing. Individual community-level changes in the microbiome were compared, as well as the presence of bacteria associated with the fermented fish. We focused on Shannon alpha and UniFrac beta diversity. We did not detect any global changes in the gut microbiome in response to Surströmming, nor were we able to recover and identify any members of Halanaerobium, which were associated with and abundant in the ingested fish, in the stool samples of the participants. Our results suggest that Surströmming consumption does not alter the microbiome of healthy individuals. However, beneficial effects on a diseased gut, impaired gut microbiome, or other effects in disease remain to be studied.

Measurement Uncertainty Impacts Diagnosis of Diabetes Mellitus: Reliable Minimal Difference of Plasma Glucose Results.

INTRODUCTION: The diagnosis of diabetes mellitus is based on suitable cut-off values of specific biomarkers, such as the concentration of glucose in plasma. The German Diabetes Association has very recently published a clinical practice guideline on the definition, classification and diagnosis of diabetes mellitus that recommends measurements of plasma glucose concentration have an imprecision defined as a minimal difference (MD) of at a fasting plasma glucose concentration of 7.0 mmol/L. To obtain reliable values for the MD, we investigated long-term and short-term measurement uncertainty. METHODS: The imprecision was determined by two approaches: (1) a long-term dataset with imprecision based on the Guideline of the German Medical Association on Quality Assurance in Medical Laboratory Examinations (Rili-BAEK), in a medical laboratory operating 24/7, using internal quality control (IQC) data for four concentrations during a 10-year period; and (2) a detailed short-term dataset with imprecision assessed by hourly measurements of control materials. These datasets were used to calculate the MD cut-off (MDcut-off) as: [Formula: see text] = 2  [Formula: see text], where SD is the standard deviation and k = 2 represents a confidence level of 95%. RESULTS: The MDcut-off of ≤ 0.7 mmol/L at a fasting plasma glucose concentration of 7.0 mmol/L (MDcut-off 7.0) for the long-term and the short-term approaches were 0.44 and 0.40 mmol/L, respectively. The MDcut-off 7.0 from both approaches was therefore below the recommended value of 0.7 mmol/L. It was noted that the variability in performance within and between instruments can be covered by reporting the long-term MDcut-off 7.0 across all connected instruments. In this study, stable results for the MDcut-off 7.0 were obtained after 1 year. CONCLUSION: Imprecision as measured by IQC data is remarkably stable over many years of operation. Current imprecision assessment usually focuses on only single instruments, whereas clinicians perceive the measurement as the result of the combined analytical performance of all instruments used for a certain assay. In the clinical setting, the MD may be a more useful measure of imprecision, and we suggest deriving the MDcut-off combined from all instruments and control cycles that are used in the patient care setting for a given analyte.

Does eGFR improve the diagnostic capability of S-Creatinine concentration results? A retrospective population based study.

The use of MDRD-eGFR to diagnose Chronic Kidney Disease (CKD) is based on the assumption that the algorithm will minimize the influence of age, gender and ethnicity that is observed in S-Creatinine concentration and thus allow a single cut-off at which further diagnostic and therapeutic actions should be considered. This hypothesis is tested in a retrospective analysis of outpatients (N=93,404) and hospitalised (N=35,572) patients in UK and Sweden, respectively. An algorithm based on the same model as the MDRD-eGFR algorithm was derived from simultaneously measured S-Creatinine concentrations and Iohexol GFR in a subset of 565 patients. The combined uncertainty of using this algorithm was estimated to about 15 % which is about three times that of the S-Creatinine concentration results. The diagnostic performance of S-Creatinine concentration was evaluated using the Iohexol clearance as the reference procedure. It was shown that the diagnostic capacity of MDRD-eGFR, as it stands, has no added value compared to S-Creatinine. The gender and age differences of the S-Creatinine concentrations in the dataset persist after applying the MDRD-eGFR algorithm. Thus, a general use of the MDRD-eGFR does not seem justified. Furthermore the claim that the eGFR is adjusted for body area is misleading; the algorithm does not include any body size marker. It is thus a dangerous marker for guiding drug administration.

Bayes' theorem, the ROC diagram and reference values: Definition and use in clinical diagnosis.

Medicine is diagnosis, treatment and care. To diagnose is to consider the probability of the cause of discomfort experienced by the patient. The physician may face many options and all decisions are liable to uncertainty to some extent. The rational action is to perform selected tests and thereby increase the pre-test probability to reach a superior post-test probability of a particular option. To draw the right conclusions from a test, certain background information about the performance of the test is necessary. We set up a partially artificial dataset with measured results obtained from the laboratory information system and simulated diagnosis attached. The dataset is used to explore the use of contingency tables with a unique graphic design and software to establish and compare ROC graphs. The loss of information in the ROC curve is compensated by a cumulative data analysis (CDA) plot linked to a display of the efficiency and predictive values. A standard for the contingency table is suggested and the use of dynamic reference intervals discussed.