High Imprecision Research Articles

Use of precision profiles to evaluate precision of the automated leukocyte differential

The commonly used methods of assessing the precision of the automated leukocyte differential have certain drawbacks that affect the validity and comparability of results. In the present report, we introduce a procedure based on building precision profiles from a large number of within-run imprecision experiments. The profiles are fitted to the function for the CV of proportions, which yields the number of theoretically differentiated leukocytes. Differences between fitted curves are evaluated for statistical significance by the F-test. As an example, we compared the precision of two hematology analyzers, a flow-cytometric technique involving fluorescence-labeled monoclonal antibodies, and the manual differential. We were able to establish definite differences in precision between different analyzers and different leukocyte classes. Our data also indicated that conventional within-run imprecision studies may completely misjudge analyzer precision. Furthermore, we could demonstrate that the precision of analyzers that analyze a fixed amount of blood rather than a fixed number of leukocytes is strongly influenced by the leukocyte count of the sample, leading to high imprecision for leukopenic samples. We believe the proposed procedure is a useful addition to currently used protocols; it yields clear results and creates a statistical basis of comparison between various instruments and techniques of differentiation.

Review of Quartz Analytical Methodologies: Present and Future Needs

Abstract A review of analytical methods for the qualitative and quantitative determination of crystalline silica is presented. The three prevalent forms of crystalline silica—quartz, cristobalite, and tridymite—are alluded to. Performance and popularity of present quartz analytical methods such as X-ray diffraction (XRD), infrared (IR) spectroscopy, and colorimetry are illustrated using Proficiency Analytical Testing (PAT) Program data. Although substantial improvements in performance have been made since the 1970s, quartz PAT sample results still continue to display high imprecision (>20% coefficient of variation). Past and present analytical methods are detailed in terms of theory and use, and strengths and weaknesses are discussed. Methods include gravimetric, chemical, microscopic, atomic absorption, XRD, IR, and colorimetry. Methods for determining bulk materials such as thermal analysis and nuclear magnetic resonance are briefly discussed. Results of a survey regarding international use of analytical methods for quartz are displayed. Both in the United States and internationally, the most popular methods are XRD and IR. Performance criteria such as detection limits, precision and accuracy, and potential future trends are shown using international data. Future needs for enhancing quartz analysis, such as a comprehensive evaluation of direct-on-filter techniques, an attempt at consensus for tridymite analysis, a consensus for defining analytical performance parameters for quartz in general, and a continued examination of methods for analyzing bulk materials for quartz, are suggested. Madsen, F.A.; Rose, M.C.; Cee, R.: Review of Quartz Analytical Methodologies: Present and Future Needs. Appl. Occup. Environ. Hyg. 10(12):991–1002; 1995.

Multicentre Evaluation of Commercial Kit Methods: Plasminogen Activator Inhibitor Activity

In order to study the analytical performance of different commercial kits for determination of plasminogen activator inhibitor (PAI) activity we distributed eight selected split samples to 11 European laboratories experienced with haemostasis testing. Three different laboratories were involved in the production of data from each of the commercial kits tested. A considerable variation of PAI activity results reported from the laboratories testing the same commercial kits was observed. The range of reported results could in individual samples exceed the median value indicating an interlaboratory variation of more than 100%. When we harmonized the results reported from different kits in different laboratories by means of an international standard from National Institute for Biological Standards and Control (NIBSC) we still observed that the results produced by some kits deviated systematically from results produced by other kits. Also, the harmonized results were used to estimate the overall coefficient of variation (CV) of PAI activity determined in various laboratories by different kits. We observed an inverse correlation between the PAI activity level and the CV with a CV of about 100% for low PAI activity levels and a CV of about 16% for high PAI activity levels. The high imprecision of the kits in the low concentration range of PAI activity indicates that unspecific factors in plasma may interfere with determination of active PAI. This was confirmed by the evaluation of the results from one of the plasma samples, which was PAI-1 depleted. The laboratories involved in the testing reported for this sample a mean value of 6.1 IU/ml.(ABSTRACT TRUNCATED AT 250 WORDS)

Detection of hypodense eosinophils by Percoll multilayer density gradient centrifugation in subjects with normal or slightly elevated eosinophilia poor reproducibility and eosinophils of density < 1.077 g/ml

In patients with marked hypereosinophilia ‘hypodense’ and ‘normodense’ eosinophils have been found after density gradient centrifugation. Subsequently this terminology has also been used in studies of patients with milder eosinophilia. However, in these cases the differentation between normo- and hypodense eosinophils was less clear. This might be due to the high imprecision of the test of density gradient centrifugation, as demonstrated in the first part of this study: the mean within-assay variance of the number of eosinophils in the different density layers was 35%. It was calculated that the test must be performed eight times to obtain an estimate of the true mean for the individual patient. In the second part of the study, the absolute number of ‘hypodense eosinophils’ in groups of patients with asthma (adults and children) and rheumatoid arthritis (adults) were compared to normal controls. Although a difference in the absolute number of hypodense eosinophils between groups of patients and controls could be demonstrated, the high imprecision of the test of density gradient centrifugation suggested that the technique used was not useful in an individual with normal or slightly elevated eosinophils in the peripheral blood.

A multi-factor experimental design for evaluating random-access analyzers.

A multi-factor experimental design for evaluating random-access analyzers has been developed and tested for the Ciba Corning "550 Express" random-access analyzer. The 12-sample design estimates imprecision, slope, nonlinearity, linear drift, and reagent carryover to the next assay. The design was constructed so that estimates of the factors' effects are almost entirely uninfluenced by each other. Use of the design is illustrated by an example in which reagent-to-assay carryover was pinpointed as an apparent cause of high imprecision. This led to a modification of the analyzer such that carryover was insignificant. The appendix contains a 27-sample design that provides additional estimates. Software to perform such calculations is available on request.

Chemiluminescent immunosorbent assay of serum myoglobin based on the luminol reaction

A simple and sensitive method for the determination of serum myoglobin is described. Myoglobin was determined in the range of 10–500 μg/1 by the chemiluminescent luminol reaction after adsorption to anti-myoglobin IgG onto a solid phase. Only 50 μl of serum has to be used and the luminescent immunoassay (LIA) has a better sensitivity and a wider linear range than the conventional radioimmunoassay (RIA). However, in its present form LIA has a higher imprecision than RIA. In clinical specimens the correlation between the two methods was excellent.

Surveys of neonatal bilirubin. An evaluation.

Experience in surveying the neonatal bilirubin determination in the Netherlands is described. Initially it was organized as a local activity (the Rotterdam area), then later enlarged to a national scale (some 150 participants). A number of items is discussed i.e. the initial improvements, which stabilized at an interlaboratory imprecision of about 8%, and further changes that could improve this rather high imprecision.