الفهرس 
Automated Hematology AnalyzersOnly 14 pages are availabe for public view
 |  | from | 32 |  |  |
| | | from | 32 | | |
Abstract
Complete blood count with leukocyte differential count is one of the first quantitative methods for blood testing, and for some time has been the most widely used of tests in clinical settings. Therefore, validation of results from a hematology analyzer has become an item of special importance that reflects the quality of laboratory work and directly influencing patient’s clinical conduct. To increase the quality of complete blood count results provided by the laboratory, it is necessary to assess the analytical performance of hematology analyzers so as to ensure that the obtained results are the most reliable possible.
In July 1981, Clinical and Laboratory Standards Institute (CLSI) developed a proposed standard for evaluation of automated and semiautomated hematology instruments for their capability to perform an acceptable leukocyte differential count (H20-P). In March 1992, CLSI published the first edition of the approved standard (H20-A). In January 2007 the second edition of the standard (H20-A2) was published.
The standard focuses on white blood cells found in blood films and presents a detailed description of an acceptable manual-visual leukocyte differential count which serves as the reference for the instrumental differential counter. The standard outlines the types of abnormalities for inclusion as well as a statistical method for determination of the performance of the test method in both qualitative and quantitative abnormalities.
Coulter LH 750 hematology analyzer is a fully automated hematology analyzer designed to improve workflow efficiency for high-volume laboratories through elimination of preanalytical sample preparation and sorting as well as reduction of postanalytical manual confirmation of results. Coulter® LH 750 hematology analyzer is equipped with the most advanced technologies available to improve overall laboratory efficiency, while maintaining the highest degree of testing accuracy.
The aim of the present work is to evaluate the leukocyte differential count of Coulter LH 750 in hematology laboratory in Ain Shams University Hospital as regard accuracy and precision. Following CLSI H20-A2 approved standard, manual leukocyte differential count was used as the reference method for evaluation of the instrument analytical performance and testing the manufacturer claims.
The study samples were routine inpatient and outpatient samples from all hospital departments except for Hematology Oncology Department which has a dedicated laboratory.
A total of 400 samples were selected randomly from daily work load over a period of 3 months including 200 samples with normal leukocyte differential count and 200 samples with abnormal leukocyte differential count. Automated complete blood counts with leukocyte differential counts were performed using Coulter LH 750 hematology analyzer. Two blood films were made for each sample and stained by Leishman stain followed by 200 cells leukocyte differential count per slide each by a different examiner or on two occasions by one examiner. This counting was done blindly (without knowing the results generated by the analyzer).
Cells counted manually were categorized into 5 main categories: neutrophils, lymphocytes, monocytes, eosinophils and basophils. Immature granulocytes and variant lymphocytes were also counted. Immature granulocytes were added to the neutrophils category while the variant lymphocytes were added to the lymphocytes category.
The mean of the manual leukocyte differential count was calculated for each cell type in each sample. The total mean of all samples was calculated for each test group; normal, abnormal and total samples. Deming regression analysis, Pearson’s coefficient of correlation and comparison of the mean difference for each cell type of each sample with the manufacturer specification were done.
Precision testing was done at two levels; within-run precision and between-run precision. Within-run precision was done by running a sample 20 times in a single run. Between run precision was done by running quality control material each day for 20 successive days. The mean, coefficient of variation and the 2 SD were calculated for each cell type.
Correlation study revealed highly significant positive correlation between the Coulter® LH 750 results and the reference manual differential results for neutrophils, lymphocytes, monocytes and eosinophils count in the abnormal and total test groups but not in the normal test group in which correlation was suboptimal. As regard basophils, correlation study revealed a significant difference between the two methods for which this type of statistical analysis has no statistical validity.
Comparison of the mean difference for each cell type with the manufacturer specifications for accuracy revealed that the manufacturer specifications were met for all cell types in the three test groups.
Within-run precision testing revealed imprecise monocytes count by the instrument when compared with the manufacturer specification or with ¼th total allowable error. Neutrophils coefficient of variation was well within the ¼th total allowable error, however it didn’t meet the manufacturer specification.
There are no specification for between-run precision published by the manufacturer, however comparing the total analytical variation for each cell type with 1/3rd total allowable error revealed imprecise monocytes, eosinophils and basophils count. The total analytical variation of other cell types was within 1/3rd total allowable error. However they were clinically accepted.