Evaluation of automated digital cell morphology system for the detection of platelet clumps / 中华检验医学杂志

ABSTRACT

Objective:To evaluate the performance of the automated digital cell morphology instrument in detecting platelet (PLT) clumps.Methods:A total of 4271 blood samples whose PLT reached the reviewing rules of thrombocytopenia were selected from inpatients having blood analysis in Xijing Hospital from January 1 st to June 30 th, 2019, including 2 200 males and 2 071 females,with a median age of (35±7.03) years old. The smears for these cases were made, stained by Wright-Giemsa, and examined to capture PLT clumps by digital cell morphology system and manual microscope separately. The digital cell analysis system (hereinafter referred to as the instrument method) as an evaluation method and the microscope method as a reference method were used to calculate the positive rate of platelet clump detection and evaluate the comparison of two methods and bias assessments. The chi-square test was used to compare counting data rates. Results:Among 4, 271 samples reaching the reviewing rule of thrombocytopenia, 128 cases with platelet clumps were detected by manual microscope(initial) with a positive detection rate of 96.24%, and a total 133 of cases with PLT clumps were detected by microscope (initial+reconfirmation) with a positive detection rate of 100 %. Meanwhile, 129 cases with platelet clumps were detected by instrument method with a positive detection rate of 96.9%. There was no significant difference in terms of positive rate of PLT clumps detection between the instrumental method and the microscope method (initial) ( χ2 =0.115, P=0.73); the positive rate of clumps detection by the instrumental method was lower than microscope method (initial+reconfirmation), and the difference was statistically significant (χ 2 =4.061, P=0.04). For instrument method, the positive rate of PLT clumps detection by simultaneous observation of RBC analysis interface+PLT aggregation interface+WBC analysis interface was higher than only observation of PLT aggregation interface, and the difference was statistically significant (χ 2 =5.090, P=0.02). The average error of the deviation of PLT counting results before and after correction of the cases with PLT plumps missed by instrument method was significantly higher than microscope method (initial), and the difference was statistically significant (χ 2 =56.26, P<0.001). Conclusion:The automated digital cell morphology system has a good consistency with manual microscope(initial) in terms of the sensitivity of platelet clumps detection and can be used as a supplementary method for detecting platelet aggregation.