Quality Control Strategy for Automated CBC: A Laboratory Point of View Deducted from an Internal Study Organised in an Emergency Laboratory

Abstract

Abstract Introduction: The aim of this study was to determine the performance of the total testing process of complete blood count (CBC) on two different instruments in an emergency setting of a county hospital, and to design an appropriate internal quality control plan. Materials and method: Two models of Statistical Quality Control (SQC) were evaluated on Sysmex XT-1800i and Cell-Dyne Ruby: 3 levels of commercial blood every 8 hours (N=9) and an alternative model using 3 levels every 12 hours (N=6) as shift changes. Total Error (TE) was calculated using the formula: TE=Bias%+1.65xCV%; Sigma score was calculated using the formula: Sigma=[(TEa%–Bias%]/CV%. Values for coefficient of variation (CV%) and standard deviation (SD) were obtained from laboratory data and Bias% from proficiency testing. For the pre-analytical phase Sigma score was calculated, while for post-analytical phase the turnaround time (TAT) was assessed. Results: TE for all directly measured parameters, for both instruments, had lower values than Total Error allowable (TEa). CV% for almost all parameters had lower values than CV% derived from biological variation except for platelets (PLT) at low level on Sysmex XT-1800i and red blood cells (RBC) on Cell-Dyne Ruby. Sigma score ranged from as low as 2 to 10. Sigma score for pre-analytical phase was 4.2 and turnaround time was 36 minutes on average. Conclusions: Given the performances of the total testing process implemented for CBC in our laboratory, performing the internal control after every 50 samples/batch seems to fulfill both the Health Ministry Order (HMO) 1301/2007 and International Organization for Standardization ISO 15189:2013 recommendation. All quality instruments must work together to assure better patient results and every laboratory should design its own control plan that is appropriate for better quality achievement.