BACKGROUNDA generally accepted guideline ("41 rules") published by the International Consensus Group for Hematology Review (ICGHR) can not be suitable for all the laboratories because the facility type, laboratory requirements, sample volume, review rate, turn around time, instrument model and characters etc. are quite different from each other, which may cause a higher workload for microscopy review or lead to false or misleading results. Therefore, we decided to develop the personalized review criteria for 4 series of hematology analyzers in the same hospital, and describe all the implement procedures in detail.

METHODSThe total 1770 blood samples were collected from Peking Union Medical College Hospital. Referring to the suggested criteria by international consensus group for hematology review ("41 rules"), the personalized review criteria for 4 series of hematology analyzers including Siemens Advia 2120, Sysmex XE-2100, Sysmex XT-1800i and Sysmex XS-800i were established and validated by adjusting the rules in order to reduce the false positive rate and keep the false negative acceptable by clinical.

RESULTSUsing the "41 rules", high review rates of 37.94%, 35.56%, 33.44% and 37.94% were got respectively in Siemens Advia 2120, Sysmex XE-2100, Sysmex XT-1800i and Sysmex XS-800i. Three false positive rules mainly were observed in all of 4 analyzers: white blood cell < 3 × 10(9)/L or >30 × 10(9)/L, platelet < 100 × 10(9)/L or > 1000 × 10(9)/L and immature granulocyte. Specialized rules were observed in different series of analyzers, atypical/variant lymphs flag were found mainly in Sysmex XE-2100, Aniso-RBC were found mainly in Sysmex XT-1800i, flag of "immature granulocyte" mainly in Sysmex XS-800i, Micro-RBC, Macro-RBC and Aniso-RBC mainly in Siemens Advia 2120. Rules of immature granulocyte, blast, and NRBC flag would be mainly triggered by hematology malignant tumor. We could not delete these rules due to the risk of false negative of serious disease, other rules were deleted or revised. After continually optimizing to the rules, we finalized the criteria suitable for Siemens Advia 2120, Sysmex XE-2100, Sysmex XT-1800i and Sysmex XS-800i in our laboratory. The false negative rates were 2.94%, 2.86%, 3.10% and 2.78%, the review rates were 31.07%, 30.00%, 30.01% and 30.09%, and there was no hematology malignant tumor missed. Validated by 547 samples, the false negative rates of our optimized rules were 0.37%, 0.55%, 0.55%, and 0.91% respectively.

CONCLUSIONThe criteria can be based on the criteria established by International Consensus Group for Hematology Review but must be optimized according to the different requirements.