The "diff-if". Use of microcomputer analysis to triage blood specimens for microscopic examination

Area of Science:

• Hematology
• Clinical Pathology
• Laboratory Medicine

Background:

• Automated blood counts often lack manual white blood cell differentials.
• The "diff-if" strategy uses automated results to decide on manual differentials.
• Comparing numeric criteria versus microcomputer analysis for this triage is crucial.

Purpose of the Study:

• To compare the sensitivity and specificity of numeric criteria and a microcomputer program for triaging blood counts for manual differentials.
• To evaluate these strategies across different patient populations.

Main Methods:

• Two sets of criteria were compared: established numeric values and a microcomputer program analysis.
• The strategies were tested on three distinct patient populations: high hematologic disorders, predominantly normal, and predominantly abnormal counts (excluding primary hematologic disorders).

Main Results:

• The microcomputer program was more sensitive than numeric criteria for detecting various abnormalities, including bands, immature granulocytes, and red blood cell fragments.
• Specificity was comparable between methods in a high-disorder population but varied based on the patient group.
• Numeric criteria were more sensitive for mild eosinophilia and resulted in fewer manual differentials ordered.

Conclusions:

• Microcomputer analysis demonstrates higher sensitivity than numeric criteria for identifying abnormal blood smears.
• The optimal triage strategy's specificity is patient-population dependent.
• Laboratories should select a triage strategy based on their specific patient demographics.