Canine complete blood counts: a comparison of four in-office instruments with the ADVIA 120 and manual differential

Elizabeth G Welles1, Amy S Hall, D Mark Carpenter

  • 1Department of Pathobiology, Auburn University, Auburn, AL 36849-5519, USA. welleeg@vetmed.auburn.edu

Insights

The Hemavet 950 and Heska CBC-Diff analyzers demonstrated the best accuracy for canine CBC results among the tested bench-top instruments. Leukocyte differential counts were generally unreliable across most in-office analyzers.

Area of Science:

  • Veterinary Hematology
  • Comparative Diagnostics
  • Laboratory Medicine

Background:

  • Increasing reliance on bench-top hematology analyzers in veterinary practice necessitates accurate results.
  • Varied methodologies in cell counting and differentiation by analyzers, with infrequent manual blood smear examination.

Purpose of the Study:

  • To compare canine complete blood count (CBC) results from four bench-top analyzers against a reference analyzer (ADVIA 120) and manual counts.
  • Evaluate the diagnostic accuracy and comparability of in-office hematology instruments.

Main Methods:

  • Analysis of 100 EDTA-anticoagulated canine blood samples on Hemavet 950, Heska CBC-Diff, IDEXX LaserCyte, and IDEXX VetAutoread.
  • Comparison with ADVIA 120 and manual 100-cell leukocyte differentials.
  • Statistical analysis including linear regression, difference plots, paired t-tests, and diagnostic equivalence.

Main Results:

  • High correlations for HCT, WBC, and platelet counts between in-office analyzers and ADVIA 120, but variable accuracy.
  • Hemavet 950 and Heska CBC-Diff showed the best comparability with ADVIA and manual differentials.
  • Leukocyte differentials from all in-office analyzers were largely unreliable, except for neutrophil counts.

Conclusions:

  • Hemavet 950 and Heska CBC-Diff offered the best performance among the evaluated bench-top analyzers.
  • Leukocyte differential counts from in-office analyzers were unreliable, offering no advantage over 3-part differentials.
  • Reticulocyte assessment on LaserCyte and VetAutoread was unreliable compared to the ADVIA 120.
Abstract