Comparisons of results between three in-house biochemistry analysers and a commercial laboratory analyser for feline plasma using multiple quality specifications

Abstract

In-house plasma biochemistry analysis is commonplace in veterinary practice but there are few studies assessing the results from such instruments. Method comparison in human clinical pathology uses results from both precision and bias studies to determine whether results fall within a margin of “acceptable” or “allowable” total error. This study aimed to assess the clinical acceptability of plasma biochemistry results determined by in-house analyzers compared to a commercial laboratory analyzer. Clinical acceptability of analytes from in-house biochemistry analyzers was assessed by determining the percentage of results from feline plasma samples within coverage ranges determined from (a) biological variation, (b) those published by the American Society for Veterinary Clinical Pathology (ASVCP), as well as (c) calculated total error and (d) expanded measurement uncertainty. Calculated total error and expanded measurement uncertainty ranges were calculated from prior quality control material analyses. Between 80 to 87 % of results fell within minimal quality standards determined from biological variation. Approximately 90 % of results fell within “acceptable total error” ranges published by ASVCP. Overall, in-house analyzers provide acceptable results, with few clinical decisions affected by results determined from any of the three in-house analyzers compared to the commercial laboratory results. Keywords: Acceptable total error, Biochemistry analyzer, Biological variation, Clinical chemistry, Diagnostic test evaluation, Method comparison