Comparison of Plasma Biochemistry Analyzers for Use in Sea Turtle Assessment
Plasma biochemistry results often vary based on the laboratory or equipment used to perform the analysis. Choosing a primary diagnostic laboratory or in-clinic analyzer is important to allow for consistency in comparing and interpreting results. When different analyzers are used, it is useful to know how results may vary among them for different analytes. Commercial diagnostic laboratories have advantages of quality assurance programs, technical support staff, and high-end equipment operating nearly continuously, but necessarily have a time delay for samples from remote facilities. In-clinic analyzers have the advantage of generating rapid results from freshly-collected samples, but generally lack the same level of quality assurance programs and depth of technical support staff as the commercial laboratories. They also require maintaining reagents in stock which may be less cost effective if not used frequently. The objective of this study was to compare results generated by two diagnostic laboratories, two in-clinic biochemistry analyzers, and one hand-held biochemistry analyzer with replicate samples of sea turtle blood.
Individual plasma samples collected from 22 sea turtles (18 loggerhead turtles, Caretta caretta; 3 green sea turtles, Chelonia mydas; and 1 Kemp's ridley, Lepidochelys kempi) were divided for analysis by the selected biochemistry analyzers. Not all analytes were assessed by each machine. The results from each analyzer were evaluated for differences in shared measured parameters. Furthermore, ease of use, reliability, and cost were subjectively evaluated. Precision and accuracy of each unit were not assessed during this study.
A Kruskal-Wallis test was used for nonparametric statistical comparison of the median plasma biochemistry parameters between the different analyzers. Statistically significant differences (p<0.05) were demonstrated for the following parameters: sodium, potassium, chloride, calcium, creatinine phosphokinase, albumin, alkaline phosphatase, and lactate dehydrogenase. Of these, differences in electrolytes and albumin could also be considered clinically significant. There were predictable trends in the way in which the analyzers differed from each other; however, the magnitude of difference for each analyte was variable. Results from the two diagnostic laboratories were in close agreement, while one of the in-clinic analyzers was most frequently an outlier. It is important to note that these differences don't necessarily mean one method is more accurate than another but may simply reflect the use of different substrates, different reaction conditions, or different analytic techniques. This means that reference intervals applicable to one method may not be directly applicable to another method. While references ranges should ideally be established that are specific to each instrument, an understanding of how results from different instruments vary can allow for more informed comparison between them.
Overall this study demonstrates the importance of reporting the type of instrument used and the analytic methods used by that instrument when making comparisons. There may not be one piece of equipment that is ideally suited for every lab or every purpose. The intended use of the equipment and needs of the facility should be considered carefully prior to making a selection.
The authors wish to thank the many volunteers at the Karen Beasley Sea Turtle Rescue and Rehabilitation Center for their support with animal care and handling during blood collection. Their dedication and assistance is greatly appreciated.