Comparison of paired serum and lithium heparin plasma samples for the measurement of serum amyloid A in horses using an automated turbidimetric immunoassay.

This research focuses on determining if equine serum amyloid A (SAA) samples can be accurately measured in plasma using an automated turbidimetric immunoassay which was originally validated for the assessment of SAA in serum.

Objective of the Study

• The main goal of the study was to determine whether the turbidimetric immunoassay, that is traditionally used to measure SAA concentrations in serum, could be reliably utilized for the measurement of SAA in plasma as well. This was to increase flexibility in sample collection.

Methodology

• Paired serum and lithium-heparin samples were taken from a total of 40 horses.
• The SAA concentrations in both the serum and plasma were then measured and compared.
• Statistical analyses, including paired t tests and Passing-Bablok regression analyses, were used to compare the SAA concentrations between the serum and plasma samples.
• Bland-Altman plots were also used to identify any potential biases present in the data.

Key Findings

• The study found no significant difference in SAA concentrations between the serum and lithium-heparin plasma samples (P=0.48), indicating that the use of either sample type would yield similar results.
• Additionally, a high correlation factor (0.97) was determined between the paired samples, further supporting this conclusion.
• Despite the Bland-Altman plots revealing a positive bias in the plasma samples compared to the serum counterparts, the difference was not deemed clinically relevant.

Conclusions and Implications

• The research concludes that lithium-heparin plasma samples can effectively be used in place of serum samples for the measurement of SAA concentrations in horses.
• This offers increased flexibility when it comes to sample collection as both serum and plasma can now be used, albeit with a need for cautious interpretation when comparing data derived from the different sample types.