Conclusions
The immunoassay demonstrated robust performance in its verification and showed good comparability with the IEC but with some biases so caution is needed if both are used interchangeably. The immunoassay offers an automated alternative to IEC in the assessment of hyperhomocysteinaemia.
Methods
Following immunoassay method verification, plasma samples of 91 patients were analysed for tHCY using immunoassay and IEC.
Results
For the Abbott immunoassay, accuracy was assessed, with UK NEQAS EQA specimens, by the correlation of our Abbott immunoassay measurements to the Abbott ARCHITECT immunoassay mean (bias = 1.6%), and to the overall immunoassay mean (bias = 2.0%). The total imprecision was 2.7% (11.00 μmol/L), 2.4% (16.80 μmol/L) and 2.8% (24.30 μmol/L) respectively. Bias in linearity assessment was 0.12%-2.58%. The inter-method correlation was strong in Passing-Bablok regression: immunoassay = IEC x0.857 + 2.445 (95% CI: slope = [0.742,0.947], intercept = [1.340,3.582]), with Spearman correlation = 0.803 (p < 0.001). The Bland-Altman plot showed an average difference of -0.284 μmol/L (95% CI: [-1.043,0.474]) with limits of agreement (mean ± 1.96SD) from -7.425 μmol/L to 6.857 μmol/L.No significant difference in tHCY was found using both methods in patients with cerebrovascular diseases and cardiovascular diseases. Most tHCY measurements were within the reference ranges of both methods. All homocystinuria patients had tHCY values above the reference ranges of both methods. Conclusions: The immunoassay demonstrated robust performance in its verification and showed good comparability with the IEC but with some biases so caution is needed if both are used interchangeably. The immunoassay offers an automated alternative to IEC in the assessment of hyperhomocysteinaemia.