Comparison of the radial immunodiffusion test and a novel lateral flow device test for measuring immunoglobulin concentrations in calf blood

Failure of transfer of passive immunity (FTPI), characterized by low serum IgG concentrations (< 10 g/L) in dairy calves, is associated with high calfhood morbidity and mortality, as well as poor productive and reproductive outcomes in later life. Calves can be blood sampled to test for FTPI between 24 h and 7 days of life. Direct measurement of IgG in calf serum using the radial immunodiffusion (RID) reference test is time-consuming, costly and requires a laboratory setting. Recently a new ‘pen-side’ test called ‘ImmunIGY’ was brought to the UK market to ‘directly measure IgG in calf whole blood’ using an immunochromatography-based lateral flow device (LFD). The objective of the current study was to compare RID and ‘ImmunIGY’ measurements. Measurements were compared using Pearson correlation, McNemar chi-squared, Bland-Altman plots, Passing-Bablok regression and receiver operating characteristics (ROC) curves. Pearson correlation was low (r = 0.35). McNemar chi-squared statistics revealed no statistically significant differences between the two tests at both IgG cut points of 10 g/L (0.25 (p = 0.61)) and 15 g/L IgG (0.15 (p = 0.69). The Passing-Bablok regression intercept value was 10.76 (95 %CI=-1.10–20.19) showing that the systematic difference between the methods was not significantly different from zero and the slope value was 0.81 (95 %CI = 0.53–1.09) so the hypothesis that the slope is equal to 1 was not rejected.

Bland-Altman plots showed wide limits of agreement −29.99–24.04 g/L, with a mean difference of −2.98 g/L (SD=13.51, 95 %CI −5.31- −0.65), with evidence of constant error. ROC curves showed area under the curve for both 10 g/L and 15 g/L cut points of 0.73 and 0.68 respectively. Overall, the test specificity of the ‘ImmunIGY’ test was superior to the sensitivity of the test which would in practice lead to a low number of false positive tests, but a relatively high number of false negatives, suggesting its appropriate use as a screening tool rather than for individual animal diagnostics.