Evaluation of a canine hemoglobin A1c point-of-care assay: analytical validation, clinical utility, and establishing a population-based reference interval
Denzel Adam, Nicholas Noto, Michael F Rosser, Sarah Connolly, Arnon GalAbstract
Background
Hemoglobin A1c (HbA1c) assay validation is essential for establishing analytical reliability, evaluating discrimination between diabetic and non-diabetic (NDM) dogs, and establishing reference intervals.
Hypothesis/Objectives
Validate a canine HbA1c point-of-care analyzer (Vet Chroma™), evaluate its diagnostic performance, and establish a reference interval in healthy dogs.
Animals
Fifty-one diabetic, 44 NDM controls, and 42 healthy client-owned dogs.
Methods
Prospective validation of precision, linearity, interference, and storage stability. Method comparison used an immunoturbidimetric comparator assay (Olympus DxC 700 AU). Diagnostic performance was evaluated via receiver operating characteristic (ROC) analysis, likelihood ratios, and odds ratios.
Results
Average intra- and inter-assay coefficients of variation were 3.92% and 8.47%. Linearity over the clinical range (0.86%-6.27% HbA1c%) was strong (R2 = 0.973) with minor constant bias. HbA1c remained stable for 5 days at 4 °C and up to 3 months at both -20 and -80 °C, but not reliably at 6 months. Lipemia and icterus exhibited variable interfering effects. Method comparison indicated non-interchangeability with the comparator assay. Diabetic dogs had significantly higher HbA1c than NDM (mean 5.10% vs 1.58%). ROC analysis yielded an AUC of 0.969, with a cutoff > 3.22% achieving 90.20% sensitivity and 100% specificity. The positive likelihood ratio was not estimable, and the odds ratio was 22.90 (95% CI, 3.60–146.90). Healthy young dogs’ reference interval was 1.25% (90% CI, 1.19–1.59) to 2.75% (90% CI, 2.56–2.90).
Conclusions and clinical importance
The Vet Chroma™ provides reliable HbA1c measurements in canine whole blood and effectively discriminates diabetic from non-diabetic dogs. Results should be interpreted alongside clinical and clinicopathologic findings.