Evaluation of a point-of-care testing platform against a laboratory HbA1c measurement method in a South African tertiary-level diabetes clinic

Background: Diabetes mellitus (DM) is a major global issue, with a prevalence of 11.1%. In Africa, South Africa is recognized as having one of the highest prevalences. Limited access to timely glycated haemoglobin (HbA1c) testing may delay treatment adjustments for diabetic patients. Point-of-care (POC) analysers provide faster results, potentially enhancing diabetes management. This study aimed to evaluate the performance of a POC device compared to a laboratory-based method, including the effect of potentially interfering haemoglobin (Hb) variants. Methods: HbA1c was measured in whole blood and capillary samples from 63 diabetic patients using the Finecare™ HbA1c POC device and compared to the Bio-Rad Variant II, a laboratory method utilising ion-exchange high-performance liquid chromatography (HPLC). 43 whole blood samples identified with Hb variants from routine laboratory HPLC analysis were also assessed on the POC device. Results: The POC device demonstrated a strong correlation with HPLC for both capillary (r = 0.94, 95% CI 0.90-0.96) and whole blood samples (r = 0.94, 95% CI 0.91-0.97). Mean bias was 1.3% (95% CI -1.2-3.8) for capillary vs. HPLC, and 4.5% (95% CI 2.1-6.9) for whole blood vs. HPLC. However, POC HbA1c in samples with Hb variants showed poor correlation with HPLC (r = 0.60, 95%CI 0.37-0.77) and a mean bias of -2.8% (95%CI -7.2-1.5). Conclusion: The Finecare™ HbA1c POC device performed acceptably for both capillary and whole blood samples, comparable to the HPLC method. We recommend capillary samples for routine monitoring but advise caution in populations with a high prevalence of Hb variants.