What is the clinical evidence from this study?

Study design: Observational. Population: Endocrine disorders (dysglycemia, prediabetes, diabetes) (n=554). Intervention: Point-of-care (POC) HbA1c testing (Afinion 2 analyzer) vs. Laboratory reference testing (HPLC, Tosoh). Primary outcome: Analytical agreement (proportion of POC values within ± 5.0% of reference method values).

What question did this study set out to answer?

To evaluate the accuracy and diagnostic potential of point-of-care HbA1c testing compared to laboratory reference methods.

May 11, 2026Open Access

Point-of-care HbA1c testing in a tertiary, university referral center: living up to the potential?

Key Result

Point-of-care HbA1c testing demonstrated insufficient analytical agreement compared to HPLC reference testing, with only 87.0% of values falling within ±5.0% of the reference method.

Key Points

To evaluate the accuracy and diagnostic potential of point-of-care HbA1c testing compared to laboratory reference methods.
Single-center observational study over 12 weeks.
Consecutive patients referred to outpatient clinics for HbA1c testing were included.
Blood samples were tested simultaneously using POC and laboratory methods.
87.0% of POC values were within ± 5.0% of reference method values.
58.7% of POC values were within ± 3.0% of reference values according to guidelines.
A medication error would have occurred in 6.2% of patients if POC was the sole testing method.

Study Design

Type

Observational (n=554)

Multicenter

Structured PICO

Does point-of-care HbA1c testing provide accurate diagnostic agreement compared to reference laboratory testing in endocrine outpatients?

Population

554 consecutive patients referred to the outpatient clinics of the Department of Endocrinology, Diabetes and Metabolism at the University Hospital Essen over 12 weeks (47.1% normoglycemic, 15.9% prediabetes, 37.0% diabetes).

Intervention

Point-of-care (POC) HbA1c testing using the Afinion 2 analyzer on venous blood samples

Comparator

Reference laboratory HbA1c testing (Tosoh HPLC) on simultaneously obtained venous blood samples

Outcome

Accuracy and diagnostic capacity of the POC HbA1c testing compared to reference laboratory testingsurrogate

Point-of-care HbA1c testing shows insufficient analytical agreement and clinically relevant misclassification compared to reference laboratory testing in a tertiary referral cohort.

Limitations

Patients with HbA1c values > 9.0% were underrepresented, limiting conclusions in severe hyperglycemia.
Absence of a formal a priori sample size calculation and restricted statistical power of subgroup analyses.
Single-center design may limit generalizability to primary care settings.
Carbamylation and hemoglobin variants were not systematically assessed.

Abstract

Background Point-of-care (POC) HbA1c analysis offers an easy and rapid approach of assessing an individual’s glycemic status. However, its accuracy and diagnostic potential remain controversial. Here we compared POC vs. reference laboratory HbA1c testing in selected endocrine patient cohorts at a tertiary, university referral center using the widely established Afinion 2 analyzer. Methods Single-center observational study of consecutive patients referred to the outpatient clinics of the Department of Endocrinology, Diabetes and Metabolism at the University Hospital Essen over 12 weeks. Venous blood samples for POC and laboratory HbA1c (Tosoh) testing were obtained simultaneously. Accuracy and diagnostic capacity of the POC HbA1c testing were evaluated. Results Among 554 patients 47.1% (n = 261) were normoglycemic, 15.9% (n = 88) had prediabetes and 37.0% (n = 205) diabetes. Comparison of POC and laboratory HbA1c testing showed that 87.0% of POC values fell within ± 5.0% of the corresponding reference method values. When applying the German Medical Association guidelines, 58.7% of the POC values fell within ± 3.0% of the corresponding reference method values. ROC-analysis demonstrated an optimal cut-off for diagnosis of prediabetes of 5.65% (38 mmol/mol) and for diabetes of 6.25% (45 mmol/mol). Considering POC HbA1c testing as only tool for treatment decision, a medication error would have occurred in 12/195 patients (6.2%). Conclusion Our study demonstrated, that POC HbA1c testing offers correlation but insufficient analytical agreement and clinically relevant misclassification compared to HPLC reference testing, highlighting that analytical validation in controlled environments does not automatically translate into diagnostic interchangeability in complex tertiary referral cohorts. Supplementary Information The online version contains supplementary material available at 10.1007/s12020-026-04615-6.

Bookmark

View Full Paper