# Impact of haemoglobin variants on the diagnostic sensitivity of glycated haemoglobin (HbA1c) assay methodologies in sub-Saharan Africa: a laboratory-based method validation study

## TL;DR

This study shows that HbA1c testing methods work reliably for diabetes diagnosis in sub-Saharan Africa, even in people with common haemoglobin variants like HbAS.

## Contribution

Demonstrates that major HbA1c assay methods are accurate and comparable in populations with high HbAS prevalence.

## Key findings

- All HbA1c methods strongly correlated with continuous glucose monitoring (CGM) average glucose levels.
- Presence of HbAS did not affect the relationship between HbA1c and CGM glucose for any method.
- Diagnostic accuracy was similar across methods for key glucose thresholds.

## Abstract

the utility of glycated haemoglobin (HbA1c) for the diagnosis and monitoring of diabetes in sub-Saharan Africa is uncertain due to limited data on the performance of the available HbA1c assay methods in this population, which has a high prevalence of haemoglobin variants. We aimed to compare the diagnostic accuracy of the major HbA1c methodologies (Boronate Affinity, Capillary Electrophoresis, High Performance Liquid Chromatography, Immunoassay) in an African population, and assess the impact of the common haemoglobin variant HbAS (sickle cell trait).

whole blood samples were obtained from 182 individuals living with type 2 diabetes in Uganda. HbA1c values for each method were compared to average glucose measured over 14 days by continuous glucose monitoring (CGM). To determine concordance, the three HbA1c assay methods were compared to the capillary electrophoresis method.

there was a strong correlation between CGM average glucose levels and all four HbA1c methodologies (r=0.81-0.89) which did not differ in those with and without HbAS (present in 37/182 participants). The presence of HbAS did not alter the relationship between HbA1c and CGM glucose for any assay (p for interaction >0.2 for all methods). Diagnostic accuracy for CGM average glucose thresholds of 7 and 10mmol/L was similar across methods (area under the receiver operating characteristic curve 0.80-0.84 and 0.76-0.84 respectively). The maximum bias between the HbA1c assay methodologies was 2 mmol/mol (2.07%).

all major HbA1c technologies offer accurate and comparable HbA1c measurement even in this population with high prevalence of haemoglobin variants.