Objectives: (1) To compare the agreement between two HbA1c measurement methods (boronate affinity chromatography and ion-exchange HPLC – IE-HPLC) in type 2 diabetic patients; (2) To determine the prevalence of hemoglobin (Hb) variants and associated clinical factors.
Methods: A cross-sectional study of 557 type 2 diabetic patients at Nguyen Tri Phuong Hospital from April to October 2025. Each sample was simultaneously assayed by affinity chromatography (A1c Hb9210 Premier, Trinity) and IE-HPLC (Adams A1c HA8190V, Arkray). Hb variants were suggested by abnormal IE-HPLC chromatograms; confirmation by hemoglobin electrophoresis or genotyping is required for definitive identification. Bland-Altman analysis and Lin's Concordance Correlation Coefficient (CCC) assessed method agreement. Multivariable logistic regression identified factors associated with Hb variants.
Results: Mean age was 60.6 ± 10.5 years; 61.8% female. Affinity chromatography HbA1c (7.98 ± 2.09%) was significantly higher than IE-HPLC (7.79 ± 1.97%; difference: 0.193 ± 0.265%, p < 0.001), with Pearson r = 0.9934 and Lin's CCC = 0.9870. The 95% limits of agreement (LoA) were −0.326% to 0.712%. Hb variant prevalence was 2.5% (95% CI: 1.4%–4.2%); the LoA was notably wider in this subgroup (−0.612% to 0.993%). Univariable analysis showed higher glucose and HDL-c in the variant group; in the multivariable model (exploratory, EPV = 2), red blood cell count (RBC) was the only independent factor (OR = 3.272; 95% CI: 1.439–7.444; p = 0.005).
Conclusions: Both methods showed good overall agreement with a low systematic bias; IE-HPLC additionally enabled Hb variant screening. Elevated RBC may be a suggestive sign of Hb variant presence, requiring confirmation in studies with larger sample sizes. Abnormal IE-HPLC chromatograms should be flagged and communicated to clinicians to support accurate HbA1c interpretation.