Objective: This study aimed to evaluate the agreement between albumin measurements obtained via capillary zone electrophoresis (CZE) and bromocresol green (BCG) methods and to determine their interchangeability in clinical practice. Additionally, it explored how these results vary across different protein patterns.

Materials and Methods: Albumin concentrations in 373 patients were measured using the CZE and BCG methods. Subgroup analyses were conducted based on albumin, total protein, IgG, gamma band percentages, and M protein concentrations. Agreement was assessed using Bland-Altman analysis, while constant, proportional, and random errors were analyzed with Passing-Bablok regression. Clinical significance was determined by comparing total allowable error (TEa) with total analytical error (TAE).

Results: CZE method yielded albumin levels that were, on average, 0.28 g/dL lower than those measured by BCG. Greater bias (12.5%) was observed at low albumin levels. Regression analysis revealed constant, proportional, and random errors. Subgroup analyses indicated that TAE exceeded the TEa threshold.

Conclusion: Significant analytical errors exist between albumin concentrations measured by CZE and BCG, making these methods non-interchangeable. Detailed subgroup analyses revealed that this discrepancy consistently persisted across different protein patterns. These findings emphasize the importance of cautious use of the CZE method in clinical decision-making.