RESUMO
Albumin is a main biomarker of denutrition severity. Since 2019, the French National Authority of Health (HAS) recommends an immunologic technique as a reference for measuring albuminemia. In a context of partial recovery by the laboratory of GHBS (using colorimetric method with bromcresol purple) of the biological activity of the dialysis center, until then carried out by the Ouest biology - Biolor laboratory (using colorimetric method with bromcresol green) and in order to standardize practices in the healthcare sector, we compared 3 albumin assay techniques: immunoturbidimetric (Diagam®), BCG (Siemens®), BCP (Siemens®) on Siemens Atellica PLCs. The albumin level of 183 patient samples was measured. We show an overestimation of the dosage by the BCG (+ 5.46 g/L compared to IT) and an interference related to the inflammatory state of the patient during the dosage with this technique, as well as an underestimation of the dosage by the BCP (- 0.91 g/L relative to the IT), increased in patients with renal injury. This biological difference is added to a discrepancy in clinical conclusions. A change in dosing technique from BCG to BCP may impact the biological follow-up of patients and the PMSI score (mean difference of - 6.67 g/L between BCG and BCP while the classification of the severity of denutrition uses thresholds set in steps of 5 g/L). This variability of results should be taken into account by the clinicians and constitutes an argument in favor of the standardization of laboratories around a reference technique.