RESUMO
Validated reference values and procedures are needed to ensure optimal diagnosis of dyslipidemia in sub-Saharan Africa. We aimed to validate an analysis method and establish reference intervals of lipid profile parameters in Cameroonians using this method. On a cross-sectional study conducted from November 2019 to August 2020 in Yaoundé, we have analyzed blood samples with Cobas® 6000. We subscribed to ASQUALAB's External Quality Assessments (EQA) and Outsourced Internal Quality Controls (IQC). Reproducibility, repeatability, correctness accuracy and uncertainty were evaluated using IQC. Consenting adult participants were conveniently sampled, excluding those with any condition that may affect lipid profile. Descriptive statistics were reported accordingly, agreement was assessed with Bland-Altman analysis, and reference intervals were defined according to CLSI and IFCC recommendations. The coefficients of variation for repeatability, reproducibility, and correctness bias ranged between 0.6% and 6%, with all values within the normal range. Expanded uncertainty of total cholesterol, HDL and triglycerides measurements were, respectively, 0.45, 0.24 and 0.18. We included 422 participants with a mean age of 30.2 (10.9) years and 248 (58.8%) females. Reference intervals for total cholesterol, HDL, triglycerides and LDL were, respectively, 2.94-6.02 mmol/L, 0.90-2.06 mmol/L, 0.35-1.36 mmol/L, 1.37-4.13 mmol/L. These intervals were similar between sex and ethnic groups, but lower in younger participants. Lipid profile measurement with Cobas® 6000 is a reliable and accurate analysis in our context. Specific reference intervals must be used in African population, with further studies need for different age subgroups.