RESUMO
Objectives: The purpose of this study was to investigate the effects of contamination of venous blood with a lipid-containing solution on parameters measured by a modern blood gas analyzer. Methods: We collected venous blood from 17 healthcare workers (46 ± 11 years; 53â¯% women) into three blood gas syringes containing 0â¯, 5 and 10â¯% lipid-containing solution. Blood gas analysis was performed within 15â¯min from sample collection on GEM Premier 5000, while triglycerides and serum indices were assays on Roche COBAS C702. Results: Triglycerides concentration increased from 1.0 ± 0.3â¯mmol/L in the uncontaminated blood gas syringe, to 39.4 ± 7.8 and 65.3 ± 14.4â¯mmol/L (both p<0.001) in syringes with 5 and 10â¯% final lipid contamination. The lipemic and hemolysis indices increased accordingly. Statistically significant variation was noted for all analytes except hematocrit and COHb in the syringe with 5â¯% lipids, while only COHb did not vary in the syringe with 10â¯% lipids. Significant increases were observed from 5â¯% lipid contamination for pO2, SO2 and lactate, while the values of pH, pCO2, sodium, potassium, chloride, ionized calcium, glucose, hematocrit (10â¯% contamination), hemoglobin and MetHB decreased. All these changes except lactate and CoHb exceeded their relative performance specifications. Conclusions: Artifactual hyperlipidemia caused by contamination with exogenous lipids can have a clinically significant impact on blood gas analysis. Manufacturers of blood gas analyzers must be persuaded to develop new instruments equipped with serum indices.