Objective: The aim of this study is to investigate the compatibility of bicarbonate (HCO3-) levels measured by blood gas meter in serum autoanalyser.

Materials and Methods: In the study, serum bicarbonate and blood gas bicarbonate levels of 186 patients were evaluated retrospectively in the last one year period (September 2018 - 2019). Serum HCO3- levels were analyzed with Olympus AU5800 autoanalyser and blood HCO3- level with Radiometer ABL800. Serum bicarbonate results were divided into three groups as 0-60, 61-120 and> 120 minutes according to the time elapsed between sampling and delivery time.

Results: The mean difference between the two methods [blood HCO3- serum HCO3-]; for 0-60, 61-120 and 120 minutes were; 0.13 ± 7.6, 0.92 ± 7.6; 1.6 ± 5.4 mmol / L, respectively. According to the literature, the maximum allowable error obtained from intra-individual coefficients of variability was reported as 0.46 mmol / L. In this study, the mean difference between the two methods examined between 0-60 minutes was less than 0.46 mmol / L; The difference between the other two groups was greater than 0.46 mmol / L. In addition, % Bias, Total CV% and total analytical error (TAH) of levels serum bicarbonate were found as 8.52, 10.27 and 25.47 respectively.

Conclusion: If the serum HCO3- level was measured in the time of 0-60 min, we think that the two methods can be considered as alternatives. However, for samples that are waiting longer, evaluation of them as an alternative may cause errors. We also believe that the TAH value of the laboratory should be taken into consideration when evaluating serum bicarbonate level results, even if it is run under optimum conditions.