A STUDY TO ESTABLISH THE AGREEMENT BETWEEN THE RESULTS OF ELECTROLYTES (SERUM SODIUM AND POTASSIUM) ESTIMATED BY A WET CHEMISTRY INSTRUMENT (EASYLYTE) WITH THAT OBTAINED BY A DRY CHEMISTRY ANALYZER (VITROS 350)

Abstract

Introduction

 The dry chemistry microslide technology for estimation of electrolytes has totally eliminated the need of water and is especially useful for paediatric patients as only a small volume of the sample is sufficient to obtain results. The authors intended to compare the results of estimation of electrolytes (namely sodium and potassium) obtained by the newly acquired dry reagent analyzer (VITROS 350) with those of the wet reagent analyzer (EASYLYTE ) which was already in use in the department.

Aims

       The aim of the study was to assess the agreement between results of electrolytes (serum sodium and potassium) estimated by a wet chemistry instrument with that obtained by a dry chemistry analyzer.

Methods and Materials

It was an observational analytical cross-sectional study done in the Departmental clinical laboratory. The samples were selected randomly from the usual lab workflow. All the samples were first run on the Easylyte (wet chemistry) and then run on the Vitros 350 (dry chemistry). The paired data thus obtained were compiled and tabulated and then statistically analysed.

 Results

       The agreement of the results between the two methods was evaluated using the Bland-Altman difference plot and the Passing-Bablok Regression Equation and the Deming regression studies. By analyzing the diagram of Bland-Altman, it is seen that for sodium, the average bias is of -2.22; limits of agreement being -26.12 to 21.77. For potassium, Bland Altman plots show a bias of -0.21; limits of agreement   -0.61 to 0.19.Passing Bablock regression calculated an intercept of -56.86, 95% CI (-100, -28) and Slope of 1.43 for sodium measurements and calculated an intercept of -0.706, 95% CI (-0.66,-0.45) and Slope of 1.2 for potassium estimation.

Conclusions:

Statistical analysis revealed conflicting solutions. There is a great discrepancy between the results of the electrolyte estimation by the two methods since the methodologies are not identical.