Original Articles
Ibrahim Sbai El Otmani, Anass
Abstract
The comparison between analytical techniques is often necessary in clinical chemistry, in particular, when a usual method is to be replaced by a new one. The aim of this work is to demonstrate the correlation between the potentiometric determination of potassium serum made on a multi-parameter automated analyzer and flame photometry considered as reference method. 50 serum samples were collected, identified and divided into two parts to be analyzed for potassium using the two methods: Indirect potentiometry using potassium ion selective electrode (ISE), performed on a DIRUI CS-1200 auto-chemistry analyzer and flame photometry performed on a JENWAY PFP7 flame photometer. Mean potassium values were found to be 3.92 mEq/L ; range 2.1-6.7 mEq/L with indirect ISE method and 4.21 mEq/L ; 2.35-7.79 mEq/L with flame photometry. A Passing-Bablok regression curve was plotted. The linear regression of matched pairs of values showed a very good correlation with R2 = 0.912. Bias plots representing the differences between the values of the two techniques against the means of both methods according to the propositions of Bland and Altman were also drawn. The comparison between the results obtained by flame photometry and indirect ISE methods showed a good agreement in potassium determination. This study showed that there is a very good correlation and agreement when serum potassium is determined by indirect potentiometry performed on the DIRUI CS-1200 auto-chemistry analyzer and by the flame photometry standard method.