Interference in Ion-Selective Electrodes Due to Proteins and Lipids

The Stability of Analytes of Ionized Magnesium Concentration and Its Reference Range in Healthy Volunteers

This study aimed to determine the stability of refrigerated analytes of iMg concentration at different time intervals and to establish iMg reference range in a cohort of healthy Omani volunteers (≥18 years). The concentrations of iMg were measured using the direct ion-selective electrode technique. Pearson's and Lin's concordance correlation coefficients along with the Bland-Altman plot were used to assess the levels of agreement between iMg concentrations of fresh and refrigerated blood samples at different time intervals. The study included 167 volunteers (51% females) with a median age of 21 (range: 20-25) years. The median, 2.5th, and 97.5th percentiles for fresh iMg reference ranges were 0.55, 0.47, and 0.68 mmol/L, respectively. The overall agreement between the fresh and refrigerated iMg concentrations was poor (rho-c = 0.51; p < 0.001). However, according to Altman's definition, iMg concentrations of the refrigerated samples for a period of ≤1 h had an excellent correlation with the fresh iMg concentrations (Lin's rho-c = 0.80), with a small average bias difference of 0.009 (95%CI; -0.025-0.043). A cut-off refrigeration period within ≤1 h at 2-8 °C can be considered an alternate time frame for the gold standard measurement (fresh or within 0.5 h).

Paraproteins and electrolyte assays: exclusion effect and effect of paraprotein elimination

Paraproteins are a potential source of error for electrolyte analyses. The exclusion effect itself causes a discrepancy between direct and indirect ion selective electrode assays (dISE and iISE, respectively). We tested the applicability of different pretreatment methods and the difference of dISE and iISE with paraprotein-rich samples. We analysed chloride (Cl-), potassium (K+), and sodium (Na+) on 46 samples with paraproteins up to 73 g/L. We compared pretreatment methods of preheating, precipitation, and filtration to the native sample. All induced a statistically significant difference (p-value <0.05). Clinically significant difference was induced by precipitation for all analytes, and filtration for Cl- and Na+, but for none by preheating. The difference in electrolyte measurements with either dISE or iISE on native samples was explained by total protein concentration (TP). There was a statistically significant difference in all electrolyte measurements. On average, there was a clinically significant difference in Na + but not in Cl- and K + measurements. Paraprotein concentration (PP) or heavy chain class did not induce a statistically significant effect. The regression analysis and comparison to the theoretical exclusion effect supported the conclusion that TP is the only explanatory factor in the difference between dISE and iISE. We conclude that preheating is a suitable pretreatment method for all the studied analytes. Precipitation is not valid for any of them, and filtration can be considered only for K+. Because the difference between dISE and iISE was explained by the exclusion effect caused by TP, dISE is the more suitable method to analyse paraprotein-rich samples.