Objective: To evaluate the analytical performance of the i-STAT(TM) system, which is designed for point of care testing and employs a hand-held chemistry analyzer and disposable cartridges, which in the configuration tested, are capable of measuring sodium, potassium, chloride, urea, glucose, and hematocrit in approximately 65 mu L of blood in 90 s. Methods: Linearity and imprecision in hematocrit measurement were assessed using whole blood, while that for the other analytes were evaluated with aqueous solutions. The accuracy of the i-STAT(TM) system was judged by assay of patient specimens obtained both by venipuncture and fingerprick and correlated with the Kodak Ektachem 700XR and the microhematocrit methods. Results: Linearity was obtained over the clinically relevant range for all analytes. Total imprecision as expressed by the coefficient of variation (CV) was less than 3.5% for all analytes at both high and low concentrations except for a low concentration of urea where a CV of 9.4% was obtained. Li near regression analysis revealed minimal systematic errors. The standard error about the regression line (S-y/x) ranged from 0.017 for hematocrit to 2.262 for chloride in the assay of venous blood, whereas in the assay of capillary blood the S-y/x ranged from 0.018 for hematocrit to 0.755 for glucose. Conclusions: The analytical performance of the i-STAT was deemed acceptable by calculation of total error and comparison with published performance standards. Our study has shown the i-STAT(TM) system to be reliable, robust, and simple to operate. Moreover, the compactness of the analyzer and the requirement for only small volumes of whole blood will make it a valuable diagnostic tool in point of care settings.
