• Christine Papadea, Joyce Foster, Sharon Grant, Sandra A Ballard, John C Cate, W Michael Southgate, and Dilip M Purohit.
• Department of Pathology, Medical University of South Carolina, Charleston 29425, USA. papadeacn@musc.edu
• Ann. Clin. Lab. Sci. 2002 Jan 1;32(3):231-43.

AbstractWe evaluated the analytical performance of the i-STAT Portable Clinical Analyzer (PCA), a point-of-care testing system consisting of a hand-held analyzer and single-use cartridges that measure different panels of electrolytes, metabolites, blood gases, and hematocrit in 65-100 microl of blood. Our objective was to determine whether PCA measurements at the bedside of patients in the neonatal and pediatric intensive care units of the MUSC Children's Hospital would be as reliable as those performed by the clinical laboratory's primary methods (Radiometer ABL 725 blood gas analyzer; Vitros 750 chemistry analyzer; and Coulter STKS hematology analyzer). Four cartridge types: (a) EC8+ (sodium; potassium; chloride; urea; glucose; pH; blood gases [PO2; pCO2]), (b) EC6+ (sodium; potassium; ionized calcium; glucose; hematocrit; pH), (c) G3+ (pH; PO2; pCO2), and (d) creatinine, were assessed for reproducibility, linearity, and method comparisons using aqueous samples, blood samples supplemented with several analytes, and -225 blood samples from patients. Reproducibility (CV) was good (< 2%) for electrolytes, glucose, urea, and pH, satisfactory (< 6.5%) for blood gases and creatinine, but poor (21%) for hematocrit. Linearity concentrations spanning the clinically relevant ranges were verified for all analytes. Method comparison studies with samples separated into 2 subgroups by patient age (> or < 3 mo) showed that agreement between the PCA and the primary methods was clinically acceptable. After the PCA was implemented for clinical testing, the observation of discrepant results of creatinine concentrations in neonatal blood samples that would have affected clinical management led to a second creatinine comparison study (59 additional samples) and to our eventual discontinuation of the PCA creatinine assay. This problem notwithstanding, the successful implementation of the PCA is attributed to careful analytical evaluations and ongoing communication with the clinical staff.

16 keywords...

hide…