Clinical Evaluation and Cross-Validation of the Point-of-Care System iStat at the Emergency Department Versus Central Laboratory Analysis

Abstract

BACKGROUND: In the emergency department (ED), large numbers of blood samples are taken on a daily basis. Owing to their ability to produce rapid results, point-of-care (POC) analysis systems are applied increasingly in the ED. In the current study, we cross validated the quality of the POC iSTAT (Abbott point of care) against the central laboratory for the most common blood samples taken at the ED of a tertiary care university hospital. METHODS: Forty-one patients older than 18 years presenting to the ED owing to shortness of breath or chest pain were enrolled in the study. In all patients, the following tests were performed on venous blood: iSTAT Chem8 + (sodium, Na; potassium, K; chloride, Cl; ionized Ca; glucose; serum urea nitrogen, BUN; creatinine; hematocrit; hemoglobin), and CG4 + (venous blood gas [VBG] + lactate). Similar tests were performed parallel to the central laboratory of the hospital. Troponin testing was performed but omitted for clinical consideration owing to the routine use of high-sensitivity troponin by our central laboratory. Statistical analysis was performed using the Pearson correlation coefficient on each analyte. RESULTS: The results of the POC analysis were available at a mean ± SD of 48.5 minutes earlier than the central laboratory results. The Chem8 + results between iSTAT and central laboratory correlated to a high degree, with the least correlation relating to K (κ = 0.77), Cl (κ = 0.82), and ionized Ca (κ = 0.86). All other items ranged between κ = 0.9 (Na) and κ andgt; 0.95 (glucose, BUN, creatinine, hematocrit, and hemoglobin). For the CG4 +, pH revealed a κ = 0.87, pCO2 κ = 0.9, base excess κ = 0.9, HCO3 κ = 0.72, pO2 κ = 0.62, and SpO2 (saturation) κ = 0.77. Lactate analysis showed a κ = 0.94. The results from the central analysis testing showed lower pH, base excess, and HCO3 values than the POC, whereas the pCO2 and lactate values were higher in the central laboratory. The discrepant values might have resulted in inadequate treatment in the clinical context. When an arterial blood gas was simultaneously analyzed by POC and the central laboratory (20 samples), we found the following results: HCO3 κ = 0.9, SpO2 κ = 0.93, and all other κ andgt; 0.95. CONCLUSION: The electrolyte analyses performed by the POC-system iSTAT correlate to a very high degree with the analysis provided by our central laboratory. The POC iSTAT system might be considered as a potential POC device for use in the ED. Furthermore, our findings suggest that POC might serve as independent control system for the clinician to find potential problems in the process of blood sample analysis.