INTRODUCTION: We compared the performance of a point-of-care measurement system to that of a conventional blood gas analyzer. METHODS: One hundred nineteen arterial blood gas samples from 2 hospitals were analyzed using the i-STAT system and Corning 278 concomitantly. Values were compared using the t lest for dependent samples and the Pearson product moment correlations (p < 0.05), and regression equations were calculated to determine the slope and intercept of the regression line for paired values of pH, P(aCO2), and P(aO2). Mean and standard deviations (SD) were also calculated for the difference between the paired results for each variable. RESULTS: Means and SD for the Corning 278 versus the i-STAT for pH were 7.40 (0.06) versus 7.41 (0.07), for P(aCO2) 43.0 (9.9) torr versus 43.4 (10.2) torr and for P(aO2) 107.9 (73.1) versus 109.7 (76.6) torr. Regression equations were: i-STAT pH = -0.2174 + 1.0301 x Corning pH; i-STAT P(aCO2) = -0.1724 + 1.0145 x Corning P(aCO2); i-STAT P(aO2) = -2.138 + 1.0373 x Corning P(aO2). Small, but statistically significant differences between the 2 instruments for pH (p = 0.015) and P(aCO2) (p = 0.017) were obtained but no significant difference for P(aO2) (p = 0.062). Significant correlations between values were obtained with the i-STAT when compared to the Corning 278 for pH (r = 0.9392; p < 0.001), P(aCO2) (r = 0.9804, p < 0.001), and P(aO2) (r = 0.9904, p < 0.001). Mean (SD) differences for the paired results between the 2 systems were -0.005 (0.02) for pH; -0.449 (2.02) torr for P(aCO2), and -138 (10.9) for P(aO2). CONCLUSIONS: Results obtained via the i-STAT system correlated well with results obtained with the Corning 278. Small, but statistically significant differences between the 2 instruments on pH and P(aCO2) are probably not clinically important.
|Original language||English (US)|
|Number of pages||5|
|State||Published - Jul 1 1997|
ASJC Scopus subject areas
- Pulmonary and Respiratory Medicine
- Critical Care and Intensive Care Medicine