OBJECTIVE: The objective of this study was to determine the accuracy of a near-patient blood gas and electrolyte analyzer when used by non-laboratory-trained clinicians in the critical care setting. METHODS: One hundred eighty-five blood samples (split samples) from 50 intensive care unit patients were analyzed by clinicians in the critical care environment using a near-patient blood gas and electrolyte analyzer (GEM Premier, Mallinckrodt Sensor Systems, Ann Arbor, MI). Near-patient measurements were compared with those obtained by laboratory technologists in an established intensive care unit laboratory. RESULTS: There was good agreement between the near-patient analyzer and the laboratory for pH, PCO2, sodium, potassium, ionized calcium, and hematocrit. Bias and precision were 0.006 and 0.03 for pH, 0.03 and 0.34 kPa for PCO2, 0.78 and 2.61 mmol/L for sodium, -0.11 and 0.12 mmol/L for potassium, -0.007 and 0.05 mmol/L for ionized calcium, and -0.99 and 1.33% for hematocrit. Bias between the laboratory instrument and the bedside analyzer was small for PO2 (-0.56 kPa). However, precision between instruments was significantly higher (2.39 kPa for all PO2 values and 1.61 kPa for PO2 < or = 13 kPa). CONCLUSIONS: The test instrument is accurate and reproducible when used in the clinical setting by non-laboratory-trained individuals. Non-laboratory-trained individuals can obtain laboratory results in the near-patient setting comparable to those obtained by trained laboratory technologists.
OBJECTIVE: The objective of this study was to determine the accuracy of a near-patient blood gas and electrolyte analyzer when used by non-laboratory-trained clinicians in the critical care setting. METHODS: One hundred eighty-five blood samples (split samples) from 50 intensive care unit patients were analyzed by clinicians in the critical care environment using a near-patient blood gas and electrolyte analyzer (GEM Premier, Mallinckrodt Sensor Systems, Ann Arbor, MI). Near-patient measurements were compared with those obtained by laboratory technologists in an established intensive care unit laboratory. RESULTS: There was good agreement between the near-patient analyzer and the laboratory for pH, PCO2, sodium, potassium, ionizedcalcium, and hematocrit. Bias and precision were 0.006 and 0.03 for pH, 0.03 and 0.34 kPa for PCO2, 0.78 and 2.61 mmol/L for sodium, -0.11 and 0.12 mmol/L for potassium, -0.007 and 0.05 mmol/L for ionizedcalcium, and -0.99 and 1.33% for hematocrit. Bias between the laboratory instrument and the bedside analyzer was small for PO2 (-0.56 kPa). However, precision between instruments was significantly higher (2.39 kPa for all PO2 values and 1.61 kPa for PO2 < or = 13 kPa). CONCLUSIONS: The test instrument is accurate and reproducible when used in the clinical setting by non-laboratory-trained individuals. Non-laboratory-trained individuals can obtain laboratory results in the near-patient setting comparable to those obtained by trained laboratory technologists.
Authors: G P Zaloga; T R Hill; R A Strickland; D Kennedy; M Visser; K Ford; J Whitley; G Holt; C Booker Journal: Crit Care Med Date: 1989-09 Impact factor: 7.598