Proficiency testing materials for pH and blood gases. The California Thoracic Society experience

The California Thoracic Society Blood Gas Proficiency Testing Program distributed ampules from three separate lots of quality control products every three months as unknowns to participating clinical (survey) laboratories and ten selected reference laboratories. For eight quarters, aqueous buffers were distributed. For each lot, PCO2 and pH measurements varied within narrow ranges between laboratories. Concurrently, the PO2 measurements varied widely between reference laboratories as well as survey laboratories, but varied minimally when repeatedly assessed on each reference laboratory machine. Change to a fluorocarbon-containing emulsion as a testing medium resulted in a significant reduction in within model and overall variability for PO2. We attribute this reduction in variability to the higher O2 content and decreased temperature sensitivity for PO2 of the fluorocarbon-containing emulsion. Because we have no evidence that the magnitude of the interinstrument differences in PO2 found with these materials would be found with fresh human blood we recommend that regulatory agencies use the results of proficiency testing for PO2 cautiously.