THE OXYGEN STATUS ALGORITHM ONLINE WITH THE PH BLOOD-GAS ANALYZER

Background: The Oxygen Status Algorithm is a computer program for inte rpretation of pH-blood gas measurements. Recently the facility for on- line transfer of measured data from the pH-blood gas analyzer (ABL 520 ) to the computer with instant calculation, graphical display, and pri nt-out of the oxygen status and the acid-base status of the blood has been implemented. Objective: To describe experiences with routine appl ication of the Oxygen Status Algorithm in an intensive care unit acid to describe recent improvements of the program. Data sources: Routine pH-blood gas analyses during a two year period with an average of 40 a nalyses per day. Results: Several unexpected results were disclosed, e specially with respect to abnormal haemoglobin-oxygen affinity, causin g suspicion of low 2,3-diphosphoglycerate concentration and hypophosph atemia. The criteria for routine administration of oxygen could be rev ised on the basis of the oxygen extraction tension. The program was ex panded to allow the printing of a cumulated patient report, and modifi ed to allow calculation of the oxygen consumption rate on the basis of simultaneous measurements on the arterial and the mixed venous blood. The program and the computer hardware functioned well in continuous o peration during the whole study. It proved to be difficult, however, t o ensure that all measurements were supplemented by information on the patient temperature and the fraction of inspired oxygen. Conclusion: The Oxygen Status Algorithm provides a more detailed description of th e oxygen status and the acid-base status of the blood than currently e mployed. The interaction between the arterial oxygen tension, the haem oglobin oxygen capacity and the haemoglobin oxygen affinity is display ed graphically and the degree of compensation among these three proper ties is expressed in terms of a single quantity: the oxygen extraction tension. The program also provides an interactive interpretation, all owing a rapid prediction of the expected effect of therapeutic interve ntion. Nevertheless, implementation of the program in the daily routin e requires an intensive educational effort among physicians as well as nurses and laboratory technicians.