[Importance of pre-analytical variables in the quality of arterial blood gas determination]

Several frequent errors in the measurement of arterial blood gases, shed doubts on their real usefulness.

To identify sampling, manipulation and transport errors in the measurement of arterial blood gases.

Three hundred and nine consecutive arterial blood samples received at the central laboratory of a public hospital were analyzed. Patient data in the order form, reception conditions at the laboratory, transport media, time of arrival and analysis of each sample were recorded.

Five percent of orders informed the hour of sampling, 0.6% the patient's temperature and 18.1% the inspired oxygen fraction. Bubbles or clots were present in 12.9% and 3.2% of samples respectively, 87.3% of samples were well sealed and the amount of blood withdrawn in relation to the syringe capacity was optimal in 47.2% of cases. Ninety three percent of syringes were transported with ice cubes that did no cover the syringe and 5.8% of samples were received at room temperature. The delay in analysis, since the time of reception at the laboratory, ranged from 0 to 55 min (mean 12.9 min).

Several deficiencies in pre analytical variables in blood gas analysis were identified, most caused by neglect and susceptible of correction. Quality controls for this determination should be performed frequently.