Frequency of blood gas analysis among intensive care patients

The factors that influence critical care nursing staff when making the decision to sample blood for arterial blood gas analysis is the subject of a recent UK study. The authors of this study specifically sought to establish if there is an association between patient values for four parameters (FO₂(I) , pCO₂, pO₂ and % sO₂ ) and frequency of blood gas analysis. 

This was a retrospective study based on the arterial blood gas records of 65 ICU patients during the first 6 days of their stay in ICU. A positive correlation was found between number of samples taken per day from each patient and maximum pCO₂ value for the first 2 days of ICU admission. That particular correlation was not evident in subsequent days. 

There was a positive correlation evident on all 6 days between abnormal pO₂ and number of samples tested. In a discussion of their study the authors suggest that their results indicate that nurses use blood gases principally to monitor oxygenation rather than ventilation. 

Given that oxygenation can be successfully monitored non-invasively by pulse oximetry, the authors imply that unnecessary blood gases might be being performed, increasing the risk of iatrogenic anemia and the need for transfusion. In this particular study the mean number of blood gases sampled from each patient per day ranged from 5.7 to 9.4. 

The maximum number per patient per day ranged from 13 to 17. As the authors concede, in the absence of clinical details for each of the 65 patients it is not possible to determine whether this frequency of blood gas analysis is clinically justified.