Zuurstoftoediening bij septische patiënten op de Spoedeisende hulp : Universitair Medisch Centrum Groningen

Background: Sepsis is a disruption of the inflammatory response, which can vary from mild to severe. The treatment consists, among other things, of the supply of sufficient oxygen, while there is a decrease in tissue oxygenation. Literature shows however, that hyperoxia (a too high oxygen tension in arterial blood) is possibly detrimental because free oxygen radicals are formed. Maybe due to this, there is an association between hyperoxia and a higher rate of neurological and pulmonary complications and mortality. This has all been studied in patients after cardiac arrest who were admitted on the Intensive Care Unit. The purpose of this study is to determine how many and which patients with sepsis are saturated sufficiently with less oxygen, so that the possible harm of hyperoxia can be prevented in them and secondly, to determine if hyperoxia may be associated with a higher rate of mortality in sepsis. Methods: Patients who came into the Emergency Room of the UMCG and who had 2 or more SIRS-criteria plus the suspicion of an infection were included prospectively over a period of 3 months. They received oxygen by a Ventimask 40% (FiO2 = 40%), after which arterial blood gas analyses took place. If the results showed a hypoxia, they then received a non-rebreathing mask (NRM) according to the current sepsis protocol (FiO2 = 65-80%). These 2 groups were compared with each other and the mortality was noted afterwards. Hyperoxia was defined as PaO2 > 13,5 mmHg. Results: There are 83 patients included totally. The average age was 60,7 (± 16,7 SD) years and 55 (66,7%) were male. Of all the patients with sepsis, there were 77 (92,8%) sufficiently saturated with a lot less oxygen than the current protocol prescribes. These were significant more often patients with a mild form of sepsis. The 6 patients in the NRM-group had significantly more often one of the 2 more severe forms of sepsis, had more often a positive history of pulmonary conditions and had initially a lower saturation and EMV-score. There were 8 patients who died altogether, of which 6 had hyperoxia. Hyperoxia however, was not significantly associated with a higher mortality. Conclusion: More than 90% of all sepsis patients are saturated sufficiently with less oxygen than current protocol prescribes. These are more often patients with a mild sepsis. For them, the protocol should be adapted, so that the possibly harmful effects of hyperoxia can be prevented. Whether hyperoxia is indeed detrimental in sepsis has never been studied and for now remains uncertain. In this study, hyperoxia showed no association with a higher rate of mortality.