Optimal Oxygenation in the Intensive Care Unit

Rationale:

Contrary to hypoxia, many physicians do not consider hyperoxia harmful for their patients. To prevent hypoxia, superfluous administration of oxygen is common practice, and hyperoxia is seen in many patients, especially on Intensive Care units. However, an increasing number of studies not only confirm the known negative pulmonary effects of chronic oxygen oversupply, but also important and more acute circulatory effects, characterised by decreased cardiac output (CO), increased systemic vascular resistance (SVR), and impaired microvascular perfusion. These phenomena can impair perfusion of organs, which may outweigh higher arterial oxygen content, resulting in a net loss of oxygen delivery and perturbed organ function. This may for example be responsible for hyperoxia-associated increased infarct size and increased mortality after myocardial infarction and cardiac arrest. The underlying mechanisms are not clarified yet, but probably involve increased oxidative stress with systemic vasoconstriction.

On the other hand, hyperoxia can also induce several favourable effects. The majority of ICU-patients have a systemic inflammatory response syndrome (SIRS) with concomitant vasoplegia due to trauma, sepsis or ischemia/reperfusion injury. Vasoconstriction could benefit these patients with severe SIRS, reducing the need for intravenous volume resuscitation and vasopressor requirements. Furthermore, hyperoxia may exert a preconditioning effect in patients with ischemia/reperfusion injury and prevent new infections due to its antibacterial properties.

Hypothesis:

Hyperoxia during SIRS ultimately has unfavourable effects on organ function, especially on a longer term.

Objectives:

• To study the short- and long-term effect of two different PaO2 targets on circulatory status, organ dysfunction and outcome.
• To study underlying mechanisms of hyperoxia by determining differences in oxidative stress response between the hyperoxic and the normoxic patients.

Study design:

Randomized, prospective multicentre clinical trial

Study population:

Patients admitted to the Intensive Care unit with ≥ 2 positive SIRS-criteria and an expected ICU stay of more than 48 hours

Intervention:

We will investigate 2 groups with PaO2 targets both within the range of current practice

Group 1: target PaO2 120 (105 - 135) mmHg (high-normal)

Group 2: target PaO2 75 (60 - 90) mmHg (low-normal)

Inclusion criteria

• ≥2 positive SIRS-criteria:

• Temperature >38 deg.C or hypothermia <36 deg.C
• Heart rate >90 bpm
• Respiratory rate >20 /min or pCO2 <32 mmHg (4.3 kPa)
• Number of leucocytes >12 x 10^9/l of <4 x 10^9/l of >10% bands
• Within 12 hours of admittance to the ICU
• Expected stay of more than 48 hours as estimated by the attending physician

Exclusion criteria

• Elective surgery
• Carbon monoxide poisoning
• Cyanide intoxication
• Methemoglobinemia
• Sickle cell anemia
• Severe pulmonary arterial hypertension (WHO class III or IV)
• Known severe Acute Respiratory Distress Syndrome (ARDS) (PaO2/FiO2 ≤100 mmHg and PEEP ≥ 5 cm H2O)
• Known cardiac right to left shunting
• Pregnancy
• Severe Chronic Obstructive Pulmonary Disease (COPD) (Gold class III or IV) or other severe chronic pulmonary disease
• Patients participating in other interventional trials