Effects of Non-invasive Ventilation (NIV) on Cerebral Oxygenation.

Near infrared spectroscopy (NIRS), is a non-invasive method for the measurement of blood flow in tissues, first used for cerebral tissue oxygenation in 1977. NIRS is a spectroscopic technique, which uses electromagnetic waves (700-950nm), an emitter and a detector. In the last 20 years there was an enormous development in the instrumentation and application of NIRS. This technique now allows to measure the oxygenation of the brain tissue. The INVOS Brain oximeter is a reliable trend monitor for changes in regional cerebral oxygenation (rSO2) and correlates with the hemoglobin saturation in venous, capillary and arterial blood, using an algorithm based upon the Beer-Lambert law. The INVOS Cerebral oximeter system uses light, with wavelengths between 730-810 nm, that penetrates layers of the human body, among them the skin, the scull and the brain. It is either scattered within the tissue or absorbed by present chromophores. In the rather transparent near infrared region, there are many absorbing light chromophores, but only three are important as far as the oxygenation is concerned, namely haemoglobin (HbO2), deoxyhaemoglobin (Hb) and cytochrome oxidase (CtOx). Oxygenated and deoxygenated haemoglobin absorb light at different wavelengths, allowing a differentiation of these two forms of haemoglobin. The sensors, ("SomaSensors"), are applied to the patients forehead with an integrated medical-grade adhesive. The method is applied by using two source-detector distances in the sensor: a "near" one (shallow), 3 cm from the source and a "far" one (deep), 4 cm from the source. Both samples penetrate the tissue beneath the light source equally well, with the difference that the 4cm source-detector measures signals deeper in the brain or other parts of the body. The subtraction of the near sample from the far one should leave a signal originating predominantly in the brain cortex.

The proper management of brain oxygenation is an essential component of all anesthesiologic and intensive-care procedures. Nevertheless, the brain remains one of the least monitored organs in the perioperative phase and intensive care therapy. Up until now, the anesthesiological application of NIRS as a method for measuring the cerebral oxygenation has only been investigated in a health population, patients undergoing cardiac surgery or cerebrovascular surgery, elderly patients undergoing major abdominal surgery and neonatal infants. If NIRS detects a decrease of rSO2 in these fields, there are specific guidelines of interventions to regain a sufficient oxygen supply in the brain.

Therefore, it is already strongly involved in the patients care in the perioperative setting. A routine use of NIRS on the ICUs has not yet been established. For this reason most studies on rSO2 were performed during surgery. These studies indicate that the measured cerebral oxygenation is affected by the relative proportion of blood in the arterial or venous part of the capillary bed, the haemoglobin concentration, and the systemic saturation.

The precise consequences of alterations in the systemic saturation (SaO2) on the cerebral oxygenation (rSO2) remain unknown. It is a current assumption that rSO2 is directly correlated to Sa02, so that an increase of SaO2 also leads to an increase of rSO2.

The amount of oxygen in the arterial blood depends on the inspired oxygen and the pulmonary gas exchange. These two starting points are primarily affected by the individual's respiratory ventilation. Patients with chronic respiratory failure, or just a temporary (acute) breathing deficiency (e.g. after general anaesthesia), are routinely treated with the continuous positive airway pressure (CPAP) therapy, a non-invasive form of ventilation. It is also commonly used in the treatment of sleep apnea and in neonates (especially premature infants). In these patients CPAP ventilation may prevent the need of tracheal (re-) intubation, or enable earlier extubation. The therapy was developed by Dr. George Gregory and colleagues in the neonatal ICU at the University of California, San Francisco in 1971, and then modified by Professor Colin Sullivan at Royal Prince Alfred Hospital in Sydney, Australia, in 1981. Initially the CPAP therapy was mainly used by patients for the treatment of obstructive sleep apnea at home. Nowadays it is commonly used in ICUs as a form of non-invasive mechanical ventilation. There it is usually reserved for subgroups of patients where the oxygen treatment via a facemask is insufficient. Patients on CPAP therapy are closely monitored in the ICU setting. The treatment supports the patient's spontaneous breathing by building up a positive end expiratory pressure (PEEP). Typically CPAP is a discontinuous therapy on the ICU that means Patients take it off during meals, or just use it if they require temporarily respiratory relief. The pressure applied by most patients ranges between 5 and 12 cmH2O. The patient can individually determine his or her own respiratory frequency as well as the depth of respiration.

The primary study goal is the evaluation of the effect of ventilation with different FiO2 settings applied via NIV on rSO2 in relation to the vital parameters, haemoglobin, SaO2 (analyzed by BGA) and SpO2. These combined measurements can be used to further describe the effect of NIV on rSO2. So far, this kind of study has not been performed on patients in the ICU. Therefore, this project aims at gaining new insights into the influence of NIV on the cerebral oxygenation.

This study will be carried out in patients with good access to the forehead. The NIRS measurement (IBO) will be performed on the temporoparietal cortex on both sides of the head and as a control on both shoulders/upper arms. During the whole investigation the SaO2, the blood pressure (continuously invasive or non invasive in 3 minute-intervals), the heartrate and the ECG will be recorded. These vital signs are part of the routine ICU monitoring.

NIRS measurement will be performed CPAP therapy (FIO2 : 21% - 50% - 100%) for each 10 minutes. The patients will be randomized into two groups, with changed FIO2 application orders.

Condition 1: CPAP 6 mbar PEEP, 0 ASB, FIO2: 21%

Condition 2: CPAP 6 mbar PEEP, 0 ASB, FIO2: 50%

Condition 3: CPAP 6 mbar PEEP, 0 ASB, FIO2: 100%

The order of condition 1 and 3 will be randomized.

Group 1: 21% -> 50% -> 100%

Group 2: 100% -> 50% -> 21%

As a standard setting of NIV with CPAP 6 mbar pressure support will be preferred, if another setup is required, it will be noted in the CRF. After each 10min-period an arterial and central venous blood sample will be drawn, to perform a blood gas analysis (BGA). CPAP and NIRS measurements are both non-invasive in nature. ABP on the other hand is an invasive procedure, which will only be performed if an arterial access is already available. It will also be used for drawing a blood sample (each 2ml) for the BGA. No additional invasive iv. line will be installed for the study.

So far there is little data concerning the effects of NIV therapy on rSO2 in patients with respiratory failure on an ICU setting.

However, there are pilot studies on other breathing deficiencies like obstructive sleep apnea syndrome. They investigated NIV in OSAS patients with the help of NIRS. A periodic decrease in HbO2, blood volume (BV) and the tissue oxygenation index (TOI) which is practically the same as rSO2, could be shown. These studies demonstrate that the mentioned periodic decrease can be eliminated through CPAP therapy.

A number of other studies investigated a different parameter, namely the cerebral blood flow. For this they used transcranial Doppler Ultrasound in patients with OSAS as well as healthy volunteers under CPAP therapy. The results of these studies are not consistent, as they show both de- and increases in the cerebral blood flow during CPAP therapy. The inconsistency of cerebral blood flow is not completely understood. A possible explanation is the relative change in PaCO2, which is already higher in OSAS patients compared to healthy individuals. That would also mean that applying CPAP therapy in healthy volunteers may be harmful, since the lowering in PaCO2 might cause a vasoconstrictive reaction in the brain supplying vessels, lowering the cerebral oxygenation. OSAS patients are more adapted to high levels of PaCO2. Therefore, the vascular response (vasoconstriction or vasodilatation) to an alteration in PaCO2 may differ between patients with pulmonary diseases in comparison to healthy individuals. 30-33

We want to investigate differences in rSO2 in patients undergoing CPAP therapy with different FiO2 settings each for 10min, in order to evaluate the immediate effect of NIV on cerebral oxygenation in ICU patients with breathing deficiency in order to conclude on the accuracy and liability of the NIRS technique.

Primary study objective:

Influence of NIV on rSO2

The following parameters will be measured, calculated and noted in the CRF:

• Blood pressure (BP)
• Heart rate (HR)
• SpO2 (FingerClip- peripheral oxygen saturation)
• rSO2 (left and right hemisphere and shoulder)
• BGA: pO2, pCO2, pH, Haemoglobin (Hb), Hk, SO2, pCO2, BE, SBC
• Ventilation: PEEP, Minute volume, Tidal volume, Respiratory rate, FiO2
• ASA Classification (1-4)

Additionally, the occurrence of vomiting and nausea, headache and agitation during the CPAP therapy will be registered with yes and no. The indication of CPAP in each patient will be documented.

Additionally other side effects will be registered (according to the CRF):

-side effects under CPAP

In our previous study (Doerr et al., 2018) a mean difference of 5.9 between COT vs. CPAP with a standard deviation of 4.2 was observed. Based on that result we expect half the difference (2.95) between both FIO2 21% vs 50% and FIO2 50% vs 100%. Based on a paired t-test with a total sample size of 25 (+ 5 accounted for drop-outs) this results in a power of 0.8.