My name is Dusan Hanijiar. I am an anesthesiologist and intensivist at Massachusetts General Hospital. I will discuss the role of inhaled anesthetics in ICU sedation. I have no disclosures. At the end of this presentation, you will be able to describe the clinical efficacy of the use of inhaled anesthetics in the critical care settings, differentiate the pros and cons of intravenous and inhaled sedatives in the ICU, and appraise the future of inhaled sedatives in the management of patient ventilator dyssynchrony. Pharmacological sedation is a complex medical intervention that is delivered to most mechanically ventilated patients in the ICUs. The management of sedation is interdisciplinary and involves collaboration between three groups of clinicians, prescribers who are physicians or nurse practitioners, approvers who are ICU pharmacists, and administrators who are ICU nurses. These ICU team members have specific roles in sedation management. Prescribers determine the desired level of sedation, typically using Richmond agitation and sedation scale. They choose sedative drug, which could be an intravenous, inhalational, oral, or transdermal agent. They specify the range of doses or infusion rates, determine the mode of administration, which in case of intravenous drugs could be intermittent or continuous, and decide when the sedation should be interrupted to facilitate spontaneous awakening trials and spontaneous breathing trials. Critical care pharmacists importantly contribute to sedation management by optimizing the choice of agents, identifying their titration parameters, investigating drug interactions, allergies, history of malignant hyperthermia, monitoring side effects, for example, QT interval prolongation, and they assist the team with resource utilization during drug shortages. Nurses are responsible for the administration of sedatives and titration of sedation based on their assessments of sedation level, typically using RAS scale and assessment of pain using CPOT scale. However, these assessments are only performed intermittently, given that nurses have multiple other patient care responsibilities, and therefore targeting the specific sedation level is oftentimes challenging. Frequent communication between nurses and the rest of the ICU team is important to provide feedback to the prescribers and optimize the sedation strategy for the individual patients. Guidelines published by SCCM in 2018 recommended that light sedation should be pursued in mechanically ventilated patients whenever feasible. On the RAS scale, this recommended sedation level would correspond to values from minus 2 to 0. This recommendation is based on studies which documented that deep levels of sedation are associated with prolonged mechanical ventilation, delirium, and mortality. However, deep sedation may be indicated in certain mechanically ventilated patients, such as those with moderate or severe ARDS. Neurologic examination alone is not sufficient to guide sedation dosing in patients with deeper sedation targets, for example, RAS-4 or RAS-5, because clinical exam cannot reliably distinguish appropriately sedated from over-sedated patients. Objective monitoring of brain activity with EEG sensors is critical to determine when patients are appropriately sedated and their EEG is dominated by slow delta oscillations versus when they are over-sedated and develop patterns of birth suppression or isoelectricity. The panel on the left illustrates the progression of EEG patterns from a wake state to unconsciousness as induced by propofol administration. The last two patterns, birth suppression and isoelectricity, indicate over-sedation, and these two patterns can be easily recognized even by clinicians who do not have significant expertise in EEG. These pathologic EEG patterns have been associated with development of delirium and increased mortality, and should be avoided in mechanically ventilated patients. Prolonged mechanical ventilation is commonly required to treat patients with severe pneumonia or ARDS, and creates many challenges for management of sedation. This problem became very apparent early during COVID-19 pandemic, when many patients were receiving mechanical ventilation for up to two or even three weeks. Prolonged sedation is often complicated by development of tolerance to intravenous sedatives and opioids, which leads to escalation of doses over time, accumulation of IV sedatives, which is associated with delayed wake up during spontaneous awakening trials, over-sedation, as already discussed, sedation-associated delirium, and side effects of sedatives, for example, elevated plasma triglycerides with propofol sedation. Due to prolonged sedation, patients are also prone to withdrawal during sedation weaning. Because of the challenges associated with the use of intravenous sedatives, COVID-19 pandemic increased the interest of intensivists in the use of inhalational sedation as an alternative. Isoflurane and sivoflurane are most commonly used volatile anesthetics. From chemical perspective, they are fluorinated ethers. These anesthetics are liquids at room pressure and temperature, and therefore they require the use of vaporizers for inhalational administration. Importantly, these agents have a long record of safety and efficacy based on their routine use in the operating rooms as general anesthetics during surgery. Isoflurane was approved by FDA in 1979 for induction and maintenance of general anesthesia. Sivoflurane was approved by FDA in 1996. In addition to their use in the operating rooms, volatile anesthetics can be used as alternative to intravenous agents in mechanically ventilated ICU patients. Isoflurane has been approved for ICU sedation in several European countries. In the United States, the use of isoflurane for ICU sedation is still off-label. Mechanism of action of volatile anesthetics is not completely understood. However, there is ample evidence that they affect neurotransmission in the brain and in the spinal cord. Volatile anesthetics interact with multiple proteins on presynaptic and postsynaptic neuronal membrane. They reduce presynaptic excitation and neurotransmitter release by inhibiting sodium channels and calcium channels and promote repolarization through activation of potassium channels. Volatile anesthetics also reduce neurotransmitter activity in the postsynaptic membrane, where they enhance GABA and glycine inhibitory activity. They additionally inhibit nicotinic acetylcholine, serotonin type 3, glutamate, and NMDA receptors. Minimum alveolar concentration, or MEK, is the most commonly used measure of volatile anesthetic potency. Relevant to surgical anesthesia is concentration of 1 MEK. This is alveolar concentration of volatile anesthetic at which 50% patients do not move in response to surgical stimulus. In the operating room, we monitor end tidal concentration of anesthetic that is displayed on anesthesia workstation, and this concentration closely reflects the alveolar concentration. In case of isoflurane, 1 MEK corresponds to 1.2% end tidal isoflurane, and this concentration is commonly used to maintain general anesthesia. MEK levels of 1.3 and higher have been associated with birth suppression, but lower concentration could cause birth suppression in patients with vulnerable brain. Therefore, EEG monitoring is increasingly used to titrate anesthetics in the operating room. Relevant for ICU sedation are concentrations between 0.3 to 0.5 MEK. These are concentrations between 0.3 to 0.5 MEK, or so-called MEK awake. This is alveolar concentration at which 50% patients do not respond to verbal command. Studies have shown that 0.3 MEK is typically sufficient to maintain light sedation in the ICU. Administration of isoflurane in the ICU requires a specialized anesthetic delivery system. Anesthetic conserving device is a small device that is integrated into the ventilator circuit and allows vaporization of isoflurane. The device is placed between ventilator circuit and endotracheal tube, and it also serves as HME filter. No modifications of breathing circuit are required. Isoflurane is delivered into the anesthetic conserving device from syringe pump. Isoflurane is then vaporized and delivered to patient's lungs. Scavenging system collects the anesthetic gases released from the ventilator exhaust to limit environmental pollution. Gas monitor measures n-tidal concentration of isoflurane. As I already mentioned, n-tidal concentration is a helpful data point, providing us with additional information on the depth of sedation or anesthesia. For instance, if n-tidal concentration was 1%, it is likely that the ICU patient is deeply anesthetized rather than sedated. However, a combination of clinical exam, n-tidal concentration, and limited channel EEG monitoring may represent an ideal approach to managing inhalational sedation. Pictured here is a cross-section of anesthetic conserving device. Anesthetic conserving device contains a miniature vaporizer and gas reflector, which allows rebreathing of approximately 90% of exhaled anesthetic. 10% of exhaled gas reaches the exhalation limb, where it should be scavenged. This reflection of anesthetic reduces the total amount of anesthetic needed to maintain sedation. Isoflurane has several pharmacokinetic advantages over intravenous sedatives. Isoflurane has very low solubility in blood, which allows for rapid equilibration of concentrations between alveoli and brain. When isoflurane delivery is stopped, isoflurane is quickly eliminated by lungs. Therefore, the onset and offset of isoflurane anesthesia is quite rapid. These properties of isoflurane are utilized in the operating rooms when rapid emergence from anesthesia is needed at the end of surgery. In the ICU, rapid offset of isoflurane sedation could facilitate spontaneous awakening trials and faster extubation. There is minimal metabolism or accumulation with isoflurane anesthesia. Isoflurane is easy to titrate. The depth of sedation rapidly follows the changes in the inspired concentration. As mentioned previously, the end tidal concentration of isoflurane can be monitored, further facilitating titration. In contrast, intravenous agents have less predictable onset and offset and are more difficult to titrate because of longer context-sensitive halftimes. The context-sensitive halftime is a function of the duration of drug administration and is difficult to predict in a clinical setting. Liver and renal dysfunction also impact the elimination of intravenous agents. Therefore, spontaneous awakening trials may oftentimes require several hours or even days after prolonged administration of propofol or midazolam. On the other hand, the equipment to administer intravenous sedatives is readily available and ICU clinicians are well familiar with intravenous drugs. Because of this pharmacokinetic exactness, the dosing of isoflurane is more straightforward than dosing of propofol. For instance, anesthesiologists in the operating room can relatively easily determine the adequate range of isoflurane that will be required during surgery. But providing general anesthesia with propofol, so-called total intravenous anesthesia or TIVA, is much more challenging. Frontal EEG monitoring is typically required to find the optimal rate of propofol infusion in the operating room. Optimal infusion rate of propofol is also difficult to determine in the ICU. Broad ranges of propofol are therefore prescribed in the ICU due to the unpredictable effects. When titrating propofol infusions in the ICU to a goal RAS, it is hard to avoid episodes of over-sedation or insufficient sedation. In patients receiving pharmacologic paralysis and who are sedated with propofol, clinicians are also concerned about the risk of awareness, and therefore EEG monitoring is typically used to confirm deep level of sedation. The efficacy and safety of isoflurane versus propofol sedation was recently compared in a Phase III randomized non-inferiority trial that was conducted in Europe. This is the largest trial on in-health sedation published to date. Patients were adult medical and surgical mechanically ventilated patients who were randomized to receive either isoflurane or propofol for sedation. 150 patients received isoflurane and 151 patients received propofol. Isoflurane was administered for up to 54 hours. The primary endpoint of this study was the percentage of time spent within target sedation range between RAS-1 and RAS-4. The non-inferiority margin was set at 15%. The results of the study showed that isoflurane was not inferior to propofol. The figure on the left shows percentage of time spent within the target range and figure on the right shows mean RAS scores on day 1 and 2 between the cohorts. The mean end tidal concentration of isoflurane in this study was 0.45%. One of the secondary endpoints was opioid exposure. This study found opioid sparing effects of isoflurane. On day one, isoflurane significantly reduced opioid requirements, which were 29% lower than in propofol group. Spontaneous breathing was also more likely with isoflurane and was identified in 50% of observations when compared to propofol, when it was identified in 37% of observations. Another secondary endpoint was time to wake up during spontaneous awakening trials. The data showed that there was no significant difference between two groups on day one. However, on day two there was statistically significant difference. Median time to wake up in isoflurane group was 20 minutes, while median time to wake up in propofol group was 30 minutes. The study team also collected data on the safety of isoflurane administration. They found that during maintenance of sedation, vasopressor use in sedated patients was similar in isoflurane group, in 79% patients, and in propofol group, in 77% patients. During emergence, the most common adverse event was hypertension, seen in 7% of patients, immediately after discontinuation of isoflurane, likely due to the rapid offset of the effect. On seven-day follow-up post-randomization, there was no statistical difference in delirium-free days and coma-free days. The impact of sedation with volatile anesthetics on ICU outcomes was also investigated in a meta-analysis of prospective randomized controlled trials, which was published in 2017. Primary outcome assessed the effect of volatile-based sedation on extubation times, which is time between discontinuing sedation and tracheal extubation. Secondary outcomes included time to obey verbal commands, proportion of time spent in target sedation, nausea and vomiting, mortality, length of intensive care unit, and length of hospital stay. The included trials are listed in the table on the left. The authors of this study found that when compared to intravenous sedatives, volatile anesthetics significantly improved extubation times, but not other studied outcomes. The reduction in extubation time was greatest when inhalational sedation was compared to sedation with midazolam. Several large trials of isoflurane sedation are currently enrolling patients. INSPIRE ICU trial is a phase 3 multicenter randomized controlled trial comparing efficacy and safety of sedation with isoflurane versus propofol. Primary outcome is the percentage of time when sedation depth is maintained within the target range. Key secondary outcomes include use of opioids during the treatment period, the wake-up time and end-of-study drug treatment, cognitive recovery at end-of-the-study drug treatment, and spontaneous breathing effort during the treatment period. SAFE-ICU is a multicenter randomized controlled trial which investigates the impact of isoflurane sedation on respiratory parameters, survival, and long-term outcomes of critically ill patients with hypoxemic respiratory failure. Primary outcome measures will be hospital mortality, ventilator-free and ICU-free days at day 30, and quality of life at 3 and 12 months after discharge. Isoflurane has been also utilized as a rescue drug in the ICU during specific situations such as refractory status epilepticus and status asthmaticus. The ability of isoflurane to induce EEG birth suppression is utilized during status epilepticus. Retrospective study, which included 45 patients in Germany, found that isoflurane terminated refractory status in 51% of patients. Concentrations up to 3.5% isoflurane were administered. Bronchodilatory effects of isoflurane are utilized to treat status asthmaticus. Retrospective study, which included 31 pediatric patients, found significant improvement in pH and PCO2 within four hours of initiating isoflurane. Mean maximum isoflurane dose was 1.1%. There are very few contraindications to administration of isoflurane. One absolute contraindication is a known genetic susceptibility or history of malignant hyperthermia. Hypersensitivity reactions to isoflurane are very rare. Isoflurane should be administered with caution in pregnant patients, given unknown impact of prolonged administration on fetus, in hemodynamically unstable patients with shock, given that isoflurane may further exacerbate hypotension, and in patients with increased intracranial pressure, given that isoflurane causes cerebral vasodilation and could further exacerbate intracranial hypertension. Malignant hyperthermia, or MH, is an autosomal dominant inherited trait. Individuals who are predisposed to this condition most commonly carry defects in ryanodine receptors in skeletal muscles. Ryanodine receptors regulate the passage of calcium from sarcoplasmic reticulum into sarcoplasm. Abnormal receptors allow excessive myoplasmic calcium accumulation in the presence of certain anesthetic triggering agents, which leads to sustained muscle contraction. Malignant hyperthermia crisis is a life-threatening complication of volatile anesthetic administration. MH crisis is a hypermetabolic state. It initially manifests as elevated CO2 levels and tachycardia. CO2 elevations are typically detected through end tidal CO2 monitoring on anesthesia workstation, and could also manifest as increased respiratory drive. Acidosis, hyperthermia, myolysis, and renal failure are later presentations of MH. Extremely high body temperatures are associated with the development of disseminated intravascular coagulation. When MH is suspected, volatile anesthetic administration needs to be discontinued. Intravenous dantrolene is the only known treatment for MH. After the MH event, patients should receive supportive care in the intensive care unit. In the final part of this presentation, I will discuss interaction between sedation and patient ventilator desynchrony. Desynchrony is an important problem in mechanically ventilated patients because it increases work of breathing and may exacerbate lung injury. Shown on the left are flow, pressure, and volume waveforms of the most common types of patient ventilator desynchrony. These waveforms that are allowing us to detect desynchrony are continuously displayed on ICU ventilators. Ineffective triggering and reverse triggering are types of desynchrony which develop due to insufficient patient effort. During ineffective triggering, inspiratory muscle efforts are not followed by ventilator breath. During reverse triggering, mechanical breaths delivered by ventilator trigger diaphragmatic contractions. Importantly, both these types of desynchrony can be caused by over sedation and decreased respiratory drive. Therefore, decreasing sedation may eliminate these forms of desynchrony. In contrast, double triggering and inspiratory airflow desynchrony are types of desynchrony which develop due to excessive patient effort. During double triggering, inspiratory effort continues beyond the ventilator inspiratory time and ventilator will be triggered to deliver another breath to the patient. During inspiratory airflow desynchrony, patient has high inspiratory flow demand, which is not matched by ventilator. These types of desynchronies may benefit from increasing sedation to suppress excessive respiratory drive. Several studies have documented that deep sedation may promote ventilator desynchrony. In 2009, a small observational study in medical ICU population found that ineffective triggering is a common form of ventilator desynchrony and the rates of ineffective triggering increase with deeper levels of sedation. A small prospective trial published in 2014 found that during pressure support ventilation, deep propofol sedation increased ineffective triggering, while light sedation with propofol did not. Propofol reduced the respiratory drive, which was evident from electrical activity of the diaphragm. Reverse triggering has been documented in comatose and deeply sedated patients. During reverse triggering, a passive ventilator delivered breath triggers a neural response resulting in involuntary patient effort and diaphragmatic contraction. The figure illustrates that patient triggered breath closely follows ventilator initiated breath. Based on our current knowledge of sedation with isoflurane, we can propose that there could be potential beneficial effects of this form of sedation on patient ventilator desynchrony. We will again divide patients with desynchrony into two groups. In patients with insufficient effort due to over sedation or weakness, isoflurane sedation may promote spontaneous breathing and facilitate spontaneous awakening trials and spontaneous breathing trials and also speed up the weaning of ventilator support. Isoflurane may also reduce opioid exposure in these patients and thereby improve patient's respiratory drive. In patients with insufficient effort due to isoflurane, in patients with excessive effort and high respiratory drive, isoflurane may serve as an adjunct sedative or even primary sedative and mitigate the problem of tolerance to intravenous sedatives and opioids. Isoflurane may also serve as reliable sedative when deep sedation is required. For example, in patients with severe ARDS who are receiving neuromuscular blockade. Recently, a panel of experts in ARDS proposed that a known A2F ICU liberation bundle could be extended by adding letter R, which stands for the control of respiratory drive. These recommendations further highlight the complexity of sedation and ventilation management in patients with ARDS. As we already discussed, management of analgesia and sedation in ARDS can be particularly challenging because patients may present with excessive respiratory drive and may require deep levels of sedation and even paralysis to facilitate lung protective ventilation. As a result, many develop tolerance to intravenous sedatives. An important point of these published recommendations is that ventilator adjustments should always be considered before adjusting sedation in patients with ARDS. The goal of ventilator adjustments is to match ventilator output with patient's respiratory demand. For instance, increases in inspiratory flow could ameliorate airflow dyssynchrony in patients with strong inspiratory effort. In those patients who require deep sedation to facilitate lung protective ventilation, inhalational sedation should be considered, especially in those circumstances when patients are refractory to the sedative effects of intravenous agents. In our institution, we have successfully utilized isoflurane for sedation of patients with COVID-19 related ARDS. Eighteen patients received isoflurane and we subsequently published our experience. Isoflurane was used as a primary anesthetic in six patients when they were receiving neuromuscular blockade or as an adjunct sedative in 12 patients who were mostly undergoing ventilator weaning. We administered isoflurane by anesthesia machines, which were brought to the intensive care unit. The inspired concentrations of isoflurane ranged between 0.27 and 0.84 percent and mean duration of isoflurane exposure was 5.6 days. In our published retrospective cohort study, we analyzed intravenous sedative drug utilization before, during, and after isoflurane exposure. We have found that the initiation of isoflurane led to significant decreases in propofol and hydromorphone infusions. We also reported that prolonged sedation with isoflurane was well tolerated. In conclusion, administration of volatile anesthetics is feasible and safe in the ICU. Volatile anesthetics have many pharmacokinetic advantages over intravenous sedatives and allow faster wake-up when ICU sedation is discontinued and their impact on neurocognitive outcomes such as delirium and cognitive impairment requires further study and randomized trials are currently ongoing. Thank you for your time and attention, please feel free to contact me with any questions you may have.

Dusan Hanijiar

Massachusetts General Hospital

ICU sedation

Inhaled anesthetics

Isoflurane

Ventilator dyssynchrony

Opioid-sparing effects