Hello, I'm Jennifer Kim, I'm an assistant professor of neurology at Yale University and I'd like to thank Hannah and SCCM for the invitation to speak today. I will be starting off our session by talking about the interplay between the physiology of the brain and lungs, and I will then discuss what might happen when potential conflicts arise affecting the management between these two organ systems. As you all know, the lungs and brain are intimately connected in a bidirectional manner. The brain is critically important for driving respiratory patterns, while the lungs dynamically affect the brain physiology through its modulation of oxygenation and ventilation. And when pathology from one organ affects the other, then inadvertent secondary damage can occur. So the brain-lung connection has both voluntary cortical and involuntary brainstem drivers of respiratory control. These pontine and medullary respiratory centers exert sympathetic and parasympathetic outflow via the spinal cord and modulate the diaphragm and respiratory muscles. These effectors then trigger mechano and chemoreceptors, such as your CO2 level, that provides feedback to the respiratory control centers of the brain. The disruption of these respiratory centers can lead to a variety of abnormal central breathing patterns that differ depending on the area of injury affected. Shane Stokes and Kuzma breathing can result from cortical injuries and are also well described from metabolic derangements that likely affect the chemoreceptor feedback loops that I just talked about. Other brainstem-related breathing patterns can also be observed and again vary depending on the exact location of the injury. These various breathing patterns can induce either hypo or hyperventilation and can make vent weaning very challenging. So severe brain injuries can also result in a large catecholamine surge, which induces significant autonomic and inflammatory changes. This is best described in Takotsubo's or neurogenic stress cardiomyopathy, but can also lead to acute lung injury, pulmonary edema, and ARDS. So what happens when a patient develops both severe brain injury and severe lung injury? With the rise of COVID, this has become an increasingly important issue to address. It turns out that there's a significant number of potential conflicts when managing a patient who has concurrent brain and lung pathology. While improving oxygenation is a common goal for both organ systems, the common methods for achieving this in ARDS may actually worsen neurologic injury and intracranial pressure. So we will step through a few common ARDS treatment strategies and how special considerations should be made in patients who have concurrent brain injuries. As most of you know, conventional high-volume ventilation has led to significant barotrauma, anelectotrauma, and inflammation, as modeled here in this rat lung tissue. Thus, low tidal volume mechanical ventilation has become a mainstay of ARDS treatment. It reduces ventilator-associated injury, reduces mortality, and increases ventilator-free days. However, one secondary effect of low tidal volume ventilation is permissive hypercapnia. While permissive hypercapnia is well-tolerated in most patients, it causes an increase in cerebral vasodilation. And when a patient has severe brain injury, this hypercapnia and resultant vasodilation can lead to a worsening of a patient's brain injury through increases in ICP. As a side note, prolonged hypocapnia can cause the opposite problem and lead to cerebral ischemia if not corrected in a timely fashion. Regarding oxygenation targets in low tidal volume ventilation, a PaO2 of 55 is often tolerated. But for patients with severe acute brain injury, this may be actually insufficient. In fact, there's some evidence to suggest that a PaO2 of less than 110 in a patient with severe brain injury can increase the odds of death by 50%. Thus, in severely brain-injured patients in which low tidal volume mechanical ventilation is used, cerebral monitoring should be strongly considered whenever possible. If not possible, then a general target of a PaCO2 of 35 to 45 and a PaO2 target of greater than 110 can be pursued. The use of high PEEP and recruitment maneuvers can be very helpful in improving oxygenation in some patients. However, high PEEP can also induce systemic hypotension related to decreased venous return. While high PEEP is often well tolerated in neurologically normal patients, in patients with brain injury, that reduction in cerebral venous drainage can actually result in elevated intracranial pressure. On the other hand, if high PEEP ends up reducing your cardiac output, then there is reduced cerebral perfusion pressure. And depending on the brain's autoregulatory capacity, this can either lead to cerebral ischemia or lead to compensatory cerebral vasodilation, which increases intracranial pressure. So how high exactly does PEEP have to be before brain-injured patients become affected? Well, the data is far from definitive, but some of the smaller studies suggest that a PEEP less than 12 and a recruitment maneuver using less than 20 are generally well tolerated without significant changes in the ICP. However, titrating specific PEEP targets using cerebral monitoring is ideal whenever possible. So sometimes even basics like fluid management can be challenging in patients with concurrent brain and lung injuries. The adage of dry lungs or happy lungs results in the use of diuretics and fluid conservative strategies for management of ARDS. In a way, neurologists similarly want to draw water out of the brain, since non-swollen brains are also happy brains. But our mainstays of achieving this are usually mannitol or hypertonic saline. However, in certain circumstances, things like hypertonic saline can actually worsen heart failure or volume overload and counter the goal of drying out the lungs. On the flip side, if diuretics are over-aggressively used and induce relative hypotension, then uninjured patients may not compensate and their cerebral perfusion may also go down. Beyond diuretics and hyperosmolar therapy, there are a number of different medication classes that can also be helpful in treating ARDS. For instance, sedation and paralytics are mainstays of ARDS management, particularly for vent synchrony and improvement in oxygenation. These medications are also helpful in reducing metabolic demand in the brain and can reduce intracranial pressure. Thus, the benefits of using these medications exist both for the brain and for lung optimization. But one caution is to remember that once you administer these medications, the ability to monitor the neurologic exam may be lost. Now, that can be okay if you already know the baseline neurologic exam and the trajectory of the patient's pathology is well understood and it's thought to be safe when consulting with your neurologic colleagues. And again, any brain monitoring or intermittent imaging that can be done to inform the safest treatment strategies should be used whenever possible. Pulmonary vasodilators are sometimes used to try and help increase oxygenation in ARDS patients by targeting vasodilation of ventilated alveoli. This improvement in oxygenation is also beneficial, obviously, to the improvement of cerebral oxygenation. Now, there is a small theoretical risk of platelet dysfunction when using inhaled PDE inhibitors, but whether this theoretical risk actually translates to being clinically impactful is not really well understood. But one could argue that caution should be exercised when used in patients with significant intracranial hemorrhage, at least until that hemorrhage has stabilized. Finally, steroids and steroid trials in both ARDS and acute brain injury have somewhat storied histories, and ultimately, it seems like it likely depends on the etiology and the sort of timing and administration of steroids as to whether the steroids will be beneficial or detrimental for a particular patient. And thus, decisions about its use should be made on a case-by-case basis. So proning, especially during COVID, has proven to be a beneficial therapy for ARDS patients. The principle behind the benefit of proning is related to improving VQ mismatch via gravity. However, proning poses some challenges for patients with concurrent brain injury. The sideways head position, laying completely flat, and the abdominal pressure from patients with large abdominal girths can all worsen cerebral venous drainage and therefore increase intracranial pressure. However, with some modifications of the proning procedure in acute brain injured patients, proning can still be beneficial even in acute brain injury. Some types of modifications include reverse Trendelenburg, keeping the head midline using cushions or pillows, and sometimes abdominal pressure related to abdominal girth can be redistributed by propping patients up with pillows. If a patient is still unable to tolerate proning, even with these modifications, then alternate strategies can also be explored, such as the use of chest weights on patients in the supine position, which can help provide some of the benefits of improved VQ mismatch with less of an impact on patients' intracranial pressure. This strategy is currently being explored in the Alterprone trial. Finally, VV ECMO is increasingly used in patients with ARDS refractory to other treatments. The ECMO risk for ischemic and hemorrhagic stroke patients are well documented, and unfortunately, most ECMO trials and institutional protocols actually exclude patients with existing brain injury. There have been case reports and case series data to suggest that ECMO can be modified even for acute brain injury patients. Some examples of modifications that might be used are femoral-femoral cannulations to reduce the risk of impairing a cerebral venous return with jugular cannulation. Another is to modify or even withhold anticoagulation in ECMO circuits to reduce the risk for hemorrhage. This has been shown in the case reports to be successful without the ECMO circuit going down. Thus, in select cases, a multidisciplinary discussion about ECMO eligibility, even for neurologically injured patients, should be strongly considered whenever possible. In summary, concurrent ARDS and acute brain injury is increasingly common, especially during the COVID pandemic. Standard ARDS treatment strategies may worsen brain injury, so understanding potential conflicts can help us avoid secondary harm. Importantly, multispecialty and multidisciplinary teams can help us devise individualized management plans to optimize recovery, and that brain monitoring tools can and should be used to help tailor any treatments administered in these complex patients whenever it's possible. Thank you.

ARDS

acute brain injury

treatment strategies

individualized management

brain monitoring tools