Podcast – Diffuse Axonal Injury with John Hell

Welcome to the St Emlyn’s podcast. I’m Iain Beardsell, and I’m delighted to say that with me today I’ve got John Hell. He’s a consultant in Neurointensive Care at the University Hospital of Southampton and was, until recently, the director of the Wessex Neuroscience’s Intensive Care Unit there. He’s also the co-author of the excellent Neurointensive Care guidelines that the team have produced and are freely available on the internet. We’ve linked to a blog post. John, welcome along, and thank you very much for joining us.

Thank you for inviting me. So we’re going to tackle today a topic that I think most emergency doctors won’t really be that familiar with but one that we come across after the trauma patients we’ve seen have been admitted to the ITU. This comes from a case we both looked after recently, which I know you presented at our major trauma meeting. We’ll just do a quick case presentation, and then I’ll let you loose with a couple of questions about it if we can because we’re going to talk today about diffuse axonal injury.

So John, the sort of case we’re thinking of, and this is a hypothetical case, is a major trauma patient, let’s say a 45-year-old cyclist. He comes off at speed and suffers a severe head injury. He’s seen by the pre-hospital team, intubated, ventilated, and transferred up to us at the major trauma centre, where he has a trauma CT. His only injury is an isolated major head injury, but there’s no intracranial bleed. He then comes along to neurointensive care for ongoing management. So, that’s the sort of patient we’re thinking about, and we want to think specifically about diffuse axonal injury.

Now, this is a bit of a mystery to me, so I wondered if we could start with some basics. Could you describe a bit more about the neuroanatomy of diffuse axonal injury? The pathophysiology of diffuse axonal injury is that of axonal shearing. The neuronal cells have their cell bodies in the grey matter at the outside of the brain, and the axons form the white matter on the interior of the brain. There is a different density between the cell bodies and the axons, the axons being myelinated sheaths. When the head is subjected to an accelerating force, the brain will accelerate at different rates within the skull. Because the grey matter and white matter have different connections and densities, they move at different speeds, causing shearing of those axons.

Initially, the primary brain injury is just stretching of those axons. It’s not thought that they are definitely ruptured at that point, but the damage causes them to swell. When they swell and start releasing mediators, that’s when the axons actually start to break down. So, a large part of diffuse axonal injury is the initial mechanical shearing force causing a release of biochemical mediators, which then cause further injury. The secondary injury comes from the cerebral edema caused by those mediators, increasing the intracranial pressure and restricting the diffusion of glucose and oxygen to the brain tissue. This results in relative hypoxia due to an increase in ICP and inadequate cerebral perfusion pressure.

It sounds like this can happen with any head injury, from the most minor to the most severe. Are we seeing this in our patients who come to the emergency department with just a concussion, or does this have to be part of a major trauma situation? I think diffuse axonal injury is far more prevalent than previously thought. This was brought home to me when my son was injured on the way back from playing basketball. Somebody threw a basketball at the back of his head. He didn’t lose consciousness but hit his nose on his knee, creating a bloody mess. He wasn’t able to form new thoughts or behave normally at college for the next 48 hours. So, I think there is an element of diffuse axonal injury even in relatively minor head injuries. In severe and moderately severe traumatic brain injuries, about 72% of survivors have a degree of diffuse axonal injury, though in about 50% of those, it will be combined with both hematomas and concussions.

So, we’ve got this patient who’s come to us with no obvious injury needing immediate surgical management. He’s down in neuro ITU. Can we predict the outcome from that sort of injury? What do we see on the imaging, and how do we know what to do next? On the initial imaging, we may see very little. If you look carefully and know the history well, you can find small petechial hemorrhages at the grey-white matter interface, which show up better on a repeat CT and even better on an MRI scan. The initial treatment is based on the suspicion from their mechanism of injury, GCS, and what’s on the CT scan. We then repeat the CT scan after 6, 12, or 24 hours to see how things have progressed.

Does initial GCS matter? If the pre-hospital team reports a reduced GCS or a low motor score, does that affect the outcome or degree of damage from diffuse axonal injury? In a population of such patients with poor initial GCS, on average, they will do worse. However, on an individual basis, we can’t predict the outcome. We see patients with a GCS of 3 at presentation who go on to make a good recovery and return to independent living. Initial GCS, especially if influenced by alcohol or drugs, doesn’t reliably predict the outcome.

So, what’s the first approach when these patients come to neuro ITU? The first step is to continue stabilizing the patient, ensuring we get everything as near normal as possible in the initial few hours after injury. We treat all major trauma similarly, aiming for complete resuscitation to normal parameters: normal blood pressure, PACO2, and PIO2. For traumatic brain injury, we want to maintain adequate venous drainage from the head by tilting it up initially and ensuring nothing obstructs the internal jugular veins. We also want to reverse any anticoagulation and maintain normal platelet function. Once stabilized, we place an intracranial pressure monitor to maintain adequate cerebral perfusion pressure as ICP increases over 72 to 96 hours.

Are we really saying that care for these patients can take three or four days before knowing the severity of the injury? It depends on the initial GCS. If it’s favorable but the patient was very agitated, we may repeat a scan after 12 or 24 hours to see progression. Initial scans can show little, with significant progression on subsequent scans. We mustn’t assume that the initial scan represents the full extent of the injury. Patients with severe injury on repeat CT scans need a longer period of sedation before waking up.

Now, about intravenous fluids, can you explain your choice and why? The volume of the central nervous system depends on the osmolality of the plasma fluid. Fluids like normal saline or Hartmann’s are options. Normal saline has an osmolality of 308 mOsm/kg but an effective osmolality of 286 mOsm/kg in the plasma, which is acceptable. Hartmann’s, however, is effectively hypo-osmolar in the body, leading to brain swelling. Therefore, we recommend normal saline or blood products for acute CNS injury due to their osmolality.

And this is based entirely on osmolality? Yes, despite the move away from normal saline in some settings, we use it exclusively in our unit for patients with isolated head injuries due to its appropriate osmolality. Over 12 years, we’ve seen no harm from raised chloride ions or impaired coagulation from using normal saline.

In the pre-hospital and early resuscitation phase, what can be done to improve outcomes? Minimizing cerebral metabolic rate for oxygen is key. Thio pento is the best drug for this, though it should be avoided in hemodynamically unstable patients. For stable patients with isolated head injury, thio pento is ideal. For unstable patients, ketamine is acceptable despite old concerns about its use in head injuries.

To simplify, for a hemodynamically stable patient with isolated head injury, thio pento is a good choice. For unstable patients, ketamine is a safe alternative. Correct? Yes, ketamine is appropriate for unstable patients. For stable ones, thio pento is best to minimize cerebral metabolic rate for oxygen and improve outcomes.

John, we’ve discussed diffuse axonal injury and management of severely head-injured patients in various settings. Finally, can we predict the outcomes for these patients? Are there prognostic factors? Diffuse axonal injury was once associated with poor outcomes, but aggressive monitoring and management in neuro-intensive care units have improved results. Outcomes vary, with some patients recovering to independent living. We aim to save as many neurons as possible and minimize secondary brain injury.

Young patients tend to do better due to more brain cells, except children who haven’t finished their education. Higher education levels also correlate with better outcomes. Initial GCS matters on average, but individual predictions are impossible. Complications like hydrocephalus or CNS infection can lead to worse outcomes. Our priority is to give every patient the best chance for recovery.

For doctors in district general hospitals, what should they do when they receive a patient with severe head injury and no operable bleed? Should they refer to a neuro-intensive care unit? Absolutely, any patient with significant traumatic brain injury, regardless of initial scan results, will benefit from immediate intubation, ventilation, and management to normal parameters. Maintaining normal oxygenation, carbon dioxide, blood pressure, osmolality, temperature, and clotting is crucial. Patients should be intubated, ventilated, and referred to an intensive care unit for further monitoring.

John, thank you for your insights. We’ve covered so many aspects of diffuse axonal injury and traumatic brain injury management. It’s been great to have you. Thank you for your time, and we hope to have you on the podcast again to discuss more on neuro-intensive care and trauma patients.

Thank you very much for inviting me.