I'll be on my tippy-toes. OK. Thank you so much. I have no disclosures. Happy to obtain some if anybody wants to help me with that. So we're going to start with a case. This is a pretty typical ARDS case of a 32-year-old with a history of alcohol use disorder who presents with acute pancreatitis. She's resuscitated and subsequently develops ARDS. Her initial ABG, you can see there, pH of 7.2, PCO2 of 55, and a PaO2 of 65. This is on high ventilatory settings of 100% FiO2 and 12 of PEEP. You can see her x-ray over here to the left and her corresponding CT scan over the right with a lot of diffuse bilateral airspace disease. So we perform typical ARDS management. Tidal volumes are set to 6 cc per kg. Her plateau pressures are maintained less than 30. To my speakers to the right here, the PEEP is adjusted according to open lung strategy and to minimize driving pressures. She is sedated. She requires neuromuscular blockade for 48 hours as well as prone positioning. We do rule out abdominal compartment syndrome. And when her hemodynamics permit, we diurese her. And on rounds, we discuss everyday ventilator-induced lung injury and the potential. And we really strictly adhere to the best of our ability to ARDS management. After two weeks of supportive care, her multi-organ system failure subsides. The ventilator settings are now weaned to minimal support. You can see her x-ray over to the right is much improved, a little bit of anelectasis at the bases there. She's awake. She's now sitting at the edge of the bed, and she is increasing strength in her extremities. In fact, she's pulling at her ET tube or trying to. And however, on her daily spontaneous breathing trials, she's hypercarbic and has shallow breathing. She requires subsequently a tracheostomy and prolonged ventilatory weaning. So the question is, what is the cause of her prolonged ventilatory needs? Is this lung injury? Is this ICU-acquired weakness? Or what I suspect, but admittedly we didn't measure, is that this patient had diaphragmatic dysfunction. It's increasingly recognized that the ventilator is a major contributor to diaphragmatic injury, otherwise known as ventilator-induced diaphragmatic dysfunction. So to quickly review before we get to the data, some of the techniques to measure diaphragmatic function and force. On the left is your gold standard of manometry, measuring transdiaphragmatic pressure using esophageal and gastric probes and often with a phrenic nerve stimulator. The middle is your continuous EMG monitoring. And to the right is airway occlusion pressure, as was discussed by my previous colleague. But increasingly being used, a technique to measure diaphragmatic size and function is that of ultrasound. One can measure diaphragmatic thickness by using a linear high frequency probe. You can measure that thickness over time with serial measurements. You can also measure contractile effort and function with each breath by your inspiratory thickening fraction, and that's with M-mode. And this basically measures the percent increase in thickness during the inspiratory cycle. You can also measure function by measuring diaphragmatic excursion. This is with a phased array probe with the B or M-mode, and one allows you to visualize the diaphragmatic displacement over time. So the suspicion that mechanical ventilation can cause atrophy and disproportionately affect the diaphragm really came from a neonatal autopsy study back in the 80s. And you can see the muscle fiber cross-sectional area on the top left. This was a neonate who had died after being on the ventilator for 47 days. If you can compare that to the fiber cross-sectional area on the bottom, this was a neonate who had died after three days being on the ventilator, and those fibers are much larger. So the authors questioned, is this disuse atrophy from prolonged mechanical ventilation, or was this just part of normal myofiber growth in a neonate? So this sparked interest in future studies. So next came a series of animal models. This one in particular were seven piglets who were mechanically ventilated, sedated, and therefore ensured that there were no spontaneous efforts. They used a phrenic nerve stimulator and measured transdiaphragmatic force as well as EMG. And you can see on the bottom left, with increasing frequency of phrenic nerve stimulation, the diaphragmatic force increased on day one, but as early as day three and day five, that weakness started to instill. On the right there, you can see the amplitude on EMG was okay on day one, but by day three and day five, weakness had set in. So the authors had determined that the diaphragm is the problem here, not the phrenic nerve, and this seems to be occurring very early after intubation. Next came the landmark study in New England Journal of Medicine, where the investigators biopsied 14 brain-dead or organ donor biopsies of their diaphragm, and these patients, because they were brain-dead, had complete diaphragmatic silence. And they compared that to eight patients who underwent thoracic surgery and had their diaphragm biopsied for other reasons and served as control. And they found that these brain-dead patients had really small fibers, regardless of the type of myocyte on cross-section compared to the controls. So the question is, critical illness, polyneuropathy, and myopathy. It's a known entity with poor outcomes and associated with prolonged weaning. So the next question, is diaphragmatic weakness just a part of critical illness, polyneuropathy, or is it its own clinical entity? In other words, is the diaphragm disproportionately affected by the ventilator? So the next study here by Dumoul and colleagues looked at diaphragmatic weakness as a separate entity and looked at its clinical relevance. So they evaluated 76 patients, and they evaluated them at their first spontaneous breathing trial and looked at diaphragmatic function as well as limb weakness as a separate entity. And what they found was diaphragmatic dysfunction was twice as frequent as limb muscle weakness with very little overlap between the two. They found diaphragmatic dysfunction was found in 80% of patients who had weaning difficulty. It had much greater impact on weaning failure than did ICU-acquired weakness. And diaphragmatic dysfunction was associated with higher ICU and hospital mortality. So what I've shown so far, we know that atrophy exists. We know that diaphragmatic dysfunction happens early. We know the ventilator seems to be disproportionately affecting the diaphragm. We seem to see that it's clinically relevant. So the next question is, is ventilator-induced diaphragmatic dysfunction only due to disuse and atrophy? In other words, is the suppression of inspiratory efforts caused by the ventilator the only culprit? So if I just decrease somebody's sedation and allow a patient to initiate their own breaths and work out the diaphragm, would I mitigate this entire problem? So this next study by Ferguson and colleagues from the Toronto group looked to establish the impact of diaphragmatic injury resulting from the ventilator on clinical outcomes. They measured ultrasound measurements, including the changes in thickness, as well as diaphragmatic functioning, the thickening fraction. And what they found was over here to the left, when diaphragmatic thickness decreased, the chance of liberating from the ventilator also decreased. And this is atrophy. But interestingly, what they also found on the right-hand side was that when diaphragmatic thickness increased, the chances of liberating from the ventilator also decreased. So they assumed that this was likely due to excessive respiratory efforts. And in fact, they looked at diaphragmatic function, specifically looking at the diaphragmatic thickening fraction compared to the duration of mechanical ventilation. And again, they found this sort of U-shaped or J-shaped concept where too little diaphragmatic activity over here on the left and too much diaphragmatic activity over here on the right was associated with increased duration of mechanical ventilation, as well as complications, including increased need for tracheostomy, length of stay in the ICU, as well as increased mortality. And authors determined that there must be an optimal amount of inspiratory efforts during mechanical ventilation to mitigate this process. So I think standardizing terminology is really important. And to help sort of define the pathophysiology and help enhance communication and really research this concept of ventilator-induced diaphragmatic dysfunction, efforts have been made to standardize terminology. And thus, borrowing from the vernacular of ventilator-induced lung injury, such as biotrauma and volutrauma, is the term myotrauma. So just to briefly review those, the first type is over-assistance myotrauma. This is essentially the excess unloading of the diaphragm by the ventilator or ECMO assistance. It's by far the most established mechanism, and this is basically disuse atrophy. The ventilator's doing too much, and this is what we saw over here on the left. The next is under-assistance myotrauma. This is when the ventilator is really providing inadequate support. So you can imagine this. We have a patient with ARDS, and we're restricting their flows to six cc's per kg, but that patient wants to initiate a larger breath because of acidemia or JACE receptors that are inflamed, or a patient with neurologic disorder who has a high respiratory drive. This enables, or what happens is the patient has excess diaphragmatic work leading to diaphragmatic injury, inflammation, and edema. And this is basically what was happening over here on the right when the ventilator is under-assisting the patient. The third form is eccentric myotrauma. This is basically when the diaphragm is contracting during the expiratory phase when the muscle is lengthening, otherwise known as eccentric contraction. And this is essentially ventilatory dyssynchrony. It also occurs during expiratory breaking, and that's when the diaphragm contracts during expiration to prevent a decrease in end-expiratory volume. In other words, when their patient is trying to mitigate its own analectasis. This is considered the most deleterious form of myotrauma. There's some evidence of eccentric myotrauma. This is a rabbit model where the investigator simulated ventilator asynchrony with breath stacking and reverse triggering, and found that the asynchronous group had much more muscle fiber damage, as well as even deleterious RNA gene expression. The final form of myotrauma is that of end-expiratory shortening myotrauma, and that's the development of longitudinal atrophy in diaphragmatic muscle fibers due to excess PEEP. So you see T0 here is the normal length of your sarcomere. When a PEEP is acutely applied in T1 here, the muscles naturally shorten, and the thin and thick filaments naturally overlap. This creates a less optimal length for tension generation. When PEEP is chronically applied, in an effort to restore sarcomere length to a better tension generation, the muscle starts to remodel, and the sarcomere starts to drop out, or longitudinal atrophy. And then by T3, when you start acutely weaning your PEEP, now the fibers have to overstretch, and this impairs diaphragmatic performance. This myotrauma is mostly theoretical. So mechanical ventilation leads to diaphragmatic weakness by one of these mechanisms over-assistance, under-assistance, eccentric, or end-expiratory myotrauma. This in turn perpetuates a dependence on mechanical ventilation, leading to further diaphragmatic weakness, and then a vicious cycle in stills. So this is a really nice schematic by Gallagher and colleagues, really demonstrating how the ventilator and sedation exerts really complex, interactive, and sometimes competing effects on the mechanisms of both lung and diaphragmatic injury. So can we protect the lung and the diaphragm at the same time? Well, if I were to create a perfect paradigm of lung and diaphragmatic protection, it would be one that maintained lung protective strategy. It would maintain appropriate level of inspiratory effort, not under-assisting nor over-assisting. It would optimize ventilatory synchrony, all in the background, while maintaining respiratory hemostasis. And while a perfect paradigm does not exist, there are a few strategies that I think can help facilitate this dual protection. Number one, it is known that lung protection, when lung protection, lung protection itself does take priority over diaphragmatic protection when management strategies are opposing. We have enough evidence to demonstrate this. Number two, I think it's really important that we start incorporating the terminology of myotrauma into our daily practice, just like I had shown in the beginning case that we were talking about ventilator-induced lung injury, but really not so much about myotrauma. I think this can influence your bedside management and can provide a rationale for the development of treatment paradigms and future studies. I like protocols. I'm a division chief and appreciate what they're worth. However, I think many things have to be nuanced and depend on the individual patients. And sedation is one of those things. Decreased sedation is sometimes necessary in order to enable a patient to initiate their own breaths. This can avoid over-assistance myotrauma. On the flip side, increasing sedation is sometimes necessary in order to enable ARDS management, to reduce high respiratory drive or dyssynchrony, and to avoid under-assistance or eccentric myotrauma. It's important that we understand the effects of analgesics and sedatives on breathing patterns. For example, propofol has much more of an effect on decreasing your respiratory drive compared to opioids, which decrease your respiratory rate more than your drive. It is advised that a multimodal analgesic approach is used. Next, fine-tuning the ventilator to match the respiratory efforts of the individual patient requires really close inspection of airway pressures and waveforms. Ventilatory management is a contact sport and requires individual management on individual patients. These are three patients with three different waveforms on one ICU. The first patient is at high risk for over-assistance myotrauma. The respiratory rate is set at 25. The patient is breathing at 25. This patient is not neuromuscularly blockaded. So in an effort to reduce this, one might decrease the respiratory rate down to 20 or 18 and allow that patient to initiate their own breath and reduce sedation. On the right, these two separate patients are at risk for eccentric myotrauma. The patient in the middle and particularly the patient on the right who's doing breath stacking, and therefore you might want to change the inspiratory time and allow cycling to occur as it should according to the patient's needs. Number five, monitoring technologies I think are really important. I think the use of ultrasound should be part of our daily practice with examining particularly respiratory efforts as well as diaphragmatic dysfunction. There are some studies where you could incorporate your ultrasound into your daily spontaneous breathing trials and use them in conjunction with your rapid spontaneous breathing trials in order to determine whether or not a patient would be successfully weaned off the ventilator. Assist modes of ventilation such as NAVA, proportional assist ventilation, in theory these modes would avoid over-assistance and help to facilitate synchrony. There are some encouraging studies although larger randomized controlled trials are necessary. Finally, future and adjuvantive strategies such as diaphragmatic pacing or phrenic nerve stimulation. A recent study of cardiac surgical patients, the investigators placed percutaneous pacers right at the end of surgery in order to pace the diaphragm. And this did demonstrate improved diaphragmatic function and the ability to extubate at 48 hours. And finally, the potential use of partial neuromuscular blockade. This would allow someone to achieve balance with muscle relaxation, for example, with ARDS, and yet still allow a patient to maintain spontaneous breathing efforts. One recent study looked at titrating rocuronium to decrease that tidal volume from nine to six cc's per kg, and yet the patient still maintained some diaphragmatic force and thus preventing diaphragmatic atrophy. So in sum, in addition to ventilator-induced lung injury, avoiding diaphragm myotrauma is an increasingly important priority in the care of mechanically ventilated patients. Finding the optimal amount of respiratory efforts remains unknown and likely requires a personalized approach at the bedside. Widespread adoption of monitoring technologies is needed. And lastly, an integrated lung and diaphragmatic protective strategy has potential to improve outcomes for our patients. Thank you.

ARDS management

diaphragmatic dysfunction

ventilator-induced myotrauma

mechanical ventilation complications

ultrasound evaluation