Editors’ Note: This article was accepted on April 14, 2020, and was accurate at that time. Because information about SARS-CoV-2 and COVID-19 is evolving rapidly, please verify these recommendations and information.
I recently spent 10 days at New York City’s Bellevue Hospital, treating patients suspected of having COVID-19. Here are some of the things I learned about managing these patients.
Position, Position, Position
The first thing to appreciate when people come in with severe hypoxia is just how much boosting fraction of inspired oxygen and positioning maneuvers help. The first thing I do is sit them straight up and put a nasal cannula on them, preferentially with a bubble jet humidifier set at 6 L. This is a small bottle of water that fits between the oxygen wall connection and the nasal cannula, and provides humidification. If you avoid intubation, these patients are going to be on oxygen for a long time. If you run high-flow nasal cannula through a standard cannula without humidification, it gets uncomfortable and may injure the mucosa. I then take the nonrebreather mask and put it over that. Part of the reason for the non-rebreather is for covering their mouth (decreasing aeroslization).
The advantage of combining those two is you’re now at 21 L minimum if you’re at 15 on the nonrebreather, and you can crank that even higher (although the manometer will only tell you it’s 15 L) and deliver significantly more oxygen. What I’ve noticed is some of these patients come in at 50 or 60 percent on the pulse oximeter and you get some up into the high 80s, maybe even 90, with this approach. Then you can take a portable chest X-ray, get your labs drawn, and then turn them over and frequently they come up to 90 percent or higher.
The first paper on awake proning in the emergency department just published in Academic Emergency Medicine.1 Looking at a convenience sample of 50 patients from Lincoln Hospital in the South Bronx, it helped three-quarters patients avoid the need for intubation in the first 24 hours, and two-thirds avoid intubation throughout their hospitalization.
This study and the value of patient positioning maneuvers have now been widely discussed in blog posts and podcasts by Salim Rezaie, MD, FACEP (@srrezaie); Scott Weingart, MD (@emcrit), Josh Farkas, MD (@Pulmcrit), and others. Positioning maneuvers include proning, but also turning patients on their right and left sides and having them sit upright in a chair. Suzanne Bentley, MD, MPH, FACEP, at the Icahn School of Medicine at Mount Sinai in New York City, created a series of proning and positioning maneuvers for COVID-19 patients that is being used at Elmhurst Hospital and other public hospitals in New York City.
These maneuvers are not difficult in skinny patients. Obese patients crunch their lower lungs sitting upright in a stretcher and they often can’t tolerate proning at all. Since two-thirds of America qualifies as obese, that’s a real problem. To help patients tolerate proning, I tried a pregnancy massage mattress that I found online, which made it much more comfortable for patients to prone. It allowed obese patients who otherwise couldn’t to tolerate proning. I subsequently started a charity with my brothers, www.prone2help.org, to help get these mattresses to health care workers who need them. In the first three weeks, we sent more than 250 cushions to 125 hospitals in 30 states. We are now shipping a new design that has better access for oxygen devices. Clinicians can go to web site and request a cushion; it is sent out at no charge.
Avoid Intubation When Possible
I view every non-intubation as an exponential win, not only for the patient but for the system from a resource perspective. Figuring out who flies and who doesn’t with a non-intubation strategy requires trying to asses where they are on the timeline of illness and to gauge carefully the work of breathing, respiratory rate, heart rate, and inflammatory markers, not just oxygen saturation improvement.
In my experience, and from the experience of my co-authors on the awake proning paper, most patients who will do well with proning show significant improvement in respiratory rate, heart rate, and subjective dyspnea quickly. If a patient is following commands and willing to work with you, I think they deserve a chance to avoid intubation. These patients need to be watched very closely, however. Disconnection from oxygen quickly causes desaturation. A high-flow nasal cannula adds much higher flows and augments the work of breathing much more than a standard nasal cannula combined with a face mask; augmentation of the work of breathing requires 40 L/min. and usually higher flows. I think these are much better tolerated than continuous positive airway pressure (CPAP) systems, especially in patients who are awake proning. Clinicians should use proning as part of an escalating oxygenation strategy, as detailed by many FOAMED sources. Hospitals effectively using a non-ventilation initial strategy round on these patients regularly—every one to two hours—checking respiratory parameters and closely communicating with patients. Patients who feel they are tiring or do not have documented improvements in respiratory rate, heart rate, etc. may need intubation. Proning seems to be very beneficial in many patients, but in our article, intubation was still required in about a third of patients.
If you do have to intubate a patient, be really careful with your personal protective equipment (PPE). I recommend “CRF4G,”an acronym for cap, respirator, full face shield, base layer of gloves and gown, and a top layer of gloves and gown. It is best do intubations in a negative pressure room and minimize the number of people who go in the room. When you come out of the patient’s room, shed the top layer, wash your base layer gloves, and then go somewhere and get out of the PPE, breathe, and cool down.
I think the best way to intubate COVID-19 patients is to sit them upright and use CPAP before intubation. I put a viral filter and a positive end-expiratory pressure (PEEP) valve on every bag-valve mask. Then push K-roc—which sounds like a radio station but stands for ketamine and rocuronium—at 1.5 mg/kg to 2.0 mg/kg of each drug. Some clinicians recommend succinylcholine instead of rocuronium for faster onset, but at these higher doses, I think the onset times are comparable. Then hold the CPAP mask against the patient’s face for 45 seconds. They will go apneic, but you have CPAP on them. Maintain the PEEP. At 45 seconds, lay them down just as far as necessary to come from above and do your video laryngoscopy. The moment you break that mask seal to intubate them, their pulse oximetry will fall. As many clinicians have noted, COVID pneumonia patients do not seem to get the bradystolic instability common in most rapid sequence intubation (RSI) patients as oxygen falls. I believe this is because they have “acclimatized” in a way to the hypoxia over days before coming in. They do require a ton of sedation, however.
Remember that these patients’ brains are fine—they aren’t encephalopathic from acute hypoxia, hypercarbic (not narcotized), or in shock with poor brain perfusion—so giving them a large dose of ketamine is important. I’ve found ketamine much better than etomidate. You also have to start multiple sedatives to keep them effectively sedated and their heart rate and blood pressure down. Most patients I encountered required three agents: propofol, fentanyl, and either midazolam or dexmedetomidine. Gauging adequate sedation is really important; there seems to be a high incidence of post-traumatic stress disorder related to patients’ being intubated, and needing mechanical ventilation for a prolonged period of time (five to 10 days is not at all uncommon).
Be very careful of your depth of tube placement. I gave up on using stethoscopes for this, but it’s likely that chest X-rays are going to be seriously delayed if the emergency department and hospital are really busy. When you hit the full pandemic spike, everything stops working or slows to a point that it becomes dysfunctional. You have to anticipate that the system is not going to be able to help you should things go south. Double- and triple-check where you secure the tube.
Post-intubation, have your ventilator ready with an inline CO2 detector so you can put it right on and you don’t have to bag.
Whether to use bougies depends on the device you’re using. At Bellevue, we mostly used hyperangulated GlideScopes, so we didn’t use bougies. However, we did have some tube delivery problems with the hyperangulated GlideScope. When you have a heavier patient with a big tongue, sometimes you have problems with tube delivery with rigid stylets. In maybe two out of 15 cases, we switched from the GlideScope to direct laryngoscopy because of significant mouth restrictions. Physicians should rotate the hyperangulated blade around the curvature of the tongue and be very conscientious to avoid the secretion pool in the hypopharynx. In general, I think Macintosh shaped video laryngoscopes have easier tube delivery (especially with a bougie), and are best for handling secretions. These conventional shaped blades are now available from all the video laryngoscope manufacturers.
I don’t believe we should fiber-optically intubate COVID-19 patients through the nose. You can’t have patients nasally intubated for more than three days because they’ll get sinusitis, so anyone started on a nasal tube will need to be switched to oral. Awake nasal intubation also exposes you and everybody in the room to much more infectious aerosolization. RSI through the mouth is the only way to go.
Remember to go slow and smooth. Go slow. Go smooth. One and done is best accomplished by slowing down and succeeding on first effort. Many patients will desaturate, but this seems to be tolerated more than in our usual RSI patients. Be careful with your introduction of the laryngoscope blade. Work your way slowly from uvula to epiglottis to laryngeal exposure to tube delivery. If you can’t get the tube on the first try, use an i-gel (or other supraglottic) attached to a viral filter. Bag the patient and switch to plan B. Safety first—be thorough and focused on proper PPE and minimizing operator and other team member exposures.
Acknowledgements: The real credit for the many lessons I learned and have since shared about COVID belongs to the ED physicians working in New York City for weeks before I arrived, who are still there weeks later, and who will be there for the long haul. I especially want to thank Dr. Nick Caputo and Dr. Reuben Strayer for allowing me to collaborate on their article in Academic Emergency Medicine, as well as “schooling” me about caring for COVID patients
- Caputo ND, Strayer RJ, Levitan R. Early self-proning in awake, non-intubated patients in the emergency department: a single ED’s experience during the COVID-19 pandemic [published online ahead of print Apr. 22, 2020]. Acad Emerg Med. doi: 10.1111/acem.13994.