COVID-19 ICU: Treat Individual Pathophysiology Not Standard ARDS 

Becky McCall


April 10, 2020

In recent days, intensive care specialists have questioned whether standard respiratory therapy protocols for Acute Respiratory Distress Syndrome (ARDS) in the management of patients with COVID-19 is really the optimal solution for all, at least on initial presentation to hospital.

The discussion rests on the findings, both data-driven and anecdotal, that suggest understanding of COVID-19 physiology may differ from that originally thought. Medscape Medical News reported on a letter published in the March 30th edition of the American Journal of Respiratory and Critical Care Medicine by Dr Luciano Gattinoni of the Medical University of Göttingen in Germany, which made the case that protocol-driven ventilator use for patients with COVID-19 could be doing more harm than good.

Dr Gattinoni noted that COVID-19 patients in intensive care units (ICUs) in Northern Italy had an atypical ARDS presentation with severe hypoxaemia and well-preserved lung gas volume. He and his colleagues suggest that instead of high positive end-expiratory pressure (PEEP), physicians should consider the lowest possible PEEP and gentle ventilation, practicing patience to "buy time with minimum additional damage".

Here, Medscape UK reports on the views of two UK-based intensive care consultants in this fast-moving discussion. Dr Nathalie Stevenson is a consultant in intensive care and anaesthetics, Royal Free London NHS Foundation Trust.  Prof Gary Mills is an NHS consultant in intensive care medicine and anaesthesia based at Sheffield Teaching Hospitals NHS Foundation Trust.  

If a patient enters A&E with respiratory failure due to COVID-19 or another disease, how do you manage them?

Dr Stevenson: The treatment of any respiratory failure depends upon the likely aetiology, the severity of the condition, and taking into account any patient-specific characteristics such as significant frailty or co-morbid disease. In COVID-19, infection control has to also be considered, and so some modalities have been avoided such as high flow nasal cannulae (HFNC) oxygen, which generates aerosols and also uses up significant oxygen supplies.

Most COVID-19 patients are hypoxaemic with type 1 respiratory failure and so

the application of oxygen and recruitment of the small airways in the lungs is the best intervention. To do this, one administers oxygen, and considers the application of less-invasive devices such as CPAP (continuous positive airway pressure). However, if the patient is tiring, or the disease is more severe, both oxygenation and ventilation can be an issue, and mechanical ventilation needs to be considered.

What ‘types’ of COVID-19 are you seeing enter ICU in Sheffield?

Prof Mills: We see many patients presenting through our infectious diseases department and when we ventilate these they often have good compliance and look like Dr Gattinoni’s Type L COVID-19 patients  [Dr Gattinoni characterised Type L phenotypes as having low elastance, low ventilation-perfusion ratio, low lung weight, and low recruitability]. Others may present late through A&E and are on the point of collapse, and can be more like the Type H or ARDS patient [Type H is characterised by high elastance, high right-to-left shunt, high lung weight, and high recruitability]. Similarly, others who have stayed on the ventilator for some time and become more like an ARDS patient with poor compliance are Type H. Others fall between the categories with many changes on chest X-ray.

What’s triggered the change in thinking around early ventilation in the treatment of COVID-19?

Dr Stevenson: The change in thinking has largely come about after acknowledging that there is a lower survival rate for those patients who end up being intubated, and seeing that some patients can perhaps avoid intubation through application of less-invasive respiratory support such as CPAP. Initially we were intubating early because data, originally sourced from China, shows that people coming into the ICU with low oxygen saturations were being ventilated early on.  Another reason is because our experiences with COVID-19 patients are changing all the time, and doctors are sharing their experiences, which prompts others to consider their practices.  

Should COVID-19 management switch from an ARDS-directed strategy to a more physiological one? 

Dr Stevenson: COVID-19 doesn’t appear to behave exactly like ARDS, particularly in the early stages, although it certainly can develop in COVID-19. Some of the ARDS strategies are therefore applicable in some patients but are not as relevant in others.

For example, in ARDS we would aim to provide tidal volumes of no more than 6 mls/kg to prevent lung trauma, but it has been observed that a good majority of the COVID-19 population have very compliant lungs at first presentation, and applying low tidal volumes to this group of patients would result in more atelectasis and collapse, leading to worsened oxygenation and ventilation.

Equally, the application of a very high PEEP (>15 cmH2O) may not be required early on in COVID-19 and may cause harm in terms of lung trauma and/or haemodynamic instability.

Another facet of ARDS management would be fluid restriction, but we are seeing many patients develop acute kidney injury and need for renal replacement therapy. This is likely to be multifactorial, but the fluid-restrictive strategies that apply at a certain point in ARDS are perhaps not as beneficial in the early stages of COVID-19.

What is the argument for using continuous positive airway pressure (CPAP) for longer?

Dr Stevenson: We’ve been seeing anecdotally that those patients started on CPAP and kept on CPAP can do better than those who are intubated early on. Once intubated, the chance of surviving is sadly much lower. 

Prof Mills:  We’ve now started using CPAP when patients present, but this can be limited by practicalities of supplying enough oxygen in hospitals. At first, we intubated relatively early, as was the original advice, but as it became clearer that something else was happening, we began using CPAP on these patients in rooms with negative pressure and good flows to clear virus in the room. CPAP helps some, but it is hard to tell if it helps more than high flow nasal oxygen. Individual variation seems to determine this. With a very compliant lung on CPAP there’s a worry of over-inflation. Unfortunately, it is hard to measure the degree of lung compliance when a patient is not on a ventilator.

How do you determine which patient receives which therapeutic strategy? 

Prof Mills: There are different phenotypes and when someone’s very ill they are likely to move from one [type] to another. We see some patients with extremely bad lungs on entry to ICU, usually because they’ve sat at home, too worried to come in, and arrive in extremis. These patients may fit the Gattinoni description of Type H, with more inflammation and stiffer, more clogged up lungs. This may be due to oedema secondary to the virus or due to high swings in pressure when working hard to breathe. You know they can no longer breathe spontaneously and need ventilation. These are easy to spot.

Some patients have high pyrexias, are hypoxic and have high metabolic demands and so are breathing hard. It’s the ones that are hypoxic but look reasonably well that are a dilemma. We need to decide how hard their lungs are working and how much damage they’re doing to them by generating high swings in pressure. Natural breathing distributes the oxygen around the lung, but on positive pressure ventilation, the non-dependent part of the lung over-stretches and the dependent slowly collapses. Proning helps to reverse this abnormal stretch and can reduce mortality.

Originally, we intubated and mechanically ventilated COVID-19 patients early. Now, from a patient point of view and a pragmatic one, it might be better to do something else. Determining who should receive which treatment is the difficulty. Our view is to give CPAP if they’re not in extremis and watch for a few hours if possible, and continue if there’s improvement. If not, or if they deteriorate, then mechanically ventilate. Once mechanically ventilated it is not easy to backtrack. They will be ventilated for at least several days.

Sedating a patient for mechanical ventilation dilates their blood vessels and lowers blood pressure, but it also sedates other organs such as the gut. Because they are not deep breathing for themselves, not coughing and clearing their chests and potentially aspirating they can develop other pneumonias, plus there’s the unequal ventilation we have already discussed.

However, if we keep a patient on CPAP, they can do worse and struggle to breathe for too long, at which point the lungs clog up and produce greater forces that damage lung tissue. Knowing the point at which to move to ventilation is hard, especially in a disease that we are only starting to understand, and sometimes it’s difficult to decipher what’s the disease, and what might be an adverse effect of a procedure. Right now, we just have to look at the patient and decide the best course of action.   

Are you seeing lung microthrombi and how do they complicate the picture?

Dr Stevenson: We’re seeing a lot of microvascular thrombi in the lungs, which affects oxygenation, and this is very different to ARDS. There’s a lot that is different with COVID-19.

Prof Mills: There could be something else going on in COVID-19 lungs in that we suspect microthrombi, which affects the way blood flows through the lungs. Knowing how to manage this is the next problem. Rather than giving our normal dalteparin dose we are deciding on doubling it in patients with high d-dimers. Most don’t seem to have pulmonary emboli but there is more general clotting. 

Are there any other findings or patient groups of particular interest? 

Prof Mills: There’s a small group with a very high degree of inflammation and we are currently trying to find a way to identify these patients, because they may need drugs that reduce inflammation, but we don’t want to give this to the wrong patient. We are measuring ferritins and interleukins, C-reactive proteins and other means of measuring inflammation as part of an effort to determine thresholds that would allow us to screen for these patients.

How effective is proning in your patients?

Prof Mills: In the Type L patients we find proning also helps patients who are spontaneously breathing. We normally consider proning for patients on ventilators and less so in those who are spontaneously breathing and awake. These patients have shown improved oxygenation when they self-prone. Proning helps by improving VQ relationships [ventilation-perfusion ratios] in Type L patients, and also helps by opening up closed air passages in Type H ventilated patients.

Is it possible to avoid multi-organ failure by adopting an individualised strategy?

Dr Stevenson: It would be wonderful if, by treating the respiratory part of the disease, we could prevent patients progressing to multi-organ failure. But that’s unfortunately never the case in real life. There are some patients who will be genetically predisposed to generating a massive hyperinflammatory response to a disease like COVID-19, or indeed many other things, for example pancreatitis or sepsis, and these people will always be predisposed to go on and develop organ failures including renal and liver failure. We never know, at the start, who this will be. There’s a lot of discussion about a subset of patients generating a cytokine storm. The existence of microvascular thrombi in the lungs means there’s almost certainly going to be thrombi elsewhere in the body too.

What might a treatment approach that accommodates all these considerations look like?

Dr Stevenson: In treating any critically unwell patient, treatment should be tailored to an individual and not to a rigid protocol, particularly with COVID-19 – which we are still learning about.  We need to respond to the patient’s pathophysiology, not just apply protocols blindly.  There’s no magic bullet in any of this, and the best way to manage this is to have strict attention to detail, provide excellent organ support as well as excellent nursing and physiotherapy support.

COVID-19 may be a new disease but it has features of disease that we are all familiar treating. We might look back in 5 years and say ‘XYZ might have been the best protocol at the time’ but we won’t know for sure. As with any critically unwell patient, the best way to treat is to provide excellent organ support, excellent nursing care and that is about as much as we can do. Every week sees new research emerge, and the intensive care community is continually updating its members.

COI: Dr Stevenson and Prof Mills have declared no relevant conflicts of interest.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: