ARDS or acute respiratory distress syndrome is describe as severe hypoxic respiratory failure (measured by comparing inspired oxygen to arterial oxygen). It is a syndrome - not a diagnosis of underlying pathology.

It can arise for multiple reasons: bacterial infection, viral infection, fungal infection, drug reactions, pancreatitis, transfusion related injury, haemorrhagic shock, neurogenic shock, inhalation injury

High altitude pulmonary oedema. Symptoms must occur within a week and include bilateral pulmonary infiltrates/ oedema that are not cardiac in nature.

What this means is that ALL studies that have looked at ARDS include a broad range of underlying causes, conditions, ages and pathologies and genetic dispositions. When we look at what helps ARDS, a condition with 40% mortality

Our landmark trial was ARDSnet almost 20 years ago and nothing since! This showed that proning, high PEEP and low tidal volumes were of value in ARDS As were low pressures

Over the years, keeping patients “dry”, steroids to reduce “lung water” and paralysing patients to improve lung compliance has been used but never shown to be conclusive.

These were “wet lungs”. Covid19 does not always present with stiff lungs or effusions or oedema - sometimes it shows patchy peripheral areas that do not look like the classic ARDS whiteout

Although the ARDSnet data is the best we have to go on, we need to recognise we have collected research on ARDS as a presentation of signs, not on the basis of its pathophysiology and phenotypes will vary in response to recommendations

It is a knife edge in medicine, of balancing gold standard evidence against innovation in new disease, without being unethical Maverick practitioners.

I was always taught to look at the population, the methodology, the data collection, before being able to blanket use information from any trials.

Lung compliance comes quite lowterslly from the pressure required to inflate lungs. This is not really affected so much by what’s happening inside blood vessels. It it affected by chest wall, and lung tissue/interstitium/alveoli.

Meanwhile VQ matching (ventilation-perfusion) IS affected by both. Another point about trials in intensive care is populations vary hugely and we are not always very good at standardising even the control arm of trials.

Learn to dig really deep into data, if you’ve been taught how, to gain the best nuggets from it.