The goals of treatment in patients with ARDS are supportive care and a protective strategy of lung ventilation using low tidal volumes to limit end inspiratory plateau pressure.[51] If the suspected underlying cause of ARDS is infection, then the source should be identified and controlled, and antibiotics started immediately. Otherwise the immediate goals are supportive care and the prevention of complications.

The mortality of patients with ARDS is usually not due primarily to respiratory failure. Most patients die from the underlying cause of ARDS, secondary infections, other organ failures, underlying comorbidities, or the complications of prolonged hospitalisation.

Oxygenation and ventilation

Although the original low tidal volume trial by the ARDS Network targeted an oxygen saturation between 88% and 95%, two subsequent clinical trials suggest that higher oxygenation targets may be associated with better clinical outcomes. A French randomised trial of oxygenation saturation target of 88% to 92% versus ≥96% in patients with ARDS was stopped early due to safety concerns, with numerically higher mortality in the low oxygen saturation target group compared with the higher saturation group at both day 28 and day 90.[52] However, one Cochrane review of oxygen targets in the intensive care unit (ICU) during mechanical ventilation for ARDS, which included this trial alone, concluded that the evidence for giving more or less oxygen to patients with ARDS remains very uncertain because of the high risk of bias (due to lack of blinding, small numbers of participants, and the trial stopping prematurely).[53] An Australian and New Zealand trial of lower versus higher oxygenation targets in critically ill mechanically ventilated patients showed non-significant trends towards worse outcomes in the lower oxygenation target group.[54]

Based on these findings, it seems prudent to target an oxygen saturation of ≥92%.

Occasionally, patients can be managed with non-invasive ventilation, but the failure rate is high and the majority will require endotracheal intubation and mechanical ventilation.[55] Data regarding the use of high-flow oxygen via nasal cannula (HFNC) in patients with acute hypoxaemic respiratory failure are unclear; the safety and efficacy of HFNC in patients with ARDS has not been studied prospectively.[56][57] Ventilator-associated lung injury may be limited by the use of a low tidal volume, plateau-pressure-limited protective ventilatory strategy. This therapy has been shown to reduce mortality.[58][59][60]

A tidal volume of 4-8 mL/kg predicted body weight should be used to maintain an inspiratory plateau pressure <30 cm H₂O.[61] Predicted body weight for men is calculated as 50 + 0.91 × (height [cm] - 152.4), and for women is 45.5 + 0.91 × (height [cm] - 152.4).[58] If the plateau pressure is >30 cm H₂O, then tidal volume should be lowered to 5 mL/kg or as low as 4 mL/kg, if needed.

Use of positive end-expiratory pressure (PEEP) titration tables

PEEP and FiO₂ should be titrated using established PEEP titration tables.[58][62] The available data suggest that higher levels of PEEP are safe and may improve oxygenation in some patients.[61][63][64] Mortality is reduced in patients who respond with improved oxygenation.[65]

Individualised PEEP titration (rather than using a PEEP titration table), lung recruitment manoeuvres in conjunction with higher PEEP levels, and PEEP titration based on radiographic classification of ARDS (as diffuse or focal) have all been evaluated in patients with ARDS.[66][67][68][69] However, consistent clinical benefits have not been demonstrated with these approaches.

Managing respiratory acidosis

Respiratory acidosis, a common complication of low tidal volume ventilation, is treated by increasing the respiratory rate. Although it is not known what level of respiratory acidosis is harmful in patients with ARDS, permissive hypercapnia is often tolerated due to low tidal volume ventilation. However, severe hypercapnia is independently associated with higher ICU mortality.[70] Normocapnia often cannot be achieved (and should not be a goal).

Clinical guidelines recommend that an arterial pH of 7.30 to 7.45 is maintained, but studies suggest patients who undergo permissive hypercapnia can tolerate a blood pH as low as 7.15. Bicarbonate infusions may be administered when the pH falls below 7.15.

Tracheal intubation animated demonstrationTracheal intubation animated demonstration

Bag-valve-mask ventilation animated demonstrationBag-valve-mask ventilation animated demonstration

Prone positioning

Prone positioning can improve oxygenation in patients with ARDS and has been shown to reduce mortality in patients with severe ARDS (PaO₂/FiO₂ <150).[61][74][75][76][77][78]

One systematic review found that reduced mortality was contingent upon patients remaining prone for at least 12 hours daily.[79] Given the potential complications of prone positioning, including facial oedema, pressure sores, and dislodgement of catheters and endotracheal tubes, prone positioning should usually only be considered in patients with severe ARDS (PaO₂/FiO₂ <150).

Proning in critical careProning in critical care

Conservative intravenous fluid management

The patient's fluid balance should be maintained as slightly negative or neutral (providing the patient is not in shock). A central line is recommended to measure the central venous pressure (CVP), with regular assessments of fluid status. The goal is to keep the CVP <4 cm H₂O. The routine use of a pulmonary artery catheter (to measure pulmonary artery occlusion pressure) is not recommended as insertion is associated with more complications than a central line.[40]

A conservative fluid strategy reduced the duration of mechanical ventilation but had no effect on mortality in a large clinical trial in patients with ARDS who were not in shock.[80] Similar results were reported in one systematic review and meta-analysis of adults and children with ARDS, sepsis, or systemic inflammatory response syndrome.[81]

Central venous catheter insertion animated demonstrationCentral venous catheter insertion animated demonstration


In patients who have an infectious cause for ARDS (e.g., pneumonia or sepsis), the prompt initiation of antimicrobials is important.[92][93]

Empirical antibiotics targeted at the suspected underlying infection should be used as soon as possible after obtaining appropriate cultures including blood, sputum, and urine cultures. Antivirals or antifungals may be appropriate in patients with suspected or confirmed viral or fungal infections. Once culture results are available, the antimicrobial regimen can be tailored for the identified organism. There is no evidence to support the use of antibiotics in patients who have ARDS without infection.

Supportive care

Standard supportive care of critically ill patients includes prevention of deep vein thrombosis, blood glucose control, prophylaxis against stress-induced gastrointestinal bleeding, haemodynamic support to maintain a mean arterial pressure >60 mmHg, and transfusion of packed red blood cells in patients with haemoglobin <70 g/L (<7 g/dL).[94][95] Nutrition should be provided enterally where possible.[96] In one large randomised trial of 1000 patients with ARDS, low-dose enteral feeding for the first 5 days of ARDS had similar clinical outcomes compared with full-calorie feeding.[97] Supplemental nutrition with omega-3 fatty acids and antioxidants is not recommended.[98]

Inhaled or intravenous beta-adrenergic agonists to promote alveolar fluid clearance and resolution of pulmonary oedema are not recommended.[99][100] Neither early nor late administration of corticosteroids has been shown to improve mortality in patients with ARDS who do not have COVID-19, and their routine use is not recommended.[101][102]

Refractory hypoxaemia

In patients with refractory hypoxaemia despite an FiO₂ of 1.0 and high levels of PEEP, rescue therapies for oxygenation should be considered.

Neuromuscular paralysis

  • Neuromuscular paralysis improves ventilator-patient synchrony and often improves oxygenation.

  • Intermittent doses of paralytics can be used as effectively as a continuous intravenous infusion. If a patient is on a continuous intravenous infusion of a paralytic, train-of-four monitoring should be used to monitor the muscle fibre twitch response to the drug.

  • Although one randomised clinical trial showed a 28-day mortality benefit with use of neuromuscular paralysis with cisatracurium besylate for the first 48 hours in severe ARDS (PaO₂/FiO₂ <150), a subsequent study with a similar approach to early neuromuscular blockade in ARDS was stopped early for futility.[103][104] Given these findings, neuromuscular blockade should be reserved for patients with ARDS and refractory hypoxaemia despite low tidal volume ventilation and adequate sedation, particularly if there is still evidence of ventilator-patient dyssynchrony.

Inhaled nitric oxide and inhaled prostacyclin

  • Inhaled nitric oxide can improve oxygenation in patients with ARDS, but does not improve mortality and has been associated with acute kidney injury.[105][106][107] [ Cochrane Clinical Answers logo ] Thus, it should be used only as a rescue therapy for refractory hypoxaemia.

  • Inhaled prostacyclin is easier to administer than inhaled nitric oxide, and also has the potential to improve oxygenation in ARDS through better ventilation perfusion matching. However, there are currently no published large randomised controlled trials of inhaled prostacyclin; thus, it should be used cautiously and only as a rescue therapy.[108]

Extracorporeal membrane oxygenation

  • Where available, extracorporeal membrane oxygenation (ECMO) should be considered (in conjunction with low tidal volume mechanical ventilation) in select patients with severe ARDS in whom standard therapies are failing (i.e., patients with profound refractory hypoxaemia).[109]

  • One multi-centre trial showed that patients with severe ARDS randomised to transfer to a tertiary care centre for consideration of ECMO (75% [n=68] of whom actually received ECMO) were more likely to survive to 6 months without disability than patients randomised to continued conventional management (RR 0.69, 95% CI 0.05 to 0.97, P=0.03).[110] One subsequent randomised multi-centre trial (n=249) did not demonstrate significantly lower 60-day mortality in the ECMO treatment group compared with standard care (35% vs. 46%, respectively; P=0.09); however, one meta-analysis pooling data from both trials reported significantly lower 60-day mortality in the venovenous ECMO group compared with the control group (RR 0.73, 95% CI 0.58 to 0.92, P=0.008) despite a moderate risk of major bleeding in the ECMO group.[111][112]

High-frequency oscillatory ventilation

  • Routine use of high-frequency oscillatory ventilation (HFOV) in moderate-to-severe ARDS is not beneficial, and may be harmful.[113][114][115][116][117]

  • HFOV may have a role as a rescue therapy for patients with severe ARDS and refractory hypoxaemia because the use of HFOV often improves oxygenation.

Coronavirus 2019 (COVID-19)

ARDS is one of the World Health Organization (WHO) criteria for the diagnosis of critical COVID-19 disease.[118] Patients with COVID-19 and ARDS should be treated in line with standard ARDS management recommendations, with the following further considerations:

  • Appropriate isolation and infection prevention and control measures.

  • Corticosteroids (low-dose intravenous or oral dexamethasone, or an alternative corticosteroid) are strongly recommended for adults with severe or critical COVID-19 disease, including those with ARDS, based on several large randomised clinical trials. The recommended duration of treatment is 7 to 10 days.[119][120]

  • Consider a trial of high-flow nasal oxygen or non-invasive ventilation in selected patients with COVID-19 and mild ARDS. Endotracheal intubation should not be delayed if there is no improvement after a short trial (1 hour).[118]

  • Prone positioning for 12 to 16 hours per day is recommended for patients with COVID-19 and severe ARDS.[118] Awake prone positioning can be considered for patients with COVID-19 receiving high-flow nasal oxygen or non-invasive ventilation.[118][121] Two small case series found that many people tolerated the prone position while awake, breathing spontaneously, or receiving non-invasive ventilation; these patients experienced an improvement in oxygenation and a decrease in respiratory rate.[122][123]

  • There are conflicting recommendations across international guidelines about the use of the antiviral remdesivir in patients with COVID-19. Local guidance and protocols should be consulted. The WHO recommends against the use of remdesivir in hospitalised patients in addition to standard care, regardless of disease severity, based on one systematic review and a network meta-analysis of four randomised trials.[120] However, remdesivir is approved by the US Food and Drug Administration for the treatment of COVID-19 in hospitalised adult and paediatric patients (aged ≥12 years and weighing ≥40 kg), based on data from a large randomised clinical trial that showed improvements in time to recovery with remdesivir treatment. Its use in selected patients is supported by several US guidelines.[124][125][126][127][121]

See our topic Coronavirus disease 2019 (COVID-19) for more information on the management of COVID-19.

Use of this content is subject to our disclaimer