Acute respiratory distress syndrome: Supportive care and oxygenation in adults

INTRODUCTION — The acute respiratory distress syndrome (ARDS) previously had a mortality rate greater than 50 percent [1]. Mortality has since declined [2-6], but the precise mortality rate is uncertain because estimates tend to be higher in observational studies than randomized trials (figure 1) [7-9]. No single change in the management of ARDS can explain the decrease in mortality, which is likely due to multiple factors (improved approaches to mechanical ventilation and supportive care) [10].

The Berlin Definition of ARDS (published in 2012) has replaced the American-European Consensus Conference’s definition of ARDS (published in 1994) [10,11]. However, it should be recognized that most evidence is based upon prior definitions. The current diagnostic criteria for ARDS are provided separately. (See "Acute respiratory distress syndrome: Clinical features, diagnosis, and complications in adults", section on 'Diagnosis'.)

Supportive care and the treatment of severe hypoxemia in patients with ARDS are discussed here. Epidemiology, diagnosis, etiologies, pathophysiology, clinical manifestations, prognosis, mechanical ventilation, and novel therapies are discussed in detail elsewhere. (See "Acute respiratory distress syndrome: Clinical features, diagnosis, and complications in adults" and "Acute respiratory distress syndrome: Epidemiology, pathophysiology, pathology, and etiology in adults" and "Ventilator management strategies for adults with acute respiratory distress syndrome" and "Acute respiratory distress syndrome: Investigational or ineffective therapies in adults".)

SUPPORTIVE CARE — A minority of patients with ARDS die from respiratory failure alone [3,12-14]. More commonly, such patients succumb to their primary illness or to secondary complications such as sepsis or multiorgan system failure. (See "Evaluation and management of suspected sepsis and septic shock in adults".)

Patients with ARDS require meticulous supportive care, including intelligent use of sedatives and neuromuscular blockade, hemodynamic management, nutritional support, control of blood glucose levels, expeditious evaluation and treatment of nosocomial pneumonia, and prophylaxis against deep venous thrombosis (DVT) and gastrointestinal (GI) bleeding.

Sedation — The use of sedative-analgesic medications in critically ill patients, including patients with ARDS, is discussed in detail separately. (See "Sedative-analgesic medications in critically ill adults: Selection, initiation, maintenance, and withdrawal".)

Sedation and analgesia can be useful in patients with ARDS to the extent that they improve tolerance of mechanical ventilation and decrease oxygen consumption [15,16]. This was illustrated by a study of seven critically ill patients, which found that the use of morphine reduced resting and total energy expenditure by 6 and 8.6 percent, respectively [16].

Patients with severe ARDS may require sedation for several days or longer. As is true for non-ARDS patients requiring sedation and analgesia, choice of agent should be driven by the patient’s specific needs. For example, narcotics may be used for pain and suppression of the respiratory drive; benzodiazepines may be used for anxiety; and antipsychotic agents may be helpful for agitated delirium [17]. (See "Sedative-analgesic medications in critically ill adults: Selection, initiation, maintenance, and withdrawal".)

Intermittent injections of sedative-analgesic agents are preferred, with continuous infusions reserved for patients who require repeated doses to achieve adequate sedation [18]. Increasing evidence suggests that dexmedetomidine may be a useful alternative, particularly when compared to infusions of benzodiazepines, the details of which are discussed separately. (See "Sedative-analgesic medications in critically ill adults: Properties, dose regimens, and adverse effects", section on 'Dexmedetomidine'.)

All these agents have potential side effects and several articles highlight significant morbidity associated with excessive sedation. Strategies such as routinely waking patients each day [19], using intermittent instead of continuous infusions of sedatives [20], following a sedation and analgesia protocol [17,18,21], and avoiding sedation altogether if tolerated [22], may lead to important benefits such as decreased time on the ventilator and fewer nosocomial infections. The avoidance of excessive sedation is discussed in detail elsewhere. (See "Sedative-analgesic medications in critically ill adults: Selection, initiation, maintenance, and withdrawal", section on 'Avoid excess sedation'.)

Using sedation scales such as the Richmond Agitation-Sedation Scale (RASS) may help clinicians meet sedation goals more effectively, decreasing the likelihood of over or under-sedation [18,23]. Most patients should be able to tolerate being kept comfortably awake or lightly sedated (eg, RASS of 0 or negative 1) although some patients with more severe lung injury or poor tolerance of mechanical ventilation may need deeper sedation. Two studies found no evidence that increased sedation is required when patients are managed with low tidal volume as opposed to more traditional higher tidal volume ventilation [24,25].

There is evidence that using no pharmacological sedation may be superior to using a continuous sedative infusion with daily interruption. In a single center study that enrolled patients requiring mechanical ventilation for more than 24 hours (including patients with ARDS), a protocol of no sedation was compared to the use of a continuous sedative infusion with daily interruption [22]. Patients managed without sedation received intensive non-pharmacological support, such as verbal comforting and reassurance. The no sedation group spent more time off the ventilator and less time in the intensive care unit (ICU) than those managed with continuous sedative infusions that were interrupted daily. Similar studies in patients with ARDS need to be performed to determine whether a strategy of no sedation is a viable approach in such patients.

Paralysis (neuromuscular blockade) — Although it is widely recognized that neuromuscular blockade (NMB) can have desirable effects (improves oxygenation [26]) and undesirable effects (prolonged neuromuscular weakness [27]) in patients with ARDS, the impact of these competing effects on patient-important outcomes has remained unclear since data are conflicting. While one older 2010 randomized trial reported a mortality benefit, a 2019 trial reported no mortality benefit in patients with moderate to severe ARDS. Until a clear benefit is demonstrated, we suggest not routinely administering NMBs to patients with moderate to severe ARDS, unless other indications are present (eg, severe ventilator dyssynchrony, particularly if it leads to double triggering, or unwanted motor movement refractory to ventilator adjustment and sedation (table 1)). (See "Neuromuscular blocking agents in critically ill patients: Use, agent selection, administration, and adverse effects".)

●In the first multicenter European trial (ACURASYS; 2010) 340 patients with ARDS were randomly assigned to receive cisatracurium besylate or placebo by continuous infusion for 48 hours [28]. At the time of enrollment, all of the patients had been mechanically ventilated using low tidal volume ventilation and had a PaO₂/FiO₂ ratio of <150 mmHg on a PEEP of ≥5 cm H₂O for less than 48 hours (median onset at 16 hours after diagnosis). Both groups were deeply sedated to a Ramsay sedation score of 6 (no response to glabellar tap). Patients treated with cisatracurium besylate had non-statistically significant lower crude 90-day, 28-day, hospital, and ICU mortality rates compared to the placebo group. A pre-specified analysis that adjusted for baseline differences in the PaO₂/FiO₂, SAPS II severity score, and plateau airway pressure, and found a statistically significant decrease in 90-day mortality in patients treated with cisatracurium besylate (HR 0.68, 95% CI 0.48-0.98). The beneficial effects on 90-day mortality were limited to patients who presented with a PaO₂/FiO₂ ratio of less than 120 mm Hg (ie, severe ARDS). Patients treated with cisatracurium besylate also had significantly more ventilator-free days during the first 28 and 90 days (defined as the number of days since successful weaning from mechanical ventilation) and were significantly less likely to experience barotrauma. There was no difference in the frequency of ICU-acquired neuromuscular weakness. Despite this encouraging study NMBs were not adopted widely by many ICU physicians, perhaps due to poor confidence in the validity of the results and persistent concerns (despite the evidence) regarding adverse effects.

●The second major randomized trial was published nine years later (ROSE). Patients had moderate to severe ARDS (PaO₂/FiO₂ ratio of <150 mmHg) and, similar to ACURASYS, were mechanically ventilated with low tidal volume ventilation but with higher levels of PEEP (≥8 cm H₂O) [29]. Similar to ACURASYS, patients were randomly assigned to receive a cisatracurium infusion (median onset eight hours after diagnosis) for 48 hours but unlike ACURASYS, the control group received light sedation (Ramsay sedation scale 2 or 3). Despite lower PEEP and FiO₂ requirements, cisatracurium did not lower hospital mortality (43 percent each), ventilator-free days, or rates of barotrauma compared with patients receiving light sedation, but did result in more adverse cardiovascular arrhythmias (2.7 versus 0.7 percent). The lack of benefit persisted at 6 and 12 months and although patients receiving cisatracurium were less physically active during their hospital stay, rates of ICU-acquired weakness were no different when compared with patients on light sedation. Limitations of this analysis include the lack of blinding of healthcare professionals administering the agent and early cessation of the trial for reasons of futility.

Reasons for the disparity between the results of these two trials are unclear but may relate to the degree of sedation in the control group, different PEEP strategies, differences in the use of prone positioning, and timing of NMB administration after enrollment.

Clinical use of neuromuscular blocking agents are discussed in detail separately. (See "Neuromuscular blocking agents in critically ill patients: Use, agent selection, administration, and adverse effects".)

Hemodynamic monitoring — Hemodynamic management guided by a central venous catheter (CVC) has been compared to that guided by a pulmonary artery catheter (PAC) in patients with ARDS [7]. In the trial, 1000 patients with ARDS were randomly assigned to receive a CVC or a PAC. There was no difference in mortality, lung function, ventilator-free days, organ failure free days, or ICU-free days at day 28. Rates of hypotension, dialysis, and vasopressor use were also the same in both groups. But, the PAC group had an approximately two-fold increase of catheter-related complications, predominantly arrhythmias. This suggests that the PAC should not be used routinely in patients with ARDS. (See "Pulmonary artery catheterization: Indications, contraindications, and complications in adults" and "Novel tools for hemodynamic monitoring in critically ill patients with shock".)

Nutritional support — The optimal approach to nutritional support in patients with ARDS is uncertain and more high quality evidence is needed to identify the best approach to optimizing patient outcomes. Patients with ARDS are intensely catabolic and nutritional support may help to offset catabolic losses and modulate the metabolic response to stress, mitigate oxidative cellular injury, and promote beneficial immune responses [30]. If the gastrointestinal tract is available for nutritional intake, enteral feedings are preferred. Possible advantages of the enteral route include fewer intravascular infections, less GI bleeding because of gastric buffering, and preservation of the intestinal mucosal barrier, which in turn may decrease bacterial translocation across the gut. Overfeeding offers no nutritional advantage and should be avoided to prevent excessive carbon dioxide production. When patients are fed, it is essential that they be kept semirecumbent with their heads in the upright position to decrease the risk of ventilator-associated pneumonia (VAP) [31]. (See "Nutrition support in critically ill patients: An overview".)

Several large randomized studies have also shown that although initial low volume (“trophic feeding”) enteral feeding does not alter the rate of ventilator-free days or mortality, it is associated with fewer side effects. Thus we prefer low-volume initial enteral feeding in patients with ARDS. (See "Nutrition support in critically ill patients: Enteral nutrition", section on 'Amount and rate'.)

Glucose control — The approach to glucose control in patients with ARDS is extrapolated from trials that enrolled patients with critical illness, including ARDS. This is discussed in detail elsewhere. (See "Glycemic control in critically ill adult and pediatric patients".)

Nosocomial pneumonia — Nosocomial pneumonia is common among patients with ARDS. One prospective study of 30 patients with severe ARDS found that nosocomial pneumonia developed in 60 percent [32]. The first episode occurred at an average of 10 days after the onset of ARDS [33].  

The impact of nosocomial pneumonia on morbidity and mortality is unclear [33]. A post hoc analysis of a multicenter prospective cohort study of mechanically ventilated patients examined the impact of ventilator-associated lower respiratory tract infections (VA-LRTIs), using strict clinical criteria and microbiological confirmation [33]. Among the 524 patients who fulfilled criteria for ARDS, 21 percent had VA-LRTIs, 10.3 percent had ventilator-associated tracheobronchitis (VAT), and 10.7 percent had VAP. Using multivariable analysis controlling for illness severity and baseline illness severity, no difference in ICU mortality was found between patients with and without VA-LRTI. In addition, no relationship was found between VA-LRTI and ICU length of stay or duration of mechanical ventilation.

Given the baseline radiographic abnormalities and frequent colonization by potential pathogens, it is difficult to diagnose pneumonia in patients with ARDS on the basis of clinical factors alone [34,35].The misdiagnosis of pneumonia in patients with ARDS may have unfortunate consequences. Inappropriate treatment of patients without pneumonia promotes the emergence of organisms with antibiotic resistance, while a missed diagnosis may be lethal.

Good clinical practice includes following standard guidelines recommended for the general population of mechanically ventilated patients to help prevent, diagnose, and treat VAP [36,37]. This topic and the Center for Disease classification of ventilator-associate events are discussed in detail separately. (See "Clinical presentation and diagnostic evaluation of ventilator-associated pneumonia" and "Treatment of hospital-acquired and ventilator-associated pneumonia in adults" and "Risk factors and prevention of hospital-acquired and ventilator-associated pneumonia in adults" and "The ventilator circuit".)

DVT prophylaxis — The frequency of DVT and pulmonary embolism (PE) in patients with ARDS is unknown, but the risk is high, despite prophylaxis. These patients often have multiple risk factors for venous thrombosis, including prolonged immobility, trauma, activation of the coagulation pathway, and predisposing illnesses, such as sepsis, obesity, and malignancy. Thus, all patients admitted to intensive care units require some form of thromboprophylaxis, the details of which are discussed separately. (See "Prevention of venous thromboembolic disease in adult nonorthopedic surgical patients".)

GI prophylaxis — Patients requiring prolonged mechanical ventilation are at increased risk for gastrointestinal bleeding [38]. Prophylaxis against stress ulcers is discussed in detail elsewhere. (See "Stress ulcers in the intensive care unit: Diagnosis, management, and prevention".)

Glucocorticoids — Our approach to the use of systemic glucocorticoid therapy in patients with ARDS is as follows:

●Glucocorticoids can be administered in the following patients:

•Patients with non-ARDS-related indications for systemic glucocorticoid therapy – When ARDS has been precipitated by a steroid-responsive process (eg, acute eosinophilic pneumonia), systemic glucocorticoid therapy should be administered. Similarly, glucocorticoids may also be administered to patients with ARDS who have refractory sepsis or community-acquired pneumonia if they meet indications. (See "Treatment of community-acquired pneumonia in adults who require hospitalization", section on 'Adjunctive glucocorticoids' and "Glucocorticoid therapy in septic shock in adults".)

•Patients with moderate to severe ARDS who are relatively early in the disease course who fail standard therapies – We recommend glucocorticoid therapy for most patients who are relatively early in the disease course (within 14 days of onset) who have persistent or refractory moderate to severe ARDS (partial arterial pressure of oxygen/fraction of inspired oxygen [PaO₂/FiO₂] ratio <200) despite initial management with standard therapies, including low tidal volume ventilation. This approach is based on available clinical trial data with suggesting a probable mortality benefit (figure 2). Based on these data, the Society of Critical Care Medicine/European Society of Intensive Care Medicine (SCCM/ESICM) issued a conditional recommendation favoring glucocorticoids in patients with early moderate to severe ARDS [39].

●Glucocorticoids are generally not administered in the following patients:

•Patients with less severe ARDS – We do not routinely use glucocorticoids in patients who have less severe ARDS.

•Patients late in the disease course – We generally avoid glucocorticoid use in patients who have persistent ARDS beyond 14 days based upon limited data suggesting glucocorticoids may be harmful in this setting. In the available clinical trials, the mortality benefit of glucocorticoid therapy was seen only among patients randomized before day 14 after ARDS onset. Among patients randomized on or after day 14, glucocorticoid therapy appeared to increase mortality, as shown in figure C of the figure (figure 2). However, in patients with persistent/refractory ARDS beyond 14 days, it may be reasonable to pursue additional diagnostic evaluation (eg, bronchoscopy, lung biopsy) for other possible etiologies, for which glucocorticoid therapy may be appropriate (eg, organizing pneumonia). (See "Cryptogenic organizing pneumonia".)

•Patients with certain viral infections – We generally avoid glucocorticoids in patients with ARDS secondary to influenza because glucocorticoid use may be associated with worse outcomes in such patients [40]. The use of glucocorticoids in patients with coronavirus disease 2019 (COVID-19)-related ARDS is discussed separately. (See "COVID-19: Management in hospitalized adults", section on 'Dexamethasone and other glucocorticoids' and "COVID-19: Management of the intubated adult", section on 'Use of glucocorticoids for non-COVID-19 reasons' and "Seasonal influenza in nonpregnant adults: Treatment".)

When glucocorticoid therapy is used in ARDS, commonly used regimens include:

●Methylprednisolone 1 mg/kg per day for 21 to 28 days followed by a taper [41,42].

●Dexamethasone 20 mg IV once daily for five days, then 10 mg once daily for five days [43].

The agents used (eg, methylprednisolone, hydrocortisone, dexamethasone) and dosing regimens varied somewhat in the available clinical trials. The efficacy of the different regimens appears to be generally similar [44].

The available evidence regarding the efficacy and adverse effects of glucocorticoid therapy in patients with ARDS is summarized in the table and figures (figure 2) [39,41,42,44-51]. The major adverse effects of systemic glucocorticoids are described in greater detail separately. (See "Major side effects of systemic glucocorticoids".)

An important limitation of the data on glucocorticoid therapy in ARDS is that low tidal volume ventilation (for example, as implemented by the ARDSNetwork) was not consistently performed or not documented in the majority of clinical trials. In the three trials that documented use of low tidal volume strategies, two trials reported similar mortality rates among patients receiving glucocorticoids compared with placebo [41,45], while the third trial reported lower mortality with glucocorticoids [43]. However, when taken together, the data suggest a probable mortality benefit, as shown in figure D of the figure (figure 2).

Ongoing trials may shed light on issues including dosing, timing, and agent selection (NCT02819453).

The effect of inhaled steroids (eg, budesonide) is subject to investigation. (See "Acute respiratory distress syndrome: Investigational or ineffective therapies in adults", section on 'Combined inhaled glucocorticoids and beta agonists'.)

Venous access — While peripheral venous access is the fastest form of access for the initial management of critically ill patients with ARDS, central venous access has additional advantages. As examples, central venous catheters can be used to administer pressors, measure central venous pressure and draw blood for laboratory testing. The indications for central line placement, technique, and complications of peripheral and central venous catheters are discussed in detail elsewhere. (See "Peripheral venous access in adults" and "Central venous access: General principles" and "Placement of jugular venous catheters" and "Overview of complications of central venous catheters and their prevention in adults" and "Routine care and maintenance of intravenous devices".)

Mucolytics — Evidence suggests that there is no role for the routine administration of mucolytics [52,53].

MANAGEMENT OF HYPOXEMIA — By definition, patients with ARDS are severely hypoxemic. Options available for improving arterial oxygen saturation (SaO₂) include:

●Use of high fractions of inspired oxygen (FiO₂)

●Decrease oxygen consumption

●Improve oxygen delivery

●Manipulate mechanical ventilatory support

These options are most frequently applied in combination. Unfortunately, each modality is associated with an element of unquantifiable risk. As a result, the clinician must ultimately choose a strategy that provides adequate oxygenation (PaO₂ ≥55 to 80 mmHg) while minimizing the inevitable risks.

Supplemental oxygen — Most patients require a high fraction of inspired oxygen (FiO₂), especially early in ARDS when pulmonary edema is most severe [54]. Prior to intubation high flow oxygen can be provided through a face mask or high flow nasal cannulae (HFNC). Preliminary data suggest that a subset of patients may be treated successfully with HFNC and potentially avoid intubation [55-57]. Although we believe that it is reasonable to attempt treatment with HFNC in carefully selected patients with ARDS who are alert, hemodynamically stable, and do not require immediate intubation, further prospective studies are needed to define the role of HFNC in patients with this population. These studies are discussed separately. (See "Heated and humidified high-flow nasal oxygen in adults: Practical considerations and potential applications", section on 'Medical patients with severe hypoxemic respiratory failure'.)

Most patients with ARDS require intubation and mechanical ventilation. During the peri-intubation period, 95 to 100 percent oxygen should be given to ensure an adequate SaO₂. Because oxygen uptake may exceed replenishment in areas with low V/Q ratios, some clinicians use slightly less than 100 percent oxygen (eg, 95 percent) in an attempt to prevent absorptive atelectasis [58]. Once well established, absorptive atelectasis is not rapidly reversed by reduction of FiO₂ to maintenance levels, emphasizing the desirability of rapid downward titration of FiO₂ to the lowest fraction necessary to maintain adequate oxygenation [59].

Although the risk of high FiO₂ supplementation has not been studied specifically in patients with ARDS, it is probably significant. Studies in animals and normal humans reveal that high concentrations of oxygen damage the lung within hours, in part by forming toxic oxygen species [60-62]. The specific threshold for oxygen toxicity is unknown but appears to begin above 50 percent, and the risk rises as concentrations approach 100 percent [63]. As a result, the FiO₂ should be decreased to the 50 to 60 percent range as soon as safely possible. Additional therapeutic measures such as fluid management may be needed to allow the FiO₂ to be decreased. (See "Adverse effects of supplemental oxygen".)

Studies that compare conservative to liberal oxygenation targets in mechanically ventilated patients (as well as other populations) have not yielded clear guidelines on optimal levels of oxygenation in ARDS patients (see "Overview of initiating invasive mechanical ventilation in adults in the intensive care unit", section on 'Fraction of inspired oxygen'). Pending the outcome of further studies, in patients with ARDS we recommend adhering to a goal PaO₂ of 55 to 80 mmHg or SaO₂ 88 to 95 percent as used in studies performed by the ARDSNET investigators [64].

Mechanical ventilation strategies in patients with ARDS, including those that may allow a decrease in the FiO₂, are discussed in detail elsewhere. (See "Ventilator management strategies for adults with acute respiratory distress syndrome".)

Fluid management — Although increased vascular permeability is the primary cause of pulmonary edema in early ARDS, the quantity of edema formed depends directly upon hydrostatic pressure, since oncotic forces are less capable of retaining fluid within the capillaries (figure 3A-D) [65-68]. As a result, pulmonary edema is more likely to develop in ARDS than in normals for any given pulmonary capillary hydrostatic pressure. (See "Noncardiogenic pulmonary edema".)

Thus, even in patients who are not volume overloaded, a strategy of conservative fluid management may help patients by reducing edema formation [8,69-71]. This was best illustrated by a trial in which 1000 patients with established ARDS were randomly assigned to either a conservative or a liberal strategy of fluid management for seven days [8]. Patients assigned to the conservative group were managed with a fluid strategy that targeted a CVP <4 mmHg or a pulmonary artery occlusion pressure (PAOP) <8 mmHg. Patients managed with the liberal strategy targeted a CVP of 10 to 14 mmHg or a PAOP of 14 to 18 mmHg. The mean cumulative fluid balance was -136 mL in the conservative strategy group and +6992 mL in the liberal strategy group. The conservative strategy improved the oxygenation index and lung injury score, while increasing ventilator-free days (15 versus 12 days) and ICU-free days (13 versus 11 days). The 60 day mortality rate was unaltered by the fluid management strategy. Despite clearly identified CVP and PAOP goals, mean CVP and PAOP remained well above the target goals in the conservative management group, suggesting that a CVP <4 mmHg or a PAOP <8 mmHg is difficult to achieve safely with the strategies outlined in this population. A clear benefit of conservative fluid management on mortality has not yet been shown [72]. Of note, one study of a subset of participants in the fluid management trial [8] suggested a possible association between conservative fluid management and cognitive impairment in ARDS survivors [70]. In a large retrospective study of critically ill patients, a positive fluid balance on day 3 was associated with increased 30-day mortality (odds ratio 1.26) whereas negative fluid balance was associated with lower mortality [68]. Further studies are warranted to explore the relationship between fluid management and long term outcomes.

Nonetheless, given the potential benefits of conservative fluid management on ventilator and ICU-free days, we prefer a conservative strategy of fluid management in patients with ARDS, as long as hypotension and organ hypoperfusion can be avoided. It is reasonable to target a central venous pressure of <4 mmHg or a pulmonary artery occlusion pressure <8 mmHg; however, it should be recognized that such goals may be difficult to achieve. Preliminary data suggests that combination therapy with albumin solution and furosemide may improve fluid balance, oxygenation, and hemodynamics [73].

Ancillary measures — The need to avoid oxygen toxicity justifies the consideration of a variety of other techniques designed to improve SaO₂ including prone positioning, and strategies to decrease oxygen consumption.

Prone positioning — Prone positioning improves oxygenation in the majority of patients with ARDS. Small uncontrolled studies, subpopulation meta-analyses and one large multicenter randomized trial suggest a survival advantage among patients with severe ARDS [74,75]. The physiologic effects, efficacy, and application of prone ventilation are discussed in detail elsewhere. (See "Prone ventilation for adult patients with acute respiratory distress syndrome".)

Extracorporeal membrane oxygenation — Extracorporeal membrane oxygenation (ECMO) is a useful mechanism employed to improve oxygenation in patients with ARDS. Indications and methods employed to apply ECMO are provided separately. (See "Ventilator management strategies for adults with acute respiratory distress syndrome", section on 'Choosing among the options' and "Extracorporeal membrane oxygenation (ECMO) in adults".)

Decrease oxygen consumption — In diseases with severe pulmonary shunting, increasing the saturation of mixed venous blood (SvO₂) may increase the SaO₂. Therapies that decrease oxygen consumption may improve SvO₂ (and SaO₂ subsequently) by decreasing the amount of oxygen extracted from the blood. Common causes of increased oxygen consumption include fever, anxiety and pain, and use of respiratory muscles; therefore, arterial saturation may improve after treatment with anti-pyretics, sedatives, analgesics, or paralytics [26,76].

Increase oxygen delivery — Oxygen delivery is determined by the following formula:

DO₂ = 10 x CO x (1.34 x Hgb x SaO₂ + 0.003 x PaO₂)

where DO₂ is oxygen delivered, CO is cardiac output, Hgb is hemoglobin concentration, SaO₂ is the arterial oxygen saturation, and PaO₂ is the partial pressure of oxygen in arterial blood. As a result, in addition to low SaO₂, DO₂ may be decreased by a low Hgb and a low CO. In turn, a low DO₂ may decrease SvO₂.

In anemic patients, attempts to increase the hemoglobin concentration may be useful, but exceeding 9 g/dL is unlikely to increase benefit. One multicenter trial randomized 838 critically ill patients to a "restrictive" transfusion strategy to maintain the hemoglobin concentration between 7 and 9 g/dL, or to a "liberal" transfusion strategy to maintain it between 10 and 12 g/dL [77]. The 30-day mortality rates of the two groups did not differ significantly, and patients randomized to the restrictive strategy had a lower mortality rate during the period of hospitalization (22 versus 28 percent; p = 0.05).

Transfusion of packed red blood cells may increase a patient's risk of developing ARDS and dying once ARDS is established [78]. For this reason, we suggest restricting transfusion of packed red blood cells in most ARDS patients, unless the hemoglobin falls below 7 g/dL or if there are other compelling reasons to justify transfusions.

Cardiac output may be augmented by raising filling pressures if they are low (if pulmonary edema is not exacerbated) or by using inotropic agents. However, raising oxygen delivery to supernormal levels is not clinically useful and may be harmful in some circumstances [10,79,80].

Investigational agents — Inhaled vasodilators may be used as a way to improve oxygenation in patients with severe ARDS who are refractory to other therapies. (See "Ventilator management strategies for adults with acute respiratory distress syndrome", section on 'Pulmonary vasodilators'.)

Inhaled vasodilators (eg, nitric oxide, prostacyclin, prostaglandin E1) selectively dilate the vessels that perfuse well-ventilated lung zones, resulting in improved oxygenation due to better ventilation/perfusion (V/Q) matching and the amelioration of pulmonary hypertension (figure 4) [81]. Inhaled vasodilators have few systemic effects and rarely cause systemic hypotension since they act locally and have short half-lives.

Choosing among these options is usually physician- and institution-dependent. For example, some institutions choose prostacyclin since it is generally less expensive and does not require sophisticated equipment for administration, while others use NO when NO is bought in bulk for use in the pediatric population. In either case, it is prudent that the physician familiarizes themselves with the advantages and disadvantages of each agent and ensures that staff are adequately trained in their use. Data thus far, suggest that neither agent is superior to the other. In a retrospective study of 239 patients with ARDS, inhaled nitric oxide and epoprostenol, a prostacyclin analogue, were associated with a similar improvement in oxygenation at four hours [82]. In addition, the proportion of patents that responded was also similar (approximately two-thirds) and there was no difference in other outcomes including mortality, duration of mechanical ventilation, or hospital length of stay.

Inhaled nitric oxide — Inhaled nitric oxide (NO) improves oxygenation, but has not been shown to reduce morbidity or mortality and is associated with a risk of renal impairment [83-89]. This lack of proven beneficial outcome and potential harm argue against the routine use of inhaled NO in settings other than refractory hypoxemia [90].

The evidence is best illustrated by three meta-analyses (each with over 1200 patients) that compared inhaled NO to either placebo or conventional management [85,86,88]. Analyses reported that inhaled NO induced a modest, transient improvement in oxygenation, without any improvement in mortality, duration of mechanical ventilation, or ventilator-free days.

While the major effect of inhaled NO appears to be improved oxygenation, oxygenation does not improve in all patients who receive inhaled NO [81,91]. The factors that determine responsiveness are uncertain, but several have been suggested in retrospective cohort studies. Patients without sepsis or septic shock may respond to inhaled NO more frequently than patients with septic shock [92]. In addition, high baseline pulmonary vascular resistance and responsiveness to positive end-expiratory pressure (PEEP) may predict a response in inhaled NO [93].

Although published guidelines endorse monitoring methemoglobin concentrations when inhaled nitric oxide is given, the risk of methemoglobinemia is rare when typical doses are used and appears to be limited to patients with methemoglobin reductase deficiency [94,95]. A meta-analysis found that inhaled NO increased the risk of renal impairment (relative risk 1.59, 95% CI 1.17 to 2.16), but did not increase the risk of bleeding, methemoglobin formation, or nitrogen dioxide formation [96].

Inhaled prostacyclin — Inhaled prostacyclin has not become routine therapy for adults with ARDS because it has not been shown to improve patient-important outcomes, despite the consistent finding that it is associated with an improvement in oxygenation and a decrease in pulmonary arterial pressure [97-104]. These effects are comparable to those associated with inhaled NO (figure 5A-B). Inhaled prostacyclin may be considered in patients with intractable, life-threatening hypoxemia despite conventional management.

The major advantage of inhaled prostacyclin compared with inhaled NO is that inhaled prostacyclin does not require sophisticated equipment for administration. Clinical experience with inhaled prostacyclin for patients with ARDS suggests that adverse effects are infrequent, although published data are limited [104]. Preliminary studies of inhaled iloprost, a prostacyclin analog, in patients with ARDS and pulmonary hypertension reported improved oxygenation without adverse effects on lung mechanics or systemic hemodynamics [105].

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Acute respiratory failure and acute respiratory distress syndrome in adults" and "Society guideline links: Assessment of oxygenation and gas exchange".)

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, “The Basics” and “Beyond the Basics.” The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on “patient info” and the keyword(s) of interest.)

●Basics topics (see "Patient education: Acute respiratory distress syndrome (The Basics)")

SUMMARY AND RECOMMENDATIONS

●Supportive care - Key components of supportive care include the following, most of which is discussed separately (see 'Supportive care' above):

•Appropriate use of sedatives and paralytics. (See "Sedative-analgesic medications in critically ill adults: Selection, initiation, maintenance, and withdrawal" and "Sedative-analgesic medications in critically ill adults: Properties, dose regimens, and adverse effects" and "Neuromuscular blocking agents in critically ill patients: Use, agent selection, administration, and adverse effects".)

•Careful hemodynamic management. (See "Pulmonary artery catheterization: Indications, contraindications, and complications in adults" and "Novel tools for hemodynamic monitoring in critically ill patients with shock".)

•Nutritional support. (See "Nutrition support in critically ill patients: An overview".)

•Control of blood glucose (See "Glycemic control in critically ill adult and pediatric patients".)

•Prevention of ventilator associated pneumonia - (See "Treatment of hospital-acquired and ventilator-associated pneumonia in adults" and "Risk factors and prevention of hospital-acquired and ventilator-associated pneumonia in adults".)

•Expeditious evaluation and treatment of nosocomial pneumonia. (See "Clinical presentation and diagnostic evaluation of ventilator-associated pneumonia" and "Treatment of hospital-acquired and ventilator-associated pneumonia in adults" and "Risk factors and prevention of hospital-acquired and ventilator-associated pneumonia in adults" and "The ventilator circuit".)

•Prophylaxis against deep vein thrombosis (DVT). (See "Prevention of venous thromboembolic disease in adult nonorthopedic surgical patients".)

•Prophylaxis against gastrointestinal (GI) bleeding. (See "Stress ulcers in the intensive care unit: Diagnosis, management, and prevention".)

•Venous access (See "Peripheral venous access in adults" and "Central venous access: General principles" and "Placement of jugular venous catheters" and "Overview of complications of central venous catheters and their prevention in adults" and "Routine care and maintenance of intravenous devices".)

●Glucocorticoid administration - Our approach to the use of systemic glucocorticoid therapy in ARDS is as follows (see 'Glucocorticoids' above):

•ARDS due to steroid-responsive process or community-acquired pneumonia - When ARDS has been precipitated by a steroid-responsive process (eg, acute eosinophilic pneumonia), systemic glucocorticoid therapy should be administered. Similarly, glucocorticoids may also be administered to patients with ARDS who have refractory sepsis or community-acquired pneumonia if they meet indications. (See "Treatment of community-acquired pneumonia in adults who require hospitalization", section on 'Adjunctive glucocorticoids' and "Glucocorticoid therapy in septic shock in adults".)

•Early moderate to severe ARDS refractory to standard therapy - In addition, for most patients who are relatively early in the disease course (within 14 days of onset) who have persistent or refractory moderate to severe ARDS (partial arterial pressure of oxygen/fraction of inspired oxygen [PaO₂/FiO₂] ratio <200) despite initial management with standard therapies, including low tidal volume ventilation, we recommend glucocorticoid therapy (Grade 1B).

A typical regimen is methylprednisolone 1 mg/kg per day for 21 to 28 days followed by a taper or dexamethasone 20 mg IV once daily for five days followed by 10 mg once daily for five days.

•Mild or late ARDS - We do not routinely use glucocorticoids in patients who have less severe ARDS and we avoid their use in patients who have persistent ARDS beyond 14 days based upon limited data suggesting glucocorticoids may be harmful in this setting (figure 2).

•Influenza-related ARDS - Similarly, we avoid glucocorticoid therapy in influenza-related ARDS.

•COVID-19-related ARDS - The role of glucocorticoids in patients with coronavirus disease 2019 (COVID-19)-related ARDS is discussed separately. (See "COVID-19: Management of the intubated adult", section on 'Use of glucocorticoids for non-COVID-19 reasons'.)

●Oxygenation strategies

•General measures - Patients with hypoxic respiratory failure may benefit from strategies that decrease oxygen utilization, such as antipyretics to control fever and sedatives to control agitation. Occasionally, neuromuscular blockade is required, particularly when asynchrony with the ventilator persists despite adequate sedation or when severe or refractory hypoxemia is present. (See 'Sedation' above and 'Paralysis (neuromuscular blockade)' above.)

•Fluid management - Data suggest that a conservative fluid management strategy that aims to minimize or eliminate positive fluid balance. Aiming for a CVP <4 mmHg or a PAOP <8 mmHg may offer clinical advantages, including improved oxygenation, increased ventilator-free days, and ICU-free days. (See 'Fluid management' above.)

•Additional measures – Additional measures for refractory hypoxemia include the following:

-Prone positioning. (See "Prone ventilation for adult patients with acute respiratory distress syndrome".)

-Extracorporeal membrane oxygenation - (See "Extracorporeal membrane oxygenation (ECMO) in adults".)

-Inhaled pulmonary vasodilators - (See "Ventilator management strategies for adults with acute respiratory distress syndrome", section on 'Pulmonary vasodilators'.)

While blood transfusion and inotropes may augment oxygen delivery, most data caution against indiscriminate use of these strategies. For most patients, packed red blood cells can be withheld until the hemoglobin concentration drops below 7 g/dL, unless there are alternative reasons for transfusion. Similarly, there is no evidence that inotropes benefit ARDS patients with a normal cardiac function. (See "Use of blood products in the critically ill" and "Indications and hemoglobin thresholds for red blood cell transfusion in the adult".)