Updates in Acute Respiratory Distress Syndrome

Abstract

Acute respiratory distress syndrome (ARDS) is a clinical entity characterized by widespread pulmonary injury following an inciting event. ARDS was first recognized in medical literature during the 1960s, and our knowledge of the disease and treatment has since then considerably advanced. The majority of patients who are diagnosed with ARDS ultimately require mechanical ventilation, and an estimated 10–15% of patients admitted to the intensive care unit (ICU) meet diagnostic criteria for severe ARDS. In this chapter, we present a comprehensive overview of ARDS with emphasis on the definition, etiology, pathophysiology, phenotypes, and management. The impact of medical innovations and scientific advances on the evolving definition of ARDS is explored through discussion of the parallels between medicine and technology. This concept is then linked to the myriad of ARDS etiologies which share a similar pathophysiological foundation. Expanding on this idea, we will focus on the ever-changing management of ARDS; importantly, this chapter will scrutinize the various viewpoints regarding mechanical ventilation strategies, prone ventilation, neuromuscular blockade, and extracorporeal membrane oxygenation (ECMO). This chapter concludes by discussion of prognosis and use of artificial intelligence in prognostication.

1. Introduction

ARDS is a clinical syndrome that is characterized by acutely worsening hypoxia with bilateral pulmonary infiltrates present on chest imaging after an inciting pulmonary injury [1, 2]. The COVID-19 pandemic has led to increased awareness and focus on ARDS due to the significant number of associated fatalities [2]. It is paramount to maintain a high index of clinical suspicion for ARDS in patients presenting with dyspnea, worsening hypoxemia, and characteristic findings on chest imaging to ensure prompt recognition and timely intervention. Unsurprisingly, up to 80% of patients who are diagnosed with ARDS will require mechanical ventilation, and an estimated 10–15% of patients will be admitted to the ICU, meeting diagnostic criteria for severe ARDS [1, 3]. In this chapter, we will be discussing etiology, pathophysiology, and management of ARDS. However, despite strides in diagnostic and therapeutic modalities on a global scale, ARDS remains a formidable, life-threatening disease process with an unwavering high mortality rate. Therefore, management guidelines are evolving as newer therapies are being explored.

2. Pathophysiology

2.1 Etiologies

The list of etiologies for ARDS is extensive and can be classified as infectious and non-infectious [1]. See Table 1. Sepsis is the most common infectious cause, while pancreatitis, aspiration of gastric contents, shock, and trauma are the most common non-infectious causes [1, 4]. There are several predisposing factors to ARDS such as smoking, alcohol, air pollution and genetics [5]. Genetic heterogeneity, although rare, may contribute towards sepsis. Haptoglobin variant Hp2 is an example of genetic association with risk of ARDS. This allele is predominant in 60% of people with European descent and thus subjecting them to an increased risk of ARDS [6].

Pulmonary causes (direct lung injury)	Extrapulmonary causes (indirect lung injury)
E-cigarettes, vaping Smoke inhalation Drowning Aspiration Pneumonia Pulmonary Contusion Ventilator-induced lung injury	Neurogenic Ischemic reperfusion after lung transplantation Pulmonary endarterectomy Drug toxicity Transfusion-related acute lung injury Acute Pancreatitis Sepsis Trauma Fat/air embolism

Table 1.

Causes of acute respiratory distress syndrome [1].

2.2 Pathologic stages

ARDS develops through the activation of injury response pathways, involving inflammation and coagulation cascades, both locally in the lung and systemically [3, 4]. The hallmark pathological finding is diffuse alveolar damage caused by neutrophilic alveolitis and hyaline membrane deposition [5, 6]. The alveolar-capillary barrier is formed by Type I (AEC I) and Type II alveolar epithelial cells (AEC II) which intersperse one another. Injury to AEC I hinders fluid transport across the epithelium leading to alveolar flooding, while damage to AEC II type II impairs surfactant production [7, 8]. It is important to recognize that high tidal volumes and inspiratory pressures can further this injury by subjecting poorly compliant alveoli to volutrauma and barotrauma highlighting the importance of proper ventilator management in ARDS patients [9].

Fluid-filled alveoli results in significant V/Q mismatch, inactivation of surfactant, and alveolar collapse [7, 8]. Subsequently, incomplete differentiation of AEC I from transitional cells leads to pulmonary fibrosis [10]. However, a recent study using explanted tissue from patients who underwent lung transplantation secondary to ARDS showed organized transitional cells (i.e., cuboidal, partially spread, flat cuboidal cells) without evidence of fibrosis. This suggests the possibility of clinical recovery and may justify the rationale of adequate treatment support including the use of extra-corporeal membrane oxygenation (ECMO), especially if patient is a candidate for lung transplant [11].

The pathogenesis of ARDS is thought to progress through three different stages: diffuse alveolar damage, proliferative stage, and finally, fibrotic stage. Diffuse alveolar damage usually happens in the first week of ARDS. By the time 2nd week of ARDS comes around, some degree of repair seems to start taking shape. In this proliferative stage, inflammation is resolved, neutrophils are removed, pulmonary edema is cleared, and alveolar-capillary membrane is restored.

However, when the inflammatory cascade persists, the myofibroblasts formed during the proliferative stage in the interstitium and responsible for deposition of extracellular proteins that help in repair, may lead to unabated fibrosis [12]. This process is like the process responsible for fibrotic damage in other organs, e.g., in kidneys [13].

2.3 Phenotyping

The process of grouping of ARDS into homogenous groups is called phenotyping and it aids in streamlining relevant treatment options [14]. Phenotypes can be further classified into sub phenotypes. This is a topic that needs considerable future research. Two sub phenotypes of biological and radiological origin are further described below.

Biologic phenotypes are further classified as hyper and hypo inflammatory. Hyperinflammatory sub-phenotypes have higher IL-6, IL-8, and TNF receptor-1 but lower bicarbonate and protein C concentration as compared to hypo inflammatory [15, 16]. Hyperinflammatory ARDS is associated with longer ICU stay, fewer ventilator-free days and higher 90- days mortality thus highlighting the importance of understanding the biological phenotypes in management [15, 16]. Radiologic phenotypes use the modality of computed tomography (CT) to further classify ARDS as diffuse and patchy loss of aeration vs. predominant dorsal-inferior consolidation. The former responds well to alveolar recruitment strategies while the latter has better outcome with proning [17]. A randomized control trial (RCT) showed personalized treatment based on radiologic phenotyping decreased mortality by 10% [18]. In future, phenotyping may allow for individually tailored treatment strategies and hopefully, better outcomes [19].

3. Diagnosis

Since its inception, the diagnostic criteria of ARDS have undergone many revisions. The 2012 Berlin definition, as noted in Table 2, played an essential role in providing feasible, reliable parameters that could objectively define ARDS. It improved the predictive validity for mortality compared to the previous definition by the American-European Consensus Conference (AECC) [20, 21]. The Berlin criteria improved specificity, established the severity of oxygenation, and acknowledged the role of noninvasive ventilation (NIV) in mild ARDS. However, after global adoption, many limitations became clear—specifically, the exclusion of patients in locations where NIV or invasive ventilation is unavailable, the lack of noninvasive oximetry as equivalent to PaO₂ on arterial blood gas (ABG), and the lack of high-flow nasal oxygen (HFNO) incorporation [20, 21, 22]. The results of the FLORALI trial have shifted the paradigm in favor of NIV in hypoxemic respiratory failure [22]. There are other limitations to Berlin 2012 definition, such as use of chest radiographs, which have compared poorly to CT chest (SN 0.73, SP 0.70, PPV 0.88, NPV 0.47) [20, 22, 23]. Also, when Berlin criteria is matched with biopsies from patients, the criteria has low specificity for ARDS [21]. Finally, the mortality prediction is poor, but improved compared to AECC. Based on a meta-analysis of 4188 patients – Berlin ROC AUC = 0.577 compared to 0.536 for AECC [20]. The Kigali modification recognized these limitations and emerged to address practical constraints in resource-limited regions, yet it did not reach the global adoption seen by the Berlin definition [20, 21]. See Figure 1.

Timing	Within 1 week of known clinical insult or new respiratory symptoms
Chest imaging	Bilateral opacities not fully explained by effusions, lobar/lung collapse, or nodules (on chest x-ray or chest computed tomography)
Origin of edema	Respiratory failure not fully explained by cardiac failure or fluid overload. Requires objective assessment (e.g., TTE) to exclude cardiogenic pulmonary edema if no risk factors present
Oxygenation	Mild: PaO2/FiO2 200-300 mmHg with PEEP or CPAP ≥5 cm H2O Moderate: PaO2/FiO2 101-199 mmHg with PEEP ≥5 cm H2O Severe: PaO2/FiO2 ≤ 100 mmHg with PEEP ≥5 cm H2O

Table 2.

Acute respiratory distress syndrome: Berlin criteria [20].

Abbreviations: CPAP: continuous positive airway pressure; FiO₂: fraction of inspired oxygen; PaO₂: partial pressure of arterial oxygen; and PEEP: positive end-expiratory pressure.

Figure 1.

Evolution and timeline of ARDS diagnostic criteria [21, 22].

In 2023, the Matthay modification (Figure 1) aimed to rectify the known limitations of the Berlin definition. Notable modifications include an extended onset of opacities following the initial insult (1–2 weeks) to encompass more indolent diseases like COVID-19, incorporation of ultrasonography for detecting bilateral opacities addresses resource limitations in regions where CT imaging or chest X-rays may not be readily available, the use of SpO2/FiO2 as an alternative to the PaO2/FiO2 ratio, with the advantage of noninvasive pulse oximetry’s widespread availability, and lastly, the inclusion of non-intubated patients on HFNO who meet ARDS criteria, requiring at least 30 L/min, expanding the criteria for invasive or noninvasive ventilation [21, 22].

4. Ventilation

4.1 Non-invasive ventilation

NIV includes simple face mask, nasal canula, non-rebreather, HFNO, continuous positive airway pressure (CPAP), bilevel positive airway pressure (BiPAP) and helmet ventilation. It is used in patients with mild ARDS who are hemodynamically stable or severe ARDS if patient can maintain oxygenation without the need for intubation. However, a large observational study, LUNG SAFE trial showed use of NIV in patients with severe ARDS was associated with higher ICU mortality [24]. Another prospective study on NIV use in ARDS showed slower recovery in patients with pulmonary ARDS vs. extrapulmonary cause [25].

4.2 Mechanical ventilation

Most patients with ARDS end up requiring mechanical ventilation. There are certain evidence-based ventilation practices that are recommended in ARDS. We will visit them next.

4.2.1 FiO2

During and after intubation, care must be taken to prevent hyperoxia to further prevent collapse of alveoli with low V/Q ratio [26]. Adverse effects of oxygen are dose and duration of exposure dependent. For e.g., in rat models mortality rapidly decreases once fio2 is lowered from 100 to 85% [27]. While Fio2 is set at 100% initially, in accordance with ARDSNet, recommendation is to maintain saturation (SpO2) of 88–95% which corelates with an arterial oxygen tension (PaO2) between 55 to 80 mmHg. The recommended goal is conservative oxygenation. No mortality benefit has been found in trials comparing conservative with liberal oxygen therapy targeting PaO2 90 to 105 mmHg: SpO2 ≥ 96% [28].

4.2.2 Low tidal volume ventilation

The gold standard approach of mechanical ventilation in ARDS has been laid down by the famous ARDSNet study (ARMA). It compared the use of standard tidal volume (Vt) of 12 mL/kg with low tidal volume ventilation (LTVV) between 4 and 8 mL/kg while aiming for a plateau pressure (P_plat) ≤ 30 cm of H2O. The weight in kg is predicted body weight. The study showed a decrease in ventilator-induced lung injury (VILI) and 22% reduction in mortality [29]. Despite the compelling evidence, LTVV is underutilized. A prospective cohort in 2016 showed only 60% clinicians successfully diagnosing ARDS and less than two thirds of those diagnosed received LTVV [1, 29]. One part of LTVV strategy is targeting high PEEP, and the other is keeping plateau pressure ≤ 30 mmHg.

4.2.3 PEEP

PEEP plays a key role in alveolar recruitment and oxygenation. As noted above, high PEEP is a part of LTVV. ALVEOLI and LOVS trials demonstrated the use of high PEEP. In ARMA trial, PEEP was high but set according to Fio2 [30, 31]. According to the ALVEOLI trial high PEEP improved oxygenation on Day 1 and Day 3 of admission but showed no difference in ventilator free days and mortality as compared to low PEEP. A meta-analysis showed mortality benefit when higher PEEP strategy was used without recruitment maneuver [32]. Another meta-analysis, a Cochrane review, showed reduced mortality in patients with moderate to severe ARDS but increased mortality in mild ARDS [33]. Based on these two meta-analyses, American Thoracic Society (ATS) made a conditional recommendation for the use of high PEEP without recruitment maneuvers in patients with moderate to severe ARDS [34].

There are many ways PEEP can be titrated such as using FiO2 as in ARMA trial, measuring esophageal pressures, looking at the inflection points on pressure-volume loops, using lung ultrasound or by calculating maximum oxygen delivery. No one measure has proven to be more beneficial than others. Optimal PEEP is still a matter of debate at this point. In general, it is recommended that initial PEEP be set at 5-8 cm of H₂0 for hypoxic respiratory failure and a higher PEEP but not higher than 34 cm of H₂0 for moderate to severe ARDS. The higher PEEP should be individualized based on patient’s compliance, oxygenation status and hemodynamics.

4.2.4 Recruitment maneuvers

To maintain or maximize alveolar oxygenation, recruitment maneuvers may be used. It usually means applying a positive pressure such as 35–40 cm of H2o for an extended duration, e.g., 20–40 seconds. Optimal time to apply recruitment maneuvers is when patient has derecruited such as after transport. Another such maneuver is Sigh Ventilation which is a cyclic maneuver. For this strategy, enough tidal volume is applied to produce a sufficient plateau pressure such as 35 cm of H2O every few minutes. The SiVent Randomized Clinical Trial was undertaken to figure out whether adding sigh ventilation improved clinical outcomes in patients with trauma who were at risk for ARDS. Although the trial did not suggest any difference in ventilator free days, prespecified secondary outcomes suggested mortality benefit [35]. Recently, there were two trials that evaluated the use of high PEEP applied for prolonged duration as a recruitment maneuver in ARDS patients. One trial found it increased mortality, and another found that it increased rates of cardiac arrythmia, but neither found any benefit [36, 37]. Based on these trials, ATS recommended against the use of lung recruitment maneuvers in moderate to severe ARDS [34].

4.2.5 Modes of ventilation

The two common modes of ventilation are Volume (VC) vs. Pressure Control (PC). Evidence does not support one mode over another. The preference usually comes down to practice pattern. The goals are to minimize patient’s work of breathing with the hope that will further prevent cascading inflammation and decrease the ongoing acute lung injury, aim for LTVV, keep airway pressures (plateau and driving pressures) in check to prevent further ventilation induced lung injury and prevent hyperoxia as described above.

There is another mode, Auto Pressure Release Ventilation (APRV) that is sometimes used in practice. The concept of APRV revolves around a specific time switching mode alternating between long duration of high airway pressure (T high) and short duration of low airway pressure (T low). By using the biphasic Positive airway pressure (BIPAP), APRV may help reduce barotrauma by minimizing alternating recruitment and de-recruitment of alveoli. Other advantages may include increased functional residual capacity, unrestricted spontaneous breathing resulting in improved ventilation/perfusion(V/Q) mismatch and decreased need for sedation and neuromuscular blockade [38, 39, 40]. A single center randomized controlled trial by Zhou et al. showed superiority of APRV over LTVV in terms of improving oxygenation and compliance, decreasing plateau pressure, and reducing the duration of mechanical ventilation and ICU stay in patients with ARDS [41]. There is still a dearth of solid data in medical patients to recommend usage of APRV over LTVV. However, it may be a suitable ventilation strategy in selective patients and may be useful depending on the experience of the providers.

4.2.6 Respiratory rate

In addition to LTVV at 6 mL/kg, the initial respiratory rate (RR) is set at ≤35, adjusted to meet the demands of minute ventilation (commonly ranging between 14 and 22 breaths/minute) [29]. Respiratory rate is an integral part of mechanical power which is discussed below.

4.2.7 Driving pressure and mechanical power

Driving Pressure (DP) calculation is useful in patients with moderate to severe ARDS. DP is equal to ventilator measured P_plat minus applied positive end-expiratory pressure (PEEP) or Tidal volume (VT)/ respiratory system compliance. A retrospective analysis of nine trials including 3562 patients on mechanical ventilation for ARDS proved that DP was the best predictor of survival when compared to VT, PEEP and P_plat. A DP increase in 7 cm of H2O was associated with increased mortality [42]. Usually, a target DP of < or = 14 cm of H2O has shown clinical benefits [43].

Lately, a unifying concept has emerged-Mechanical Power (MP). In simple words, it is the collective energy delivered by the ventilator to the patient. Therefore, it can be hypothesized that lower MP may increase the odds for better outcomes.

The mechanical power in Volume and Pressure Controlled mode are described in the following equations (Eqs. (1) and (2)):

$$\text{VC}:\text{MP}=\text{RR}.\left\{\Delta {\text{V}}^2.\left[1/2.{\text{EL}}_{\text{rs}}+\text{RR}.\left(1+\text{I}:\text{E}\right)/60.\text{I}:\text{E}.{\text{R}}_{\text{aw}}\right]+\Delta \text{V}.\text{PEEP}\right\}$$E1

where ∆V is the tidal volume, EL_rs is the elastance of the respiratory system, I:E is the inspiratory-to-expiratory time ratio, and R_aw is the airway resistance [44].

$$\text{PC}:\text{MP}=0.098^{\ast }{\text{ RR}}^{\ast }{\text{ V}}_{\text{t}}{}^{\ast }\left[\text{PEEP}+\Delta {\text{P}}_{\text{insp}}{}^{\ast }\left(1–{\text{e}}^{\ast }–\text{Tinsp}/{\text{R}}^{\ast }\text{C}\right)\right]$$E2

where 0.098 is a conversion factor to J/min, RR is the respiratory rate in beats/min, V_t is the tidal volume in L, PEEP is the positive-end expiratory pressure in cmH₂O, ∆P_insp is the inspiratory pressure in cmH₂O, T_insp is the inspiratory time in s, R is the resistance in cmH₂O/L/s and C is the compliance in L/cmH₂O [45].

To figure out each ventilator variable’s impact, 4549 patients from a pooled database were analyzed [26]. Results showed only two variables, DP, and RR, had a significant association with mortality. The effect of every 1 cm H2O increase in driving pressure produced 4 times the effect of increase in each breath/min of RR. This can be summed into a bivariate model represented by the equation: {(4*DP) + RR}; this was a significant predictor of mortality even more so than mechanical power and may hold the independent variables that truly predict mortality in the mechanical power equation [26].

Chiumello et al. derived simplified equations for both VC and PC that can be used for bedside calculation and found they had good correlation with the above-mentioned equations [46]. The equations are described below (Eqs. (3) and (4)).

$$\text{VC : MP=}\left\{{\text{VE}}^{\ast }(\text{Peak Pressure + PEEP + Inspiratory flow}/6)\right\}/20$$E3

where VE is the minute ventilation expressed in l/min. Peak pressure and PEEP are expressed in cmH₂O.

$$\text{PC}:\text{MP}=0.098^{\ast }{\text{RR}}^{\ast }{\text{V}}_{\text{t}}{}^{\ast }\left[\text{PEEP}+\Delta {\text{P}}_{\text{insp}}\right]$$E4

where 0.098 is a conversion factor from cmH₂O l min⁻¹ in J/min, RR is the respiratory rate, and V_t is the tidal volume in liters. PEEP and ΔP_insp is the pressure (cmH₂O) above PEEP during pressure-controlled ventilation.

In a proof-of-concept study, Petra J et al., showed that VC without pause time had the lowest mechanical power [47]. Overall, mechanical power remains an exciting concept that may offer a unifying variable that can be targeted to decrease the risk of VILI. Further research will be needed to validate this concept for bedside use.

4.2.8 Side effects of mechanical ventilation

The four common pathophysiological mechanisms associated with VILI include atelectrauma, barotrauma, volutrauma and biotrauma [48]. Atelectrauma involves damage to the alveolar unit caused by high-shear forces that help in the recruitment of alveoli. Interalveolar septae and fluid-filled nonaerated alveoli mediate the deformation of neighboring alveolar units, resulting in trauma [48]. Barotrauma is caused by high lung inflation pressure, leading to regional overdistention of the alveolar unit resulting in alveolar rupture, pneumothorax, pneumomediastinum, and subcutaneous emphysema. Volutrauma results from alveolar over-distension. Biotrauma results from the release of cytokine and inflammatory mediators due to mechanical injury. This not only affects normal and diseased lungs, but may result in multi organ dysfunction [48, 49].

LTTV is associated with potential side effects including but not limited to hypercapnia, auto-PEEP, ventilator dyssynchrony and as previously described, VILI. Hypercapnic respiratory acidosis is the result of alveolar hypoventilation to avoid overdistention. Studies have shown that permissive hypercapnia may be beneficial when LTVV is applied [50].

Theoretically, LTVV needs a higher respiratory rate which may result in auto-PEEP due to insufficient time of expiration. Subgroup analysis from ARDSNet trial detected insignificant auto-PEEP (needs source). LTVV may result in increased work of breathing that may result in increased sedation requirement at the initiation of LTVV. The need for increased sedation does not usually persist [51].

A common phenomenon experienced in one quarter of mechanically ventilated patients is dyssynchrony described as incongruity between the patients’ breathing efforts and ventilator- delivered breaths [52, 53, 54]. Factors affecting dyssynchrony can be divided into patient- centered and ventilator centered. Patient’s factors include respiratory drive and lung mechanics (compliance and airflow resistance). Ventilator factors include respiratory rate, inspiratory flow, and trigger sensitivity. Double triggering (breath stacking) commonly seen in ARDS occurs when the second breath is taken before the ventilator completes the first breath [55]. Increasing the VT while maintaining the 4 to 8 mL/kg and recommended P_plat may help with the problem. Ineffective triggering results from a failed effort on patient’s part to trigger the ventilator and may suggest presence of auto PEEP. Reverse triggering involves mechanical ventilation induced contractions. Decreasing sedation, increasing sedation, neuromuscular blockade or adjusting the respiratory rate or tidal volume may help alleviate some of the problems.

5. Fluids

Fluids are a common topic of discussion in ARDS, and in the past have often been the source of some controversy. Studies have shown that a conservative, rather than a more liberal approach, leads to more favorable outcomes [56, 57]. Given increased vascular permeability, a conservative approach to fluids is needed, with the goal of minimizing or eliminating a positive fluid balance, if patients are hemodynamically stable [57]. Although difficult to achieve in clinical practice, literature suggests aiming for a central venous pressure (CVP) of <4 mmHg or a pulmonary artery occlusion pressure (PAOP) of <8 mmHg [56]. Conservative fluid strategy, consisting of fluid restriction and use of diuretics, leads to improvement in oxygenation index, lung injury score and increased numbers of ventilator-free days and ICU-free days [57]. Although a clear mortality benefit has not been proven, retrospective data suggests a positive fluid balance is associated with increased 30-day mortality compared to a negative fluid balance [57, 58]. The high-inflammatory phenotype, which is associated with higher mortality, seems to benefit from conservative fluid strategy the most [58].

6. Prone ventilation

Prone ventilation is a technique in which ventilation is delivered while a patient is placed in a prone position. Prone position reduces the difference between dorsal and ventral transpulmonary pressure, recruits alveoli that collapse during supine ventilation and decreases medial posterior lung compression [59, 60]. These effects improve ventilation and oxygenation by increased lung recruitment [60]. Proning may also decrease systemic inflammation.

Current recommendations stemming from PROSEVA trial suggest that after a 12–24 hours of stabilization period with supine ventilation, prone ventilation can be initiated for up to 36 hours for patients with severe ARDS [61]. Per the PROSEVA trial, the mean duration of time in prone position was 17 hours per session with an average of 4 sessions. Most patients who show a response to prone positioning usually do so within the first hour after being placed into the prone position [61]. A response can be [determined] via improvement in ABG of >10 mmHg PaO2 on ventilator after 1 hour or notable increase in lung compliance based on a decrease in plateau pressure [62].

PROSEVA trial showed decrease in mortality in patients with severe ARDS who undergo prone ventilation [61]. It also showed improvements in ventilator-free days and time to extubation. Proning does not seem to prevent organ dysfunction or lead to reduced ICU stay [61, 62]. Several meta-analyses concur on the findings of PROSEVA trial, with one meta-analysis even reporting that proning of 12 hours or more in patients with severe ARDS led to lower mortality [63]. This suggests that at least 12-hours of proning session daily is sufficient to provide benefit of reducing mortality [63].

Contraindications and complications of proning are outlined below in Table 3 [63]. Of note, use of prone positioning in an ICU is a labor-intensive process as staff need experience and training as they should be able to quickly put patient back into supine positioning, for e.g., to perform cardiopulmonary resuscitation [64].

Absolute contraindications	Relative contraindications	Complications
Spinal instability, including patients at risk of spinal instability (i.e., History of rheumatoid arthritis) Unstable fractures Anterior burns Open wounds in places that would limit proning Pregnancy Acute bleeding	Severe hemodynamic instability (i.e., on multiple pressors) such that patient is not able to tolerate change in positioning Active arrythmias that may require cardioversion Recent thoracic &/or abdominal surgeries	Pressure injuries Facial edema Transient reduction in oxygen saturation Transient arrhythmias

Table 3.

Contraindications and complications of prone ventilation [63].

In the light of COVID-19 pandemic, prone positioning in awake patients who are not intubated has garnered much attention. Recent data shows that prone positioning reduced the need for intubation for patients with COVID-19 induced acute hypoxemic respiratory failure and should be used in patients with COVID-19 that are in the ICU or are requiring advanced respiratory support [65]. Awake COVID-19 patients who are on supplemental oxygen but do not require mechanical ventilation may not have the same clinical benefits from prone positioning and have a high probability of worse clinical outcomes [65]. Therefore, prone positioning should not be used on all patients with COVID-19 but rather, be individualized based on degree of hypoxic respiratory failure. The considerations for initiation of the use of prone positioning, the length and its discontinuation in COVID-19 patients are like non-COVID-19 ARDS patients. However, prone positioning in awake patient does not have enough data to extrapolate to those with non-COVID ARDS and further trials are called for [66].

Recently completed, PRONECMO trial concluded that prone ventilation did not significantly reduce time to successful weaning of ECMO for patients with severe ARDS supported by VV-ECMO when compared to supine position [67].

7. Steroids

The proposed indications glucocorticoid in ARDS varies among professional society guidelines. It is widely accepted that glucocorticoids, specifically dexamethasone, are likely to benefit patients with moderate to severe ARDS during the early course of disease. Before the DEXA-ARDS trial in 2020, many studies had demonstrated fewer ventilator days and shorter ICU length of stay with steroid use [68]. This was further supported by the DEXA-ARDS trial in 2020 which once again showed that in patients with moderate to severe ARDS as defined by PaO₂/FiO₂ of 100–200, early dexamethasone usage improved ventilatory-free days and decreased mortality [68]. The intervention of DEXA-ARDS, which dictates most of current practice, was dexamethasone 20 mg IV daily for days 1–5, followed by 10 mg daily from days 6–10; following extubation, dexamethasone was stopped [68]. It should be noted that steroids should not be used in Influenza related ARDS as it is associated with higher mortality [69].

8. Neuromuscular blockade

The mitigating effects of neuromuscular blockade agents (NMBAs) on inflammation, oxygen consumption, and ventilatory dyssynchrony intrinsic to the pathophysiology of ARDS have long been theorized. However, earlier research has not substantiated this hypothesis. Despite its arguable potential, the widespread adoption of NMBAs among ICUs is still limited, presenting an intriguing avenue for further investigation.

The first multicenter trial was ACURASYS 2010 [70]. It showed that in patients with severe ARDS (i.e., PaO2/FiO2 ≤ 120 mmHg), the use of cisatracurium improved adjusted 90-day mortality and increased ventilatory-free days. However, no statistically significant difference was seen in crude 90-day, 28-day, in-hospital, or ICU mortality rates. Furthermore, whether cisatracurium or heavy sedation caused these differences could not be concluded as all patients were heavily sedated.

The second multicenter trial was ROSE in m019 [71]. See Table 4. It randomized patients with moderate to severe ARDS (i.e., PaO2/FiO2 ≤ 150 mmHg) with PEEP of ≥8 cm H2O to receive cisatracurium vs. placebo randomly, but the trial stopped early after enrolling 1006 patients due to futility. ROSE 2019 concluded that compared to placebo, cisatracurium did not significantly reduce 90-day in-hospital mortality (42% cisatracurium vs. 43% control; P = 0.93). Furthermore, secondary outcomes at day 28 (e.g., in-hospital mortality, ventilatory-free days, and ICU-free days) also showed no significant difference. Moreover, the cisatracurium group had significantly more complications of ICU-acquired weakness and serious adverse cardiovascular events [71]. ROSE not only cast doubt on the beneficial effects of early NMBAs for patients with moderate to severe ARDS but also suggested potential harm.

	ROSE	ACURASYS
Prone ventilation	15.8%	44.8%
Sedation targets	Light sedation	Heavy sedation
PEEP strategies	Higher	Lower

Table 4.

Differences between ROSE and ACURASYS trials [70, 71].

Abbreviation: PEEP = positive end-expiratory pressure.

European Society of Intensive Care Medicine (ESICM) published its guidelines in 2023 and recommended against the routine use of NMBAs in moderate-t0-severe ARDS [72]. ESICM noted that given differing ventilatory approaches between the two trials and heavy use of sedation in ACURASYS, NMBA cannot be recommended in ARDS patients.

9. Extracorporeal membrane oxygenation (ECMO)

Despite the other evidence-based modalities mentioned in this chapter, severe ARDS is still associated with high mortality [73]. In the last decade, with the widespread adoption of ECMO around U.S., it has become an intervention that is often considered for patients with severe ARDS. However, lack of universal availability of ECMO and limited circuits of ECMO at ECMO centers has prevented ECMO from becoming a routine standard of care for patients with severe ARDS.

COVID-19 pandemic has accelerated the use of ECMO [74]. The goal of implementing ECMO in ICU care is to provide advanced mechanical life support to patients with acute respiratory or cardiac failure [75]. This support is achieved by draining deoxygenated blood from the venous system via cannulation, pumping the blood through a membrane that facilitates oxygen and carbon dioxide exchanges, and returning the newly oxygenated blood to the body for circulation and organ perfusion. The two most utilized ECMO cannulation strategies are venovenous (VV) and venoarterial (VA), which provide isolated respiratory support and combined cardiopulmonary support, respectively [75]. In the setting of ARDS, VV ECMO is the strategy that is commonly implemented. Because of this, the current landmark trials evaluating implementation of ECMO in ARDS have focused on VV cannulation (Figures 2 and 3).

Figure 2.

Common cannulation strategy for veno-venous ECMO [75].

Figure 3.

Common cannulation strategy for veno-arterial ECMO [75].

9.1 The ECMO trials

CESAR (Conventional Ventilation or ECMO for Severe Adult Respiratory Failure) trial showed a statistically significant improvement in survival rates by 33–35% without disability in the group evaluated for ECMO when compared to conventional ventilation [76]. The other trial, EOLIA (ECMO to Rescue Lung Injury in Severe ARDS) randomized patients to conventional ventilation or VVECMO in severe ARDS [77]. It showed no mortality benefit. However, multiple secondary outcomes did show significant improvement with the early ECMO intervention group. These outcomes included ICU length of stay, days free from mechanical ventilation, and days free from renal replacement therapy. In the intention-to-treat analysis, mortality at 60 days was 35% (44 out of 124) in the ECMO arm and 46% (57 out of 125) in the control arm; a relative risk of 0.76 (95% CI 0.55–1.04) with p-value of 0.09. This meant an absolute risk reduction of 11% in the ECMO arm, but this did not reach statistical significance [77]. A post hoc analysis of EOLIA trial showed mortality benefit, but the range of benefit varied depending on the analysis [78].

Subsequently, two meta-analyses comparing the two trials showed mortality benefit for VVECMO in severe ARDS and one meta-analysis showed improvement in cardiac, renal, and neurological dysfunction (Table 5) [81, 82].

CESAR trial	EOLIA trial
Issues with blinding and standardization of treatment protocols. The control group did not receive uniform treatment, leading to variability in care. Some participants referred for ECMO did not receive it, affecting the study’s validity [79].	Underpowering of the study. High crossover rate to ECMO in control group may have affected outcomes. Variability in ECMO implementation across centers raised concerns about consistency [80].

Table 5.

Criticisms of landmark ECMO trials [76, 77].

9.2 ESICM guidelines for ECMO in ARDS

In 2023, ESICM gave updates to their 2017 clinical practice guidelines on the management of ARDS. In their discussion on ECMO’s role in ARDS, the society gave recommendations for selecting proper patient for ECMO [72]. The guidelines gave a strong recommendation for patients with severe ARDS not due to COVID-19 to be treated with ECMO with management mirroring the EOLIA trial. While noting a low level of evidence, they also extended this recommendation to patients with severe ARDS due to COVID-19 since a randomized control trial of ECMO in severe ARDS due to COVID-19 is unlikely to occur. This decision was guided by multiple observational studies in COVID-19 patients that suggested improved short-term survival with ECMO. However, the available data did not show significant differences in 0-to-90-day mortality (Table 6) [72].

Indication Criteria for ECMO in ARDS
Severe ARDS with PaO2/FIO2 < 80 mmHg for more than 6 hours
PaO2/FiO2 < 50 mmHg for more than 3 hours
pH <7.25 with PaCO2 > 60 mmHg for more than 6 hours

Table 6.

Indication criteria for ECMO in ARDS per ECISM, implemented directly from the inclusion criteria of the EOLIA trial [72].

10. Prognostication

Various prognostication models have been developed over the years. The predictive values of disease-, patient-, and management-related factors have been studied, yet none reign supreme. For example, the Murray Lung Injury Score, the PaO2/FIO2 ratio, and the Oxygenation Saturation Index [80]. Recently, two studies have tried to take another evidence-based approach to this.

In 2021, the Stratification for Identification of Prognostic Categories In the Acute RESpiratory Distress Syndrome (SPIRES) Score was formulated. SPIRES identified three variables: patient’s age, number of extrapulmonary organ failures, and the PaO2/FiO2 ratio assessed at 24 hours of ARDS diagnosis. Those variables predicted ICU mortality significantly better than many previously proposed models, such as the PaO2/FiO2 ratio alone or the SOFA assessment scale [80].

In 2023, the Predicting ICU Mortality in Acute Respiratory Distress Syndrome Patients Using Machine Learning: The Predicting Outcome and STratifiCation of severity in ARDS (POSTCARDS) Study was published. POSTCARDS used a machine learning program (MLP) and identified seven variables listed in Table 7 with the highest predictive accuracy for 24-hour ICU mortality performance was comparable to traditional methods [83]. The POSTCARDS study did note that two different MLP did not outperform the SPIRES scoring system (AUC, 0.91; 95% CI, 0.82–0.91; SN, 0.85: SP0.84).

7 Potential Predictors Identified in POSTCARDS
Age Cancer Immunosuppression Baseline & 24 hr Pplat	PaO2/FIO2 ratio Inspiratory plateau pressure Number of extrapulmonary organ failures

Table 7.

Predictors in POSTCARDS study [83].

11. Complication, Prognosis & Long-term Outcomes

11.1 Complications

Since ARDS patients often end up requiring mechanical ventilation and intensive care unit (ICU) admission, they are at risk for VILI, nosocomial infections, such as ventilator-associated pneumonia, critical care myopathies, venous thromboembolic events, stress ulcerations, decreased nutrition and delirium [84, 85, 86]. These complications are likely multifactorial in nature; however, they lead to high morbidity and mortality in these patients.

11.2 Mortality

The Lung Safe study directly observed the outcomes of patients who had ARDS. Patients had a median duration of 8 days of ventilation, ICU stay of 10 days and total hospital stay of 17 days [84]. Mortality was directly proportional to the severity of the ARDS with 35% mortality for mild ARDS increasing up to 46% mortality for patients with severe ARDS [84, 85, 86]. Comparing ICU patients, ARDS increased mortality rate by 15% compared to those who did not have ARDS in the ICU [87]. Mortality is also increased in patients with ARDS in low-income and middle-income countries when compared to high-income countries [88]. Most of the deaths in the first 3 days are related to the underlying cause and later deaths were mostly attributed to sepsis [88, 89].

11.3 Outcomes in survivors

Among the survivors of ARDS, the cardiopulmonary function often reaches back to baseline by 6 months after the initial lung injury. Patients are, however, usually left with new or worsening cognitive, psychiatric, and physical deficits [88]. Severity of ARDS and resulting hypoxemia is associated with increased risk of cognitive deficits, including executive reasoning, verbal reasoning, memory issues and attention deficits [88]. There is an increase in depression, anxiety, and post-traumatic stress disorder (PTSD) reported in survivors [90]. The physical deficits include reduced exercise tolerance and increased disabilities, are further complicated by muscle weakness [89, 90]. There are also reports of increased hospital readmissions for up to 40% of survivors, including a third who need ICU admissions [91]. Overall, ARDS survivors experience noticeable decline in quality of life. Growing evidence shows that survivors may benefit from multidisciplinary-led post-intensive care clinics with regular intervals of follow-ups to help optimize functional status of the patient [92]. Further research is warranted into methods to improve outcomes for these patients, including the benefits of recovery programs and support groups.

12. Conclusion

Acute respiratory distress syndrome diagnosis does not adequately describe what the syndrome is. While the patient is typically in respiratory distress due to this syndrome, the diagnosis usually means acute diffuse lung injury with imaging showing bilateral infiltrates, blood gas showing severe hypoxia and/or hypercapnia while the pathology shows diffuse alveolar damage. It can portend significant morbidity and mortality. As we touched on, regardless of how ARDS is defined, and definitions are evolving, there are certain golden rules for treatment. NIV, Proning, ventilation with LTVV while keeping P_plat < 30 mmHg and DP <15 mmHg, steroids, restrictive fluid usage and ECMO are the usual interventions available. Our understanding of ARDS and its pathophysiology is evolving. There is increasing attention being given to ARDS phenotypes. On this front, much research remains to be done. The management and understanding of ARDS remains a dynamic and complex topic, further necessitating a multidisciplinary approach with prudent application of evidence-based strategies. Future research is likely to shift the current paradigms dramatically.