The consensuses on adult ARDS (American-European Consensus Conference,9 Berlin10) and the Pediatric Acute Lung Injury Consensus Conference (PALICC and PALICC-2)11,12 both agree on the diagnosis of adult acute respiratory distress syndrome (ARDS). However, there are also some important differences in the diagnostic criteria for pediatric patients due to the particular characteristics inherent to the pediatric patient.

According to the most recent pediatric consensus,12 PARDS is defined as a set of symptoms of hypoxemia and radiological changes that appear within seven days after pulmonary or extrapulmonary disease. This includes cases of pulmonary edema that cannot be fully explained by heart failure, volume overload, and/or perinatal respiratory disease (Table 1). A recent consensus among adults13 categorizes patients into three groups: non-intubated (including high-flow oxygen therapy [HFOT]), intubated, and resource-poor scenarios. The utilization of S/F in the absence of PaO2/FiO2 (P/F) is proposed to assess oxygenation, and lung ultrasound can be employed for radiological imaging (Table 2).

The diagnosis of PARDS does not require bilateral involvement on the chest X-ray. Rather, it is characterized by the presence of new opacities in the lung parenchyma that are not explained by atelectasis or pleural effusions. The consideration of this clinical criterion has not increased the diagnosis of PARDS, as 87% of these patients exhibit bilateral infiltrates within three days.2 Conversely, chest radiography is susceptible to significant interobserver variability, even when interpreting the number of affected quadrants. Research indicates that patients with infiltrates in all four quadrants and a large intrapulmonary shunt (P/F < 100) have a high mortality rate.14 In summary, oxygenation levels offer a more objective measurement of severity in PARDS cases than radiological imaging.

The methodology used in the radiographic assessment of lung edema (RALE) score developed by Yan et al. in PARDS attempts to more objectively delineate the lung quadrants and parenchymal density.15 However, it remains imprecise.

The PALICC-2 definition has been shown to have equivalent mortality predictive capacity in pediatric patients as the Berlin definition. Its prognostic ability increases the longer the time elapsed since the onset of the disease.16 Therefore, it is an effective predictor of mortality on the third day (odds ratio [OR], 256.5, 95% confidence interval [95%CI]: 27.1–2.424, P < .001), with an oxygenation index (OI) cut-off value of 17 (AUC = 0.92 [95%CI = 0.83–1], S 0.97; specificity [E] 0.87; P = .0001).17

In terms of diagnostic accuracy, if a patient meets the PALICC-2 criteria, the probability of PARDS increases almost ninefold. Conversely, if the patient does not meet the established criteria, the probability decreases to less than 10%.18–20

Furthermore, the diagnosis of PARDS is accepted in patients with congenital heart disease or chronic lung disease, even if severity stratification using the hypoxemia criteria is not feasible.

This new category includes those patients receiving non-invasive nasal respiratory support in the context of non-invasive ventilation (NIV) with bi-level positive airway pressure (BiPAP), nasal continuous positive airway pressure (nasal CPAP), or HFO ≥ 1.5 l/kg/m or ≥30 l/m. This category permits the exclusion of the radiological criterion in settings with limited medical resources, provided that the other criteria are met.12

These are patients who receive oxygen to maintain SpO2 ≥ 88%, but do not meet the definition of PARDS or “possible PARDS”. This category is instrumental in evaluating disease progression and implementing prevention strategies.

It is essential to stratify patients based on severity at least four hours after the diagnosis of ARDS. Two prognostic categories are described: mild-moderate oxygenation index (OI) < 16 or severe OI ≥ 16.12

For patients on invasive mechanical ventilation (IMV), the OI should be calculated as F × mean airway pressure (MAP) × 100/P, and the oxygenation-saturation index (OSI) as F × MAP/S. For patients on non-invasive ventilation (NIV), the P/F ratio or S/F should be used. The selection of P/F for severity stratification is based on a study that compared intrapulmonary shunt measured with clinical parameters (P/F) and the translation into lung density on the CT scan of adult patients with ARDS.21 However, OI and OSI do not convey this meaning. MAP, included in both formulas, can be modified by several parameters that do not influence oxygenation, such as inspiratory time or the I:E ratio.22,23 Therefore, given that positive end-expiratory pressure (PEEP) and not MAP is associated with lung recruitment, and since it does not measure the degree of intrapulmonary shunt, the use of OI and OSI could be called into question. However, studies have been conducted using the P/F ratio as the gold standard for diagnosing ARDS. For example, a study by Thomas et al. reported a high accuracy of OI and the S/F ratio, although not of OSI.24

The use of the S/F ratio as a marker of lung damage is very useful in pediatrics, since sedation and cannulation of an arterial access can have a detrimental effect on the patient's baseline condition. Therefore, non-invasive monitoring is ideal in patients with mild-moderate involvement and subjected to NIV. Such measurement should be performed under stable conditions and not during transient desaturation episodes. To find the area of the hemoglobin dissociation curve where the linear relationship between P/F and S/F is found, we must adjust FiO2 to maintain SpO2 ≤ 97%.25,26

To date, no pharmacological treatment has demonstrated efficacy in the management of ARDS. The primary objective of respiratory support is to ensure adequate oxygen delivery to the tissues while preventing ventilator-induced lung injury (VILI).

The identification of the mechanical factors that induce greater lung injury has influenced the decrease in mortality. The type of support provided should align with the severity of hypoxemia (Fig. 1).

Figure 1.

Diagnostic and therapeutic scheme applicable to acute respiratory distress syndrome (ARDS).

ECMO: extracorporeal membrane oxygenation; RM: recruitment maneuvers; iNO: inhaled nitric oxide; P/F: partial arterial pressure of oxygen/inspired oxygen fraction; S/F: oxygen saturation measured by pulse oximetry/inspired oxygen fraction; HFOV: high frequency oscillatory ventilation; MV: mechanical ventilation; NIV: non-invasive ventilation.

Reproduced from Ref.80

At this time, there is no evidence to recommend pressure or volume ventilation in patients with PARDS. Similarly, high-frequency oscillatory ventilation is not recommended except when the ventilatory and oxygenation goals are not achieved after optimizing the lung protection strategy in conventional modalities.12

The application of optimal PEEP aims to keep as many alveolar units as possible “open” throughout the respiratory cycle.39 This strategy is aimed at reducing intrapulmonary shunt and ensuring sufficient gas exchange. PEEP titration should be performed on an individualized basis, with close monitoring of the effect on P/F, OI, compliance, and hemodynamics.40 At present, the table proposed in the ARDS Network protocol is recommended, where a PEEP level in relation to FiO2 is selected as a starting point (Table 3).12 It has been demonstrated that there is a higher mortality rate in children who received less PEEP than the amount indicated in this table. Conversely, a PEEP level that exceeded the recommended range by up to 5 cmH2O did not increase mortality.41

The Open Lung Strategy (OLS), which includes high PEEP levels and low tidal volume, is considered the gold standard for adult patients with ARDS. However, studies have not demonstrated a reduction in mortality with this strategy. The latter may be particularly advantageous for patients who have not been successfully recruited by conventional methods (P/F < 150) and for whom applying a higher PEEP results in reduced energy application to the lung (i.e., decreasing DP or MP).38

Although the level of evidence is low, ventilation with a tidal volume (Vt) of 6–8 ml/kg is recommended. It may be necessary to adjust Vt to below 6 ml/kg to avoid exceeding the Pplat or DP limits. However, it is important to note that volumes lower than 4 ml/kg have been associated with an increased mortality rate.27 The calculations should be based on the lowest weight (ideal or actual).12

In the adult population, studies have demonstrated that limiting Pplat significantly reduces mortality. A Cochrane systematic review indicates that for a Pplat below 32 cmH2O, there is no observed difference in mortality rates for ARDS. However, given the absence of pediatric studies in this regard, it is advisable to limit Pplat to 28 cmH2O, allowing up to 32 cmH2O in patients with low compliance, as well as DP < 15 cmH2O.42

Although the evidence is insufficient, the PALICC-2 guidelines recommend maintaining SpO2 ranges between 92%–97%. In critically ill patients, SpO2 levels between 88%–92% are considered acceptable, but central venous saturation and other indicators related to oxygen transport and consumption should be closely monitored.

With regard to the permissive hypercapnia strategy (with pH > 7.20), it may be necessary to maintain protective ventilation, and its application has been shown to reduce mortality in pediatric patients.3 Greater control is indicated in disorders such as intracranial hypertension, severe pulmonary hypertension, some congenital heart diseases, and/or patients with severe ventricular dysfunction and/or hemodynamic instability. Conversely, the administration of bicarbonate is recommended exclusively in severe metabolic acidosis or pulmonary hypertension with severe cardiac and hemodynamic involvement.43

In the case of adult patients with very severe ARDS (P/F 75), the use of extracorporeal membrane oxygenation (ECMO) in the first 48 h has been shown to significantly reduce the probability of death or severe disability at 6 months. The number of patients needed to treat (NNT) to achieve this outcome is four (relative risk [RR] 0.65; 95%CI: 0.52–0.80) (Table 5). In addition, it has proven to be cost-effective. To date, pediatric studies on mortality in ARDS patients undergoing ECMO, although limited, show scientific evidence of its benefit.44

In children, there is currently no evidence to support the use of strict criteria for selecting patients who will benefit from ECMO. Veno-venous (VV) ECMO should be considered as a rescue strategy for patients with refractory hypoxemia, following a structured evaluation by an expert team, when other treatment options have failed, including fluid restriction, prone positioning, and protective ventilation.45

Research has demonstrated that patients receiving VV ECMO exhibit enhanced hemodynamic stability and reduced reliance on vasoactive medications.46 Therefore, the choice of the type of assistance should be based not only on the necessity for vasoactive drugs47 but also on the presence of cardiac dysfunction. In the event of recovery, early conversion to VV ECMO is a viable option.48 During ECMO therapy, close monitoring is essential, and the necessity for veno-arterial (VA) ECMO should be assessed in cases of right ventricular failure.49 The survival rate is higher in patients undergoing VV ECMO compared to VA ECMO.50

The most common complications of VV ECMO are bleeding (29.3%), neurological complications (7.1%), including intracranial hemorrhage, ischemic stroke, brain death, and seizures.47

In the initial stages of extracorporeal support, it is essential to avoid hyperoxia and instead induce a gradual decrease in the partial pressure of CO2. This approach is crucial to prevent neurological complications, as abrupt decreases have been associated with an increased risk for such complications.47 The initiation of ECMO typically results in a reduction of ventilator parameters to protective levels. Even when adequate support is provided, extubation of the patients is considered as they recover from pulmonary complications. Pediatric studies have reported increased mortality in those patients with greater FiO2, Pplat, and DP or with Vt < 4 ml/kg during ECMO support.51,52 Therefore, lung protection measures should be maintained (Pplat < 28–32 cmH2O, DP < 15 cmH2O, and general reduction of MP).53

A recent study has indicated that the prone position during ECMO does not appear to offer any advantages in terms of reducing the duration of support when compared to the supine position.54

Prone ventilation is regarded as a primary strategy for severe ARDS in adult patients. It is a lung protective measure that enhances compliance, gas exchange, and overall lung mechanics. If performed within 36 h of diagnosis and for a minimum of 16 h per day, there is a reduced risk of mortality.54

While the data are insufficient in pediatrics, its use is considered to optimize recruitment in patients who could not be recruited by other means (P/F < 150).55 Given the current evidence, it is not possible to specify the length of time the pediatric patient should remain prone.56

Improvement of the ventilation-perfusion ratio has been observed in patients with PARDS undergoing prone ventilation and studied using electrical impedance tomography.57 However, the temporary enhancement in oxygenation did not result in a substantial decrease in mechanical ventilation duration.58–60 A multicenter study is currently underway to clarify the benefits of this practice.61

There is insufficient evidence to recommend the application of recruitment maneuvers (RM) in pediatrics. Recruitment maneuvers consist of transient increases in insufflation pressure to open collapsed alveolar units, followed by fine-tuning of PEEP to the optimal level (P/F > 150–175), since reaching higher P/F values may increase MP and, therefore, mortality.

In adult patients with diffuse lesions, a tendency to hyperinsufflation of well-ventilated areas has been observed in those with patchy lesions.62 Conversely, deleterious effects were observed with RM with progressive increases in PEEP of 5 cmH2O.63

In selected pediatric patients with severe PARDS, slow incremental or decremental PEEP changes (typically PEEP steps of 2 cmH2O) can be made while continuously monitoring the patient's oxygenation, compliance, and hemodynamics. It is not advisable to implement sustained insufflation maneuvers in children. To date, there are no studies that establish the mode of application, safety, and usefulness of RM. The implementation of point-of-care ultrasound has the potential to enhance the safety of these maneuvers.64

The PALICC considers that inhaled nitric oxide (iNO) cannot be routinely advised in children with ARDS but can be recommended in selected cases. It produces a transient improvement in oxygenation, but does not reduce mortality, and increases the risk of renal failure.65

The concentrations of iNO reaching the alveolus depend on two factors: the type of device used for administration and the mode of delivery of the gas into the lungs.66

An iNO test is acceptable when there is suspicion of other shunt mechanisms contributing to poor oxygenation (altered hypoxic pulmonary vasoconstriction, pulmonary hypertension in patients with patent foramen ovale). These cases are very prevalent in pediatrics (bronchiolitis, viral pneumonias, etc.).67 The use of iNO is also considered in patients with documented pulmonary hypertension or severe right ventricular dysfunction.12

A pediatric clinical trial was unable to demonstrate improvement in mortality; however, the use of iNO reduced the need for ECMO in PARDS.68 If it is not effective, it should be withdrawn within 4 h in order to avoid side effects.

The application of scales to monitor pain, sedation, withdrawal, and delirium is recommended as a good clinical practice standard.69

The use of minimal effective sedation is crucial to ensure the successful implementation of the ventilatory strategy. Drug dose titration should be performed with specific objectives, such as facilitating mobility and some activity, while also ensuring the promotion of quality sleep during the nighttime hours. In regard to delirium, it is essential to perform daily assessments utilizing validated pediatric scales. It is crucial to note that delirium cannot be detected in states of deep sedation (State Behavioral Scales ≤ 2 points).69

The administration of muscle relaxants is recommended if adequate sedoanalgesia is insufficient to maintain protective ventilation. The dosage should be titrated, administering the minimum effective dose, and if possible, by means of the train of four.69

For patients with PARDS, the intubation process should be performed electively using cuffed endotracheal tubes (ETT) to prevent de-recruitment.55 ETT with subglottic secretion drainage, from size 6.0 and above, is an option for patients over 10 years of age.70

There is no difference in morbidity and mortality with respect to the choice between programmed or necessary aspirations. It is imperative that suction catheters be positioned such that they occlude less than 70% of the ETT lumen. Furthermore, it is crucial to maintain suction pressure below −120 mmHg. Deep aspiration should be performed only when superficial suctioning proves ineffective. The FiO2 required for preoxygenation should be titrated, and saline instillation should be avoided.71

Although there is little evidence to support it, a pediatric clinical trial found that restricting fluid during the resuscitation phase and reducing total body fluid balance thereafter improved the number of days without mechanical ventilation (MV-free days) compared to the liberal strategy.72 Furthermore, volume overload ≥ 5% within the first 24 h of admission has been associated with a higher risk of acute renal failure and prolonged IMV.73

It is important to note that different ARDS phenotypes may respond differently to fluid management, and focused studies are needed to determine the most effective approach.74

Given its protective role, the start of enteral nutrition within the first 72 h is recommended. In addition, the administration of at least 1.5 g/kg/day of protein per day is necessary to ensure a positive protein balance. Similarly, it is advisable to maintain a nutrition plan that addresses metabolic requirements, promotes recovery, and facilitates sustained growth.75

Transfusion is not recommended if the patient has a hemoglobin level of at least 7 g/dl, unless there is hemodynamic instability, baseline chronic cyanotic disease, severe ARDS, or hemolytic anemia.75

For adult patients with ARDS due to SARS-CoV-2, corticosteroid treatment is considered a standard treatment option.76 In the field of pediatrics, there is currently no evidence to support the use of corticosteroids or surfactant administration in cases of PARDS.77