Mechanical ventilation (MV) is essential for children with acute respiratory distress syndrome (ARDS). However, it may also contribute to lung inflammation and injury via ventilator-induced lung injury (VILI) [1]. Barotrauma, volutrauma due to alveolar overdistension, and atelectrauma following ventilation at low lung volumes all contribute to VILI. Understanding these mechanisms has resulted in lung protective ventilation strategies including pressure limitation, delivery of low tidal volume (V t), and application of positive end-expiratory pressure (PEEP) in adults with ARDS.

Pediatric intensivists have readily adopted these practices. However, many uncertainties remain. Numerous differences in physiological systems between children and adults may invalidate extrapolation of adult practice to children. In addition, the susceptibility to VILI may differ with age [2]. Specifically the association between V t and mortality is unclear in pediatric ARDS [3]. Hence, studies need to be undertaken to understand and improve the management of pediatric ARDS, during both the acute and weaning phases (Fig. 1). Various compelling issues need to be addressed, such as a better definition of pediatric ARDS, individualized titration of ventilator settings, and the role of non-invasive ventilation (NIV) or high frequency oscillatory ventilation (HFOV).

Fig. 1
figure 1

Schematic representation of mechanical ventilation (MV) course in pediatric acute respiratory distress syndrome (PARDS). There is a need to better define the PARDS diagnosis criteria and use of the ventilation modes and monitoring methods available. NIV non-invasive ventilation, HFOV high frequency oscillatory ventilation, CMV conventional mechanical ventilation, EaDi electrical activity of the diaphragm, EIT electrical impedance tomography, TPP transpulmonary pressure

The incidence and mortality of pediatric ARDS differ from those in adults. Pediatric ARDS is relatively rare at 2–12.8 per 100,000 persons per year. Mortality rates are between 18 and 35 % [4]. Despite this low incidence, Spanish investigators have shown that a substantial number of ventilated children may develop ARDS during their stay in the pediatric intensive care unit (PICU) [4]. To date the North-American European Consensus Conference (NAECC) definition has been used to identify children with ARDS with the inherent risk of underestimating its true incidence, especially because it requires an invasive marker of oxygenation (PaO2) [5]. The Berlin definition of ARDS has replaced the NAECC definition but does not include specific pediatric considerations. However, the Section Respiratory Failure of the European Society for Pediatric and Neonatal Intensive Care (ESPNIC) have since demonstrated good performance of the Berlin definition among infants younger than 24 months [6]. More recently, the Pediatric Acute Lung Injury Consensus Conference (PALICC) brought together investigators from all over the world with the aim of improving the definition of pediatric ARDS, its management, and providing directions for further research [7]. This new definition is based on analysis of observational data from various pediatric studies, taking primarily into account the association between hypoxemia/lung injury severity scoring and outcome for pediatric ARDS [8]. A better taxonomy of pediatric ARDS will most likely result in a better patient selection for randomized controlled trials.

Specific pediatric guidelines on how to ventilate a child with ARDS are lacking. As a consequence, pediatric critical care physicians have adapted to varying degrees lung-protective ventilation strategies based on adult recommendations. In general V t is targeted in the range of 5–7 mL/kg and a certain level of PEEP is applied at the discretion of the attending physician [9]. The question is if such a ‘one size fits all’ approach makes sense. The physiologic V t is in the range of 6–8 mL/kg body weight, but patients with more severe lung injury have a smaller residual inflatable lung available for alveolar ventilation (i.e., the baby lung). There is no specific threshold for V t associated with mortality in pediatrics [3]. Hence, the optimal V t might need to be smaller in severe cases, but perhaps higher in less sick or improving lungs. Thus, the underlying disease and severity of lung disease need to be taken into account instead of targeting 6 mL/kg predicted body weight (PBW) in every patient. This more physiologic approach has not been tested in clinical trials but does make sense. A subgroup analysis of the ARDS Network trial showed that randomization to 6 mL/kg PBW was only beneficial in patients with poor respiratory system compliance at study entry [10]. Likewise, pediatric patients with higher lung injury score managed with pressure-limited ventilation displayed lower V t [11].

Simple bedside tools to titrate ventilation are much needed, such as transpulmonary pressure (TPP) measurements and electrical impedance tomography (EIT). The esophageal pressure may be used as a surrogate for the pleural pressure and therefore be used to set PEEP. TPP-guided PEEP titration resulted in improved oxygenation and compliance in adults with ARDS [12]. EIT is a non-invasive imaging technique displaying real-time changes in ventilation. It provides information on which parts of the lung the V t is distributed to, and allows for a quantification of regional overdistension and atelectasis. It may thus be used to determine the appropriate V t and level of PEEP. EIT-guided PEEP titration resulted in improved respiratory mechanics, improved gas exchange, and reduced histologic evidence of VILI in an animal model of acute lung injury [13]. Pediatric clinical studies are awaited.

Other areas in need of study in pediatric ARDS include neutrally adjusted ventilator assist (NAVA). NAVA might result in better oxygenation, less sedation, and less patient–ventilator asynchrony. Its use is gaining interest in pediatric critical care as discussed recently in this journal [14]. However, the PALIVE study showed that less than 10 % of pediatric ARDS is managed with NIV [15]. This can most likely be explained by the fact that the efficacy of NIV in pediatric ARDS is unknown. HFOV is often used when conventional ventilation fails. However, the safety and benefit of HFOV are questioned following the OSCILLATE trial in adult ARDS and a retrospective observational pediatric study [16, 17]. Hence, a pediatric HFOV trial is eagerly awaited. Finally, we do not know the optimum weaning strategy for children with ARDS. A big question is whether we should use clinical decision-making protocols to decrease practice variation and standardize patient care. Recently, a small pediatric study showed that implementation of such a protocol reduced weaning duration by 1.5 days [18].

In summary, many uncertainties remain concerning the ventilation of pediatric ARDS. This should encourage investigators worldwide to join forces and address these questions. It is absolutely clear that we need to improve the scientific basis of one of the most practiced interventions in pediatric critical care. The challenging research agenda for the next decades is set.