Year in Review 2015: Pediatric ARDS

Pediatric ARDS: Consensus

PALICC was a 2-year process that consisted of 27 experts from 8 countries on 3 continents (North America, Europe, and Australia). This interdisciplinary group published their recommendations for the definition of pediatric ARDS in 2015 (Table 1). Some of the key PALICC recommendations included: (1) no age criteria for the definition of pediatric ARDS to better understand the pathophysiology of pediatric ARDS across the spectrum of age groups in future studies, although perinatal-related lung injury was excluded; (2) stratification of the severity of lung injury based on an oxygenation deficit, as defined by the oxygenation index (OI) or oxygen saturation index, if an arterial blood gas was not available; (3) inclusion of infants and children requiring noninvasive ventilation as well as those with congenital heart disease and chronic lung disease; and (4) inclusion of patients with unilateral lung disease.

The severity of pediatric ARDS was based on OI/oxygen saturation index criteria, with pediatric ARDS being described as mild (OI = 4–8), moderate (OI = 8–16), or severe (OI > 16).¹⁵ It should be noted that, similar to the Berlin definition,³ PALICC eliminated the previous category of “acute lung injury.” The recommendation to allow S_pO2 criteria to define pediatric ARDS was made with the caveat that the F_IO2 be titrated to achieve S_pO2 of 88–97%. PALICC included children receiving noninvasive ventilation (minimum CPAP/expiratory positive airway pressure of 5 cm H₂O) in the pediatric ARDS definition by use of the P_aO2/F_IO2 or S_pO2/F_IO2 ratio to recognize the increasing use of noninvasive ventilation and allow for the possibility of earlier intervention for those with pediatric ARDS. Similar to the prior adult-based definitions, PALICC included the onset of pediatric ARDS within 7 days of a known clinical insult as well as the presence of respiratory failure not fully explained by fluid overload and/or cardiac failure.

PALICC went beyond defining pediatric ARDS and offered 151 recommendations in a series of articles, also published in June 2015, which covered a wide range of topics, including invasive ventilatory support, noninvasive ventilation, pulmonary-specific ancillary approaches, non-pulmonary treatments, respiratory monitoring, extracorporeal support, and others.^16–21 Also offered by PALICC was a comprehensive discussion of the epidemiology of pediatric ARDS as well as its associated morbidities and long-term outcomes.^22,23 The reader is directed to the references provided because a comprehensive discussion of the PALICC recommendations is beyond the scope of this review article.

The PALICC guidelines were developed based on peer-reviewed, published data as available. When pediatric data were not available, data from adult patients with ARDS and/or neonates with acute lung injury were used. In the event that no data were available, expert opinion was utilized. The common lack of pediatric ARDS data is probably related to the challenges in conducting randomized clinical trials in pediatric subjects with ARDS, including a relatively low overall incidence of pediatric ARDS, heterogeneity in physiology and patient age, subjectivity of the definition of pediatric ARDS (before PALICC), and a lack of consensus for standardized approaches for mechanical ventilation strategies and adjunctive therapies.

If accepted by the pediatric critical care community, the PALICC definition of pediatric ARDS will probably facilitate prompt recognition and diagnosis of pediatric ARDS in clinical practice, improve prognostication and stratification of disease severity to help guide management strategies, and aid in the design of randomized, controlled trials. Multi-center data will be required to validate (or refute) the 2015 PALICC pediatric ARDS definition across all pediatric age groups as well as the management recommendations offered.

Sedation Management

One of the most challenging aspects of mechanical ventilation for the pediatric patient with acute lung injury is sedation management. On the one hand, pharmacologic sedation is generally required to help ensure the safety and comfort of mechanically ventilated infants and children. On the other hand, the clinician must minimize the adverse effects of such medications, which may include limiting spontaneous ventilation, prolonging the course of mechanical ventilation, inducing critical illness neuromyopathy, increasing the risk for delirium, and causing the development of iatrogenic withdrawal syndromes when sedation is no longer necessary.^24–26

The RESTORE (Sedation Management for Pediatric Patients with Acute Respiratory Failure) study⁹ was a multi-center cluster-randomized clinical trial that studied the effect of a nurse-implemented, goal-directed sedation management protocol on clinical outcomes in infants, children, and adolescents with acute respiratory failure requiring invasive positive-pressure ventilation. In this largest randomized controlled trial in pediatric acute respiratory failure, 2,449 subjects were enrolled, 50% from intervention sites and 50% from control sites. Protocolized sedation did not reduce the length of mechanical ventilation; however, those in the intervention group demonstrated a greater number of awake days and fewer days of opioid administration as well as fewer categories of sedatives prescribed. An increase in adverse events compared with subjects receiving usual sedation management was not seen despite these subjects being more awake.

Although a negative study from the perspective of the primary end point, RESTORE⁹ is important to the pediatric critical care community. The safety of a bedside-driven sedation protocol was demonstrated. More importantly, it showed that mechanically ventilated pediatric subjects can safely be maintained in a more awake state. Such an approach may lead to a reduction in the occurrence of iatrogenic withdrawal because the intervention group had a significantly lower need for methadone administration. Last, it should be mentioned that the RESTORE algorithm did not include dexmedetomidine, which has become more commonly used in the pediatric critical care environment since the initiation of the RESTORE investigation.^27,28 Further study of a bedside-driven sedation protocol that includes the use of dexmedetomidine should be considered.

High-Frequency Oscillatory Ventilation

The OSCAR⁵ and OSCILLATE⁶ trials of high-frequency oscillatory ventilation in adult subjects with ARDS have shed a negative light on this ventilatory modality. The OSCAR trial reported no difference in all-cause mortality between HFOV- and conventional mechanical ventilation-managed subjects.⁵ The OSCILLATE trial was prematurely discontinued due to increased mortality in the HFOV group.⁶ However, with a void of definitive pediatric data, clinicians have been left with a difficult decision regarding the use of HFOV for pediatric ARDS. A recent publication on HFOV use for pediatric ARDS helps to shed new light on this topic⁷ by using prospectively collected data from the RESTORE study.⁹ The RESTORE HFOV study compared the length of mechanical ventilation in pediatric subjects managed with early HFOV (initiated within 24–48 h of intubation) as compared with those receiving conventional ventilation or late HFOV by means of a propensity score analysis.

Of the 2,449 subjects enrolled in RESTORE, 353 (14%) were managed at some time during their ventilatory course with HFOV. Two hundred ten (59%) had HFOV initiated within 24–48 h of intubation. The propensity score model predicting the probability of receiving early HFOV included 1,064 subjects (n = 181 early HFOV vs n = 883 conventional mechanical ventilation/late HFOV) with moderate to severe pediatric ARDS based on the PALICC criteria (OI ≥ 8). The most significant contributor to the propensity score model was the degree of hypoxia. After adjusting for risk category, early HFOV was associated with a longer duration of ventilation but not mortality as compared with those managed with conventional mechanical ventilation/late HFOV.

An important question remains: Is the increased length of ventilation with early HFOV use as seen in the RESTORE HFOV study related to HFOV per se or to the manner in which this approach is employed? On the one hand, low-tidal volume ventilation during conventional mechanical ventilation, as has been increasingly employed over recent years, may simply be better than HFOV. However, it must be noted that the general management approaches to mechanical ventilation may differ with HFOV as compared with conventional mechanical ventilation. Specifically, clinicians seem less hesitant to wean conventional ventilator settings than mean airway pressure during HFOV based on a fear that lung de-recruitment will occur in the HFOV setting. Additionally, endotracheal tube suctioning is generally performed less frequently with HFOV with uncertain sequelae.

The HFOV RESTORE study seems to have raised more questions than it has answered. However, the data provided may help in the design of future studies of HFOV for pediatric ARDS. In the meantime, one may speculate that HFOV has a role in the management of pediatric ARDS at the time of peak lung injury followed by aggressive weaning as lung recovery occurs and an earlier transition back to conventional ventilation than has traditionally occurred.

Delirium

Delirium is the behavioral manifestation of acute cerebral dysfunction associated with an underlying medical illness and presents with acute, fluctuating changes in a patient's mental status. This clinical entity has been well-reported in the adult critical care environment and is associated with increased mortality, morbidity (both short- and long-term), and health-care costs.^29–31 Recently, increased recognition of this important clinical entity has occurred in the pediatric critical care setting.^8,32

In a 2015 single-center observational study,⁸ the incidence of delirium in the pediatric ICU setting was demonstrated to be 21%. Risk factors were identified as need for mechanical ventilation, developmental delay, and age between 2 and 5 y. Beyond the need for additional research to better characterize pediatric delirium and identify additional predisposing risk factors is the need to learn more in regard to optimal treatment options. In a retrospective review, quetiapine,³² an atypical antipsychotic that has been used for the treatment of delirium in the adult critical care setting, was shown to be safe in a cohort of pediatric subjects with ICU delirium. Further investigation, preferably in a prospective fashion, is clearly needed.

Noninvasive Respiratory Support

Much attention has been paid to noninvasive ventilation in the pediatric population over the past couple of years. Unfortunately, despite the rapid growth in the use of noninvasive ventilation in the pediatric population, definitive data to support its use remain limited. A recent pilot, prospective, randomized clinical trial¹⁰ comparing noninvasive ventilation and standard oxygen therapy in children <3 y of age with respiratory failure after extubation showed no difference between the groups; however, the study was limited by a small sample size and a large number of excluded subjects. This report is important to note because it demonstrates the existence of equipoise (despite the rapidly growing use of noninvasive ventilation) and the ability to study noninvasive ventilation in a randomized, controlled fashion.

Similar to noninvasive ventilation, the use of high-flow nasal cannula continues to increase in the pediatric critical care environment, again without definitive data. Although pediatric studies are limited, it is important to note a recent publication of high-flow oxygen therapy in acute hypoxemic respiratory failure in adult subjects.¹¹ Three hundred ten subjects with acute hypoxemic respiratory failure, defined as a P_aO2/F_IO2 ratio of ≤300, without hypercapnia, were randomly assigned to high-flow oxygen via a nasal cannula, standard oxygen delivery by a face mask, or noninvasive ventilation. Intubation rates among the 3 groups did not differ; however, ventilator-free days at day 28 were higher, and mortality at 90 d was lower in the high-flow nasal cannula group. Although important findings, it should be noted that the primary end point did not show a difference, and this study was limited to those ≥18 y old. A systematic investigation of the use of high-flow nasal cannula support for acute respiratory distress/failure in the pediatric population is needed.

Lung-Protective Ventilation

Although lung-protective ventilation has been a focus of ARDS management since the ARDS Network publication on low-tidal volume ventilation in 2002,³³ there has been much discussion regarding the more important mechanism for ventilator-induced lung injury: barotrauma (ie, elevated peak/plateau airway pressures) or volutrauma (ie, increased tidal volume). Although pediatric data are only observational,^34–36 it is important to review the recent publication by Amato et al¹² describing the association between the driving pressure and survival in adults with ARDS. Among ventilation parameters, the drive pressure (difference between peak inspiratory pressure and positive end-expiratory pressure) was most strongly associated with survival. A 1-SD increment in drive pressure (approximately 7 cm H₂O) was associated with increased mortality with a relative risk of 1.4 (CI 1.31–1.51, P < .001). Although the pediatric practitioner must be cognizant of the inherent physiologic and anatomic differences between the adult and pediatric lung, the findings of this study cannot be ignored. Until more definitive pediatric data are available, one should consider reducing drive pressure when ventilating the pediatric patient with ARDS.

Adjunct Therapies for the Management of ARDS

PALICC provides a comprehensive review of the current state of pulmonary and non-pulmonary adjunct therapies for the management of pediatric ARDS.^16,17 Of particular interest are 2 studies from 2015 that focus on corticosteroid administration and use of exogenous surfactant.

Yehya et al¹³ performed a single-center, observational study in a prospectively enrolled cohort of children with pediatric ARDS. The authors demonstrated that corticosteroid exposure for >24 h was independently associated with fewer ventilator-free days at 28 d and a longer duration of mechanical ventilation even after adjustment for the primary potential confounding variables. It must be noted that as an observational study, it is impossible to distinguish cause and effect from association because confounding variables are likely. This study should be viewed as hypothesis-generating for future prospective, randomized investigations rather than definitive.

In yet another study, the administration of exogenous surfactant was found to have no benefit for adult subjects with ARDS.¹⁴ The findings of the adult CARDS (Calfactant in Acute Respiratory Distress Syndrome) trial are similar to those of the previous pediatric surfactant study.³⁷ In both studies, exogenous surfactant administration did not improve oxygenation or outcome (hospital stay or survival). At this point, it would seem reasonable to conclude that exogenous surfactant cannot be recommended for routine clinical use for ARDS in either the adult or pediatric populations.

Summary

In 2015, the medical literature was greatly augmented with regard to the field of pediatric ARDS. The advance of available information was largely led by the work of PALICC as well as the RESTORE investigators. Despite the recent progress, additional investigation in each of the areas discussed in this review article is essential to the continued advancement of our knowledge and, more importantly, improvements in the outcome for pediatric patients with ARDS. It is hopeful that the publications discussed here as well as the topics of research proposed by PALICC will stimulate further basic science, translational, and clinical investigations to continue the advancement of the fields of pediatric respiratory care and pediatric critical care medicine.