Restrictive lung diseases (RLD) are a heterogeneous group of pulmonary disorders characterized by reduced distensibility of the lungs, impaired lung expansion, decreased lung volumes, and a diminished total lung capacity (1). Causes are intrinsic, including interstitial lung diseases, and extrinsic, due to limitations in chest wall movement, resulting in decreased thoracic compliance (1). In pediatrics, extrinsic causes predominate, and include genetic neuromuscular diseases (2), neuromuscular scoliosis associated with cerebral palsy (3–5), and obesity (6). As RLD can contribute to hypoventilation and pump failure of the diaphragm, when this form of chronic lung disease is exacerbated by concomitant acute respiratory infection, these patients are particularly prone to acute on chronic hypoxic and hypercarbic respiratory failure and need for invasive mechanical ventilation (2, 7, 8). Despite this knowledge and consensus regarding the need for respiratory care in this population, there is no guidance regarding what ventilator mode by which to invasively ventilate children with RLD.

Airway pressure release ventilation (APRV) was first described in 1987 as a mode of invasive pressure control mechanical ventilation utilizing an inverse inspiratory: exhalation time ratio, by which continuous positive airway pressure is maintained in the airways, with only short spontaneous ventilation releases throughout the respiratory cycle (9). This ventilator mode works physiologically by maintaining a higher mean airway pressure, recruiting available lung units with varying time constants, and allowing for the improvement of refractory hypoxemia (3, 4), APRV has become synonymous with the “open lung concept”, in which shear stress and atelectrauma are minimized in the mechanically ventilated patient by preventing the lung from fully de-recruiting (5, 10). These properties have helped to label APRV as a rescue mode of ventilatory support in moderate to severe pediatric acute respiratory distress syndrome (PARDS), in whom conventional modes of ventilation fail to improve hypoxemia, or in whom providers are concerned for ventilator-induced lung injury due to volutrauma or barotrauma (3, 6, 10). Despite knowledge regarding the physiology of APRV, and understanding the pulmonary mechanics of RLD, to our knowledge, use of APRV as a recruitment maneuver in mechanically ventilated children with RLD has not previously been described in the literature.

In our center, we have utilized APRV for patients with RLD due to neuromuscular conditions or truncal obesity receiving invasive mechanical ventilation. Anecdotally, we have seen an increase in both patient types in recent years. and observed an improvement in gas exchange and resolution of atelectasis within 24–48 h of transitioning to APRV. Therefore, we aim to test our hypothesis that APRV is beneficial as a recruitment strategy in the pediatric population with RLD. This is a retrospective study to evaluate the effect of APRV on ventilation and atelectasis in pediatric patients with RLD.

This is a single-center retrospective case series in a 28-bed pediatric intensive care unit (PICU). Patients were identified via the EPIC electronic medical record (EMR) from those patients admitted to the Nemours Children's Hospital, Florida Pediatric Intensive Care Unit (PICU) in Orlando, FL during the time period between March 1, 2013 and March 1, 2024. Patients were included if they fulfilled the following inclusion criteria: (1) between the ages of 0–18 years old, (2) restrictive lung disease (scoliosis or kyphosis, obesity, cerebral palsy, or genetic neuromuscular disease: e.g., spinal muscular atrophy or Duchenne muscular dystrophy), (3) diagnosed with acute or acute on chronic respiratory failure, (4) required invasive positive pressure ventilation for a minimum of 48 h, at least 24 h of which on APRV. Patients were excluded if diagnosed with PARDS.

A total of 18 patient encounters were identified that met inclusion criteria. In the neuromuscular group, there were 14 patient encounters, comprised of 11 unique patient encounters and 1 patient with 3 different encounters. In the obese group, there were 4 unique patient admission encounters. All patients were started on CMV (Control data) and were switched to APRV based on the clinical need for recruitment of atelectasis.

We divided the participants into two groups based on the causes of RLD physiology: the neuromuscular group and the obese group, as shown in Table 1. The obese group was older than the neuromuscular group, with an average age of 14 years (± 2.3) compared to 10.45 years (± 3.7); however, this difference was not statistically significant (p = 0.195). Demographic data indicated that height, weight, and body mass index (BMI) were significantly different between the two groups. The obese group was heavier and taller, with a weight of 127 kg (range: 94–169 kg) compared to 24.7 kg (range: 22.1–34.1 kg) in the neuromuscular group (p < 0.001). The average height for the obese group was 166.95 cm (±4.16 cm) vs. 112.72 cm (±15.2 cm) for the neuromuscular group (p < 0.001). The BMI for the obese group was 44.85 (range: 33.3–60.85) compared to 21.3 (range: 17.21–22.4) for the neuromuscular group (p = 0.003). Four patients in our neuromuscular group received continuous neuromuscular blockade. Three of the neuromuscular group were rotated to prone positioning during their intubation vs. one in the obese group. Bronchoscopy was performed to address mucous plugging and/or to obtain bronchoalveolar lavage samples in 7 patients in the neuromuscular group vs. one in the obese group (Table 1).

No differences were observed in other categorical variables between the two groups, including gender, race, initial blood gases, oxygenation, ventilation, and mode of ventilatory support. PICU LOS and LOS were not compared due to many confounding reasons. Our hospital does not have a step-down unit, and the majority of patients remain in the PICU due to shortage of acute care beds or due to their need of noninvasive ventilation or ventilation through tracheostomy, which prevents transfer. Individual patient-level data are provided in the Supplementary Materials (Supplementary Table 1).

Our case series of 18 patient encounters, 15 unique patients, showed that patients with RLD and with acute respiratory failure, without the diagnosis of PARDS, who were transitioned from conventional mechanical ventilation (CV) to APRV showed an increase in lung volume, decrease in atelectasis, and improvement in oxygenation within the first 24 h of transition. APRV is a ventilator mode that when employed in pediatrics, is most often utilized as a rescue modality to improve gas exchange in PARDS. Our study showed that for children with RLD and atelectasis impairing gas exchange and/or their ability to liberate from mechanical ventilation, APRV can be successfully utilized as a recruitment modality. Use of APRV in this patient population reduces atelectasis and thereby increases the amount of lung surface area able to participate in gas exchange. This results in improved oxygenation for up to 24 h after the introduction of APRV without a clinically significant worsening of existing ventilation impairment. Our study also demonstrates that ARPV is safe and well-tolerated, even in patients receiving neuromuscular blockade.

RLD in the pediatric age group is most often encountered in children with neuromuscular disease, either due to cerebral palsy or muscular dystrophies. It is often in association with neuromuscular scoliosis, which imposes additional restrictions. With the increase in childhood obesity, critical care providers are seeing more adolescents with morbid obesity, with RLD resulting from their chest wall and abdominal adiposity. Despite this increasingly common group of patients, children with neuromuscular disease still comprised the majority of our cohort.

Alveolar recruitment maneuvers are performed by respiratory therapists to reduce atelectasis and improve gas exchange in children with acute lung injury. Methods in the literature by which to transiently provide increased transpulmonary pressure for recruitment include sustained high-pressure inflation techniques, intermittent sigh breaths, and brief periods of increased PEEP (11). Criticisms of these recruitment maneuvers include the risk of barotrauma, air leak, and hemodynamic instability, therefore prone positioning has emerged as an attractive alternative. There is currently no consensus regarding the use of recruitment maneuvers in pediatrics, with the literature showing conflicting results and inconsistent survival benefits (12). Our study showed that patients with atelectasis due to RLD experienced on average a 14% reduction in atelectasis after 12–24 h of APRV, which they maintained after 48 h of APRV despite weaning the mean airway pressure. This suggests that recruitment is sustained with APRV, even after decreases in ventilatory support.

In terms of oxygenation, both PaO2 and the P/F ratio were significantly improved at T1, findings that were most significant in children whose RLD was attributable to obesity. While the OI also improved at T1, as expected, the improvement was just short of achieving statistical significance. A possible explanation for this finding is that in efforts to re-expand atelectatic lung, we may have provided more mean airway pressure than was needed, influencing the OI equation. We postulate that concomitant obstructive lung disease from mucous plugging and microatelectasis may explain why children with neuromuscular disease did not have the same improvement in oxygenation as obese children in whom this modality was used. Use of APRV did not significantly worsen ventilation impairment, which is important to note since this modality is sometimes criticized for its inefficient ventilation.

Complications occurred rarely in our cohort. One obese child developed pneumomediastinum, which was not clinically significant, and one neuromuscular child developed worsening hypercapnic respiratory failure, necessitating a change in ventilatory mode. A second neuromuscular child developed an air leak prior to being placed in APRV, which, importantly, did not preclude the decision to utilize this modality.

APRV is a safe, effective way to provide alveolar recruitment and lung volume optimization in children with RLD from both neuromuscular conditions and obesity, and thus, improves oxygenation. Our results suggest APRV should be considered for lung recruitment in patients with RLD and recurrent atelectasis delaying extubation attempts. In order to provide recommendations and clear guidelines as to its use in this population, however, additional high-powered prospective studies utilizing an APRV protocol are necessary. To overcome our limitations, we propose conducting a prospective case control feasibility study to evaluate the benefits of APRV over conventional mechanical ventilation in the pediatric RLD population in increasing ventilator-free days and shortening the PICU LOS.