Abstract

Objective

To evaluate the physiological rationale and clinical impact of a protocolized rescue ventilation strategy targeting chronic lung mechanics in infants with established severe bronchopulmonary dysplasia (sBPD) who failed to respond to low-dose dexamethasone therapy (DART).

Method

This retrospective, exploratory case series included infants with sBPD who showed no improvement in oxygenation 72 hours after initiating the regimen low-dose DART. Identifying a mismatch between conventional ventilator settings and the pathophysiology of sBPD, we transitioned patients to a physiology-based strategy using volume-guarantee combined with pressure support ventilation: Low respiratory rates (20-25/min), prolonged inspiratory times (≥0.8 s), and higher tidal volumes (10-12 mL/kg).

Results

Seven infants [median GA 25 (24-26) weeks; median birth weight 680 (590-810) g] met the criteria for steroid refractoriness. Baseline PaO₂/FiO₂ was critically low [median 66 (61-70) mmHg]. Following the physiological reset, PaO₂/FiO₂ markedly improved to a median of 110 (90-140) mmHg by day 14. Ventilatory efficiency improved despite lower rates, as reflected by a reduction in median pCO₂ from 55 (54-63) to 45.5 (45-47) mmHg. All subjects were successfully weaned from invasive ventilation.

Conclusion

In steroid-refractory sBPD, respiratory failure may be driven by heterogeneity in mechanical time constants rather than by active inflammation alone. A ventilation strategy strictly aligned with these mechanics resulted in a reduction in respiratory failure and facilitated extubation, warranting prospective validation.

Introduction

Bronchopulmonary dysplasia (BPD) affects nearly 20% of extremely preterm infants, and established severe BPD (sBPD) remains a primary driver of long-term morbidity (1). As the disease progresses from the acute phase to a chronic phenotype, the lung undergoes a profound physiological transition. The mechanics of established sBPD exhibit marked heterogeneity: Hyperinflated areas coexist with atelectasis, airway resistance rises significantly, and respiratory time constants become pathologically prolonged (2, 3).

Despite these well-documented changes, clinicians often manage sBPD using ventilatory strategies extrapolated from acute respiratory distress syndrome (ARDS), specifically rapid rates and low tidal volumes. While this approach is protective during the acute phase, it often becomes maladaptive in the chronic phase (4). High respiratory rates in lungs with long time constants prevent complete exhalation, leading to air trapping and failure to ventilate “slow” alveolar compartments, which ultimately worsens ventilation-perfusion (V/Q) mismatch (3).

Systemic corticosteroids, particularly the low-dose dexamethasone (DART) regimen, remain a standard therapy to facilitate extubation (5-7). However, a distinct subset of infants fails to respond to this treatment. We hypothesize that in these “steroid-refractory” patients, active inflammation is no longer the primary driver of respiratory failure. Instead, the failure likely stems from a severe mismatch between conventional ventilator settings and the lung’s structural abnormalities, which include fibrosis, acquired tracheobronchomalacia, and fixed airflow obstruction (8, 9).

In this study, we present a physiology-based rescue ventilation protocol tailored to overcome these specific mechanical barriers. We evaluated this strategy in a small cohort of infants with sBPD who had already failed pharmacologic rescue. Given the limited sample size, this study was designed as an exploratory case series to generate hypotheses for future prospective trials.

Materials and Methods

Study Design and Population

We conducted a retrospective, exploratory case series in a Level IV neonatal intensive care unit. The cohort included infants born at less than 32 weeks of gestation who had established sBPD and received the DART regimen but who failed to show clinical improvement.

We strictly defined “non-response” as the absence of any improvement in the PaO₂/FiO₂ ratio 72 hours after initiating dexamethasone. This 72-hour window was selected because the genomic and anti-inflammatory effects of systemic dexamethasone typically manifest within 24 to 48 hours of initiation (10, 11). This timeframe provides a sufficient therapeutic window to observe a clinical response while preventing unnecessary exposure to prolonged steroid therapy in infants whose respiratory failure is primarily driven by structural and mechanical constraints rather than active inflammation. Advanced imaging techniques have increasingly highlighted these fixed structural deficits as primary drivers of ongoing respiratory failure (12).

This study was approved by University of Health Sciences Turkey, İstanbul Bağcılar Training and Research Hospital Ethics Committee (approval no: 2025/04/10/042, date: 11.04.2025). Because of the retrospective design, the requirement for parental informed consent was waived in accordance with the Declaration of Helsinki.

Ventilation Protocol

Once we confirmed steroid failure, we adjusted the ventilator settings to reflect the “two-compartment” mechanical model of sBPD. Patients were transitioned to volume-guarantee combined with pressure support ventilation (VG-PSV) mode, a strategy increasingly recognized for chronic management (13). We reduced the respiratory rate to 20-25 breaths per minute, extended the inspiratory time (Ti) to ≥0.8 seconds, increased the tidal volume target to 10-12 mL/kg, and titrated positive end-expiratory pressure (PEEP) to 8-10 cmH₂O to stent open airways prone to malacia (14).

Statistical Analysis

Given the exploratory nature of this case series and the heavily restricted sample size (n=7), formal statistical hypothesis testing was deliberately avoided to prevent the overestimation of treatment effects and false-positive conclusions. Instead, descriptive statistics were used to characterize the cohort. Continuous non-parametric variables are reported as medians with minimum and maximum values (minimum-maximum). In alignment with the study’s primary objective, analytical focus was strictly placed on the evaluation of individual physiological trajectories and paired longitudinal mechanical ventilation trends rather than population-level comparative statistics.

Results

Seven infants met the inclusion criteria. The cohort had a median birth weight of 680 g (590-810) and a median gestational age of 25 weeks (24-26).

As detailed in Table 1, which presents individual patient trajectories, all patients exhibited severe baseline gas-exchange impairment: Median pH, 7.25 (7.20-7.32); pCO₂, 55 (54-63) mmHg; PaO₂/FiO₂, 66 (61-70) mmHg.

Transitioning these patients to the physiology-based ventilation protocol resulted in improvements in gas exchange (Table 2). The median PaO₂/FiO₂ ratio increased from 66 (61-70) mmHg at baseline to 110 (90-140) mmHg by day 14. Individual patient trends in PaO₂/FiO₂ ratios and mean airway pressures over time are illustrated in Figure 1.

Despite a reduced respiratory rate [median 22 (20-25) breaths per minute], the median arterial pCO₂ decreased from a baseline of 55 (54-63) mmHg to 45.5 (45-47) mmHg by day 14, suggesting improved alveolar clearance (Figure 2). Individual mechanical ventilation parameters (peak inspiratory pressure, PEEP, mean airway pressure) are detailed in Table 3.

Clinically, the mechanical reset was associated with successful weaning from invasive mechanical ventilation in these seven infants. None of the patients required a tracheostomy.

Discussion

In this exploratory cohort of infants with steroid-refractory sBPD, a ventilation strategy explicitly targeted at chronic lung mechanics was associated with a reduction in respiratory failure and with facilitated extubation. These preliminary observations support individualized therapy rather than relying on a “one-size-fits-all” approach. Recent advanced imaging techniques have increasingly highlighted fixed structural deficits as primary drivers of ongoing respiratory failure in these non-responders (12), necessitating adapted, long-term respiratory support strategies (13).

Our findings align closely with recent shifts in ventilatory paradigms for established sBPD. As emphasized by Sindelar et al. (3) and Gibbs et al. (4), standard ARDS-like strategies utilizing rapid rates become maladaptive in the chronic phase due to pathologically prolonged time constants. Furthermore, while Miller et al. (9) previously noted a lack of conclusive evidence for an optimal ventilator mode, our implementation of targeted volume-guarantee combined with pressure support ventilation (VG-PSV) provides a direct physiological countermeasure to the structural abnormalities of sBPD. This includes utilizing adequate PEEP to stent open airways, which is particularly critical for ventilator-dependent children with acquired tracheobronchomalacia (14), and aligns with adapting management practices to changing ventilator technologies (15).

The improvement in oxygenation may stem directly from addressing the time-constant heterogeneity. By extending Ti ≥0.8 seconds, our protocol may have facilitated the recruitment of these slow units. This targeted mechanical reassessment is crucial to mitigate the neurodevelopmental risks associated with severe BPD and its invasive interventions (16). Furthermore, early mechanical optimization may reduce the reliance on recurrent anti-inflammatory therapies (17), which is particularly relevant given the variable respiratory outcomes following late postnatal DART administration (18).

In established sBPD, a “standard” tidal volume might be largely lost to anatomical dead space in such conditions. Our observations suggest that higher delivered volumes (10-12 mL/kg) may be beneficial in select cases. Furthermore, the absence of air leaks in this small cohort suggests that the sBPD lung may tolerate larger volumes under close physiological monitoring. Avoiding tracheostomy in this high-risk population remains a critical therapeutic milestone to improve long-term trajectories (19). It should be noted that while genetic or syndromic factors (such as Bardet-Biedl syndrome) can independently complicate respiratory outcomes, our cohort specifically represented acquired sBPD (20, 21).

Study Limitations

This study has several important limitations. First, the lack of a control group makes it difficult to definitively attribute outcomes to the intervention; each patient effectively served as their own historical control.

Our preliminary findings suggest that for infants with established sBPD who are refractory to corticosteroids, a physiology-based ventilation strategy targeting prolonged time constants and dead space washout may improve gas exchange. Prospective studies are warranted to validate these findings and explore interpretable predictive models that can better define diagnostic criteria and optimize the timing of mechanical rescue.

Conclusion

Our results suggest that if respiratory failure is primarily mechanical, escalating anti-inflammatory therapy may be of limited benefit.