124 - A Novel Use of Risk Stratification Tools to Identify Quality Improvement Targets for the Reduction of Bronchopulmonary Dysplasia Rates

Publication Number: 3118.124

Richard A. Hutchinson, Liverpool Women's Hospital, Liverpool, England, United Kingdom; Catherine Graham, Liverpool Women's Hospital, Liverpool, England, United Kingdom; Alex J. Cleator, Liverpool Women’s Hospital, Liverpool, England, United Kingdom

Background: Bronchopulmonary dysplasia (BPD) is a common adverse consequence of prematurity. It is associated with increased mortality and, among survivors, ongoing respiratory and neurodevelopmental morbidity.

Objective: It has been shown that improved post-term lung development is associated with the extended provision of non-invasive positive pressure support to support lung and pulmonary blood vessel growth.
This study examines whether premature progression to lower levels of non-invasive respiratory support is temporally associated with increased rates of BPD in preterm life. The longitudinal risk of BPD development will be correlated with ventilatory support modes, to identify periods of respiratory management where the risk of progression to subsequent BPD has increased.

Design/Methods: In a cohort of < 28/40 infants admitted to the NICU at Liverpool Women's Hospital in 2024, BPD risk was calculated longitudinally, using the NICHD Neonatal BPD Outcome Estimator (2022) at days 1, 3,7,14, and 28. In patients who subsequently received a diagnosis of death/BPD (defined as the requirement for respiratory support at 36/40), the respiratory mode in use at the lowest inflexion point of risk probability (see Figure 1) was noted. With death/BPD risk subsequently increasing after this period, this was identified as an insufficient level of respiratory support level to maintain lung growth and prevent BPD.

Results: Patient demographics are illustrated in Table 1. 55 patients were studied, of whom 45 (76%) died or developed BPD. 9 patients were excluded from analyses due to death before all analysis points could be assessed; 1 patient was additionally excluded due to concurrent NEC confounding respiratory progression in the analysis period.
The point at which risk of developing BPD increased was not uniformly distributed across all modes of respiratory support. High-flow nasal cannula (HFNC) was noted as a significant outlier (Grubb's Test p< 0.05), being in use when BPD rates escalated in 16/45 cases (36%). It was also notable that in 8/45 cases (18%), babies advanced to no respiratory support, before regressing to require such support by 36/40.

Conclusion(s): These findings support the hypothesis that premature progression to lower levels of respiratory support may be associated with increasing the subsequent risk of developing BPD, through theorised insufficient intra-pulmonary pressures to support lung growth. This provides evidence to support local initiatives to meet higher thresholds for respiratory health indices, before progressing to lower levels of respiratory support, as part of a wider respiratory care bundle.

Example longitudinal BPD risk progression
Idealised longitudinal course of BPD risk progression plotted against time, overlaid on ventilatory mode. Inflexion point of risk progression highlighted, following progression to HFNC.

Patient Demographics
PAS 2026 table.pdf