648 - Data-Driven Quality Improvement Program to Reduce Bronchopulmonary Dysplasia in Very Low Birth Weight Infants

Poster Number: 648
Publication Number: 648.138

Vineet Lamba, University of Tennessee Health Science Center College of Medicine, Memphis, TN, United States; Omoloro Adeleke, University of Tennessee Health Science Center, Memphis, TN, United States; Ajay J. Talati, UTHSC, Memphis, TN, United States

Background: Bronchopulmonary dysplasia (BPD) is a common morbidity in preterm infants and there are few interventions which can led to a reduction in its incidence. We have seen a gradual increase in the incidence of BPD in very low birth-weight (VLBW) infant population at our level III NICU.

Objective: Achieve a 20% reduction in the combined outcome of BPD or Death by 36 weeks post menstrual age (PMA). Secondary aim is to increase the number alive ventilator-free days in the first 28 days of life.

Design/Methods: A QI team formulated and implemented consensus guidelines using the Plan-Do-Study-Act (PDSA) method of quality improvement. We conducted sequential improvement cycles focusing on our key drivers of reducing extubation failure, early surfactant administration, gentle ventilation strategies, optimizing medication use (Fig1). Overall outcome measure was rate of BPD by Vermont Oxford Network (VON) definition or death by 36 weeks PMA. Individual outcome and process measures were defined for each PDSA cycle. Balancing measures were pneumothorax and death which was built into primary measure as competing outcome.

Results: We conducted an extensive review of our practices over a 9-month period which helped identify areas of improvement and formulate our key-drivers. The median gestation age of our population is 28 weeks and median birth weight is 980 grams. The baseline rate of BPD of Death was 36.1% and average alive ventilator free days (in 1st 28 DOL) was 16.9 days. The first PDSA cycle was aimed at reducing our rates of extubation failure, defined as need for re-intubation within 7 days of planned extubation. With the implementation of an extubation bundle which recommended use of non-invasive positive pressure ventilation, higher post extubation peep and peri-extubation caffeine, we reduced extubation failure from 34% as baseline to < 10% (Fig 2). Second PDSA focused on improving early surfactant use (< 2 hours) when indicated with which we improved early surfactant delivery from 70% to 90%. Subsequent PDSA cycles were conducted focusing on delivery room CPAP use and gentle ventilation led to improvement in delivery room CPAP rates and adoption of gentle ventilation guidelines. Thus far, while our individual PDSA processes have been successful at improving key measures, it has not led to a meaningful reduction in our outcomes of BPD or death or alive ventilator-free days (Fig 3).

Conclusion(s): This QI initiative has been able to improve some of the pathophysiological contributors of BPD but has not yet successfully reduced the rate of BPD or death in our level III NICU