Thrasher Research Fund - Medical research grants to improve the lives of children

Project Details

Early Career

Status: Funded - Closed

Impact of CPAP level on ventilation/perfusion mismatch and shunt in premature infants; a pilot randomized controlled trial

Nicolas Bamat, MD

Summary

Background: The use of nasal continuous positive airway pressure (CPAP) over mechanical ventilation (MV) increases survival without bronchopulmonary dysplasia (BPD) in very preterm infants. Strategies to identify individualized, best CPAP levels that minimize ventilation/perfusion (V/Q) mismatch could help reduce CPAP failure. Objective: To measure within-subject changes in V/Q mismatch in response to an individualized CPAP level selection strategy. Design/Methods: We performed a single-arm trial (NCT02983825) in infants >24 hours of age, 27-35 weeks postmenstrual age, and requiring a fraction of inspired oxygen (FiO2) ≥ 25% while on a CPAP level of 4-7 cm H2O. The study intervention was a protocol of stepwise CPAP level changes, with the direction and magnitude guided by the V/Q mismatch response at each step. V/Q mismatch was measured by the degree of right-shift (RS) in a curvilinear best-fit of multiple FiO2/SpO2 pairs, relative to the oxyhemoglobin dissociation curve. In each subject, RS was measured at 2 baseline periods 30 minutes apart, and then following each CPAP level change. “Best CPAP” was defined as the lowest CPAP level where a response (> 5% improvement in RS) was observed, and assigned as baseline when not observed. Our primary outcome was the within-subject change in RS between baseline and best CPAP, using a Wilcoxon signed-rank test for nonparametric matched-pairs. Results: Of 21 enrolled subjects, 12 were evaluable; the remainder were intubated for MV or fell outside of the eligible FiO2 and/or CPAP range during or after informed consent. Nine of 12 subjects completed the study protocol without meeting pre-specified stopping criteria. Six of nine (67%) subjects had a response. The median [IQR] within-subject change in RS between baseline 2 and “best CPAP” was 1.2 [0-3.2] kPa; p = 0.02. A similar magnitude change in RS was observed between baseline periods at the same CPAP level 30 minutes apart; 1.3 [-1.3 - 3.5] kPa. Conclusions: Non-invasive measures of V/Q mismatch are feasible, but challenged by the dynamically changing respiratory status of preterm infants requiring supplemental oxygen. It is unclear if the improved RS observed in response to an individualized CPAP level selection strategy reflects improvements in pulmonary gas exchange or random variation over time. The utility of this strategy for reducing CPAP failure requires further research. Our experience can help guide the development of future protocols. Website Link: http://www.chop.edu/doctors/bamat-nicolas

Supervising Institution:
Children's Hospital of Philadelphia

Mentors
Haresh Kirpalani

Project Location:
Pennsylvania

Award Amount:
$26,750