1. Exhaled volatile organic compounds (VOCs) at day 3 and day 7 of life in preterm infants (<30 weeks) are predictive of bronchopulmonary dysplasia (BPD) at 36 weeks of life.
2. Adding VOC prediction into existing BPD models increases the predictive power over the existing model alone in infants on non-invasive ventilatory support.
Evidence Rating Level: 2 (Good)
Study Rundown: Bronchopulmonary dysplasia (BPD) is a common complication associated with prematurity that can have longstanding impacts on pulmonary and neurocognitive function. Despite its increasing incidence, there are few prediction models available to identify premature infants at risk for developing BPD. One area of active research is the detection of volatile organic compounds (VOCs) in exhaled breath, which has previously been shown to be a prognostic biomarker for respiratory diseases in both adults and children. This study aimed to investigate the use of VOCs in a prediction model for BPD and compare the predictive power when alone or in combination with other prediction models. In this prospective cohort of 117 infants, VOCs compounds were predictive of developing BPD with moderate-to-high predictive value at days 3 and 7 of life, although the performance characteristics were lower in validation analyses. Adding VOC compounds to existing BPD prediction models increased the predictive power in non-invasively ventilated patients, but did not add power to all patients’ cohorts (both invasive and non-invasive ventilation). One limitation of the study is the inclusion of patients <30 weeks gestational age and does not address the predictive power of VOCs in other age groups such as late-preterm infants. Overall, this study provides evidence that VOC profiles in exhaled breath of preterm infants on non-invasive ventilatory support are predictive of developing BPD and add additional predictive power to existing models.
Click here to read the article in Journal of Pediatrics
Relevant Reading: Evaluation of Trends in Bronchopulmonary Dysplasia and Respiratory Support Practice for Very Low Birth Weight Infants: A Population-Based Cohort Study
In-Depth [prospective cohort]: This prospective cohort study enrolled 129 patients that were <30 weeks gestational age at two NICU sites in the Amsterdam University Medical Centers between January 2017 and November 2020. Exhaled breath samples were collected on postnatal days 3 and 7 and detected by gas-chromatography–mass-spectrometry analysis and utilized in multivariate logistic regression to drive a clinical prediction model. The comparison model was the updated clinical BPD prediction model of the National Institute of Child Health and Human Development (NICHD). The diagnosis and assessment of BPD severity occurred at 38 weeks corrected gestational age according to international criteria, with neonatologists blinded to VOC results. Of the 117 survivors from the 129 enrolled patients, 39/117 developed BPD at 36 weeks of age. In the derivation model, at day 3 of life (N=73), VOC had a c-statistic (95% CI) of 0.89 (0.80, 0.97) with a mean accuracy (min-max) for the internally validated model of 0.66 (0.45, 0.77). Similarly, on day 7 of life (N= 87), the derivation model had a c-statistic of 0.92 (0.84, 0.99) with a mean accuracy (min-max) for the internally validated model of 0.72 (0.54, 0.85). In contrast, the NICHD model had a c-statistic of 0.83 (95% CI 0.73-0.93) on day 3 and 0.82 (95% CI 0.72-0.92) on day 7. When adding VOC to the NICHD model for non-invasive ventilation patients, the c-statistic increased to 0.92 (95% CI 0.86-0.98) (p-value=0.04) on day 3 and increased to 0.94 (95% CI 0.88, 1.00) (p-value=0.03) on day 7. There was no statistical difference between VOC vs VOC+NICHD models when all patients were included.
Image: PD
©2023 2 Minute Medicine, Inc. All rights reserved. No works may be reproduced without expressed written consent from 2 Minute Medicine, Inc. Inquire about licensing here. No article should be construed as medical advice and is not intended as such by the authors or by 2 Minute Medicine, Inc.
