Fig. 1: Improved performance of BPD prediction by novel plasma protein biomarkers detected in the first week of life.
From: Validation of disease-specific biomarkers for the early detection of bronchopulmonary dysplasia
Protein markers (SIGLEC-14, BCAM, ANGPTL-3) significantly improve performance of BPD prediction models solely based on clinical variables. AIC vs. AUC for 31 of the 32 analyzed models (excluding the null model from the graphs) including protein expression levels (pr), gestational age (GA), sex, birth weight (weight) and early-onset infection (eoi) for a no BPD vs. moderate and severe BPD (<32 weeks GA), for b no BPD vs. moderate and severe BPD (<28 weeks GA), and for c no BPD vs. all BPD Grades (<32 weeks GA). AIC Akaike’s Information Criterion, AUC area under the curve, BPD bronchopulmonary dysplasia, BPD grades: 0 = no BPD, 1 = mild BPD, 2 = moderate BPD, 3 = severe BPD.
