| Summary: | Background. KL-6 is a biomarker of interstitial lung injury and increases during repair. Aim. Our aim was to determine the predictive value of plasma KL-6 for the development of bronchopulmonary dysplasia (BPD) in preterm infants. Methods. Ninety-five extremely preterm infants (EPIs), born at <28 gestational age (GA), were divided into two main BPD groups as follows: the moderate/severe and the no/mild group. KL-6 was analyzed on days 7 and 14. Binary logistic regression analyses and ROC curve analyses were performed. Results. Infants <26 + 0 weeks’ GA have higher mean KL-6 than infants >25 + 6 weeks’ GA on 7 and 14 days (335 vs. 286 U/ml and 378 vs. 260 U/ml; p=0.005 and 0.018, respectively). In the binary regression model at KL-6 day 7, three of the prognostic factors remained significant—mechanical ventilation OR: 10.38 (95% CI: 3.57–30.14), PDA OR: 6.39 (95% CI: 0.87–46.74), and KL-6 OR: 4.98 (95% CI: 1.54–16.08). The AUC was 0.86 with a sensitivity and specificity of 79% at a cutoff value ≥0.34. In the binary regression model at KL-6 day 14, six of the prognostic factors were significant—PDA OR: 23.34 (95% CI: 2.14–254.24), KL-6 OR: 13.59 (95% CI: 3.19–57.96), GA OR: 4.58 (95% CI: 1.16–18.06), mechanical ventilation OR: 4.45 (95% CI: 1.23–16.16), antenatal steroids OR: 0.19 (95% CI: 0.04–0.95), and gender (female OR: 0.30 (95% CI 0.08–1.12)). The AUC was 0.91, and the sensitivity and accuracy for a cutoff ≥0.37 were 89% and 85%, respectively. Conclusion. KL-6 could be a useful screening biomarker for early detection of infants at increased risk for developing BPD.
|
|---|