Accuracy of point-of-care nasopharyngeal Interleukin 6 and lung ultrasound in predicting the development of bronchopulmonary dysplasia in preterm infants born before 30 weeks of gestation

Abstract

Background Bronchopulmonary dysplasia (BPD) is a common cause of morbidity in preterm infants, leading to long-term respiratory complications and risk of neurodevelopmental impairment. Although it has a multifactorial etiology, local inflammation may play a major role. Objectives. We aimed to analyze the relationship between nasopharyngeal aspirate (NA) interleukin 6 (IL6) levels and clinical and imaging findings of BPD. Methods: Pilot study in preterm infants < 30 weeks. NA was collected at 7 days of life (DOL) and serial lung ultrasounds (LUS) were performed during admission. NA-IL6 levels were measured using an automated electrochemiluminescence immune-analyzer Cobas-e602 and an IL6 ELISA method. Results. Forty-two patients were studied. Infants with BPD had significantly lower gestational age and higher levels of NA-IL6 at DOL 7. Both methods showed good agreement: ICC = 0.937 (95%CI 0.908-0.957); p < 0.001) and Passing-Bablok Regression (R2 = 0.961). LUS score (AUC = 0.83) and NA-IL6 (AUC = 0.81) at DOL 7 predicted BPD. The AUC of NA-IL6 as a stand-alone marker of BPD was 0.808 (95% CI 0.67 – 0.94); p = 0.002, with 24 pg/ml being the best cutoff with a sensitivity and specificity of 83.3%. A model including birth weight, LUS score at DOL7, NA-IL6 at DOL7, and days of mechanical ventilation predicted BPD with R2 = 0.600 (p < 0.001). Conclusions. Point-of-care assessment of NA-IL6 is feasible and reliable compared with a reference method and can be useful in managing BPD. Predictive models of BPD in the first week of life, including clinical, biological, and imaging biomarkers must be tested in larger cohorts.