Objective:To evaluate the efficacy of high-flow nasal cannula (HFNC) for assisted weaning from nasal continuous positive airway pressure (NCPAP) in extremely preterm infants (<28 weeks of gestation) with bronchopulmonary dysplasia (BPD).Methods:This multicenter retrospective cohort study enrolled extremely preterm infants with BPD receiving NCPAP beyond one month postnatally at three tertiary hospitals (Fujian Children's Hospital, Fujian Maternity and Child Health Hospital, and Fujian Obstetrics and Gynecology Hospital) from December 2017 to June 2024. After meeting weaning criteria, these infants transitioned to low-flow nasal cannula (LFNC group, n=51) or HFNC (HFNC group, n=60). Primary outcomes were initial weaning success and total NCPAP duration. Secondary outcomes included oxygen therapy duration, time to full enteral feeding, hospital stay, and oxygen therapy at discharge. Safety outcomes encompassed pneumothorax, nasal trauma, new-onset necrotizing enterocolitis, retinopathy of prematurity, sepsis, and grade Ⅲ-Ⅳ intraventricular hemorrhage. Group differences were analyzed using independent t-tests, Wilcoxon rank-sum, and Chi-square tests, with inverse probability of treatment weighting (IPTW) adjusting confounders followed by logistic/linear regression. Results:After IPTW (HFNC group: 54.8 cases; LFNC group: 53.0 cases), no statistically significant differences were observed in baseline demographic and clinical characteristics between groups, including gestational age, birth weight, BPD severity, duration of invasive mechanical ventilation, and duration of antibiotic exposure (all P>0.05). HFNC group demonstrated higher initial weaning success rate [83.2% (45.6/54.8) vs. 61.6% (32.6/53.0), OR=3.087, 95% CI: 1.027-9.281, P=0.044] and a significant reduction of 13.7 days in total NCPAP duration [25.0 (21.0-35.3) vs. 42.0 (29.4-61.0) days, β=－13.700, 95% CI:－21.372 to－6.027, P<0.001]. Secondary outcomes showed no significant differences between HFNC and LFNC groups: oxygen therapy duration [70.0 (51.8-81.9) vs. 68.0 (59.2-83.1) days], time to full enteral feeding [41.0 (31.0-59.0) vs. 45.0 (40.0-61.6) days], hospital stay [87.0 (74.6-98.5) vs. 93.7 (74.0-102.7) days], and oxygen therapy at discharge [16.6% (9.1/54.8) vs. 13.8% (7.3/53.0)] (all P>0.05). Safety outcomes were comparable between HFNC and LFNC groups: pneumothorax [2.8% (1.5/54.8) vs. 5.1% (2.7/53.0)], nasal trauma [3.8% (2.1/54.8) vs. 11.6% (6.2/53.0)], new-onset necrotizing enterocolitis [6.9% (3.8/54.8) vs. 5.5% (2.9/53.0)], and retinopathy of prematurity [14.4% (7.9/54.8) vs. 17.9% (9.5/53.0)] (all P>0.05). Conclusion:HFNC-assisted weaning may improve initial success and reduce NCPAP duration in BPD extremely preterm infants without increasing adverse events versus LFNC.

Objective:To evaluate the efficacy of high-flow nasal cannula (HFNC) for assisted weaning from nasal continuous positive airway pressure (NCPAP) in extremely preterm infants (<28 weeks of gestation) with bronchopulmonary dysplasia (BPD).Methods:This multicenter retrospective cohort study enrolled extremely preterm infants with BPD receiving NCPAP beyond one month postnatally at three tertiary hospitals (Fujian Children's Hospital, Fujian Maternity and Child Health Hospital, and Fujian Obstetrics and Gynecology Hospital) from December 2017 to June 2024. After meeting weaning criteria, these infants transitioned to low-flow nasal cannula (LFNC group, n=51) or HFNC (HFNC group, n=60). Primary outcomes were initial weaning success and total NCPAP duration. Secondary outcomes included oxygen therapy duration, time to full enteral feeding, hospital stay, and oxygen therapy at discharge. Safety outcomes encompassed pneumothorax, nasal trauma, new-onset necrotizing enterocolitis, retinopathy of prematurity, sepsis, and grade Ⅲ-Ⅳ intraventricular hemorrhage. Group differences were analyzed using independent t-tests, Wilcoxon rank-sum, and Chi-square tests, with inverse probability of treatment weighting (IPTW) adjusting confounders followed by logistic/linear regression. Results:After IPTW (HFNC group: 54.8 cases; LFNC group: 53.0 cases), no statistically significant differences were observed in baseline demographic and clinical characteristics between groups, including gestational age, birth weight, BPD severity, duration of invasive mechanical ventilation, and duration of antibiotic exposure (all P>0.05). HFNC group demonstrated higher initial weaning success rate [83.2% (45.6/54.8) vs. 61.6% (32.6/53.0), OR=3.087, 95% CI: 1.027-9.281, P=0.044] and a significant reduction of 13.7 days in total NCPAP duration [25.0 (21.0-35.3) vs. 42.0 (29.4-61.0) days, β=－13.700, 95% CI:－21.372 to－6.027, P<0.001]. Secondary outcomes showed no significant differences between HFNC and LFNC groups: oxygen therapy duration [70.0 (51.8-81.9) vs. 68.0 (59.2-83.1) days], time to full enteral feeding [41.0 (31.0-59.0) vs. 45.0 (40.0-61.6) days], hospital stay [87.0 (74.6-98.5) vs. 93.7 (74.0-102.7) days], and oxygen therapy at discharge [16.6% (9.1/54.8) vs. 13.8% (7.3/53.0)] (all P>0.05). Safety outcomes were comparable between HFNC and LFNC groups: pneumothorax [2.8% (1.5/54.8) vs. 5.1% (2.7/53.0)], nasal trauma [3.8% (2.1/54.8) vs. 11.6% (6.2/53.0)], new-onset necrotizing enterocolitis [6.9% (3.8/54.8) vs. 5.5% (2.9/53.0)], and retinopathy of prematurity [14.4% (7.9/54.8) vs. 17.9% (9.5/53.0)] (all P>0.05). Conclusion:HFNC-assisted weaning may improve initial success and reduce NCPAP duration in BPD extremely preterm infants without increasing adverse events versus LFNC.