Weaning Very Preterm Infants from Nasal Continuous Positive Airway Pressure: A Noninferiority Trial Comparing Weaning from Continuous Positive Airway Pressure with Weaning Using Heated Humidified High-Flow Nasal Cannula: The CHiPS Study

Introduction The best strategy for weaning from continuous positive airway pressure (CPAP) has not been clearly identified. Converting to heated humidified high-flow nasal cannula (HHHFNC) may provide advantages including ease of application, reduced need for intensive nursing, improved parental interaction and satisfaction, and earlier introduction of suck feeding.

Materials and Methods A single center, prospective, randomized control trial of 120 infants born at <30 weeks’ gestation was performed. Infants requiring CPAP pressures of 6 cm H₂O in room air and who were stable (by predetermined criteria) for 48 hours were eligible. They were randomized (stratified by gestation) to either weaning off CPAP (by reducing pressure to 5 cm H₂O and removing the CPAP after 96 hours) or by use of HHHFNC and stepwise weaning over 96 hours. There was a strict weaning protocol with provision for the escalation of therapy if required. Primary outcome was the duration (in hours) of respiratory support from randomization to achieving 72 hours free of support. A total of 120 infants was estimated to provide ≥90% power to conclude noninferiority of weaning from HHHFNC (threshold 15%).

Results Total 59 infants were randomized to HHHFNC and 61 to CPAP. Demographics were similar in both groups ([Table A011]) .The mean difference in hours to successfully wean from HHHFNC versus CPAP was −27 (95% confidence interval: −133 to 80) hours which was within the specified noninferiority margin (with correction for covariates). Infants in both groups required a similar number of weaning attempts. There were very few treatment complications in either group. Overall parental satisfaction was similar for both groups.

Conclusion HHHFNC was shown to be noninferior to CPAP in regards to hours of weaning from support with similar numbers of weaning attempts and treatment complications.

Table A011

Patient characteristics (data are shown as median (interquartile range) or number (%))
	CPAP (n = 61)	HHHFNC (n = 59)	p-Value
Birth gestation (wk)	28 (27–29)	28 (27–29)	0.58
Birthweight (gm)	1,115 (932–1,347)	1,132 (965–1,290)	0.23
Male	34 (56)	26 (44)	0.27
Antenatal steroids
Completed	48 (79)	34 (58)	0.18
Partial	11 (18)	20 (34)	0.06
Results: parental satisfaction survey
	CPAP	HHHFNC	p-Value
Parental satisfaction at 1 week postrandomization (Likert)	3 (2–4.25)	3 (2–5)	0.9
Parental satisfaction at 36-week corrected gestational age)	2.5 (2–5)	4 (2.75 -5)	0.8
Results: primary outcome
	CPAP (n = 61)	HHHFNC (n = 59)	p-Value
Hours to successful wean (mean, CI)	372 (295–450)	342 (274–410)	0.57
Infants requiring rescue CPAP (%)	13 (21.3)	14 (23.7)	0.86
Weaning attempts	2 (1–4)	2 (1–4)	0.23

Abbreviations: Cl, confidence interval; CPAP, continuous positive airway pressure; HHHFNC, heated humidified high-flow nasal cannula; wGA, weeks’ gestational age.

Conflict of Interest

None declared.