EBNEO COMMENTARY: Weaning from cpap using high-flow is non-inferior to using cpap alone in very preterm infants

April 19, 2023

Weaning from cpap using high-flow is non-inferior to using cpap alone in very preterm infants


Clements J, Christensen PM, Meyer M. A randomised trial comparing weaning from CPAP alone with weaning using Heated humidified high flow nasal cannula in very preterm infants: the ChiPS study. Arch Dis Child Fetal Neonatal Ed 2922; 0:F1- F6. DOI: 1136 /arch dis child-2021-323636. PMID: 35851035


Nonye Ijeoma Ezeh; MBBCh, FWACP (Paed), MRCPCH
Department of Neonatology
New Cross Hospital,
Wolverhampton, WV10 0QP
United Kingdom

Teim Jengoa Eyo; MBBS, MRCPCH
Department of Neonatology
Heartlands Hospital
Birmingham, B9 5SS
United Kingdom

Kalyana Gurusamy
Department of Neonatology
New Cross Hospital,
Wolverhampton, WV10 0QP
United Kingdom


Nonye Ijeoma Ezeh
+44 7925 938797




(P)In preterm infants, less than 30 weeks gestation at birth who have been stable for at least 48hours on bubble nasal Continuous Positive Airway Pressure (nCPAP), (I) is weaning from nCPAP using heated humidified high flow nasal cannula (nHF) (C )non-inferior to weaning using nCPAP alone (O) in terms of the duration on respiratory support (T) when weaning over 96hours or till 36 weeks corrected gestation?


Design: Single-center, randomised control, non-inferiority trial.

Allocation: Eligible infants were randomised using a computer-generated sequence and stratified based on the gestation (<27 weeks or > 27 weeks GA)

Blinding: The allocation sequence was concealed from the investigators.

Follow-up period: Infants were weaned over 96 hours. Follow-up at 1 week and at 36 weeks postnatal age to assess Parental Stress Perception.

Setting: Middlemore Hospital, Neonatal Intensive Care Unit, New Zealand from 2015 to 2017.

Patients: 120 infants were randomly assigned to the nCPAP or nHF group (61 and 59 respectively). All infants <30 weeks gestational age who were stable on nCPAP of 6cm of water for at least 48hours were included. Stability criteriafor weaning were no oxygen requirement, absence of tachypnoea and absence of significant desaturations or bradycardia episodes.


Intervention group: Infants randomised to the nHf group after meeting stability criteria were weaned over 96 hours. The flow was reduced every 24 hours starting at 6L/min on day 1 and weaned to 4L by day 2, then reduced by one every 24 hours till 2L and discontinued on day 5.

Control group: Infants in the control group commenced weaning at 6cm of water and weaned by 1 every 48 hours before discontinuing on day 5 after reaching 5cm of water.

Criteria for escalation: Pressure or flow was increased if failure criteria were met. That is up to 6 episodes of desaturations or bradycardia requiring intervention over 6 hours, an episode where a T-piece was required to provide intermittent positive pressure, or 30 minutes of sustained tachycardia above 70/min. Infants on nHF could be ‘rescued’ to nHF 8L/min before being placed back on nCPAP. Infants were then recommenced on weaning once meeting set stability criteria.


Primary outcome: Duration (in hours) of respiratory support from randomisation until completion of weaning.

Secondary outcomes:

Duration of respiratory support for neonates < 27 weeks GA, Chronic Lung Disease (CLD), nasal trauma, age at the neonate achieved first sucking feed, age when full sucking feeds was achieved, type of feeding at discharge, weight z scores, and other neonatal morbidities.

Analysis and Sample Size: Using pilot data from an observational study and bootstrapping, the sample size needed to provide 92% power, was determined to be 100 infants with 50 per group. A non-inferiority margin of 15% was chosen.  Thus, 120 infants (60 per group) were recruited to achieve the sample size. Statistical analysis was performed using an R Version 3.6.1. The primary outcome was done using time-to-event analyses. The Parental Stress Scale: Neonatal Intensive Care Unit was analyzed using linear mixed modelling with Turkey correction. A non-inferiority one-sided alternative (significance level 2.5%) was used to test the primary hypothesis and a superiority two-sided alternative for the secondary hypotheses

Patient follow-up (% included in the analysis):  204 infants were assessed for eligibility of which 84 were excluded; 38 due to not meeting criteria, 36 due to non-consent, and 10 on account of transfers. 120 were randomized, 61 to the nCPAP group and 59 to the high flow group. Per-protocol analyses were used for the primary outcome. Four infants were lost to follow-up; 3 due to inter-hospital transfer and 1 due to withdrawn consent. Hence 95.8% were included in the pre-protocol analyses. Infants were weaned from randomisation till 96 hours or 36 weeks PMA. A follow-up survey was done a week post-randomisation and at 36weeks post-menstrual age to assess Parental Stress and Perception of the infant’s treatment.


Primary Outcome

The restricted mean hours from randomization to 72 hours off respiratory support or 36 weeks’ CGA was 401 hours in the nCPAP group (upper bound of 461 hours) and 375 hours (with an upper CI of 413 hours). Using the 15% non-inferiority margin, meant weaning using nHF was non-inferior to weaning directly from nCPAP. An intention-to-treat analysis gave similar results. Thirty-two infants (15 nHF, 17nCPAP) achieved primary outcomes at 36 hours. 

Secondary Outcomes.

Outcomes  high-flow group


nasal CPAP group


Odds ratio (95% CI)
CLD 11 (18%) 22(36%) 0.42 (0.18 to 0.99)
Nasal trauma 2(3%) 4(6%) 0.51 (0.09 to 2.9)
Age at first sucking feed in days 38.6 39.1 0.63
Age at full sucking feed 65 66.2 0.35
Feeding at discharge
Tube: 13(22%) 12(20%) 1.12 (0.47 to 2.65)
Exclusive breastfeeding: 12(21%) 13(22%) 0.95(0.40 to 2.26)
Weight z scores 0.18 (+/- 0.60) 0.22 (+/- 0.79) 0.19 (0.41 to -0.02)
Late-onset sepsis 6(10%) 8(13%) 0.77 (0.25 to 2.37)
Retinopathy of prematurity grade 3 1 (1.7%) 3 (5%) 0.34 (0.03 to 3.4)
IVH grade 3/4 1(1.7%) 0(0%) 1.02 (0.98 to 1.05)
NEC 0(0%) 2(3%) 0.97 (0.92 to 1.01)

For infants under 27 weeks, restricted mean hours from randomisation to 72 hours off respiratory support or 36 weeks PMA was 304 (with an upper bound of 350 hours) for the nCPAP group and 362 (with an upper CI of 408) for the nHF group.  Hence weaning using nHF could not be concluded as non-inferior to weaning directly from nCPAP in this group.

Infants in the nHF arm had significantly less CLD than those in the nCPAP arm (11 versus 22 with an odds ratio of 0.42 and 95% CI between 0.18 to 0.99). Other secondary outcomes, however, were not significantly different.


The authors concluded that for infants older than 27 weeks but less than 30 weeks gestation, who were clinically ready to wean from nCPAP, weaning from 2l/min of high-flow therapy was non-inferior to discontinuing nCPAP at 5cm of water.


Continuous positive airway pressure (nCPAP) is widely used in neonatal units both as a primary mode of non-invasive respiratory support and post-extubation (1). Despite increased uptake of non-invasive ventilation, time on oxygen therapy and oxygen requirement at 36 weeks CGA is higher than before (2).  In contrast to mechanical ventilation, benefits of nCPAP include decreased combined death or chronic lung disease (CLD) rates at 36 weeks corrected gestational age (CGA) (3).


Nonetheless, the use of nCPAP is not without risks namely pneumothorax, nasal trauma, increased abdominal distension and impeded venous return, therefore early weaning is crucial(4). However, a lack of consensus over the best way to achieve this remains(4) with options ranging from gradually decreasing the pressure, interval-based weaning off CPAP, transition to high flow nasal cannula (nHF) to cycling between different modes of non-invasive support (5).


Notable advantages of nHF includes ease of use, reduced nasal trauma and pneumothorax rates, and improved infant comfort and parental stress(6) thus, the ChiPS study aimed at providing evidence that weaning from nCPAP with nHF was non-inferior to weaning from nCPAP alone with regards to duration of respiratory support.


With restricted mean hours from randomisation to 72 hours off respiratory support or 36 weeks’ CGA of 401 hours in the nCPAP group (upper bound of 461 hours) and 375hours (with an upper CI of 413 hours), nHF was shown to be non-inferior to weaning directly from nCPAP within a 15% margin which is like findings of previous studies. (7)


The study was appropriately powered and randomised to address the clinical question. It was methodically written, with detailed results and balanced baseline demographics in study groups. There were clear inclusion, exclusion, stability and failure criteria.


The ChiPS study has some constraints including being a single-centre study with a small sample size limiting its representativeness. The study’s approach of a minimum 96-hour weaning period with nCPAP, as opposed to the 1cm of water reduction every 12 to 24 hours recommended by some guidelines (8), may have impacted the primary outcome. The study also lacks a cost analysis of weaning over a minimum of 96 hours. The exclusion of stable babies on nCPAP with oxygen needs up to 30%, who would have met weaning criteria in clinical practice, reduces the generalisability of the study in supporting changes in clinical practice. Adjusting for mechanical ventilation duration as well as the gestation at birth may have produced different results, as nCPAP group had longer mechanical ventilation which increases CLD risk. Furthermore, small numbers and insufficient power prevented the study from proving non-inferiority in the <27 weeks subgroup.


In conclusion, the ChiPS study provided evidence for the use of nHF in weaning stable preterm infants aged between 27 and 29 weeks gestation from nCPAP, nevertheless, clinicians need to assess its suitability in their individualized settings.  Further research is needed to determine its safety and applicability in weaning premature babies  < 27 weeks gestation from nCPAP as well as the comparability of nHF when weaned over a shorter time frame.


  1. Chowdhury O, Wedderburn CJ, Duffy D, Greenough A. CPAP review. Vol. 171, European Journal of Pediatrics. 2012.

  2. Doyle LW, Carse E, Adams AM, Ranganathan S, Opie G, Cheong JLY. Ventilation in Extremely Preterm Infants and Respiratory Function at 8 Years. New England Journal of Medicine. 2017;377(4).

  3. Schmölzer GM, Kumar M, Pichler G, Aziz K, O’Reilly M, Cheung PY. Non-invasive versus invasive respiratory support in preterm infants at birth: Systematic review and meta-analysis. BMJ (Online). 2013;347.

  4. Ting J, Kissoon N. Continuous Positive Airway Pressure: Method of Discontinuing in Neonates, Unresolved. The Indian Journal of Pediatrics. 2015 Sep 5;82(9):775–6.

  5. Van Delft B, Van Ginderdeuren F, Lefevere J, Van Delft C, Cools F. Weaning strategies for the withdrawal of non-invasive respiratory support applying continuous positive airway pressure in preterm infants: A systematic review and meta-analysis. BMJ Paediatr Open. 2020;4(1).

  6. Roberts CT, Hodgson KA. Nasal high flow treatment in preterm infants. Matern Health Neonatol Perinatol. 2017 Dec 6;3(1):15.

  7. Lavizzari A, Colnaghi M, Ciuffini F, Veneroni C, Musumeci S, Cortinovis I, et al. Heated, Humidified High-Flow Nasal Cannula vs Nasal Continuous Positive Airway Pressure for Respiratory Distress Syndrome of Prematurity. JAMA Pediatr. 2016 Aug 8;

  8. Kakkilaya V, Tang A, Wagner S, Ridpath J, Ibrahim J, Brown LS, et al. Discontinuing Nasal Continuous Positive Airway Pressure in Infants ≤32 Weeks of Gestational Age: A Randomized Control Trial. Journal of Pediatrics. 2021;230.

Leave a comment