Comparison of CPAP Machines With Reusable vs Disposable Circuits

Randomised Trial Comparing CPAP Machines With Reusable vs Disposable Circuits

Status

Terminated

Phases

Study type

Interventional

Source

ClinicalTrials.gov

Registry ID

NCT03121612

Enrollment

Registered

2017-04-20

Start date

2017-08-01

Completion date

2018-10-15

Last updated

2019-11-20

For informational purposes only — not medical advice. Sourced from public registries and may not reflect the latest updates. Terms

Conditions

Respiratory Distress Syndrome, Newborn

Keywords

RDS

Brief summary

The study aims to assess the basic functionality of a newly designed CPAP machine with reusable circuits to existing machines with disposable circuits, for treatment of newborn infants diagnosed with respiratory distress syndrome. The assessment will compare a comprehensive list of physiological parameters over the first 72 hours of treatment, and will also monitor rates of side effects and adverse events. The null hypothesis is that infants treated on the two categories of machine (reusable vs disposable) will not differ in relation to key physiological parameters by more than 0.63 standard deviations.

Detailed description

One of the commonest sources of serious newborn morbidity and mortality is difficulty with breathing. When this occurs, three main types of supportive therapy are available to increase the provision of oxygen to cells: a) passive provision of oxygen-enriched gases (i.e., higher than the 21% O2 found in the earth's atmosphere) through tubes in the nostrils, or by putting a hood over the baby's head and enriching the gases under that hood; b) provision of room air or oxygen-enriched gasses under pressure, frequently performed using a method called continuous positive airway pressure \[CPAP\] therapy; and/or c) by using a machine that is able to breath on behalf of the baby, most commonly referred to as mechanical ventilation \[MV\]. Passive therapy is the least invasive method but is also of limited benefit, particularly for infants born preterm. CPAP is more effective than passive methods because continuous distending pressure to the lungs allows better oxygen exchange; however, the distending pressure increases the risk of damage to the lung. MV is the only method that can be used on babies without a neurological impulse to breath, but the mechanical breathing action can damage the lungs, and MV is usually provided through a tube inserted into the lungs which increases the risk of lung infection; MV machines are also significantly more expensive than CPAP machines. In high resource settings, CPAP is now the preferred method of providing oxygen for infants where passive therapy is insufficient, because of the lower infection risk, lower risk of lung damage, and relative ease of clinical care. CPAP is increasingly recommended for low resource settings, but the CPAP machines used in high resource settings are too expensive for low resource settings due to high-priced consumables ($US50-200/baby), and are usually unusable in low resource settings because they require 'medical air' (clean air in a cylinder, or through a piped wall system) with which to blend 100% oxygen. Low cost 'indigenous' machines ('jury-rigged' by hospital staff) have also been developed, but these do not provide the heated, humidified and blended gasses, that are recommended for CPAP. This study seeks to evaluate a novel CPAP machine that provides heated, humidified, blended gasses, in line with recommendations for high-resource settings, while massively reducing costs by including re-usable tube sets and humidifiers that can be autoclaved, and with an on-board air-compressor to allow use in a broader range of clinical settings. By reducing the cost per CPAP treatment, such a machine can dramatically increase the number of hospitals in low resource settings that can provide high quality CPAP treatment.

Interventions

DEVICECPAP

CPAP therapy

Study design

Allocation

RANDOMIZED

Intervention model

PARALLEL

Primary purpose

TREATMENT

Masking

NONE

Masking description

As the machines are physically distinct, blinding of participants, care providers and investigators is not feasible. The assessment of most outcomes cannot therefore be blinded, with one exception: x-rays for neonates with suspected pneumothorax will be provided to a radiologist for independent verification; the radiologist will not be in the nursery and will therefore be blinded to treatment allocation.

Eligibility

Sex/Gender

ALL

Age

1 Minutes to 24 Hours

Healthy volunteers

Inclusion criteria

1. Infants born at Ramaiah Medical College hospital ('inborn infants'); 2. Infants born elsewhere, and admitted to Ramaiah Medical College hospital under 6 hours of age (outborn infants); 3. Infants with a gestational age at birth (weeks +days) in the range ≥ 28+0 to ≤ 36+6; 4. Infants thought to have RDS (clinically diagnosed after onset of respiratory distress \<6 hours of age, sometimes confirmed by X-ray showing homogenous bilateral opacity) who would routinely be provided CPAP therapy; and 5. Infants \<24 hours old at the time of fulfilling other inclusion criteria.

Exclusion criteria

1. Infants with a 1-minute Apgar score \<3 (as a marker of severe birth asphyxia); 2. Infants who received MV prior to randomisation; 3. Infants with suspected meconium aspiration syndrome will be excluded to avoid any imbalance in this condition across groups; 4. Infants clinically suspected to have another specified serious condition as their main disease process, diagnosed prior to randomisation, specifically: cardiac anomaly, other congenital malformation with respiratory sequelae, septicaemia, pulmonary haemorrhage, pneumothorax, meningitis, poor respiratory effort or recurrent apnoea, or brain haemorrhage (IVH Grades III or IV); 5. Infants who have an airway abnormality precluding the use of the standard CPAP interface proposed for this study (e.g., Pierre-Robin sequence, cleft lip or cleft palate) or who have a neuromuscular condition that interferes with respiration; 6. Any infant whose treating clinician believes should not be randomised due to some other condition, or for any other reason (reason to be documented).

Design outcomes

Primary

Measure	Time frame	Description
FiO2	6, 12, 24, 48 and 72 hours after treatment commencement	Fraction of inspired oxygen (FiO2), measured as a change from baseline as shown on the two machines. Note. Outcome contaminated. An SpO2 target of 90-95% reflects currently recommended practice for neonates; when SpO2 exceeds the target, FiO2 should be reduced. The majority of readings were at SpO2 \> 95%, so FiO2 for these SpO2 readings reflects oxygen provided, not oxygen required to achieve the recommended SpO2 target range (i.e., the FiO2 provided was excessive, by an unknown amount).

Secondary

Measure	Time frame	Description
Respiratory Rate	6, 12, 24, 48 and 72 hours after treatment commencement	Respiratory rate (breaths/minute), measured as a change from baseline
Arterial pH	6, 12, 24, 48 and 72 hours after treatment commencement (where available)	pH measured as a change from baseline (where measured)
Partial Pressure of Arterial Oxygen (PaO2)	6, 12, 24, 48 and 72 hours after treatment commencement (where available)	PaO2 measured as a change from baseline (where measured)
Partial Pressure of Arterial Carbon Dioxide (PaCO2)	6, 12, 24, 48 and 72 hours after treatment commencement (where available)	PaCO2 measured as a change from baseline (where measured)
Number of Participants Who Died or Needed Intubation and/or Mechanical Ventilation, as a Measure of CPAP Failure, Measured to Date and Time of Cessation of CPAP Treatment	From date and time of randomization to date and time of cessation of CPAP treatment, or until date and time of hospital discharge, if infant is on CPAP treatment until discharged (i.e., dies or is transferred), assessed to a maximum of 2 months of age.	Death or need for intubation and mechanical ventilation as demonstrated by an FiO2 requirement ≥ 60% for ≥ 1 hour to maintain SpO2 at 90-95% Note. An SpO2 target of 90-95% reflects currently recommended practice for neonates. FiO2 and SpO2 at time of intubation and ventilation not separately recorded, so we report this endpoint as recorded by clinicians, assuming that they have verified FiO2 \[≥ 60% for ≥ 1 hour\] and SpO2 \[targeting 90-95%\] requirements at the time of intubation and ventilation. If SpO2 was actually \>95% (not 90-95%) at the time of intubation and ventilation, then FiO2 would be higher than required to meet the target range, by an unknown amount (possibly not meeting the FiO2 threshold of ≥ 60% for ≥ 1 hour, required to justify intubation and ventilation). We note this potential source of contamination because the results of Outcome #2 show that the majority of SpO2 readings were \>95%.
Surfactant Provided When FiO2 > 40% to Maintain SpO2 at 90-95% for ≥ 30 Minutes, With Respiratory Distress Syndrome Confirmed by Chest X-Ray	From date and time of randomization to date and time of cessation of CPAP treatment, or until date and time of hospital discharge, if infant is on CPAP treatment until discharged (i.e., dies or is transferred), assessed to a maximum of 2 months of age.	Surfactant provided when FiO2 \> 40% to maintain SpO2 at 90-95% for ≥ 30 minutes, with Respiratory Distress Syndrome confirmed by chest X-Ray Note that endpoint likely corrupted because SpO2 was not routinely targeting 90-95%, and large number of infants received surfactant before achieving \>40% or without having x-ray confirmation of RDS.
CPAP Failure or Surfactant Provision	From date and time of randomization to date and time of cessation of CPAP treatment, or until date and time of hospital discharge, if infant is on CPAP treatment until discharged (i.e., dies or is transferred), assessed to a maximum of 2 months of age.	Outcomes 6 or 7 Note that endpoint likely corrupted because SpO2 was not routinely targeting 90-95%, and large number of infants received surfactant before achieving \>40% or without having x-ray confirmation of RDS.
CPAP Duration	From date and time of randomization to date and time of cessation of CPAP treatment, or until date and time of hospital discharge, if infant is on CPAP treatment until discharged (i.e., dies or is transferred), assessed to a maximum of 2 months of age.	Duration of CPAP treatment (hours) in infants that do not fail CPAP
SpO2	6, 12, 24, 48 and 72 hours after treatment commencement	Oxygen saturation by pulse oximetry (SpO2), measured as a change from baseline Note. Outcome contaminated. An SpO2 target of 90-95% reflects currently recommended practice for neonates. The majority of readings were at SpO2 \> 95%, above SpO2 target (90-95%).
Sentinel Outcome #1	From date and time of randomization to date and time of CPAP cessation, assessed to a maximum of 2 months of age.	Damage to the nasal septum
Sentinel Outcome #2	From date and time of randomization to date and time of CPAP cessation, assessed to a maximum of 2 months of age.	Damage to the nares of the infant
Sentinel Outcome #3	From date and time of randomization to date and time of CPAP cessation, assessed to a maximum of 2 months of age.	Pneumothorax as diagnosed by X-ray
Sentinel Outcome #4	From date and time of randomization to date and time of CPAP cessation, assessed to a maximum of 2 months of age.	Intra-ventricular haemorrhage (IVH), intra-cranial haemorrhage (ICH), or periventricular leukomalacia (PVL) as diagnosed by cranial ultrasound scan
Sentinel Outcome #5	From date and time of randomization to date and time of CPAP cessation, assessed to a maximum of 2 months of age.	FiO2 ≥ 60% to maintain SpO2 at 90-95% for one hour or more, during CPAP treatment
Sentinel Outcome #6	From date and time of randomization to date and time of CPAP cessation, assessed to a maximum of 2 months of age.	Intubation and mechanical ventilation Note. An SpO2 target of 90-95% reflects currently recommended practice for neonates. FiO2 and SpO2 at time of intubation and ventilation not separately recorded, so we report this endpoint as recorded by clinicians, assuming that they have verified FiO2 \[≥ 60% for ≥ 1 hour\] and SpO2 \[targeting 90-95%\] requirements at the time of intubation and ventilation. If SpO2 was actually \>95% (not 90-95%) at the time of intubation and ventilation, then FiO2 would be higher than required to meet the target range, by an unknown amount (possibly not meeting the FiO2 threshold of ≥ 60% for ≥ 1 hour, required to justify intubation and ventilation). We note this potential source of contamination because the results of Outcome #2 show that the majority of SpO2 readings were \>95%.
Sentinel Outcome #7	From date and time of randomization to date and time of CPAP cessation, assessed to a maximum of 2 months of age.	For infants born at 28+0 to 33+6 weeks' gestation, oxygen dependent at 36 weeks' gestation
Sentinel Outcome #8	From date and time of randomization to date and time of hospital discharge, assessed to a maximum of 2 months of age.	Death of the infant before hospitals discharge
Serious Adverse Event	From date and time of birth to date and time of hospital discharge, assessed to a maximum of 2 months of age.	A Serious adverse Event (SAEs) is any untoward medial occurrence that: * Results in death; * Is life-threatening; \[NOTE: the term life-threatening in the definition of serious refers to an event in which the patient was at risk of death at the time of the event; it does not refer to an event which hypothetically might have caused death if it were more severe.\] * Requires inpatient hospitalisation or prolongation of existing hospitalisation; * Results in persistent or significant disability/incapacity, or * Is a congenital anomaly/birth defect

Countries

India

Participant flow

Pre-assignment details

There were no wash-out or run-in procedures. No enrolled participant was excluded from the study before randomisation.

Participants by arm

Arm	Count
Dolphin Continuous Positive Airway Pressure Continuous Positive Airway Pressure (CPAP) machine with oxygen intake and an on-board air compressor, and integrated blender and humidifier, delivering heated humidified blended gases through a Fisher-Paykel nasal mask. \[Also includes built-in Massimo pulse oximeter, though this is not used for this study, to prevent differential measurement error in SpO2 measurement\] CPAP: CPAP therapy	26
Fisher-Paykel Continuous Positive Airway Pressure Fisher-Paykel (F&P) Continuous Positive Airway Pressure (CPAP) machine with separate oxygen and medical air intakes, with third-party FP-compliant blender, and F&P blender, delivering heated humidified blended gases through a Fisher-Paykel nasal mask. CPAP: CPAP therapy	25
Total	51

Baseline characteristics

Characteristic	Fisher-Paykel Continuous Positive Airway Pressure	Total	Dolphin Continuous Positive Airway Pressure
Age, Customized Gestational age	31.57 Weeks	31.71 Weeks	31.86 Weeks
Birthweight	1434 Grams STANDARD_DEVIATION 417	1421 Grams STANDARD_DEVIATION 363	1409 Grams STANDARD_DEVIATION 311
FiO2 at baseline	40 Percentage of inspired oxygen	40 Percentage of inspired oxygen	40 Percentage of inspired oxygen
Partial pressure of arterial carbon dioxide at baseline	27.5 Torr	37.2 Torr	38.9 Torr
Partial pressure of arterial oxygen at baseline	139 Torr	115 Torr	115 Torr
pH at baseline	7.34 pH	7.29 pH	7.29 pH
Race and Ethnicity Not Collected	—	0 Participants	—
Region of Enrollment India	26 Participants	51 Participants	25 Participants
Respiratory rate at baseline	43 Breaths/minute	43 Breaths/minute	42.5 Breaths/minute
Sex: Female, Male Female	9 Participants	17 Participants	8 Participants
Sex: Female, Male Male	16 Participants	34 Participants	18 Participants
SpO2 at baseline	97 Percentage capillary oxygen saturation	97 Percentage capillary oxygen saturation	96 Percentage capillary oxygen saturation

Adverse events

Event type	EG000 affected / at risk	EG001 affected / at risk
deaths Total, all-cause mortality	1 / 26	0 / 25
other Total, other adverse events	4 / 26	2 / 25
serious Total, serious adverse events	1 / 26	0 / 25

Outcome results

Primary

FiO2

Fraction of inspired oxygen (FiO2), measured as a change from baseline as shown on the two machines. Note. Outcome contaminated. An SpO2 target of 90-95% reflects currently recommended practice for neonates; when SpO2 exceeds the target, FiO2 should be reduced. The majority of readings were at SpO2 \> 95%, so FiO2 for these SpO2 readings reflects oxygen provided, not oxygen required to achieve the recommended SpO2 target range (i.e., the FiO2 provided was excessive, by an unknown amount).

Time frame: 6, 12, 24, 48 and 72 hours after treatment commencement

Population: Number missing at each time point:~6 hrs: 1 Dolphin (failed), 0 FP~12 hrs: 1 Dolphin (failed), 1 FP (recovered)~24 hrs: 6 Dolphin (2 failed + 4 recovered), 4 FP (all recovered)~48 hrs: 16 Dolphin (3 failed + 13 recovered), 11 FP (1 failed + 10 recovered)~72 hrs: 21 Dolphin (3 failed + 18 recovered), 21 FP (1 failed + 20 recovered)

Arm	Measure	Group	Value (MEDIAN)
Dolphin CPAP	FiO2	Change at 12 hours	0 Percentage of inspired oxygen
Dolphin CPAP	FiO2	Change at 48 hours	-14.5 Percentage of inspired oxygen
Dolphin CPAP	FiO2	Change at 24 hours	-10 Percentage of inspired oxygen
Dolphin CPAP	FiO2	Change at 72 hours	-10 Percentage of inspired oxygen
Dolphin CPAP	FiO2	Change at 6 hours	0 Percentage of inspired oxygen
Fisher-Paykel CPAP	FiO2	Change at 72 hours	-7.5 Percentage of inspired oxygen
Fisher-Paykel CPAP	FiO2	Change at 6 hours	0 Percentage of inspired oxygen
Fisher-Paykel CPAP	FiO2	Change at 12 hours	0 Percentage of inspired oxygen
Fisher-Paykel CPAP	FiO2	Change at 24 hours	-15 Percentage of inspired oxygen
Fisher-Paykel CPAP	FiO2	Change at 48 hours	-12.5 Percentage of inspired oxygen

Comparison: Analysis restricted to change at 6 hours, as least affected by missing data.~Distribution non-normal, so non-parametric test (Wilcoxon) used.p-value: 0.65van Elteren's extension to Wilcoxon

Comparison: Planned sub-group analysis for infants born at 28-33 gestational weeks (n=41). No adjustment for multiple comparisons.p-value: 0.64Wilcoxon (Mann-Whitney)

Comparison: Planned sub-group analysis for infants born at 34-36 gestational weeks (n=10). No adjustment for multiple comparisons.p-value: 1Wilcoxon (Mann-Whitney)

Secondary

Arterial pH

pH measured as a change from baseline (where measured)

Time frame: 6, 12, 24, 48 and 72 hours after treatment commencement (where available)

Population: Only 31/51 infants had a baseline blood gas taken, and only 7 of these had a second or subsequent test, with five of them at 6 hours, two each at 12 and 24 hours and one each at 48 and 72 hours. No statistical testing performed.

Arm	Measure	Group	Value (MEDIAN)
Dolphin CPAP	Arterial pH	Change at 24 hours	0.015 pH
Dolphin CPAP	Arterial pH	Change at 12 hours	0.19 pH
Dolphin CPAP	Arterial pH	Change at 6 hours	0.135 pH
Fisher-Paykel CPAP	Arterial pH	Change at 72 hours	-0.01 pH
Fisher-Paykel CPAP	Arterial pH	Change at 12 hours	0.23 pH
Fisher-Paykel CPAP	Arterial pH	Change at 6 hours	0.05 pH
Fisher-Paykel CPAP	Arterial pH	Change at 48 hours	0.01 pH

Secondary

CPAP Duration

Duration of CPAP treatment (hours) in infants that do not fail CPAP

Time frame: From date and time of randomization to date and time of cessation of CPAP treatment, or until date and time of hospital discharge, if infant is on CPAP treatment until discharged (i.e., dies or is transferred), assessed to a maximum of 2 months of age.

Population: Excludes 3 Dolphin and 2 FP randomised infants that failed treatment (escalated to mechanical ventilation)

Arm	Measure	Value (MEDIAN)
Dolphin CPAP	CPAP Duration	46.9 Hours
Fisher-Paykel CPAP	CPAP Duration	56.5 Hours

p-value: 0.81van Elteren's extension to Wilcoxon

Secondary

CPAP Failure or Surfactant Provision

Outcomes 6 or 7 Note that endpoint likely corrupted because SpO2 was not routinely targeting 90-95%, and large number of infants received surfactant before achieving \>40% or without having x-ray confirmation of RDS.

Arm	Measure	Value (COUNT_OF_PARTICIPANTS)
Dolphin CPAP	CPAP Failure or Surfactant Provision	6 Participants
Fisher-Paykel CPAP	CPAP Failure or Surfactant Provision	3 Participants

p-value: 0.31Cochran-Mantel-Haenszel

Secondary

Number of Participants Who Died or Needed Intubation and/or Mechanical Ventilation, as a Measure of CPAP Failure, Measured to Date and Time of Cessation of CPAP Treatment

Death or need for intubation and mechanical ventilation as demonstrated by an FiO2 requirement ≥ 60% for ≥ 1 hour to maintain SpO2 at 90-95% Note. An SpO2 target of 90-95% reflects currently recommended practice for neonates. FiO2 and SpO2 at time of intubation and ventilation not separately recorded, so we report this endpoint as recorded by clinicians, assuming that they have verified FiO2 \[≥ 60% for ≥ 1 hour\] and SpO2 \[targeting 90-95%\] requirements at the time of intubation and ventilation. If SpO2 was actually \>95% (not 90-95%) at the time of intubation and ventilation, then FiO2 would be higher than required to meet the target range, by an unknown amount (possibly not meeting the FiO2 threshold of ≥ 60% for ≥ 1 hour, required to justify intubation and ventilation). We note this potential source of contamination because the results of Outcome #2 show that the majority of SpO2 readings were \>95%.

Arm	Measure	Value (COUNT_OF_PARTICIPANTS)
Dolphin CPAP	Number of Participants Who Died or Needed Intubation and/or Mechanical Ventilation, as a Measure of CPAP Failure, Measured to Date and Time of Cessation of CPAP Treatment	3 Participants
Fisher-Paykel CPAP	Number of Participants Who Died or Needed Intubation and/or Mechanical Ventilation, as a Measure of CPAP Failure, Measured to Date and Time of Cessation of CPAP Treatment	2 Participants

p-value: 1Fisher Exact

Secondary

Partial Pressure of Arterial Carbon Dioxide (PaCO2)

PaCO2 measured as a change from baseline (where measured)

Time frame: 6, 12, 24, 48 and 72 hours after treatment commencement (where available)

Arm	Measure	Group	Value (MEDIAN)
Dolphin CPAP	Partial Pressure of Arterial Carbon Dioxide (PaCO2)	Change at 24 hours	-12 Torr
Dolphin CPAP	Partial Pressure of Arterial Carbon Dioxide (PaCO2)	Change at 12 hours	-18.6 Torr
Dolphin CPAP	Partial Pressure of Arterial Carbon Dioxide (PaCO2)	Change at 6 hours	-17 Torr
Fisher-Paykel CPAP	Partial Pressure of Arterial Carbon Dioxide (PaCO2)	Change at 72 hours	6.4 Torr
Fisher-Paykel CPAP	Partial Pressure of Arterial Carbon Dioxide (PaCO2)	Change at 12 hours	-5.9 Torr
Fisher-Paykel CPAP	Partial Pressure of Arterial Carbon Dioxide (PaCO2)	Change at 6 hours	-9 Torr
Fisher-Paykel CPAP	Partial Pressure of Arterial Carbon Dioxide (PaCO2)	Change at 48 hours	-0.1 Torr

Secondary

Partial Pressure of Arterial Oxygen (PaO2)

PaO2 measured as a change from baseline (where measured)

Time frame: 6, 12, 24, 48 and 72 hours after treatment commencement (where available)

Arm	Measure	Group	Value (MEDIAN)
Dolphin CPAP	Partial Pressure of Arterial Oxygen (PaO2)	Change at 24 hours	-42.5 Torr
Dolphin CPAP	Partial Pressure of Arterial Oxygen (PaO2)	Change at 12 hours	-60.2 Torr
Dolphin CPAP	Partial Pressure of Arterial Oxygen (PaO2)	Change at 6 hours	-138 Torr
Fisher-Paykel CPAP	Partial Pressure of Arterial Oxygen (PaO2)	Change at 72 hours	15.6 Torr
Fisher-Paykel CPAP	Partial Pressure of Arterial Oxygen (PaO2)	Change at 12 hours	12.1 Torr
Fisher-Paykel CPAP	Partial Pressure of Arterial Oxygen (PaO2)	Change at 6 hours	-47 Torr
Fisher-Paykel CPAP	Partial Pressure of Arterial Oxygen (PaO2)	Change at 48 hours	-9.3 Torr

Secondary

Respiratory Rate

Respiratory rate (breaths/minute), measured as a change from baseline

Time frame: 6, 12, 24, 48 and 72 hours after treatment commencement

Population: Number missing (and thus imputed):~6 hrs: 1 Dolphin (failed), 0 FP~12 hrs: 1 Dolphin (failed), 1 FP (recovered)~24 hrs: 6 Dolphin (2 failed + 4 recovered), 4 FP (all recovered)~48 hrs: 16 Dolphin (3 failed + 13 recovered), 11 FP (1 failed + 10 recovered)~72 hrs: 21 Dolphin (3 failed + 18 recovered), 21 FP (1 failed + 20 recovered)

Arm	Measure	Group	Value (MEDIAN)
Dolphin CPAP	Respiratory Rate	Change at 6 hours	-2 Breaths per minute
Dolphin CPAP	Respiratory Rate	Change at 48 hours	-1 Breaths per minute
Dolphin CPAP	Respiratory Rate	Change at 12 hours	0 Breaths per minute
Dolphin CPAP	Respiratory Rate	Change at 72 hours	0 Breaths per minute
Dolphin CPAP	Respiratory Rate	Change at 24 hours	-2 Breaths per minute
Fisher-Paykel CPAP	Respiratory Rate	Change at 72 hours	0 Breaths per minute
Fisher-Paykel CPAP	Respiratory Rate	Change at 24 hours	0 Breaths per minute
Fisher-Paykel CPAP	Respiratory Rate	Change at 6 hours	0 Breaths per minute
Fisher-Paykel CPAP	Respiratory Rate	Change at 12 hours	0 Breaths per minute
Fisher-Paykel CPAP	Respiratory Rate	Change at 48 hours	0 Breaths per minute

Comparison: Analysis restricted to change at 6 hours, as least affected by missing data.~Distribution non-normal, so non-parametric test (Wilcoxon) used.p-value: 0.86van Elteren's extension to Wilcoxon

Secondary