Continuous positive airway pressure (CPAP) provides extra gas flow through the nose and helps reduce breathing problems in preterm babies after the tube used to assist breathing is removed from their windpipe.
Preterm babies (born before 37 weeks) may need help to breathe properly. Sometimes this is given via a tube placed into the windpipe, through the mouth or nose, to give oxygen from a machine (mechanical ventilation). This method helps restore breathing but when the tube is removed (this process of removal is called extubation), breathing problems can occur. Nasal continuous positive airways pressure (NCPAP) provides extra gas flow through the nose. The review of trials found NCPAP is effective in preventing failure of extubation after a period of mechanical ventilation.
Preterm infants being extubated following a period of intermittent positive pressure ventilation via an endotracheal tube are at risk of developing respiratory failure as a result of apnea, respiratory acidosis and hypoxia. Nasal continuous positive airway pressure appears to stabilise the upper airway, improve lung function and reduce apnea and may therefore have a role in facilitating extubation in this population.
In preterm infants having their endotracheal tube removed following a period of intermittent positive pressure ventilation (IPPV), does management with nasal continuous positive airways pressure (NCPAP) lead to an increased proportion remaining free of additional ventilatory support, compared to extubation directly to headbox oxygen?
Searches were made of the Oxford Database of Perinatal Trials, MEDLINE up to November 2002, Cochrane Controlled Trials Register (The Cochrane Library, Issue 4, 2002), previous reviews including cross references, abstracts of conferences and symposia proceedings, expert informants and journal handsearching mainly in the English language.
All trials utilising random or quasi-random patient allocation, in which NCPAP (delivered by any method) was compared with headbox oxygen for post-extubation care were included. Methodological quality was assessed independently by the two authors.
Data were extracted independently by the two authors. Prespecified subgroup analysis to determine the impact of different levels of NCPAP, differences in duration of IPPV and use of aminophylline were also performed using the same package. Data were analysed using relative risk (RR), risk difference (RD) and number needed to treat (NNT).
Nasal CPAP, when applied to preterm infants being extubated following IPPV, reduces the incidence of adverse clinical events (apnea, respiratory acidosis and increased oxygen requirements) indicating the need for additional ventilatory support [RR 0.62 (0.49, 0.77), RD -0.17 (-0.24,-0.10), NNT 6 (4,10)].
Implications for practice: nasal CPAP is effective in preventing failure of extubation in preterm infants following a period of endotracheal intubation and IPPV.
Implication for research: further definition of the gestational age and weight groups in whom these results apply is required. Optimal levels of NCPAP as well as methods of administration remain to be determined.
Preterm infants frequently receive intermittent positive pressure ventilation (IPPV) in the days following birth. Efforts to limit the duration of IPPV, and thereby minimise its attendant risk, lead to attempts at early weaning of ventilatory support and removal of the endotracheal tube. Failure to make the transition to spontaneous breathing following extubation is a common occurrence and may be due in part to a tendency towards alveolar atelectasis and to relatively poor respiratory drive, both characteristics of increasing prematurity. Clinical indications of respiratory failure post-extubation include: 1) apneic episodes, defined by the National Institute of Health (NIH 1987) as pauses greater than 20 seconds duration or shorter episodes associated with bradycardia, cyanosis or pallor and 2) respiratory acidosis or 3) increasing oxygen requirements. These problems may lead to reintubation and ventilation with their attendant morbidity and economic costs.
Nasal continuous positive airways pressure (NCPAP) appears to stabilise the upper airway, improve lung function and reduce apnea. A number of randomised trials address the clinically important question of whether CPAP, applied prophylactically, reduces the need for additional ventilatory support.
The main objective was to assess whether NCPAP, applied after extubation of preterm infants being weaned from IPPV, reduces the incidence of respiratory failure (as defined above) leading to the need for additional ventilatory support, without clinically important side effects. Additional ventilatory support comprised the reintroduction of endotracheal intubation for IPPV, or rescue therapy with NCPAP in infants failing initial treatment with headbox in trials where this was offered. The need for endotracheal reintubation alone was evaluated for all trials and also for the subset of trials in which rescue NCPAP was permitted.
In prespecified subgroups, analyses were conducted to determine whether
the results differed for those infants:
1) also given methylxanthines
2) extubated in the first 14 days
3) given less than 5 cms of water CPAP compared with those given higher
pressures
In post hoc subgroups, analyses were conducted to determine whether the
results differed for those infants:
1) weighing less than 2000g
All trials utilising random or quasi-random patient allocation.
Preterm infants being weaned from IPPV.
NCPAP compared with head box oxygen after extubation.
1) failure of extubation as defined by respiratory acidosis, increased
oxygen requirement or apnea that is frequent or severe leading to additional
ventilatory support;
2) the need for endotracheal reintubation
3) oxygen use at 28 days of age.
Searches were made of the Oxford Database of Perinatal Trials, Cochrane
Controlled Trials Register (The Cochrane Library, Issue 4, 2002), MEDLINE
up to November 2002 [infant, newborn (exp), positive pressure respiration
(exp) and random*], previous reviews including cross references, abstracts,
conferences and symposia proceedings of the Society for Pediatric Research
and the European Society for Pediatric Research from 1993-2002, expert informants
and journal handsearching mainly in the English language.
Criteria and methods used to assess the methodological quality of the included trials: standard method of the Cochrane Collaboration and its Neonatal Review Group was used. Methodological quality of each trial was reviewed by the second author blinded to trial authors and institutions. Additional information was requested from the authors of each trial to clarify methodology.
Methods used to collect data from the included trials: each author extracted data separately before comparison and resolution of differences. Additional data requested from authors.
Methods used to synthesise data: The statistical analysis used a fixed effect model and calculated the relative risk (RR), risk difference (RD) and number needed to treat (NNT) with 95% confidence intervals.
Details are given in the table of included studies and in the reference list. Trials included; Engelke 1982, Higgins 1991, Chan 1993, So 1995, Annibale 1994, Tapia 1995, Davis 1998 and Dimitriou 2000.
Four studies (Higgins 1991, Engelke 1982, Davis 1998 and Dimitriou 2000) allowed crossover from headbox to NCPAP if an infant was deemed to have failed. These infants were classified as treatment failures, even if they did not have IPPV reintroduced.
Methodological quality was assessed using the criteria described in 'Effective Care of the Newborn Infant' (Sinclair 1992). Additional details of each study appear in the table of included studies.
Method of subject allocation:
All trials, with the exception of Tapia (1995), used random allocation for
group assignment. Tapia used a method of quasi-random allocation, described
as 'systematic sampling' which the author clarified as sequential case allocation.
Chan (1993), Higgins (1991), Engelke (1982), Annibale (1994), Davis (1998)
and Dimitriou (2000) specified the use of sealed envelopes.
Masking of care-givers:
Because of the nature of the interventions, masking of group assignment
was not undertaken. The care-givers were therefore aware of the group to
which each infant was allocated and the possibility of co-intervention bias
exists.
Completeness of outcome assessment:
This was achieved for all patients randomised in six of the studies. Tapia
(1995) did not report the outcome in three of 90 patients because of 'accidental
extubation'. Two of the 126 patients in Annibale's (1994) study were withdrawn,
one at parents request and one following recognition of a predefined exclusion
criterion. Outcome assessment was therefore complete or near complete in all
studies.
Masking of outcome assessors:
The treatment method to which each baby was assigned was known to those
assessing whether or not the baby met 'failure' criteria. Although the time
period over which this was assessed varied from one day (Engelke 1982) to
seven days (Annibale 1994 and Davis 1998), the criteria for failure were
objective and consistent across the studies. These included respiratory acidosis
(defined in terms of pH and pCO2), increased oxygen requirements or recurrent
minor or single major apnoeic events. The use of such predetermined criteria
addresses but does not eliminate the potential for bias in these unblinded
trials. Annibale (1994) and Engelke (1982) included the less objective 'clinical
deterioration' in their failure criteria. The other important outcome, assessed
by Higgins (1991), Annibale (1994), So (1995), Davis (1998) and Dimitriou
(2000) was chronic lung disease, defined as oxygen dependency at 28 days of
life. This assessment was also made in an unblinded manner.
The trial of Engelke (1982) had 18 participants, the other studies involved between 50 and 150 infants.
Independent assignment of inclusion/exclusion status, methodological quality scores and data extraction resulted in no disagreement between the two reviewers.
Infants extubated to NCPAP experience a reduction in respiratory failure leading to the need for additional ventilatory support [RR 0.62 (0.49, 0.77), RD -0.17 (-0.24, -0.10)]. Hence treatment of 6 (4,10) infants with NCPAP is required to prevent one extubation failure. The use of NCPAP as rescue therapy for infants failing headbox in 4 trials "dilutes" this effect for the outcome of need for endotracheal intubation. There is a trend towards a decreased requirement for endotracheal reintubation when NCPAP is used which does not reach statistical significance [RR 0.93 (0.72, 1.19), RD -0.02 (-0.09, 0.05)]. Exclusion of the trial of Engelke from these analyses (larger infants, shorter time to 'failure', no aminophylline) does not change these findings. There was no significance difference in rates of oxygen use at 28 days of age in patients allocated to NCPAP in the five trials where this outcome was examined [RR 1.00 (0.81, 1.24), RD 0.00 (-0.09, 0.09)].
If analysis is restricted to those infants who were extubated in the first 14 days of life, there is also a reduction in failure of extubation in the NCPAP group [RR 0.58 (0.44, 0.75), RD -0.19 (-0.27, -0.10)] but no significant difference in oxygen use at 28 days of age [RR 1.17 (0.91, 1.50), RD 0.06 (-0.03, 0.16)]. Pooled analysis of the 2 trials (66 infants) in which infants were extubated after 14 days showed no significant difference in failure of extubation [RR 1.12 (0.58,2.15), RD 0.04 (-0.19, 90.27)].
Secondary analysis of the studies excluding that of Tapia (on the basis that patient allocation was not truly random) strengthens the conclusion that extubation to NCPAP leads to a lower 'failure' rate than extubation to headbox oxygen [RR 0.57 (0.46, 0.72), RD -0.20 (-0.28, -0.13), NNT 5 (4,8]. Excluding this study reduces but does not remove the heterogeneity apparent in the meta-analysis.
While no direct comparison has been made between different levels of NCPAP, it is notable that in the two trials using levels less than 5 cm water pressure no significant difference in failure rate was demonstrated [RR 1.09 (0.69, 1.73), RD 0.024 (-0.104, 0.152)] whereas in trials permitting levels of 5 cm and above, there was a significant result favouring NCPAP [RR 0.49 (0.37, 0.66) RD -0.29 (-0.40, -0.19), NNT 3 (3,5)] . Dimitriou 2000 was not included in this subanalysis as the CPAP level administered (3 to 5 cm water) included the predefined cutpoint.
In the four trials allowing crossover of treatment groups, there was no significant difference in requirement for endotracheal reintubation [RR 1.04 (0.70, 1.53), RD 0.01 (-0.08, 0.10)].
Although results varied from trial to trial, there is a clinically important advantage for infants extubated to nasal CPAP. The heterogeneity apparent in the overall analysis can be partly explained by the prespecified subgroup analysis into trials that used higher or lower levels of pressure. However, in the absence of trials directly comparing level of NCPAP, caution should be exercised in the interpretation of these findings.
The mechanism of action of NCPAP is uncertain but could be one or more of the following: maintenance of lung volume, ensuring maximum upper airway patency or airflow stimulation of the nose or upper airway, decreasing apnea of prematurity.
It may be anticipated that the benefits of NCPAP might vary between infants of differing weights and degrees of immaturity. Subgroup analysis of the studies involving infants weighing less than 2000g shows the same advantage of NCPAP. Further refinement of weight and gestational age specific recommendations pertaining to the method of postextubation care could be undertaken in future trials.
Sensitivity analysis, including only the trials that permitted rescue therapy for infants failing headbox treatment, raises the question of whether the delayed use of NCPAP confers any disadvantage to infants allocated to headbox. It may be argued that the stability afforded by NCPAP, that is the prevention of failure criteria (apnea, increased oxygen requirements, respiratory acidosis), make it the desirable form of initial postextubation care. However there is no statistically significant difference in rates of endotracheal reintubation [RR 1.04 (0.70, 1.53), RD 0.01 (-0.08, 0.10)]. Robertson (1998) compared "elective" with rescue NCPAP after extubation and concluded that rescue therapy could be safely used because there were no differences in number of ventilator days, days on CPAP or rates of successful extubation. Individual units, particularly those with limited resources, may use these findings to justify a policy of having infants "earn" their NCPAP.
The appropriate duration of treatment with NCPAP remains uncertain, as does the method of its weaning. There was heterogeneity across studies with respect to method of delivery of NCPAP. Direct comparison of different methods of delivery should be undertaken in future studies.
NCPAP is effective in preventing respiratory failure in preterm infants following a period of endotracheal intubation and IPPV but does not influence the rate of supplemental oxygen dependency at 28 days of age. NCPAP of 5 cm water or more appears to be more effective than NCPAP at lower pressures, although direct comparison has not been made.
Further definition of the gestational age and weight groups in whom these results apply is required. Optimal levels of nasal CPAP as well as methods and duration of administration remain to be determined.
Additional information was kindly provided by Prof. Anne Greenough for the Chan 1993 trial, Dr Tapia for the Tapia 1995 trial and Dr Dimitriou for the Dimitriou 2000 trial.
Peter Davis is first author of a trial included in this review.
| Study | Methods | Participants | Interventions | Outcomes | Notes | Allocation concealment |
| Annibale 1994 | Random allocation by opaque sealed envelopes, using random number
table. Stratification by birth-weight (3 groups) and surfactant usage to give
6 strata. Blinding of intervention and outcome assessment not attempted. Adequate followup. |
Eligible if birth weight 600 to 1500 g, endotracheal intubation
within 48 hours of life, eligible for extubation by day 14 of life and stable
or improving clinical course. Excluded if chromosomal anomalies, neuromuscular
diseases, congenital anomalies or other surgical conditions. Number enrolled = 82. |
Three treatment regimens: 1) nasopharyngeal CPAP at 6 cm water until reached (defined) goal criteria, 2) nasopharyngeal CPAP for 6 hours then headbox oxygen and 3) headbox oxygen | Success defined as the ability to remain free of the need for mechanical ventilation for 7 days after extubation and the achievement of specified goal criteria. Failure defined as an oxygen requirement of 80% or more, pH < 7.20, severe apnea and bradycardia or clinical deterioration. CLD defined as oxygen use at 28 days. | Extubation from low rate ventilation. | A |
| Chan 1993 | Randomised using sealed envelopes. Blinding of intervention and outcome assessment not attempted. Adequate followup. |
Ventilated infants less than 1800g with no congenital abnormalities.
Stratified into 2 groups: postnatal age less than 14 days or greater than
or equal to 14 days. Number enrolled = 120. |
NCPAP of 3 cm water or headbox oxygen | Success defined as not requiring reintubation for 48 hours after extubation. Criteria for reintubation - ph < 7.25 with pCO2 >6.67 kPa, inspired oxygen >60% or recurrent minor or one major apnea. | Aminophylline load given. Extubated after 1 hour ETT CPAP. | A |
| Davis 1998 | Random allocation using sequentially numbered, sealed, opaque
envelopes using a random number table. Stratified by gender, birthweight
and postnatal age giving 8 strata. Blinding of intervention and outcome assessment not attempted. Adequate followup. |
Inclusion criteria: inborn and outborn infants admitted to a single
centre with 1) birthweight 600 to 1250 g 2) endotracheal tube for > 12 hours 3) stable or improving respiratory status - ventilator rate < 20/minute and inspired oxygen requirement < 50%. Number enrolled = 92. |
Allocated to receive either nasal CPAP, set at 7 cm water, via a size 2.5 or 3.0 Portex tube inserted 2.5 cm into one nostril and connected to a Bear Cub ventilator or headbox oxygen | Successful extubation determined over 7 days postextubation. Failure criteria comprised apnea (recurrent minor or 1 major), increased oxygen requirement (>15% absolute increase above that required pre-extubation or respiratory acidosis (pH<7.25 with pCO2>50 mm Hg) | Extubated from low rate ventilation after loading with aminophylline. Trial stopped early after second planned interim analysis demonstrated significant benefit to NCPAP group using the O'Brien -Fleming stopping rule. | A |
| Dimitriou 2000 | Random allocation using sequentially numbered, sealed opaque envelopes. Blinding of intervention and outcome assessment not attempted. Adequate followup. |
Included inborn and outborn infants considered by the clinician
to be ready for extubation. Gestational age less than or equal to 34 weeks
and postnatal age less than or equal to 14 days. Methylxanthines were commenced
universally before extubation. Number enrolled = 150. |
Allocated to receive either nasal CPAP, set at 3 to 5 cm water or headbox oxygen. CPAP provided either by single prong (n=40), Argyle prongs (n=22) or Flow Driver prongs (n=13) depending on availability in individual units. | Succesful extubation determined over 48 hours post extubation. Failure criteria comprised oxygen requirement > 60%, respiratory acidosis (pH<7.25) and one major or frequent minor apnoeic episodes. | Extubation occurred after a period of endotracheal CPAP up to one hour. | A |
| Engelke 1982 | Random allocation using shuffled envelopes. Blinding of intervention and outcome assessment not attempted. Adequate followup. |
Inclusion criteria - all neonates intubated for more than 72 hours
but less than 14 days. Exclusion criteria - multiple or traumatic intubations, tracheal stenosis, chronic lung changes, neurological conditions, congestive heart failure, birthweight less than 1.0 kg, surgical conditions or major congenital anomalies. Enrolled 18 infants, mean birthweight 2 kg and mean duration of ventilation 6 days. Number enrolled = 18. |
Allocated to either NCPAP at 6 cm H2O or headbox oxygen. | Extubation failure defined as 1) progressive atelectasis and respiratory distress, 2) pCO2 > 60 mm Hg with pH < 7.20, or increased oxygen requirement by >0.15 - 0.20. Crossover from hood to NCPAP allowed. | This study performed in the presurfactant era and the use of aminophylline was 'avoided'. Enrolment was discontinued after 18 patients because of 'considerably better' postextubation courses in the CPAP group. | A |
| Higgins 1991 | Random allocation using sealed envelopes. Blinding of intervention and outcome assessment not attempted. Adequate followup. |
Infants weighing < 1 kg at time of first elective extubation.
Intubated at least 24 hours, requiring < 35% oxygen, MAP <7 cm H2O,
ventilator rate < 20 breaths per minute and weight at least 80% of birth-weight. Number enrolled = 58. |
Silicone binasal pharyngeal airway prongs using 4 to 6 cm H2O of NCPAP or headbox oxygen | Success defined as requiring no additional respiratory support for at least 5 days. Failure criteria: FiO2 >0.60 to maintain oxygen saturations > 93%, pCO2 >60 or pH <7.23, or moderate to severe apnea (defined). CLD defined as oxygen use at 28 days. | A | |
| So 1995 | Random allocation. Blinding of intervention and outcome assessment not attempted. Adequate followup. |
Birth weight < 1500g, GA < 34 weeks, mechanical ventilation
within first few hours of life, weaning started within 7 days of life, absence
of overwhelming infection and congenital abnormalities. Number enrolled = 50. |
Single prong NCPAP at 5 cm water or headbox oxygen | Success defined as reintubation not required for 72 hours after extubation. Reintubation if pO2 <50 mm Hg with FiO2 >70%, pH <7.25 and pCO2 >60, severe or frequent apnea. CLD defined as oxygen use at 28 days with an abnormal chest xray. | Aminophylline load given. Extubation from ETT CPAP. | B |
| Tapia 1995 | Assigned by alternate case. Blinding of intervention and outcome assessment not attempted. Adequate followup. |
Preterm infants < 1500g requiring ventilation for more than
48 hours. Criteria for entry: FiO2 <0.40, rate < 20/min, peak pressure
< 15 cm water. Excluded infants with sepsis, PDA, NEC, grade 3 or 4 IVH
or other severe neurological damage. Number enrolled = 59. |
Three groups: 1) direct extubation to headbox oxygen 2) pre-extubation ETT CPAP (3 to 4 cm H2O) for 12 -24 hours 3) Postextubation nasopharyngeal CPAP (3 to 4 cm H2O) for 12 - 24 hours. |
Failure defined as need to return to mechanical ventilation within 72 hours of extubation because of either 1) frequent or severe apnea, 2) pH < 7.25 and pCO2 >60 mm Hg or 3)FiO2 >0.60 | Extubated from low rate ventilation. Loaded with aminophylline. | C |
| Study | Reason for exclusion |
| Andreasson 1988 | This well conducted cross-over trial evaluated the short term benefits of CPAP administered via a face-mask. Infants experienced less apnea and were better oxygenated during their 10 minute trial periods of CPAP. This study was not included because it did not address the issues of longer term therapy with NCPAP vs headbox. |
| Robertson | |
| Robertson 1998 | This randomised trial compared two strategies of postextubation delivery of CPAP: weaning vs rescue. The criteria for rescue CPAP were set "lower" than those for endotracheal reintubation. In fact, all or nearly all infants in the rescue group receivedCPAP. This differed from other included trials in which rescue of headbox infants was permitted. In these trials, infants were deemed o have reached reintubation criteria and counted as failures of therapy but could be tried on CPAP at the clinician's discretion. The question asked by Robertson is "does it matter when CPAP is started after extubation" rather than that posed by this review. |
Annibale D, Hulsey T, Engstrom P, Wallin L, Ohning B. Randomised, controlled trial of nasopharyngeal continuous positive airways pressure in the extubation of very low birthweight infants. J Pediatr 1994;124:455-60.
Chan 1993 {published data only}
Chan V, Greenough A. Randomised trial of methods of extubation in acute and chronic respiratory distress. Arch Dis Child 1993;68:570-572.
Davis 1998 {published data only}
Davis P, Jankov R, Doyle L, Henschke P. Randomised, controlled trial of nasal continuous positive airway pressure in the extubation of infants weighing 600 to 1250g. Arch Dis Child Fetal Neonatal Ed 1998;79:F54-F57.
Dimitriou 2000 {published data only}
Dimitriou G, Greenough A, Kavvadia V, Laubscher, Alexiou C, Pavlou V, Mantagos S. Elective use of nasal continuous positive airways pressure following extubation of preterm infants. Eur J Pediatr 2000;159:434-439.
Engelke 1982 {published data only}
Engelke SC, Roloff DW, Kuhns LR. Postextubation nasal continuous positive airway pressure. Am J Dis Child 1982;136:359-361.
Higgins 1991 {published data only}
Higgins RD, Richter SE, Davis JM. Nasal continuous positive pressure facilitates extubation of very low birthweight neonates. Pediatrics 1991;88:999-1003.
So B-H, Tamura M, Mishina J, Watanabe T, Kamoshita S. Application of nasal continuous positive airway pressure to early extubation in very low birthweight infants. Arch Dis Child 1995;72:F191-193.
Tapia 1995 {published data only}
Tapia J, Bancalari A, Gonzalez A, Mercado M. Does continuous positive airways pressure (CPAP) during weaning from intermittent mandatory ventilation in very low birthweight infants have risks or benefits? A controlled trial. Ped Pulmonol 1995;19:269-279.
Andreasson B, Lindroth M, Svenningsen NW, Jonson B. Effects on respiration of CPAP immediately after extubation in the very preterm infant. Pediatr Pulmonol 1988;4:213-218.
Robertson {published data only}
Robertson NJ and Hamilton PA. Randomised trial of elective continuous positive airway pressure (CPAP) compared with rescue CPAP after extubation. Arch Dis Child Fetal Neonatal Ed 1998;79:F61-63.
Robertson NJ, Hamilton PA. Randomized trial of elective continuous positive airway pressure (CPAP) compared with rescue CPAP after extubation. Arch Dis Child Fetal Neonatal Ed 1998;79:F58-F60.
Robertson 1998 {published data only}
* indicates the primary reference for the study
National Institutes of Health. National Institutes of Health Consensus Development Conference on Infantile Apnea and Home Monitoring. Pediatrics 1987:292 -299.
Sinclair JC, Bracken MB, editors. Effective care of the newborn infant. Oxford: Oxford University Press, 1992.
Davis PG, Henderson-Smart DJ. Nasal continuous positive airways pressure immediately after extubation for preventing morbidity in preterm infants (Cochrane Review). In: The Cochrane Library, Issue 1, 1997. Oxford: Update Software.
Davis PG, Henderson-Smart DJ. Nasal continuous positive airways pressure immediately after extubation for preventing morbidity in preterm infants (Cochrane Review). In: The Cochrane Library, Issue 3, 2000. Oxford: Update Software.
| Comparison or outcome | Studies | Participants | Statistical method | Effect size |
|---|---|---|---|---|
| 01 Nasal CPAP vs Headbox | ||||
| 01 Failure | 8 | 629 | RR (fixed), 95% CI | 0.62 [0.49, 0.77] |
| 02 Endotracheal reintubation | 8 | 629 | RR (fixed), 95% CI | 0.93 [0.72, 1.19] |
| 03 Oxygen use at 28 days | 5 | 433 | RR (fixed), 95% CI | 1.00 [0.81, 1.24] |
| 02 Nasal CPAP vs Headbox (methylxanthines given) | ||||
| 01 Failure | 7 | 611 | RR (fixed), 95% CI | 0.64 [0.51, 0.81] |
| 02 Oxygen use at 28 days | 4 | 283 | RR (fixed), 95% CI | 0.86 [0.67, 1.10] |
| 03 Nasal CPAP vs Headbox (extubation < or = 14 days of age) | ||||
| 01 Failure | 6 | 446 | RR (fixed), 95% CI | 0.58 [0.44, 0.75] |
| 02 Oxygen use at 28 days | 4 | 367 | RR (fixed), 95% CI | 1.17 [0.91, 1.50] |
| 04 Nasal CPAP vs Headbox (extubation > 14 days of age) | ||||
| 01 Failure | 2 | 66 | RR (fixed), 95% CI | 1.12 [0.58, 2.15] |
| 05 Nasal CPAP vs Headbox (CPAP < 5cm water) | ||||
| 01 Failure | 2 | 179 | RR (fixed), 95% CI | 1.09 [0.69, 1.73] |
| 06 Nasal CPAP vs Headbox (CPAP > or = 5 cm water) | ||||
| 01 Failure | 5 | 300 | RR (fixed), 95% CI | 0.49 [0.37, 0.66] |
| 02 Oxygen use at 28 days | 4 | 283 | RR (fixed), 95% CI | 0.86 [0.67, 1.10] |
| 07 Nasal CPAP vs Headbox (trials allowing rescue CPAP) | ||||
| 01 Endotracheal reintubation | 4 | 318 | RR (fixed), 95% CI | 1.04 [0.70, 1.53] |
| 08 Nasal CPAP vs Headbox (excluding quasi-randomised trial of Tapia) | ||||
| 01 Failure | 7 | 570 | RR (fixed), 95% CI | 0.57 [0.46, 0.72] |
| 02 Oxygen use at 28 days | 5 | 435 | RR (fixed), 95% CI | 1.00 [0.81, 1.24] |
| 09 Nasal CPAP vs Headbox (weight < 2000g) | ||||
| 01 Failure | 6 | 461 | RR (fixed), 95% CI | 0.65 [0.51, 0.84] |
| 02 Oxygen use at 28 days | 4 | 283 | RR (fixed), 95% CI | 0.86 [0.67, 1.10] |
Prof David J Henderson-Smart
Director
NSW Centre for Perinatal Health Services Research
Queen Elizabeth II Institute for Mothers and Infants
Building DO2
University of Sydney
Sydney
NSW AUSTRALIA
2006
Telephone 1: +61 2 93517318
Telephone 2: +61 2 93517728
Facsimile: +61 2 93517742
E-mail: dhs@perinatal.usyd.edu.au