Tracheal gas insufflation for the prevention of morbidity and mortality in mechanically ventilated newborn infants

Davies MW, Woodgate PG

Background - Methods - Results - Characteristics of Included Studies - References - Data Tables & Graphs

Cover sheet

Title

Tracheal gas insufflation for the prevention of morbidity and mortality in mechanically ventilated newborn infants

Reviewers

Davies MW, Woodgate PG

Dates

Date edited: 26/02/2002
Date of last substantive update: 28/12/2001
Date of last minor update: 14/12/2001
Date next stage expected / /
Protocol first published: Issue 1, 2001
Review first published: Issue 2, 2002

Contact reviewer

Dr Mark William Davies, MB BS FRACP DCH
Staff Neonatologist
Grantley Stable Neonatal Unit
Royal Women's Hospital
Butterfield St
Herston
Brisbane
Queensland AUSTRALIA
4029
Telephone 1: +61 7 3636 2245
Telephone 2: 0408 159 138
Facsimile: +61 7 3636 5259
E-mail: mwdavies@ozemail.com.au

Contribution of reviewers

MWD - instigated the review, wrote the protocol, searched for studies, extracted data from included studies, entered data into RevMan, wrote the review.

PGW - wrote the protocol, searched for studies, extracted data from included studies, wrote the review.

Intramural sources of support

Grantley Stable Neonatal Unit, Royal Women's Hospital, Brisbane, AUSTRALIA
Perinatal Research Centre, Royal Women's Hospital, Brisbane, AUSTRALIA
Dept of Paediatrics and Child Health, University of Queensland, Brisbane, AUSTRALIA
Centre for Clinical Studies, Mater Hospital, Brisbane, AUSTRALIA
Cochrane Perinatal Team, Brisbane, AUSTRALIA
Royal Children's Hospital Foundation, Brisbane, AUSTRALIA

Extramural sources of support

None

What's new

Dates

Date review re-formatted: / /
Date new studies sought but none found: / /
Date new studies found but not yet included/excluded: / /
Date new studies found and included/excluded: / /
Date reviewers' conclusions section amended: / /
Date comment/criticism added: / /
Date response to comment/criticisms added: / /

Text of review

Synopsis

Synopsis pending.

Abstract

Background

Tracheal gas insufflation (TGI) is a technique where a continuous flow of gas is instilled into the lower trachea during conventional mechanical ventilation. TGI can improve carbon dioxide removal with lower ventilation pressures and smaller tidal volumes, potentially decreasing secondary lung injury and chronic lung disease (CLD).

Objectives

To assess whether, in mechanically ventilated neonates, the use of tracheal gas insufflation reduces mortality, CLD and other adverse clinical outcomes without significant side effects.

Search strategy

Searches were made of MEDLINE 1966 to December 2001, CINAHL 1982 to December 2001, the Cochrane Controlled Trials Register (Cochrane Library, Issue 4, 2001) and conference and symposia proceedings.

Selection criteria

Randomised controlled trials (RCT) that include newborn infants who are mechanically ventilated, and compare TGI during conventional mechanical ventilation (CMV) with CMV alone. Primary outcomes - mortality, CLD and neurodevelopmental outcome; secondary outcomes - air leak, intraventricular haemorrhage, periventricular leukomalacia, duration of mechanical ventilation, duration of respiratory support, duration of oxygen therapy, duration of hospital stay, retinopathy of prematurity, immediate adverse effects.

Data collection & analysis

Each reviewer assessed eligibility, trial quality and extracted data separately. Study authors were contacted for additional information if necessary.

Main results

Only one small study was found to be eligible. This study found no evidence of effect on mortality, CLD or age at first extubation. The total duration of ventilation was 9.3 days shorter in the TGI group (95% CI from 15.7 to 2.9 days shorter). The age at complete weaning from ventilation was 26 days shorter in the TGI group (95% CI from 46 to 6 days shorter). There was no evidence of effect on the total duration of respiratory support, oxygen therapy or hospital stay.

Reviewers' conclusions

There is evidence from a single RCT that TGI may reduce the duration of mechanical ventilation in preterm infants - although the data from this small study do not give sufficient evidence to support the introduction of TGI into clinical practice. The technical requirements for performing TGI (as performed in the single included study) are great. There is no statistically significant reduction in the total duration of respiratory support or hospital stay. TGI cannot be recommended for general use at this time.

Background

Tracheal gas insufflation (TGI), or dead space washout, is a technique where, in addition to conventional mechanical ventilation (CMV), a continuous flow of gas (air/oxygen) is instilled into the lower trachea. This additional flow, or insufflation, of gas can occur throughout the ventilator cycle or only during expiration, with or without the addition of active aspiration of gas from the trachea during expiration (De Robertis 1999). TGI can be achieved with the insufflation of gas via extra lumens in the endotracheal tube (ETT) or a catheter placed down the primary lumen of an ETT.

TGI has been demonstrated to increase carbon dioxide (CO2) removal and lower arterial carbon dioxide (PaCO2) in animal studies (Bernath 1997, De Robertis 1999), and in uncontrolled studies in human adults (Richecoeur 1999) and preterm neonates (Danan 1996, Dassieu 1998). It allows the use of lower peak inspiratory pressures and smaller tidal volumes to achieve similar PaCO2s, potentially decreasing volu- and baro-trauma, and secondary lung injury (Bernath 1997, De Robertis 1999, Richecoeur 1999). In neonates this could potentially lead to decreased chronic lung disease (CLD).

Preterm infants have a very high dead space to tidal volume ratio which can limit the amount of CO2 clearance possible with conventional mechanical ventilation (Dassieu 1998). This high dead space to tidal volume ratio leads to a relatively greater amount of CO2 trapped in the dead space in the smallest of infants (e.g. extremely low birth weight infants who are most at risk of developing chronic lung disease). Increased CO2 clearance may only be possible by increasing tidal volume, with possible adverse effects such as volu-trauma, acute lung injury and chronic lung disease. In this situation washing CO2 out of the trachea may be of particular benefit.

Two reports in the literature describe the use of TGI in preterm neonates (Danan 1996, Dassieu 1998). Both studies were uncontrolled and investigated short-term respiratory outcomes only. They both confirmed that TGI increases CO2 removal and allows the use of smaller tidal volumes and lower peak inspiratory pressures, but neither assessed the effect on the incidence of CLD.

The response to, or benefit from, TGI may not be the same for all neonates but may vary according to birth weight and/or gestational age. Furthermore, there may be variation in effect dependent upon the strategy used: 1) TGI administered throughout the respiratory cycle or only during expiration, 2) tracheal gas aspiration used simultaneously with TGI or TGI used alone, 3) TGI used as an elective procedure or as rescue treatment.

Potential side effects relate to the smaller internal diameter of the endo-tracheal tubes required for the technique or the use of intra-luminal ETT catheters, possibly increasing the incidence of ETT tube obstruction; increased ETT handling, possibly increasing the incidence of accidental extubation; and uncontrolled manipulations of minute ventilation, possibly resulting in fluctuations in PaCO2 and altered haemodynamics, cerebral blood flow or oxygenation with consequent adverse neurological sequelae.

Objectives

The primary objective was to assess whether, in mechanically ventilated neonates, the use of tracheal gas insufflation reduces mortality, duration of respiratory support, chronic lung disease, adverse neurodevelopmental outcome, and other adverse clinical outcomes.

A secondary objective was to assess whether the use of tracheal gas insufflation is associated with significant side effects.

Sub-group analyses were planned to determine whether the results differ by:

Population:
i. gestational age
ii. birth weight

Intervention:
i. tracheal gas insufflation throughout the respiratory cycle or only during expiration
ii. tracheal gas aspiration simultaneously with tracheal gas insufflation or tracheal gas insufflation alone
iii. whether used as elective or rescue treatment

Criteria for considering studies for this review

Types of studies

Randomised controlled trials. Crossover studies will be excluded.

Types of participants

Newborn infants who are mechanically ventilated.

Types of interventions

Tracheal gas insufflation during conventional mechanical ventilation compared with conventional mechanical ventilation alone.

Types of outcome measures

One or more of the following outcomes must be reported:

- mortality (neonatal and/or before discharge)
- chronic lung disease (requirement for respiratory support and/or oxygen at 28 days of age and at 36 weeks corrected gestational age with or without changes on the chest x-ray)
- neurodevelopmental outcome (cerebral palsy, sensorineural hearing loss, visual impairment and/or developmental delay) at 1, 2, 3, 5 or 7 years.
- air leak
- intraventricular haemorrhage (any, grade 3-4)
- periventricular leukomalacia
- duration of mechanical ventilation (IPPV)
- duration of respiratory support (IPPV or CPAP)
- duration of oxygen therapy
- duration of hospital stay
- retinopathy of prematurity (any, stage 3 or greater)

Immediate adverse effects such as:
- endo-tracheal tube obstruction
- re-intubation whilst still ventilated
- altered haemodynamics, cerebral blood flow or oxygenation, including hypotension, bradycardia, tachycardia, increased cerebral blood flow, decreased cerebral blood flow, periods of decreased arterial oxygen saturation (duration and severity)
- severe hypocapnia (PaCO2 <25mmHg)
- severe hypercapnia (PaCO2 >70mmHg)

Search strategy for identification of studies

Using textword search terms 'tracheal gas insufflation', 'dead space washout', 'expiratory washout' and the MeSH search term 'exp infant, newborn' searches were made of MEDLINE 1966 to December 2001, CINAHL 1982 to December 2001, the Cochrane Controlled Trials Register (Issue 4, 2001), expert informants and journal hand searching mainly in the English language, previous reviews including cross references, abstracts, and conference and symposia proceedings published in Pediatric Research from 1990 to 1994. There were no limits on the search with respect to language or whether published data or not.

Methods of the review

Criteria and methods used to assess the methodological quality of the trials: standard method of the Cochrane Collaboration and its Neonatal Review Group were used. The two reviewers worked independently to search for and assess trials for inclusion and methodological quality. Studies were assessed using the following key criteria: blinding of randomisation, blinding of intervention, completeness of follow up and blinding of outcome measurement. Data were extracted independently by the reviewers. Differences were resolved by discussion and consensus of the reviewers. Study investigators were contacted for additional information or data.

For individual trials, where possible, mean differences (and 95% confidence intervals) were reported for continuous variables such as duration of oxygen therapy. For categorical outcomes such as mortality, the relative risk and risk difference (and 95% confidence intervals) were reported.

Description of studies

Only one study was identified and found eligible for inclusion in this review. Dassieu et al (Dassieu 2000) performed a randomised, controlled trial comparing conventional pressure-controlled ventilation and continuous tracheal gas insufflation with conventional pressure-controlled ventilation alone in preterm infants. The trial included infants <30 weeks gestational age who were ventilated for hyaline membrane disease and had received surfactant. Included infants had to be randomised by six hours of age. Forty-one infants were randomised and after seven post-randomisation exclusions 34 infants were studied - 15 TGI and 19 control. The mean (SD) gestational age was 27.8 (1.3) weeks and the mean (SD) weight was 965 (202) grams. Babies were randomised at a mean (SD) age of 226 (81) minutes and at a mean (SD) interval of 50 (42) minutes after surfactant administration. The primary outcomes of the study were the ventilatory pressures required and oxygenation status during ventilation. Secondary outcomes included mortality, chronic lung disease and duration of mechanical ventilation.

In the experimental group TGI was achieved via a specially constructed ETT with eight secondary lumens in the wall of the ETT. The secondary lumens allow continuous gas flow of 0.5 L/min to produce TGI (6 lumens) and pressure monitoring (1 lumen) - one lumen is left free to use for the introduction of surfactant. A pump is required to draw warmed humidified gases from the inspiratory line for TGI. A separate computerised monitoring system is also needed to monitor pressure at the distal ETT to enable emergency cut-off of the TGI flow with any abnormal rise in intratracheal pressure (to prevent dangerous lung hyperinflation). TGI was used in this group of infants until extubation. TGI was discontinued during any periods of high-frequency ventilation. In both groups conventional mechanical ventilation (with or without TGI) was continued if the oxygenation index remained within the range of two to 12. An oxygenation index of >12 resulted in a switch to high-frequency ventilation. Infants were weaned from ventilation when the ventilatory rate was below 30 breaths per minute and the oxygenation index was less than two.

In response to our request, the study investigators provided unpublished data for some outcomes not included in the primary report (Dassieu 2000) including the total duration of respiratory support, oxygen therapy and hospital stay. They also provided data on neurodevelopmental outcomes (cerebral palsy, sensorineural hearing loss, visual impairment and/or developmental delay); however, the details of exact definitions and timing of assessments are not clear and these outcomes have not been included in this review. Other data were published as medians and the study investigators have provided the data as means and standard deviations.

Methodological quality of included studies

In Dassieu et al's study (Dassieu 2000):
- treatment allocation was randomised (exact method not stated);
- treatment allocation was concealed using 'sealed envelopes';
- treatment was not blinded - however in both groups ventilation settings were adjusted according to a predefined protocol to target set oxygen saturation and PCO2; and there were predefined criteria for weaning ventilation and extubation;
- follow-up was not complete - 41 patients were randomised and there were seven (17%) post-randomisation exclusions. Three of these exclusions were because of technical difficulties with the equipment which prevented commencement of TGI; therefore, analysis is not an 'intention to treat' analysis. Follow-up rates for the remaining 34 infants were generally better than 90% (except for ROP and occurrence of severe hypo- or hypercapnia);
- most of the published outcomes were not assessed by blinded evaluators.

Results

The results of this review are based on one eligible study with small numbers.

Mortality (neonatal and/or before discharge):
There was no statistically significant difference between groups.

Chronic lung disease (requirement for respiratory support and/or oxygen at 28 days of age and at 36 weeks corrected gestational age:
There was no statistically significant difference between groups.

Neurodevelopmental outcome (cerebral palsy, sensorineural hearing loss, visual impairment and/or developmental delay):
No data available.

Air leak:
There was no statistically significant difference between groups.

IVH:
There was no statistically significant difference between groups.

PVL:
There was no statistically significant difference between groups.

Duration of mechanical ventilation:
The duration of mechanical ventilation was measured and reported as three different variables:
1. the age (in days) at first extubation - this was not significantly different between groups.
2. total duration of ventilation (defined as the sum of the periods of ventilation until the infant remains extubated for seven consecutive days post-extubation) was 9.3 days shorter in the TGI group (95% CI from 15.7 to 2.9 days shorter).
3. age at complete weaning from ventilation was 26 days shorter in the TGI group (95% CI from 46 to 6 days shorter).

Duration of respiratory support (IPPV or CPAP):
There was no statistically significant difference between groups. However, this was six days longer in the treatment group (mean difference 6 days, 95% CI -6.4 to 18.4) and the 95% confidence interval is wide.

Duration of oxygen therapy:
There was no statistically significant difference between groups.

Duration of hospital stay:
There was no statistically significant difference between groups. However this was 10 days shorter in the treatment group (mean difference -10 days, 95% CI -30 to 10) and the 95% confidence interval is wide.

ROP:
There was no statistically significant difference between groups.

Endo-tracheal tube obstruction:
There was no statistically significant difference between groups.

Re-intubation whilst still ventilated:
There was no statistically significant difference between groups.

Altered haemodynamics, cerebral blood flow or oxygenation, including hypotension, bradycardia, tachycardia, increased cerebral blood flow, decreased cerebral blood flow, periods of decreased arterial oxygen saturation (duration and severity):
No data available.

Severe hypocapnia (PaCO2 <25mmHg):
There was no statistically significant difference between groups.

Severe hypercapnia (PaCO2 >70mmHg):
There was no statistically significant difference between groups.

Discussion

This review is limited in its scope given that only one study was identified which met the inclusion criteria. The single included study (Dassieu 2000) was a well conducted randomised controlled trial with only one moderately worrisome methodological shortcoming - the post-randomisation exclusions (although the proportion is <20% of those randomised). The study was a small study with only 34 subjects randomised and followed-up. Therefore, whilst most of the comparisons favour the treatment group, only two outcomes, both related to duration of mechanical ventilation, reached statistical significance. The potential for type 2 error is considerable and the 95% confidence intervals are wide.

Nevertheless this study provides some evidence that TGI may be useful in reducing the duration of mechanical ventilation for infants <30 weeks gestation with hyaline membrane disease. It is important to note, however, that the total duration of respiratory support (i.e. conventional ventilation, high frequency ventilation and continuous positive airway pressure) was six days longer in the treatment group although this difference was not statistically significant. The duration of hospital stay was reduced in the treatment group by 10 days but this also did not reach statistical significance. Much larger studies would be required to determine whether the advantage of a reduction in the duration of mechanical ventilation is not lost by an increased total duration of respiratory support, or augmented by a reduction in hospital stay. Benefits also need to be weighed against an intervention that is technically demanding, requires additional specialised equipment and may add considerably to the cost of ventilation.

Reviewers' conclusions

Implications for practice

There is evidence from a single small RCT that TGI may reduce the duration of mechanical ventilation in preterm infants - although the data from this single small study do not give sufficient evidence to support the introduction of TGI into clinical practice. The technical requirements for performing TGI (as performed in the single included study) are great, and there is no statistically significant reduction in the total duration of respiratory support or hospital stay. TGI cannot be recommended for general use at this time.

Implications for research

Appropriately sized RCTs are required to determine whether the advantage of TGI in reducing the duration of mechanical ventilation is not lost by an increased total duration of respiratory support, or augmented by a reduction in hospital stay. Future studies should also investigate the effects on long term neuro-developmental outcomes (i.e. cerebral palsy, sensorineural hearing loss, visual impairment and/or developmental delay) to determine that any short term gains are not lost with an increased incidence of disability in survivors.

Acknowledgements

We would like to thank Prof Claude Danan, from the Service de Reanimation Neonatale, Hopital Intercommunal de Creteil, for providing extra data from Dassieu et al's study (Dassieu 2000).

Potential conflict of interest

Nil

Characteristics of included studies

Study Methods Participants Interventions Outcomes Notes Allocation concealment
Dassieu 2000 RCT
(sealed envelopes).

Randomised within 6 hours of birth.

Randomised - YES (exact method of sequence generation not stated).
Blindness of randomisation (allocation) - YES (sealed envelopes, although not stated whether envelopes were opaque).
Blindness of intervention - NO.
Completeness of follow up - NO.
Blinding of outcome measurement - NOT ALL.

 

 Newborn infants <30 weeks gestational age.
Ventilated for hyaline membrane disease and had received surfactant.
Exclusions
1. HIE
2. refractory hypotension
3. lethal malformations
4. congenital malformations with pulmonary consequences.		 Experimental group - conventional pressure-controlled ventilation with continuous tracheal gas insufflation. (Number randomised = 19) TGI was used in this group of infants until extubation. TGI was discontinued during any periods of high-frequency ventilation. 

Control group - conventional pressure-controlled ventilation alone. (Number randomised = 22)

Treatment was not blinded - however in both groups ventilation settings were adjusted according to a predefined protocol to target set oxygen saturation and PCO2; and there were predefined criteria for weaning ventilation and extubation.

In both groups conventional mechanical ventilation (with or without TGI) was continued if the oxygenation index remained within the range of 2 to 12. An oxygenation index of >12 resulted in a switch to high-frequency ventilation. Infants were weaned from ventilation when the ventilatory rate was below 30 breaths per minute and the oxygenation index was <2.

 Published - primary
-ventilatory pressures
-oxygenation status

Published - secondary
-mortality at 28 days
-mortality at discharge
-oxygen at 28 days
-oxygen at 36 weeks
-pneumothorax
-ETT obstruction
-IVH any
-IVH grade 3+
-PVL
-need for HFOV (OI >12)
-death or CLD at 28 days
-death or CLD at 36 weeks
-days to first extubation
-total duration IPPV
 

Unpublished (as at September 2001)
-pneumothorax
-emphysema
-need for re-intubation whilst ventilated (plug)
-need for re-intubation whilst ventilated (other)
-severe hypocapnia (<25) even once
-severe hypercapnia (>70) even once
-age of complete weaning from ventilatory support IPPV or HFO
-age of complete weaning from nasal ventilation
-age of complete weaning from oxygen
-retinopathy (any)
-retinopathy (stage 3 or greater)
-duration of hospital stay (NICU + general)
-cerebral palsy
-sensorineural hearing loss
-visual impairment
-developmental delay

 There were 7 post-randomisation exclusions, 4 from the experimental group and 3 from the control group. These exclusions were because of:
technical difficulties with equipment so that TGI could not be commenced (3);
post-randomisation withdrawal of consent (2);
no consent obtained (1);
oxygenation index >12 at the time of randomisation (1).
In the case of the last 3 reasons for exclusion, it is not clear from which group the exclusions occurred.		 A

References to studies

References to included studies

Dassieu 2000 {published data only}

Dassieu G, Brochard L, Benani M, Avenel S, Danan C. Continuous tracheal gas insufflation in preterm infants with hyaline membrane disease. Am J Respir Crit Care Med 2000;162:826-831.

* indicates the primary reference for the study

Other references

Additional references

Bernath 1997

Bernath MA, Henning R. Tracheal gas insufflation reduces requirements for mechanical ventilation in a rabbit model of respiratory distress syndrome. Anaesth Intens Care 1997;25:15-22.

Danan 1996

Danan C, Dassieu G, Janaud J-C, Brochard L. Efficacy of dead-space washout in mechanically ventilated premature newborns. Am J Respir Crit Care Med 1996;153:1571-1576.

Dassieu 1998

Dassieu G, Brochard L, Agudze E, Patkai J, Janaud J-C, Danan C. Continuous tracheal gas insufflation enables a volume reduction strategy in hyaline membrane disease: technical aspects and clinical results. Intensive Care Med 1998;24:1076-1082.

De Robertis 1999

De Robertis E, Sigurdsson SE, Drefeldt B, Jonson B. Aspiration of airway dead space. A new method to enhance CO2 elimination. Am J Respir Crit Care Med 1999;159:728-732.

Richecoeur 1999

Richecoeur J, Lu Q, Vieira SRR, Puybasset L, Kalfon P, Coriat P, Rouby J-J. Expiratory washout versus optimization of mechanical ventilation during permissive hypercapnia in patients with severe acute respiratory distress syndrome. Am J Respir Crit Care Med 1999;160:77-85.

Comparisons and data

01 Continuous TGI during conventional mechanical ventilation (CMV) vs CMV alone
01.01 Mortality before day 28
01.02 Mortality before discharge
01.03 In oxygen at day 28 in survivors to 28 days
01.04 In oxygen at 36 weeks corrected GA in survivors to 36 weeks corrected GA
01.05 Death or CLD at 28 days
01.06 Death or CLD at 36 weeks corrected GA
01.07 Pneumothorax
01.08 Endotracheal tube obstruction
01.09 Intraventricular haemorrhage (any) in survivors to 28 days
01.10 Intraventricular haemorrhage (grade 3+) in survivors to 28 days
01.11 Periventricular leukomalacia in survivors to 28 days
01.12 Pulmonary interstitial emphysema among babies examined
01.13 Retinopathy of prematurity (any) among babies examined
01.14 Retinopathy of prematurity (grade 3+) among babies examined
01.15 Need for re-intubation because of ETT obstruction in babies where data are available
01.16 Need for re-intubation for any reason other than ETT obstruction in babies where data are available
01.17 One or more episodes of severe hypocapnia (PaCO2 <25 mmHg) in babies where data are available
01.18 One or more episodes of severe hypercapnia (PaCO2 >70 mmHg) in babies where data are available
01.19 Time to first extubation (days) in survivors to 28 days
01.20 Total duration of ventilation until remains extubated for 7 consecutive days (days) in survivors to 28 days
01.21 Age at complete weaning from ventilation - IPPV or HFO (days) in survivors to discharge
01.22 Age at complete weaning from respiratory support - ventilation or CPAP (days) in survivors to discharge
01.23 Age at complete weaning from oxygen (days) in survivors to discharge
01.24 Duration of hospital stay (days) in survivors to discharge
 

Notes

Published notes

Amended sections

None selected

Contact details for co-reviewers

Dr Paul G Woodgate, MB BS FRACP MMedSc
Consultant Neonatologist
Department of Neonatology
Mater Mother's Hospital
Raymond Terrace
South Brisbane
Brisbane
Queensland AUSTRALIA
4101
Telephone 1: +61 7 3840 1911
Facsimile: +61 7 3840 1949
E-mail: PWOODGAT@mater.org.au
Secondary contact person's name: Mark W Davies