Sven M Schulzke1, Jane Pillow2, Ben Ewald3, Sanjay K Patole2
Background - Methods - Results - Characteristics of Included Studies - References - Data Tables and Graphs
1Department of Neonatology, University of Basel Children's Hospital (UKBB), Basel, Switzerland
2School of Women's and Infant's Health, University of Western Australia, King Edward Memorial Hospital, Perth, Australia
3Centre for Clinical Epidemiology and Biostatistics, University of Newcastle, Newcastle, Australia
Citation example: Schulzke SM, Pillow J, Ewald B, Patole SK. Flow-cycled versus time-cycled synchronized ventilation for neonates. Cochrane Database of Systematic Reviews 2010, Issue 7. Art. No.: CD008246. DOI: 10.1002/14651858.CD008246.pub2.
Department of Neonatology
University of Basel Children's Hospital (UKBB)
Spitalstrasse 21
4031 Basel
Switzerland
E-mail: sven.schulzke@unibas.ch
| Assessed as Up-to-date: | 20 April 2010 |
|---|---|
| Date of Search: | 20 October 2009 |
| Next Stage Expected: | 20 April 2012 |
| Protocol First Published: | Issue 1, 2010 |
| Review First Published: | Issue 7, 2010 |
| Last Citation Issue: | Issue 7, 2010 |
| Date / Event | Description |
|---|
| Date / Event | Description |
|---|
Synchronized ventilation of neonates is standard care in industrialized countries. Both flow-cycled and time-cycled modes of synchronized ventilation are in widespread use for assisted ventilation of neonates.
To determine the effect of flow-cycled versus time-cycled synchronized ventilation on the risk of bronchopulmonary dysplasia (BPD) at 36 weeks postmenstrual age (PMA) in neonates requiring assisted ventilation.
We used the standard methods of the Cochrane Neonatal Review Group to search the Cochrane Central Register of Controlled Trials (CENTRAL) in The Cochrane Library Issue 4, 2009, PubMed (January 1966 to October 2009), EMBASE (January 1974 to October 2009) and CINAHL (January 1982 to October 2009). We checked references and cross-references from identified studies. Abstracts from the proceedings of the Pediatric Academic Societies Meetings (from January 1990 to October 2009) were handsearched. We placed no restrictions on language.
Randomized or quasi-randomized clinical trials comparing flow-cycled with time-cycled synchronized endotracheal ventilation in neonates, reporting on at least one outcome of interest were eligible for inclusion in the review.
One author (SMS) searched the literature as described above. Selection of studies and data extraction were done separately by two authors (SMS and SKP). Any disagreements were resolved by discussion involving all authors.
The use of breathing machines that match their output with the breathing effort of the baby (that is, they synchronize their output to a baby's spontaneous effort) is standard care in industrialized countries. Both the level of airflow generated by the baby during a breath ('flow-cycling') and the baby's intention to take a breath ('time-cycling') can be used for synchronization. Theoretically, flow-cycling allows for better synchronization because it synchronizes not only the beginning but also the duration of a breath. The main aim of this review was to find out whether flow-cycled compared to time-cycled synchronized ventilation offers advantages for babies requiring assisted ventilation.
Only two small short-term studies on a total of 19 preterm babies were identified in this review. These studies did not report on important outcomes such as the frequency of long lasting breathing disorders. Therefore, there is insufficient evidence to decide whether flow-cycled, compared to time-cycled synchronized ventilation offers advantages for babies who need assisted ventilation.
Neonates requiring intensive care are frequently placed on mechanical ventilation in order to assist their breathing. Although mechanical ventilation can be life-saving for neonates with respiratory failure, it may cause lung injury due to excess airway pressure (barotrauma), delivery of high tidal volumes (volutrauma) and repetitive closing/re-opening of lung units (atelectotrauma) (Attar 2002). Preterm neonates at an early stage of lung development are especially vulnerable to ventilator induced lung injury and may require respiratory support for considerable time (Donn 2006). Ventilator associated lung injury is one of several factors contributing to the burden of chronic lung disease of infancy, a condition also referred to as bronchopulmonary dysplasia (BPD) (Jobe 2001). Damage to the airways (Sherman 1986) and respiratory infection (Aly 2008) are additional risks associated with prolonged mechanical ventilation.
Synchronizing patient effort with ventilator assistance is one among several established lung protective strategies. Until the 1980s, synchronized modes of ventilation in neonates primarily included high frequency positive pressure ventilation (HFPPV) or time-cycled modes of synchronized ventilation such as synchronized intermittent mandatory (SIMV) or assist-control (ACV) ventilation. From the 1990s, flow-cycled modes of synchronized ventilation such as pressure support ventilation (PSV) have become widely available.
Studies comparing synchronized with non-synchronized ventilation in neonates focus on time-cycled synchronized versus non-synchronized ventilation. Randomized controlled trials (RCT) suggest that HFPPV compared to non-synchronized ventilation is advantageous in reducing the risk of air leaks in neonates (Heicher 1981; Pohlandt 1992) while SIMV versus non-synchronized ventilation in preterm neonates is associated with reduced duration of mechanical ventilation and lower risk of extubation failure (Chen 1997). Decreased work of breathing in ACV versus non-synchronized ventilation was demonstrated in a small randomized cross-over trial in preterm neonates (Jarreau 1996). The sole small RCT (N = 30) comparing flow-cycled synchronized ventilation with non-synchronized ventilation in neonates reported a reduced time to extubation in the former without evidence of adverse events (Donn 1994).
A systematic review of fourteen randomized and quasi-randomized trials comparing HFPPV, SIMV or ACV with non-synchronized ventilation in neonates reported a reduced risk of air leaks using HFPPV and shorter duration of mechanical ventilation using SIMV or ACV, but no significant effect on the incidence of BPD or death; no significant adverse events attributable to synchronized ventilation were noted (Greenough 2008). This systematic review also included comparison of different modes of time-cycled synchronized ventilation, but no comparison of flow-cycled with time-cycled synchronized ventilation. Meta-analysis did not demonstrate significant benefits or harms favoring any particular time-cycled mode of synchronized ventilation; however, the authors noted that there was a trend towards shorter duration of mechanical ventilation using ACV compared to SIMV. While inappropriate (e.g. delayed or insensitive) triggering or auto-triggering is a recognized complication of synchronized ventilation, especially when there is excessive condensation in the ventilatory circuit or a large leak around the endotracheal tube, none of these outcomes were reported in Greenough 2008.
Although the evidence to support the use of time-cycled synchronized over non-synchronized ventilation in neonates is limited, time-cycled synchronized ventilation of neonates is standard care in industrialized countries.
Time-cycled modes of ventilation such as SIMV or ACV enable the patient to trigger a ventilator-assisted breath which is delivered over a preset inflation time. Such modes synchronize the onset but not the duration and end of a ventilator-assisted breath with patient effort. Given that the spontaneous inspiratory time is highly variable within and between individual infants (Schmalisch 2005), preset inflation time of a time-cycled ventilator may not match the spontaneous inspiratory time of an infant. This may result in increased airway pressure if spontaneous expiration occurs prior to the end of the preset inflation time, potentially aggravating ventilator-associated lung injury and delaying weaning from the ventilator (Donn 2006). In order to better match spontaneous breathing with ventilatory output, some ventilators incorporate flow-cycled modes of synchronized ventilation. Depending on the ventilator in use, such modes may be called PSV (Bear Cub 750PSV, Viasys Healthcare, Conshohocken, PA, US; Draeger Babylog 8000, Draeger AG, Luebeck, Germany; Leonie plus, Heinen & Löwenstein, Bad Ems, Germany; Maquet Servo-i, Maquet Critical Care, Solna, Sweden; SLE 5000, SLE Limited, South Croydon, UK), assisted spontaneous breathing (ASB; Draeger Evita, Draeger AG, Luebeck, Germany), or flow synchronization (V.I.P. Bird, Bird Products Corporation, Palm Springs, CA). Using flow-cycled synchronized ventilation, an inflation is terminated depending on the peak inspiratory flow generated by the patient, i.e. the ventilator terminates an inflation when inspiratory flow drops below a predefined threshold proportion of the peak inspiratory flow. Given that inspiratory flow depends on a patient's spontaneous inspiratory effort, flow-cycling allows for variable inspiratory time and enhanced patient-control of breathing by synchronizing not only the onset, but also the duration and end of an inspiration. This may result in decreased airway pressure, reduced ventilator-associated lung injury, less asynchrony, and may facilitate weaning from the ventilator (Abd El-Moneim 2005). Furthermore, flow-cycling with variable inflation time may help to mimic the highly variable, non-linear dynamics of physiological lung inflation (Suki 1994) and may enable a ventilated neonate to establish a breathing pattern that includes sighs. The ability to sigh seems to be particularly important for maintaining oxygenation in preterm neonates who use frequent sighs to recruit and restore their lung volume (Poets 1997). On the other hand, delivery of inflations using flow-cycled synchronized ventilation is highly dependent on accurate flow measurement that might be disturbed in the presence of a large leak around the endotracheal tube or air flow obstruction. Under these circumstances, flow-cycled synchronized ventilation may be detrimental in terms of inadequate gas exchange and increased work of breathing. Factors including the proportion of peak inspiratory flow used to terminate flow-cycled inflation, presence of tidal volume targeting, magnitude of ventilator bias flow, and presence of a leak adaption algorithm may influence the accuracy and reliability of delivery of inflations, especially in the presence of leak or airflow obstruction. Subgroup analyses were planned to find out whether any of these factors affects outcomes of the review.
Key to abbreviations used in this review:
AaDO2: Alveolar-arterial oxygen gradient; ACV: Assist-control ventilation; ASB: Assisted spontaneous breathing; BPD: Bronchopulmonary dysplasia; CPAP: Continuous positive airway pressure; GA: Gestational age; HFPPV: High frequency positive pressure ventilation; HHHF: Heated, humidified high-flow nasal cannula therapy; IVH: Intraventricular hemorrhage; NEC: Necrotizing enterocolitis; OI: Oxygenation index; pCO2: Partial pressure of carbon dioxide; PVL: Periventricular leukomalacia; PMA: Postmenstrual age; PSV: Pressure support ventilation; ROP: Retinopathy of prematurity; SaO2: Oxygen saturation; SIMV: Synchronized intermittent mandatory ventilation; SIPPV: Synchronized intermittent positive pressure ventilation (equivalent to assist-control ventilation).
Flow-cycled and time-cycled modes of synchronized ventilation are in widespread use for mechanical ventilation of neonates. These modes of ventilation need to be formally assessed in order to provide caregivers with clinically relevant data on their efficacy and safety. The aim of this systematic review was to summarize the current evidence on benefits and harms of flow-cycled versus time-cycled synchronized ventilation in neonates.
The primary objective is to determine the effect of flow-cycled inspiration versus time-cycled inspiration on the prevalence of BPD at 36 weeks postmenstrual age (PMA) in neonates requiring assisted ventilation.
The secondary objectives include potential benefits of flow-cycled versus time-cycled synchronized ventilation in terms of:
Subgroup analyses were planned considering whether one of the following factors influence outcomes of the systematic review:
Degree of prematurity at birth (preterm neonates < 37 weeks gestational age (GA); very preterm neonates < 32 weeks GA); proportion of peak inspiratory flow used to terminate flow-cycled inflation (> 10%); availability of volume targeting; magnitude of the ventilator bias flow (< 8 L/min); and presence of a leak adaption algorithm in flow-cycled synchronized ventilation.
Randomized or quasi-randomized clinical trials comparing flow-cycled with time-cycled synchronized ventilation in neonates were eligible for inclusion in the review. Parallel-group, cross-over, and cluster randomized trials were eligible.
Neonates requiring assisted ventilation (i.e. onset of assisted ventilation < 28 days).
The following forms of ventilation were considered with or without volume targeting for both experimental and control groups:
Experimental group:
Control group:
Trials considered for inclusion in the review needed to assess at least one of the following clinical outcomes:
We searched the Cochrane Neonatal Review Group's specialized register, The Cochrane Central Register of Controlled Trials (CENTRAL) in The Cochrane Library Issue 4, 2009, PubMed (January 1966 to October 2009), EMBASE (January 1974 to October 2009) and CINAHL (January 1982 to October 2009). We checked references and cross-references from identified studies. Abstracts from the proceedings of the Pediatric Academic Societies Meetings (from January 1990 to October 2009) were handsearched. We placed no restrictions on language.
Appendix 1 lists the search terms used.
We used the standard methods of the Cochrane Neonatal Review Group and The Cochrane Collaboration.
One review author (SMS) independently searched the literature as described above. Only randomized and quasi-randomized controlled trials fulfilling the above inclusion criteria were considered for inclusion in the review. Studies published only in abstract form were not included unless the final results of the trial were reported and all necessary information could be ascertained from the abstract and/or the authors. Selection of studies was done separately by two review authors (SMS and SKP). Any disagreements were resolved by discussion involving all review authors.
Two review authors (SMS and SKP) independently extracted, assessed and coded all data for each study, using standardized data extraction forms. In order to reduce the risk of coding errors, data were entered into Review Manager software by one review author (SMS) using double data entry. Data were checked for accuracy by all review authors. We sought additional information from the authors of retrieved trials if further data or clarification of existing data was required.
Risk of bias was assessed according to selection bias (quality of randomization, allocation concealment/blinding of randomization), performance bias (blinding of intervention), attrition bias (completeness of follow-up) and detection bias (blinding of outcome measurement). Assessments were specified as "Yes", "No", or "Can't tell". Any disagreement was resolved by discussion involving all review authors. This information is included in the table Characteristics of included studies.
Statistical analyses were planned using the standard methods of the Cochrane Neonatal Review Group. Briefly, we planned to use relative risk (RR), risk difference (RD), number needed to treat (NNT) or number needed to harm (NNH) for categorical variables and weighted mean difference (WMD) for continuous variables. Any within-group standard error of the mean (SEM) reported in a trial was to be replaced by its corresponding standard deviation (SD) using the formula SD = SEM* square root (N). Ninety-five percent confidence intervals (95% CI) were to be reported for each statistic.
Cross-over trials were considered a priori for inclusion in the review. For pooling of data from cross-over trials we planned to seek statistical advice. Where pooling of data was not possible we decided to report these trials as given by the primary authors.
We planned to request additional information on study design and/or missing data when necessary.
The magnitude of heterogeneity was to be estimated by the I2 statistic. A P value of 0.1 (rather than 0.05) for the Chi2 statistic was considered as indication of significant heterogeneity in order to account for low statistical power of the Chi2 statistic when assessing heterogeneity in meta-analyses of trials with small sample size. Additionally, we planned to inspect each forest plot carefully for heterogeneity, as indicated by lack of overlapping confidence intervals of individual trials. Given that intensity of mechanical ventilation may influence any potentially observed differences between modes of ventilation, we were aiming at grouping studies depending on their target pCO2 range as well as their tidal and minute ventilation targets in order to assess whether these targets were a source of heterogeneity.
We examined the possibility of within-study selective outcome reporting for each study included in the review. In order to assess whether outcome reporting seemed to be sufficiently complete and transparent, we searched for trial protocols of included trials on PubMed, ClinicalTrials.gov, and the Clinical Trials Search Portal of the World Health Organisation.
Given the significant baseline risk of prolonged mechanical ventilation and/or BPD in preterm neonates, the following subgroup analyses were planned:
Depending on the type of ventilator and specific ventilator settings, the proportion of peak inspiratory flow used as a cut-off to terminate flow-cycled inflation may affect reliability of the trigger mechanism and inflation time. Subgroup analysis was planned to determine whether the proportion of peak inspiratory flow terminating flow-cycled inflation influenced outcomes (> 10% versus ≤ 10%).
The use of volume targeting as an add-on mode to both flow-cycled and time-cycled synchronized ventilation may affect outcomes. We considered whether volume targeting in both experimental and control groups affected the outcomes.
The bias flow of the ventilator circuit may influence inflation time. We considered whether low bias flow (< 8 L/min versus ≥ 8 L/min) affected the results of the systematic review.
Leak adaption algorithms may improve the reliability of delivering flow-cycled synchronized ventilation. Subgroup analysis of studies using a leak adaption algorithm was planned.
See: Included studies; Excluded studies below.
The search identified 768 abstracts. Screening these abstracts, we detected nine potentially eligible studies. Of these, seven were excluded. Details of the excluded studies along with the reasons for exclusion are listed in the table Characteristics of excluded studies. Two studies were included in this review (De Luca 2009; Kapasi 2001). No ongoing trials were identified.
Details of the two included studies (De Luca 2009; Kapasi 2001) are given in the table Characteristics of included studies. Both studies were short-term, randomized, individual cross-over trials in preterm neonates. Both trials reported on lung mechanics and short-term respiratory physiology data but not on major clinical outcomes. De Luca 2009 compared flow-cycled with time-cycled synchronized ventilation. Kapasi 2001 compared non-synchronized ventilation with two modes of time-cycled synchronized ventilation and one mode of flow-cycled synchronized ventilation.
Additional information on study design and data analysis was requested and obtained from Prof Kacmarek (Kapasi 2001) (randomization by random number table, random allocation not concealed, exclusion of two infants after randomization due to problems in data transfer between pneumotachograph and computer, care givers and assessors of outcomes not blinded) and Dr De Luca (De Luca 2009) (randomization using computer-based random table, random allocation concealed using sealed opaque envelopes, care givers and assessors of outcomes not blinded). This information is included in the table Characteristics of included studies.
De Luca 2009 (N = 10) was a small, randomized, single-center cross-over trial comparing flow-cycled (PSV) with time-cycled synchronized ventilation (ACV). The purpose of this cross-over trial was to evaluate short-term effects of flow-cycled versus time-cycled ventilation in sick preterm neonates. The trial enrolled 10 preterm neonates (interquartile GA range 26 to 30.5 weeks, interquartile birth weight range 735 to 1475 grams) with respiratory distress syndrome (RDS) on day one of life. One infant was excluded pre-randomization due to intraventricular hemorrhage and seizures. Patients were ventilated using Bear Cub 750PSV ventilators (Viasys Healthcare, Conshohocken, PA, USA). Patients were randomly assigned to one hour periods of flow-cycled and time-cycled synchronized ventilation using a random number table with codes retrieved from sealed opaque envelopes. Each ventilation modality was used for a period of 15 minutes before pulmonary function data was collected over the following 45 minutes. All patients crossed over to the alternative mode after one hour of ventilation. Airway pressure and airflow were measured using the ventilators' hot wire anemometer and manometer, both placed at the y-piece between ventilator circuit and endotracheal tube. Prior to each study window, central venous pressure was measured and capillary blood gas analysis was performed. Pulse oximetry, transcutaneous oxygen and carbon dioxide levels were monitored every 10 minutes. Outcomes reported in this trial included respiratory rate, expiratory tidal volume, minute ventilation, mean airway pressure, spontaneous inspiratory time, dynamic compliance, total resistance, rate x mean airway pressure product, rate volume ratio, pulse oximetry, heart rate, blood gases, transcutaneous oxygen and carbon dioxide levels, central venous pressure, oxygenation index, and work of breathing generated by the ventilator (but not work of breathing of patients).
Kapasi 2001 was a small (N = 9), randomized single-center cross-over trial to test the hypothesis that synchronized ventilation would result in better synchrony, less patient effort, and less work of breathing than non-synchronized ventilation (IMV). The authors further hypothesized that among synchronized modes, flow-cycled synchronized ventilation (PSV) would perform better than time-cycled synchronized ventilation (SIMV and ACV). The trial enrolled nine preterm neonates (GA range 28 to 36 weeks, birth weight range 925 to 2100 grams) with RDS between the second and eighth day of life. Two infants were excluded post-randomization due to computer problems preventing data analysis. Patients were ventilated using Bird VIP ventilators (Bird Product, Riverside, CA, USA). In each patient, the mode of ventilation was randomly changed between IMV, SIMV, ACV, and PSV based on a random number table. Each ventilation modality was used for a period of 20 minutes before pulmonary function data were collected. Airway pressure and airflow were measured with a pneumotachograph between ventilator circuit and endotracheal tube. Esophageal pressure was measured with an esophageal balloon. Pulse oximetry and transcutaneous carbon dioxide levels were monitored continuously. Data were collected over two to three minutes. Outcomes reported in this trial included trigger delay, airway and esophageal trigger pressure change, endotracheal tube leak volume, heart rate, respiratory rate, arterial oxygen saturation, transcutaneous carbon dioxide level, minute ventilation, inspiratory time, inspiratory work of breathing, and pressure-time products.
Subgroup and sensitivity analyses
The two trials that met inclusion criteria did not report outcomes suitable for pre-specified subgroup and sensitivity analyses.
See the table Characteristics of excluded studies.
A search of PubMed, ClinicalTrials.gov, and the Clinical Trials Search Portal of the World Health Organisation revealed no protocols or ongoing trials of flow-cycled versus time-cycled synchronized ventilation in neonates.
Both included trials (De Luca 2009; Kapasi 2001) were small, randomized, short-term, individual cross-over trials reporting on physiological but not clinical outcomes. In both trials, total duration of follow-up was approximately two hours. Further assessment of risk of bias in included studies is given in the table Characteristics of included studies.
Treatment allocation was concealed in De Luca 2009 but not in Kapasi 2001. Given that Kapasi 2001 was a short-term, individual cross-over trial, the risk of selection bias, though minimal, cannot be excluded.
Both included studies reported on short-term respiratory physiological outcomes and lung mechanics but not on major clinical outcomes.
Both De Luca 2009 and Kapasi 2001 tested this comparison.
Neither of the included studies assessed treatment effects on any of the following secondary outcomes of this review: Prevalence of BPD (mild, moderate, severe BPD or supplemental oxygen at 28 days of life), duration of endotracheal ventilation at 40 weeks PMA, duration of endotracheal ventilation to hospital discharge, incidence of air leaks, duration of respiratory support, duration of supplemental oxygen at 40 weeks PMA, supplemental home oxygen requirement, incidence of extubation failure, ventilator-associated pneumonia, mortality at 36 weeks PMA, mortality to hospital discharge, duration of hospital stay, neurodevelopmental impairment.
Neither of the included studies reported on adverse events.
Complications of preterm birth including intraventricular hemorrhage, periventricular leukomalacia, necrotizing enterocolitis, retinopathy of prematurity, and sepsis were not reported in either of the studies.
Only Kapasi 2001 tested this comparison.
Kapasi 2001 did not report on the primary outcome of BPD at 36 weeks PMA.
Kapasi 2001 found no difference in work of breathing between PSV and SIMV (mean (SD) inspiratory work of breathing 2.02 (2.08) mJ per breath versus 3.91 (2.94) mJ per mechanical SIMV breath, P = 0.19). The authors reported lower inspiratory effort (pressure-time product per minute) in PSV compared to SIMV (no numeric values given, outcome reported in figure).
Kapasi 2001 found no difference in respiratory rate between PSV and SIMV groups (mean (SD) 67.3 (9.2) *min-1 versus 58.3 (13.9) *min-1, P = 0.18).
Kapasi 2001 found no difference in oxygen saturation and transcutaneous CO2 levels between PSV and SIMV groups (no group-specific numeric values given).
Kapasi 2001 did not assess treatment effects on any of the following secondary outcomes of this review: Prevalence of BPD (mild, moderate, severe BPD or supplemental oxygen at 28 days of life), duration of endotracheal ventilation at 40 weeks PMA, duration of endotracheal ventilation to hospital discharge, incidence of air leaks, duration of respiratory support, duration of supplemental oxygen at 40 weeks PMA, supplemental home oxygen requirement, incidence of extubation failure, ventilator-associated pneumonia, mortality at 36 weeks PMA, mortality to hospital discharge, duration of hospital stay, neurodevelopmental impairment.
Adverse events and complications of preterm birth including intraventricular hemorrhage, periventricular leukomalacia, necrotizing enterocolitis, retinopathy of prematurity, and sepsis were not reported in Kapasi 2001.
Only two small, randomized, short-term cross-over trials were eligible for inclusion in this review (De Luca 2009; Kapasi 2001). Both trials only reported lung mechanics and short-term respiratory physiology outcomes. Clinical morbidities and mortality were not reported.
The two included trials provided insufficient evidence to address the objectives of this review, which concerned the effects of flow-cycled versus time-cycled synchronized ventilation on clinical outcomes including major and minor morbidities and mortality. Thus, this review cannot establish whether the use of flow-cycled versus time-cycled ventilation is associated with clinically important benefits or harms. Given that several non-randomized studies excluded from this review similarly focused on lung mechanics, the widespread use of flow-cycled modes of synchronized ventilation is not based on clinical evidence from clinical trials.
The quality of available evidence to assess safety and efficacy of flow-cycled compared to time-cycled synchronized ventilation in neonates is poor due to extremely small samples, lack of blinding, very short follow-up, and the lack of clinical endpoints.
Only randomized and quasi-randomized trials were eligible for inclusion in the review. Three small controlled trials were excluded due to lack of randomization (Abubakar 2001; Migliori 2003; Scopesi 2007). The combined sample size of these excluded trials (N = 53) is considerably larger than that of the two trials included in the review (N = 19). However, it is unlikely that important clinical results were missed given that the concerned excluded studies only reported short-term physiological outcomes (follow-up time 2 to 16 hours) but no clinical outcomes. Abubakar 2001 compared PSV with SIMV in 23 neonates and reported shorter inspiratory time and lower mean airway pressure in PSV but no difference in tidal volume or blood gases. Migliori 2003 compared PSV with SIMV in 20 preterm neonates and found higher mean airway pressure, higher tidal and minute volume, and lower respiratory rate in PSV but no difference in blood gases or SaO2. Scopesi 2007 studied PSV versus ACV versus SIMV in 10 preterm neonates and showed greater variability of tidal volume in SIMV versus PSV and ACV, and lower peak inspiratory pressures in ACV versus PSV and SIMV but no difference in blood gases.
Thus, the strict inclusion criteria of this review don't seem to have introduced major bias in terms of missing important clinical findings.
There is insufficient evidence to determine the safety and efficacy of flow-cycled compared to time-cycled synchronized ventilation in neonates.
Although there is biological plausibility for the use of flow-cycled synchronized ventilation in neonates, and many clinicians are using ventilators providing this mode of ventilation, randomized clinical trials of flow-cycled versus time-cycled synchronized ventilation are warranted. Future studies should enroll large numbers of neonates, use a parallel-group design, and report on clinically important outcomes including those specified in the criteria for considering studies for this review. Given the health burden of very preterm birth and evidence from a previous systematic review indicating a tendency towards a benefit of ACV over SIMV, priority should be given to studies in very preterm infants comparing flow-cycled synchronized ventilation with ACV.
We thank Dr. Daniele De Luca and Dr. Robert Kacmarek for clarification of study methodology.
The Cochrane Neonatal Review Group has been funded in part with Federal funds from the Eunice Kennedy Shriver National Institute of Child Health and Human Development National Institutes of Health, Department of Health and Human Services, USA, under Contract No. HHSN267200603418C.
SMS: Designed the review, assessed methodological quality, performed data extraction, wrote the manuscript.
JJP: Contributed to design of the review and revised the manuscript.
BE: Contributed to design of the review and revised the manuscript.
SKP: Contributed to design of the review, assessed methodological quality, performed data extraction, revised the manuscript.
| Methods | Single-center study performed in Rome, Italy. Allocation concealment: Yes Blinding of caretakers to intervention: No Blinding of outcome ascertainment: No Complete follow-up: Yes |
|---|---|
| Participants | Ten preterm neonates (interquartile GA range 26 - 30.5 weeks, interquartile birth weight range 735 - 1475 g) with RDS were enrolled on day one of life. One eligible infant was excluded pre-randomization due to intraventricular hemorrhage and seizures. Data from ten clinically stable infants (median postnatal age, 12 h; median FiO2, 0.21) were analyzed. |
| Interventions | Patients were ventilated using Bear Cub 750PSV ventilators (Viasys Healthcare, Conshohocken, PA, USA). Using a random number table with codes retrieved from sealed opaque envelopes, flow-cycled or time-cycled synchronized ventilation was randomly assigned to each patient for a 1 h period. After 1 h, the mode ventilation was changed to the alternative mode. Each ventilation modality was used for a period of 15 minutes prior to collecting pulmonary function data over the following 45 min. Airway pressure and airflow were measured using the ventilator's hot wire anemometer and manometer, both placed at the y-piece between ventilator circuit and endotracheal tube. Prior to each study window, central venous pressure was measured and capillary blood gas analysis was performed. Pulse oximetry, transcutaneous oxygen and carbon dioxide levels were monitored every 10 min. |
| Outcomes | Short-term physiological outcomes: Respiratory rate Expiratory tidal volume Minute ventilation Mean airway pressure Spontaneous inspiratory time Dynamic compliance Total resistance Rate x mean airway pressure product Rate volume ratio Oxygen saturation Heart rate Blood gases Transcutaneous oxygen and carbon dioxide levels Central venous pressure Oxygenation index Work of breathing generated by the ventilator (but not work of breathing of patients) |
| Notes | The authors clarified methodological details of the study. |
| Item | Judgement | Description |
|---|---|---|
| Adequate sequence generation? | Yes | Randomization sequence was generated using a random number table |
| Allocation concealment? | Yes | |
| Blinding? | No | Blinding of caretakers was not possible due to the nature of the interventions. Blinding of outcome ascertainment was not attempted |
| Incomplete outcome data addressed? | Yes | All prespecified outcomes reported |
| Free of selective reporting? | Yes | None detected |
| Free of other bias? | Yes | None detected |
| Methods | Single-center study performed in Boston, MA, USA. Allocation concealment: No Blinding of caretakers to intervention: No Blinding of outcome ascertainment: No Complete follow-up: Yes |
|---|---|
| Participants | Nine clinically stable preterm neonates (GA range 28 - 36 weeks, birth weight range 925 - 2100 g) with respiratory distress syndrome (RDS) were enrolled between the second and eighth day of life. Data from two infants were excluded after randomization due to technical problems in data transfer between pneumotachograph and computer. Data from seven infants (mean ± SD GA 31.4 ± 2.0 w, weight 1.49 ± 0.38 kg) were analyzed. |
| Interventions | Patients were ventilated using Bird VIP ventilators (Bird Product, Riverside, CA, USA). In each patient, the mode of ventilation was randomly changed between IMV, SIMV, ACV, and PSV based on a code from a random number table. Each ventilation modality was used for a period of 20 minutes before pulmonary function data was collected over a period of 2 - 3 min. Airway pressure and airflow were measured with a pneumotachograph between ventilator circuit and endotracheal tube. Esophageal pressure was measured with an esophageal balloon. Pulse oximetry and transcutaneous carbon dioxide levels were monitored continuously. |
| Outcomes | Short-term physiological outcomes: Trigger delay Airway trigger pressure change Esophageal trigger pressure change Endotracheal tube leak volume Heart rate Respiratory rate Arterial oxygen saturation Transcutaneous carbon dioxide level Minute ventilation Inspiratory time Inspiratory work of breathing Pressure-time products |
| Notes | The authors clarified methodological details of the study. |
| Item | Judgement | Description |
|---|---|---|
| Adequate sequence generation? | Yes | Randomization sequence was generated using a random number table |
| Allocation concealment? | No | Codes were not concealed |
| Blinding? | No | Blinding of caretakers was not possible due to the nature of the interventions. Blinding of outcome ascertainment was not attempted |
| Incomplete outcome data addressed? | Yes | All prespecified outcomes reported |
| Free of selective reporting? | Yes | None detected |
| Free of other bias? | Yes | None detected |
RDS - respiratory distress syndrome
GA - gestational age
Fi02 - Fraction of inspired Oxygen
h - hour
| Reason for exclusion | Controlled short-term individual cross-over trial (N = 23) in neonates comparing SIMV, ACV, and PSV with and without volume targeting option. Treatment allocation (i.e. order of using different ventilation modes) not randomized or quasi-randomized. The order in which mode of ventilation was changed was as follows: Comparison of PSV versus SIMV: SIMV - PSV - PSV+VG - SIMV - SIMV+VG - SIMV; Comparison of PSV versus ACV: ACV - PSV - PSV+VG - ACV - ACV+VG - ACV. |
|---|
| Reason for exclusion | Controlled short-term individual cross-over trial (N = 20) comparing PSV with SIMV in preterm neonates. Treatment allocation (i.e. order of using different ventilation modes) not randomized or quasi-randomized. The order in which mode of ventilation was changed was as follows: SIMV - PSV - SIMV - PSV. |
|---|
| Reason for exclusion | Randomized clinical trial (N = 34) comparing PSV with volume targeting to SIMV without volume targeting in preterm neonates. Excluded because volume targeting was only used in PSV group. |
|---|
| Reason for exclusion | Randomized short-term individual cross-over trial (N = 14) comparing PSV and volume targeting with SIMV without volume targeting in preterm neonates. Excluded because volume targeting was only used in PSV mode. |
|---|
| Reason for exclusion | Controlled short-term individual cross-over trial (N = 10) comparing PSV with SIMV and ACV in preterm neonates. Treatment allocation (i.e. order of using different ventilation modes) not randomized or quasi-randomized. The order in which mode of ventilation was changed was as follows: Comparison of PSV versus ACV versus SIMV: SIMV - ACV+VG - SIMV - PSV+VG - SIMV - SIMV+VG - SIMV. |
|---|
| Reason for exclusion | Randomized clinical trial (N = 50) comparing volume-controlled with time-cycled synchronized ventilation in preterm neonates. Excluded because there was no comparison of flow-cycled versus time-cycled synchronized ventilation. |
|---|
| Reason for exclusion | Randomized short-term individual cross-over trial comparing different levels of pressure support in non-intubated neonates assisted by CPAP. Excluded because neonates were not intubated and there was no control group with time-cycled synchronized ventilation. |
|---|
Abubakar KM, Keszler M. Patient-ventilator interactions in new modes of patient-triggered ventilation. Pediatric Pulmonology 2001;32(1):71-5.
Migliori C, Cavazza A, Motta M, Chirico G. Effect on respiratory function of pressure support ventilation versus synchronised intermittent mandatory ventilation in preterm infants. Pediatric Pulmonology 2003;35(5):364-7.
Nafday SM, Green RS, Lin J, Brion LP, Ochshorn I, Holzman IR. Is there an advantage of using pressure support ventilation with volume guarantee in the initial management of premature infants with respiratory distress syndrome? A pilot study. Journal of Perinatology 2005;25(3):193-7.
Olsen SL, Thibeault DW, Truog WE. Crossover trial comparing pressure support with synchronized intermittent mandatory ventilation. Journal of Perinatology 2002;22(6):461-6.
Scopesi F, Calevo MG, Rolfe P, Arioni C, Traggiai C, Risso FM, et al. Volume targeted ventilation (volume guarantee) in the weaning phase of premature newborn infants. Pediatric Pulmonology 2007;42(10):864-70.
Abd El-Moneim ES, Fuerste HO, Krueger M, Elmagd AA, Brandis M, Schulte-Moenting J, et al. Pressure support ventilation combined with volume guarantee versus synchronized intermittent mandatory ventilation: a pilot crossover trial in premature infants in their weaning phase. Pediatric Critical Care Medicine 2005;6(3):286-92.
Aly H, Badawy M, El-Kholy A, Nabil R, Mohamed A. Randomized, controlled trial on tracheal colonization of ventilated infants: can gravity prevent ventilator-associated pneumonia? Pediatrics 2008;122(4):770-4.
Attar MA, Donn SM. Mechanisms of ventilator-induced lung injury in premature infants. Seminars in Neonatology 2002;7(5):353-60.
Chen JY, Ling UP, Chen JH. Comparison of synchronized and conventional intermittent mandatory ventilation in neonates. Acta Paediatrica Japonica 1997;39(5):578-83.
Donn SM, Nicks JJ, Becker MA. Flow-synchronized ventilation of preterm infants with respiratory distress syndrome. Journal of Perinatology 1994;14(2):90-4.
Donn SM, Sinha SK. Minimising ventilator induced lung injury in preterm infants. Archives of Disease in Childhood. Fetal and Neonatal Edition 2006;91(3):F226-30.
Greenough A, Dimitriou G, Prendergast M, Milner AD. Synchronized mechanical ventilation for respiratory support in newborn infants. Cochrane Database of Systematic Reviews 2008, Issue 1. Art. No.: CD000456. DOI: 10.1002/14651858.CD000456.pub3.
Heicher DA, Kasting DS, Harrod JR. Prospective clinical comparison of two methods for mechanical ventilation of neonates: rapid rate and short inspiratory time versus slow rate and long inspiratory time. Journal of Pediatrics 1981;98:957-61.
Jarreau PH, Moriette G, Mussat P, Mariette C, Mohanna A, Harf A, et al. Patient-triggered ventilation decreases the work of breathing in neonates. American Journal of Respiratory and Critical Care Medicine 1996;153(3):1176-81.
Jobe AH, Bancalari E. Bronchopulmonary dysplasia. American Journal of Respiratory and Critical Care Medicine 2001;163(7):1723-9.
Poets CF, Rau GA, Neuber K, Gappa M, Seidenberg J. Determinants of lung volume in spontaneously breathing preterm infants. American Journal of Respiratory and Critical Care Medicine 1997;155(2):649-53.
Pohlandt F, Saule H, Schroder H, Leonhardt A, Hornchen H, Wolff C, et al. Decreased incidence of extra-alveolar air leakage or death prior to air leakage in high versus low rate positive pressure ventilation: results of a randomised seven-centre trial in preterm infants. European Journal of Pediatrics 1992;151(12):904-9.
Schmalisch G, Wilitzki S, Wauer RR. Differences in tidal breathing between infants with chronic lung diseases and healthy controls. BMC Pediatrics 2005;5:36.
Sherman JM, Lowitt S, Stephenson C, Ironson G. Factors influencing acquired subglottic stenosis in infants. Journal of Pediatrics 1986;109(2):322-7.
The following search strategy was used:
("Infant, newborn" [MeSH] OR "Infant, premature" [MeSH] OR infant* OR neonate* OR newborn*) AND ("Respiration, Artificial"[MeSH] OR "synchronized intermittent mandatory ventilation" OR "synchronised intermittent mandatory ventilation" OR "assist-control ventilation" OR "assist control ventilation" OR "pressure support ventilation" OR "assisted spontaneous breathing" OR "A-C ventilation" OR "A-C mode" OR "AC ventilation" OR "AC mode" OR SIMV OR SIPPV OR PSV OR ASB) AND ("controlled clinical trial" [Publication Type] OR "randomized controlled trial" [Publication Type]).
| This review is published as a Cochrane review in The Cochrane Library, Issue 7, 2010 (see http://www.thecochranelibrary.com for information). Cochrane reviews are regularly updated as new evidence emerges and in response to feedback. The Cochrane Library should be consulted for the most recent version of the review. |
|---|