Pediatric Pulmonology 42:704–710 (2007)

Effects of Non-Invasive Pressure Support Ventilation(NI-PSV) on Ventilation and Respiratory Effort in

Very Low Birth Weight Infants

Nabeel Ali, MD, Nelson Claure, PhD,* Ximena Alegria, MD, Carmen D’Ugard, RRT,Roberto Organero, MD, and Eduardo Bancalari, MD

Summary. Background: Nasal continuous positive airway pressure (NCPAP) is used to provide

support to non-intubated infants, but it often fails. Pressure support ventilation (PSV) is a mode of

synchronized ventilation that can supplement the spontaneous breathing effort, but it is unknown if

it is effective in non-intubated very low birth weight (VLBW) infants. Objectives: To compare the

acute physiological effects of non-invasive PSV (NI-PSV) versus NCPAP on tidal volume (VT),

minute ventilation (VE), gas exchange, breathing effort, and chest wall distortion in VLBW infants.

Methods: Stable preterm infants of birth weight less 1,250 g were studied during consecutive 2 hr

periods of NCPAP and NI-PSV in random sequence. VT, VE, and thoraco-abdominal synchrony

weremeasured using respiratory inductance plethysmography. Breathing effort wasmeasured by

esophageal manometry. Gas exchange was measured by pulse oximetry and transcutaneous

PCO2. Results: Fifteen infants of birth weight (mean�SD) 808�201 g and 25.9�1.8 weeks

gestational age were studied while on NCPAP 5.3�0.6 cm H2O and on NI-PSV with 7.9� 1.3 cm

H2O above NCPAPof pressure support. There were no differences in VT, VE, PCO2 or hypoxemia

episodes. Peak and minute inspiratory effort were significantly reduced in NI-PSV mode as

compared to NCPAP. There was a significant reduction in indices of chest wall asynchrony in NI-

PSV mode. Conclusion: When compared to NCPAP, NI-PSV did not increase minute ventilation,

but it effectively unloaded the patient’s respiratory pump as indicated by a lower inspiratory effort

and reduced chest wall distortion. Pediatr Pulmonol. 2007; 42:704–710. � 2007 Wiley-Liss, Inc.

Key words: preterm infant; breathing effort; continuous distending pressure; non-

invasive ventilation.

INTRODUCTION

Nasal continuous positive airway pressure (NCPAP) iswidely used in neonatal intensive care to supportpremature infants with respiratory distress1 or apnea.2,3

NCPAP has been shown to relieve upper airway obstruc-tion,4,5 improve the synchrony of thoracic andabdominal movements,6,7 and increase end-expiratorylung volume.8,9 However, infants managed with NCPAPsometimes fail, and require endotracheal intubation, thusexposing them to the complications associated withinvasive mechanical ventilation. Rates of failure of post-extubation NCPAP may reach 40%,10–12 while failure as aprimary treatment of RDS may reach 60%.13

Pressure support ventilation (PSV) is a ventilatorymode where positive pressure is used to assist each of thepatient’s spontaneous breaths. A pressure support breath isinitiated and terminated in synchrony with the patient’sinspiratory effort. PSV has been shown to successfullyaugment spontaneous breathing and reduce thoraco-abdominal asynchrony in intubated infants,14 as well asreduce inspiratory effort in preterm neonates.15

The non-invasive application of PSV (NI-PSV) is arecognized mode of support in adults with respiratory

failure16 where it has been used successfully to treatCOPD,17 atelectasis and post-surgical patients.18 Adultstudies show that NI-PSV improves tidal volume, gasexhange17 and reduces patient effort.19 However applica-tion of NI-PSV in neonates has not been reported.

Frequent failure of NCPAP in neonates has led tointerest in developing additional methods of non-invasiveventilatory support. Due to weak musculature and a

Division of Neonatology, Department of Pediatrics, University of Miami

Miller School of Medicine, Miami, Florida.

Grant sponsor: University of Miami Project NewBorn.

*Correspondence to: Nelson Claure, Department of Pediatrics, Division of

Neonatology, University of Miami Miller School of Medicine, PO Box

016960 R-131, Miami, FL 33101. E-mail: nclaure@miami.edu

Received 10 October 2006; Revised 4 February 2007; Accepted 24

February 2007.

DOI 10.1002/ppul.20641

Published online 26 June 2007 in Wiley InterScience

(www.interscience.wiley.com).

� 2007 Wiley-Liss, Inc.

compliant ribcage, preterm infants are prone to chest walldistortion during breathing. The underlying lung diseasefurther increases their respiratory workload and theprovision of continuous distending pressure alone is oftennot sufficient to support their breathing. NI-PSV canpotentially boost their spontaneous inspiratory effort andprovide a degree of unloading not achievable by NCPAPalone.

The objective of this study was to evaluate thephysiological effects of NI-PSV in premature infants onNCPAP. The hypothesis of the present study was thatNI-PSV would improve minute ventilation when com-pared to NCPAP, and that secondarily, it would improvegas exchange, reduce patient effort, and reduce chest walldistortion.

METHODS

Patient Population

Clinically stable preterm infants, appropriate forgestational age with birth weights between 500 and1,250 g currently on CPAP and requiring less than 50%supplemental oxygen were eligible for the study. Infantswith major congenital anomalies, neuromuscular disease,lung hypoplasia, hemodynamic instability and pulmonaryair leaks were excluded. The study was conducted at theUniversity of Miami/Jackson Memorial Hospital Neona-tal Intensive Care Unit. All infants were recruited andstudied after written informed parental consent wasobtained. The study was approved by the University ofMiami Human Subjects Research Office and the WesternInstitutional Review Board, Olympia, WA. It wasestimated that enrollment of at least 15 infants was neededto detect a 30% change in minute ventilation between thetwo modes, with a power of 0.8 and a two-tailed alphalevel of 0.05.

Study Protocol

Infants were studied supine while in their incubators.After a 15-min stabilization period, each infant wasstudied either in NCPAP or NI-PSV mode for 2 hr.Thereafter, the infant was crossed over to the other modeand allowed to stabilize for 15 min. The second 2-hr periodof recording ensued. The sequence of NI-PSVand NCPAPwas assigned at random using sealed opaque envelopes.Investigators were not blinded to the treatment mode. Theinfants were left undisturbed during both periods.

NCPAP and NI-PSV were provided with a Sechrist IV-200 SAVI ventilator (Sechrist Industries, Anaheim, CA).NCPAP was delivered using the flow rate and pressure setby the clinical team and administered through binasalINCA prongs (Ackrad Labs, Copper Surgical, Trumbull,CT). NI-PSV was delivered using the ventilator in SAVImodewhich provides for synchronization of both the onsetand end of inspiration. Pressure support breaths weretriggered and terminated in synchrony with the abdominalwall movement using the signal obtained from respiratoryinductance plethysmography (see below). During NI-PSV, CPAP was maintained at the same level and pressuresupport was added to assist each breath. The level ofpressure support was adjusted to match 100–150% of thepatient’s spontaneous inspiratory effort as measured byesophageal pressure (see below). Maximal inspiratorytime was limited to 0.45 sec. Trigger sensitivity wasadjusted in each infant to achieve a minimal responsedelay without auto-triggering. No backup rate was usedduring the NI-PSV mode, since none is used routinelyduring NCPAP and to avoid interfering with the measure-ment of the frequency of apneic episodes.

Infants were monitored at the bedside throughout thestudy by the research team. In case of prolonged apneicspells infants received gentle stimulation. In the event ofhypoxemia, defined as drop in oxygen saturation below85%, FiO2 was increased by 0.2 after 30 sec, followed by0.1 increments every 20 sec thereafter, and weaned to theprevious baseline level after recovery. Nursery guidelinesfor assistance during a severe event of apnea or hypoxemiawere followed if needed. The infant’s existing feedingtube was left in place throughout the study for feeds andgastric decompression.

Measurements

Esophageal pressure (PES) measurements with a size6 French water filled feeding tube placed in the loweresophagus were used to estimate changes in pleuralpressure. Airway pressure (PAW), that is, the pressuredelivered by the ventilator during NCPAP and NI-PSV,was measured at the nasal adapter. Both PES and PAW weremeasured with pressure transducers (Sorenson transpac42586-1, Abbot critical care systems, North Chicago, IL)connected to medical grade transducer couplers (Gould

ABBREVIATIONS

NCPAP nasal continuous positive airway pressure

NI-PSV non-invasive pressure support ventilation

N-SIMV nasal synchronized intermittent mandatory ventilation

PAW airway pressure

PES esophageal pressure

PES area area under esophageal pressure curve

PESminute area area under esophageal pressure curve per minute

PES peak peak esophageal pressure

SpO2 arterial oxygen saturation by pulse oximetry

TCD total compartment displacement

TcPCO2 transcutaneous carbon dioxide tension

TcPO2 transcutaneous oxygen tension

V0E minute ventilation

VLBW very low birth weight

VT tidal volume

WOBspont spontaneous work of breathing

Non-Invasive Pressure Support Ventilation in Preterm Infants 705

Electronics, Inc., Valley View, OH) and calibrated bywater manometry. Esophageal pressure transmission wasverified by a brief airway occlusion.

Respiratory activity and chest wall instability weremeasured by (RIP) respiratory inductance plethysmogra-phy (Respitrace Plus, Sensormedics Corporation, YorbaLinda, CA) using two soft elastic bands placed around theabdomen and chest. These bands were secured withtape. The chest wall and abdominal signal voltages wereadded to yield a third signal, denoted the sum signal.Calibration of RIP was achieved using the built-inqualitative diagnostic calibration procedure as describedpreviously.20

Arterial oxygen saturation (SpO2) was continuouslymeasured by pulse oximetry (Masimo Radical, MasimoCorp., Irvine, LA). The averaging time in the pulseoximeter was set at 8 sec for all infants. TranscutaneousPCO2 (TcPCO2) was measured using a Microgas 7650transcutaneous carbon dioxide monitor (SensormedicsCorporation). The fraction of inspired oxygen (FiO2) wasmeasured with an oxygen analyzer (O2000 Maxtec,Thermo respiratory group, Salt Lake City, UT). Allsignals were digitized at 100 samples per second, andrecorded on a personal computer (AT-CODAS, DataqInstruments, Akron, OH).

Data Analysis

Mean FiO2, PCO2, SpO2, as well as the number ofepisodes of hypoxemia with SpO2 below 85% and thepercent of time spent with SpO2 <85% were determinedby automated analysis of the entire recording periodsduring NCPAP and NI-PSV. The number of apneic spellslasting more than 20 sec was determined by reviewing allrecordings. Apnea was defined as a flat RIP signal withoutany negative deflection of esophageal pressure. Tidalvolume (VT), minute ventilation (VE), and respiratory rate(RR) were determined by automated analysis of the sumsignal of RIP throughout each of the 2 hr recordingperiods and reported in arbitrary units, arbitrary units perminute and breaths per minute respectively. Relativechange in VE for each patient was defined as the differencebetween VE during NI-PSV and VE during NCPAP overthe VE during NCPAP expressed as a percentage. Toavoid bias during the data analysis, the above-mentionedoutcome parameters were calculated by automated soft-ware throughout each of the recording periods withoutoperator intervention.

Per-breath spontaneous inspiratory effort was deter-mined by peak esophageal pressure (PES peak) as well asthe area under the esophageal pressure curve (PES area).Minute effort (PES area minute) was calculated as the productof PES area multiplied by the average respiratory rate perminute. This variable was used to account for cases wherethe per-breath inspiratory effort may change in an opposite

direction of the respiratory rate resulting in a net zero oropposite change in breathing effort over time.

Spontaneous patient work of breathing (WOBspont) wascalculated as the product of tidal volume and esophagealpressure throughout inspiration. Esophageal pressurerather than transpulmonary pressure was used in thecalculation of work to determine the infant’s owncontribution to the work of breathing. Thoraco-abdominalmotion was analyzed by determining the phase delaybetween abdominal and thoracic expansion expressed indegrees with the full respiratory cycle represented by3608. The ratio of the displacement of the ribcage andabdomen compartments to the net tidal volume wasdenoted as the total compartment displacement ratio(TCD ratio). TCD ratio was calculated as describedpreviously by Musante et al.20 Briefly, the areas underthe curve for the abdominal and thoracic band signalswere determined. The sum of the absolute values of theabdominal and thoracic areas were divided by theiralgebraic sum (sign included), yielding the TCD ratio. Incases where no chest wall distortion is present, theTCD ratio is equal to 1. Increasing values greater than1 indicated higher degrees of chest-wall distortion.

Data on outcome parameters of breathing effort, workof breathing and chest wall distortion require operatoridentification of artifacts. To avoid bias a criterion forbreath selection was established. With this, the operatorincluded the first three consecutive artifact-free breathsfor every 5 min throughout each of the recording periods.The above-mentioned parameters were calculated from asample of approximately 72 breaths per 2 hr period perinfant. Breaths were considered unsuitable for analysis ifthere was drift of the baseline PES signal, drift of the RIPsignal or movement artifact.

Statistical Analysis

Within subjects comparisons of the data were obtainedusing paired t-test or the Wilcoxon signed rank test whendifferences were not normally distributed. A two-tailedalpha value of 0.05 (P< 0.05) was considered assignificant. Comparisons of continuous variables betweenpost hoc groups were made using the Mann–WhitneyU-test for difference in medians. Statistical comparisonswere done using Sigmastat version 3.10 (Systat Software,Inc., Point Richmond, CA) and NCSS 2001 (NumberCruncher Statistical Systems, Kaysville, UT).

RESULTS

Fifteen preterm infants were studied. Their birth weight(mean� SD) was 808� 201 g and gestational age at birthwas 25.9� 1.8 weeks. At the time of study, they were25.0� 20.0 days old and weighed 972� 215 g. Of thefifteen infants studied, seven were male. Twelve ofthe infants had been mechanically ventilated prior to the

706 Ali et al.

study. All infants were receiving caffeine for apnea. Eightinfants were assigned to start in CPAP mode, while theother seven started in NI-PSV. Mean NCPAP level duringthe study was 5.3� 0.6 cm H2O and the pressure supportused during the NI-PSV phase was 7.9� 1.3 cm H2Oabove the NCPAP level.

All infants completed the 4-hr study and there were noadverse events including death, pneumothorax, bowelperforation or intraventricular hemorrhage. Averageabdominal circumference was 20.2� 2.0 cm duringNCPAP and 20.3� 2.1 cm at the end the of NI-PSVperiod.

Table 1 shows the results for ventilation and gasexchange parameters in both modes. VT and VE remainedunchanged across both modes of support. There was nodifference in TcPCO2 or in the number of hypoxemiaepisodes with SpO2 below 85% between both modes ofsupport. There was a small reduction in the required FiO2

during NI-PSV. This reduction however may not beclinically relevant. Apneic episodes lasting longer than 20sec were rare during both periods, with a total of 10 beingrecorded throughout the whole study among all patients (7during NCPAP and 3 during NI-PSV). Some apneic spellsrequired gentle stimulation before recovery. Nonerequired bag and mask ventilation.

Table 2 shows data for inspiratory effort and thoraco-abdominal motion across both modes. There was asignificant reduction in per breath (PES peak and PES area)and per minute inspiratory effort (PES minute area) duringNI-PSV. This is illustrated in the sample tracing ofFigures 1 and 2 shows that the reduction in PES peak wasconsistent in all infants. Spontaneous work of breathing(WOBspont) during NI-PSV was also reduced.

There was a significant reduction in chest walldistortion with the application of NI-PSVas demonstratedby the reductions in phase shift angle and TCD ratio(Table 2). This is also illustrated in Figure 1.

The post hoc comparison of the relative change in VE

from NCPAP to NI-PSV in infants with an initial TcPCO2

in NCPAP <55 mmHg to those with an initial TcPCO2 inNCPAP �55 mmHg is shown in Figure 3. Infants with an

initial TcPCO2 �55 mmHg had greater change in VE

(26.0% versus �8.4%, P¼ 0.042).

DISCUSSION

The aim of this study was to compare the acute effectsof NI-PSV to NCPAP on minute ventilation and gasexchange in VLBW infants. Contrary to the initialhypothesis, the results show that NI-PSV did not producea change in minute ventilation, respiratory rate or tidalvolume. Oxygenation and carbon dioxide elimination alsoremained unchanged. It is possible that the infants’spontaneous ventilation was already close to adequatebefore entry into the study, considering their state of lungdisease, and that they reacted to the addition of pressuresupport by reducing their effort instead of improving theirtidal volume. The fact that there is a striking 34%reduction in per breath inspiratory effort (PES peak) and a42% drop in total minute effort (PES minute area) supportsthis explanation. These findings suggest that NI-PSVeffectively unloads the patient.

It can be argued that the measured reduction in negativeesophageal pressure is in part artifactual and simply theintra-thoracic transmission of the positive pressure breathsduring NI-PSV. However, previous studies have shownthat airway pressure transmission to the esophagus isapproximately 17% in intubated preterm infants.21 Theobserved reduction in PES during NI-PSV exceeded thislevel and therefore cannot be explained by transmission ofairway pressure alone.

Chest wall distortion was also improved with applica-tion of NI-PSV, as demonstrated by the significantreductions in TCD ratio and phase shift angle. Similarimprovements in thoraco-abdominal asynchrony wereobserved in intubated infants receiving proportional assistventilation compared to endotracheal CPAP.20 Thisfurther supports the hypothesis that NI-PSV effectivelyunloads the infant’s respiratory pump resulting in a moreefficient use of spontaneous breathing effort. It is possiblethat some of the observed effects of NI-PSV are due to areduction in upper airway resistance. Support breaths

TABLE 1— Ventilation and Gas Exchange

NCPAP NI-PSV P

FiO2 (%) 28 (21–46) 27 (21–41) 0.009

SpO2 (%) 92 (89–98) 92 (90–98) 0.115

Frequency of hypoxemia episodes (SpO2 <85%, episodes per hour) 5.1 (0.8–19.5) 8.3 (0–21.8) 0.649

Duration of hypoxemia (% of time with SpO2 <85%) 8.0� 7.8 7.0� 7.1 0.165

PCO2 (mmHg) 54 (35–82) 51 (32–73) 0.348

Respiratory rate (breaths/min) 42.9� 9.0 42.8� 8.9 0.942

VT (arbitrary units) 13.1� 4.6 13.0� 3.8 0.913

VE (arbitrary units/min) 565� 246 556� 190 0.837

Data in mean� SD or median and (10th–90th percentile).

Non-Invasive Pressure Support Ventilation in Preterm Infants 707

delivered in synchrony with patients breathing effortsmay prove more effective in splinting the soft tissues of theupper airway than continuous distending pressure.

The data suggest that some infants do responddifferently to NI-PSV than others. While all infantsreduced their effort in NI-PSV, post hoc analysis showedthat infants with baseline levels of TcPCO2 �55 mmHgin NCPAP had a greater increase in minute ventilation.Therefore, infants with increased ventilatory demands orthose infants whose respiratory pump is unable to fullymeet their needs were probably more likely to benefit fromadditional support and thus show an increase in VE.

This is the first study to investigate the effects of NI-PSVin VLBWinfants. Two previous studies report similarfindings using non-invasive assist/control ventilation inlarger preterm infants over shorter periods of time.22,23

The observed reduction in breathing effort appears to berelated to the non-invasive application of positive pressurebreaths. Pressure support ventilation and assist-controlventilation are similar since they both support everyspontaneous inspiration. However, there may be added

benefits to the synchronous breath termination offered byNI-PSV, such as a reduction in the risk of abdominaldistention from swallowed gas, avoidance of a prolongedinspiratory hold and a greater ability to regulate breathingfrequency due to a greater control of breath duration. Thishowever remains to be explored.

While this study does suggest that NI-PSV has benefitsas a mode of support, it is not known whether the observedacute reduction in breathing effort would translate into along-term improved success rate of non-invasive ventila-tion. There is also no data on the long-term safety of thismode of respiratory support. In this study, backup rate inthe case of apnea was not used during either mode. Thefrequency of apnea observed was small during the study.However, in infants with more immature control ofbreathing, or infants who are not receiving respiratorystimulants, the use of a backup ventilator rate may berequired. Air leaks due to an inadequate seal at the nostrilsor due to an open mouth play an important role in themaintenance of a continuous distending pressure duringNCPAP. This may affect the transmission of positive

TABLE 2— Inspiratory Effort, Work of Breathing and Chest Wall Distortion

NCPAP NI-PSV P

PES peak (cm H2O) 7.72� 3.76 5.08� 3.01 <0.001

PES area (cm H2O sec) 3.60� 1.73 2.18� 1.39 <0.001

PES minute area (cm H2O sec/min) 137 (77–220) 79 (43–133) <0.001

WOBspont (cm H2O arbitrary unit) 79.9� 59.8 44.8� 42.1 <0.001

Phase angle (8) 63.2� 21.9 47.7� 18.6 0.002

TCD ratio 2.98� 1.64 1.67� 0.54 0.003

Data in mean� SD or median and (10th–90th percentile).

Fig. 1. A representative recording from an infant during NCPAP (A) and NI-PSV (B) showing

airway pressure (Paw), esophageal (PES), chest wall (RIP CW) and abdominal (RIP AB) respiratory

inductance. Onset and termination of the pressure support breath occur in synchrony with the

rise and fall of RIP AB. Note the reduction in PES deflection and the attenuation in chest wall

distortion (reduced inward motion and shorter lag in RIP CW denoted by the dashed lines) during

NI-PSV.

708 Ali et al.

pressure during NI-PSV thus partially limiting itseffectiveness in providing mechanical unloading duringprolonged use in a clinical setting.

Studies have looked at the application of nasalsynchronized intermittent mandatory ventilation (N-SIMV) to support preterm infants after extubation10–12

and have shown that it is an effective modality in reducingextubation failure. N-SIMVonly assists a fixed number ofthe infants’ breaths. Assistance of all breaths, as providedby NI-PSV, may provide further unloading which maybenefit infants whose respiratory pump cannot cope withthe increased loads induced by their underlying lungdisease while combining NI-PSV with a fixed number ofN-SIMV breaths may alleviate some of the limitations of

NI-PSV in infants with immature respiratory drive. Theassistance of every breath in a non-invasive manner is notnovel. However, the reported data on the effects of thistype of support is useful when evaluating the prospectiveuse of this form of support for preterm infants. These datashould be considered when planning future larger trials.

In summary, NI-PSV successfully provides respiratoryunloading to preterm infants. While tidal volume, minuteventilation and PCO2 remained unchanged, there was aconsistent reduction in inspiratory effort and thoraco-abdominal asynchrony. Although this remains to beproven, NI-PSV may be effective in supporting preterminfants following extubation or in avoiding intubationaltogether.

REFERENCES

1. Verder H, Albertsen P, Ebbesen F, Greisen G, Robertson B,

Bertelsen A, Agertoft L, Djernes B, Nathan E, Reinholdt J. Nasal

continuous positive airway pressure and early surfactant therapy

for respiratory distress syndrome in newborns of less than

30 weeks’ gestation. Pediatrics 1999;103:E24.

2. Lin CH, Wang ST, Lin YJ, Yeh TF. Efficacy of nasal intermittent

positive pressure ventilation in treating apnea of prematurity.

Pediatr Pulmonol 1998;26:349–353.

3. Ryan CA, Finer NN, Peters KL. Nasal intermittent positive-

pressure ventilation offers no advantages over nasal continuous

positive airway pressure in apnea of prematurity. Am J Dis Child

1989;143:1196–1198.

4. Miller MJ, Carlo WA, Martin RJ. Continuous positive airway

pressure selectively reduces obstructive apnea in preterm infants.

J Pediatr 1985;106:91–94.

5. Miller MJ, DiFiore JM, Strohl KP, Martin RJ. Effects of nasal

CPAP on supraglottic and total pulmonary resistance in preterm

infants. J Appl Physiol 1990;68:141–146.

6. Locke R, Greenspan JS, Shaffer TH, Rubenstein SD, Wolfson

MR. Effect of nasal CPAP on thoracoabdominal motion in

neonates with respiratory insufficiency. Pediatr Pulmonol

1991;11:259–264.

7. Kiciman NM, Andreasson B, Bernstein G, Mannino FL, Rich W,

Henderson C, Heldt GP. Thoracoabdominal motion in newborns

during ventilation delivered by endotracheal tube or nasal prongs.

Pediatr Pulmonol 1998;25:175–181.

8. Courtney SE, Pyon KH, Saslow JG, Arnold GK, Pandit PB, Habib

RH. Lung recruitment and breathing pattern during variable

versus continuous flow nasal continuous positive airway pressure

in premature infants: An evaluation of three devices. Pediatrics

2001;107:304–308.

9. Elgellab A, Riou Y, Abbazine A, Truffert P, Matran R, Lequien P,

Storme L. Effects of nasal continuous positive airway pressure

(NCPAP) on breathing pattern in spontaneously breathing

premature newborn infants. Intensive Care Med 2001;27:1782–

1787.

10. Barrington KJ, Bull D, Finer NN. Randomized trial of nasal

synchronized intermittent mandatory ventilation compared with

continuous positive airway pressure after extubation of very low

birth weight infants. Pediatrics 2001;107:638–641.

11. Khalaf MN, Brodsky N, Hurley J, Bhandari V. A prospective

randomized, controlled trial comparing synchronized nasal

intermittent positive pressure ventilation versus nasal continuous

positive airway pressure as modes of extubation. Pediatrics 2001;

108:13–17.

Fig. 2. PES peak during NCPAP and NI-PSV for each of the 15

infants (^). The mean PES peak for all 15 infants (&) is plotted in

bold.

Fig. 3. Percent change in VE from NCPAP to NI-PSV in infants

with TcPCO2 <55 and �55 mmHg. Box plots depict median

values, 25th–75th percentiles, and 10th–90th percentiles.

Non-Invasive Pressure Support Ventilation in Preterm Infants 709

12. Friedlich P, Lecart C, Posen R, Ramicone E, Chan L, Ramanathan

R. A randomized trial of nasopharyngeal-synchronized inter-

mittent mandatory ventilation versus nasopharyngeal continuous

positive airway pressure in very low birth weight infants after

extubation. J Perinatol 1999;19:413–418.

13. Finer NN, Carlo WA, Duara S, Fanaroff AA, Donovan EF, Wright

LL, Kandefer S, Poole WK. Delivery room continuous positive

airway pressure/positive end-expiratory pressure in extremely

low birth weight infants: A feasibility trial. Pediatrics 2004;114:

651–657.

14. Tokioka H, Nagano O, Ohta Y, Hirakawa M. Pressure support

ventilation augments spontaneous breathing with improved

thoracoabdominal synchrony in neonates with congenital heart

disease. Anesth Analg 1997;85:789–793.

15. Osorio W, Claure N, D’Ugard C, Athavale K, Bancalari E. Effects

of pressure support during an acute reduction of synchronized

intermittent mandatory ventilation in preterm infants. J Perinatol

2005;25:412–416.

16. Wysocki M, Tric L, Wolff MA, Gertner J, Millet H, Herman B.

Noninvasive pressure support ventilation in patients with acute

respiratory failure. Chest 1993;103:907–913.

17. Appendini L, Patessio A, Zanaboni S, Carone M, Gukov B,

Donner CF, Rossi A. Physiologic effects of positive end-

expiratory pressure and mask pressure support during exacerba-

tions of chronic obstructive pulmonary disease. Am J Respir Crit

Care Med 1994;149:1069–1076.

18. Pasquina P, Merlani P, Granier JM, Ricou B. Continuous positive

airway pressure versus noninvasive pressure support ventilation to

treat atelectasis after cardiac surgery. Anesth Analg 2004;99:

1001–1008.

19. Fauroux B, Louis B, Hart N, Essouri S, Leroux K, Clement A,

Polkey MI, Lofaso F. The effect of back-up rate during non-

invasive ventilation in young patients with cystic fibrosis.

Intensive Care Med 2004;30:673–681.

20. Musante G, Schulze A, Gerhardt T, Everett R, Claure N, Schaller

P, Bancalari E. Proportional assist ventilation decreases thor-

acoabdominal asynchrony and chest wall distortion in preterm

infants. Pediatr Res 2001;49:175–180.

21. Gerhardt T, Bancalari E. Chestwall compliance in full-term

and premature infants. Acta Paediatr Scand 1980;69:359–364.

22. Moretti C, Gizzi C, Papoff P, Lampariello S, Capoferri M,

Calcagnini G, Bucci G. Comparing the effects of nasal

synchronized intermittent positive pressure ventilation (nSIPPV)

and nasal continuous positive airway pressure (nCPAP) after

extubation in very low birth weight infants. Early Hum Dev 1999;

56:167–177.

23. Aghai ZH, Saslow JG, Nakhla T, Milcarek B, Hart J, Lawrysh-

Plunkett R, Stahl G, Habib RH, Pyon KH. Synchronized nasal

intermittent positive pressure ventilation (SNIPPV) decreases

work of breathing (WOB) in premature infants with respiratory

distress syndrome (RDS) compared to nasal continuous positive

airway pressure (NCPAP). Pediatr Pulmonol 2006;41:875–881.

710 Ali et al.