Mechanical Ventilation of the Neonate: Should We … · target variable, pressure or volume, is...

Mechanical Ventilation of the Neonate:Should We Target Volume or Pressure?

Steven M Donn MD and Win Boon MD

IntroductionGas DeliveryLimitations of Volume VentilationVolume-Targeted ModalitiesHybrid Modalities Providing Volume-Targeted Ventilation

Volume-Guarantee VentilationPressure-Regulated Volume Control VentilationVolume-Assured Pressure SupportVolume-Support VentilationPressure Augmentation

The EvidenceVolume-Controlled VentilationVolume-Guarantee VentilationPressure-Regulated Volume Control

Meta-analysisWhy Does It Work?Summary

For more than 40 years conventional mechanical ventilation has been used for the treatment ofneonatal respiratory failure. Until relatively recently, this was accomplished with time-cycled pres-sure-limited ventilation, using intermittent mandatory ventilation. Earlier attempts at volume-targeted ventilation were largely ineffective because of technological limitations. The advent ofmicroprocessor-based devices gives the clinician an option to choose either target variable to treatneonatal patients. This paper reviews the principles of each and the accumulated evidence. Keywords: newborn, respiratory failure, mechanical ventilation, pressure-targeted, volume-targeted. [RespirCare 2009;54(9):1236–1243. © 2009 Daedalus Enterprises]

Introduction

The traditional method of mechanical ventilation to treatneonatal respiratory failure has been time-cycled pressure-limited ventilation. Originally this was accomplished bymodification of adult ventilators. Time-cycled pressure-limited ventilation was easy to use and was felt to safe-guard against barotrauma because the peak inspiratory pres-sure (PIP) could be limited and the ventilator would notexceed this pressure. Unfortunately, the delivered tidal vol-ume (VT) would fluctuate according to pulmonary com-pliance; a smaller volume of gas would be delivered at low

Steven M Donn MD and Win Boon MD are affiliated with the Divisionof Neonatal-Perinatal Medicine, Department of Pediatrics, CS Mott Chil-dren’s Hospital, University of Michigan Health System, Ann Arbor, Mich-igan.

Dr Donn has disclosed relationships with Discovery Laboratories, Dey,Cardinal Health, and Hamilton Medical. Dr Boon has disclosed no con-flicts of interest.

Dr Donn presented a version of this manuscript at the New HorizonsSymposium, “Neonatal Respiratory Care,” at the International Respira-tory Congress of the American Association for Respiratory Care, at the54th International Respiratory Congress of the American Association forRespiratory Care, held December 13-16, 2008, in Anaheim, California.

1236 RESPIRATORY CARE • SEPTEMBER 2009 VOL 54 NO 9

compliance, whereas a higher volume would be deliveredas compliance improved, unless the clinician vigilantlyadjusted the PIP.

In the mid-1970s an infant ventilator was specificallydesigned to deliver volume-controlled breaths to newborninfants. The clinician would target a volume of gas to bedelivered to the baby, and the pressure was permitted tofluctuate to deliver the desired VT. Although the principleseemed sound, there were severe technological limitations,including the inability to measure VT, compliant ventilatorcircuits that increased compressible volume loss, slow re-sponse times, incompatibility with continuous flow, and,most importantly, insufficient knowledge about neonatallung mechanics. In the early 1980s the manufacturerstopped producing the ventilator, and for the next decadealmost all neonatal ventilation was pressure-limited.

The incorporation of microprocessor-based technologyinto neonatal ventilators and the development of small,lightweight, and low-dead-space transducers and accuratereal-time monitoring enabled the reintroduction of vol-ume-targeted ventilation into neonatal intensive care.1 Thispaper will examine the current controversy as to whichtarget variable, pressure or volume, is better for the treat-ment of neonatal respiratory failure.

Gas Delivery

The major difference between pressure and volumebreaths is the way in which inspiratory gas flow is deliv-ered to the patient. In both time-cycled pressure-limitedventilation and pressure-control ventilation there is a rapid

acceleration of gas flow at the onset of inspiration, result-ing in rapid pressurization of the circuit and the achieve-ment of peak pressure and volume delivery early in inspi-ration. Flow then decelerates. This creates a peaked flowwaveform and rapidly rising and falling pressure wave-form (Fig. 1). Thus, breaths may be considered as being“front-end loaded.” Theoretically this might be advanta-geous when the primary pathophysiology is diffuse andcharacterized by the need for a high opening pressure. Incontrast, flow delivery during volume-controlled ventila-tion (VCV) accelerates at the start of inspiration but is heldconstant throughout inspiration, creating a square wave-form (Fig. 2). This results in a ramping effect, where peakpressure and peak volume delivery occur late in inspira-tion. These breaths are “back-end loaded,” resulting inslower inflation of the lung. VCV might be advantageouswhen lung disease is more heterogeneous and in situationswhere compliance changes quickly, such as following theadministration of surfactant.2

Limitations of Volume Ventilation

The major drawback to the use of volume ventilation innewborns is related to the use of uncuffed endotrachealtubes. This results in a variable degree of leak around theendotracheal tube, which affects the volume of gas thebaby actually receives. Ventilators are set to deliver atargeted VT, but this is based on how much gas leaves themachine. A certain amount of this gas will be compressedwithin the ventilator, and this is referred to as compress-ible volume loss. This is affected by the compliance ofboth the lungs and the circuit, in addition to other factors,such as humidification. Because of compressible volumeloss, it is imperative to measure the delivered VT as closeto the airway as possible.3

Correspondence: Steven M Donn MD, Division of Neonatal-PerinatalMedicine, Department of Pediatrics, F5790, Mott Children’s’ Hospital,University of Michigan Health System, 1500 E Medical Center Drive,Ann Arbor MI 48109-0254. E-mail: [email protected].

Fig. 1. Graphic representation of flow waveform (middle panel) inpressure-control ventilation. Note the rapid acceleration and de-celeration of inspiratory flow, producing a peaked waveform.

Fig. 2. Graphic representation of flow waveform (middle panel) involume-targeted ventilation. Note the constant inspiratory flow,producing a square waveform.

MECHANICAL VENTILATION OF THE NEONATE: SHOULD WE TARGET VOLUME OR PRESSURE?

RESPIRATORY CARE • SEPTEMBER 2009 VOL 54 NO 9 1237

Volume-Targeted Modalities

Volume-targeted ventilation has been achieved in sev-eral ways. True volume-targeting involves all of the fea-tures described earlier, including constant flow delivery(square waveform) and auto-adjustment of pressure as com-pliance changes (Fig. 3). There are several hybrid forms ofvolume-targeting, where pressure adjustments are made toassure the delivery of a minimal VT. One might also arguethat volume-targeting of time-cycled pressure-limited ven-tilation is possible if the clinician is vigilant and constantlyattends to the ventilator so that delivered VT stays withina tight range. However, delivered breaths will still havethe characteristics of pressure-targeted breaths, and it seemsdoubtful that even the most astute clinician could makeadjustments as fast as a microprocessor.

Hybrid Modalities ProvidingVolume-Targeted Ventilation

Both VCV and time-cycled pressure-limited ventilationhave certain advantages and disadvantages. Hybrid mo-dalities try to combine the most desirable features of each.These include volume guarantee, pressure-regulated vol-ume control, and volume-assured pressure support. Theyare essentially pressure-targeted forms of ventilation, bututilize microprocessor servo-controlled ventilation with analgorithm that adjusts the rise and fall of pressure for VT

delivery within a desired range. Volume guarantee andpressure-regulated volume control use the VT of previousbreaths as a reference, with follow-up adjustments in PIPon averages of 4–6 breaths. Volume-assured pressure sup-port makes intra-breath adjustments of inspiratory timeand/or pressure, until the desired volume has been deliv-ered. All of these try to optimize VT delivery, althougheach does so differently.

Volume-Guarantee Ventilation

Volume-guarantee ventilation, available on the Babylog8000-plus ventilator (Drager Medical, Telford, Pennsylva-nia), is a popular form of volume targeting. It is bestdescribed as a dual-loop synchronized modality that ven-tilates with time-cycled pressure-limited ventilation butallows pressure to be adjusted to deliver a VT in a clini-cian-chosen range. The clinician selects a target VT andlimits the pressure (referred to as the working pressure).The device measures the exhaled VT of the previous breathas a reference and adjusts the working pressure to reachthe target volume over the next few breaths. The feedbackloop has some limitations. For example, because adjust-ments to PIP are based on the exhaled VT and are made insmall increments to avoid overcompensation, the deliveredVT may not compensate for substantial breath-to-breathfluctuations in the presence of large leaks. When the leak

exceeds 40%, volume guarantee is less reliable because ofthe inability to accurately measure the real VT. Catch-upadjustments in pressure occur every few breaths, so if theventilatory rate is set too low, its effectiveness may bequestionable.

Potential advantages of volume guarantee include de-creased volutrauma because the clinician sets a VT that isnot exceeded, and as lung compliance improves, pressureis automatically decreased.4 Volume guarantee can be usedonly with patient-triggered modes: assist-control, synchro-nized intermittent mandatory ventilation (SIMV), or pres-sure-support ventilation (PSV). The addition of volumeguarantee to one of these modes enables the clinician to seta mean VT to be delivered, as well as the standard venti-lator settings of PIP, positive end-expiratory pressure, in-spiratory time, and respiratory rate. No more than 130% ofthe target VT is supposed to be delivered. The usual initialtarget is 4–5 mL/kg. The pressure limit is set approxi-mately 15% to 20% above the peak pressure needed toconsistently deliver the desired VT. Because the adjust-ment of PIP is in reference to exhaled VT and changes aremade in small increments to prevent overcompensation,the PIP cannot be adjusted instantaneously to compensatefor large breath-to-breath fluctuations. Consequently, al-though the delivered VT is more consistent with volumeguarantee than without it, there is fluctuation around thetarget value. Most infants can be extubated when theymaintain VT at or above the target value at a PIP � 10–12 cm H2O with reliable respiratory effort.

Pressure-Regulated Volume Control Ventilation

Pressure-regulated volume control is another modalitythat attempts to combine the benefits of time-cycled pres-sure-limited ventilation and VCV. It is available on theServo 300A and the Servo-i ventilators (Maquet, Bridge-water, New Jersey). It is flow-cycled and offers the “vari-able” flow rate of pressure-control ventilation and volume

Fig. 3. Schematic comparison of pressure-volume loops in pres-sure-targeted (left) and volume-targeted (right) ventilation. Duringpressure-targeting a decrease in pulmonary compliance (movingfrom light loop to dark loop) results in a loss of tidal volume,despite consistency of inspiratory pressure. During volume-target-ing, a decrease in compliance (moving from the dark loop to thelight loop) results in an automatic increase in pressure to assureconstant volume delivery.



targeting. Like volume guarantee, pressure-regulated vol-ume control is a form of closed-loop ventilation; pressureis adjusted according to the delivered VT. The new Servo-iventilator measures pressure at the proximal airway, as-suring better accuracy.5

The clinician selects a target VT and the maximum al-lowable pressure. The microprocessor attempts to use thelowest pressure (with a decelerating-flow waveform) todeliver the set VT. The first breath is delivered 10 cm H2Oabove the baseline setting (positive end-expiratory pres-sure) and is utilized as a test breath to calculate the pres-sure needed to deliver the desired VT in accordance withthe patient’s compliance. The next 3 breaths are deliveredat a pressure that is 75% of the calculated necessary pres-sure. If the targeted VT fails to be delivered, the inspiratorypressure is increased by 3 cm H2O for each breath until thetargeted volume is reached. If the targeted VT is exceeded,the converse occurs, and pressure is decreased by 3 cm H2O.Inspiratory pressure is regulated between the positive end-expiratory pressure and 5 cm H2O below the clinician-limited PIP. Variations in delivered volume also occurbecause pressure adjustments are based on a 4-breath mov-ing average.

Volume-Assured Pressure Support

Volume-assured pressure support is available on theVIP Bird Gold infant/pediatric ventilator (CareFusion, SanDiego, California). This modality combines the advan-tages of pressure-targeted and volume-targeted ventilationwithin a single breath. It can be used with both assist-control and SIMV, or by itself in babies with reliablerespiratory drive. It may be described as “variable-flowvolume ventilation.” This is a true hybrid modality, withdecelerating, variable inspiratory flow, and guaranteed vol-ume delivery. The breath is patient-triggered and begins asa PSV breath. The device measures delivered volume wheninspiratory flow has decelerated to the set minimum. If thedelivered volume exceeds the desired level, the breath isterminated by flow-cycling. However, if the preset volumehas not been delivered, the breath transitions to a volume-targeted breath; the set flow will persist and the inspiratorytime will be prolonged until the desired volume has beenreached. In this situation the sinusoidal flow waveformtransitions to a square waveform. If the delivered volumeis considerably below the desired level, the pressure mayalso increase slightly.6

Volume-Support Ventilation

Volume-support ventilation is another hybrid modality,available on the Servo-i device (Maquet, Bridgewater, NewJersey), which is similar to PSV and pressure-regulatedvolume control.5 The volume-supported breath is patient-triggered, pressure-limited, and flow-cycled. It is used in

spontaneously breathing patients with adequate respiratorydrive. Breath rate, VT, and minute volume are chosen bythe clinician; inspiratory time is determined by the patient.The algorithm adjusts the pressure limit by no more than3 cm H2O at a time. Adjustments are made in consecutivebreaths until the targeted VT is reached. The flow, pres-sure, and volume waveforms for a volume-supported breathare similar to those of pressure-supported breaths; how-ever, the efficacy of volume support can be determinedonly by evaluating sequential graphic presentations. TheVT waveform will increase in a stepwise fashion until thetarget volume is reached.

Pressure Augmentation

Pressure augmentation, available on the Bear 1000 ven-tilator (CareFusion, San Diego, California), is another hy-brid modality that matches the patient’s flow demand whileguaranteeing a minimum VT.2 Pressure augmentation isunique in the following ways: the pre-set VT is only aminimum, and the patient can go above this; minimum VT

is guaranteed by adjustment in flow rather than pressure,which is fixed; and adjustment to flow is made within asingle breath, like volume-assured pressure support. Pres-sure augmentation matches the patient’s flow demand andlung dynamics and can be delivered in either assist-controlor SIMV modes.

The Evidence

Table 1 summarizes studies on volume-targeted and pres-sure-limited ventilation.

Volume-Controlled Ventilation

One of the first trials evaluating VCV was conducted bySinha and colleagues.7 Fifty pre-term infants with respi-ratory distress syndrome (RDS) and birth weights of atleast 1,200 g requiring mechanical ventilation and surfac-tant therapy were randomized to receive either VCV ortime-cycled pressure-limited ventilation. VT was tightlymaintained between 5–8 mL/kg for both treatment groups.Primary outcome measures included the time from studyentry until achievement of either an alveolar-arterial oxy-gen difference less than 100 mm Hg, or a mean airwaypressure less than 8 cm H2O for at least 12 hours. Infantsreceiving VCV reached success criteria faster than thetime-cycled pressure-limited ventilation group, with a meantime of 65.5 hours versus 125.6 hours (P � .001). Thetotal duration of ventilation for the VCV group was loweras well (122.4 h vs 161.9 h, P � .001). Secondary out-comes included the incidence of intraventricular hemor-rhage, periventricular leukomalacia, patent ductus arterio-sus, and bronchopulmonary dysplasia (BPD, defined asoxygen requirement beyond 36 weeks post-menstrual age,



Tab

le1.

Sum

mar

yof

Key

Tri

als

Eva

luat

ing

Vol

ume-

Tar

gete

dV

ersu

sPr

essu

re-L

imite

dV

entil

atio

n

Firs

tA

utho

rY

ear

Ran

dom

izat

ion

Part

icip

ants

Inte

rven

tion

Out

com

eM

easu

res

Che

ema1

020

01R

ando

miz

eddo

uble

cros

sove

rtr

ial

40in

fant

s�

34w

kge

stat

ion

requ

irin

gM

Vfo

rR

DS.

Exc

lusi

ons:

Infa

nts

requ

irin

gm

uscl

ere

laxa

nts

orw

ithle

thal

cong

enita

lan

omal

ies

Sync

hron

ized

inte

rmitt

ent

posi

tive-

pres

sure

vent

ilatio

nal

one

vsth

esa

me

mod

epl

usvo

lum

egu

aran

tee,

orSI

MV

alon

evs

SIM

Vpl

usvo

lum

egu

aran

tee

Prim

ary:

Peak

airw

aypr

essu

reSe

cond

ary:

Mea

nai

rway

pres

sure

,ex

pire

dV

T,

min

ute

volu

me,

F IO

2,an

dtr

ansc

utan

eous

CO

2an

dO

2pr

essu

re

D’A

ngio

2220

05R

ando

miz

edco

ntro

lled

tria

l21

3in

fant

sw

hore

quir

edM

Vw

ere

atle

ast

24w

kge

stat

iona

lag

e,an

dw

eigh

ed50

0–1,

249

gat

birt

h

Pres

sure

-reg

ulat

edvo

lum

eco

ntro

lvs

time-

cycl

edpr

essu

re-l

imite

dSI

MV

Prim

ary:

Prop

ortio

nof

infa

nts

who

wer

eal

ive

and

extu

bate

dat

14d

ofag

eSe

cond

ary:

Prop

ortio

nof

infa

nts

who

wer

eal

ive

and

extu

bate

dat

28d

ofag

eor

36w

kpo

st-c

once

ptua

lag

e,ag

eat

fina

lex

tuba

tion,

mor

talit

y,fa

ilure

ofve

ntila

tory

mod

e,in

cide

nce

ofB

PD,

air

leak

s,pu

lmon

ary

hem

orrh

age,

PDA

,IV

H,

peri

vent

ricu

lar

leuk

omal

acia

,ne

crot

izin

gen

tero

colit

is,

retin

opat

hyof

prem

atur

ity,

hom

eox

ygen

use,

and

num

erou

sot

her

vari

able

sH

erre

ra11

2002

Ran

dom

ized

cros

sove

rtr

ial

17in

fant

s,60

0–1,

200

g,w

ithre

spir

ator

yfa

ilure

.E

xclu

sion

s:Se

vere

cong

enita

lan

omal

ies,

peri

nata

las

phyx

ia,

seps

is,

sym

ptom

atic

PDA

,gr

ade

3–4

IVH

,se

datio

n,an

dcl

inic

alin

stab

ility

asde

fine

dby

atte

ndin

gne

onat

olog

ist

Firs

t9

infa

nts:

SIM

Val

one

vsSI

MV

plus

volu

me

guar

ante

e(4

.5m

L/k

g)N

ext

8in

fant

s:SI

MV

alon

evs

SIM

Vpl

usvo

lum

egu

aran

tee

(4.5

mL

/kg)

vsSI

MV

plus

volu

me

guar

ante

e(3

mL

/kg)

Peak

insp

irat

ory

pres

sure

,m

ean

airw

aypr

essu

re,

num

ber

ofve

ntila

tor-

gene

rate

dan

dpa

tient

-gen

erat

edbr

eath

s,V

T,

F IO

2,S p

O2,

and

tran

scut

aneo

usO

2pr

essu

re

Kes

zler

1220

04R

ando

miz

edco

ntro

lled

tria

l18

infa

nts

�34

wk

gest

atio

n,w

ithR

DS,

onM

VE

xclu

sion

s:Pa

tient

sw

ithco

ngen

ital

anom

alie

s,re

ceiv

ing

neur

omus

cula

rpa

raly

sis

orna

rcot

icag

ents

,or

with

�30

%en

dotr

ache

altu

bele

ak

Ass

ist-

cont

rol

only

vsas

sist

-con

trol

plus

volu

me

guar

ante

ePr

imar

y:Pe

rcen

tage

oftim

eth

atV

Tan

dP a

CO

2w

ere

outs

ide

targ

etra

nge

(4–6

mL

/kg,

and

35–4

5m

mH

g,re

spec

tivel

y)

Lis

ta16

2004

Ran

dom

ized

cont

rolle

dtr

ial

stra

tifie

dby

trea

tmen

tce

nter

and

gest

atio

nal

age

(25–

28w

k,an

d29

–32

wk)

53in

fant

s,25

–32

wk

gest

atio

nal

age,

onm

echa

nica

lve

ntila

tion

for

seve

reR

DS

Exc

lusi

ons:

Let

hal

anom

alie

s,us

eof

para

lytic

agen

ts,

grad

e3–

4IV

H,

seps

is,

orsu

spec

ted

infe

ctio

n

Pres

sure

-sup

port

vent

ilatio

nal

one

vspr

essu

re-s

uppo

rtve

ntila

tion

plus

volu

me

guar

ante

e(5

mL

/kg)

Prim

ary:

Con

cent

ratio

nsof

inte

rleu

kin-

6,in

terl

euki

n-8,

and

tum

orne

cros

isfa

ctor

�in

trac

heal

aspi

rate

son

days

oflif

e1,

3,an

d7.

Seco

ndar

y:D

urat

ion

ofve

ntila

tion,

airw

aypr

essu

re,

inci

denc

ean

dra

teof

trea

tmen

tfo

rPD

A,

num

ber

ofsu

rfac

tant

dose

s,in

cide

nce

ofai

rle

aks,

IVH

,pe

rive

ntri

cula

rle

ukom

alac

ia,

retin

opat

hyof

prem

atur

ity,

oxyg

ende

pend

ency

at28

dan

d/or

36w

kpo

st-

conc

eptu

alag

e,an

dsu

rviv

alM

cCal

lion2

320

05M

eta-

anal

ysis

ofth

efo

llow

ing

4tr

ials

:K

eszl

eret

al,

2004

Lis

taet

al,

2004

Piot

row

ski

etal

,19

97Si

nha

etal

,19

97

178

infa

nts

�37

wk

gest

atio

nal

age

Exc

lusi

ons:

Let

hal

cong

enita

lan

omal

ies,

mus

cle

rela

xatio

n,su

spec

ted

seps

is,

lack

ofar

teri

alac

cess

,na

rcot

icus

e,en

dotr

ache

altu

bele

aks

�30

%,

seve

reIV

H,

asph

yxia

,pn

eum

otho

rax,

and

mec

oniu

mas

pira

tion

synd

rom

e

Vol

ume-

targ

eted

vspr

essu

re-l

imite

dve

ntila

tion

Prim

ary:

Hos

pita

lm

orta

lity,

and

deat

hor

need

for

supp

lem

enta

lox

ygen

atei

ther

28d

oflif

eor

36w

kpo

st-c

once

ptua

lag

e.Se

cond

ary:

Failu

reof

vent

ilato

rym

ode

orne

edfo

rne

wus

eof

mus

cle

rela

xant

s,du

ratio

nof

resp

irat

ory

supp

ort,

adve

rse

bloo

dga

sva

lues

,PD

A,

air

leak

s,gr

owth

,IV

H,

peri

vent

ricu

lar

leuk

omal

acia

,ne

urod

evel

opm

enta

lou

tcom

e,ne

edfo

rsu

pple

men

tal

oxyg

enat

eith

er28

dof

life

or36

wk

post

-con

cept

ual

age

amon

gsu

rviv

ors,

and

impa

ctof

mod

eof

volu

me-

targ

eted

vent

ilatio

nPi

otro

wsk

i21

1997

Ran

dom

ized

cont

rolle

dtr

ial

60in

fant

sw

ithR

DS

orco

ngen

ital

pneu

mon

ia,

requ

irin

gM

V,

and

wei

ghin

g�

2,50

0g

Exc

lusi

ons:

Ter

min

alst

ate

ofin

fant

atad

mis

sion

,ai

rle

aks,

cong

enita

lan

omal

ies,

seps

is,

and

mec

oniu

mas

pira

tion

Pres

sure

-reg

ulat

edvo

lum

eco

ntro

lvs

time-

cycl

edpr

essu

re-l

imite

din

term

itten

tm

anda

tory

vent

ilatio

n

Prim

ary:

Dur

atio

nof

vent

ilatio

nan

din

cide

nce

ofB

PD.

Seco

ndar

y:In

cide

nce

ofai

rle

aks,

IVH

,hy

pote

nsio

n,ne

crot

izin

gen

tero

colit

is,

PDA

,an

dne

edfo

rse

datio

n

Sing

h820

06R

ando

miz

edco

ntro

lled

tria

lA

prio

rist

ratif

icat

ion

into

2gr

oups

acco

rdin

gto

birt

hw

eigh

t:60

0–1,

000

g,an

d1,

001–

1,50

0g

109

infa

nts,

600–

1,50

0g,

gest

atio

nal

age

24–3

1w

k,re

quir

ing

MV

and

surf

acta

ntth

erap

yE

xclu

sion

s:Se

vere

cong

enita

lm

alfo

rmat

ions

Vol

ume-

cont

rolle

dve

ntila

tion

vstim

e-cy

cled

pres

sure

-lim

ited

vent

ilatio

n.V

Tm

aint

aine

dat

4–6

mL

/kg

inbo

thgr

oups

Prim

ary:

Tim

efr

omst

udy

entr

yun

tilac

hiev

emen

tof

eith

eran

alve

olar

-art

eria

lox

ygen

diff

eren

ce�

100

mm

Hg

ora

mea

nai

rway

pres

sure

�8

cmH

2Ofo

rat

leas

t12

h.Se

cond

ary:

Dur

atio

nof

vent

ilatio

nor

resp

irat

ory

supp

ort,

surv

ival

todi

scha

rge,

inci

denc

eof

chro

nic

lung

dise

ase,

IVH

,pe

rive

ntri

cula

rle

ukom

alac

ia,

PDA

,or

necr

otiz

ing

ente

roco

litis

Sing

h920

09L

ong-

term

outc

omes

from

prio

rra

ndom

ized

cont

rolle

dtr

ial

(Sin

ghet

al,

2006

)

90of

the

109

infa

nts

inth

e20

06st

udy

wer

efo

llow

ed.

Med

ian

corr

ecte

dag

eat

follo

w-u

pw

as22

mon

ths

Vol

ume-

cont

rolle

dve

ntila

tion

vstim

e-cy

cled

pres

sure

-lim

ited

vent

ilatio

n.Pa

tient

spr

ospe

ctiv

ely

follo

wed

with

med

ical

asse

ssm

ents

and

pare

ntal

inte

rvie

ws

via

ast

ruct

ured

ques

tionn

aire

Mor

talit

y,re

-adm

issi

onra

te,

pulm

onar

you

tcom

es(i

nclu

ding

the

freq

uenc

yof

coug

hor

whe

eze,

and

use

ofpu

lmon

ary

med

icat

ions

),an

dgr

oss

neur

odev

elop

men

tal

outc

ome

Sinh

a719

97R

ando

miz

edco

ntro

lled

tria

l50

pre-

term

infa

nts

with

RD

San

dbi

rth

wei

ght

ofat

leas

t1,

200

g,re

quir

ing

MV

and

surf

acta

ntth

erap

yE

xclu

sion

s:Pn

eum

onia

,se

psis

,co

ngen

ital

mal

form

atio

ns,

lack

ofar

teri

alac

cess

Vol

ume-

cont

rolle

dve

ntila

tion

vstim

e-cy

cled

pres

sure

-lim

ited

vent

ilatio

n.V

Tm

aint

aine

dat

5–8

mL

/kg

inbo

thgr

oups

Prim

ary:

Tim

efr

omst

udy

entr

yun

tilac

hiev

emen

tof

eith

eran

alve

olar

-art

eria

lox

ygen

diff

eren

ce�

100

mm

Hg

ora

mea

nai

rway

pres

sure

�8

cmH

2Ofo

rat

leas

t12

h.Se

cond

ary:

Inci

denc

eof

IVH

,pe

rive

ntri

cula

rle

ukom

alac

ia,

PDA

,or

BPD

at36

wk

post

-con

cept

ual

age

MV

�m

echa

nica

lve

ntila

tion

RD

S�

resp

irat

ory

dist

ress

synd

rom

eSI

MV

�sy

nchr

oniz

edin

term

itten

tm

anda

tory

vent

ilatio

n

VT

�tid

alvo

lum

eF I

O2

�fr

actio

nof

insp

ired

oxyg

enB

PD�

bron

chop

ulm

onar

ydy

spla

sia

PDA

�pa

tent

duct

usar

teri

osus

IVH

�in

trav

entr

icul

arhe

mor

rhag

eS p

O2

�ox

ygen

satu

ratio

nm

easu

red

via

puls

eox

imet

ry



with correlating radiographic findings). The frequency ofmajor scan abnormality (large intraventricular hemorrhage,ventriculomegaly, and intraparenchymal echodensities)was lower in the VCV group than in the time-cycled pres-sure-limited ventilation group (P � .5). While there wasless BPD in the VCV group (4% vs 25%), this did notreach statistical significance. There were no significantdifferences in the incidence of air leaks or patent ductusarteriosus requiring therapy. The authors concluded thatVCV appeared to be both safe and effective in this groupof patients. At the time of this study, technological limi-tations of the ventilator precluded enrolling smaller infants.

A subsequent study by the same group further bolsteredthe efficacy of VCV.8 Refinements in ventilator designeventually led to the ability to deliver even smaller VT,therefore providing VCV as an option to smaller babiesthan before. The second trial enrolled 109 infants, weigh-ing 600–1,500 g, with gestational ages of 24–31 weeks.These babies also had RDS requiring mechanical ventila-tion and surfactant therapy, and were randomized to re-ceive either VCV or time-cycled pressure-limited ventila-tion. A priori stratification into 2 groups according to birthweight was performed (600–1,000 g, and 1,001–1,500 g).Target VT was 4–6 mL/kg for both groups. Primary out-comes evaluating the alveolar-arterial oxygen differenceor mean airway pressure were identical to their prior study.There were no statistically significant differences in pri-mary outcomes between the 2 groups in aggregate, al-though there was a trend toward faster weaning from theventilator in the VCV group. However, in the cohort ofinfants weighing 600–1,000 g, those assigned to VCVreached their primary end point faster than those assignedto time-cycled pressure-limited ventilation, with a meantime of 21 hours versus 58 hours (P � .03). No statisti-cally significant differences were found among the sec-ondary outcomes, which included duration of ventilationand respiratory support, survival to neonatal intensive careunit discharge, and frequency of other complications, suchas intraventricular hemorrhage, periventricular leukomala-cia, patent ductus arteriosus, and necrotizing enterocolitisBell stage II or greater. The authors concluded that VCVis safe and efficacious in very-low-birth-weight infantsand may have advantages over time-cycled pressure-lim-ited ventilation, especially in smaller infants.

Singh and colleagues later reported the long-term out-comes of this study population.9 Ninety of the 109 infantsenrolled in the initial study were prospectively followedusing medical assessments and parental interviews via astructured questionnaire, to obtain masked evaluation ofhealth status and gross neurodevelopment. The mediancorrected age at follow-up was 22 months. Fewer childrenin the VCV group required inhaled medications at the timeof follow-up (odds ratio of 0.3, with a 95% confidenceinterval [CI] of 0.09–0.98, P � .04). Mortality, readmis-

sion rate, and frequency or severity of respiratory dysfunc-tion were not affected by the type of ventilation received.The authors noted that the trend toward less chronic lungdisease at 36 weeks post-conceptual age has a clinicalcorrelate at 22 months, with fewer children requiring in-haled medications for chronic lung disease.

Volume-Guarantee Ventilation

Cheema and Ahluwalia were among the first to evaluatethe efficacy of volume-guarantee ventilation in the neona-tal population.10 In their crossover trial, 40 infants� 34 weeks gestation requiring mechanical ventilation forRDS received synchronized intermittent positive-pressureventilation alone, versus synchronized intermittent posi-tive-pressure ventilation plus volume guarantee, or SIMValone versus SIMV plus volume guarantee. Patients whoreceived synchronized intermittent positive-pressure ven-tilation were in the acute phase of RDS, while those whoreceived SIMV were in the recovery (weaning) phase.There were slight decreases in PIP and mean airway pres-sure in both volume-guarantee groups, which reached sta-tistical significance. There were no significant differencesin overall gas exchange or adverse outcomes between thosewith and without volume guarantee. While this was a smallshort-term study, it did suggest that volume-targeted mo-dalities are feasible for the neonatal population.

In another crossover trial, Herrera and colleagues notedthat SIMV plus volume guarantee, compared to SIMValone, had a statistically significant reduction in peak andmean airway pressure.11

Keszler and Abubakar performed the first randomizedcontrolled trial of volume guarantee.12 Eighteen infants� 34 weeks gestation, who had RDS, were randomized toreceive assist-control alone, or assist-control plus volumeguarantee. Patients who received assist-control plus vol-ume guarantee had fewer breaths outside the targeted VT

range, compared to assist-control alone (37% vs 61%,P � .001). The volume-guarantee group also had signifi-cantly fewer episodes of inadvertent hypocapnia, furthersupporting the potential for volume guarantee.

A more recent trial by Cheema and colleagues evaluatedinitial arterial blood gases in infants with RDS shortlyafter they were placed on synchronized intermittent posi-tive-pressure ventilation versus synchronized intermittentpositive-pressure ventilation plus volume guarantee.13

Stratified analysis revealed that infants � 25 weeks gesta-tion had significantly less hypocarbia when volume guaran-tee was used (27% vs 61%, P � .05).

Further studies have addressed the efficacy of the var-ious volume-guarantee modes. Fully supported breaths ap-pear to be more effective, as demonstrated by Abubakarand Keszler’s crossover trial of assist-control plus volumeguarantee versus SIMV plus volume guarantee.14 Another



crossover trial by Scopesi and colleagues suggested that inthe weaning phase, volume guarantee in modes that sup-port every breath may be more effective than a combina-tion of SIMV plus volume guarantee.15

The protective effects of volume guarantee are high-lighted by the trial done by Lista et al, in which 53 pre-mature infants with RDS were randomized to receive ei-ther PSV alone or PSV plus volume guarantee.16 A VT of5 mL/kg was targeted for all study patients. Tracheal as-pirates were obtained on days of life 1, 3, and 7, andpro-inflammatory cytokines interleukin 6 (IL-6), IL-8, andtumor necrosis factor alpha were measured. Statisticallysignificant elevations in IL-6 and IL-8 were found in pa-tients who received only PSV, compared to PSV plus vol-ume guarantee. The volume-guarantee group also had lowerairway pressures and decreased duration of ventilation,though the latter did not reach statistical significance. Nev-ertheless, this study suggests that volume-targeted venti-lation may reduce ventilator-induced lung injury.

A subsequent study by Lista and colleagues examinedthe inflammatory effects of ventilation with variable vol-umes.17 The targeted volume of 5 mL/kg used in theirearlier study was compared to a smaller 3.5 mL/kg vol-ume. Interestingly, the lower volume had a statisticallysignificant increase in cytokine levels. Duration of venti-lation was also prolonged for the low-VT group (9.2 d vs16.8 d, P � .05). It was hypothesized that these findingsresulted from atelectrauma.

Lista et al most recently compared volume guarantee(5 mL/kg) with high-frequency oscillatory ventilation.18

This analysis revealed higher pro-inflammatory cytokineconcentrations and longer oxygen dependence in patientsreceiving high-frequency oscillatory ventilation, comparedto those receiving assist-control plus volume guarantee.These findings raise more questions regarding optimal mo-dalities and what constitutes the ideal VT.

Clinical studies of conventional ventilation have thusfar used target VT of 3–6 mL/kg.19 However, it is unlikelythat “one size fits all,” and more research is warranted inthis area.20

Pressure-Regulated Volume Control

The evidence for pressure-regulated volume control ven-tilation is quite sparse. Piotrowski and colleagues com-pared pressure-regulated volume control to time-cycledpressure-limited intermittent mandatory ventilation in arandomized controlled trial.21 Sixty patients with RDS orcongenital pneumonia requiring ventilatory support, weigh-ing � 2,500 g, were enrolled in this study. Primary out-comes included the duration of ventilation and the inci-dence of BPD. No statistically significant differences werefound between the 2 groups; however, subgroup analysisof infants � 1,000 g revealed a significant decrease in both

the duration of ventilation and hypotension in the pres-sure-regulated volume-control group compared to the time-cycled pressure-limited ventilation group (P � .05). Ad-ditional outcomes included a statistically significantdecrease in the incidence of severe (grade 3–4) intraven-tricular hemorrhage in the pressure-regulated volume-con-trol group (11% vs 35%, P � .05). Air leaks were also lessfrequent in the pressure-regulated volume-control group,although this did not reach statistical significance.

D’Angio and colleagues compared pressure-regulatedvolume control to time-cycled pressure-limited SIMV.22

They studied 213 infants who required mechanical venti-lation and were at least 24 weeks gestational age andweighed 500–1,249 g at birth. Primary outcome analysisrevealed no statistically significant differences betweenthe 2 groups for the number of infants who were alive andextubated at 14 and 28 days of life, and at 36 weekspost-menstrual age. There was no difference in the inci-dence of BPD at 36 weeks post-menstrual age. The authorsnoted that infants in the pressure-regulated volume-controlgroup were less likely to fail their assigned modality thanthose in the SIMV group (relative risk 0.61, 95% CI 0.38–0.97). The authors concluded that pressure-regulated vol-ume control did not provide any clear benefits over time-cycled pressure-limited SIMV.

Meta-analysis

There has been one meta-analysis comparing volume-targeted to pressure-limited ventilation in the neonate thusfar.23 Four of the trials discussed above, consisting of 178patients, met the inclusion criteria for this meta-analy-sis.7,12,16,22 The 2 primary outcome measures were hospitalmortality, and death or need for supplemental oxygen ateither 28 days of life or 36 weeks post-conceptual age.There were no statistically significant differences noted inthe first outcome measure. As for secondary outcomes,there was a significant reduction in the mean number ofdays of intermittent positive-pressure ventilation in thevolume-targeted group (weighted mean difference �2.93,with a 95% CI of �4.28 to �1.57). This volume groupalso had a lower incidence of pneumothorax (relative riskof 0.23, with a 95% CI of 0.07 to 0.76). Grade 3–4 intra-ventricular hemorrhage was less common in the volume-targeted group (relative risk of 0.32, 95% CI of 0.11 to0.9). Furthermore, there was a trend toward decreased BPDin the volume-targeted group that approached—but didnot reach—statistical significance (relative risk 0.34,95% CI of 0.11 to 1.05). The authors concluded that sta-tistically significant effects favoring volume-targeting wereshown, but further trials are still needed. When the studyby Singh et al8 is added to these trials, there is a statisti-cally significant reduction in the duration of ventilationand the incidence of pneumothorax, and the relative risk of



BPD is 0.7 (95% CI 0.5 to 1.05) (personal communication,Peter G Davis, Department of Paediatrics, Royal Women’sHospital, Parkville, Australia).

Why Does It Work?

The revival of volume-targeted ventilation in the treat-ment of neonatal respiratory failure and its early successraises questions about mechanisms of efficacy. While it isreasonable to believe that faster reduction in pressure mightobviate both barotrauma and volutrauma, other factors areno doubt involved.

A recent study by Swamy and colleagues24 examinedthe patients in the Singh et al trial.8 They looked at dif-ferences in VT delivery in babies assigned to either VCVor time-cycled pressure-limited ventilation. Although themedian VT was quite similar, there was a statistically sig-nificant reduction in the variability of VT delivery for ba-bies receiving VCV. This could explain why they wereable to be weaned faster and perhaps why there was lesslung injury.

In a study of adult patients, Kallet and colleagues25

showed differences in the work of breathing among pa-tients randomized to VCV (1.09 � 0.59 J/L), pressure-control ventilation (1.27 � 0.58 J/L), and pressure-regu-lated volume control (1.35 � 0.60 J/L). While comparabledata do not exist yet for neonatal patients, it does suggestanother mechanism of benefit.

Summary

In summary, volume-targeted ventilation of newbornswith respiratory failure leads to small but clinically im-portant improvements in short-term pulmonary outcomes,including BPD. Further investigation is warranted to bettercomprehend and refine the technique and to determine theoptimal clinical applications. These should also include anassessment of clinically relevant long-term outcomes.

REFERENCES

1. Bandy KP, Nicks JJ, Donn SM. Volume-controlled ventilation forsevere neonatal respiratory failure. Neonatal Intensive Care 1992;5(3):70-73.

2. Sinha SK, Donn SM. Volume-controlled ventilation. In: GolsmithJP, Karotkin EH, editors. Assisted ventilation of the neonate, 4thedition. Philadelphia: WB Saunders; 2003:171-182.

3. Donn SM, Bandy KP. Volume-controlled ventilation. In: Donn SM,Sinha SK, editors. Manual of neonatal respiratory care, 2nd edition.Philadelphia: Mosby-Elsevier; 2006:206-209.

4. Keszler M, Abubakar KM. Volume guarantee ventilation. Clin Peri-natol 2007;34(3):107-116.

5. Bushell MK. Servo-i ventilator. In: Donn SM, Sinha SK, editors.Manual of neonatal respiratory care, 2nd edition. Philadelphia: Mosby-Elsevier; 2006:273-278.

6. Becker MA, Donn SM. Bird VIP Gold ventilator. In: Donn SM,Sinha SK, editors. Manual of neonatal respiratory care, 2nd edition.Philadelphia: Mosby-Elsevier; 2006:249-255.

7. Sinha SK, Donn SM, Gavey J, McCarty M. Randomised trial ofvolume controlled versus time cycled, pressure limited ventilation in

preterm infants with respiratory distress syndrome. Arch Dis Child1997;77(3):F202-F205.

8. Singh J, Sinha SK, Clarke P, Byrne S, Donn SM. Mechanical ven-tilation of very low birth weight infants: is volume or pressure abetter target variable? J Pediatr 2006;149(3):308-313.

9. Singh J, Sinha S, Alsop E, Gupta S, Misra A, Donn S; Heart of EnglandNHS Foundation Trust. Long term follow up of VLBW infants from aneonatal volume versus pressure mechanical ventilation trial. Arch DisChild Fetal Neonatal Ed, 2009. Mar 25 [Epub ahead of print].

10. Cheema IU, Ahluwalia JS. Feasibility of tidal volume-guided ven-tilation in newborn infants: a randomized, crossover trial using thevolume guarantee modality. Pediatrics 2001;107(6):1323-1328.

11. Herrera CM, Gerhardt T, Claure N, Everett R, Musante G, ThomasC, Bancalari E. Effects of volume-guaranteed synchronized intermit-tent mandatory ventilation in preterm infants recovering from respi-ratory failure. Pediatrics 2002(3):110:529-533.

12. Keszler M, Abubakar K. Volume guarantee: stability of tidal volumeand incidence of hypocarbia. Pediatr Pulmonol 2004;38(3):240-245.

13. Cheema IU, Sinha AK, Kempley ST, Ahluwalia JS. Impact of vol-ume guarantee ventilation on arterial carbon dioxide tension in new-born infants: a randomised controlled trial. Early Hum Devel 2007;83(3):183-189.

14. Abubakar K, Keszler M. Effect of volume guarantee combined withassist/control vs synchronized intermittent mandatory ventilation. JPerinatol 2005;25(10):638-642.

15. Scopesi F, Calevo MG, Rolfe P, Arioni C, Traggiai C, Risso FM,Serra G. Volume targeted ventilation (volume guarantee) in the wean-ing phase of premature newborn infants. Pediatr Pulmonol 2007;42(10):864-870.

16. Lista G, Colnaghi M, Castoldi F, Condo V, Reali R, Compagnoni G,Mosca F. Impact of targeted-volume ventilation on lung inflamma-tory response in preterm infants with respiratory distress syndrome(RDS). Pediatr Pulmonol 2004;37(4):510-514.

17. Lista G, Castoldi F, Fontana P, Reali R, Reggiani A, Bianchi S,Compagnoni G. Lung inflammation in preterm infants with respira-tory distress syndrome: effects of ventilation with different tidalvolumes. Pediatr Pulmonol 2006;41(4):357-363.

18. Lista G, Castoldi F, Bianchi S, Battaglioli M, Cavigioli F, BosoniMA. Volume guarantee versus high-frequency ventilation: lung in-flammation in preterm infants. Arch Dis Child Fetal Neonatal Ed2008;93(4):F252-F256.

19. Polimeni V, Claure N, D’Ugard C, Bancalari E. Effects of volume-targeted synchronized intermittent mandatory ventilation on sponta-neous episodes of hypoxemia in preterm infants. Biol Neonate 2006;89(1):50-55.

20. Keszler M. Volume guarantee and ventilator-induced lung injury:Goldilocks’ rules apply. Pediatr Pulmonol 2006;41(4):364-366.

21. Piotrowski A, Sobala W, Kawczynski P. Patient-initiated, pressure-regulated, volume-controlled ventilation compared with intermittentmandatory ventilation in neonates: a prospective, randomised study.Intensive Care Med 1997;23(9):975-981.

22. D’Angio CT, Chess PR, Kovacs SJ, Sinkin RA, Phelps DL, KendigJW, et al. Pressure-regulated volume control ventilation vs synchro-nized intermittent mandatory ventilation for very low-birth-weightinfants: a randomized controlled trial. Arch Pediatr Adolesc Med2005;159(9):868-875.

23. McCallion N, Davis PG, Morley CJ. Volume-targeted versus pres-sure-limited ventilation in the neonate. Cochrane Database Syst Rev2005;(3):D003666.

24. Swamy R, Gupta S, Singh J, Donn SM, Sinha SK. Tidal volumedelivery and peak inspiratory pressure in babies receiving volumetargeted or time cycled, pressure limited ventilation: a randomizedcontrolled trial. J Neonatal-Perinatal Med 2008;1(2):239-243.

25. Kallet RH, Hemphill JC III, Dicker RA, Alonso JA, Campbell AR,et al. The spontaneous breathing pattern and work of breathing ofpatients with acute respiratory distress syndrome and acute lunginjury. Respir Care 2007;52(8):989-995.



Mechanical Ventilation of the Neonate: Should We … · target variable, pressure or volume, is...

Documents

Transcript of Mechanical Ventilation of the Neonate: Should We … · target variable, pressure or volume, is...