Evidence-based weaning - Anaesthesia UK

22 Pages·2001·0.13 MB·English

Special Articles
Evidence-Based Guidelines for Weaning
and Discontinuing Ventilatory Support
A Collective Task Force Facilitated by the American College of Chest
Physicians, the American Association for Respiratory Care, and the American
College of Critical Care Medicine
Introduction
Pathophysiology of Ventilator Dependence
Criteria to Assess Ventilator Dependence
Managing the Patient Who Has Failed a Spontaneous Breathing Test
Role of Tracheotomy in Ventilator-Dependent Patients
The Role of Long-Term Facilities
[Respir Care 2002;47(1):69–90]
Introduction quicklyaspossible.Althoughthisprocessoftenistermed
“ventilatorweaning”(implyingagradualprocess),wepre-
The discontinuation or withdrawal process from me- fer the more encompassing term “discontinuation.”
chanical ventilation is an important clinical issue.1,2 Pa-
tients are generally intubated and placed on mechanical SEE THE RELATED EDITORIAL ON PAGE 29
ventilatorswhentheirownventilatoryand/orgasexchange
capabilitiesareoutstrippedbythedemandsplacedonthem
Unnecessary delays in this discontinuation process in-
from a variety of diseases. Mechanical ventilation also is
crease the complication rate from mechanical ventilation
required when the respiratory drive is incapable of initi-
(eg, pneumonia, airway trauma) as well as the cost. Ag-
ating ventilatory activity, either because of disease pro-
gressivenessinremovingtheventilator,however,mustbe
cesses or drugs. As the conditions that warranted placing
balanced against the possibility that premature discontin-
thepatientontheventilatorstabilizeandbegintoresolve,
uation may occur. Premature discontinuation carries its
attention should be placed on removing the ventilator as
ownsetofproblems,includingdifficultyinreestablishing
artificial airways and compromised gas exchange. It has
been estimated that as much as 42% of the time that a
TheWritingCommittee,onbehalfofthePanel,includesNeilRMac-
medical patient spends on a mechanical ventilator is dur-
IntyreMDFAARC(Chairman),DeborahJCookMD,EWesleyElyJr
MDMPH,ScottKEpsteinMD,JamesBFinkMScRRTFAARC,John ing the discontinuation process.3 This percent is likely to
EHeffnerMD,DeanRHessPhDRRTFAARC,RolfDHubmayrMD, be much higher in patients with more slowly resolving
andDavidJScheinhornMD.OthermembersofthepanelincludeSu- lung disease processes.
zanneBurnsRNMSNCCRNRRN,DavidChaoMD,AndresEsteban
There are a number of important issues involved in the
MD,DouglasRGraceyMD,JesseBHallMD,EdwardFHaponikMD,
management of a mechanically ventilated patient whose
MarinHKollefMD,JordiManceboMD,ConstantineAManthousMD,
ArthurSSlutskyMD,MegAStearn-HassenpflugMSRD,andJamesK diseaseprocesshasbeguntostabilizeand/orreversesuch
StollerMDMScFAARC. thatthediscontinuationofmechanicalventilationbecomes
a consideration. First, an understanding of all the reasons
Reprinted with permission from Chest (Chest 2001;120(6)375–395).
that a given patient required a mechanical ventilator is
©AmericanCollegeofChestPhysicians.Copiescanbeorderedfromthe
AmericanCollegeofChestPhysicians,at1-800-343-2227or1-847-498- needed. Only with this understanding can medical man-
1400.www.chestjournal.org. agement be optimized. Second, assessment techniques to
identify patients who are capable of ventilator discontin-
Correspondence:NeilRMacIntyreMDFAARC,RespiratoryCareSer-
uation need to be utilized. Ideal assessment techniques
vices,DukeUniversityMedicalCenter,Box3911,DurhamNC27710.
E-mail:[email protected]. should be able to easily and safely distinguish which pa-
RESPIRATORY CARE•JANUARY 2002 VOL 47 NO 1 69
EVIDENCE-BASED GUIDELINES FOR WEANING AND DISCONTINUING VENTILATORY SUPPORT
Table1. GradesofEvidence wasreliedonto“fillinthegaps.”Consensuswasreached,
first, by team discussions and, later, through the re-
Grade Description
peated cycling of the draft through all members of the
A Scientificevidenceprovidedbywell-designed,well-conducted, task force.
controlledtrials(randomizedandnonrandomized)with Both the McMaster AHCPR group and the task force
statisticallysignificantresultsthatconsistentlysupportthe
recognized the needs for the future. These include more
guidelinerecommendation
randomizedcontrolledtrialstolookatanumberofissues.
B Scientificevidenceprovidedbyobservationalstudiesorby
Amongthemoreimportantquestionsthatneedanswering
controlledtrialswithlessconsistentresultstosupportthe
guidelinerecommendation arethefollowing:(1)Whichcriteriaarethebestindicators
C Expertopinionsupportedtheguidelinerecommendation,but ofreversalofrespiratoryfailureinthescreeningprocess?
scientificevidenceeitherprovidedinconsistentresultsorwas (2)Whatfactorsareinvolvedinventilatordependenceand
lacking
which measurement techniques are most useful in deter-
mining ultimate success in the discontinuation process?
(3)Inbalancingdiscontinuationaggressivenessagainstthe
tients need prompt discontinuation and which need con- risks of premature discontinuation, what is a reasonable
tinued ventilatory support. Third, ventilator management reintubation rate in patients recently removed from venti-
strategies for stable/recovering patients who still require latory support? (4) What is the value of trying to reduce
some level of ventilatory support need to be employed. levelsofpartialventilatorsupportinstable/recoveringpa-
Thesestrategiesneedtominimizebothcomplicationsand tients who have failed a discontinuation assessment? (5)
resourceconsumption.Fourth,extendedmanagementplans What role do tracheotomies have in facilitating the dis-
(includingtracheotomyandlong-termventilatorfacilities) continuationprocess?(6)Whatistheroleofthelong-term
need to be considered for the long-term ventilator-depen- facility, and when should patients be transferred to such
facilities?
dent patient.
Toaddressmanyoftheseissues,theAgencyforHealth-
carePolicyandResearch(AHCPR)chargedtheMcMaster
University Evidence Based Practice Center to do a com- Pathophysiology of Ventilator Dependence
prehensive evidence-based review of many of the issues
involved in ventilator weaning/discontinuation. Led by Introduction
DeborahCookMD,anexhaustivereviewofseveralthou-
sand articles in the world literature resulted in a compre-
Patients require mechanical ventilatory support when
hensive assessment of the state of the literature in 1999.4
the ventilatory and/or gas exchange capabilities of their
At the same time, the American College of Chest Physi-
respiratory system fail. This failure can be the result of
cians (ACCP), the Society for Critical Care Medicine
processes both within the lung as well as in other organ
(SCCM), and the American Association for Respiratory
systems, most notably the central nervous and the cardio-
Care (AARC) formed a task force to produce evidence-
vascularsystems.Althoughpatientsmaybedependenton
based clinical practice guidelines for managing the venti-
ventilatory support for brief periods of anesthesia or neu-
lator-dependentpatientduringthediscontinuationprocess.
romuscular blockade, the term “ventilator-dependent” is
The charge of this task force was to utilize the McMaster
usually reserved for patients with a need for mechanical
AHCPR report as well as their own literature review to
ventilation beyond 24 hours or by the fact that they have
addressthefollowing5issues:(1)thepathophysiologyof failed to respond during discontinuation attempts. Under
ventilator dependence; (2) the criteria for identifying pa- these circumstances, the clinical focus should be not only
tients who are capable of ventilator discontinuation; (3) onventilatormanagementbutalsoshouldincludeasearch
ventilatormanagementstrategiestomaximizediscontinu- for all of the possible reasons (especially potentially re-
ation potential; (4) the role of tracheotomy; and (5) the versibleones)thatmayexplaintheventilatordependency.
role of long-term facilities. Review/writing teams were
formed for each of these issues. Recommendation 1. In patients requiring mechanical
From these evidence-based reviews, a series of rec- ventilationfor(cid:1)24hours,asearchforallthecausesthat
ommendations were developed by the task force, which may be contributing to ventilator dependence should be
are the basis of this report. Each recommendation is undertaken.Thisisparticularlytrueinthepatientwhohas
followed by a review of the supporting evidence, in- failed attempts at withdrawing the mechanical ventilator.
cluding an assessment of the strength of the evidence Reversing all possible ventilatory and nonventilatory is-
(Table 1). As there were many areas in which evidence sues should be an integral part of the ventilator discon-
wasweakorabsent,theexpertopinionofthetaskforce tinuation process.
70 RESPIRATORY CARE•JANUARY 2002 VOL 47 NO 1
EVIDENCE-BASED GUIDELINES FOR WEANING AND DISCONTINUING VENTILATORY SUPPORT
Table2. CausesofVentilatorDependency below). Finally, dynamic hyperinflation can put ventila-
torymusclesinamechanicallydisadvantageousposition.33
Causes Description
Inanumberofstudies,patientswhofailedtorespondtoa
Neurologiccontroller Centraldrive;peripheralnerves withdrawal from mechanical ventilation tended to be
Respiratorysystem Mechanicalloads:respiratorysystem weaker (ie, they had a lower performance capacity) than
mechanics;imposedloading
thosewhosucceeded,34–49but,ingeneral,thewithin-group
Ventilatorymuscleproperties:inherent
variability in respiratory muscle strength was too large to
strength/endurance;metabolic
state/nutrients/oxygendeliveryand justify general conclusions.
extraction Ventilatorymusclefatiguealsocouldcontributetopoor
Gasexchangeproperties:vascularproperties muscle performance. However, the role of fatigue in ven-
andventilation/perfusionmatching
tilator dependence is not well understood, and the studies
Cardiovascularsystem Cardiactoleranceofventilatorymuscle
performed to date21,26,50–54 have failed to delineate the
work;peripheraloxygendemands
Psychologicalissues sensitivity and specificity of specific fatigue tests in ven-
tilator-dependent patients. Ventilatory support reduction-
relatedchangesintransdiaphragmaticpressure,respiratory
rate, and thoracoabdominal dyssynchrony are clearly not
Evidence (Grade B)
specific manifestations of respiratory muscle fatigue.55–60
There are a number of specific reasons why patients Themostpromisingdiagnostictestofdiaphragmcontrac-
maybeventilator-dependent(Table2).Determiningwhich tility to date is the transdiaphragmatic pressure measure-
factor or factors may be involved in a given patient re- ment during twitch stimulation of the phrenic nerves.21,61
quires both clinical awareness of these factors as well as However, too few patients have been studied with this
focused clinical assessments. The search for the underly- technique to draw any meaningful conclusions about the
ing causes for ventilator dependence may be especially prevalenceofdiaphragmfatiguethatisattributabletoven-
importantifpreviouslyunrecognized,butreversible,con- tilator dependence.
ditions are discovered.
The load on the ventilatory muscles is a function of
Neurologic Issues: The ventilatory pump controller in ventilationdemandsandrespiratorysystemmechanics(ie,
the brainstem is a rhythm and pattern generator, which primarilycomplianceandresistance).Normalminuteven-
receivesfeedbackfromcortical,chemoreceptiveandmech- tilation during spontaneous breathing is generally (cid:2) 10
anoreceptive sensors. The failure of this controller can L/min,normalrespiratorysystemcompliance(ie,tidalvol-
comefromseveralfactors.5–12Thesefactorscanbeeither ume/staticinflationpressure)isgenerally(cid:1)50–100mL/cm
structural(eg,brainstemstrokesorcentralapneas)ormet- H O, and normal airway resistance (ie, peak-static infla-
2
abolic(eg,electrolytedisturbancesorsedation/narcoticus- tionpressure/constantinspiratoryflow)isgenerally(cid:2)5–15
age13,14). The failure of the peripheral nerves also can be cm H O/L/s. Ventilation demands can increase as a con-
2
the result of either structural factors15 or metabolic/drug sequence of increased oxygen demands in patients with
factors.16,17Auniqueneurologicdysfunctionthatalsocould sepsisorincreaseddeadspaceinpatientswithobstructive
cause ventilator dependence is obstructive sleep apnea, in diseases. Compliance worsening can be a consequence of
which an artificial airway may be necessary to maintain lung edema, infection, inflammation, or fibrosis, and of
airway patency.10,11 chest wall abnormalities such as edema or surgical dress-
ings.Resistanceworseningcanbeaconsequenceofbron-
RespiratorySystemMuscle/LoadInteractions:Often,pa-
choconstrictionandairwayinflammation.Additionalload
tients who exhibit ventilator dependence do so because
canalsobeimposedbynarrowendotrachealtubesandby
there appears to be a mismatch between the performance
insensitiveorpoorlyresponsiveventilatordemandvalves.
capacityofventilatorypumpandtheloadplacedonit(ie,
thecapacity/loadimbalancehypothesis).18–23Thereisam- The load imposed by ventilation demands interacting
ple evidence that ventilatory pump performance may be with respiratory system mechanics can be expressed as
impairedinventilator-dependentpatientsbecauseventila- respiratory work, pressure time integral, or the change in
tory muscles are weak. This may be a consequence of metabolism(eg,theoxygencostattributabletobreathing).
atrophyandremodelingfrominactivity.2,24Itmayalsobe Manystudies19,35,62–66showthatpatientswhoareventila-
aconsequenceofinjuryfromoveruseandofinsultsasso- tor-dependenttendtohavelargerrespiratorymuscleloads
ciated with critical illness neuropathy and myopathy.25–29 than do patients who can be withdrawn from mechanical
A number of drugs (eg, neuromuscular blockers, amino- ventilation. In patients with airways obstruction, the load
glycosides,andcorticosteroids)alsocancontributetomy- imposedbydynamichyperinflationhasreceivedparticular
opathy,17,30–32ascanvariousmetabolicderangements(see attention as an important contributor to ventilator depen-
RESPIRATORY CARE•JANUARY 2002 VOL 47 NO 1 71
EVIDENCE-BASED GUIDELINES FOR WEANING AND DISCONTINUING VENTILATORY SUPPORT
dence.23,33,65,67–71 As is true for measures of ventilatory propermusclefunction.85,86Impairedoxygendeliverycan
pumpcapacity,however,mostinvestigatorsreportconsid- be a consequence either of inadequate oxygen content or
erable overlap in load parameters between patients with of inadequate cardiac output.87 Impaired oxygen uptake
different discontinuation outcomes. occursmostcommonlyduringsystemicinflammatorysyn-
Patients who go on to fail to respond to ventilator dromes such as sepsis.88
withdrawal attempts because of a capacity/load imbal-
Gas Exchange Factors: Gas exchange abnormalities
ance tend to display rapid, shallow breathing pat-
can develop during ventilatory support reductions for
terns.2,72,73 This pattern is advantageous from an ener-
several reasons. Various lung diseases produce ventila-
geticsperspective,butitisalsoassociatedwithincreased
tion-perfusionimbalancesandshunts.Ventilatordepen-
dead space and wasted ventilation and, hence with im-
dence thus may be a consequence of a need for high
paired carbon dioxide elimination. Chemoreceptive and
levels of expiratory pressure and/or the fraction of in-
mechanoreceptive feedback into the neural control of
spired oxygen (F ) to maintain an adequate oxygen
breathing is not well understood, and thus it is difficult IO
2
to distinguish whether this breathing pattern is a con- content.5,9Apatientwithhypoxemiaalsocandevelopa
sequenceofareducedrespiratorydriveperbreathoran fall in mixed venous PO levels from the cardiovascular
2
inability of ventilatory muscles to respond to an appro- factors described below.
priately increased neural stimulus.19,62,65,71,72
Cardiovascular Factors: Several groups of investiga-
Metabolic Factors and Ventilatory Muscle Function: tors have drawn attention to cardiovascular responses in
Nutrition, electrolytes, hormones, and oxygen transport ventilator-dependentpatientsandhaveemphasizedthepo-
areallmetabolicfactorsthatcanaffectventilatorymuscle tential for ventilatory support reductions to induce isch-
function. Inadequate nutrition leads to protein catabolism emia or heart failure in susceptible patients with limited
and loss of muscle performance.74,75 The normal hypoxic cardiacreserve.9,89–93Putativemechanismsincludethefol-
ventilatory response and the hypercapnic ventilatory re- lowing: (1) increased metabolic demand and hence circu-
sponse also have been shown to deteriorate under condi- latorydemandsthatareassociatedwiththetransitionfrom
tionsofsemistarvation.76Incontrast,overfeedingalsocan mechanicalventilationtospontaneousbreathinginpatients
impairtheventilatorwithdrawalprocessbyleadingtoex- withlimitedcardiacreserve;(2)increasesinvenousreturn
cesscarbondioxideproduction,whichcanfurtherincrease as the contracting diaphragm displaces blood from the
the ventilation loads on ventilatory muscles. Studies77,78 abdomen to the thorax; and (3) the increased left ventric-
havesuggestedthatpropernutritionalsupportcanincrease ularafterloadthatisimposedbynegativepleuralpressure
thelikelihoodofsuccessofventilatorwithdrawal.Anum- swings.Lemaireandcolleaguesdemonstratedleftventric-
ber of electrolyte imbalances also can impair ventilatory ular dysfunction (ie, the pulmonary capillary wedge pres-
musclefunction.5,9,79–81Phosphatedeficiencyhasbeenas- sure increased from 8 to 25 mm Hg) during failed venti-
sociated with respiratory muscle weakness and ventilator lator withdrawal attempts in 15 patients with COPD.
withdrawalfailure.Astudydemonstratingimprovedtrans- Following diuresis, 9 of these 15 patients were success-
diaphragmaticpressurevalueswiththerepletionofserum fully withdrawn from the ventilator.90
phosphorus levels in patients receiving mechanical venti-
Psychological Factors: Psychological factors may be
lation, however, did not specifically address the issue of
among the most important nonrespiratory factors leading
ventilator withdrawal.79 Magnesium deficiency also has
to ventilator dependence. Fear of the loss of an apparent
been reported to be associated with muscle weakness,82
lifesupportsystemaswellassocial/familial/economicis-
althoughtherelationshiptoventilatordependencehasnot
beenspecificallyaddressed.Finally,bicarbonateexcretion sues all may play a role. Stress can be minimized by
frominappropriateoverventilation(oftenoccurringinpa- frequent communication among the staff, the patient, and
tientswithchronicobstructivepulmonarydisease(COPD) the patient’s family.94 Environmental stimulation using
with chronic baseline hypercapnia) can impair ventilator television, radio, or books also appears to improve psy-
withdrawaleffortsasthepatienthasadiminishedcapacity chologicalfunctioning.2Ambulationusingaportableven-
tocompensateforhypercapnia.Severehypothyroidismand tilator(orbagging)hasbeenshowntobenefitattitudesand
myxedema directly impair diaphragmatic function and outlooksinlong-termventilator-dependentpatients.Sleep
bluntventilatoryresponsestohypercapniaandhypoxia.83,84 deprivation may cause impairment of the respiratory con-
Otherhormonalfactorsthatareimportantforoptimalmus- trolsystem,95althoughthismayberelatedtoaccompany-
clefunctionincludeinsulin/glucagonandadrenalcortico- ing factors rather than to sleep deprivation per se.96 Fi-
steroids. nally,biofeedbackmaybehelpfulindecreasingtheweaning
As in other organs, adequate oxygen delivery and oxy- time in patients who are having difficulty withdrawing
gen uptake by the ventilatory muscles is necessary for from ventilatory support.97–98
72 RESPIRATORY CARE•JANUARY 2002 VOL 47 NO 1
EVIDENCE-BASED GUIDELINES FOR WEANING AND DISCONTINUING VENTILATORY SUPPORT
Criteria to Assess Ventilator Dependence ventilator discontinuation. These “entry” criteria often in-
cludesomeformofclinicaljudgmentorintuition,making
Introduction results from one study difficult to compare to another. In
addition, clinician/investigators deciding to proceed with
Theprocessofdiscontinuingmechanicalventilatorysup- ventilator discontinuation/extubation often have not been
port begins with a recognition of adequate recovery from blinded to the parameters being analyzed as possible pre-
acute respiratory failure. Thereafter, careful clinical as- dictors. Indeed, the parameter being analyzed may often
sessmentsarerequiredtodeterminethepatient’sreadiness enter into the clinical decision on whether either to con-
forsubsequentdiscontinuationofventilatorysupportand, tinueortodiscontinueventilatorysupport.Othermethod-
ultimately, extubation. To facilitate this process, investi- ologicalproblemswiththeseobservationalstudiesinclude
gators have focused on identifying objective criteria to different measurement techniques of a given parameter
determine the answers to the following questions: When from study to study, large coefficients of variation with
can efforts to discontinue ventilation be initiated? What repeated measurements or from study to study of a given
assessment strategies will best identify the patient who is parameter,111,112 different patient populations (eg, long-
ready for ventilator discontinuation? When should extu- term vs short-term ventilator dependence),113,114 and the
bation be carried out, and how can extubation outcome absence of objective criteria to determine a patient’s tol-
best be predicted? erance for a trial of either discontinuation or extubation.
Evidencetoanswerthesequestionscomeslargelyfrom Third, assessed outcomes differ from study to study.
observational studies in which a certain parameter (or set Someinvestigatorshaveexaminedsuccessfultoleranceof
of parameters) is compared in a group of patients who aspontaneousbreathingtrial(SBT),othershaveusedper-
either successfully or unsuccessfully have been removed manent discontinuation of the ventilator, and others have
fromtheventilator.Thegeneralgoalofthesestudiesisto combinedsuccessfuldiscontinuationandextubation.This
find “predictors” of outcome. Evaluating the results from latterapproachisnotoptimal,giventhedifferencesinthe
these types of studies can be difficult for several reasons. pathophysiologyofdiscontinuationversusextubationfail-
First,the“aggressiveness”oftheclinician/investigator’s ure (see below).102,106 In addition, different studies use
weaning and discontinuation philosophy needs to be un- differentdurationsofventilatordiscontinuationorextuba-
derstood, as it will affect the performance of a given pre- tiontodefinesuccessorfailure.Although24–48hoursof
dictor. A very aggressive clinical philosophy will maxi- unassistedbreathingoftenisconsideredtodefinethesuc-
mize the number of patients withdrawn from ventilatory cessfuldiscontinuationofventilatorsupport,manystudies
support but could also result in a number of premature use shorter time periods to indicate success and often do
discontinuations,withasubsequentneedforreintubations not report subsequent reintubation rates or the need to
and/or reinstitution of support. In contrast, a less aggres- reinstitute mechanical ventilatory support.
sive clinical philosophy will minimize premature discon- Fourth, a number of ways have been used to express
tinuationsbutcouldalsounnecessarilyprolongventilatory predictorperformance,andmanycanbeconfusingormis-
supportinotherpatients.Unfortunately,therearenogood leading. Traditional indexes of diagnostic test power in-
data to help clinicians to determine the best balance be- clude sensitivity/specificity and positive/negative predic-
tween premature and delayed discontinuations in evaluat- tive values. These indexes are limited, however, in that
ing a given discontinuation strategy. Clearly, extubation theyrelyonasinglecutpointorthresholdandthattheydo
failure should be avoided whenever possible because the not provide an easy way to go from pretest likelihood or
needforreintubationcarriesan8-foldhigheroddsratiofor probability, through testing, to a posttest probability. The
nosocomialpneumonia99anda6-foldto12-foldincreased McMaster AHCPR report4 recommends the use of likeli-
mortality risk.100–103 In contrast, the maintenance of un- hood ratios (LRs), and these will be used in this report to
necessary ventilator support carries its own burden of pa- describe predictor performance. The LR is an expression
tient risk for infection and other complications.104,105 Re- of the odds that a given test result will be present in a
portedreintubationratesrangefrom4%to23%fordifferent patientwithagivenconditioncomparedtoapatientwith-
intensive care unit (ICU) populations,100,101,103,104,106–110 outthecondition.AnLR(cid:1)1indicatesthattheprobability
and may be as high as 33% in patients with mental status of success increases, while values (cid:2) 1 indicate that the
changesandneurologicalimpairment.103Althoughtheop- probability of failure increases. LRs between 0.5 and 2
timal rate of reintubation is not known, it would seem indicate that a weaning parameter is associated with only
likely to rest between 5% and 15%. small,clinicallyunimportantchangesintheposttestprob-
Second,anumberofmethodologicalproblemsexistwith ability of success or failure. In contrast LRs from 2 to 5
most of these observational studies. For instance, patients and from 0.3 to 0.5 correlate with small but potentially
are recruited into these studies because investigators be- importantchangesinprobability,whileratiosof5to10or
lieve that there is some reasonable chance of success for 0.1to0.3correlatewithmoreclinicallyimportantchanges
RESPIRATORY CARE•JANUARY 2002 VOL 47 NO 1 73
EVIDENCE-BASED GUIDELINES FOR WEANING AND DISCONTINUING VENTILATORY SUPPORT
Table3. CriteriaUsedinWeaning/DiscontinuationStudiesto Rationale and Evidence (Grade B)
DetermineWhetherPatientsReceivingHighLevelsof
VentilatorySupportCanBeConsideredfor While some investigators argue that the process of dis-
Discontinuation(ie,EnteredIntotheTrials)* continuation starts as soon as the patient is intubated, it
would seem reasonable that an appropriate level of venti-
ObjectiveMeasurements Adequateoxygenation(eg,P (cid:2)60mm
HgonF (cid:1)0.4;PEEP(cid:1)O25–10cm latory support should be maintained until the underlying
H O;P IO/F2 (cid:2)150–300) cause of acute respiratory failure and any complicating
2 O2 IO2
Stablecardiovascularsystem(eg,HR(cid:1) issues have shown some sign of reversal. Indeed, patients
140beats/min;stablebloodpressure; withunresolvingrespiratoryfailurewhorequirehighlev-
noorminimalvasopressors) els of ventilatory support are probably at high risk for
Afebrile(eg,temperature(cid:2)38°C)
respiratory muscle fatigue (and the consequent prolonga-
Nosignificantrespiratoryacidosis
tion of the need for mechanical ventilation) if aggressive
Adequatehemoglobin(eg,Hgb(cid:2)8–10
g/dL) reductions in support are undertaken.50,52,116–118
Adequatementation(eg,arousable,GCS Thecriteriausedbyclinicianstodefinedisease“rever-
(cid:2)13,nocontinuoussedativeinfusions sal,”however,havebeenneitherdefinednorprospectively
Stablemetabolicstatus(eg,acceptable evaluatedinarandomizedcontrolledtrial.Rather,various
electrolytes)
combinations of subjective assessment and objective cri-
SubjectiveClinical Resolutionofdiseaseacutephase;
teria (eg, usually gas exchange improvement, mental sta-
Assessments physicianbelievesdiscontinuation
possible;adequatecough tusimprovement,neuromuscularfunctionassessments,and
radiographicsigns)thatmayserveassurrogatemarkersof
PO2(cid:3)partialpressureofoxygen recoveryhavebeenemployed(Table3).101–103,107–109,119,120
FIO2(cid:3)fractionofinspiredoxygen
PEEP(cid:3)positiveend-expiratorypressure Itshouldbenoted,however,thatsomepatientswhohave
HR(cid:3)heartrate
not ever met one or more of these criteria still have been
GCS(cid:3)GlasgowComaScale
*AdaptedfromReferences101–103,107–109,119,and120 shown to be capable of eventual liberation from the ven-
tilator.104
These “clinical assessments” of the status of the pa-
inprobability.Ratiosof(cid:1)10or(cid:2)0.1correlatewithvery
tient’srespiratoryfailure,however,arenotenoughtomake
large changes in probability.115 decisions on the discontinuation of support. For example,
Finally, because some investigators report data as con-
one survey121 of intensivists using clinical judgment to
tinuous values (eg, means) rather than providing defined
assessthepotentialfordiscontinuationfoundasensitivity
threshold values, combining studies using meta-analytic
of only 35% (6 of 17 patients who were successfully dis-
techniques often cannot be done.
continuedwereidentified)andaspecificityof79%(11of
14whofaileddiscontinuationwereidentified).Moreover,
Recommendation2. Patientsreceivingmechanicalven-
in 2 large trials,107,109 despite the presence of apparent
tilation for respiratory failure should undergo a formal
disease stability/reversal prior to performing a screening
assessment of discontinuation potential if the following
SBT, the managing clinicians did not recognize that dis-
criteria are satisfied:
continuation was feasible in almost two thirds of the sub-
1.Evidenceforsomereversaloftheunderlyingcauseof
jects. Thus, the conclusion is that some evidence of “clini-
respiratory failure;
2. Adequate oxygenation (eg, P /F (cid:1) 150–200; cal” stability/reversal is a key first step in assessing for
requiringpositiveend-expiratorypraeOss2ureIO[2PEEP](cid:1)5–8 discontinuation potential but that more focused assessments
cm H O; F (cid:1) 0.4–0.5) and pH (eg, (cid:2) 7.25); are needed before deciding to continue or discontinue ven-
2 IO
2 tilatorysupport.
3. Hemodynamic stability as defined by the absence of
active myocardial ischemia and the absence of clinically
Recommendation3. Formaldiscontinuationassessments
importanthypotension(ie,aconditionrequiringnovaso-
pressor therapy or therapy with only low-dose vasopres- for patients receiving mechanical ventilation for respira-
sors such as dopamine or dobutamine (cid:2) 5 (cid:3)g/kg/min); toryfailureshouldbedoneduringspontaneousbreathing
and rather than while the patient is still receiving substantial
4. The capability to initiate an inspiratory effort. ventilatorysupport.Aninitialbriefperiodofspontaneous
The decision to use these criteria must be individual- breathingcanbeusedtoassessthecapabilityofcontinu-
ized. Some patients not satisfying all of the above the ing onto a formal SBT. The criteria with which to assess
criteria (eg, patients with chronic hypoxemia below the patienttoleranceduringSBTsaretherespiratorypattern,
thresholds cited) may be ready for attempts at discontin- adequacy of gas exchange, hemodynamic stability, and
uation of mechanical ventilation. subjective comfort. The tolerance of SBTs lasting 30 to
74 RESPIRATORY CARE•JANUARY 2002 VOL 47 NO 1
EVIDENCE-BASED GUIDELINES FOR WEANING AND DISCONTINUING VENTILATORY SUPPORT
Table4. MeasurementsPerformedEitherWhilePatientWasReceivingVentilatorySupportorDuringaBriefPeriodofSpontaneousBreathing
ThatHaveBeenShowntoHaveStatisticallySignificantLRstoPredicttheOutcomeofaVentilatorDiscontinuationEffortinMoreThan
OneStudy
Parameter Studies(n) ThresholdValues PositiveLRsRange
MeasuredonVentilator
V˙ 20 10–15L/min 0.81–2.37
E
NIF 10 (cid:4)20to(cid:4)30cmHO 0.23–2.45*
2
P 16 (cid:4)15to(cid:4)30cmH O 0.98–3.01
Imax 2
P /P 4 0.30 2.14–25.3
0.1 Imax
CROPscore 2 13 1.05–19.74
MeasuredDuringaBriefPeriodofSpontaneousBreathing
RR 24 30–38breaths/min 1.00–3.89
V 18 325–408mL(4–6mL/kg) 0.71–3.83
T
f/V ratio 20 60–105/L 0.84–4.67
T
LR(cid:3)likelihoodratio
V˙E(cid:3)minuteventilation
*1studyreportedalikelihoodratio(LR)of35.79.
NIF(cid:3)negativeinspiratoryforce(maximuminspiratorypressure)
PImax(cid:3)maximuminspiratorypressure
P0.1(cid:3)mouthocclusionpressure0.1saftertheonsetofinspiratoryeffort
CROP(cid:3)indexincludingcompliance,respiratoryrate,oxygenation,andpressure
RR(cid:3)respiratoryrate
VT(cid:3)tidalvolume
f/VT(cid:3)respiratoryrate/tidalvolumeratio
Table5. FrequencyofToleratinganSBTinSelectedPatientsandRateofPermanentVentilatorDiscontinuationFollowingaSuccessfulSBT
Patients PatientsHaving
Patients PatientsDiscontinuing
Study ToleratingSBT VentilationReinstituted
ReceivingSBT Ventilation
n(%) n(%)
Estebanetal105 546 416(76) 372 58(16)
Elyetal108 113 88(78) 65 5(4)
Dojatetal110 38 22(58) 22 5(23)
Estebanetal101 246 192(78) 192 36(19)
Estebanetal102 270* 237(89) 237 32(14)
Estebanetal102 256† 216(84) 216 29(13)
SBT(cid:3)spontaneousbreathingtrial
*30minSBT
†120minSBT
120 minutes should prompt consideration for permanent The McMaster AHCPR report4 found evidence in the
ventilator discontinuation. literature supporting a possible role for 66 specific mea-
surements as predictors. Some of these (eg, the negative
Rationale and Evidence (Grade A)
effects of the duration of mechanical ventilation, and the
Because clinical impression is so inaccurate in deter- lengthof/difficultyofsurgery44,122–124)werederivedfrom
miningwhetherornotapatientmeetingthecriterialisted general clinical observations, but most were from studies
in Table 3 will successfully discontinue ventilatory sup- onfocusedassessmentsofthepatient’srespiratorysystem.
port,amorefocusedassessmentofdiscontinuationpoten- Fromthese,theMcMasterAHCPRgroupidentified8pa-
tial is necessary. These assessments can be performed ei- rameters that had consistently significant LRs to predict
ther during spontaneous breathing or while the patient is successfuldiscontinuationinseveralstudies.Someofthese
still receiving substantial ventilatory support. These as- measurementsaremadewhilethepatientisstillreceiving
sessmentscanbeusednotonlytodrivedecisionsonwean- ventilatorysupport;othersrequireanassessmentduringa
ing and discontinuation (ie, functioning as predictors) but brief period of spontaneous breathing. These parameters,
also to offer insight into mechanisms of discontinuation their threshold values, and the range of reported LRs are
failures. given in Table 4. It should be noted that despite the sta-
RESPIRATORY CARE•JANUARY 2002 VOL 47 NO 1 75
EVIDENCE-BASED GUIDELINES FOR WEANING AND DISCONTINUING VENTILATORY SUPPORT
Table6. CriteriaUsedinSeveralLargeTrials*toDefineTolerance value) cannot be formally assessed. Indeed, it is conceiv-
ofanSBT able that iatrogenic factors such as endotracheal tube dis-
comfort or continuous positive airway pressure (CPAP)
Objectivemeasurements Gasexchangeacceptability(S (cid:2)
indicatingtolerance/success 85–90%;P (cid:2)50–60mmpHO2g; demand-valve insensitivity/unresponsiveness, rather than
pH(cid:2)7.32;O2increaseinP (cid:1) true ventilator dependence, caused the failure of the SBT
aCO2
10mmHg) inatleastsomeofthesepatients.125Thus,itisthusunclear
Hemodynamicstability(HR(cid:2) how many patients who are unable to tolerate an SBT
120–140beats/min;HRnot wouldstillbeabletotoleratelong-termventilatordiscon-
changed(cid:1)20%;systolicBP(cid:2)
tinuation. Although the number is likely to be small, it is
180–200mmHgand(cid:1)90mm
Hg;BPnotchanged(cid:1)20%,no probably not zero, and this needs to be considered when
pressorsrequired) dealing with patients who repeatedly fail an SBT.
Stableventilatorypattern(eg,RR(cid:1) The criteria used to define SBT “tolerance” are often
30–35breaths/min,RRnot integratedindexessince,asnotedabove,singleparameters
changed(cid:1)50%) aloneperformsopoorly.Theseintegratedindexesusually
Subjectiveclinicalassessments Changeinmentalstatus(eg,
include several physiologic parameters as well as clinical
indicatingintolerance/failure somnolence,coma,agitation,
judgment, incorporating such difficult-to-quantify factors
anxiety)
as“anxiety,”“discomfort,”and“clinicalappearance.”The
Onsetorworseningofdiscomfort
Diaphoresis criteriathathavebeenusedinseverallargetrialsaregiven
Signsofincreasedworkof in Table 6.
breathing(useofaccessory A potential concern about the SBT is safety. Although
respiratorymuscles, unnecessary prolongation of a failing SBT conceivably
thoracoabdominalparadox)
couldprecipitatemusclefatigue,hemodynamicinstability,
SBT(cid:3)spontaneousbreathingtrial discomfort, or worsened gas exchange,19,50,117,126,127 there
SpO2(cid:3)oxygensaturationmeasuredbypulseoximetry are no data showing that SBTs contribute to any adverse
PPHOaRC2O(cid:3)(cid:3)2(cid:3)hpeaaartrrittaerlraipatelrepsasrutriealopfroesxsyugreenofcarbondioxide onuiztecdo.mIensdeifedt,erimnianactoedhoprtroomfp(cid:1)tly1,w00h0enpfaatiielunrtes iins rwehcoogm-
RR(cid:3)respiratoryrate
SBTswereroutinelyadministeredandproperlymonitored
*References101,102,105,108–110,119,and120.
as part of a protocol, only one adverse event was thought
to be even possibly associated with the SBT.104
tisticalsignificanceoftheseparameters,thegenerallylow There is evidence that the detrimental effects of venti-
LRsindicatethattheclinicalapplicabilityoftheseparam- latorymuscleoverload,ifitisgoingtooccur,oftenoccurs
eters alone to individual patients is low. early in the SBT.73,108,110,128 Thus, the initial few minutes
Although assessments that are performed while a pa- ofanSBTshouldbemonitoredclosely,beforeadecision
tient is receiving substantial ventilatory support or during ismadetocontinue(thisisoftenreferredtoasthe“screen-
a brief period of spontaneous breathing (Tables 3 and 4) ing”phaseofanSBT).Thereafter,thepatientshouldcon-
canyieldimportantinformationaboutdiscontinuationpo- tinuethetrialforatleast30min,butfornot(cid:1)120min,102
tential, assessments that are performed during a formal, to assure maximal sensitivity and safety. It also appears
carefullymonitoredSBTappeartoprovidethemostuseful that whether the SBT is performed with low levels of
information to guide clinical decision-making regarding CPAP(eg,5cmH O),lowlevelsofpressuresupport(eg,
2
discontinuation.Indeed,becauseoftheefficacyandsafety 5–7 cm H O), or simply as “T-piece” breathing has little
2
ofaproperlymonitoredSBT(seebelow),theassessments effectonoutcome.101,129–131CPAP,however,conceivably
inTable4thatareperformedtopredictSBToutcomeare couldenhancebreathtriggeringinpatientswithsignificant
generally unnecessary. auto-PEEP.132,133
Inconcept,theSBTshouldbeexpectedtoperformwell,
as it is the most direct way to assess a patient’s perfor- Recommendation4. Theremovaloftheartificialairway
mance without ventilatory support. Indeed, the evidence from a patient who has successfully been discontinued
forthisconceptisquitestrong.AscanbeseeninTable5, from ventilatory support should be based on assessments
multiplestudieshavefoundthatpatientstolerantofSBTs of airway patency and the ability of the patient to protect
that are 30 to 120 min in length were found to have suc- the airway.
cessful discontinuations at least 77% of the time.
Rationale and Evidence (Grade C)
Becausethe12to42%ofpatientsintheTable5studies
failing the SBT were not systematically removed from Extubation failure can occur for reasons distinct from
ventilatory support, the ability of a failed SBT to predict thosethatcausediscontinuationfailure.Examplesinclude
theneedforventilatordependence(ie,negativepredictive upperairwayobstructionorinabilitytoprotecttheairway
76 RESPIRATORY CARE•JANUARY 2002 VOL 47 NO 1
EVIDENCE-BASED GUIDELINES FOR WEANING AND DISCONTINUING VENTILATORY SUPPORT
andclearsecretions.Theriskofpostextubationupperair- tilatorysupportbemanaged?Specifically,shouldanSBT
wayobstructionincreaseswiththedurationofmechanical be tried again? If so, when? What form of ventilatory
ventilation, female gender, trauma, and repeated or trau- support should be provided in between SBTs and should
matic intubation.106 The detection of an air leak during support be at a constant high level or should efforts be
mechanicalventilationwhentheendotrachealtubeballoon madetoroutinelyreducethelevelofsupportgradually(ie,
is deflated can be used to assess the patency of the upper to wean support)?
airway(cuffleaktest).134Inastudyofmedicalpatients,135 Evaluating evidence addressing mechanical ventilatory
a cuff leak (cid:2) 110 mL (ie, average of 3 values on 6 con- support strategies is particularly problematic. This is be-
secutive breaths) measured during assist control ventila- cause trials comparing 2 or more approaches to ventilator
tion within 24 hours of extubation identified patients at management compare not only the modes of ventilation
highriskforpostextubationstridor.Althoughothershave but also how those modes are used. Ideally, trial design
notconfirmedtheutilityofthecuffleaktestforpredicting shouldbesuchthatmanagementphilosophiesandtheag-
postextubation stridor,136 many patients who develop this gressivenessofsupportreductionaresimilarineachstrat-
can be treated with steroids and/or epinephrine (and pos- egy being evaluated. Unfortunately, this is often not the
sibly with noninvasive ventilation and/or heliox) and do case, as investigator experience with one approach has a
not necessarily need to be reintubated. Steroids and/or tendency to result in more favorable “rules” of support
epinephrine could also be used 24 hours prior to extuba- reduction for that approach compared to others.
tion in patients with low cuff leak values. It is also im-
portanttonotethatalowvalueforcuffleakmayactually Recommendation5. Patientsreceivingmechanicalven-
beduetoencrustedsecretionsaroundthetuberatherthan tilation for respiratory failure who fail an SBT should
to a narrowed upper airway. Despite this, reintubation have the cause for the failed SBT determined. Once re-
equipment (including tracheostomy equipment) should be versiblecausesforfailurearecorrected,andifthepatient
readilyavailablewhenextubatingpatientswithalowcuff still meets the criteria listed in Table 3, subsequent SBTs
leak values. should be performed every 24 hours.
The capacity to protect the airway and to expel secre-
Rationale and Evidence (Grade A)
tions with an effective cough would seem to be vital for
extubation success, although specific data supporting this AlthoughfailedSBTsareoftenareflectionofpersistent
conceptarefew.Successfulextubationshavebeenreport- respiratory system abnormalities,52 a failed SBT should
ed137 in a select group of brain-injured comatose patients prompt a search for other causes or complicating factors
whowerejudgedtobecapableofprotectingtheirairways. (see the “Patholophysiology of Ventilator Dependence”
However, it is difficult to extrapolate this experience to section).Specificissuesincludetheadequacyofpaincon-
more typical ICU patients, and many would argue that trol,theappropriatenessofsedation,fluidstatus,broncho-
some capability of the patient to interact with the care dilator needs, the control of myocardial ischemia, and the
team should be present before the removal of an artificial presenceofotherdiseaseprocessesthateithercanbereadily
airway. Airway assessments generally include noting the addressedorelsecanbeconsideredwhendecidingtopro-
quality of cough with airway suctioning, the absence of ceed further with discontinuation attempts.
“excessive”secretions,orthefrequencyofairwaysuction- Assuming medical management is optimized and that
ing(eg,every2hormore).34,108,138Coplinetal137devised the patient who has failed an SBT still meets the criteria
an“airwaycarescore,”whichsemiquantitativelyassesses listedinTable3,thefollowing2questionsinvolvingsub-
cough, gag, suctioning frequency, and sputum quantity, sequentSBTsarise:First,shouldSBTsbeattemptedagain
viscosity,andcharacterthatpredictedextubationoutcome. or should another approach to ventilator withdrawal be
Peakcoughflowsof(cid:1)160L/minpredictsuccessfultrans- attempted? Second, if an SBT is attempted again, when
laryngeal extubation or tracheostomy tube decannulation should that be?
in neuromuscular- or spinal cord-injured patients.139 Therearesomedataonwhichtobaseananswertothe
firstquestion.Theonelargerandomizedtrial107thatcom-
Managing the Patient Who Has Failed pared routine SBTs to 2 other weaning strategies that did
a Spontaneous Breathing Test notincludeSBTsprovidescompellingevidencethatSBTs
administered at least once daily shorten the discontinua-
Introduction tionperiodcomparedtostrategiesthatdonotincludedaily
SBTs. In addition, 2 studies108,119 showing the success of
The failure of a patient to complete an SBT raises 2 protocol-driven ventilator discontinuation strategies over
important questions. First, what caused the SBT failure, “usual care” both included daily SBTs. The subsequent
and are there readily reversible factors that can be cor- use of routine subsequent SBTs in this patient population
rected? Second, how should subsequent mechanical ven- thus seems appropriate.
RESPIRATORY CARE•JANUARY 2002 VOL 47 NO 1 77
EVIDENCE-BASED GUIDELINES FOR WEANING AND DISCONTINUING VENTILATORY SUPPORT
Thereareseverallinesofevidencethatsupportwaiting Table7. ModesofPartialVentilatorSupport
24hoursbeforeattemptinganSBTagaininthesepatients.
First, except in patients recovering from anesthesia, mus- Mode PatientWorkAdjustedBy
clerelaxants,andsedatives,respiratorysystemabnormal- SIMV Numberofmachinebreathssupplied(ie,thefewer
ities rarely recover over a short period of hours and thus thenumberofmachinebreaths,themore
frequent SBTs over a day may not be expected to be spontaneousbreathsarerequired)
PSV Levelofinspiratorypressureassistancewith
helpful.SupportingthisaredatafromJubranandTobin52
spontaneousefforts
showing that failed SBTs often are due to persistent re-
SIMV(cid:5)PSV CombiningtheadjustmentsofSIMVandPSV
spiratorysystemmechanicalabnormalitiesthatareunlikely
VS PSVwitha“guaranteed”minimumV (PSVlevel
T
toreverserapidly.Second,therearedatasuggestingthata adjustsautomaticallyaccordingtoclinicianV
T
failedSBTmayresultinsomedegreeofrespiratorymus- setting)
clefatigue.50,117,118Ifso,studies126,140conductedinhealthy VAPS(PA) PSVwith“guaranteed”minimumVT(additional
subjects suggest that recovery may not be complete for flowissuppliedatendinspirationifnecessary
anywhere from several hours to (cid:1) 24 hours. Third, the toprovideclinicianVTsetting)
MMV SIMVwitha“guaranteed”V˙ (machinebreath
trial by Esteban et al107 specifically addressed this issue E
rateautomaticallyadjustsaccordingtoclinician
and provided strong evidence that twice-daily SBTs offer V˙ setting)
E
no advantage over a single SBT and, thus, would serve APRV Pressuredifferencebetweeninflationandrelease
only to consume unnecessary clinical resources. (ie,thelessthepressuredifference,themore
spontaneousbreathsarerequired)
Recommendation6. Patientsreceivingmechanicalven-
SIMV(cid:3)synchronizedintermittentmandatoryventilation
tilation for respiratory failure who fail an SBT should PSV(cid:3)pressuresupportventilation
receive a stable, nonfatiguing, comfortable form of venti- VS(cid:3)volumesupport
VAPS(PA)(cid:3)volume-assuredpressuresupport(pressureaugmentation)
latory support. VT(cid:3)tidalvolume
MMV(cid:3)mandatoryminuteventilation
Rationale and Evidence (Grade B) V˙E(cid:3)minuteventilation
APRV(cid:3)airwaypressure-releaseventilation
Thereareanumberofventilatormodesthatcanprovide
substantial ventilatory support as well as the means to
reduce partial ventilatory support in patients who have strategiesusingdifferentmodesbutnotroutinedailySBTs.
failed an SBT (Table 7). A key question, however, is The study by Brochard et al109 was the most similar in
whether attempts at gradually lowering the level of sup- design to the study by Esteban et al107 that was noted
port (weaning) offer advantages over a more stable, un- before, and it included a pressure-support group and an
changing level of support between SBTs. The arguments IMV group. A third group received gradually increasing
forusinggradualreductionsare(1)thatmusclecondition- fixedperiodsofspontaneousbreathingthatweredesigned
ing might occur if ventilatory loads are placed on the only to provide brief periods of work and not specifically
patient’s muscles and (2) that the transition to extubation to test for discontinuation (ie, they were not routine daily
or to an SBT might be easier from a low level of support SBTs,asdefinedabove).Theresultsshowedthattheiruse
thanfromahighlevelofsupport.Datasupportingeitherof of gradually lengthening spontaneous breathing periods
these claims, however, are few. However, maintaining a was inferior to other strategies and, like the Esteban trial,
stable level of support between SBTs reduces the risk of thepressuresupportstrategywaseasiertoreducethanthe
precipitating ventilatory muscle overload from overly ag- IMV strategy. The other 2 randomized trials141,142 that
gressive support reduction. It also offers a substantial re- wereidentifiedbytheMcMasterAHCPRreportweremuch
source consumption advantage in that it requires far less smallerthantheEstebanetal107andBrochardetal,109and
practitioner time. The study by Esteban et al107 partially both suggested that pressure support was easier to re-
addressedthisissueinthatitcompareddailySBTs(anda duce than IMV alone. Because none of these studies
stable level of support in those who failed) to 2 other offer evidence that gradual support strategies are supe-
approachesusinggradualreductionsinsupport(ie,wean- rior to stable support strategies between SBTs, the clin-
ing with pressure support or intermittent mandatory ven- ical focus for the 24 hours after a failed SBT should be
tilation [IMV]) and demonstrated that the daily SBT with on maintaining adequate muscle unloading, optimizing
stable support between tests permitted the most rapid dis- comfort (and thus sedation needs), and avoiding compli-
continuation. What has not been addressed, however, is cations, rather than aggressive ventilatory support reduc-
whether gradual support reductions coupled with daily tion.
SBTs offer any advantages. Ventilator modes and settings can affect these goals.143
The McMaster AHCPR report4 identified 3 other ran- Assisted modes of ventilation (as opposed to machine-
domized trials109,141,142 that compared gradual reduction controlled modes) are generally preferable in this setting
78 RESPIRATORY CARE•JANUARY 2002 VOL 47 NO 1