-
Innovations in Mechanical Ventilation
Richard D Branson MSc RRT FAARC and Jay A Johannigman MD
IntroductionClosed-Loop Control of Weaning
HistorySmartCare/PSInvention or Innovation
Automated Measurement of Functional Residual Capacity
DuringMechanical Ventilation
HistoryAutomated FRC Measurements Using Nitrogen
WashoutPrinciple of MeasurementFRC INviewAccuracy and
ApplicationInvention or Innovation
Neurally Adjusted Ventilatory AssistHistory and Principle of
OperationApplicationCurrent LiteratureInvention or Innovation
Information PresentationHistory and PrinciplesCurrent
LiteratureImplementationInvention or Innovation
Summary
New features of mechanical ventilators are frequently
introduced, including new modes, monitoringtechniques, and
triggering techniques. But new rarely translates into any
measureable improvementin outcome. We describe 4 new techniques and
attempt to define what is a new invention versuswhat is innovativea
technique that significantly improves a measurable variable. We
describe andreview the literature on automated weaning, automated
measurement of functional residual capac-ity, neural triggering,
and novel displays of respiratory mechanics. Key words: mechanical
ventila-tion, weaning, neurally adjusted ventilatory assistance,
ventilator, functional residual capacity. [RespirCare
2009;54(7):933947. 2009 Daedalus Enterprises]
Richard D Branson MSc RRT FAARC and Jay A Johannigman MD
areaffiliated with the Department of Surgery, Division of
Trauma/CriticalCare, University of Cincinnati, Cincinnati,
Ohio.
Mr Branson has disclosed relationships with Ikaria, Cardinal,
Newport,and Covidien. Dr Johannigman has disclosed no conflicts of
interest.
Mr Branson presented a version of this paper at the symposium
Cur-rent and Evolving Concepts in Critical Care, at the 54th
International
Respiratory Congress of the American Association for
RespiratoryCare, held December 1316, 2008, in Anaheim, California.
The sym-posium was made possible by an unrestricted educational
grant fromIkaria.
Correspondence: Richard D Branson MSc RRT FAARC, Department
ofSurgery, University of Cincinnati, 231 Albert Sabin Way,
Cincinnati OH45267-0558. E mail: [email protected].
RESPIRATORY CARE JULY 2009 VOL 54 NO 7 933
-
Introduction
Mechanical ventilators are among the most sophisti-cated and
most expensive devices in the intensive care unit(ICU). In fact,
modern day ICUs can be characterized asthe common space where
mechanical ventilation of criti-cally ill patients is provided.
Ventilators have evolved fromsmall, pneumatically powered and
pneumatically controlleddevices to marvels of microprocessor
technology capableof closed-loop control.
The competition in the mechanical-ventilator market-place is
fierce and new generation ventilators are intro-duced frequently or
new innovations are added. Theseinnovations include new techniques,
new modes, new mon-itoring, new displays, and new trigger and cycle
variables.1-4Whether these represent real advances based on science
orsimply new bells and whistles to attract attention is oftenlost
in the lexicon of strategic marketing and device-spe-cific
education.
According to Wikipedia, innovation is:
a new way of doing something. It may refer toincremental,
radical, and revolutionary changes inthinking, products, processes,
or organizations. Adistinction is typically made between
inventionanidea made manifestand innovationideas
appliedsuccessfully. In many fields, something new mustbe
substantially different to be innovative, not aninsignificant
change (eg, in the arts, economics, busi-ness, and government
policy). In economics thechange must increase value, customer
value, or pro-ducer value. The goal of innovation is
positivechange, to make someone or something better.5 [ital-ics
mine]
It is an interesting insight, to include separate defini-tions
for innovation and invention. Clearly, invention hasbeen the
hallmark of mechanical ventilation evolution todate. While we might
all agree that changes in ventilatorshave made them safer, easier
to use, and more accurate inmonitoring the patient, there have been
few real innovations.
This paper reviews several new commercially availablefeatures of
mechanical ventilators. We will try to keep inmind the difference
between invention and true innova-tion. Not every new feature can
be covered. In an effort tobe complete, we discuss several features
that are quitedifferent from one another, including closed-loop
controlof ventilator weaning, automated measurement of func-tional
residual capacity (FRC), neural triggering/cycling,and information
display.
Closed-Loop Control of Weaning
Weaning from mechanical ventilation has undergonewholesale
changes over the last decade. Improved under-
standing of sedation, delirium, and weaning predictors hasled to
evidence-based development of ventilator-discon-tinuation
guidelines based on the weaning-readiness screen-ing and daily
spontaneous breathing trails.6 Discontinua-tion of mechanical
ventilation does not always involveweaning, which is the slow
withdrawal of support, butthe term weaning seems to be embedded in
the respira-tory-care lexicon and we will use it here.
History
Automated weaning is not a microprocessor-based in-vention; it
has roots in the introduction of mandatory minutevolume (MMV), in
1977, by Hewlett.7 With an Engstromventilator, Hewlett devised a
system to guarantee a pre-setminute volume (V E), based on a
mechanical system. Con-ceptually, MMV would allow the patient to
take over ven-tilation as lung mechanics improved and the patient
as-sumed the work of breathing.8-11 In that initial version ofMMV,
the V E was guaranteed by an automatic increase inthe set
respiratory rate.
Subsequent versions of MMV have used increases inpressure
support and/or the number of mandatory breathsto guarantee a
minimum V E. More recently, ventilatorshave been developed that
allow an increase of the rate orthe pressure-support level, based
on the underlying sup-port mode.
Though MMV was introduced over 30 years ago, wehave no evidence
of its efficacy. Claure et al studied MMVin infants and found that,
compared to intermittent man-datory ventilation, there were fewer
mandatory breaths,lower peak airway pressure, and smaller tidal
volume(VT).10 The only paper published on the use of MMVduring the
present decade was by Guthrie et al, who ran-domly applied
synchronized intermittent mandatory ven-tilation and MMV in
sequence for 2-hour periods. MMVresulted in fewer mandatory breaths
and more spontaneousbreaths, and, consequently, a lower mean airway
pres-sure.11 However, they did not evaluate the time to
discon-tinuation of mechanical ventilation.
Adaptive support ventilation (ASV) is also a techniquecapable of
automated weaning.12-15 ASV can also chooseinitial ventilator
settings and escalate ventilatory supportwhen ventilation targets
are not met. ASV uses a V E targetbased on predicted body weight
and a clinician-set per-centage of the predicted V E. As an
example, in adults,predicted V E is 0.1 L/kg/min (eg, a 70-kg
patient wouldreceive a V E of 7.0 L at a percent setting of 100%).
Chang-ing the percent V E setting to 150% would change thetarget to
10.5 L/min. VT and respiratory rate are portionedbased on the
minimum-work-of-breathing algorithm de-scribed by Otis et al.13 A
complete description of the nu-ances of ASV can be found
elsewhere.16,17
INNOVATIONS IN MECHANICAL VENTILATION
934 RESPIRATORY CARE JULY 2009 VOL 54 NO 7
-
ASV has been studied as a method to speed
ventilator-discontinuation after cardiac surgery.18-22 Those
studiesfound that ASV is safe and effective, but the time to
ex-tubation differed substantially. In patients who had fast-track
surgery, ASV shortened the time to extubation.18-21A recent trial
by Dongelmans and colleagues found nodifference in weaning time
between ASV and traditionalweaning in non-fast-track
cardiac-surgery patients.22 Theseconflicting data may well
represent the importance of choos-ing the appropriate V E target.
If the percent V E is too highthe patient may be prone to increased
pressure support,periodic breathing, and slower weaning. A more
aggres-sive approach that is tailored toward pushing the
patienttoward spontaneous breathing may be warranted when ASVis
applied as a weaning tool.
SmartCare/PS
Dojat et al initially described a closed-loop system
formechanical ventilation known as NeoGanesh in 1992.12Ganesh is
the Hindu elephant-deity often depicted with 4arms and riding a
mouse. Ganesh is known as the removerof obstacles, the patron of
arts and sciences, and the divaof intellect and wisdom.23 NeoGanesh
is perhaps aptlynamed for a technique to automatically withdraw
mechan-ical ventilation, although, as we will see, the
commercialversion only changes the pressure-support setting,
negat-ing the need for many arms. Introduced as
SmartCare/PS(Drager, Telford, Pennsylvania) in 2008, this system
usesmeasurements of the respiratory rate, VT, and partial pres-sure
of end-tidal carbon dioxide (PETCO2) to control thepressure-support
ventilation (PSV). Unlike previously in-troduced adaptive-control
modes, this system uses severalinputs in an effort to maintain
patient comfort in a definedrespiratory-rate range. Initially, the
system adjusts pres-sure support to maintain a respiratory rate of
1228 breaths/min, a VT above a minimum clinician-set threshold
(250300 mL), and PETCO2 below a clinician-set threshold(55 mm Hg
for normals, 65 mm Hg for patients withchronic obstructive
pulmonary disease). If ventilation re-mains within the prescribed
range for a predeterminedperiod, the system automatically reduces
the pressure-sup-port setting in an effort to facilitate
weaning.
The commercial version of SmartCare/PS identifies thepatients
breathing pattern as normal ventilation, insuffi-cient ventilation,
hypoventilation, hyperventilation, andtachypnea. Table 1 describes
the breathing patterns, theparameters that define those patterns,
and the response ofthe SmartCare/PS algorithm to patterns outside
the normalventilation parameters.
The first trial of SmartCare/PS was with 19 patients.The system
effectively maintained respiratory rate in theprescribed range for
up to 24 hours.24 The patients wereclassified as weanable or
unweanable based on results
of weaning parameters. The weanable patients remained inthe
prescribed comfort range for 95% of the duration ofventilation,
whereas the unweanable patients were in theprescribed comfort range
for 72% of the duration ofventilation. These initial results were
considered a successfor the technique. Dojat et al studied the
SmartCare/PSsystems ability to predict weaning success, compared to
aconventional weaning technique,24 and found that theSmartCare/PS
system had a positive predictive value of89%, compared to 77% for
conventional weaning.
Another study by Dojat et al studied 10 patients whorequired PSV
following acute lung injury (ALI).25 PSVwas implemented with a
modified Veolar ventilator (Ham-ilton, Reno, Nevada), which
continuously monitored pa-tient respiratory rate, VT, and airway
pressure. A stand-alone, mainstream CO2 monitor measured PETCO2.
Allvariables were measured every 10 seconds and averagedover a
2-min period.
The principle of control was to maintain ventilationwithin an
acceptable range by automatic adjustments ofpressure support.
Acceptable ventilation was defined asa respiratory rate between 12
and 28 breaths/min, a VT 250 mL ( 300 mL in patients 50 kg), and
PETCO2 55 mm Hg ( 65 mm Hg in patients with chronicobstructive
pulmonary disease). If the respiratory rate was2835 breaths/min and
VT and PETCO2 were in the accept-able ranges, the breathing pattern
was denoted as inter-mediate respiratory rate. In that situation,
the pressuresupport was increased by 2 cm H2O. If the respiratory
ratewas 35 breaths/min, the pattern was denoted as highrespiratory
rate and pressure support was increased by4 cm H2O. If respiratory
rate was 12 breaths/min, thepattern was denoted as low respiratory
rate and the pres-sure support was reduced by 4 cm H2O. If VT was
low( 250 mL) or PETCO2 was high, pressure support wasincreased by 2
cm H2O. Apnea caused the ventilator torevert to continuous
mandatory ventilation.
Additionally, if the ventilation was acceptable for 30 min-utes,
pressure support was reduced by 2 cm H2O if thepressure-support
level was 15 cm H2O, or by 4 cm H2Oif the pressure-support level
was 15 cm H2O and ven-tilation had been stable for 60 minutes. The
system toler-ated transient instabilities of 24 min, depending on
thepressure-support level. Tachypnea or inadequate ventila-tion
longer than 2 minutes caused the pressure-supportlevel to be
increased by 2 cm H2O if the pressure-supportlevel was 15 cm H2O,
or by 4 cm H2O if the pressure-support level was 15 cm H2O. If 3
consecutive obser-vation periods failed to have adequate
ventilation de-spite changes in the pressure-support level, the
ventilatordisplayed a message. When the pressure-support level
hadbeen 10 cm H2O for 2 hours and ventilation had beenstable, the
ventilator displayed a message suggesting aspontaneous breathing
trial and discontinuation of mechan-
INNOVATIONS IN MECHANICAL VENTILATION
RESPIRATORY CARE JULY 2009 VOL 54 NO 7 935
-
ical ventilation. Dojat et al found that closed-loop PSVresulted
in more time in acceptable ventilation and lesstime in critical
ventilation.25
Lellouche et al recently completed a randomized con-trolled
trial of SmartCare/PS versus traditional weaningin 144 patients, in
5 centers in Europe.27 They foundsignificant reductions in the time
to extubation, durationof mechanical ventilation until extubation,
time to suc-cessful extubation, total duration of mechanical
venti-lation, and ICU stay. Criticisms of that trial include
thefact that not all of the 5 centers used a weaning proto-col, and
at least 2 centers appear not to have used dailyspontaneous
breathing trials. In this instance it may bethat the local care
pattern was not up to the standard ofcare, which would bias the
results in favor of the com-
puter-directed weaning. However, given the knowledgewe have
about the time that elapses between the publi-cation of new
evidence and the translation of that evi-dence into practice, this
study may depict the real stan-dard across the world. Despite
best-practice evidence,protocols are not always implemented or
followed. Theintroduction of SmartCare/PS eliminates practice
vari-ations and implements the weaning protocol.
Bouadma and colleagues evaluated SmartCare/PS in 33patients.
They compared the times to recognition of wean-ing readiness by the
algorithm and by the physicians.28They found that the algorithm
detected weaning readinessearlier than the physicians in 17
patients, the physiciansdetected weaning readiness earlier than the
algorithm in 4patients, and detection was simultaneous in 11
patients.
Table 1. Ventilation Patterns, Parameters That Define Those
Patterns, and Response of the SmartCare/PS Algorithm
Breathing Pattern Variable Range Qualifier Response of
SmartCare/PSNormal ventilation f (breaths/min) 1530
1534No neurologic diseaseNeurologic disease
No change
VT (mL) 300 250
55 kg3655 kg
Pressure support decreased by 24 cm H2O at 15-min, 30-min, or
60-min intervals, based on currentlevel of pressure support
PETCO2 (mm Hg) 55 65
No COPDCOPD
Insufficient ventilation f (breaths/min) Acceptable None
Pressure support increased by 24 cm H2O,VT (mL) 300 Weight
depending on current level of pressure supportPETCO2 (mm Hg) 55
COPD
Hypoventilation f (breaths/min) 15 None Pressure support
increased immediately by 4 cm H2OVT (mL) 300 WeightPETCO2 (mm Hg)
55 COPD
Central hypoventilation f (breaths/min) 15 None No change. If
detected at 3 consecutive evaluations,VT (mL) 300 Weight alarm
requests that clinician evaluate patient.PETCO2 (mm Hg) 55 COPD
Tachypnea f (breaths/min) 3036 Neurologic disease Pressure
support increased immediately, by 24 cmVT (mL) 300 None H2O,
depending on current level f pressurePETCO2 (mm Hg) 55 None
support. If detected at 3 consecutive evaluations,
alarm requests that clinician evaluate patient.
Severe tachypnea f (breaths/min) 36 Neurologic disease Immediate
increase in pressure support. by 4 cmVT (mL) 300 Weight H2O.PETCO2
(mm Hg) 55 COPD If detected at 3 consecutive evaluations, alarm
requests that clinician evaluate patient.
Hyperventilation f (breaths/min) 15 Weight Pressure support
decreased immediately, by 4 cmVT (mL) 300 COPD H2OPETCO2 (mm Hg)
55
Unexplained hyperventilation f (breaths/min) 30 Neurologic
disease No change. If detected at 3 consecutive evaluations,VT (mL)
300 Weight alarm requests that clinician evaluate patient.PETCO2
(mm Hg) 45 COPD
f respiratory frequency per minuteVT tidal volumePETCO2
end-tidal CO2 pressureCOPD chronic obstructive pulmonary
disease(Adapted from Reference 26.)
INNOVATIONS IN MECHANICAL VENTILATION
936 RESPIRATORY CARE JULY 2009 VOL 54 NO 7
-
They concluded that SmartCare/PS was successful in man-aging PSV
for up to one week and proposed weaningreadiness earlier than
physicians. Though this study wasnot designed to evaluate duration
of mechanical ventila-tion, it suggests that the elimination of
practice variationmay be of benefit and could lead to earlier
discontinuationof mechanical ventilation.
Very recently, Rose et al compared SmartCare/PS
tonursing-directed weaning in an Australian intensive careunit.29
Australia does not include respiratory therapists intheir medical
system, and nursing care in that study was ata 1:1 ratio with
mechanically ventilated patients. The studyfound no differences in
time to extubation between thegroups, but noted a trend toward
earlier detection of wean-ing readiness by the experienced nursing
staff. Rose et alconcluded that SmartCare/PS did not reduce the
durationof weaning, in stark contrast to the finding by
Lelloucheand colleagues. Rose et al suggested that SmartCare/PShas
no obvious advantage over existing weaning methods,which consist of
frequent assessment of weaning readinessand titration of
ventilatory support by qualified and expe-rienced nurses in a
closed ICU model. These results givecredence to the criticisms
leveled at the European trial,where the care model was different.
However, a larger trialof current practice of a large variety of
ICU modelswould be required to determine when SmartCare/PS islikely
to be of benefit.
Invention or Innovation
Closed-loop ventilation holds the promise of reducingpractice
variation and responding to changes in patientcondition with a
speed and vigilance not usually availablefrom the ICU staff. There
is no doubt that SmartCare/PS isinnovative if judged against our
definition of making apositive change. As with the introduction of
all new sys-tems, we often learn more about our current practice
thanwe do about the new article under review.
SmartCare/PSchallenges earlier versions of closed-loop control by
al-lowing several minutes or longer in between changes, ascompared
to the typical breath-to-breath changes. Whilethis is still far
more frequent than the clinician can provide,we believe this less
aggressive approach may be a key tothe success. Despite the long
history of SmartCare/PS, theavailable literature remains quite
limited. Further researchshould help determine the scenarios where
computer con-trol of weaning will be beneficial.
Automated Measurement of Functional ResidualCapacity During
Mechanical Ventilation
History
Monitoring the mechanically ventilated patient has rou-tinely
revolved around the dynamic changes in airway
pressure, volume, and flow. Blood gases and
noninvasivemeasurements of oxygenation (transcutaneous
oxygen,oximetry) and ventilation (PETCO2, transcutaneous
carbondioxide) are also commonly used. Lung volume measure-ments
during mechanical ventilation are typically limitedto continuous
monitoring of VT and intermittent evalua-tions of vital capacity.
The latter measurement has beenconsidered as a parameter to
evaluate cough effectivenessprior to discontinuation of mechanical
ventilation.
FRC measurement during mechanical ventilation hasbeen made with
helium dilution, nitrogen washout, andtracer gases (sulfur
hexafluoride).30-35 In all of those stud-ies the systems were
homemade amalgamations assembledby the research team to evaluate
the effects of positiveend-expiratory pressure (PEEP) on
end-expiratory lung vol-ume.
East and colleagues produced the majority of work inthis arena,
following a research plan with the possibility ofproducing a
commercially available system.31,36,37 In theirseries of
investigations they found that sulfur hexafluorideFRC measurements
were accurate and aided in closed-loop control of PEEP in
traditional ventilation and inde-pendent lung ventilation. But
sulfur hexafluoride does nothave regulatory approval for human use,
and this and otherissues precluded further advancement of this
technique.
Automated FRC Measurements Using NitrogenWashout
Measurement of FRC in the pulmonary function labo-ratory can be
accomplished via body plethysmography,helium dilution, and nitrogen
washout. In each case, acooperative, spontaneously breathing
patient is the sub-ject. Helium dilution can be used in
mechanically venti-lated patients, but requires the addition of the
helium cyl-inder and a helium analyzer. Additionally, the effects
ofhelium on ventilator performance are quite variable andcan lead
to ventilator malfunction.38-40
Nitrogen washout is the simplest FRC measurementmethod in the
mechanically ventilated patient, as there isno need for additional
equipment or gases. However, thenitrogen-washout method requires
accurate measurementof inspired and expired oxygen and carbon
dioxideinessence a metabolic monitoring module must be integral
tothe ventilator. Under normal conditions FRC is the resultof the
opposing forces of lung and chest wall compliance.In the
mechanically ventilated patient FRC is a function ofthose forces,
air-flow obstruction, and the application ofPEEP. It may be more
appropriate to consider FRC in theventilated patient as
end-expiratory lung volume (EELV).Equally important is the fact
that the use of PEEP in theventilated patient does not attempt to
normalize FRC topredicted values. Measurements in the ventilated
patientmore likely are able to evaluate changes in EELV created
INNOVATIONS IN MECHANICAL VENTILATION
RESPIRATORY CARE JULY 2009 VOL 54 NO 7 937
-
by alterations in therapy. For example, following a suc-cessful
recruitment maneuver the EELV measurement mayincrease, compared to
pre-procedure values. Similarly, fol-lowing delivery of a
bronchodilator in a patient with in-trinsic PEEP (auto-PEEP), EELV
may decrease.
Principle of Measurement
The calculation of EELV is based on a step-change inthe fraction
of inspired oxygen (FIO2) and the assumptionthat in the ventilator
system, N2 is the balance gas in thesystem, where N2 1 FIO2. The
following is a simplifiedexplanation of the measurement of EELV.
Inspired andexpired concentrations of N2 are not measured, but
deter-mined from the presence of O2 and CO2. Inspired N2 (FIN2)and
end-tidal N2 (FETN2) are calculated as:
FIN2 1 FIO2
and
FETN2 1 FETO2 FETCO2
where FETO2 is the fraction of end-tidal oxygen.
Inspiredalveolar VT (VTalv(I)) and expired alveolar VT
(VTalv(E))are calculated with energy-expenditure measurements
foroxygen consumption (V O2) and carbon dioxide production(V
CO2):
V O2 V CO2 / RQ
where RQ is respiratory quotientand
VTalvE V CO2/FETCO2 f
and
VTalvI VTalvE V O2 V CO2 / f
where f is respiratory frequency per minute.Remember, that the
respiratory exchange ratio results in
expired volumes that are typically smaller than inspiredvolumes,
because the volume of oxygen consumed ex-ceeds the volume of carbon
dioxide produced. The single-breath nitrogen volumes associated
with expiration andinspiration are then calculated as:
VEN2 FETN2 VTalv(E)
and
VIN2 FIN2 VTalvI
The resulting values are then used to calculate the changein
nitrogen concentration during a single breath:
VN2 VEN2 VIN2
At this point in the measurement, a baseline determina-tion is
made, which includes the values of V O2, V CO2, andFETN2 baseline.
During this time the device assumesthat the V O2 and V CO2 remain
constant throughout themeasurement, but that assumption may not
always holdtrue: changes in V O2 and V CO2 associated with
activity,agitation, or changes in patient condition can cause
errors.A step-change in FIO2 is then accomplished and the EELVis
calculated as:
EELV VN2/FETN2
where FETN2 is the last recorded value following the step-change
in FIO2. The breath-to-breath changes are calcu-lated over
approximately 20 breaths.
EELV
breathsVN2
baseline FETN2 last FETN2
FRC INview
The EELV measurements performed by the EngstromCarestation (GE
Healthcare, Waukesha, Wisconsin),known as FRC INview can be made on
demand or pre-programmed to make a series of measurements. Each
EELVdetermination is made with 2 measurements in a series of20
breaths. The step-change in FIO2 required to determineEELV is
typically 10%. Accuracy is best at FIO2 of 0.40.65. Prior to the
step-change the patient should be stableand the FIO2 should be
constant for at least 5 minutes.Because patient stability is
required for accuracy, and VO2and VCO2 are assumed to stay
constant, the device willterminate the measurement if there is any
change in theventilation mode, ventilator settings, or performance
of aprocedure (eg, airway suctioning). Following the
EELVmeasurement, the FIO2 returns to the original setting. EELVis
displayed numerically and graphically over the 20-breathmeasurement
period. Table 2 lists the conditions and/oralterations in patient
status that can affect the accuracy ofthe EELV measurement.
Accuracy and Application
The current EELV measurement technique (FIO2 step-change then
nitrogen washout) was pioneered by Stenqvist
INNOVATIONS IN MECHANICAL VENTILATION
938 RESPIRATORY CARE JULY 2009 VOL 54 NO 7
-
and colleagues,41,42 and several early studies with lungmodels,
animals, and patients found the method accurateand
reproducible.42-44 However, the evidence regardingthe value of EELV
in critically ill patients remains sparse.Bikker et al recently
compared EELV in mechanicallyventilated patients to reference
equations for FRC.45 In 3groups (one group with normal lungs, one
group withprimary lung disorders, and one group with secondarylung
disorders) they used 3 PEEP settings while measuringEELV and
pulmonary compliance. They found that in me-chanically ventilated
and sedated patients, EELV was mark-edly lower than the predicted
sitting FRC values. Impor-tantly, they identified the critical
issue in the use of EELVmeasurements to guide PEEP: PEEP-induced
changes inEELV may not only detect recruitment or derecruitment:the
change can result from inflation or deflation of alreadyventilated
lung units. More simply, an EELV increase mayrepresent
overdistention of lung units that were alreadyopen, as compared to
recruitment of previously closedlung units. The EELV changes can be
similar, but theeffect on the patient is markedly different. Bikker
et al
further suggested that the use of EELV measurements
inconjunction with lung-compliance measurements is morelikely to
determine which EELV changes represent re-cruitment and which
represent overdistention. We wouldadd that, since metabolic
measurements are integral to thenitrogen washout system, the
addition of volumetric cap-nography and dead-space calculations
along with EELVand compliance may be the optimal system for
adjustingPEEP. This concept is supported by preliminary work
byRylander et al and others.46-48
Invention or Innovation
While the technique of monitoring EELV using the
ni-trogen-washout method is clearly inventive, clinical
dataregarding the usefulness of the technique are lacking.
Ad-mittedly, this is predominantly a function of its very
recentintroduction to the market. We predict that, when coupledwith
compliance measurement and capnography, this tech-nique will
provide data not currently available and willobtain the definition
of innovative.
Neurally Adjusted Ventilatory Assist
History and Principle of Operation
Neurally adjusted ventilatory assist (NAVA) was intro-duced by
Sinderby in 1999, and introduced on the Servo-iventilator (Maquet,
Bridgewater, New Jersey) in 2007.49NAVA is a mode of partial
ventilatory support that usesthe electrical activity of the
diaphragm to control patient-ventilator interaction. The electrical
activity of the dia-phragm represents the final neural output of
the respiratorycenters to the diaphragm and is therefore able to
bothtrigger and cycle a breath. Breaths remain
pressure-con-trolled. With NAVA the pressure delivered during
inspi-ration is proportional to the electrical activity of the
dia-phragm. The pressure level can be adjusted based on
patienteffort and the proportionality setting. Depending on
thepatients response to the delivered pressure, an increase inthe
NAVA level may increase the delivered pressure (ifelectrical
activity of the diaphragm is unchanged), or itcould suppress
electrical activity of the diaphragm andinstead deliver a constant
level of assist, or some levelin between.
As NAVA is relatively new, a comparison to PSV maybe helpful to
highlight the differences. PSV delivers afixed pressure,
independent of changes in respiratory drive.PSV is
patient-triggered and uses either flow or pressuremeasurements. The
absolute pressure is set by the clinicianand is unaffected by
alterations in patient effort. PSV isnormally cycled when flow
reaches a predetermined per-centage of the initial peak flow. PSV
can also be cycled bytime (in the presence of leaks) or pressure
(in the presence
Table 2. Factors That Can Affect the Accuracy of the
End-Expiratory-Lung-Volume Measurement
Factor Cause
Respiratory rate 35breaths/min
Agitation, anxiety
Variable respiratory rate Head injury, metabolic syndromesLarge
variations in VT Head injury, disorders of ventilation
controlFebrile SepsisChange in ventilator mode
or settingsSettings result in alterations of EELV
during the measurementPosition change Position alterations of
EELV during
the measurementProcedure Suctioning, nebulization. Changes
in
EELV and gas concentrations altermeasurement.
Leak Circuit leak, cuff leak, chest-tube leakalter inspiratory
and expiratoryvolume measurements, and gasesare lost to the
measurement system.
Humidity Excessive humidity can decrease theaccuracy of the flow
sensor. Aheat-and-moisture exchanger ispreferred during
measurement.
High FIO2 High FIO2 alters accuracy of V O2, andV CO2
measurements and makes thestep change in FIO2 more difficult.
Elevated bias flow Decreases accuracy of volumemeasurements
VT tidal volumeEELV end-expiratory lung volumeFIO2 fraction of
inspired oxygenV O2 oxygen consumptionV CO2 carbon dioxide
production
INNOVATIONS IN MECHANICAL VENTILATION
RESPIRATORY CARE JULY 2009 VOL 54 NO 7 939
-
of patient expiratory effort).50 When the pressure-supportlevel
is set too high, a large VT may be delivered, whichcan cause
hyperinflation and missed triggers because ofauto-PEEP.51
NAVA requires an esophageal catheter that measuresthe electrical
signal to the diaphragm (Edi). The Edi cath-eter is similar to a
standard nasogastric tube in diameterand length, but has a series
of electrodes that measure theEdi. The electrodes appear as black
circumferential stripes.There is a reference electrode (the most
proximal elec-trode) and 9 measuring electrodes. NAVA is triggered
bythe Edi signal, and the sensitivity can be set, like
pressuretriggering or flow triggering. The minimum Edi signal
isrecorded and the sensitivity is based on an increase in thesignal
above that reference value. NAVA cycling occurswhen Edi drops to
approximately 70% of its peak value.During NAVA, pressure
triggering and flow triggeringremain as redundant systems that
activate if the catheter isnot placed properly or if airway
triggering is sensed first.A proportionality factor (known as the
NAVA level) de-termines the delivered pressure for a given
electrical ac-tivity of the diaphragm amplitude (ie, cm H2O per
unit ofelectrical activity of the diaphragm).
Based on the principle of operation then, NAVA mayavoid the
overinflation associated with PSV. Since NAVAuses the electrical
activity of the diaphragm to trigger andcycle, patient-ventilator
interaction should be improved.NAVA should be unaffected by the
most common factorsthat confound traditional triggering: auto-PEEP
and leaks(in noninvasive ventilation or uncuffed endotracheal
tubes).
Application
The Edi catheter is placed with guidance from the Edicatheter
positioning screen, which displays a series of 4electrocardiograph
leads and the Edi signal. Proper posi-tioning of the Edi catheter
is assumed when the Edi signaland the corresponding second and
third electrocardiographsignals display in blue. Blue signals from
the first or fourthlead suggest the Edi catheter needs to be either
pulled back oradvanced forward. This system simplifies placement,
but evenwith perfect placement the signal can be absent or very
lowin the presence of excessive sedation, muscle relaxants,
hy-perventilation, high PEEP, or a neural disorder.
If used simultaneously for Edi measurement and naso-gastric
drainage, the conflicting goals can make the cath-eter fail to
optimally perform either one of those functions.In our experience
the catheter is not easily placed, nor isthe signal always found.
We have also seen that, thoughinitial placement can achieve
excellent signal quality, pa-tient movement and routine ICU care
can result in signalloss. These are preliminary observations and
demonstratethat, despite proper engineering design and controls,
clin-ical factors can confound implementation.
Setting the NAVA level is accomplished by selectingthe Neural
Access and NAVA preview screen. This screenallows the clinician to
set the PEEP, FIO2, and Edi triggerlevel (in microvolts). The NAVA
pressure-support andbackup ventilation settings are set by the
clinician andensure safe ventilation in the absence of patient
effort orloss of communication with the Edi catheter. The NAVAlevel
is set by observing the airway pressure, delivered VT,and Edi. The
initial NAVA level is generally that whichreduces Edi and
corresponds to patient comfort.
Current Literature
Since the initial paper in 1999,49 a series of
originalinvestigations have been published, all of which were
au-thored or co-authored by the NAVA inventors. Allo et alevaluated
the effects of NAVA on rabbits with ALI. Theyfound that ALI caused
a vagally mediated atypical dia-phragm-activation pattern in
spontaneously breathing an-imals. The addition of PEEP restored
phasic activity, andNAVA efficiently maintained respiratory-muscle
unload-ing while delivering safe VT.52 Those experiments lasted4
hours.
Beck and colleagues found that NAVA reduced the num-ber of
missed triggers, compared to PSV, and reduced thetransdiaphragmatic
pressure-time product and Edi in a rab-bit model.53 At an excessive
pressure-support level, only66% of efforts triggered the
ventilator, compared to 100%with NAVA. The large VT resulted in
auto-PEEP, missedtriggers, and failure to adequately unload the
respiratorymuscles. In volunteers, Sinderby et al found that
NAVAunloaded the respiratory muscles and that Edi was detect-able
and triggered the ventilator even at low amplitudes.54
In another animal study, Beck and colleagues found thatNAVA can
be successfully applied with noninvasive ven-tilation and that
synchrony was maintained in the presenceof large leaks.55 Moerer et
al found that NAVA success-fully triggered and cycled the
ventilator during the use ofa helmet (noninvasive ventilation
interface) in normal sub-jects.56
More recently, studies with adult and neonatal patientshave been
published.57,58 Brander and colleagues studied15 patients with ALI
and systematically increased theNAVA level over a 3-hour period,
which reduced the re-spiratory drive, unloaded the respiratory
muscles, and al-lowed the clinician to identify an assist level
that resultedin sustained unloading, appropriate VT (5.47.2 mL/kg
ofpredicted body weight), and normal hemodynamics.57
Inlow-birth-weight infants, Beck et al found that NAVAimproved
patient-ventilator synchrony, even in the pres-ence of leaks around
the endotracheal tube (ET).58
INNOVATIONS IN MECHANICAL VENTILATION
940 RESPIRATORY CARE JULY 2009 VOL 54 NO 7
-
Invention or Innovation
NAVA is clearly an innovative design. Detecting
respira-tory-muscle effort earlier than changes occur at the airway
isa positive change (Fig. 1). NAVA remains in a very earlystage of
development, and there have been no long-term stud-ies to determine
if NAVA has outcome benefits. However,we know that patients with
greater patient-ventilator asyn-chrony have worse outcomes and
longer duration of mechan-ical ventilation.59 What we dont know is
if that relationshipis one of association or causality. If missed
triggers and asyn-chrony lead to prolonged mechanical ventilation,
then a sys-tem that reduces missed triggers might improve
outcome.However, at present it is more likely that the severity
ofillness results in asynchrony, not the other way around.
Over the last 20 years traditional ventilator triggeringhas
improved markedly, to the point where cardiacchanges in
intrathoracic pressure can cause triggering.60On several occasions
cardiac activity has been responsiblefor triggering in patients
with brain death and delayedorgan donation.61 This means that in
many patients trig-gering is not the issue. Thille et al found that
simply re-ducing the pressure-support level significantly
reducedmissed triggers.51
NAVA would seem to be ideal when factors confoundtraditional
triggering, the most likely of these being in thepresence of
hyperinflation and/or leaks. This suggests thatin a neonate with an
uncuffed ET, particularly at a higherrespiratory frequency, NAVA
should be advantageous.Similarly, NAVA should have value in the
patient withsubstantial auto-PEEP. Finally, in a patient with leaks
andhyperinflation (eg, a patient with chronic obstructive
pul-monary disease undergoing noninvasive ventilation),NAVA would
seem ideal, but the NAVA requirement of anasogastric tube may be
counter to the noninvasive goal.Gaining the patients trust and
assuring patient comfortand cooperation are key to successful
initiation of nonin-
vasive ventilation. How placement of a nasogastric tubewould
impact this is unknown. We await further researchon NAVA and look
forward to definitive answers aboutthe proposed advantages.
Information Presentation
History and Principles
Every elementary school student has heard the adagethat a
picture is worth a thousand words, which is based onthe Chinese
proverb, A pictures meaning can express tenthousand words.62 This
concept was embraced by thegroup led by Westenskow at the
University of Utah63 andcommercially implemented as the Ventilation
Cockpiton the G5 ventilator (Hamilton Medical, Reno, Nevada).The
goal of information presentation is to build on thesuccess of
graphical displays of pressure, volume, andflow into pictorial
representations of common clinical con-ditions.64,65 Theoretically
these displays might improvesafety by allowing faster detection of
untoward events.Given the current concern regarding errors in
medicine, asystem that alerts the clinician to changes in patient
con-dition seems ideal.
Current Literature
Drews and Westenskow reviewed data regarding theuse of graphical
displays to depict changes in cardiovas-cular function, compared to
numeric displays, and foundthat graphical displays can improve
situational awareness,enhance clinician performance, and improve
patient safe-ty.63 Wachter et al studied 19 clinicians ability to
manage5 scenarios: obstructed ET; endobronchial intubation;
au-to-PEEP; hypoventilation; and a normal condition.64 Thegraphical
pulmonary display changed shape and color ac-cording to measures of
pulmonary function. The displaywas generated from data from a
respiratory monitor and apatient simulator. The monitor measured
airway pressure,air flow, respiratory rate, PETCO2, and VT. Airway
resis-tance and total lung compliance were calculated from
mea-surements of airway pressure, volume, and flow. The
par-ticipants were assigned alternately to one of 2
groups:intervention pulmonary display present, or control
display.Both groups had access to standard displays of
numericaldata. In a simulated operating room, with a simulated
pa-tient, the participants assumed the role of an anesthesiol-ogist
called because of an unspecified problem midwaythrough a surgery.
At completion of each scenario, thevolunteers responded to a
National Aeronautics and SpaceAdministration (NASA) Task Load Index
questionnaire toassess perceived work load and performance.
Fig. 1. Comparison of trigger signals in traditional mechanical
ven-tilation and neurally adjusted ventilatory assist
ventilation.
INNOVATIONS IN MECHANICAL VENTILATION
RESPIRATORY CARE JULY 2009 VOL 54 NO 7 941
-
Figure 2 shows the graphical displays used in the sim-ulation.
The graphical displays resulted in earlier detectionof the
airway-associated issues in 2 of the 4 abnormalconditions. During
the normal scenario simulation, 3 cli-nicians using the graphical
display, and 5 clinicians usingthe conventional display gave
unnecessary treatments. Par-ticipants reported significantly lower
subjective work loadwith the graphical display during the
obstructed-ET andthe auto-PEEP scenarios.
Implementation
The Hamilton G5 uses 2 main displays to reflect changesin
pulmonary compliance and airway resistance. Figure 3shows the
normal display. This display is known as thedynamic lung. In the
normal display the lungs havesmooth, rounded edges and the airways
appear pink and ofuniform diameter. As lung compliance falls, the
lungs takeon an angular appearance. Figure 4 demonstrates a fall
incompliance to 50 mL/cm H2O. The lungs have 6 sides andno longer
fill the full volume. This display denotes thereduction in
compliance and loss of lung volume. Figure 5demonstrates a further
reduction in compliance to 30 mL/cm H2O and the lungs are reduced
to 5 sides. At the lowestlung compliance ( 20 mL/cm H2O) the lungs
have only4 sides and appear the most angular (Fig. 6). These are
incontrast to the display of hyperinflation or overdistention
(Fig. 7). Figure 7 represents a pulmonary compliance of 75 mL/cm
H2O. The lungs are overly rounded andexceed the normal boundaries
of the lung display.
Figures 810 show changes in airway resistance. Nor-mal airway
resistance is displayed as pink airways withuniform walls (see Fig.
8). As resistance increases (see
Fig. 2. Examples of the pulmonary display used by Wachter et
al.64 The display anatomically represents the bellows, airway,
lungs, and inspiredand expired gas. In each part of the figure, the
left box represents the fraction of inspired oxygen (FIO2); the
middle box is similar to the bellowsof the ventilator andmoves
along the vertical axis, representing tidal volume; and the right
box represents end-tidal carbon dioxide (PETCO2). A: Theobstructed
endotracheal tube event; the upper airway has black restrictive
fingers. B: The endobronchial intubation event has a
thickenedcompliance cage surrounding the lung icon. C: The
intrinsic positive end-expiratory pressure event shows an
overinflated lung icon, whichextends past the normal boundary of
the lung icon and the compliance cage. D: The hypoventilation event
is shown with a short bellows icon thatrepresents low tidal volume.
E: The normal event: all parameters are within normal limits. (From
Reference 63, with permission.)
Fig. 3. Dynamic lung display on the Hamilton G5 ventilator,
demon-stratingnormalcompliance.The
lungsaresmooth,withoutanyangles.
INNOVATIONS IN MECHANICAL VENTILATION
942 RESPIRATORY CARE JULY 2009 VOL 54 NO 7
-
Fig. 9), the color becomes deeper and the colored portiondoes
not fill the gray space that outlines the airways. Whenairway
resistance increases, the color becomes deeper andthe gray space is
noticeably larger (see Fig. 10). Thesechanges alert the clinician
of changes in pulmonary compli-ance and airway resistance without
numerical values or theneed for clinicians to remember the normal
ranges for thesevalues.
Figure 11 is a screen shot from the ventilator. Note
thedepiction of increased airway resistance. This is confirmedby
the numerical display, which shows an airway resis-tance of 21 cm
H2O/L/s. Alerted to the change, the clini-cian found that the
heat-and-moisture exchanger was oc-cluded with secretions. Removal
of the heat-and-moistureexchanger returned the display and
numerical values tonormal (Fig. 12)
Fig. 4. Dynamic lung display on the Hamilton G5 ventilator,
dem-onstrating a reduction in compliance to 50 mL/cm H2O. Note
theangular appearance and 6-sided lung.
Fig. 5. Dynamic lung display on the Hamilton G5 ventilator,
dem-onstrating a reduction in compliance to 30 mL/cm H2O. Note
theangular appearance and 5-sided lung.
Fig. 6. Dynamic lung display on the Hamilton G5 ventilator,
dem-onstrating a reduction in compliance to 20 mL/cm H2O. Note
theangular appearance and 4-sided lung.
Fig. 7. Dynamic lung display on the Hamilton G5 ventilator,
dem-onstrating an increase in compliance to 100 mL/cm H2O. Note
theballooning appearance and rounded lungs stretching beyond
thenormal area.
INNOVATIONS IN MECHANICAL VENTILATION
RESPIRATORY CARE JULY 2009 VOL 54 NO 7 943
-
Clearly, increases in airway resistance and decreasesin
compliance are also associated with increases in peakinspiratory
pressure and alterations in the shape of thepressure and flow
waveforms. Importantly, a change inthe graphical display that
denotes reduced pulmonarycompliance does not differentiate between
pneumotho-
rax and a distended abdomen. Both chest-wall and lungcompliance
can affect the display. Similarly, a changein airway resistance
cannot distinguish between a kinkedET and bronchospasm. Despite
these criticisms, giventhe overload of information and alarms in
the ICU,simplified graphical displays may play a role in
alertingclinicians of changes in patient condition earlier
thantraditional measures.
Invention or Innovation
Though a simple invention, compared to the otherfeatures we have
discussed, we believe that these graph-ical displays are
innovative. The presentation of infor-mation in a simple picture
can enhance patient safetyand reduce time to recognition of adverse
events. Oneadvantage is that such graphical display is applicable
toevery ventilated patient, whereas some other featuresare used
only for weaning, certain disease states, orimproving
synchrony.
Summary
In our nearly 30 years of studying mechanical venti-lation, many
new techniques have been introduced. Inmost cases these were
brought to market with little re-search or scientific background.
It is refreshing to seethe newest techniques being developed on
stronger sci-entific footing. Each of the innovations we
describehere is in fact early enough in development that the
jury
Fig. 8. Dynamic lung display on the Hamilton G5 ventilator,
dem-onstratingnormal airway resistance.Note thepink,
uniformairways.
Fig. 9. Dynamic lung display on the Hamilton G5 ventilator,
dem-onstrating an increase in airway resistance to 15 cm H2O/L/s.
Notethe deepening color of the airway and the narrowing within
thenormal outlined gray space.
Fig. 10. Dynamic lung display on the Hamilton G5 ventilator,
dem-onstrating an increase in airway resistance to 25 cm H2O/L/s.
Notethe deepening color of the airway and the narrowing within
thenormal outlined gray space.
INNOVATIONS IN MECHANICAL VENTILATION
944 RESPIRATORY CARE JULY 2009 VOL 54 NO 7
-
is still out on the role and efficacy of each. Given ourpast
skepticism about new ventilator techniques, thiscurrent group of
offerings gives us hope that advantageswill be proven with further
study.66
REFERENCES
1. Johannigman JA, Barnes SA, Muskat P, Davis K, Beck G,
BransonRD. Autonomous control of oxygenation. J Trauma 2008;64(4
Suppl):S295-S301.
2. Johannigman JA, Barnes SA, Muskat P, Davis K, Branson
RD.Autonomous control of ventilation. J Trauma 2008;64(4
Suppl):S302-S320.
3. Johannigman JA, Branson RD, Barnes SL, Muskat P, Beck G.
Closedloop control of oxygenation and ventilation. Int J
Gravitational Physiol2007;14:P35-P38.
4. Burns KE, Lellouch FA, Lessard MR. Automating the weaning
pro-cess with advanced closed loop systems. Intensive Care Med
2008;34(10):1757-1765.
5. Wikipedia contributors. Innovation. Wikipedia. The free
encyclope-dia. http://en.wikipedia.org/wiki/innovation. Accessed
May 12, 2009.
Fig. 11. Screen shot demonstrating the plethora of information
available on a typical ventilator. Note the increased airway
resistance bothgraphically and numerically. (Courtesy of Paul
Gabarini MSc RRT.)
Fig. 12. Same simulated patient as in Figure 11, following
removal of a plugged heat-and-moisture exchanger. (Courtesy of Paul
GabariniMSc RRT.)
INNOVATIONS IN MECHANICAL VENTILATION
RESPIRATORY CARE JULY 2009 VOL 54 NO 7 945
-
6. MacIntyre NR, Cook DJ, Ely EW Jr, Epstein SK, Fink JB,
HeffnerJE, et al. Evidence-based guidelines for weaning and
discontinuingventilatory support: a collective task force
facilitated by the Amer-ican College of Chest Physicians; the
American Association for Re-spiratory Care; and the American
College of Critical Care Medicine.Chest 2001;120(6
Suppl):375S-395S.
7. Hewlett AM, Platt AS, Terry VG. Mandatory minute volume: a
newconcept in weaning from mechanical ventilation. Anaesthesia
1977;32(2):163-169.
8. Ravenscroft PJ. Simple mandatory minute volume. Anaesthesia
1978;33(3):246-249.
9. Davis S, Potgieter PD, Linton DM. Mandatory minute volume
wean-ing in patients with pulmonary pathology. Anaesth Intensive
Care1989;17(2):170-174.
10. Claure N, Gerhardt T, Hummler H, Everett R, Bancalari E.
Com-puter-controlled minute ventilation in preterm infants
undergoingmechanical ventilation. J Pediatrics
1997;131(6):910-913.
11. Guthrie SO, Lynn C, Lafleur BJ, Donn SM, Walsh WF. A
crossoveranalysis of mandatory minute ventilation compared to
synchronizedintermittent mandatory ventilation in neonates. J
Perinatol 2005;25(10):643-646.
12. Dojat M, Brochard L, Lemaire F, Harf A. A knowledge based
systemfor assisted ventilation of patients in intensive care units.
Int J ClinMonit Comput 1992;9(4):239-250.
13. Otis AB, Fenn WO, Rahn H. Mechanics of breathing in man. J
AppPhysiol 1950;2(11):592-607.
14. Laubscher TP, Frutiger A, Fanconi S, Jutzi H, Brunner JX.
Auto-matic selection of tidal volume, respiratory frequency and
minutevolume in intubated ICU patients as startup procedure for
closed-loop controlled ventilation. Intern J Clin Monit Comp
1994;11(1):19-30.
15. Laubscher TP, Frutiger A, Fanconi S, Brunner JX. The
automaticselection of ventilation parameters during the initial
phase of me-chanical ventilation. Intensive Care Med
1996;22(3):199-207.
16. Campbell RS, Branson RD, Johannigman JA. Adaptive support
ven-tilation. Respir Care Clin N Am 2001;7(3):425-440.
17. Brunner JX, Iotti GA. Adaptive support ventilation (ASV).
MinervaAnaestesiol 2002;68(5):365-368.
18. Sulzer CF, Chiolero R, Chassot PG, Mueller XM, Revelly JP.
Adap-tive support ventilation for fast tracheal extubation after
cardiac sur-gery: a randomized controlled study. Anesthesiology
2001;95(6):1339-1345.
19. Cassina T, Chiolero R, Mauri R, Revelly JP. Clinical
experiencewith adaptive support ventilation for fast-track cardiac
surgery. J Car-diothorac Vasc Anesth 2003;17(5):571-575.
20. Petter AH, Chiolero RL, Cassina T, Chassot PG, Muller XM,
Rev-elly JP. Automatic respirator/weaning with adaptive support
ven-tilation: the effect on duration of endotracheal intubation and
patientmanagement. Anesth Analg 2003;97(6):1743-1750.
21. Gruber PC, Gomersall CD, Leung P, Joynt GM, Ng SK, Ho
KM,Underwood MJ. Randomized controlled trial comparing
adaptive-support ventilation with pressure-regulated
volume-controlled ven-tilation with automode in weaning patients
after cardiac surgery.Anesthesiology 2008;109(1):81-87.
22. Dongelmans DA, Veelo DP, Paulus F, de Mol BA, Korevaar
JC,Kudoga A, et al. Weaning automation with adaptive support
venti-lation: a randomized controlled trial in cardiothoracic
surgery pa-tients. Anesth Analg 2009;108(2):565-571.
23. Wikipedia contributors. Wikipedia. The free encyclopedia.
Ganesha.http://en.wikipedia.org/wiki/ganesha. Accessed May 12,
2009.
24. Dojat M, Harf A, Touchard D, Laforest M, Lemaire F, Brochard
L.Evaluation of a knowledge based system providing ventilatory
man-agement and decision for extubation. Am J Respir Crit Care
Med1996;153(3):997-1004.
25. Dojat M, Harf A, Touchard D, Lemaire F, Brochard L.
Clinicalevaluation of a computer controlled pressure support mode.
Am JRespir Crit Care Med 2000;161(4 Pt 1):1161-1166.
26. Burns KE, Lellouche F, Lessard MR. Automating the weaning
pro-cess with advanced closed loop systems. Intensive Care Med
2008;34(10):1154-1160.
27. Lellouche F, Mancebo J, Jolliet P, Roeseler J, Schortgen F,
Dojat M,et al. A multicenter randomized trial of computer driven
protocolizedweaning from mechanical ventilation. Am J Respir Crit
Care Med2006;174(8):894-900.
28. Bouadma L, Lellouche F, Cabello B, Taille S, Mancebo J,
Dojat M,Brochard L. Computer-driven management of prolonged
mechanicalventilation and weaning: a pilot study. Intensive Care
Med 2005;31(10):1446-1450.
29. Rose L, Presneill JJ, Johnston L, Cade JF. A randomized
controlledtrial of conventional versus automated weaning from
mechanicalventilation using SmartCare PS. Intensive Care Med
2008;34(10):1788-1795.
30. Hedenstierna G. The recording of FRCis it of importance and
canit be made simple? Intensive Care Med 1993;19(7):365-366.
31. East TD, Wortelboer PJ, van Ark E, Bloem FH, Peng L, Pace
NL, etal. Automated sulfur hexafluoride washout functional residual
ca-pacity measurement system for any mode of mechanical
ventilationas well as spontaneous respiration. Crit Care Med
1990;18(1):84-91.
32. Di Marco F, Rota SL, Milan B, Stucchi R, Centanni S,
Brochard L,Fumagalli R. Measurement of functional residual capacity
by heliumdilution during partial support ventilation: in vitro
accuracy and invivo precision of the method. Intensive Care Med
2007;33(12):2109-2115.
33. Fretschner R, Deusch H, Weitnauer A, Brunner JX. A simple
methodto estimate functional residual capacity in mechanically
ventilatedpatients. Intensive Care Med 1993;19(7):372-376.
34. Patroniti N, Bellani G, Manfio A, Maggioni E, Giuffrida A,
Foti G,Pesenti A. Lung volume in mechanically ventilated patients:
mea-surement by simplified helium dilution compared to quantitative
CTscan. Intensive Care Med 2004;30(2):282-289.
35. Zinserling J, Wrigge H, Varelmann D, Hering R, Putensen C.
Mea-surement of functional residual capacity by nitrogen washout
duringpartial ventilator support. Intensive Care Med
2003;29(5):720-726.
36. East TD, Veen JC, Pace NL, McJames S. Functional residual
capac-ity as a noninvasive indicator of optimal positive
end-expiratorypressure. J Clin Monit 1988;4(2):91-98.
37. East TD, Veen JC, Jonker TA, Pace NL, McJames S.
Computer-controlled positive end-expiratory pressure titration for
effective ox-ygenation without frequent blood gases. Crit Care Med
1988;16(3):252-257.
38. Brown MK, Willms DC. A laboratory evaluation of 2
mechanicalventilators in the presence of helium-oxygen mixtures.
Respir Care2005;50(3):354-360.
39. Oppenheim-Eden A, Cohen Y, Weissman C, Reuven P. The effect
ofhelium on ventilator performance: study of five ventilators and
abedside pitot tube spirometer. Chest 2001;120(2):582-588.
40. Venkataraman ST. Heliox during mechanical ventilation.
Respir Care2006;51(6):632-639.
41. Stenqvist O, Olegrd C, Sndergaard S, Odenstedt H, Karason
S,Lundin S. Monitoring functional residual capacity (FRC) by
quan-tifying oxygen/carbon dioxide fluxes during a short apnea.
ActaAnaesthesiol Scand 2002;46(6):732-739.
42. Olegrd C, Sndergaard S, Houltz E, Lundin S, Stenqvist O.
Esti-mation of functional residual capacity at the bedside using
standardmonitoring equipment: a modified nitrogen washout/washin
tech-nique requiring a small change of the inspired oxygen
fraction. AnesthAnalg 2005;101(1):206-212.
INNOVATIONS IN MECHANICAL VENTILATION
946 RESPIRATORY CARE JULY 2009 VOL 54 NO 7
-
43. Heinze H, Sedemund-Adib B, Heringlake M, Gosch UW, GehringH,
Eichler W. The impact of different step changes of
inspiratoryfraction of oxygen on functional residual capacity
measurementsusing the oxygen washout technique in ventilated
patients. AnesthAnalg 2008;106(5):1491-1494.
44. Choncholas G, Sndergaard S, Heinonen E. Validation and
clinicalapplication of a first order step response equation for
nitrogen clear-ance during FRC measurement. J Clin Monit Comput
2008;22(1):1-9.
45. Bikker IG, van Bommel J, Reis Miranda D, Bakker J, Gommers
D.End-expiratory lung volume during mechanical ventilation: a
com-parison with reference values and the effect of positive
end-expira-tory pressure in intensive care unit patients with
different lung con-ditions. Crit Care 2008;12(6):R145.
46. Rylander C, Hogman M, Perchiazzi G, Magnusson A,
HedenstiernaG. Functional residual capacity and respiratory
mechanics as indi-cators of aeration and collapse in experimental
lung injury. AnesthAnalg 2004;98(3):782-789.
47. Rylander C, Tylen U, Rossi-Norrlund R, Herrmann P, Quintel
M,Bake B. Uneven distribution of ventilation in acute respiratory
dis-tress syndrome. Crit Care 2005;9(2):R165-R171.
48. Maisch S, Reissmann H, Fuellekrug B, Weismann D, Rutkowski
T,Tusman G, Bohm SH. Compliance and dead space fraction indicatean
optimal level of positive end-expiratory pressure after
recruitmentin anesthetized patients. Anesth Analg
2008;106(1):175-181.
49. Sinderby C, Navalesi P, Beck J, Skrobik Y, Comtois N,
Friberg S,Gottfried SB, Lindstrom L. Neural control of mechanical
ventilationin respiratory failure. Nat Med
1999;5(12):1433-1436.
50. Kondili E, Prinianakis G, Georgopoulos D. Patient ventilator
inter-action. Br J Anaesth 2003;91(1):106-119.
51. Thille AW, Cabello B, Galia F, Lyazidi A, Brochard L.
Reduction ofpatient-ventilator asynchrony by reducing tidal volume
during pres-sure-support ventilation. Intensive Care Med
2008;34(8):1477-1486.
52. Allo JC, Beck JC, Brander L, Brunet F, Slutsky AS, Sinderby
CA.Influence of neurally adjusted ventilatory assist and positive
end-expiratory pressure on breathing pattern in rabbits with acute
lunginjury. Crit Care Med 2006;34(12):2997-3004.
53. Beck J, Campoccia F, Allo JC, Brander L, Brunet F, Slutsky
AS,Sinderby C. Improved synchrony and respiratory unloading by
neu-rally adjusted ventilator assist (NAVA) in lung-injured
rabbits. Pe-diatr Res 2007 Mar;61(3):289-294.
54. Sinderby C, Beck J, Spahija J, de Marchie M, Lacroix J,
Navalesi P,Slutsky AS. Inspiratory muscle unloading by neurally
adjusted ven-tilatory assist during maximal inspiratory efforts in
healthy subjects.Chest 2007;131(3):711-717.
55. Beck J, Brander L, Slutsky AS, Reilly MC, Dunn MS, Sinderby
C.Non-invasive neurally adjusted ventilatory assist in rabbits with
acutelung injury. Intensive Care Med 2008;34(2):316-323.
56. Moerer O, Beck J, Brander L, Costa R, Quintel M, Slutsky AS,
et al.Subject-ventilator synchrony during neural versus
pneumatically trig-gered non-invasive helmet ventilation. Intensive
Care Med 2008;34(9):1615-1623.
57. Brander L, Leong-Poi H, Beck J, Brunet F, Hutchison SJ,
SlutskyAS, Sinderby C. Titration and implementation of neurally
adjustedventilatory assist in critically ill patients. Chest
2009;135(3):695-703.
58. Beck J, Reilly M, Grasselli G, Mirabella L, Slutsky AS, Dunn
MS,Sinderby C. Patient-ventilator interaction during Neurally
AdjustedVentilatory Assist in very low birth weight infants.
Pediatr Res 2009[Epub ahead of print].
59. Thille AW, Rodriguez P, Cabello B, Lellouche F, Brochard L.
Pa-tient-ventilator asynchrony during assisted mechanical
ventilation.Intensive Care Med 2006;32(10):1515-1522.
60. Willatts SM, Drummond G. Brainstem death and ventilator
triggersettings. Anaesthesia 2000;55(7):676-684.
61. Haldar M, Farrimond J. Ventilator trigger setting in an
intensive careunit (letter). Anaesthesia 2000;55(12):1213-1235.
62. Wikipedia contributors. Wikipedia. The free encyclopedia. A
pictureis worth a thousand words.
http://en.wikipedia.org/wiki/a_picture_is_worth_a_thousand_words.
Accessed May 12, 2009.
63. Drews FA, Westenskow DR. The right picture is worth a
thousandnumbers: data displays in anesthesia. Hum Factors
2006;48(1):59-71.
64. Wachter SB, Johnson K, Albert R, Syroid N, Drews F,
WestenskowD. The evaluation of a pulmonary display to detect
adverse respira-tory events using high resolution human simulator.
J Am Med In-form Assoc 2006;13(6):635-642.
65. Wachter SB, Markewitz B, Rose R, Westenskow D. Evaluation of
apulmonary graphical display in the medical intensive care unit:
anobservational study. J Biomed Inform 2005;38(3):239-243.
66. Branson RD. Dual control modes, closed loop ventilation,
handguns,and tequila (editorial). Respir Care
2001;46(3):232-233.
INNOVATIONS IN MECHANICAL VENTILATION
RESPIRATORY CARE JULY 2009 VOL 54 NO 7 947