Performance Comparison of 15 Transport Ventilators

Daniel W Chipman RRT, Maria P Caramez MD, Eriko Miyoshi MD,

Joseph P Kratohvil RRT, and Robert M Kacmarek PhD RRT FAARC

BACKGROUND: Numerous mechanical ventilators are designed and marketed for use in patient
transport. The complexity of these ventilators differs considerably, but very few data exist to compare
their operational capabilities. METHODS: Using bench and animal models, we studied 15 currently
available transport ventilators with regard to their physical characteristics, gas consumption (duration
of an E-size oxygen cylinder), battery life, ease of use, need for compressed gas, ability to deliver set
ventilation parameters to a test lung under 3 test conditions, and ability to maintain ventilation and
oxygenation in normal and lung-injured sheep. RESULTS: Most of the ventilators tested were relatively
simple to operate and had clearly marked controls. Oxygen cylinder duration ranged from 30 min to
77 min. Battery life ranged from 70 min to 8 hours. All except 3 of the ventilators were capable of
providing various FIO2 values. Ten of the ventilators had high-pressure and patient-disconnect alarms.
Only 6 of the ventilators were able to deliver all settings as specifically set on the ventilator during the
bench evaluation. Only 4 of the ventilators were capable of maintaining ventilation, oxygenation, and
hemodynamics in both the normal and the lung-injured sheep. CONCLUSIONS: Only 2 of the venti-
lators met all the trial targets in all the bench and animal tests. With many of the ventilators, certain of
the set ventilation parameters were inaccurate (differed by > 10% from the values from a cardiopul-
monary monitor). The physical characteristics and high gas consumption of some of these ventilators
may render them less desirable for patient transport. Key words: transport, mechanical ventilation, ven-
tilator, positive end-expiratory pressure, PEEP, fraction of inspired oxygen, FIO2. [Respir Care 2007;52(6):
740 –751. © 2007 Daedalus Enterprises]

Introduction transport ventilators are also required in ambulances and

in forward military positions. In addition, the threat of
Patients who require ventilatory support are frequently bioterrorism requires health care systems to be able and
transported from one hospital location to another. Portable rapidly with very little notice to accept and ventilate large
numbers of patients.
To provide ventilatory support under the above-defined
Daniel W Chipman RRT, Maria P Caramez MD, Eriko Miyoshi MD,
conditions requires that transport ventilators be appropri-
Joseph P Kratohvil RRT, and Robert M Kacmarek PhD RRT FAARC are
affiliated with the Department of Respiratory Care, Massachusetts Gen- ately designed, though this does not mean they must be
eral Hospital, Boston, Massachusetts. Maria P Caramez MD is also af- equivalent to intensive care unit ventilators.1 Transport
filiated with the Department of Anesthesia, Massachusetts General Hos- ventilators must incorporate certain characteristics to be of
pital, Boston, Massachusetts. Robert M Kacmarek PhD RRT FAARC is use in the above-defined settings.1–3 First, they must be
also affiliated with Harvard Medical School, Boston, Massachusetts.
able to ventilate patients with healthy or acutely or chron-
This study was partly funded by a grant from the United States Army, ically injured lungs. Second, they must be portable and
DAMD 17–02-2–0006. easy to operate. Third, they must be able to deliver a high
Daniel W Chipman RRT presented a version of this report at the OPEN fraction of inspired oxygen (FIO2). Fourth, in forward mil-
FORUM of 51st International Respiratory Congress of the American As- itary positions they must be able to operate on an internal
sociation for Respiratory Care, held December 3–6, 2005, in San Anto- battery for a long period, and without compressed gas.
nio, Texas.

Conflict of interest: Robert M Kacmarek PhD RRT FAARC has received

research grants and honoraria from Respironics, Maquet Medical, Tyco
Puritan Bennett, and Hamilton Medical. Daniel W Chipman RRT is a Correspondence: Robert M Kacmarek PhD RRT FAARC, Respiratory
paid consultant for and has received lecture honoraria from Maquet Med- Care, Ellison 401, Massachusetts General Hospital, 55 Fruit Street, Bos-
ical. ton MA 02114. E-mail: [email protected].

740 RESPIRATORY CARE • JUNE 2007 VOL 52 NO 6

 PERFORMANCE COMPARISON OF 15 TRANSPORT VENTILATORS

Table 1. Comparison of Evaluated Ventilators by Power Source, Physical Dimensions, Modes, FIO2, and PEEP

Approved Size FIO2 PEEP

as Power Weight (range or Available?
Ventilator Available Modes
Transport Requirement Height Width Depth (kg) available (range
Ventilator (cm) (cm) (cm) values) cm H2O)

Sophisticated
Univent Eagle 754 Yes Electricity 29 23 11 4.5 A/C, SIMV 0.21–1.0 Yes, 0–20
VersaMed iVent Yes Electricity 33 24 26 10 A/C (VP), SIMV (VP), 0.21–1.0 Yes, 0–20
PSV, CPAP
Newport HT50 Yes Electricity 26 27 20 6.8 A/C (VP), SIMV (VP), 0.21–1.0 Yes, 0–30
PSV, CPAP
Pulmonetic Systems LTV 1000 Yes Electricity 8 25 30 6.1 A/C (VP), SIMV (VP), 0.21–1.0 Yes, 0–30
PSV, CPAP
Simple
Oceanic Medical Products Yes Gas 12.7 17.8 10.2 2.1 CMV, IMV 1.0 No
Magellan
Bio-Med Devices IC2A Yes Gas 26 16 9 4.1 A/C, SIMV 1.0 Yes, 0–30
Pneupac Parapac Medic Yes Gas 9.2 22 16.2 2.4 CMV, SIMV 0.5, 1.0 No
Pneupac Parapac Transport Yes Gas 9.2 22 16.2 3.1 CMV, SIMV 0.5, 1.0 No
200D
Life Support Products Auto Yes Gas 15 4.5 9 0.68 CMV 1.0 No
Vent 2000
Carevent ATV⫹ Yes Gas 23.5 11.1 16.2 4.1 CMV, IMV 0.6, 1.0 Yes, 0–20
Vortran RespirTech Pro No Gas 16.76 6.35 8.38 0.165 CMV, IMV 1.0 No
Percussionaire TXP Yes Gas 10.6 10.6 16.5 0.68 CMV, IMV 0.5 No
Bio-Med Devices Crossvent 3 Yes Gas ⫹ electricity 22.9 28 12.7 4.32 A/C (VP), SIMV (VP), 0.5, 1.0 No
PSV, CPAP
Bird Avian Yes Gas ⫹ electricity 25 30 12.7 4.5 A/C, SIMV 1.0 No
Pneupac Compac 200 Yes Gas or electricity 36 21 21 8.5 CMV, IMV 0.45, 1.0 No

FIO2 ⫽ fraction of inspired oxygen

PEEP ⫽ positive end-expiratory pressure
A/C ⫽ assist/control
SIMV ⫽ synchronized intermittent mandatory ventilation
VP ⫽ volume-controlled and pressure-controlled modes available
PSV ⫽ pressure support ventilation
CPAP ⫽ continuous positive airway pressure
CMV ⫽ controlled mechanical ventilation
IMV ⫽ intermittent mandatory ventilation

Fifth, they should be able to provide both assisted and sions, available modes, available FIO2 ranges and settings,
controlled ventilation. Sixth, they must incorporate alarms and positive end-expiratory pressure (PEEP) range for the
that identify catastrophic conditions. 15 tested ventilators. All 15 ventilators were provided by
Previous evaluations of transport ventilators included their respective manufacturers specifically for this evalu-
only up to 8 ventilators.4 –10 Many of the ventilators pre- ation. All 15 ventilators are approved by the U.S. Food and
viously evaluated have since been modified by the man- Drug Administration for use in transport, except the Vor-
ufacturers, and new ventilators have entered the market. tran RespirTech Pro, which is marketed as a resuscitator.
We present an evaluation of 15 transport ventilators for
use during intrahospital or ambulance transport and in for- Bench Protocol
ward military positions. The goals of this study were (1) to
determine if these transport ventilators could ventilate both
We evaluated gas consumption, battery life, ease of use,
healthy and injured lungs, and deliver tidal volumes (VT)
physical characteristics, need for compressed gas, and the
and respiratory rates (RR) as specifically set, and (2) to
ability to deliver set ventilation parameters under 3 differ-
identify which ventilators would be most appropriate in
ent test conditions. Gas consumption was defined as the
which transport settings.
amount of time the ventilator could function on one full
Methods E-size oxygen cylinder (capacity 660 L of oxygen), with
the ventilator set to deliver a VT of 1,000 mL at an RR of
Table 1 shows power requirements, physical dimen- 10 breaths/min and an FIO2 of 1.0. Battery life was defined

RESPIRATORY CARE • JUNE 2007 VOL 52 NO 6 741

 PERFORMANCE COMPARISON OF 15 TRANSPORT VENTILATORS

as the amount of time the ventilator could function on a bilization and data collection were repeated. All other set-
fully charged battery with the ventilator set to deliver a VT tings remained unchanged. These settings were repeated
of 1,000 mL at an RR of 10 breaths/min and an FIO2 of for each of the compliance and resistance combinations.
0.21. The cardiopulmonary monitor was interfaced with a laptop
A ventilator was considered easy to use if all the pa- computer, on which the flow, volume, and pressure data
rameters were clearly labeled and easily set to deliver a were collected and analyzed. The set ventilation parame-
precise variable (eg, VT or RR). We assumed the manu- ters, ventilator-displayed values, and cardiopulmonary-
facturer’s published weight and dimensions to be accurate. monitor-measured values were simultaneously recorded.
Ability to ventilate without compressed gas was met if the All measurements during the bench assessment were at
ventilator could deliver the set minute volume (V̇E) under atmospheric-temperature-and-pressure-dry conditions.
each of the test conditions without a compressed gas source
or an external compressor. Laboratory Protocol
The ability to deliver set parameters was evaluated with
a test lung (Training and Test Lung, Michigan Instru- This protocol was approved by the animal care commit-
ments, Grand Rapids, Michigan) under 3 different test tee of Massachusetts General Hospital, Boston, Massachu-
conditions: high resistance with normal compliance; nor- setts.
mal resistance with normal compliance; and normal resis- Using 30-kg sheep, we evaluated each ventilator’s abil-
tance with low compliance. High and normal resistance ity to ventilate both healthy and saline-lavage lung-injured
was achieved with resistors (Pneuflo Rp20 and Rp5, Mich- sheep. In both settings we evaluated the ventilator’s ability
igan Instruments, Grand Rapids, Michigan). Normal and to maintain normal arterial blood gas values and cardio-
low compliance were set on the test lung (0.05 L/cm H2O pulmonary hemodynamics. We studied 12 sheep: 6 with
and 0.02 L/cm H2O, respectively). For each condition the normal lungs and 6 with saline-lavage lung injury. Five
tested ventilator was set to deliver a VT of 500 mL at ventilators were evaluated on each sheep (healthy and in-
15 breaths/min and 30 breaths/min, and a VT of 1 L at jured), and each group of 5 ventilators was studied on 2
10 breaths/min and 20 breaths/min. healthy and 2 injured sheep. Three groups of 5 ventilators
VT, RR, peak inspiratory pressure (PIP), and positive end- were randomly selected.
expiratory pressure (PEEP) were measured and analyzed with
a cardiopulmonary monitor (NICO, Respironics, Walling- Healthy Lung Evaluation
ford, Connecticut) and its software (Analysis Plus, Respiron-
ics, Wallingford, Connecticut). Ventilator performance was Each group of ventilators was randomly applied for a
determined by comparing the set parameters to the measure- 60-min period to a healthy sheep. Initially, each ventilator
ments from the cardiopulmonary monitor. was set at a VT of 9 mL/kg and an RR of 20 breaths/min,
Each ventilator was bench tested as follows. With re- with a TI or peak flow setting to maintain a TI of 1.0 s. If
sistance set at 20 cm H2O/L/s and compliance set at 0.05 the device was capable of applying PEEP, PEEP of
L/cm H2O, the ventilator was connected to the test lung, 5 cm H2O was applied with 50% oxygen. The ventilator
and the cardiopulmonary monitor’s flow sensor was placed was attached to the animal’s airway, followed by a 15-min
between the ventilator circuit and the flow resistor. VT was stabilization period. After stabilization we collected arte-
initially set at 500 mL, RR at 15 breaths/min, inspiratory rial and mixed venous blood samples, and measured sys-
time (TI) at 1.0 s (if setting the TI was possible on that temic arterial pressure, pulmonary artery pressure, pulmo-
ventilator), and PEEP at 5 cm H2O (if PEEP was available nary capillary wedge pressure, and heart rate. Airway
on the ventilator). The FIO2 was set at the lowest available pressure and VT were measured at the endotracheal tube
setting, which may have been 0.21, air mix (entrainment), (ETT). Cardiac output was measured in triplicate, using
or 1.0, depending on the ventilator’s capabilities. Follow- the thermodilution technique. The ventilator was adjusted
ing a 10-breath stabilization period, we recorded PIP, mean and oxygen added if needed to attempt to reach the target
airway pressure, PEEP, VT, V̇E, RR, and the pressure, blood gas values (PaO2 60 –100 mm Hg, PaCO2 30 –
flow, and volume graphics. After that data collection, the 50 mm Hg, pH 7.30 –7.50). Once we determined whether
RR was increased to 30 breaths/min and the TI was de- the targets could be met, the next ventilator was attached
creased to 0.5 s. All other settings remained unchanged. to the animal’s airway for evaluation.
Following another 10-breath stabilization period, we again
recorded PIP, mean airway pressure, PEEP, VT, V̇E, RR, Injured Lung Evaluation
and graphics. The VT was then increased to 1,000 mL, RR
was decreased to 10 breaths/min, and TI was increased to During the lung-injury tests, the ventilator was initially
1 s. Following another stabilization period and data col- set at a VT of 6 mL/kg, an RR of 30 breaths/min, PEEP of
lection, the RR was increased to 20 breaths/min, and sta- 15 cm H2O (if available), and FIO2 of 0.50. Again, blood

742 RESPIRATORY CARE • JUNE 2007 VOL 52 NO 6

 PERFORMANCE COMPARISON OF 15 TRANSPORT VENTILATORS

gases and hemodynamics were evaluated to determine if digital converter (DI-220, Dataq Instruments, Akron, Ohio),
the target blood gas values (PaO2 60 –100 mm Hg, PaCO2 and recorded at a sampling rate of 100 Hz, with data-
30 –50 mm Hg, pH 7.30 –7.50) had been met, then the acquisition software (Windaq/200, version 1.36, Dataq
ventilator was adjusted as necessary to attempt to meet the Instruments, Akron, Ohio). Ventilatory measurements
targets. Once we determined whether the targets could be made during the animal tests were all made at body-tem-
met, the next ventilator was attached to the animal’s air- perature-and-pressure-saturated conditions. All infusions,
way for evaluation. including the anesthetic, were given via volumetric infu-
sion pump. A heating blanket was used to maintain a core
Instrumentation temperature of 38 –39°C. An orogastric tube was placed to
empty the stomach.
We used 12 female Dorset sheep (21–31 kg), each fasted
for 24 hours. Orotracheal intubation, with an 8-mm inner- Lung Injury
diameter ETT, was performed during deep halothane an-
esthesia via mask. The external jugular vein was then can- Severe lung injury was produced with bilateral lung
nulated, and an 8 French sheath introducer was inserted. lavage via instillations of 1 L of isotonic saline, warmed to
After line placement, the anesthesia delivery was changed 39°C, repeated every 30 min, until PaO2 decreased to
to intravenous only, with a loading dose of 10 mg/kg ⱕ 100 mm Hg at an FIO2 of 1.0 and a PEEP of 5 cm H2O.
pentobarbital, 4 mg/kg ketamine, and 0.1 mg/kg pancuro- A stable lung injury was defined as a PaO2 change of ⬍ 10%
nium, followed by continuous infusion of pentobarbital after 60 min. It took 2– 4 lavages and 2–3 hours to estab-
(4 mg/kg/h), ketamine (8 mg/kg/h), and pancuronium lish a stable lung injury. During development and stabili-
(0.1 mg/kg/h) to provide surgical anesthesia with paraly- zation of lung injury, the animals were ventilated with the
sis. After intubation, the basic ventilatory settings were Puritan Bennett 840 ventilator. After a stable lung injury
volume control ventilation at a VT of 10 mL/kg, inspira- was established, 5 transport ventilators were randomly ap-
tory-expiratory ratio of 1:2, FIO2 of 1.0, and PEEP of plied.
5 cm H2O, delivered by an intensive care ventilator (840, On completion of the protocol, the animals were sacri-
Puritan Bennett, Carlsbad, California). RR was adjusted to ficed under deep anesthesia (10 mg/kg pentobarbital) with
achieve eucapnia (PaCO2 35– 45 cm H2O). rapid infusion of 50 mL saturated potassium chloride so-
An 18-gauge catheter was then placed into the carotid lution. Electrocardiogram and arterial blood pressure read-
artery for continuous measurement of arterial blood pres- ings confirmed cardiac standstill.
sure and sampling of arterial blood gas values. Arterial and
mixed venous blood samples were drawn for blood gas Statistical Analysis
analysis. PO2, PCO2, pH, oxyhemoglobin saturation, and
hemoglobin content were assessed with a blood gas ana- Formal statistical analysis was not performed. Lung
lyzer (282, Ciba Corning Diagnostics, Norwood, Massa- model data were compared to the ventilator settings. A
chusetts). Flow at the ETT was measured by a heated difference ⬎ 10% was considered excessive, because most
pneumotachometer (Hans Rudolph, Kansas City, Mis- of the ventilator manufacturers indicate that the normal
souri) connected to a differential pressure transducer range of operation is within 10% of the set parameters.
(MP-45 ⫾ 2 cm H2O, Validyne, Northridge, Califor- The mean ⫾ SD VT was calculated from 5 breaths.
nia). Volume was determined via digital integration of PIP, PEEP, and RR did not change with any ventilator
the flow signal. A differential pressure transducer during the bench evaluation. During the normal and in-
(MP-46 ⫾ 100 cm H2O, Validyne, Northridge, Califor- jured-lung animal evaluations, the ability of each ventila-
nia) was used to measure airway opening pressure. Car- tor to achieve the target blood gas values was evaluated.
diac output and pulmonary arterial pressure were mea- The oxygen cylinder duration and battery life were re-
sured via a 7.5 French pulmonary artery catheter corded in minutes.
(831 HF 7.5, Edwards Life Sciences, Irvine, California) Results
inserted into the left external jugular vein. Proper po-
sition of the catheter was confirmed via pressure wave- Bench Test
form analysis before and after balloon occlusion. Fol-
lowing instrumentation and a 30-min stabilization period, The 15 ventilators evaluated can be classified as either
5 transport ventilators were randomly applied. “simple” or “sophisticated” transport ventilators, and as
All signals (flow at the ETT, airway opening pressure, those that require compressed gas (pneumatic), those that
arterial blood pressure, and pulmonary arterial pressure) can operate without compressed gas but require electrical
were amplified (8805C, Hewlett Packard, Waltham, Mas- power, and those that require both or either power source.
sachusetts), converted to digital signals with an analog-to- The data in the tables and figures are organized with that

RESPIRATORY CARE • JUNE 2007 VOL 52 NO 6 743

 PERFORMANCE COMPARISON OF 15 TRANSPORT VENTILATORS

Table 2. Operational Features of Evaluated Ventilators: Battery, Gas Consumption, Alarms, and Ease of Use

Oxygen Able to Able to

High- Ease
Battery External Gas Cylinder Disconnect Ventilate Ventilate
Ventilator Battery Life* Pressure of
Powered Required Duration Alarm Normal Injured
Alarm Use‡
(min)† Lungs Lungs

Sophisticated
Univent Eagle 754 Yes 4h No 35 Yes Yes Yes No 1
VersaMed iVent Yes 90 min No 52 Yes Yes Yes Yes 1
Newport HT50 Yes 8 h, 10 min No 46 Yes Yes Yes Yes 1
Pulmonetic Systems LTV 1000 Yes 75 min No 32 Yes Yes Yes No 1
Simple
Oceanic Medical Products Magellan No NA Yes 60 Yes Yes Yes No 1
Bio-Med Devices IC2A No NA Yes 30 No No Yes Yes 2
Pneupac Parapac Medic No NA Yes 68§ Yes Yes Yes No 1
Pneupac Parapac Transport 200D No NA Yes 62 Yes Yes Yes No 1
Life Support Products AutoVent 2000 No NA Yes 60 No No Yes No 1
Carevent ATV⫹ No NA Yes 65 Yes Yes Yes No 1
Vortran RespirTech Pro㛳 No NA Yes Variable No No No No 3
Percussionaire TXP㛳 No NA Yes 77¶ No No Yes Yes 3
Bio-Med Devices Crossvent 3** Yes NA Yes 53 No No Yes No 1
Bird Avian** Yes NA Yes 30 Yes Yes Yes No 1
Pneupac Compac 200 Yes 4h No 65 Yes Yes Yes No 2

*Battery life is based on tidal volume (VT) of 1 L, respiratory rate of 10 breath/min, and fraction of inspired oxygen (FIO2) of 0.21.
†The oxygen cylinder duration is the time it took to consume 1 full E-size oxygen cylinder with the ventilator set at VT of 1 L, respiratory rate of 10 breaths/min, and FIO2 of 1.0
‡Ease of use: 1 ⫽ clearly labeled and easy to access; 2 ⫽ clearly labeled but difficult to access; 3 ⫽ not clearly labeled and difficult to access.
§VT gradually decreased as cylinder became depleted to 200 mL just before the ventilator shut down.
㛳True pressure-cycled ventilator. Changes in resistance or compliance significantly altered the delivered VT. Difficult to set at desired parameters.
¶FIO2 fixed at 0.5.
**Pneumatically powered, electronically controlled ventilator. Battery and/or alternating current electricity are required for electronic controls, monitoring, etc. Compressed gas is required for
ventilation.
NA ⫽ not applicable

schema. Five of the ventilators tested can be classified as Battery life ranged from 75 min to 8 hours and 10 min. All
sophisticated transport ventilators; the other ten are simple except 5 ventilators (Bio-Med Devices IC2A, Bio-Med
transport ventilators. Devices Crossvent 3, Life Support Products AutoVent
Eight of the ventilators were purely pneumatic (they 2000, Percussionaire TXP, and Vortran RespirTech Pro)
rely completely on compressed gas and do not require any incorporated both a low-pressure/disconnect alarm and a
other power source). Most of these incorporate an air- high-inspiratory-pressure alarm.
entrainment device to provide different FIO2 values. The One interesting finding was with the Bio-Med Devices
Oceanic Medical Products Magellan and the Life Support Crossvent 3. When changing from the “Air Mix” setting to
Products AutoVent 2000 are the exceptions; all breaths are 100% oxygen, the delivered VT approximately doubled
delivered with 100% source gas. Four ventilators were above the set VT. According to the manufacturer, this is
capable of ventilating with only battery or alternating- expected, and the operations manual instructs to adjust the
current power. Oxygen was not required, but may be added flow accordingly.
to increase FIO2. The Percussionaire TXP was the most difficult to oper-
The third group consisted of the 3 ventilators that re- ate, and its controls were not clearly identified. The Bio-
quire both compressed gas and electricity (battery or al- Med Devices IC2 and the Oceanic Medical Products Ma-
ternating current). These ventilators incorporate a pneu- gellan were the only ventilators operable near a magnetic
matic gas-delivery system and also require battery or resonance imaging device.
alternating current to operate their electronic controls. Tables 3 and 4 show the bench performance data. Most
Table 1 shows the ventilators’ physical dimensions and of the ventilators performed at or close to specifications
ventilation modes. There are considerable differences in under bench test conditions. The measured VT of most of
the size and weight. In some ventilators the modes are very the ventilators was less than the set VT, and this discrep-
limited, whereas others have multiple pressure and volume ancy increased under conditions of increased resistance or
modes. decreased compliance. Seventy-eight individual tests were
Gas consumption (Table 2) ranged from 30 min to 77 min. performed at a VT of 1,000 mL. In 32 of these tests the

744 RESPIRATORY CARE • JUNE 2007 VOL 52 NO 6

 Table 3. Bench Performance at a Tidal Volume of 500 mL

High Resistance and Normal Resistance and Normal Resistance and

Normal Compliance Normal Compliance Low Compliance

VT VT VT
RR PIP PEEP RR PIP PEEP RR PIP PEEP
(mean ⫾ (mean ⫾ (mean ⫾
(br/min) (cm H2O) (cm H2O) (br/min) (cm H2O) (cm H2O) (br/min) (cm H2O) (cm H2O)
SD mL) SD mL) SD mL)
Sophisticated Transport
Ventilators
Univent Eagle 754* 554 ⫾ 3.1† 15 19 5 583 ⫾ 4.5† 15 12 5 571 ⫾ 8.4† 15 32 5
Univent Eagle 754‡ 535 ⫾ 2.9 29 31 5 570 ⫾ 5.1† 30 16 5 555 ⫾ 3.6† 29 34 5
VersaMed iVent* 557 ⫾ 3.2† 15 19 4 541 ⫾ 11.5 15 12 5 515 ⫾ 16.3 15 32 4
VersaMed iVent‡ 401 ⫾ 7.1† 30 25 4 555 ⫾ 25.2† 30 16 4 496 ⫾ 6.1 30 32 5
Newport HT50* 480 ⫾ 9.2 15 18 4 477 ⫾ 12.1 15 14 4 498 ⫾ 4.5 15 29 5
Newport HT50‡ 468 ⫾ 4.8 29 21 9 459 ⫾ 7.4 30 14 5 494 ⫾ 6.7 30 30 4
Pulmonetic Systems 483 ⫾ 3.5 15 16 5 486 ⫾ 2.0 15 14 5 476 ⫾ 3.5 15 29 5
LTV 1000*
Pulmonetic Systems 483 ⫾ 4.2 30 20 7 484 ⫾ 3.8 30 16 6 477 ⫾ 1.3 30 29 5

RESPIRATORY CARE • JUNE 2007 VOL 52 NO 6

LTV 1000‡
Simple Transport
Ventilators
PERFORMANCE COMPARISON

Oceanic Medical 411 ⫾ 1.3† 23† 16 5 462 ⫾ 2.9 21† 11 5 441 ⫾ 10.4† 21† 26 4
Products Magellan*
OF

Oceanic Medical 428 ⫾ 3.1† 30 26 5 410 ⫾ 0.9† 31 12 5 418 ⫾ 1.6† 32 27 5

Products Magellan‡
Bio-Med Devices 487 ⫾ 0.6 17† 17 6 487 ⫾ 0.8 17† 13 6 474 ⫾ 1.5 18† 31 6
IC2A*
Bio-Med Devices 483 ⫾ 1.7 30 31 7 519 ⫾ 1.8 30 16 6 487 ⫾ 3.1 30 32 6
IC2A‡
Pneupac Parapac 507 ⫾ 1.0 15 17 5 530 ⫾ 1.2 15 16 6 483 ⫾ 2.9 16 28 5
Medic*
Pneupac Parapac 306 ⫾ 0.9† 30 14 6 324 ⫾ 1.0† 30 13 6 312 ⫾ 2.6† 30 20 5
Medic‡
Pneupac Parapac 472 ⫾ 1.6 16 18 5 485 ⫾ 1.8 16 14 5 462 ⫾ 10.2 16 27 4
Transport 200D*
15 TRANSPORT VENTILATORS

Pneupac Parapac 456 ⫾ 1.5 30 20 7 479 ⫾ 0.8 30 15 6 461 ⫾ 11.7 30 27 5

Transport 200D‡
Life Support Products 475 ⫾ 1.1 15 13 NA 504 ⫾ 2.7 15 12 NA 445 ⫾ 7.5† 15 23 NA
AutoVent 2000*
Life Support Products 472 ⫾ 5.5 17† 12 NA 497 ⫾ 5.3 18† 12 NA 438 ⫾ 9.8† 18† 24 NA
AutoVent 2000‡
Carevent ATV⫹* 535 ⫾ 1.3 13† 16 5 560 ⫾ 3.1† 12† 15 5 549 ⫾ 2.0 12† 31 6
Carevent ATV⫹‡ 555 ⫾ 5.1† 25† 24 5 570 ⫾ 3.8† 25† 16 5 575 ⫾ 10.7† 25† 34 6
Vortran RespirTech NA NA NA NA NA NA NA NA NA NA NA NA
Pro
Percussionaire TXP NA NA NA NA NA NA NA NA NA NA NA NA

745
(Continued)
 PERFORMANCE COMPARISON OF 15 TRANSPORT VENTILATORS

measured VT was ⬎10% different than the set VT, and in

(cm H2O)
PEEP
19 of these tests the measured RR was ⬎10% different

6
6
4
5
than the set RR.
(cm H2O) Seventy-eight individual tests were performed at a VT
of 500 mL. In 28 of these tests the measured VT was
Normal Resistance and

PIP

30
⬎10% different than set VT, and in 15 of these tests the
31

29
32
23
26
Low Compliance

RR was ⬎10% different than the set RR. The inability to

meet the target VT is partially explained by compressible
(br/min)

volume loss, since none of these units compensates for the

RR

15

30

15
30
14
29
volume loss due to compression. The compressible vol-
ume was 0.91 ⫾ 0.39 mL/cm H2O (range 0.43–1.65 mL/cm
H2O), and since the PIP values were at most in the mid-
390 ⫾ 8.7†
424 ⫾ 6.9†
(mean ⫾

507 ⫾ 1.1

520 ⫾ 1.1

475 ⫾ 1.3
488 ⫾ 2.9
SD mL)

30s, only about one third of the volume loss can be attrib-
VT

uted to compressible volume. Two of the ventilators (Vor-

tran RespirTech Pro and Percussionaire TXP) were not
included in the bench test because their design was incom-
(cm H2O)

patible with our protocol; they did not incorporate readily

PEEP

7
10
5
6

identifiable parameter settings.

Three of the ventilators were unable to meet some of the
test conditions, due to their designs. The Carevent ATV⫹
(cm H2O)
Normal Resistance and

VT ⫽ tidal volume. RR ⫽ respiratory rate. br/min ⫽ breaths/min. PIP ⫽ peak inspiratory pressure. PEEP ⫽ positive end-expiratory pressure. NA ⫽ not applicable.

achieves VT by setting V̇E and RR, with a maximum V̇E of

Normal Compliance

PIP

16

18

13
16
13
15

14 L/min and an inspiratory-expiratory ratio of 1:2 to es-

tablish TI. The Pneupac Compac 200 also achieves VT by
setting V̇E and RR with a maximum V̇E of 14 L/min. The
(br/min)

Life Support Products AutoVent 2000 has 2 controls: VT

RR

15

30

15
30
14
29

and RR, with a maximum RR of 17–18 breaths/min. In-

spiratory time and flow vary with changes in V̇E. At high
RR settings the inspiratory-expiratory ratio is ⬎1:1.
429 ⫾ 5.8†
442 ⫾ 5.8†
(mean ⫾

532 ⫾ 0.9

523 ⫾ 0.8

467 ⫾ 3.2
489 ⫾ 2.7
SD mL)
VT

Animal Evaluations

We evaluated 14 ventilators with healthy and saline-

(cm H2O)
PEEP

lavage-injured sheep. We were unable to evaluate the Vor-

5

6
7
6
6

tran RespirTech Pro in either animal model because the

weight of the sheep (20 –31 kg) was below the operating
range of this ventilator (minimum 40 kg). All of the 14
(cm H2O)
High Resistance and
Normal Compliance

PIP

ventilators tested were capable of ventilating the healthy

20

25

15
29
14
20

lungs (Fig. 1). Ventilators without FIO2 control met the pH

and PaCO2 targets, but exceeded the PaO2 target. With all the
†Measured value was ⬎ 10% different from the setting on the ventilator
(br/min)

ventilators, hemodynamics were stable throughout the tests.

RR

15

30

15
30
15
29

Four ventilators (VersaMed iVent, Newport HT50, Bio-

Med Devices IC2A, and Percussionaire TXP) met all the
targets when ventilating injured lungs (Fig. 2). Seven ven-
410 ⫾ 4.0†
436 ⫾ 5.0†
(mean ⫾

526 ⫾ 2.0

507 ⫾ 1.5

472 ⫾ 0.6
473 ⫾ 1.6
SD mL)

tilators (Pulmonetic Systems LTV 1000, Bird Avian, Oce-

VT

anic Medical Products Magellan, Pneupac Parapac Trans-

port 200D, Pneupac Parapac Medic, Bio-Med Devices
*Respiratory rate set at 15 breaths/min

‡Respiratory rate set at 30 breaths/min

Crossvent 3, and Carevent ATV⫹) successfully ventilated

Pneupac Compac 200*
Pneupac Compac 200‡

the lungs of only one of the 2 lung-injured animals. The

(Continued)

remaining 3 ventilators (Univent Eagle 754, Pneupac Com-

Bio-Med Devices

Bio-Med Devices
Crossvent 3‡

pac 200, and Life Support Products AutoVent 2000) failed

Crossvent 3*

Bird Avian*
Bird Avian‡

to meet the ventilation targets in either of the 2 lung-

injured sheep used. In all cases of failure, the PaCO2 and/or
Table 3.

pH targets were not met. In most cases, the RR setting was

the limiting parameter. All 14 ventilators met the oxygen-

746 RESPIRATORY CARE • JUNE 2007 VOL 52 NO 6

 Table 4. Bench Performance at a Tidal Volume of 1,000 mL

High Resistance and Normal Resistance and Normal Resistance and

Normal Compliance Normal Compliance Low Compliance

VT M RR PIP PEEP VT RR PIP PEEP VT RR PIP PEEP

(mean ⫾ (br/ (cm (cm (mean ⫾ (br/ (cm (cm (mean ⫾ (br/ (cm (cm
SD mL) min) H2O) H2O) SD mL) min) H2O) H2O) SD mL) min) H2O) H2O)
Sophisticated Transport
Ventilators
Univent Eagle 754* 1,028 ⫾ 4.7 10 36 5 1,102 ⫾ 3.1† 10 22 5 1,055 ⫾ 4.7 10 56 5
Univent Eagle 754‡ 1,027 ⫾ 2.8 20 35 5 1,097 ⫾ 6.8 20 22 5 1,009 ⫾ 5.7 20 55 5
VersaMed iVent* 989 ⫾ 74.9 10 33 5 996 ⫾ 58.8 10 20 5 906 ⫾ 17.3 10 51 5
VersaMed iVent‡ 932 ⫾ 6.3 20 40 5 972 ⫾ 66.6 20 20 4 929 ⫾ 0.8 20 51 5
Newport HT50* 926 ⫾ 8.9 10 36 4 940 ⫾ 24.9 10 23 3 962 ⫾ 20.2 10 52 5
Newport HT50‡ 920 ⫾ 14.8 20 37 5 944 ⫾ 14.4 20 24 4 967 ⫾ 15.4 20 54 5
Pulmonetic Systems LTV 1000* 955 ⫾ 1.1 10 36 5 945 ⫾ 6.7 10 24 5 927 ⫾ 10.8 10 52 5
Pulmonetic Systems LTV 1000‡ 954 ⫾ 2.5 20 37 6 949 ⫾ 1.7 20 24 5 926 ⫾ 2.1 20 52 5
Simple Transport Ventilators
Oceanic Medical Products 789 ⫾ 2.2† 12† 29 5 752 ⫾ 1.6† 12† 16 4 753 ⫾ 6.1† 12† 42 4
Magellan*

RESPIRATORY CARE • JUNE 2007 VOL 52 NO 6

Oceanic Medical Products 870 ⫾ 3.3† 24† 34 6 842 ⫾ 4.2† 25† 17 5 850 ⫾ 7.6† 25† 47 5
Magellan‡
Bio-Med Devices IC2A* 910 ⫾ 1.7 9 35 6 952 ⫾ 2.1 10 20 6 898 ⫾ 4.4† 10 52 6
Bio-Med Devices IC2A‡ 887 ⫾ 1.2† 21 37 7 937 ⫾ 1.1 21 20 6 886 ⫾ 4.1† 21 52 6
PERFORMANCE COMPARISON

Pneupac Parapac Medic* 942 ⫾ 2.9 11 28 5 973 ⫾ 3.5 11 24 6 899 ⫾ 6.5† 10 53 5

OF

Pneupac Parapac Medic‡ 635 ⫾ 2.0 20 24 5 653 ⫾ 2.2† 20 19 6 614 ⫾ 4.6† 20 36 5

Pneupac Parapac Transport 784 ⫾ 2.7† 12† 27 5 840 ⫾ 1.7† 12† 21 4 750 ⫾ 15.5† 12† 41 4
200D*
Pneupac Parapac Transport 774 ⫾ 1.6† 21 28 6 835 ⫾ 2.4† 21 21 5 755 ⫾ 3.2† 22† 42 5
200D‡
Life Support Products AutoVent 925 ⫾ 1.5 10 23 NA 948 ⫾ 4.7 10 20 NA 892 ⫾ 2.6† 10 44 NA
2000*
Life Support Products AutoVent 913 ⫾ 4.8 17† 22 NA 939 ⫾ 5.4 17† 20 NA 896 ⫾ 4.1† 17† 44 NA
2000‡
Carevent ATV⫹* 1,132 ⫾ 1.9† 9 28 5 1,164 ⫾ 1.3† 9 25 5 1,125 ⫾ 5.2† 9 56 6
Carevent ATV⫹‡ 1,142 ⫾ 1.3† 13† 33 5 1,179 ⫾ 5.2† 13† 26 5 1,149 ⫾ 1.4† 13† 58 5
Vortran RespirTech Pro NA NA NA NA NA NA NA NA NA NA NA NA
Percussionaire TXP NA NA NA NA NA NA NA NA NA NA NA NA
Bio-Med Devices Crossvent 3* 975 ⫾ 0.7 10 38 4 1,016 ⫾ 1.7 10 25 5 948 ⫾ 1.1 10 51 5
15 TRANSPORT VENTILATORS

Bio-Med Devices Crossvent 3‡ 972 ⫾ 0.9 20 39 5 1,017 ⫾ 0.8 20 25 4 953 ⫾ 9.7 20 51 4

Bird Avian* 950 ⫾ 2.6 10 33 6 970 ⫾ 3.1 10 20 6 958 ⫾ 4.7 10 52 5
Bird Avian‡ 947 ⫾ 2.7 20 35 8 972 ⫾ 3.1 20 21 7 958 ⫾ 4.0 20 53 6
Pneupac Compac 200* 917 ⫾ 1.3 9 24 6 917 ⫾ 9.8 9 21 5 794 ⫾ 13.6† 9 42 4
Pneupac Compac 200‡ 654 ⫾ 6.0† 18† 23 6 654 ⫾ 7.3† 18† 15 5 629 ⫾ 12.2† 18† 36 4

*Respiratory rate set at 15 breaths/min

†Measured value was ⬎ 10% different from the setting on the ventilator
‡Respiratory rate set at 30 breaths/min
VT ⫽ tidal volume. RR ⫽ respiratory rate. br/min ⫽ breaths/min. PIP ⫽ peak inspiratory pressure. PEEP ⫽ positive end-expiratory pressure. NA ⫽ not applicable.

747
 PERFORMANCE COMPARISON OF 15 TRANSPORT VENTILATORS

Fig. 1. PaCO2, arterial pH, and ratio of PaO2 to fraction of inspired oxygen (FIO2) in healthy (no lung injury) sheep during ventilation with 14
transport ventilator models. Each set of values represents data from a single sheep. Assessment was performed on 2 sheep with each
ventilator. The large variability in the ratio of PaO2 to fraction of inspired oxygen (PaO2/FIO2) is because some of the ventilators only offer only
1 or 2 FIO2 settings, and because of the level of gas exchange in each sheep. The Vortran RespirTech Pro could not be used on the animals
we tested. The Percussionaire TXP and the Vortran RespirTech Pro could not be set to the specifications required by the lung model.
U ⫽ Univent Eagle 754. P ⫽ Pulmonetic Systems LTV 1000. V ⫽ VersaMed iVent. B ⫽ Bird Avian. M ⫽ Oceanic Medical Products Magellan.
N ⫽ Newport HT50. PT ⫽ Pneupac Parapac Transport 200D. PM ⫽ Pneupac Parapac Medic. C ⫽ Pneupac Compac 200. BI ⫽ Bio-Med
Devices IC2A. BC ⫽ Bio-Med Devices Crossvent 3. CV ⫽ Carevent ATV⫹. A ⫽ Life Support Products AutoVent 2000. PC ⫽ Percussionaire
TXP.

ation target under all conditions with the lung-injured sheep. Use of Transport Ventilators
As with the uninjured sheep, with all the ventilators the
hemodynamics were stable throughout the tests.
Transport ventilators are required in various settings:
intra-hospital, inter-hospital, pre-hospital, and in the field
Discussion
by military or civilian authorities.3 Each of these settings
has different priorities regarding ventilator design. In for-
The major findings of this study are: ward military or field use by civilian groups, the ideal
1. All the evaluated ventilators were able to maintain ventilator would be simple to operate, battery powered,
normal ventilation and hemodynamics in healthy sheep. compact, lightweight, and would operate without com-
2. In the lung-injured sheep, few of the ventilators could pressed gas. In that setting it is unlikely that the patient
be set to meet the PaCO2 or pH targets. The ventilators will be breathing spontaneously, so versatility of available
unable to meet these targets were limited by the RR set-
modes is unnecessary. Similar issues exist during pre-hos-
ting.
pital transport, but compressed gas is readily available in
3. In the bench study, only 6 of the ventilators met the
most ambulances, so a pneumatically operated ventilator is
VT and RR settings under all the test conditions.
as acceptable as a battery operated unit. During inter-hos-
4. Only 5 of the ventilators (Univent Eagle 754, Ver-
saMed iVent, Newport HT50, Pulmonetic Systems LTV pital transport the patient may be breathing spontaneously,
1000, and Pneupac Compac 200) can operate without a which necessitates patient-triggered ventilation, and fre-
compressed gas source, and their battery life differed con- quently these patients require high FIO2.
siderably. The most common use of transport ventilators is in in-
5. A full E-size cylinder of oxygen allowed ventilation tra-hospital transport. At Massachusetts General Hospital,
with 100% oxygen for only 30 –77 min. the respiratory care department performs about 30 one-
6. The 2 ventilators most suitable for use in front-line way patient transports per day and another 10 –15 are per-
rescue situations, where oxygen may not be available, are formed by the anesthesia department, all of which require
the Newport HT50 and the Univent Eagle 754. continuous mechanical ventilation. Most of these trans-

748 RESPIRATORY CARE • JUNE 2007 VOL 52 NO 6

 PERFORMANCE COMPARISON OF 15 TRANSPORT VENTILATORS

Fig. 2. PaCO2, arterial pH, and ratio of PaO2 to fraction of inspired oxygen (FIO2) in lung-injured sheep during ventilation with 14 transport
ventilator models. Each set of values represents data from a single sheep. Assessment was performed on 2 sheep with each ventilator. The
large variability in the ratio of PaO2 to fraction of inspired oxygen (PaO2/FIO2) is because some of the ventilators only offer 1 or 2 FIO2 settings,
and because of the level of gas exchange in each sheep. The Vortran RespirTech Pro could not be used on the animals we tested. The
Percussionaire TXP and the Vortran RespirTech Pro could not be set to the specifications required by the lung model. U ⫽ Univent Eagle
754. P ⫽ Pulmonetic Systems LTV 1000. V ⫽ VersaMed iVent. B ⫽ Bird Avian. M ⫽ Oceanic Medical Products Magellan. N ⫽ Newport
HT50. PT ⫽ Pneupac Parapac Transport 200D. PM ⫽ Pneupac Parapac Medic. C ⫽ Pneupac Compac 200. BI ⫽ Bio-Med Devices IC2A.
BC ⫽ Bio-Med Devices Crossvent 3. CV ⫽ Carevent ATV⫹. A ⫽ Life Support Products AutoVent 2000. PC ⫽ Percussionaire TXP.

ports are to and from diagnostic areas, the operating room, tors, simple transport ventilators, and sophisticated trans-
or the emergency department. port ventilators. They defined a simple transport ventilator
It is well documented that transport ventilators provide as one that provides a specified rate and volume with a
more stable gas exchange and hemodynamics than manual high-pressure relief valve,3 whereas a sophisticated trans-
ventilators.10 –13 Gervais et al11 observed severe respiratory port ventilator has modes that allow spontaneous breath-
alkalosis during transport with manual ventilation. In 20 pa- ing, and additional alarms and monitors of gas delivery.
tients transported to diagnostic areas, Braman et al12 found Five of the ventilators we evaluated were sophisticated:
substantial respiratory acidosis or alkalosis and hemodynamic Newport HT50, Univent Eagle 754, VersaMed iVent, Pul-
compromise in 16 patients receiving manual ventilation. Hurst monetic Systems LTV 1000, Bird Avian.
et al13 also documented respiratory alkalosis during intra- We considered the Newport HT50 and the Univent
hospital transport of 28 patients receiving manual ventilation. Eagle 754 most suited for use in forward military po-
Nakamura et al10 also observed greater variability of gas sitions, because they have longer battery life (8 hours
exchange and hemodynamics during transport with manual
and 4 hours, respectively). The VersaMed iVent was the
ventilation than with a transport ventilator.
heaviest of the units evaluated (10 kg). However, the
Types of Transport Ventilators Newport HT50, Univent Eagle 754, VersaMed iVent,
and Pulmonetic Systems LTV 1000 clearly could func-
Austin et al3 classified transport ventilators into 3 cat- tion exceptionally well in all transport settings if they
egories, based on their capabilities: automatic resuscita- had longer battery life. The Bird Avian was the most

RESPIRATORY CARE • JUNE 2007 VOL 52 NO 6 749

 PERFORMANCE COMPARISON OF 15 TRANSPORT VENTILATORS

limited in this regard, because it lacks a battery and controls, and the manufacturer’s specified patient-weight
needs compressed gas to operate. range was outside the weight range of the animals we
used. However, it is the smallest and lightest of the ven-
Issues/Problems With Specific Ventilators tilators we tested, and it is only for single-patient use.
It may be necessary with some of these ventilators to
The choice of a ventilator is also determined by other monitor gas delivery with a secondary monitor because of
specific design issues. Only 2 of the ventilators (Bio-Med the large difference between the set and actual VT and RR.
Devices IC2 and Oceanic Medical Products Magellan) are Since we did not assess these ventilators during spontane-
designed for use during magnetic resonance imaging. The ous breathing, we cannot comment on patient-ventilator
following ventilators had no alarms: Vortran RespirTech synchrony or the difference between the set and delivered
Pro, Bio-Med Devices IC2, Percussionaire TXP, Oceanic parameters during spontaneous ventilation.
Medical Products Magellan, and Life Support Products
AutoVent 2000. Comparison With Other Studies
Many of the ventilators allow very few FIO2 values: Vor-
tran RespirTech Pro (FIO2 1.0), Bio-Med Devices IC2 (FIO2 Nolan et al5 evaluated the performance of 6 pneumati-
1.0), Oceanic Medical Products Magellan 2000 (FIO2 1.0), cally operated ventilators. Similar to our results, they noted
Pneupac Parapac Transport 200D (FIO2 0.5 or 1.0), Pneupac that the overall ability of the ventilators they tested to
Parapac Medic (FIO2 0.5 or 1.0), Bio-Med Devices Cross- maintain delivered VT, V̇E, and RR consistent with the set
vent 3 (FIO2 0.5 or 1.0), Carevent ATV⫹ (FIO2 0.8 or 1.0), levels diminished as resistance increased or compliance
and Life Support Products AutoVent 2000 (FIO2 1.0). decreased. McGough et al6 observed the same problem
The oxygen cylinder life of the Pneupac Parapac Medic with 8 pneumatically operated ventilators they evaluated
exceeded the maximum estimated time (66 min), because with a test lung. The Univent 750 was evaluated by Camp-
VT gradually decreased as the cylinder became depleted to bell et al,7 with a test lung, during controlled and patient-
200 mL just before the ventilator shut down. triggered ventilation. They observed, as we did, that with
With the Percussionaire TXP, the maximum FIO2 deliv- the Univent Eagle 754, gas delivery was not markedly
ered was 0.5, which accounts for its 77-min cylinder life. affected by a decrease in compliance or an increase in
Note, however, that the volume of gas in E-size cylinders resistance.
does vary, because filling pressure varies, which adds to More recently, Miyoshi et al8 evaluated 4 ventilators
the variability in cylinder life. with transport capabilities, all with internal batteries. How-
With the Newport HT50 and its nondisposable propri- ever, at least 3 of these units (Puritan Bennett 740, Bird
etary circuit, intrinsic PEEP developed at higher RR be- T-Bird, and Respironics Espirit) would not be considered
cause of high expiratory resistance. With the Bio-Med typical transport ventilators. However, all of these units,
Devices IC2, Life Support Products Magellan 2000, Pneu- along with the Pulmonetic Systems LTV 1000, were ca-
pac Parapac Transport 200D, and Pneupac Parapac Medic pable of ventilating a test lung during assisted ventilation,
the VT is set with the flow rate and TI controls, and RR is at various ventilation settings.
controlled by those two plus an expiratory time control. Zanetta et al9 evaluated 5 transport ventilators and 3
With the Carevent ATV⫹, VT is determined by V̇E and intensive care unit ventilators during controlled and pa-
RR. The Life Support Products AutoVent 2000 has 2 con- tient-triggered ventilation with a test lung. They deter-
trols (RR and VT), its maximum RR is 18 breaths/min, and mined that VT varied ⬍ 10% as delivered VT varied from
it does not have any alarms. The Pneupac Compac 200 is 300 mL to 800 mL and compliance and resistance were
designed for military use. It has a sturdy case, and VT is varied. However, they noted that, because of high resis-
adjusted by setting V̇E and RR. It has a fixed TI of 1 s and tance to exhalation, all the portable ventilators they eval-
a maximum RR of 26 breaths/min. The Percussionaire uated trapped gas at high V̇E.
TXP is a pressure-limited and time-cycled ventilator, and
its VT varied with changes in impedance, but we found Limitations
that even with constant impedance the VT drifted upwards.
The maximum RR with the Life Support Products The primary limitation of the present study is that it was
AutoVent 2000 is 18 breaths/min, and with the Carevent not performed with patients. However, the bench and an-
ATV⫹ it is 40 breaths/min. With the Oceanic Medical imal evaluations did simulate common settings required
Products Magellan, setting RR at 15 breaths/min and by patients during controlled ventilation. In addition, the
20 breaths/min resulted in measured RR of 23 breaths/min animal model evaluations were consistent with pediatric
and 30 breaths/min, respectively. patients, not adults. This limited the assessment of some of
The most difficult ventilator to evaluate was the Vortran the ventilators. We also did not evaluate any of these
RespirTech Pro. This ventilator has few clearly labeled ventilators during spontaneous breathing, which is clearly

750 RESPIRATORY CARE • JUNE 2007 VOL 52 NO 6

 PERFORMANCE COMPARISON OF 15 TRANSPORT VENTILATORS

a major issue in transport within and between hospitals. As 5. Nolan JP, Baskett PJF. Gas-powered and portable ventilators: an
a result, we cannot comment on their performance during evaluation of six models. Prehospital Disaster Med 1992;7(1):25–
34.
spontaneous triggering. Also, we evaluated only one ven-
6. McGough EK, Banner MJ, Melker RJ. Variations in tidal volume
tilator from each company, and we cannot be sure that the with portable transport ventilators. Respir Care 1992;37(3):233–
single ventilator we tested reflects the operation of all 239.
ventilators of that model. 7. Campbell RS, Davis K Jr, Johnson DJ, Porembka D, Hurst JM.
Laboratory and clinical evaluation of the Impact Uni-Vent 750 por-
Conclusions table ventilator. Respir Care 1992;37(1):29–36.
8. Miyoshi E, Fujino Y, Mashimo T, Nishimura M. Performance of
Only 2 of the transport ventilators evaluated met the trial transport ventilator with patient-triggered ventilation. Chest 2000;
118(4):1109–1115.
targets in all bench and animal settings. With some ventila-
9. Zanetta G, Robert D, Guerin C. Evaluation of ventilators used during
tors the settings were inaccurate. The physical characteristics transport of ICU patients: a bench study. Intensive Care Med 2002;
and high gas consumption of some of these ventilators may 28(4):443–451.
render them less desirable for patient transport. 10. Nakamura T, Fujino Y, Uchiyama A, Mashimo T, Nishimura M.
Intrahospital transport of critically ill patients using ventilator with
REFERENCES patient-triggering function. Chest 2003;123(1):159–164.
1. Branson RD, McGough EK. Transport ventilators. Probl Crit Care 11. Gervais HW, Eberle B, Konietzke D, Hennes HJ, Dick W. Compar-
1990;4:254–274. ison of blood gases of ventilated patients during transport. Crit Care
2. Branson RD. Intrahospital transport of critically ill, mechanically ven- Med 1987;15(8):761–763.
tilated patients. Respir Care 1992;37(7):775–793; discussion 793–795. 12. Braman SS, Dunn SM, Amico CA, Millman RP. Complications of
3. Austin PN, Campbell RS, Johannigman JA, Branson RD. Transport intrahospital transport in critically ill patients. Ann Intern Med 1987;
ventilators. Respir Care Clin N Am 2002;8(1):119–150. 107(4):469–473.
4. Johannigman JA, Branson RD, Campbell R, Hurst JM. Laboratory 13. Hurst JM, Davis K Jr, Branson RD, Johannigman JA. Comparison of
and clinical evaluation of the MAX Transport Ventilator. Respir blood gases during transport using two methods of ventilatory sup-
Care 1990;35(10):952–959. port. J Trauma 1989;29(12):1637–1640.

RESPIRATORY CARE • JUNE 2007 VOL 52 NO 6 751