Thorax 1991;46:131-135

Assisted ventilation 5 Series editor: John Moxham

Non-invasive and domiciliary ventilation:negative pressure techniques

John M Shneerson

Historical developmentNegative pressure ventilation first came intouse during the second half of the nineteenthcentury. It was recognised that air would bedrawn into the lungs through the mouth andnose if a subatmospheric pressure could bedeveloped around the thorax and abdomen.When the pressure around the -chest wallreturned to that of the ambient air, expirationoccurred passively owing to the elastic recoilof the lungs and chest wall.The chest and abdomen are enclosed in an

airtight, rigid chamber in all types of negativepressure ventilator, but in most of the earlierdesigns the whole of the body up to the neckwas also contained in the chamber. This hadthe advantage that chest wall expansion wasnot limited by contact with the sides of thenegative pressure device and that only oneairtight seal, that around the neck, wasrequired. The first of these negative pressureventilators to be of clinical value was thatdeveloped at Harvard University MedicalSchool by Drinker, an engineer, in 1928.1 Hedesigned severaI modified versions of his tankventilatof or irohlung and-these were widelyused during the poliomyelitis epidemics of thenext 30 years. Tank ventilators were producedthat could be constructed rapidly whenneeded and small enough for children.2 Mostof the early designs were telescopic in that thepatient was pulled in and out of the mainchamber of the tank on the mattress. In thelater designs the upper part of the chamberwas hinged towards the foot end and openedupwards (alligator type).

Simpler, non-tank negative pressure ven-tilators were first developed around the end ofthe nineteenth century. The most successfulof these was Eisenmenger's biomotor, acuirass designed in 1904.' This was super-seded by various negative pressure shells orcuirasses introduced from 1930 to 1960.These were not as effective as the tank ven-tilators or the modern individually mouldedcuirasses, but because of their simplicity theybecame widely used, particularly duringrecovery from acute poliomyelitis.

In the 1950s the jacket (wrap or poncho)design of negative pressure ventilator wasproduced,4 in which the properties of rigidityand imperviousness to air were separated intotwo structures. The rigidity was provided byan inner framework of metal or plastic andthis was covered by an airtight anorak likegarment with seals around the neck, arms, and

usually the waist. The jacket, like the tank andcuirass, was connected to a pump, which gen-erated a negative pressure between it and thepatient's chest wall.

All these types of negative pressure ven-tilator were used both in hospital and at hometo treat acute and chronic ventilatory failure.Ironically, as substantial improvements intheir design were being made intermittentpositive pressure ventilation using a trans-laryngeal endotracheal tube or a tracheostomywas shown to be more successful in thepoliomyelitis epidemics, such as that inCopenhagen in 1952.' The superiority ofintermittent positive pressure ventilation was

balbly tobetter protection of tie air-way rom asp ti gaNive pressure ven-tilators rapidly fell out of favour. In the lastdecade, however, negative pressure tech-niques have been used once again for variousconditions, particularly neuromuscular andskeletal disorders, and have been shown tohave a place, particularly for long term noc-turnal ventilation in the home.6

Tank ventilationMost of the modern tank ventilators are con-structed of aluminium, though some, such asthe Portalung, are made of plastic and aretherefore lighter. The patient's body rests ona mattress within the chamber and a head andneck rest is provided in most designs to ensurecomfort and to prevent kinking and obstruc-tion of the upper airway. Most designs havewindows that allow some observation of thepatient and portholes through which cathetersand monitor leads can be passed. They alsoenable some physiotherapy to be carried outwhile the patient is in the chamber andprocedures such as arterial blood gas samplingto be performed. In several of the mode<@either the head or the feet can be raised and inthe Kelleher design the whole tank can rotatethrough 1800 so that postural drainage can becarried out without removing the patient fromthe ventilator.7The older tank ventilators have a separate

bellows pump with a large stroke volume. Insome designs this is incorporated into thestructure of the ventilator. Newer modelshave separate rotary pumps of sufficientcapacity to evacuate the large volume of airfrom within the tank chamber.Tank ventilators are effective, do not need to

be constructed individually to fit each patient

Newmarket GeneralHospital, Newmarket,Suffolk CB8 7JGJM ShneersonReprint requests to:Dr Shneerson

131

on February 11, 2020 by guest. P

rotected by copyright.http://thorax.bm

j.com/

Thorax: first published as 10.1136/thx.46.2.131 on 1 F

ebruary 1991. Dow

nloaded from

Shneerson

and require only one airtight seal, aroundthe neck. These advantages are, however,balanced by the lack of access to the patientand their size, weight, and cost. In addition,they are available in few centres in the UnitedKingdom and considerable experience isreSpired to settle an ill patient successfully inthe ventilator. An airtight, comfortable necksemay be difficult to achieve and patientso -nfid it awkward to lie in one position ontheir back for long periods.Tank ventilation, like other forms of

negative pressure and nasal intermittentpositive pressure ventilation, does not protectthe airway. Aspiration of material from thepharynx into the trachea and bronchi mayoccur, but in practice this is uncommonexcept in neuromuscular disorders associatedwith abnormalities of the swallowing mechan-ism. Tank ventilators are capable of maintain-ing normal blood gas tensions even if there islittle or no spontaneous respiratory effort. The

Figure 1 Tankventilator: (a) closed,showing collar and headrest; (b) open, showingmattress and patientcontrol panel.(Reproduced by permissionof Penlon Ltd.)

I

tidal volume is linearly related to the peaknegative pressure within the chamber.8 Apressure of around 30 cm H20 is usuallyrequired but up to 40 cm H20 may be neces-sary.The models of tank ventilators that are used

most frequently in the United Kingdom arethose previously produced by Cape Warwick.There are several models, of which the"alligator" is the most common. A smallnumber of Kelleher rotating ventilatorssurvive and a simplified portable model wasintroduced in 1986 (fig 1). A few Both tankventilators are also in use. These were firstproduced in 1938 and constructed of lamin-ated wood. Most of them were modified in the1950s but they are less satisfactory than theCape ventilators. The most satisfactory ven-tilator of the newer, lighter, and more port-able design is the Lifecare Portalung. Thispolycarbonate tank ventilator is available inthree sizes. It is probably as effective as thetraditional designs and the current cost isapproximately £5000.

Jacket ventilationThe first effective jacket ventilator was theTunnicliffe jacket, developed in the 1950s.4 Itis still produced in two sizes, but even thelarger is too small for many patients, par-ticularly those with scoliosis. Several newermodels have been marketed in which the innerframework is made of a metal or plastic grid(fig 2). Some of these designs also have a backplate and they are all enclosed in an airtightsynthetic garment. This may cover the legs aswell as the trunk but in most designs thegarment finishes below the hips. The airwithin the jacket is intermittently evacuatedby a pump similar to those used for cuirassventilation (see below).The jackets do not restrain the expansion of

the rib cage or abdomen, but they are awk-ward for many patients to put on and oftencold to wear because of air leaks. Pressureareas are not a problem but jackets areinvariably larger and more cumbersome thancuirasses. They are preferable to tank ven-tilators for home use but less satisfactory fortreating the more severely ill patients inhospital. The tidal volume that they developat any given pressure is less than that of a tankventilator and the peak pressure that patientscan tolerate is also usually slightly less.

Several jacket designs are currently avail-able, costing about £500 plus the cost of thenegative pressure pump. The most commonlyused are the Tunnicliffe jacket, the LifecarePulmo-Wrap, and the Lifecare Nu-mo Gar-ment.

Cuirass ventilationMany of the earlier designs of cuirasses wereineffective because they were apoor fitand badlydesigned. Standard sized models rarely fitindividual patients well enough both to becomfortable and to provide an airtight sealaround the chest wall. Attention to detail in

132

---

:...

t

.1

T

I

I

mm x

on February 11, 2020 by guest. P

rotected by copyright.http://thorax.bm

j.com/

Thorax: first published as 10.1136/thx.46.2.131 on 1 F

ebruary 1991. Dow

nloaded from

Non-invasive and domiciliary ventilation: negative pressure techniques

Figure 2 Jacket ventilator showing rigid internalframework surrounding thorax andabdomen, garment, and connecting tubing leading to negative pressure pump.

constructing the cuirass is important so that,for instance, the anterior abdominal wall isfree to expand during inspiration and themovement of the lateral aspect and the upperchest is not restrained.

Standard sized designs of cuirass are stillavailable but should seldom be used. Somesuppliers will produce a cuirass from anindividual patient's chest and abdominaldimensions, but these are rarely satisfactory.

It is usually best to construct the cuirass froma cast ofthe patient. This can be made ofplasterof Paris, from which the cuirass is moulded.9The cuirass itself is constructed of light, air-tight, synthetic material, such as vitrathene orfibreglass. The edges are padded and coveredwith airtight material such as neoprene (fig 3). Aback strap may be needed to maintain anairtight seal between the cuirass and patient.

Cuirasses can be made to fit patients even ifthere is a severe thoracic deformity. The con-struction of individually moulded cuirasses is

Figure 3 Cuirassconstructed of vitrathene,with back strap attachedand connectorfor tubingleading to negativepressure pump.

analogous to the use of individually mouldednasal masks, which are widely used, par-ticularly in Europe, for nasal intermittentpositive pressure ventilation. The cost of thematerials for a cuirass is less than £50.

Disadvantages of cuirass ventilation are thatpressure areas may develop at the points ofcontact between the cuirass and the patient,and that if the patient grows or changes inweight a new cuirass may be needed. Thepatient has to sleep on his back or tilted slightlyto one side and, particularly in patients withscoliosis, extra padding may be needed aroundthe spine. Cuirasses are light and durable, andmost patients can put them on without assis-tance.The tidal volume achieved with a cuirass is

linearly related to the peak negative pressurewithin it'" and is considerably greater if theabdomen as well as the thorax is enclosedwithin the cuirass. Respiratory muscle activityis reduced" unless the patient is unable tocoordinate with the pump.

Various negative pressure pumps are avail-able in the United Kingdom and are suitablefor jacket or cuirass ventilation. The Capecuirass pumps are still in use but are no longermanufactured. The most suitable are the New-market pump and the Lifecare 170-C pump.The Newmarket pump is a rotary pump andable to compensate for variable air leaks be-tween the cuirass and the patient so that anypreset pressure within the cuirass can be main-tained.'2 The standard form provides con-trolled ventilation but there is also a patienttriggered model.

Indications for negative pressureventilationACUTE RESPIRATORY FAILURENeuromuscular and skeletal disorders Negativepressure ventilation may be invaluable inavoiding the need for endotracheal intubationduring acute episodes of respiratory failure inthese conditions." Cuirass or jacket ventilationmay be sufficient but many patients requiretreatment in a tank ventilator either contin-uously or at least overnight for a few days.Any of the three types of negative pressure

ventilator may be effective in weaning patientsfrom translaryngeal positive pressure ventila-tion. Once a tracheostomy has been construc-ted, however, use of a cuirass is preferable tothe other techniques as these may cause ob-struction of the tracheostomy. After the acuteepisode has been effectively treated it is impor-tant to assess whether long term ventilatorysupport is needed and, if so, what form thisshould take.Chronic lung disorders The place of negativepressure ventilation in acute infective exacer-bations of chronic airflow obstruction has notbeen definitely established. There is someevidence that tank ventilation combined withenergetic treatment of the underlying condi-tion is effective and may avoid the need forendotracheal intubation."4 Jacket and cuirassventilators are much less effective in thesecircumstances. There have been no com-

133

on February 11, 2020 by guest. P

rotected by copyright.http://thorax.bm

j.com/

Thorax: first published as 10.1136/thx.46.2.131 on 1 F

ebruary 1991. Dow

nloaded from

Shneerson

parative studies of the value of tank and nasalpositive pressure ventilation.

CHRONIC RESPIRATORY FAILURENeuromuscular and skeletal disorders Thesedisorders are the main indication for negativepressure ventilation. Tank and jacket ven-tilators are effective but in most cases theirinconvenience outweighs this advantage and a

cuirass is preferable for long term use. Arterialblood gas tensions improve both during the dayand at night,'0 and the prognosis once treat-ment is instituted is usually good unless theunderlying disease is progressive.'5Chronic lung disease Recent studies haveindicated that respiratory muscle strength'6and endurance'7 may be increased by regularnegative pressure ventilation. The evidence,however, is conflicting and the clinical impor-tance of the physiological improvements thathave been found is still uncertain.'8 There are

no studies of the effectiveness of this type oftreatment in improving prognosis but, as withintermittent positive pressure ventilation,'9 theresults would seem likely to be no better thanlong term oxygen treatment.

Negative or positive pressure ventilation?Many patients requiring mechanical res-

piratory support in the home need this onlyduring sleep. When they are awake their res-

piratory drive is sufficient to maintain more or

less normal blood gas tensions. In more mildlyaffected patients mechanical assistance may berequired only during acute illnesses or post-

operatively. Some patients with severe ven-tilatory failure, however, require mechanicalsupport for their breathing for part ofthe day as

well as each night.There have been no comparative studies of

cuirass and nasal positive pressure ventilationin these groups of patients. Less experience hasbeen obtained with nasal intermittent positivepressure ventilation but comparison of theresults of reported series suggests that the twomethods are about equally effective. Upperairway obstruction may occur with all types ofnegative pressure ventilation, especially duringrapid eye movement (REM) sleep. This may bedue to inhibition of the upper airway abductormuscles as part of a generalised inhibition ofinspiratory muscle activity, but loss of thenormal sequence of upper and lower res-

piratory muscle activation may also be impor-tant. Obstruction is often reduced by loweringthe peak negative pressure, or by giving thetricyclic antidepressant protryptyline, whichreduces the duration of REM sleep. Nasalventilation is preferable if upper airway ob-struction during use of a cuirass cannot beovercome, and a cuirass is preferable if nasalintermittent positive pressure ventilationforces air through the cricopharyngeal sphinc-ter into the stomach, causing abdominal disten-sion. In most other patients either system issatisfactory and the choice depends on

acceptability to the patient, local experience,availability ofequipment, and cost. The capitalcost of a cuirass and pump is slightly less than

that of a nasal ventilator, and the servicing andmaintenance are considerably cheaper. An ass-isted ventilation unit should have the facility toprovide both types of treatment so that patientsare not confined to either nasal intermittentpositive pressure ventilation or negative pres-sure ventilation for lack of an alternative.There is a smaller group of more severely

affected patients, such as those with quadri-plegia, who require continuous or almost con-tinuous mechanical ventilatory support. Inthese circumstances positive pressure ventila-tion through a tracheostomy is usually re-quired, though in selected cases alternativetechniques, such as mouth intermittentpositive pressure ventilation or phrenic nervepacing,2 may be of value for some or all of thetime. Intermittent positive pressure ventilationthrough a tracheostomy with a cuffed endo-tracheal tube is also indicated if aspiration intothe tracheobronchial tree is a problem owing todisordered swallowing, a poor cough, or both.Long term treatment in the home may bedifficult to organise in these severely disabledand highly dependent patients.

Indications for long term domiciliaryassisted ventilationThe decision to institute mechanical assistanceto ventilation is an important one because thisis usually a long term treatment. It is essentialto establish that respiratory failure is presentand that all conventional methods of treatmentof the underlying cause have been tried. Unfor-tunately, there are no firm guidelines for judg-ing whether respiratory failure is severe enoughto require long term mechanical support. Ingeneral, however, this should be considered ifrespiratory failure has caused troublesomesymptoms or potentially serious complicationssuch as polycythaemia or pulmonary hyperten-sion, or is likely to lead to these problems or topremature death.

Besides showing that the patient has severerespiratory failure, it is important to assess theprobability of improving the outlook if treat-ment is instituted. In many neuromuscular andskeletal disorders the quality of life and theprognosis may be greatly improved but inchronic airflow obstruction there is no evidenceyet that mechanical ventilatory support is moreeffective than long term oxygen treatment. Inthese patients ventilatory support does notprevent the slow and often inexorable progres-sion ofthe underlying airway disease. The needto use a home ventilator reduces the quality oflife and, as prognosis may not be improved, thistreatment should be recommended only forcarefully selected patients.

In some progressive neuromuscular disor-ders, such as Duchenne's muscular dystrophy,intermittent positive pressure ventilationthrough a tracheostomy may prolong life con-siderably but lead to a long period duringwhich the quality of life is extremely poorbecause of severe and widespread muscle weak-ness. In these circumstances it may be con-sidered ethical not to construct a tracheostomyfor intermittent positive pressure ventilation

134

on February 11, 2020 by guest. P

rotected by copyright.http://thorax.bm

j.com/

Thorax: first published as 10.1136/thx.46.2.131 on 1 F

ebruary 1991. Dow

nloaded from

Non-invasive and domiciliary ventilation: negative pressure techniques

but to provide a negative or nasal intermittentpositive pressure ventilation system, neither ofwhich protects the airway.

Funding and organisation of a homeventilator serviceThere are wide geographical differences in theuse of domiciliary ventilatory support in theUnited Kingdom. These are due partly tovariations in awareness of the value of treat-ment but also to the patchy availability of ahome ventilator service.The initiation of home ventilator treatment

requires much more than simply providingsuitable equipment and discharging the patientinto the community.20 A successful outcomerequires explanation, education, and supportfor both the patient and the family or atten-dants. An understanding of the underlyingcause of the respiratory failure and the equip-ment that has been provided, and what to do ifproblems arise, is needed. The choice ofwhether a negative or a positive pressure sys-tem is preferable often depends as much on thepatient's personality and the degree of supportin the home as on purely medical factors.

It is also essential that a rapid replacementservice is available in case equipment fails.Within the supervising hospital there should befacilities and trained staffavailable continuouslyin case emergency readmission is needed.At present there is no consistent pattern of

NHS funding for domiciliary ventilatory sup-port. Studies in the United Kingdom, France,and the United States have all shown that it isconsiderably cheaper than hospital care andjust as safe.'9 The British government'sproposals in the NHS and Community CareAct 1990 may rectify this once hospitals arereimbursed for the cost of treating individualpatients. There is, however, the risk that theinevitable pressure to reduce costs will alsoreduce the standards of care and support, andlimit the range of techniques available fortreating this vulnerable group of patients.

Appendix: Suppliers of negative pressureequipmentTANK VENTILATORSPortalung, available from Medicaid Ltd, Hook Lane,Pagham, Sussex P021 3PP.JACKETSTunnicliffejacket, available from Watco Services (BasingInstruments Ltd), PO Box 86, Basingstoke, HantsRG24 OGZ.Lifecare Pulmo-Wrap, available from Medicaid Ltd,Hook Lane, Pagham, Sussex P021 3PP.

Lifecare Nu-mo Garment, available from Medicaid Ltd,Hook Lane, Pagham, Sussex P021 3PP.cuiRAssShould be individually constructed.PUMPSNewmarket pump, available from Si-Plan ElectronicsResearch Ltd, Avenue Farm Industrial Estate, Bir-mingham Road, Stratford-on-Avon, Warwicks CV37OHP (about £2500).Lifecare 170-C pump, available from Medicaid Ltd,Hook Lane, Pagham, Sussex P021 3PP (about C5000).

1 Drinker PA, McKhann CF III. The iron lung. Firstpractical means of respiratory support. JAMA1985;225:1476-80.

2 Shneerson JM. Disorders of ventilation. Oxford: Blackwell,1988:233.

3 Anonymous. Apparatus for maintaining artificial respiration.Lancet 1904;i:515.

4 Spalding JMK, Opie L. Artificial respiration with theTunnicliffe breathing-jacket. Lancet 1958;i:613-5.

5 Lassen HCA. A preliminary report on the 1952 epidemic ofpoliomyelitis in Copenhagen with special reference to thetreatment of acute respiratory insufficiency. Lancet1953;i:37-41.

6 KinnearWJM, Shneerson JM. Assisted ventilation at home:is it worth considering? Br J Dis Chest 1985;79:313-51.

7 Kelleher WH. A new pattern of "iron lung" for theprevention and treatment of airway complications inparalytic disease. Lancet 1961;ii:1 113-6.

8 Whittenberger JL, Ferris BG Jr. Alterations of respiratoryfunction in poliomyelitis. Am JPhysMed 1952;31:226-37.

9 Brown L, Kinnear W, Sergeant K-A, Shneerson JM.Artificial ventilation by external negative pressure-amethod for manufacturing cuirass shells. Physiotherapy1985;71:181-3.

10 Kinnear W, Petch M, Taylor G, Shneerson JM. Assistedventilation using cuirass respirators. Eur Respir J1988;1:198-203.

11 Goldstein RS, Molotiu N, Skrastins R, Long S, ContrerasM. Assisting ventilation in respiratory failure by negativepressure ventilation and by rocking bed. Chest 1987;92:470-4.

12 Kinnear WJM, Shneerson JM. The Newmarket pump: anew suction pump for external negative pressure ventila-tion. Thorax 1985;40:677-81.

13 Libby DM, Briscoe WA, Boyce B, Smith JP. Acute res-piratory failure in scoliosis or kyphosis. Prolongedsurvival and treatment. Am JMed 1982;73:532-8.

14 Lovejoy FW, Yu PNG, Nye RE, Joos HA, Simpson JH.Pulmonary hypertension. III. Physiologic studies in threecases of carbon dioxide narcosis treated by artificialrespiration. Am J Med 1954;16:4-1 1.

15 Splaingard ML, Frates RC Jr, Jefferson LS, Rosen CL,Harrison GM. Home negative pressure ventilation: reportof 20 years of experience in patients with neuromusculardisease. Arch Phys Med Rehabil 1985;66:239-42.

16 Scano G, Gigliotti F, Duranti R, Spinelli A, Gorini M,Schiavina M. Changes in ventilatory muscle function withnegative pressure ventilation in patients with severeCOPD. Chest 1990;97:322-7.

17 Cropp A, Dimarco AF. Effects of intermittent negativepressure ventilation on respiratory muscle function inpatients with severe chronic obstructive pulmonary dis-ease. Am Rev Respir Dis 1987;135:1056-61.

18 Zibrak JD, Hill NS, Federman EC, Kwa SL, O'Donnell C.Evaluation of intermittent long-term negative pressureventilation in patients with severe chronic obstructivepulmonary disease. Am Rev Respir Dis 1988;138:1515-8.

19 Robert D, Gerard M, Leger P, et al. La ventilationmecanique a domicile definitive par tracheotomie del'insuffisant respiratoire chronique. Rev Fr Mal Respir1983;1 1:923-36.

20 Goldberg AI. Home care for life-supported persons: theFrench system of quality control, technology assessmentand cost containment. Publ Health Rep 1989;104:329-35.

135

on February 11, 2020 by guest. P

rotected by copyright.http://thorax.bm

j.com/

Thorax: first published as 10.1136/thx.46.2.131 on 1 F

ebruary 1991. Dow

nloaded from