a comparison of the effects of manual and ventilator ... ventilator hyperinflation on static lung...

100 Physiotherapy Research International, 7(2) 100–108, 2002 © Whurr Publishers Ltd

A comparison of the effects of manualand ventilator hyperinflation on staticlung compliance and sputum productionin intubated and ventilated intensive carepatients

SUSAN BERNEY Department of Physiotherapy, Austin and Repatriation Medical Centre,Victoria, Australia

LINDA DENEHY School of Physiotherapy, The University of Melbourne, Australia

ABSTRACT Background and Purpose. Lung hyperinflation is a technique used by phys-iotherapists to mobilize and remove excess bronchial secretions, reinflate areas ofpulmonary collapse and improve oxygenation. Hyperinflation may be delivered by the ven-tilator or manually, by use of a manual resuscitation circuit, depending upon therespiratory and cardiovascular status of the patient. The effects of manual hyperinflation,with respect to excess bronchial secretions and static lung compliance, have been well-established. There is, however, only limited evidence as to the efficacy of ventilatorhyperinflation as a physiotherapy treatment technique. The purpose of the present studywas to compare the effects of manual hyperinflation and ventilator hyperinflation on staticpulmonary compliance and sputum clearance in stable intubated and ventilated patients.Method. Twenty patients who met the inclusion criteria were studied. This was a doublecrossover study where all patients were randomly allocated to one of two treatmentsequences over two days. The first sequence involved manual hyperinflation followed twohours later by ventilator hyperinflation and the order was reversed on the second day. Inthe second sequence, ventilator hyperinflation preceded manual hyperinflation. The vari-ables of static pulmonary compliance and sputum wet weight were analysed by use of ananalysis of variance (ANOVA) for repeated measures. Statistical significance was set atp < 0.05. Results. There was no significant difference in sputum wet weight productionbetween either technique or on either day of treatment. Static pulmonary complianceimproved with both hyperinflation techniques (p < 0.05). Conclusions. Hyperinflation aspart of a physiotherapy treatment can be performed with equal benefit using either amanual resuscitation circuit or a ventilator. Both methods of hyperinflation improve staticpulmonary compliance and clear similar volumes of pulmonary secretions.

Key words: lung hyperinflation, sputum wet weight, static pulmonary compliance

PRI 7(2)_3rd/crc 20/5/02 2:00 PM 00

101Manual and ventilator hyperinflation: a comparison of effects

INTRODUCTION

Atelectasis is common in intubated andventilated patients (Johnson et al., 1987).The reasons for this are multifactorial andinclude low volume ventilation strategies. Ithas been suggested that the use of intermit-tent large tidal volume inflations of thelungs may help prevent the development ofatelectasis (Bendixen et al., 1963; Suter etal., 1978; Blanch et al., 1994; Dorringtonand Radcliffe, 1999; Pelosi et al., 1999).

Manual hyperinflation involves deliver-ing tidal volumes to airway pressures of 40cm H

2O (Rothen et al., 1993) or a tidal

volume (VT) that is 50% greater than that

delivered by the ventilator (Singer et al.,1994). This is followed by a quick releaseof pressure on expiration, leading to a rapidflow of air, simulating the effect of a cough(Clement and Hubsch, 1968). Physiothera-pists have used manual hyperinflation formany years for mobilizing excess pul-monary secretions, reinflating areas ofatelectasis and improving oxygenation(Stiller et al., 1990; Tweed et al., 1993;Webber and Pryor, 1993; Singer et al.,1994; Stiller et al., 1996). It can be deliv-ered using an anaesthetic circuit or theventilator.

The short-term effects of manual hyper-inflation on pulmonary compliance andresolution of atelectasis have been well doc-umented (Marini et al., 1979; Rhodes,1987; Stiller et al., 1990; Jones et al., 1992;Rothen et al., 1993; Stiller et al., 1996;Hodgson et al., 2000). However, manualhyperinflation is contraindicated in severalpatient groups who may otherwise benefitfrom the technique. These patients mayrequire higher levels of positive end expira-tory pressure (PEEP) or may be agitatedand intolerant of manipulation of theirendotracheal tube. Ventilator hyperinflation

is achieved by altering the ventilatory set-tings to gradually increase tidal volume(Imle and Klemic, 1989). It may producethe same effects as manual hyperinflationwhilst maintaining the PEEP level and con-trolling airway pressure limits (Brown et al.,1987; Imle and Klemic, 1989). There is,however, no evidence as to the efficacy ofventilator hyperinflation as a physiotherapytreatment technique. As there are no con-trolled trials comparing the technique tomanual hyperinflation, the clinical use ofventilator hyperinflation is not widespread.Therefore, the purpose of the present studywas to compare the effects of manualhyperinflation and ventilator hyperinflationon static lung compliance and sputum wetweight in stable intubated and ventilatedpatients.

METHOD

This was a prospective, randomized, con-trolled crossover study of patients who wereintubated, ventilated and cardiovascularlystable.

Subjects

The study was approved by the Ethics andHuman Research Committee at the Austinand Repatriation Medical Centre (ARMC),and informed consent was obtained from thenext of kin and from the treating intensivecare physician. Sample size calculation wasbased on sputum wet weight using the meandifference and standard deviation from Hodg-son et al. (2000). For the given effect size,alpha = 0.05 (two-tailed) and power of 0.8,the sample size estimate was 20 subjects.

Subjects were included if they were:

• Intubated and ventilated.• Would normally receive hyperinflation as

a part of their physiotherapy treatment.

PRI 7(2)_3rd/crc 20/5/02 2:00 PM Page 101

102 Berney and Denehy

Patients were excluded from the trial ifthey:

• Required a fraction of inspired oxygen(FiO

2) ≥ 0.6.

• Had a PEEP ≥ 10 cm H2O.

• Had pulmonary pathology where lunghyperinflation was contraindicated (forexample, adult respiratory distress syn-drome, exacerbation of chronicobstructive pulmonary disease).

• Were prescribed a head-up position forbrain disease.

• Had an unstable cardiovascular condi-tion as def ined by a mean arterialpressure (MAP) ≤ 75 mmHg with a fluc-tuation of 15 mmHg with positionchange, a heart rate >130.

• Had an arterial oxygen saturation (SaO2)

≤ 90%.

Patients were withdrawn from the studyif they suffered cardiovascular compromiseduring the treatment, as def ined by theabove variables.

Procedure

All subjects were randomly allocated usingsealed envelopes to a treatment sequence Aor B on Day 1. Patients received two treat-ments, at least two hours apart, on twoconsecutive days. The first treatment was

given in the morning and the alternate treat-ment was performed in the afternoon. OnDay 2 the order of the treatments wasreversed using the same patient positionsequence (Figure 1).

Treatment A consisted of gravityassisted drainage with the foot of the bedelevated to 35˚–45˚ from the horizontal, sixsets of six manual hyperinflation breathsand endotracheal suctioning. Treatment Bconsisted of the same treatment except thatventilator hyperinflation was used insteadof manual hyperinflation. The initial treat-ment position of the subject was decidedupon by consultation between the intensivecare physician and treating physiotherapist.

Subjects were turned into the appropriatesidelying position and left undisturbed for 10minutes before treatment. The foot of the bedwas then elevated and treatment started. Oneresearcher, blinded to all outcome measures,performed all hyperinflation treatments. Anassistant physiotherapist collected sputumand recorded all measurements. It was notpossible for the second physiotherapist to beblinded as the assistant physiotherapists wereassisting in the treatment of the patient.

Manual hyperinflation

Manual hyperinflation breaths were deliv-ered using a Mapleson C anaesthetic circuitusing a 10 l/min fresh gas flow (Ohmeda

FIGURE 1: Flowchart of the treatment procedure: MHI = manual hyperinflation; VHI = ventilator hyperinflation.

PRI 7(2)_3rd/crc 20/5/02 2:00 PM Page 102


MA 105). These breaths had a slow inspira-tion for 3 s duration to a peak airwaypressure of 40 cm H

2O, as measured by an

inline manometer (Bird pressure manome-ter, Tag Medical). A 2-s end inspiratorypause was followed by an uninterruptedexpiration during which the bag was heldcompressed. The manual hyperinflationtreatment consisted of six sets of six hyper-inflation breaths, each set being followed bysix tidal breaths to a peak airway pressureof 20 cm H

2O. The manual hyperinflation

treatment was of 20 minutes duration.

Ventilator hyperinflation

In volume control, ventilators were set tosix breaths per minute, an inspiratory flowof 20 l/min, incorporating a square waveform and a 2-s end inspiratory pause. Tidalvolume was increased in 200 ml incrementsuntil a peak airway pressure of 40 cm H

2O

was achieved. Once this pressure wasreached, six mechanical breaths were deliv-ered to the patient. After this, the ventilatorwas reset to pre-treatment variables and thepatient rested for 30 s. The sequence wasthen repeated. The treatment lasted 20 min-utes and consisted of six sets of sixventilator hyperinflation breaths.

Endotracheal suction was performedusing size 12 Baxter catheters (BaxterHealth Care Corporation, Edward CriticalCare Division, Irvine, CA 92714–5686,USA). The same catheter was used foreach suction pass. The patients were suc-tioned three times throughout theprocedure following every second set ofhyperinflation breaths. Four millilitres ofnormal saline was used for lavaging downthe endotracheal tube and 1 ml was used towash through the suction catheter at theend of the treatment. An arterial line, ECGand pulse oxymetry were used to monitorthe subject continuously (Hewlett Packard

monitoring systems M1046-9001b, HPGmBH, Boedhingn, Germany).

Measurement

Sputum

Once subjects were initially turned into theappropriate flat sidelying position they weresuctioned once, and the sputum was dis-carded as it was thought to be related topositional change rather than the treatment.In order to measure the wet weight ofsputum suctioned, the secretions were col-lected in a disposable sputum trap (40 ccspecimen trap, Sherwood Medical, StLouis, MO 63103, USA; model no.8884–724500). This was weighed beforeand after the treatment with a digital weigh-ing scale (PC400, Hewlett Packard,Boedhingn, Germany). Before each mea-surement the scale was calibrated accordingto manufacturer’s guidelines.

Static pulmonary compliance

After 10 minutes’ rest before raising thefoot of the bed, baseline measures of staticpulmonary compliance were recorded.Once the treatment was completed subjectswere returned to the horizontal sidelyingposition and remained there for up to twohours in accordance with their pressure careschedules. Measurements of static pul-monary compliance were recordedimmediately and 30 minutes after treatment.All measures were performed with subjectsin a flat sidelying position.

Static lung compliance was calculatedby use of the formula V

T/ IP – PEEP, where

VT

is tidal volume, IP is inspiratory pres-sure and PEEP is positive end expiratorypressure (Nunn, 1993). The exhaled tidalvolumes, end inspiratory plateau pressure

PRI 7(2)_3rd/crc 20/5/02 2:00 PM Page 103


and PEEP were read from the display on theventilator by the assistant physiotherapist.Three readings of static pulmonary compli-ance were taken, end inspiratory plateaupressure was achieved using a 1.5 s pause atend inspiration which was programmed intothe ventilator. The ventilators used wereeither the Bear 1000 (Bear) or the BennettsStar Ventilator (Bennetts).

Data management

The sputum wet weight produced duringmanual hyperinflation and ventilator hyper-inflation on each day of measurement wasadded to give a total value of sputum wetweight for each of the techniques. Thesevalues were then used for comparisons ofsputum wet weight for the two techniquesby use of a one-way ANOVA. The data forstatic pulmonary compliance was combinedover the two days of measurement giving amean value for manual hyperinflation andventilator hyperinflation at each time point.These variables were analysed by use of anANOVA for repeated measures. Probabilityvalues of p < 0.05 were deemed to be sig-nificant. Data is expressed as mean (95%confidence interval).

RESULTS

Seventeen males and three females fulfilledthe criteria for inclusion in the study. Nosubjectsts were withdrawn. The mean ageof the subjects was 45.2 years (range 16–86years). All subjects were intubated andmechanically ventilated with a mean FiO

2

of 0.44 (range 0.3–0.6). The descriptivedata for subjects are presented in Table 1.

There was no significant difference inthe sputum wet weight between treatments(p = 0.11) (Table 2). The mean difference insputum production and 95% confidenceinterval was 2.65 grams (range 1.79–3.54).

Both hyperinflation techniques significantlyimproved static pulmonary compliance (p <0.001) (Table 3). Manual hyperinflationproduced an average of 11.47% and 9.75%improvement in static pulmonary compli-ance immediately and 30 minutespost-treatment and ventilator hyperinflationproduced a 9.8% and a 11.58% increase atthe same time intervals.

There were no adverse changes in MAP,heart rate or SaO

2nor was there an increase

in inotropic support during any treatmentover the two days of measurement.

DISCUSSION

Physiotherapists use manual hyperinflationfor the treatment and prevention of atelecta-sis and the mobilization of secretions(Rothen et al., 1993; Tweed et al., 1993;Webber and Pryor, 1993; Singer et al.,1994; Stiller et al., 1996; Hodgson et al.,2000). Although manual hyperinflation hasbeen shown to be an effective technique inthe management of intubated patients it hasseveral limitations. These include discon-nection of the patient from the ventilatorresulting in loss of PEEP, poor control ofairway pressure and flow and the fraction ofinspired oxygen being 1.0 (Ciesla, 1996;Clarke et al., 1999). These limitations areeliminated with ventilator hyperinflation.We found that hyperinflation using the ven-tilator is as effective as manualhyperinflation in clearing excess pulmonarysecretions and improving static pulmonarycompliance.

Ten of the subjects were acute quadri-plegic patients. However, it was not feltthat this influenced the results of the pre-sent study, as there is no evidence thatthese patients respond differently fromothers to physiotherapy treatment. Acrossover design was used to control forthe effect of time of day of treatment, and

PRI 7(2)_3rd/crc 20/5/02 2:00 PM Page 104


TABLE 1: Subject data

Subject/sex Age Diagnosis CXR PaO2/FiO

2APACHE II Outcome

(years)

1 m 86 Resp failure LLL coll 400.0 19 Died2 f 18 C5 quad NAD 467.3 10 Survived3 f 33 Guillain-Barré RLL coll 372.5 12 Survived4 m 79 Sepis/pneum Ling/LLL cons 460.0 22 Survived

RLL coll5 m 77 Total gastrectomy LLL coll 210.0 22 Survived6 m 38 C5 quad LLL/LML coll 247.5 14 Survived

RUL cons7 m 74 C sp fusion LLL coll 365.0 14 Survived8 m 57 C1 fusion RML/RLL coll 277.5 12 Survived9 m 39 C4 quad NAD 292.5 12 Survived10 m 16 C6 quad LLL coll 295.8 10 Survived11 m 32 C1 quad RLL cons 375.0 10 Survived12 m 61 C5 quad RUL cons 256.0 14 Survived

LLL cons13 m 48 C1 quad RLL coll 287.5 12 Survived

LLL coll14 m 20 C5 quad RUL coll 235.0 14 Survived15 m 42 Multi-trauma LLL/LUL cons 118.0 16 Survived

RLL cons16 f 42 Multi-trauma LLL coll 197.5 16 Survived

RLL cons17 m 41 C4 quad LLL coll 200.0 12 Survived18 m 19 Multi-trauma RUL col 230.0 14 Survived

LUL collLing cons

19 m 39 C5 quad LLL/LML cons 144.0 12 SurvivedRUL coll

20 m 43 Multi-trauma LLL coll 353.5 10 Survived

m = male; f = female; quad = quadraplegic; resp = respiratory; pnuem = pneumonia; C sp = cervial spine;coll = collapse; cons = consolidation; NAD = no abnormality detected; LLL = left lower lobe; LUL = leftupper lobe; LML = left middle lobe; RLL = right lower lobe; RUL = right upper lobe; RML = right middlelobe; ling = lingula.

TABLE 2: Mean sputum production (95% confidence interval) for the two days of physiotherapy and themean total sputum wet weight (95% confidence interval)

Day Mean hyperinflation Ventilator hyperinflation(g) (g)

1 6.87 (5.60–8.11) 5.41 (414–6.68)2 6.18 (4.83–7.53) 6.62 (5.38–7.86)Mean (CI) 6.53 95.86–7.20) 6.01 (4.83–7.19)

PRI 7(2)_3rd/crc 20/5/02 2:00 PM Page 105


also to perform four treatments on one daywas impractical.

Static lung compliance and sputum wetweight were used as outcome measures toevaluate the eff icacy of physiotherapytreatment. These short-term outcome mea-surements have been used in previousstudies (Jones et al., 1992; Hodgson et al.,2000) and were chosen because they reflectphysiological changes in patients in inten-sive care who are immobilized andintubated. These patients may have areduction in pulmonary compliance due tomechanical ventilation (Oh, 1988),increased mucous production (Jones et al.,1997) and impaired mucociliary clearancemechanisms (Konrad et al., 1994). Thesefactors have been associated with anincreased risk of sputum retention, atelec-tasis and pneumonia (Anderson andJenkins, 1993; Konrad et al.,1994). Bothdry and wet weights of sputum have beenpreviously used to assess the outcome ofphysiotherapy intervention. It has beensuggested that the wet weight may be influ-enced by the presence of saliva (Rossmanet al., 1982). The measurement of driedsputum weight may eliminate this problem.There is, however, no evidence for reliabil-ity of specif ic drying protocols in theliterature. In addition, a recent studyreported a strong linear relationshipbetween the wet and dry weight of sputum(Cecins et al., 1999). For these reasons and

because the subjects were intubatedpatients in whom the contamination ofsputum with saliva was minimized, it wasdecided to use sputum wet weight as theoutcome measure.

The aim of lung hyperinflation is to re-expand atelectasis (Stiller et al., 1990;Stiller et al., 1996), mobilize secretions(Hodgson et al., 2000) and prevent orreduce the incidence of nosocomial pneu-monia in intubated patients. Currently,there is only limited evidence with regardto the effectiveness of physiotherapy treat-ment in the prevention of nosocomialpneumonia. In a previous study of manualhyperinflation twice as many patients in thecontrol group (no manual hyperinflation)developed nosocomial pneumonia(Ntoumenoupolous et al., 1988). Althoughthis result did not reach statistical signifi-cance, it was an important clinical finding.

Atelectasis is a common clinical prob-lem in the intubated and ventilated patient(Oh, 1988; Jones 1997) and, if prolonged,may lead to hypoxaemia, pulmonary infec-tion and fibrosis (Marini et al., 1979). Inaddition, loss of lung volume and atelecta-sis lead to a progressive reduction inpulmonary compliance making ventilationmore difficult (Oh, 1988).

Recruitment manoeuvres such as hyper-inflation have been shown to improve bothatelectasis (Scholten et al., 1985; Stiller etal., 1990; Rothen et al., 1993; Tweed et al.,

TABLE 3: Means (mean percentage improvement) (95% confidence interval)in static pulmonary compliancebefore and after physiotherapy

Hyperinflation Pre-treatment Post-treatment 30 min post-treatment(ml/cm H

2O) (ml/cm H

2O) (ml/cm H

2O)

Manual 46.2 51.5 (11.5%) 50.7 (9.7%)(41.5–50.9) (46.6–56.4) (46.5–54.9)

Ventilator 44.9 49.3 (9.8%) 50.1 (11.6%)(40.5–49.3) (45.7–52.9) (45.5–54.7)

PRI 7(2)_3rd/crc 20/5/02 2:00 PM Page 106


1993) and static pulmonary compliance(Jones et al., 1992; Hodgson et al., 2000).The results of the present study were con-sistent with previous findings showing animprovement in static pulmonary compli-ance with both forms of hyperinflation(Jones et al., 1992; Hodgson et al., 2000).This is the f irst study to examine theeffects of ventilator hyperinflation on staticpulmonary compliance.

Although no cardiovascular changeswere observed in the current study, manualhyperinflation has previously been associ-ated with alterations in cardiac output andarterial blood pressure (Laws and McIntyre,1969; Gormezano and Branthwaite, 1972;Singer et al., 1994). Singer et al. (1994)suggested that the increased respiratory rateused by physiotherapists while manuallyhyperinflating, led to the development ofgas trapping which reduced cardiac output.In contrast ventilator hyperinflation allowsthe physiotherapist to observe the flow pat-terns generated by each breath, therebyensuring full exhalation before the nextbreath is delivered. The risk of cardiovascu-lar compromise due to gas trapping shouldtherefore be reduced.

Research regarding the effects of manualhyperinflation has been difficult to under-take because of the variety of circuits usedin clinical practice, confusion regarding thedefinition of hyperinflation and inconsis-tency in physiotherapist technique(McCarren and Chow, 1996; Rusterholz andEllis, 1998; Clarke et al., 1999). Ventilatorhyperinflation is an alternative for bothphysiotherapy treatment and research. Thecomputer software of newer ventilators pro-vides accurate measurement of pressure,volume, flow and fraction of inspiredoxygen. The variables of each ventilatorhyperinflation breath are quantified leadingto improved consistency in clinical practiceand accurate data collection for research.

Ventilator hyperinflation was found to beas good as manual hyperinflation in sputumclearance and improving static pulmonarycompliance. This suggests that both shouldbe equally effective in treating atelectasis. Italso enables both ventilator and manualhyperinflation, as defined in this study, tobe used in future research examining patientoutcomes in intensive care.

REFERENCES

Anderson J, Jenkins S. Physiotherapy problems andtheir management. In: Webber B, Pryor J (eds),Physiotherapy for Respiratory and Cardiac Prob-lems. London: Churchill Livingstone, 1993;226–227.

Bendixen H, Hedley-White J, Laver MB. Impairedoxygenation in surgical patients during generalanaesthesia with controlled ventilation. A conceptof atelectasis. New England Journal of Medicine1963; 269: 991–996.

Blanch L, Fernandez R, Valles J, Sole J, Roussos C,Artigas A. Effect of two tidal volumes on oxy-genation and respiratory system mechanicsduring the early stages of adult respiratory dis-tress syndrome. American Journal of Respiratoryand Critical Care Medicine 1994; 9: 151–158.

Brown S, Stansbury D, Merrill E Linden G, Light R.Prevention of suction related arterial oxygendesaturation. Chest 1987; 83: 621–627.

Cecins N, Jenkins S, Pengelly J, Ryan G. The activecycle of breathing techniques — to tip or not totip? Respiratory Medicine 1999; 93: 660–665.

Ciesla N. Chest physical therapy for patients in theintensive care unit. Physical Therapy 1996; 76:609–625.

Clarke R, Kelly B, Convery P, Fee J. Ventilatorycharacteristics in mechanically ventilated patientsduring manual hyperinflation for chest physio-therapy. Anaesthesia 1999; 54: 936–940.

Clement A, Hubsch S. Chest physiotherapy by the‘bag squeezing’ method. Physiotherapy 1968; 54:355–359.

Dorrington K, Radcliffe F. Effect of single inflationof the lungs on oxygenation during extracorporealcarbon dioxide removal in experimental respira-tory distress syndrome. Intensive Care Medicine1999; 17: 469–474.

Gormezano J, Branthwaite M. Effects of physiother-apy during intermittent positive pressure

PRI 7(2)_3rd/crc 20/5/02 2:00 PM Page 107


ventilation. Changes in arterial blood gas ten-sions. Anesthesia 1972; 27: 258–264.

Hodgson C, Denehy L, Ntoumenopoulos G, Santa-maria J, Carroll S. An investigation of the earlyeffects of manual lung hyperinflation in criticallyill patients. Anaesthesia and Intensive Care 2000;28: 255–261.

Imle P, Klemic N. Methods of airway clearance:coughing and suctioning. In: Mackenzie C (Ed.),Chest Physiotherapy in the Intensive Care Unit(second edition). Baltimore, MD: Williams &Wilkins, 1989; 181.

Johnson N, Marini JJ, Pierson D, Huson L. Acutelobar atelectasis. Effect of chest percussion andpostural drainage on resolution. AmericanReview of Respiratory Diseases 1987; 135:A433.

Jones A. Physiotherapy in intensive care. In: Oh T(Ed.), Intensive Care Manual (fourth edition).London: Butterworth-Heinemann, 1997; 28–32.

Jones A, Hutchinson R, Oh T. Effects of bagging andpercussion on total static compliance of the respi-ratory system. Physiotherapy 1992; 78: 661–666.

Konrad F, Schreiber T, Brecht-Kraus D, GeorgieffM. Mucociliary transport in ICU patients. Chest1994; 105: 237–241.

Laws AK, McIntyre RW. Chest physiotherapy: aphysiological assessment during IPPV in respira-tory failure. Canadian Anaesthetic SocietyJournal 1969; 16: 487–493.

McCarren B, Chow MC. Manual hyperinflation: adescription of the technique. Australian Journalof Physiotherapy 1996; 42: 203–208.

Marini J, Pierson D, Hudson L. Acute lobar atelecta-sis: a prospective comparison of fibreopticbronchoscopy and respiratory therapy. AmericanReview of Respiratory Diseases 1979; 119:971–978.

Ntoumenopoulos G, Gild A, Cooper D. The effect ofmanual lung hyperinflation and postural drainageon pulmonary complications in mechanicallyventilated patients. Anaesthesia and IntensiveCare 1988: 26: 492–496.

Nunn J. Nunn’s Applied Respiratory Physiology(fourth edition). Oxford: Butterworth-Heine-mann, 1993; 58.

Oh T. Ventilation — matching man, mode andmachine. British Journal of Hospital Medicine1988; 40: 216–220.

Pelosi P, Cadringher P, Bottino N, Panigada M, Car-rieri F, Riva E, Lissoni A, Gattinoni L. Sigh inacute respiratory distress syndrome. AmericanJournal of Respiratory and Critical Care Medi-

cine 1999; 159: 872–880.Rhodes J. A pressure regulated bagging device — a

pilot study. South African Journal of Physiother-apy 1987; 43: 117–120.

Rossman C, Waldes R, Sampson D, Newhouse M.Effect of chest physiotherapy on the removal ofmucus in patients with cystic fibrosis. AmericanReview of Respiratory Diseases 1982; 126:131–135.

Rothen H, Sporre B, Engberg G, Wegenius G,Hedenstierna. Re-expansion of atelectasisduring general anaesthesia: a computed tomo-graphy study. British Journal of Anaesthesia1993; 71: 788–795.

Rusterholz B, Ellis E. The effect of lung complianceand experience on manual hyperinflation. Aus-tralian Journal of Physiotherapy 1998; 44: 23–28.

Scholten D, Novak R, Snyder J. Directed manualrecruitment of collapsed lung in intubated andnon-intubated patients. Annals of Surgery 1985;51: 330.

Singer M, Vermaat J, Hall G, Latter G, Patel M.Haemodynamic effects of manual hyperinflationin critically ill mechanically ventilated patients.Chest 1994; 106: 1182–1187.

Stiller K, Geake T, Taylor R, Hall B. Acute lobaratelectasis: a comparison of two chestphysiotherapy regimens. Chest 1990; 98:1336–1340.

Stiller K, Jenkins S, Grant R, Geake T, Taylor J, HallB. Acute lobar atelectasis: a comparison of fivephysiotherapy regimens. Physiotherapy Theoryand Practice 1996; 12: 197–207.

Suter P, Fairley H, Isenberg M. Effect on tidalvolume and positive end expiratory pressure oncompliance during mechanical ventilation. Chest1978; 7: 158–162.

Tweed W, Phua W, Chong E, Lim E, Lee TL. Tidalvolume, lung hyperinflation and arterial oxygena-tion during general anaesthsesia. Anaesthesia andIntensive Care 1993; 21: 806–810.

Webber B, Pryor J. Physiotherapy skills: techniquesand adjuncts. In: Webber B, Pryor J (eds),Physiotherapy for Respiratory and CardiacProblems. London: Churchill Livingstone,1993; 157–158.

Address for correspondence: Susan Berney, Physio-therapy Department, Austin and RepatriationMedical Centre, Studley Road, Heidelberg, Victoria3084, Australia (E-mail: [email protected]).

Submitted March 2001; accepted November 2001

PRI 7(2)_3rd/crc 20/5/02 2:00 PM Page 108

a comparison of the effects of manual and ventilator ... ventilator hyperinflation on static lung...

Documents