ISSUES IN PULMONARY NU

RSING

Comparison of effects of manual versus ventilator

hyperinflation on respiratory compliance

and arterial blood gases in patients undergoing

mitral valve replacement

Faizan Ahmed, MPT,a Aamir Mohammed Shafeeq, MPT,b Jamal Ali Moiz, MPT,c

and Muhammed Abid Geelani, MS, MChd

FromIndia;MilliaCardiotal, Un

CorresColony

0147-9� 2010doi:10.

HEAR

OBJECTIVE: To compare the effects of manual hyperinflation (MHI) and ventilator hyperinflation (VHI)

delivered to completely sedated and paralyzed patients undergoingmitral valve replacement (MVR) while

maintaining minute ventilation.

METHODS: This was a randomized study with a 2-group, pre-test, post-test experimental design. Effects

of hyperinflation were studied on static compliance (Cstat), dynamic compliance (Cdyn), oxygenation

(PaO2:FIO2), partial pressure of carbon dioxide in arterial blood (PaCO2), and cologarithm of activity of

dissolved hydrogen ions in arterial blood (pH). A sample of 30 patients in the immediate postoperative

phase of MVR surgery were included in the study.

RESULTS: No significant differences were found between the groups. Significant improvements were

found in oxygenation at both 1 minute and 20 minutes after MHI, but only at 1 minute after VHI (P <.05). VHI led to improved Cdyn (P < .05).

CONCLUSION: In the immediate postoperative phase of MVR, both techniques produced similar effects

on respiratory compliance and oxygenation. MHI produced longer lasting improvements in oxygenation

than VHI, whereas VHI produced better improvements in dynamic compliance. PaCO2 and pH were

maintained by both. (Heart Lung� 2010;39:437–443.)

Patients who have undergone mitral valve re-

placement (MVR) are said to be at a greater

risk of developing serious respiratory compli-

cations than those who have undergone cardiac sur-

gery for coronary artery disease.1 Patients who have

undergone MVR and have a history of rheumatic

heart disease usually have long-standing pulmonary

hypertension. During this surgery, the patients are

subjected to general anesthesia, median sternot-

the aHamdard University (Jamia Hamdard), New Delhi,bPushpanjali Crosslay Hospital, Vaishali, UP, India; cJamiaIslamia (University), New Delhi; and dDepartment of

thoracic and Vascular Surgery, Govind Ballabh Pant Hospi-iversity of Delhi, New Delhi.

ponding author: Faizan Ahmed, V-102, Taj Enclave, Geeta, Delhi 110031, India. E-mail: faizanmpt@gmail.com

563/$ - see front matterElsevier Inc. All rights reserved.

1016/j.hrtlng.2009.10.006

T & LUNG VOL. 39, NO. 5

omy, mechanical ventilation, and cardiopulmonary

bypass. All of these are known to considerably affect

lung function1 and respiratory compliance.2 Induc-

tion of anesthesia is closely followed by formation

of atelectasis, and it may take weeks for the lung to

recover to the preoperative status.2

Manual hyperinflation (MHI) is known to improve

pulmonary compliance and oxygenation (PaO2:

FIO2).3,4 Ventilator hyperinflation (VHI), on the other

hand, is a newer technique and comparatively less

popular than MHI. Yet, few studies comparing MHI

and VHI have found VHI to be an equally beneficial5

alternative of MHI or even a better alternative than

MHI.6 It needs to be investigated whether any of

these techniques can be effectively used to improve

respiratory compliance and oxygenation in the im-

mediate postoperative phase in patients who have

undergone MVR surgery. Also, literature related to

www.heartandlung.org 437

Manual versus ventilator hyperinflation Ahmed et al

the isolated effects of any of these techniques, the ef-

fects of hyperinflation without hyperventilation, or

the effects of hyperinflation at constant FIO2 of 1.0,

is scarce. Compliance is said to be best measured

at the ideal state of complete paralysis, but literature

on effects of hyperinflation at complete paralysis

could not be found by the authors. Therefore, this

study was conducted to investigate and compare

the efficacy of MHI and VHI techniques in improving

compliance and oxygenation when these techniques

were performed in isolation and at the state of com-

plete paralysis in the immediate postoperative phase

in patients who have undergone MVR surgery.

MATERIALS AND METHODS

This was a randomized, 2-group interventional

study using MHI or VHI as the intervention. A sample

of 30 patients who underwent MVR because of

rheumatic heart disease were enrolled from the

Cardiothoracic and Vascular Surgery department.

Institutional Board of Studies approval and written in-

formed consent from each patient were obtained. The

sample consisted of patients who were between 18 and

40 years old, had cardiopulmonary bypass time be-

tween 50 and 120 minutes, were completely sedated

and paralyzed under the influence of anesthesia,

were between 3 and 4 hours after surgery, were intu-

bated and mechanically ventilated (at volume control

mode), and were cardiovascularly stable. Cardiovascu-

lar stability was based on hemodynamics and medica-

tions. A patient was considered cardiovascularly

unstable when the patient required inotropic support

of more than 10 mg/kg body weight/minute of dopa-

mine and/or.1 mg/kgbody weight/minute of adrenaline,

the patient was on intraaortic balloon pump, there was

continuous mediastinal drainage of more than 6 mL/kg

body weight/min, the systolic blood pressure was less

than 80 mm Hg, or the heart rate was more than 140

beats/min. Only patients with severe mitral stenosis

and moderate to severe pulmonary hypertension

were included in the study. The patients had the

same type of lesion of mitral valve (stenotic lesions

were included, and regurgitant lesions were excluded).

The diagnoses of rheumatic heart disease were based

on echocardiography by 2-dimensional echocardiogra-

phy and Doppler studies. All patients were adminis-

tered the same anesthesia during the surgery and

received similar medications after surgery, that is, do-

pamine, adrenaline, and nitroglycerine were adminis-

tered to all the patients, but the dosage of each was

variable as per the hemodynamic status of the patient.

Patients with obstructive lung disease, patients

with positive end-expiratory pressure (PEEP) 10

438 www.heartandlung.org

cmH2O or more, and patients with any of the compli-

cations for which the intervention (hyperinflation)

was contraindicated were excluded.

The withdrawal criteria included unstable cardio-

vascular status of the patient, including arrhythmias

compromising the cardiovascular status of the pa-

tient and any other serious complication arising dur-

ing hyperinflation.

Procedure

Postsurgically, between 3 and 4 hours after sur-

gery, all patients were placed in a supine position.

The breath frequency was set as 12 breaths/min on

the ventilator at volume control mode. Care was

taken to ensure that no patient was triggering a spon-

taneous breath and the measured breath frequency

equalled the preset breath frequency on volume

control mode of mechanical ventilation. Therefore,

any patient capable of triggering a spontaneous

breath was considered not to be completely sedated

and paralyzed, and was not included in the study.

Thus, the effects were studied only for hyperinflation

under the state of pure positive pressure ventilation.

Tidal volume delivered was 10 mL/kg body weight or

550 mL, whichever was less. If any patient was sus-

pected of having secretions in the central airways,

endotracheal (ET) suctioning was performed approx-

imately 30 minutes before intervention, so that

excess secretions did not influence the dependent

variables. Fifteen minutes before intervention, the

FIO2 of the ventilator was set as 1.0. After 15 minutes

of being ventilated at FIO2 1.0, a pre-intervention (ie,

pre-hyperinflation) measure of the dependent vari-

ables (ie, static and dynamic compliance, PaO2:FIO2,

PaCO2, and pH) was performed. After recording the

pre-intervention (baseline) measurements of the

dependent variables, the intervention (ie, VHI or

MHI) was given. The preset breath frequency was de-

creased from 12 breaths/min to 8 breaths/min during

hyperinflation, and the tidal volume was increased

to 150% of the tidal volume the patient had been

breathing at the baseline to maintain the minute

ventilation. During the entire procedure, all precau-

tions were undertaken to prevent infection to the pa-

tient. MHI was performed by the same person for all

the patients. Heart rate, electrocardiogram, arterial

blood pressure (invasive), arterial blood oxygen

saturation (SpO2), and temperature were monitored

using bedside monitor (Philips Intelli Vue MP 40,

Philips International B.V., The Netherlands).

Manual hyperinflation. MHI was given by ‘‘bag

squeezing’’ using the self-inflating Intersurgical 1.5-

liter manual resuscitation bag with oxygen reservoir

SEPTEMBER/OCTOBER 2010 HEART & LUNG

Ahmed et al Manual versus ventilator hyperinflation

(Intersurgical Ltd, Berkshire, UK) by the 2-handed

technique. The bag was connected to oxygen supply

at a flow rate of 15 L/min. A disposable PEEP valve

(Vital Signs Inc, Totowa, NJ) was connected to the

resuscitation bag. Hyperinflation breaths with a

2-second inspiration, 2-second inspiratory pause,

and 1-second expiration were given for 3 minutes,

at a rate of 8 breaths/min and FIO2 1.0.

Ventilator hyperinflation. VHI was given by a Sie-

mens Servo-300 (Siemens-Elema AB, Solna, Sweden)

mechanical ventilator by increasing the inspiratory

tidal volume to 150% of the tidal volume the subject

had been breathing at baseline. Breathing rate was

decreased to 8 breaths/min, and FIO2 was maintained

at 1.0 for the duration of the technique (3 minutes).

The upper pressure limit was set at 35 mm Hg.

Measurements

Postintervention measures of the dependent vari-

ables were taken at 1 minute after the intervention

and again 20 minutes postintervention. PaO2, pH,

and PaCO2 were measured by the Arterial Blood Gas

analyzer (Radiometer ABL 800 Basic, Copenhagen,

Denmark). Fresh arterial blood samples were taken

from the arterial line in situ at all 3 times of data col-

lection. A discard sample was removed each time to

clear the tubing. Peak inspiratory pressure, end in-

spiratory plateau pressure, PEEP, and exhaled tidal

volume were read from the Siemens Servo ventilator

for calculating static and dynamic compliance. A

2-second ‘‘inspiratory hold’’ was incorporated with

the use of the inspiratory hold knob on the ventila-

tor, after complete inspiration, for the measurement

of exact plateau pressure. The formula used for static

compliance (Cstat) was corrected eVt/Plateau pres-

sure – PEEP, and the formula used for dynamic com-

pliance (Cdyn) was corrected eVt/Peak Inspiratory

pressure – PEEP, where corrected eVt is exhaled tidal

volume with compensation for tubing compression.

Data analysis

Measured values were compared for the 2 tech-

niques by the use of a t test. Changes within the

groups 1 minute and 20 minutes after hyperinflation

were analyzed using repeated-measures analysis of

variance. Probability values of P < .05 were deemed

to be significant for PaO2:FIO2, PaCO2, Cstat, and Cdyn.

For pH, significance was set at P < .01.

RESULTS

The demographic data of the selected patients are

presented in Table I. The results are shown in Tables II

and III. There was no significant difference between

HEART & LUNG VOL. 39, NO. 5

groups in terms of age, height, weight, cardiopulmo-

nary bypass time, and body mass index. Baseline

values were not found to be significantly different be-

tween the groups in case of any dependent variable.

Also, there were no significant differences between

the groups in terms of changes in any of the depen-

dent variables (Figs 1-3). There was a significant im-

provement (P = .0001) in PaO2:FIO2 ratio in the MHI

group 1 minute and 20 minutes after MHI (Table II,

Fig 3), whereas in the VHI group there was a signifi-

cant improvement (P = .0001) in PaO2:FIO2 only at

1 minute after VHI (Table III, Fig 3). Cdyn showed

a significant improvement at 1 minute after MHI

(P = .001) (Table II, Fig 2). The remaining dependent

variables (except PaCO2) in both the groups showed

improvements that were statistically nonsignificant.

pH and PaCO2 showed nonsignificant and negligible

change after both techniques (Tables II and III). There

were no adverse changes in blood pressure, heart

rate, or heart rhythm during the intervention.

DISCUSSION

This study compared the effects of MHI and VHI on

arterial blood gas values, which included PaO2, PaCO2,

and pH, and the respiratory mechanics, which in-

cluded static and dynamic compliance, in sedated,

paralyzed, volume control mode mechanically venti-

lated, hemodynamically stable patients in the imme-

diate postoperative phase of MVR. This is perhaps

the first study to investigate the effects of hyperinfla-

tion while maintaining constant minute ventilation,

to examine the effects of VHI on dynamic compli-

ance, and to compare the effects of hyperinflation

techniques at the ideal state of complete paralysis.

Although MHI has been shown to be an effective

technique in the management of intubated patients,

it has several known limitations such as the need of

disconnection of the patient from the ventilator re-

sulting in loss of PEEP, poor control of airway pres-

sure and flow, and lack of accurate control of FIO2

delivered.5 These limitations are eliminated with

the use of VHI because it can give accurate tidal vol-

ume, easily limit airway pressure at the desired limit

of 35 mm Hg, maintain constant flow, and deliver an

accurate FIO2. The present study found that VHI is as

effective as MHI in improving pulmonary compli-

ance, although the improvements found were statis-

tically not significant (P < .05). For the purpose of

convenience, the effects of MHI or VHI that were ob-

served immediately 1 minute after the hyperinflation

are referred here to as ‘‘immediate effects,’’ and the

effects that were observed 20 minutes after the

hyperinflation are referred to as ‘‘delayed effects.’’

www.heartandlung.org 439

Table I

Patients’ demographic data

Variable MHI (mean ± SD) (N = 15) VHI (mean ± SD) (N = 15)

Age (y) 27.8� 6.9 29� 7.8Height (cm) 151.4� 7 153� 7.4Weight (kg) 44.3� 4.9 47.13� 8.3BMI (kg/m2) 19.3� 1.24 20.14� 3.42CPB time (min) 79.13� 16.76 81.86� 17.52Male 5 7Female 10 8

BMI, body mass index; CPB, cardiopulmonary bypass; SD, standard deviation.

Manual versus ventilator hyperinflation Ahmed et al

Because the improvements in Cstat after VHI were

better maintained at 20 minutes posthyperinflation

compared with MHI, if VHI is given with intention

of improving compliance, maintenance of Cstat

above the baseline for a longer duration may be

expected. The maintenance of the improved compli-

ance in the VHI group may be due to recruitment of

the collapsed alveoli by VHI initially, and later, main-

tenance of reinflation and resumption of almost nor-

mal functioning of the alveoli in the 20 minutes

post-VHI. Although a few alveoli might have recol-

lapsed after resumption of normal tidal volume

breaths, the majority of the alveoli may have re-

mained inflated, leading to maintenance of Cstat.

Constant maintenance of PEEP may be responsible

for this maintenance of improvement because VHI

does not require disconnection from the ventilator.

There may have been a better and greater recruit-

ment of the collapsed alveoli immediately after

MHI than by VHI as is shown by the better Cstat im-

mediately post-MHI than post-VHI, but the decline

in Cstat in the delayed effects of MHI signifies that

there may have been a cascade decrease in the num-

ber of reinflated alveoli due to recollapse, thus lead-

ing to a decline in compliance in the delayed effects

phase. This may mean that VHI has better late effects

on pulmonary mechanics than MHI, whereas MHI

has better immediate effects. Although improve-

ments in static and dynamic compliance were not

found to be statistically significant (except in the

immediate effects after VHI), a further study needs

to be conducted to determine whether there is any

statistically significant change in compliance or

not if the procedure, as done in this study, is

repeated 3 to 4 times within a period of 2 hours. If-

such a procedure results in a summation of

improvements achieved after each set of treatment,

440 www.heartandlung.org

it may dramatically improve the compliance. Such

an improvement would be of clinical value because

it may help reduce atelectasis and improve oxygen-

ation considerably.

The elastic properties of respiratory system deter-

mine the static compliance (Cstat), whereas dynamic

compliance (Cdyn), in addition to dependence on

elastic properties of respiratory system, also reflects

flow and resistance of the airways.7 Cdyn improved

more after VHI than MHI. The reason may be that

MHI requires disconnection from the ventilator

and connection to the bag and then disconnection

from the bag and reconnection to the ventilator.

Such connections and disconnections usually cause

disturbance of the ET tube, which may in turn irritate

the trachea. Tracheal irritation may cause broncho-

spasm or the disturbance of the ET tube may disturb

the mucus, which is usually collected at the opening

of the ET in situ near the carina. This disturbance of

mucus may cause narrowing of the ET tube, in turn

leading to an increase in airway resistance. Accord-

ing to the Poiseuille equation, the airway resistance

offered to the flow of air in the airways is inversely

proportional to the fourth power of the effective ra-

dius of the airway. This means that if the effective ra-

dius of the airway is reduced to half, the airway

resistance will increase by 16 times. The effective

radius denotes the radius of the airway that is actu-

ally available for the flow of air after the radius of the

airway has been compromised by the collected se-

cretions in that part of the airway.8 However small

the obstruction in the airway may be, the effects it

has on the airway resistance are exaggerated. This

was not the case with VHI, in which the ET tube is

not disturbed and considerably smoother flow may

be expected, which keeps airway resistance low. Par-

atz et al4 also found improvements in Cdyn after MHI.

SEPTEMBER/OCTOBER 2010 HEART & LUNG

Table II

Measurements of dependent variables in MHI group

Variable Pre-MHI (mean ± SD)1min post-MHI

(mean ± SD)20min post-MHI

(mean ± SD)

pH 7.425 � .02 7.438 � .02 (.04%)P= .65

7.429 � .02 (.03%)P= .79

PaO2:FIO2 508.86� 67.33 533.06� 66.13 (4.88%)a

P= .0001518.33� 65.88 (1.93%)a

P= .0001PaCO2 (mm Hg) 33.98� 1.77 33.38� 1.48 (�1.7%)

P= .1533.71� 1.45 (�.73%)

P= .65Cstat (mL/cm H2O) 35.07� 8.67 37.16� 6.45 (7.66%)

P= .1835.31� 6.61 (2.15%)

P= 1Cdyn (mL/cm H2O) 26.7� 5.90 27.91� 4.59 (5.49%)

P= .0727.24� 5.28 (2.59%)

P= 1

pH, power of hydrogen (cologarithm of the activity of dissolved hydrogen ions in arterial blood); PaO2:FIO2, oxygenation; PaCO2,partial pressure of carbon dioxide in arterial blood; Cstat, static compliance; Cdyn, dynamic compliance; SD, standard deviation.

Values in parentheses are mean percentage changes from the baseline measurements.aMeasurements are statistically significant at P < .05.

Ahmed et al Manual versus ventilator hyperinflation

The effects also were maintained at above baseline

at 30 minutes after MHI in their study. Suter et al7

also found improvements in Cstat and Cdyn on in-

creasing tidal volume from 10 to 15 L/min on the me-

chanical ventilator in mechanically ventilated

patients, but the measurements were taken during

ventilation at the increased tidal volume of 15 L/

min and not after returning to 10 L/min.

The absence of significant change in PaCO2 in any

group post-hyperinflation is in agreement with the

attempts made to maintain alveolar ventilation by

decreasing the respiratory rate and increasing the

tidal volume for hyperinflation. A maintained venti-

lation of the alveoli during hyperinflation (MHI or

VHI) may have prevented a marked washout of CO2

from the blood. The nonsignificant difference in

change of PaCO2 (immediate effects) between the

groups indicates that both the techniques were

equally effective in maintaining the minute ventila-

tion. Nonsignificant changes (significance accepted

at P < .01 for pH) observed in pH may be because

the value of PaCO2 was maintained because of con-

stant minute ventilation, and therefore, no corre-

sponding changes occurred in pH. Hyperinflation

achieved without hyperventilation by maintaining

minute ventilation has clinical importance in those

intensive care unit patients who require hyperinfla-

tion for secretion mobilization but have already

decreased levels of PaCO2. If hyperventilation is pro-

duced by hyperinflation in such patients, it may lead

to further respiratory alkalosis.

HEART & LUNG VOL. 39, NO. 5

It is important to note that the increase in PaO2:-

FIO2 was not due to increased FIO2 during MHI

because FIO2 was 1.0 throughout the duration of

the study. Also, PaO2 cannot be expected to increase

with increased ventilation of the lungs during hyper-

inflation, because minute ventilation was kept con-

stant by decreasing the respiratory rate from 12 to

8 breaths/min and increasing the tidal volume to

approximately 150% of baseline tidal volume. The

probable reason is that with the increase in the re-

cruitment of functional alveolar units after hyperin-

flation, there may have been an improvement in

the ventilation of that area of the lung, resulting in

improved ventilation-perfusion ratio, decreased

shunting of blood in the lungs, and improved oxygen

transport in the blood (ie, increased PaO2 of the pul-

monary venous blood). Most clinical interventions

apply ET suctioning after MHI. Because ET suction

may have hypoxic effects,9 the effects of improve-

ment in PaO2:FIO2 by MHI done before ET suction

may be contaminated by ET suctioning. Because

no ET suction or any other intervention was done

in the present study before or after hyperinflation,

the results are not expected to be influenced by

any intervention other than MHI or VHI.

In a similar study, Patman et al3 investigated effects

of MHI on Cstat and PaO2:FIO2 in patients within 4 hours

of coronary artery bypass graft surgery. They found ap-

proximately 4 times greater improvement in PaO2:FIO2

and approximately double the improvement in Cstat.

The great difference in the improvement in PaO2:FIO2

www.heartandlung.org 441

Fig 1 Changes in Cstat between groups. Cstat = static

compliance; MHI=manual hyperinflation; VHI= ven-

tilator hyperinflation; Post 1=measurement of depen-

dent variable (Cstat) at 1 minute after hyperinflation;

Post 20=measurement of dependent variable (Cstat) at

20 minutes after hyperinflation.

Fig 2 Changes in Cdyn between groups. Cdyn = dynamic

compliance;MHI=manual hyperinflation; VHI= ven-

tilator hyperinflation; Post 1=measurement of depen-

dent variable (Cdyn) at 1 minute after hyperinflation;

Post 20=measurement of dependent variable (Cdyn) at

20 minutes after hyperinflation.

Table III

Measurements of dependent variables in VHI group

Variable Pre-VHI (mean ± SD)1 min post-VHI

(mean ± SD)20 min post-VHI

(mean ± SD)

pH 7.42 � .05 7.41 � .06 (�.06%)P= 1

7.42 � .05 (.048%)P= .54

PaO2:FIO2 467.4� 56.08 490.6� 48.99 (5.27%)a

P= .0001477.1� 47.23 (2.46%)

P= .57PaCO2 (mm Hg) 34.9� 4.04 34.6� 4.44 (�1.05%)

P= .934.6� 4.78 (�.87%)

P= 1Cstat (mL/cm H2O) 33.2� 7.39 34.8� 6.92 (5.57%)

P= .0834.5� 6.30 (4.79%)

P= .13Cdyn (mL/cm H2O) 25.5� 4.42 27.2� 4.50 (6.62%)a

P= .00126.4� 4.49 (3.37%)

P= .12

pH, power of hydrogen (cologarithm of the activity of dissolved hydrogen ions in arterial blood); PaO2:FIO2, oxygenation; PaCO2,partial pressure of carbon dioxide in arterial blood; Cstat, static compliance; Cdyn, dynamic compliance; SD, standard deviation.

Values in parentheses are mean percentage changes from the baseline measurements.aMeasurements are statistically significant at P < .05.

Manual versus ventilator hyperinflation Ahmed et al

between the present study and the study by Patman et

al may be due to the difference in the average age of

the patients and the fact that minute ventilation was

maintained in the present study.

In a previous study, Berney and Denehy5 com-

pared MHI and VHI in intubated and ventilated

patients. Their results support the results of the

present study, except in the case of 30 minutes

post-VHI measurements of Cstat, when Cstat contin-

ued to increase until 30 minutes after VHI, making

the measured static compliance better at 30 minutes

post-VHI compared with the measurements re-

442 www.heartandlung.org

corded immediately after VHI. This was not observed

in the present study. Also, the other results that are

supported by Berney and Denehy’s study are not

found to be statistically significant. The reasons

may be different MHI bag type, different patient

population, and pressure-dependent hyperinflation

protocol in their study.

Savian et al6 found no significant difference

between the effects of MHI and VHI in general inten-

sive care patients. They used a self-inflating Laerdal

(Laerdal Medical, Norway) resuscitation bag and

found no significant improvements in oxygenation.

SEPTEMBER/OCTOBER 2010 HEART & LUNG

Fig 3 Changes in PaO2:FIO2 between groups. Pao2:

FIO2 = oxygenation; MHI=manual hyperinflation;

VHI= ventilator hyperinflation; Post 1=measurement

of dependent variable (PaO2:FIO2) at 1 minute after

hyperinflation; Post 20=measurement of dependent

variable (PaO2:FIO2) at 20 minutes after hyperinflation.

Ahmed et al Manual versus ventilator hyperinflation

The studies3,4 in which oxygenation showed im-

provement used a Mapleson circuit (Mapleson "C"

circuit and a 2 L reservoir bag [BS 3352, Warne Sur-

gical Products Ltd, Lurgan, Ireland]. In the present

study, although a self-inflating Intersurgical resusci-

tation bag (Intersurgical Ltd, Berkshire, UK) was

used, significant improvements in oxygenation

were found after MHI, which persisted significantly

above baseline at 20 minutes after hyperinflation.

LIMITATIONS

The tidal volume delivered to the lungs of the

patients may be related to the changes in PaO2:FIO2

ratio, but the authors were unable to measure the ex-

act tidal volume delivered during MHI. An esopha-

geal balloon was not used in the measurement

of compliance. Also, both the hyperinflation proce-

dures were volume dependent; therefore, the results

found may not be applicable to pressure-dependent

hyperinflation procedures. The findings of the pres-

ent study where MHI was delivered by Intersurgical

resuscitation bag may not be generalized to all the

bags because variability exists in the effects of differ-

ent bags, especially when an Intersurgical or Laerdal

bag is compared with the Mapleson-C circuit.

CONCLUSIONS

Both MHI and VHI produced similar effects on

respiratory compliance and oxygenation. Both tech-

niques improve static and dynamic compliance and

oxygenation. However, VHI seems to better improve

dynamic compliance. Both techniques are equally

effective in producing hyperinflation without caus-

HEART & LUNG VOL. 39, NO. 5

ing hyperventilation. Longer-lasting improvements

in oxygenation may be obtained by MHI, whereas

better dynamic compliance may be achieved using

VHI in the immediate postoperative phase of MVR.

The improved oxygenation may be of great advan-

tage in intubated patients who have atelectasis

and pulmonary shunting, which compromise the ox-

ygenation, whereas improved dynamic compliance

holds value in patients who have stiffer lungs result-

ing from atelectasis or segmental collapse. There-

fore, on the basis of the patient’s condition and

the need of a specifically targeting method, a choice

between the 2 techniques can be made by the clini-

cian. However, the clinical significance and benefits

of these improvements on outcomes are unclear.

Therefore, further studies are required to investigate

the effects of improved oxygenation and dynamic

compliance on clinical outcomes.

The authors thank Fuzail Ahmed for advice on statisti-

cal analysis, Kushal Madan of the Department of Physiol-

ogy, All India Institute of Medical Sciences, for

manuscript preparation, D. K. Tempe and Monica S. Tan-

don of the Department of Anesthesiology, Govind Ballabh

Pant Hospital, and Farah Khaliq of Department of Physiol-

ogy, University College of Medical Sciences, University

of Delhi, for valuable advice, and the staff of Department

of Cardiothoracic and Vascular Surgery, Govind Ballabh

Pant Hospital, for cooperation and assistance.

REFERENCES

1. Hachenberg T, Tenling A, Hansson HE, Tyden H,Hedenstierna G. The ventilation-perfusion relation and gasexchange in mitral valve disease and coronary artery diseaseimplications for anesthesia, extracorporeal circulation, andcardiac surgery. Anesthesiology 1997;86:809-17.

2. Weismann C. Pulmonary function after cardiac and thoracicsurgery. Anesth Analg 1999;88:1272-9.

3. Patman S, Jenkins S, Stiller K. Manual hyperinflation–effects onrespiratory parameters. Physiother Res Int 2000;5:157-71.

4. Paratz J, Lipman J, McAuliffe M. Effect of manual hyperinflationon hemodynamics, gas exchange, and respiratory mechanics inventilated patients. J Intensive Care Med 2002;17:317-24.

5. Berney S, Denehy L. A comparison of the effects of manual andventilator hyperinflation on static lung compliance and sputumproduction in intubated and ventilated intensive care patients.Physiother Res Int 2002;7:100-8.

6. Savian C, Paratz J, Davies A. Comparison of the effectiveness ofmanual and ventilator hyperinflation at different levels ofpositive end-expiratory pressure in artificially ventilated andintubated intensive care patients. Heart Lung 2006;35:334-41.

7. Suter PM, Fairley HB, Isenberg MD. Effect of tidal volumeand positive end expiratory pressure on compliance duringmechanical ventilation. Chest 1978;73:158-62.

8. Cloutier MM. Dynamic lung mechanics. In: Respiratory physiol-ogy. New Delhi: Elsevier; 2008. p. 33-49.

9. Ciesla ND. Chest physical therapy for patients in the intensivecare unit. Phys Ther 1996;76:609-25.

www.heartandlung.org 443