Critical Care Medicine

Volume 28 • Number 5 • May 2000

Copyright © 2000 Lippincott Williams & Wilkins, Inc.

1648

REVIEWS

Chest physiotherapy in mechanically ventilated children: A review

Martin F. Krause MD 1
Thomas Hoehn MD 1
1
From the Children's Hospital, Albert-Ludwigs-University, Freiburg, Germany.

Address requests for reprints to: Martin F. Krause, MD, University Children's Hospital, Mathildenstr. 1, D-79106
Freiburg, Germany. E-mail: Krause@kkl200. ukl.uni-freiburg.de

Copyright © 2000 by Lippincott Williams & Wilkins

Objective: Many physicians, nurses, and respiratory care practitioners consider chest
physiotherapy (CP) a standard therapy in mechanically ventilated children beyond the
newborn period. CP includes percussion, vibration, postural drainage, assisted coughing, and
suctioning via the endotracheal tube.

Data Sources: We searched the medical literature by using the key words "chest
physiotherapy" and "chest physical therapy" (among others) by means of the MEDLINE and
Current Contents databases.

Study Selection: Because of the paucity of objective data, we examined all reports dealing with
this topic, including studies on adult patients. For data extraction, not enough material existed
to perform a meta-analysis.

Data Synthesis: Despite its widespread use, almost no literature dealing with this treatment
modality in pediatric patients exists. Studies with mechanically ventilated pediatric and adult
patients have shown that CP is the most irritating routine intensive care procedure to
patients. An increase in oxygen consumption often occurs when a patient receives CP
accompanied by an elevation in heart rate, blood pressure, and intracranial pressure. CP
leads to short-term decreases in oxygen, partial pressure in the blood, and major fluctuations
in cardiac output. Changes in these vital signs and other variables may be even more
pronounced in pediatric patients because the lung of a child is characterized by a higher
closing capacity and the chest walls are characterized by a much higher compliance, thus
predisposing the child to the development of atelectasis secondary to percussion and
vibration.

Conclusion: CP in mechanically ventilated children may not be considered a standard

therapy. Controlled studies examining the impact of CP on the duration of mechanical
ventilatory support, critical illness, and hospital stay are needed. (Crit Care Med 2000;
28:1648-1651)

K EY W ORDS: chest physiotherapy; mechanical ventilatory support; children; vital signs;

closing capacity; chest wall compliance; suctioning; oxygen consumption; cardiac output;
intracranial pressure

Chest physiotherapy (CP) has been advocated in the field of pediatrics for >20 yrs and has emerged
as a standard treatment modality in young patients with pulmonary diseases [1] . Different techniques
of CP have been studied in a variety of different lung diseases in children with an almost
unpredictable effect.

A "positive effect" of CP has, in general, been defined as an acute improvement in the patient's
overall appearance, lung auscultation, blood gas tensions, and chest radiographs [2] . The volume of
expectorated lung secretions after CP will be underestimated because parts of the secretions will be
swallowed by the pediatric patient [3] .

CP has produced different effects on spontaneously breathing patients with a variety of acute
pulmonary diseases. No improvement in lung function occurred in children with exacerbated
bronchial asthma who received CP [4] . CP seems to be of little value in the treatment of
bronchiolitis. No favorable influence on lung mechanics [5] , a rapid recovery, or earlier discharge
from the hospital [6] was detected. Studies in adult patients with pneumonia [7] failed to detect any
influence on the duration of fever, clearing of radiodensities on chest radiographs, duration of
hospitalization, or mortality rate. In a controlled study with young adults [8] , CP led to prolonged
fever and hospital stay in the treated group.

In contrast, CP has emerged as a standard treatment modality for young patients with cystic
fibrosis. Its effectiveness has been proven by numerous studies [9] [10] . However, CP does not
improve pulmonary status in all chronic lung diseases. Wollmer et al. [11] studied adult patients with
acutely exacerbated chronic bronchitis. Percussion, assisted coughing, and postural changes did not
lead to improvements in this condition.

A beneficence has been reported in pediatric lung abscess in children aged >7 yrs [12] , acute lobar
atelectasis [13] , and lung complications after traumatic quadriplegia [14] . The latter two studies
included only adult patients. Therefore, we searched the literature for reports of the effectiveness of
CP in mechanically ventilated pediatric patients. Children usually develop respiratory failure
because of acute pulmonary diseases for which they receive mechanical ventilatory support. We
limited our search to studies involving patients treated with mechanical ventilatory support to
overcome respiratory failure caused by acute pulmonary diseases.

Literature Search.
We searched the literature by using the key words "chest physiotherapy" or "chest physical
therapy," "mechanical ventilatory support," "children," and "vital signs" with the help of the
MEDLINE and Current Contents databases. Reference lists of all related articles were checked for
further articles of interest.

Chest Physiotherapy Techniques.

The following CP techniques are thoroughly described in the international literature for use in
mechanically ventilated patients [1] [15] : a) manual percussion with proper cupping of hand or with a
face mask; b) vibration of the chest wall transmitting

1649

energy through the chest wall to loosen or move bronchial secretions; c) postural drainage using
gravity to move secretions from peripheral airways to the larger bronchi; d) assisted coughing to
transport loosened secretions from the bronchial tree to the trachea; and e) suctioning of secretions
via the endotracheal tube. Other CP techniques, such as with contact breathing or guided breathing,
have been advocated. No studies are available evaluating their effectiveness.

Studies in Mechanically Ventilated Children.

Despite our extensive literature search, we found only the following two studies that involve
mechanically ventilated children beyond the neonatal age. Reines et al. [16] describe the effects of CP
in 44 cardiac patients aged 3 months through 9 yrs. Vibration and percussion, postural drainage
(including the head-down position), and endotracheal suctioning were all used. In the CP group,
13/19 (68%) vs. 8/25 (32%) patients of the control group developed atelectasis after cardiac
surgery, with the consequence of a longer hospital stay for those patients in the CP group. The two
groups of patients showed no differences in age, incision site (sternotomy vs. lateral thoracotomy),
and percentage of cardiac lesions with left-to-right shunt. Reines et al. speculated that this apparent
reversal of expected results could be caused by a number of factors, such as an increase in
ventilation-perfusion mismatch resulting from mucus moving from peripheral to central airways,
lung compression from percussion, decrease in functional residual capacity because of pain induced
by CP and hypoventilation in the head-down position. This study did not provide data on the length
of mechanical ventilatory support postoperatively or the number of CP treatments during or after
mechanical ventilatory support. Zach et al. [17] observed a high solution quota (>95%) of atelectases
in children treated with CP. A minority of patients received mechanical ventilatory support. This
study used a historical control group for comparison of results, omitted distinguishing children
receiving mechanical ventilatory support from those breathing naturally, and did not describe the
CP techniques used.

Studies in Mechanically Ventilated Adult Patients.

Applying the data from studies evaluating the effectiveness of CP in mechanically ventilated adult
patients when considering benefits and burdens of CP in pediatric patients under these clinical
conditions may be inappropriate because the specific physiologic properties of lung and chest wall
of children differ from adults. All of the following studies used a variety of CP techniques
mentioned above.

Laws and McIntyre [18] were first to investigate the influence of CP on oxygenation and cardiac
output in six mechanically ventilated patients in 1969. Their introduction states: "Chest
physiotherapy has become accepted as an essential and valuable adjunct to the care of patients
undergoing artificial ventilation for respiratory failure." No significant differences in oxygenation
during and immediately after CP could be detected when comparing the results to the situation
before initiating CP. However, a rise or fall by as much as 50% of the initial cardiac output was
observed.

Gormezano and Branthwaite [19] looked for changes in gas exchange 5 to 30 mins after CP in 42
patients on ventilatory support for reasons of major general surgery, respiratory failure, and cardiac
diseases. Significant decreases in Pa O 2 and increases in Pa CO2 could be detected, especially in
those patients with low cardiac output or respiratory failure. Gormezano and Branthwaite postulated
that the increase in intrathoracic pressure exerted by CP could lower the cardiac output so that for a
given shunt effect, mixed-venous P O 2 and, hence, Pa O2 would fall.

Similar results were obtained by Connors et al. [20] in 22 mechanically ventilated patients. Connors
et al. claimed that an increase in ventilation-perfusion mismatch as responsible for the deterioration
in oxygenation. Decrease in Pa O 2 was more pronounced in a subgroup of patients with less
production of lung secretions. These findings were backed up by the results of Tyler et al. [21]
describing a decrease of Pa O2 by 18 mm Hg (75-57 mm Hg; 2.4, 9.9 to 7.6 kPa) during the
sequence of CP. Tyler et al. emphasized the necessity of a thorough monitoring of the patient's vital
signs during CP, especially for those patients with a low baseline Pa O 2 .

Results vary in studies involving the influence of CP on gas exchange in mechanically ventilated
patients. Mackenzie et al. [22] could not detect any negative effect of CP on oxygenation during or
after intervention in 47 patients. They were convinced that the use of an endexpiratory pressure of
5-10 cm H2 O (0.5-1.0 kPa) prevented the fall in Pa O2 observed by other authors. Mackenzie and
Shin [23] found an improvement in oxygenation immediately and 2 hrs after CP, accompanied by
small improvements in lung compliance and intrapulmonary shunt. They reported marked
individual changes for cardiac index and intrapulmonary shunt among the patients.

Weissman and Kemper [24] calculated a mean increase in oxygen consumption during CP of +62% ±
29% above the baseline level in 16 critically ill surgical patients. Weissman et al. [25] were able to
identify two different groups of patients regarding their hemodynamic response during CP. The
intervention resulted in an increase in oxygen consumption in all patients as verified by a
significantly higher oxygen extraction (35% to 50%) in the vascular periphery or by a higher
oxygen supply through the cardiorespiratory response that was reflected by an overall increase in
heart rate, systemic blood pressure, and minute ventilation. Those patients reacting with an
increase in oxygen extraction in the vascular periphery were those with an almost unchanged
cardiac output reflecting absent cardiac reserve.

The increased oxygen demand during CP is mainly because of increased muscle activity because
administration of a muscle relaxant (in combination with a sedative) leads to a complete abolition
of this side effect [26] . Increases in heart rate and systemic blood pressure are not influenced by
muscle relaxants because these side effects are a result of increased sympathetic output.

Our literature search detected a single clinical study dealing with physiologic aspects other than gas
exchange and hemodynamics after CP. MacLean et al. [27] proved that CP positively influences
maximum expiratory flow rates in mechanically ventilated patients, and this effect could even be
augmented when administering an increased pressure on the epigastric area of the patient treated.

Weissman et al. [28] demonstrated in 23 mechanically ventilated patients that CP met with the most
pronounced variations of vital signs when compared with other routine daily intensive care
activities. CP caused an average increase in oxygen consumption of 38% and an 35% increase in
CO2 production as well as marked increases in heart rate and systemic blood pressure when
compared with spontaneous movements, physical exam, presence of visitors, bathing, dressing, and
taking chest radiographs. Administration

1650

of 3 mug/kg fentanyl immediately before beginning CP clearly attenuated the hemodynamic

response (heart rate, systemic blood pressure, cardiac output, Pa O 2 , Pa CO2 and minute
ventilation) to CP [29] . In contrast, 1.5 mug/kg fentanyl was not superior to placebo when
comparing those above-mentioned vital signs.

We were unable to find a single study dealing with the influence of CP on the weaning process or
weaning strategy and the duration of mechanical ventilatory support. All available data are
confined to short-term effects such as hemodynamics, gas exchange, or lung function as acute
changes after CP. Therefore, we do not know from clinical data whether CP really contributes to
shortening the duration of mechanical ventilatory support.

Study with Research Animals.

Zidulka et al. [30] demonstrated in mechanically ventilated dogs after muscle relaxation that
percussion of the lungs with cupped hands (comparable with the clinical setting) leads to
esophageal pressure swings of 10-17 cm H2 O (1.0-1.7 kPa). Consecutive histopathologic
examinations of the lungs immediately after termination of the treatment and killing of the animals
showed large atelectatic areas adjacent to the areas of the chest wall where CP was administered.
Moreover, there were atelectatic areas on the surface of the opposite lung to be seen that could be
best described as a contra-coup effect. Zidulka et al. speculated that worsening of oxygenation,
demonstrated in numerous clinical studies during and immediately after CP, happens because of the
induction of atelectasis in treated lungs. They recommended giving patients a couple of deep
breaths after termination of CP to overcome ventilation-perfusion mismatch caused by the
formation of atelectases.

Special Properties of Lungs in Pediatric Patients.

The special properties of the lungs in pediatric patients must be considered when applying therapy
for respiratory failure found to be beneficial in adult patients to children. First, closing capacity
refers to the volume of gas present in the lung at the point that small conducting airways begin to
collapse. In children aged >6 yrs, functional residual capacity exceeds closing capacity. In infants
and in children <6 yrs of age, however, the closing capacity exceeds functional residual capacity
[31]
.

Second, the increasing chest wall stiffness with age leads to a diminishing ratio of compliance of
the chest wall:lung compliance (Cw /C l ) with age. A value of ~3-6 occurs in infancy, 2 in early
childhood, unity in young adulthood, and 0.5 in the elderly [32] . For these two reasons, infants and
children are more prone to developing atelectasis associated with a variety of pulmonary diseases
and CP may lead to atelectasis in young patients after the administration of vibration and percussion
to the chest wall.

Burden of Chest Physiotherapy in Children.

Vandenplas et al. [33] demonstrated a significant increase in gastroesophageal reflux prevalence
while administering CP in 63 spontaneously breathing infants, ranging in age from 1 to 4 months.
Gastroesophageal reflux was defined as an episode of esophageal pH <4 as measured by an
esophageal pH probe. All three components of CP (percussion, vibration, and postural drainage)
were linked with an increased prevalence of gastroesophageal reflux when compared with a control
group. Vandenplas et al. observed that there was no timely relation of gastroesophageal reflux
episodes and coughing during treatment. They concluded that CP should only be administered in
the fasting infant because of the potential risk of developing aspiration pneumonia.

Dependent head position as part of the postural drainage regimen in CP leads to an increase in
intracranial pressure (ICP) as demonstrated by Emery and Peabody [34] in 14 mechanically ventilated
newborns with or without perinatal asphyxia. Use of a 30° head-down position resulted in an
increase in ICP from 13.8 ± 2.5 cm H2 O (1.38 ± 0.25 kPa) to 17.2 ± 2.0 cm H2 O (1.72 ± 0.2 kPa)
and was even augmented by a concomitant lateral position to a maximum of 25.0 ± 2.6 cm H2 O
(2.5 ± 0.26 kPa).
Perlman and Volpe [35] described a significant increase in cerebral blood flow velocity and ICP after
routine suctioning of 34 preterm infants. These increases subsided only gradually after cessation of
the procedure. Similar results were obtained in 12 comatose adult patients after suctioning or
hyperinflation of the lungs as part of a regimen to clear lung secretions. ICP rose by 34% and 31%,
respectively, in comparison with baseline values before commencement of these procedures. In
some patients, ICP increased by as much as +70 mm Hg (+9.3 kPa) [36] .

Raval et al. [37] reported a frequency rate of grades III/IV intracranial hemorrhages in 5/10 preterm
infants treated with CP within the first 24 hrs of life vs. a frequency rate of 0/10 in an untreated
control group. McCulloch et al. [38] observed a marked increase in skin blood flow in mechanically
ventilated newborns who received CP.

In a recent study from Bloomfield et al. [39] , 220 newborn infants of all gestational ages with
respiratory failure received periextubation CP or no CP to assess its effect on the prevention of
postextubation atelectasis. Postextubation atelectasis occurred in 23% of the treated and in 15% of
the untreated infants (difference statistically not significant). Moreover, a larger subgroup of infants
receiving mechanical ventilatory support beyond 28 days of life could be identified in the group
treated with CP.

Summary of the Literature Reviewed.

CP in mechanically ventilated adult patients is associated with acute changes in the following
physiologic variables: a) gas exchange--decrease in Pa O2 , increase in Pa CO2 ; b) sympathetic
output--increases in heart rate and systemic blood pressure; c) lung physiology--increase in minute
ventilation, increase in maximum expiratory flow rate; d) increases in oxygen consumption and
oxygen extraction in the vascular periphery (the latter especially in patients who are unable to
increase their cardiac output); and e) increases in intracranial pressure.

CP administered in pediatric patients is associated with the following: a) a higher rate of atelectasis
and a longer hospital stay; b) a higher frequency rate of gastroesophageal reflux; and c) increases in
intracranial pressure, cerebral blood flow velocity and a higher rate of intracranial hemorrhage in
mechanically ventilated newborn infants.

Closing Remarks.
In mechanically ventilated children, CP cannot be regarded as a standard treatment modality. CP
must be considered as the most stimulating and disturbing intensive care procedure in mechanically
ventilated patients and

1651

should not be administered in children with low cardiopulmonary reserve attributable to an

increased oxygen consumption and increases in intracranial pressure. CP should only be
administered to critically ill patients whose respiratory gas exchange and hemodynamics are
monitored. CP should be evaluated in controlled studies that consider the length of mechanical
ventilatory support, influence on weaning the patient from the ventilatory support, occurrence of
atelectases, and length of hospital stay. Variations of response to CP in different age groups because
of changing physiologic properties such as closing capacity and chest wall compliance should also
be assessed.

REFERENCES

1. Lewis R: Chest physical therapy. In: Perinatal and Pediatric Respiratory Care. Barnhart SL, Czervinske MP (Eds).
Philadelphia, WB Saunders, 1995, pp 218-238

2. Mellins RB: Pulmonary physiotherapy in the pediatric age group. Am Rev Respir Dis 1974; 110:137-142 citation

3. Etches
PC, Scott B: Chest physiotherapy in the newborn: Effect on secretions removed. Pediatrics 1978; 62:713-
715 abstract

4. Asher MI, Douglas C, Airy M, et al: Effects of chest physical therapy on lung function in children recovering
from acute severe asthma. Pediatr Pulmonol 1990; 9:146-151 abstract

5. Quittell
LM, Wolfson MR, Schidlow DV: The effectiveness of chest physical therapy (CPT) in infants with
bronchiolitis. Am Rev Respir Dis 1988; 137:406A

6. Webb MSC, Martin JA, Cartlidge PHT, et al: Chest physiotherapy in acute bronchiolitis. Arch Dis Child 1985;
60:1078-1079 abstract

7. Graham WGB, Bradley DA: Efficacy of chest physiotherapy and intermittent positive-pressure breathing in the
resolution of pneumonia. N Engl J Med 1978; 299:624-627 abstract

8. Britton S, Bejstedt M, Vedin L: Chest physiotherapy in primary pneumonia. BMJ 1985; 290:1703-1704

9. Thomas J, Cook DJ, Brooks D: Chest physical therapy management of patients with cystic fibrosis. A meta-
analysis. Am J Respir Crit Care Med 1995; 151:846-850 abstract

10. Coates AL: Chest physiotherapy in cystic fibrosis: Spare the hand and spoil the cough? J Pediatr 1997; 131:506-
508 full text

11. Wollmer P, Ursing K, Midgren B, et al: Inefficiency of chest percussion in the physical therapy of chronic
bronchitis. Eur J Respir Dis 1985; 66:233-239 abstract

12. Kosloske AM, Ball WS, Butler C, et al: Drainage of pediatric lung abscess by cough, catheter, or complete
resection. J Pediatr Surg 1986; 21:596-600 abstract

13. Stiller K,
Geake T, Taylor J, et al: Acute lobar atelectasis. A comparison of two chest physiotherapy regimens.
Chest 1990; 98: 1336-1340 abstract

14. McMichan JC, Michel L, Westbrook PR: Pulmonary dysfunction following traumatic quadriplegia. Recognition,
prevention and treatment. JAMA 1980; 243:528-531 abstract
15. Kirilloff LH, Owens GR, Rogers RM, et al: Does chest physical therapy work? Chest 1985; 88:436-444
abstract

16. Reines HD, Sade RM, Bradford BF, et al: Chest physiotherapy fails to prevent postoperative atelectasis in
children after cardiac surgery. Ann Surg 1982; 195:451-455 abstract

17. ZachM, Oberwaldner B, Purrer B, et al: Thoraxphysiotherapeutische Behandlung bronchopulmonaler

Erkrankungen des Kindesalters. Monatsschr Kinderheilkd 1981; 129: 633-636 abstract

18. Laws AK, McIntyre RW: Chest physiotherapy: A physiological assessment during intermittent positive pressure
ventilation in respiratory failure. Can Anaesth Soc J 1969; 16: 487-493 citation

19. Gormezano, J, Branthwaite MA: Effects of physiotherapy during intermittent positive pressure ventilation.
Changes in arterial blood gas tensions. Anaesthesia 1972; 27: 258-264 citation

20. Connors AF Jr, Hammon WE, Martin RJ, et al: Chest physical therapy. The immediate effect on oxygenation in
acutely ill patients. Chest 1980; 78:559-564 abstract

21. Tyler ML, Hudson LD, Grose BL, et al: Prediction of oxygenation during chest physiotherapy in critically ill
patients. Am Rev Respir Dis 1980; 121:218A

22. Mackenzie CF,Shin B, McAslan TC: Chest physiotherapy: The effect on arterial oxygenation. Anesth Analg
1978; 57:28-30 abstract

23. Mackenzie CF, Shin B: Cardiorespiratory function before and after chest physiotherapy in mechanically
ventilated patients with post-traumatic respiratory failure. Crit Care Med 1985; 13:483-486 abstract

24. Weissman C, Kemper M: The oxygen uptake: Oxygen delivery relationship during ICU interventions. Chest
1991; 99:430-435 abstract

25. Weissman
C, Kemper M, Harding J: Response of critically ill patients to increased oxygen demand:
Hemodynamic subsets. Crit Care Med 1994; 22:1809-1816 abstract

26. Horiuchi
K, Jordan D, Cohen D, et al: Insights into the increased oxygen demand during chest physiotherapy.
Crit Care Med 1997; 25:1347-1351 full text

27. MacLean D, Drummond G, Macpherson C, et al: Maximum expiratory airflow during chest physiotherapy on
ventilated patients before and after the application of an abdominal binder. Intensive Care Med 1989; 15: 396-399
abstract

28. Weissman C, Kemper M, Damask MC: Effect of routine intensive care interactions on metabolic rate. Chest
1984; 86:815-818 abstract

29. Klein
P, Kemper M, Weissman C, et al: Attenuation of the hemodynamic responses to chest physical therapy.
Chest 1988; 93: 38-42 abstract

30. ZidulkaA, Chrome JF, Wight DW, et al: Clapping or percussion causes atelectasis in dogs and influences gas
exchange. J Appl Physiol 1989; 66:2833-2838 abstract
31. Venkataraman
ST, Orr RA: Mechanical ventilation and respiratory care. In: Pediatric Critical Care. Fuhrman
BP, Zimmerman JJ (Eds). St. Louis, CV Mosby, 1992, pp 519-543

32. Allen
JL, Sivan Y: Measurements of chest wall function. In: Infant Respiratory Function Testing. Stocks J, Sly
PD, Tepper RS, et al. (Eds). New York, John Wiley & Sons, 1996, pp 329-354

33. Vandenplas Y, Diericx A, Blecker U, et al: Esophageal pH monitoring data during chest physiotherapy. J Pediatr
Gastroenterol Nutr 1991; 13:23-26 abstract

34. EmeryJR, Peabody JL: Head position affects intracranial pressure in newborn infants. J Pediatr 1983; 103:950-
953 abstract

35. Perlman JM, Volpe JJ: Suctioning in the preterm infant: Effects on cerebral blood flow velocity, intracranial
pressure, and arterial blood pressure. Pediatrics 1983; 72:329-334 abstract

36. ErssonU, Carlson H, Mellstrom A, et al: Observations on intracranial dynamics during respiratory physiotherapy
in unconscious neurosurgical patients. Acta Anaesthesiol Scand 1990; 34:99-103 abstract

37. RavalD, Yeh TF, Mora A, et al: Chest physiotherapy in preterm infants with RDS in the first 24 hours of life. J
Perinatol 1987; 7:301-304 abstract

38. McCulloch KM, Ji SA, Raju TNK: Skin blood flow changes during routine nursery procedures. Early Hum Dev
1995; 41:147-156 abstract

39. Bloomfield FH, Teele RL, Voss M, et al: The role of neonatal chest physiotherapy in preventing postextubation
atelectasis. J Pediatr 1998; 133:269-271 full text

MD Consult L.L.C. http://www.mdconsult.com

Bookmark URL: /das/journal/view/N/11348447?ja=179186&PAGE=1.html&ANCHOR=top&source=HS,MI