In this thesis, the effects of physiotherapy techniques of percussion/vibration and manual hyperinflation, and the respiratory care technique of endotracheal airway suctioning were investigated in an endotoxaemia-induced acute lung injury (ALI) model.
Studies concerning the use of physiotherapy in ALI and its more severe form, acute respiratory distress syndrome (ARDS), are lacking. Recent findings indicate that alveolar collapse could develop from low-tidal-volume ventilation and ARDS of extrapulmonary causes. Retained secretions, whilst not a feature of ARDS, could occur in patients ventilated by low tidal volume. Physiotherapy with the aim to re-expand collapsed alveoli or remove excess secretions could be helpful. Study 1, a questionnaire survey seeking physiotherapists' opinions on the role of physiotherapy in ARDS showed that only 59% of the respondents would treat ARDS with the aim to re-expand collapsed alveoli, and 70% to remove excess secretions. Half of the questionnaire respondents were familiar with the current research work on ARDS. Over 80% of the physiotherapists in this survey identified manual hyperinflation as contraindicated in the presence of ARDS. Another about 20% of the sample considered percussion or vibration similarly contraindicated. From the findings of this survey, it became apparent that there was a need to develop evidence for or against the effects of physiotherapy in ALI or ARDS. The main aim of this thesis was to establish whether physiotherapy techniques might have adverse effects in a condition already characterized by hyperdynamic and hypermetabolic states, with increased cytokine production. Haemodynamic effects of some of these techniques in healthy ventilated subjects were investigated in Study 2 to compare with those effects in subjects with ALI.
Study 2 reports on the haemodynamic effects of percussion and vibration in healthy ventilated sheep, as these techniques were identified as poorly investigated. Haemodynamic variables were measured during the application of these techniques to verify claims of adverse events. Physiotherapists in this study consistently applied percussion at a rate of 6.2 ± 0.9 Hz, vibration at 10.5 ± 2.3 Hz and shaking at 6.2 ± 2.3 Hz, and caused oesophageal pressure swings of 8.8 ± 5.0 mm Hg, 0.7 ± 0.3 mm Hg and 1.4 ± 0.7 mm Hg respectively. These techniques were found to produce no significant effects on haemodynamic and most ventilatory variables in these healthy anaesthetized animals. Data from this study involving healthy subjects form the basis for further studies involving subjects with ALI.
Subjects with ALI were healthy ventilated anaesthetized sheep given bacterial endotoxin. Study 3 reports the haemodynamic and respiratory changes following endotoxin infusion. The changes associated with endotoxaemia-induced ALI, such as four-fold increase in pulmonary artery pressure but decrease in cardiac output, systemic vascular resistance and mean arterial pressure, were similar to those reported in previous studies and mirrored those observed in early extrapulmonary ALI or ARDS in human, making this experimental model suitable for the current study aims. Animal subjects were used to overcome any confounding variables that might arise from heterogeneous aetiological factors precipitating ARDS, treatment preferences, pharmacologic interventions and inherent patient characteristics (for example, age, lung compliance, co-morbidities, etc).
In Study 4, effects on haemodynamic and oxygenation variables were investigated. Ten anaesthetized sheep with endotoxaemia-induced ALI were randomly allocated into treatment (n = 6) or control group (n = 4). Treatment animals received percussion/vibration, manual hyperinflation and endotracheal suctioning in random order. No significant differences in haemodynamics and oxygenation were found between the groups. However, oxygenation indices were significantly altered by individual techniques: arterial oxygen partial pressure to fraction of inspired oxygen (PaO2/FiO2) ratio was increased by manual hyperinflation by 118%, reduced by percussion/vibration by 6%, and reduced by endotracheal suctioning by 24% (p < 0.05). No major adverse haemodynamic complication associated with individual technique was identified.
Study 5 reports the effects on serum tumour necrosis factor (TNF)-α and lactate. Treatment and control groups were significantly different in TNFα and lactate during the most acute inflammatory phase of the lung injury (p < 0.05) but not at the end of the study (p = 0.085). Open endotracheal suctioning, but not manual hyperinflation, was associated with greater release of TNFα and lacates (p = 0.016).
These results provided experimental evidence that physiotherapy techniques of percussion/vibration and manual hyperinflation would not cause significant changes to the haemodynamics nor increased cytokine production in an anaesthetized animal model of extrapulmonary ALI. Open endotracheal suctioning, however, was associated with greater short-term cytokine release, although no significant haemodynamic alterations were demonstrated. The combined results of these studies indicate that physiotherapy techniques of percussion/vibration and manual hyperinflation would unlikely be associated with adverse effects on the haemodynamics, metabolic demand and lung injury in an animal model of extrapulmonary ALI/ARDS.