Physiological responses to artificial ventilation

A review of the literature revealed that there was no comprehensive history of artificial ventilation and the historical section of the thesis has been expanded to fill this gap. The history of artificial ventilation has been traced from myth to reality and from hesitant attempts to confident application. In the last two decades intermittent positive pressure ventilation has become increasingly commonly used in the management of patients with respiratory failure. There is a large number of mechanical ventilators available for this purpose and in many instances the manufacturers claim superiority for their machines because of the inspiratory flow waveform and inspiratory-expiratory time ratio (I:E ratio). There is, however, very little scientific evidence available on which to base such assertions and the present study is an attempt to assess the physiological implications of changing only the inspiratory flow waveform or I:E ratio in artificial ventilation. A variable waveform ventilator was built so that selected flow waveforms could be administered with independent control of the duration of inspiration and expiration. The ventilator consists of a variable electronic signal generator which acts through an attenuator upon an electropneumatic converter which linearly converts the electrical signal to a pressure signal of up to 150 p.s.i. This pressure is then transformed to a flow by passing the gas through a choke, and the size of the choke may be changed by substitution. The resulting flow is linearly related to the original electrical signal so that faithful reproduction of the signal is achieved. The patchboard of the signal generator allows a large number of waveforms to be studied but in this study only sine, tophat, ramp and reversed ramp waveforms were used. This ventilator includes a new mechanical approach to the production of variable flow waveforms and allows greater flexibility than any previously designed. Before embarking on the physiological study, an analogue study of the lung/thorax system was performed using electrical, mathematical and computer models. The analogue study predicted some interesting phenomena. One of these predictions was that the chest wall resistance was of importance in the response of the alveolar pressure. This resistance causes an instantaneous pressure step to occur in alveolar pressure, and also in oesophageal pressure, when an instantaneous step change is made in either the pressure or flow being applied to the patient in artificial ventilation. This observation was confirmed experimentally in humans and dogs. The analogue study also predicted that sine, triangular or reversed ramp waveforms with an inspiratory time of 1 to 1.5 seconds in a 4 second cycle would be the least harmful in the general application of artificial ventilation. This conclusion was, however, based on compromise and may be in error when any particular physiological variable is being considered. The analogue study also showed that some previous studies of lung analogues in artificial ventilation were in error. Physiological studies were performed on mongrel dogs and measurements were made of cardiac output by two methods, blood pressures, airway and oesophageal pressures, blood gas partial pressures, gas flow and tidal volume, end-tidal carbon dioxide concentration, electrocardiograph, haemoglobin, barometric pressure and temperatures. Certain physiological indices of cardio-respiratory efficiency were then computed and analysed. The experimental procedure was arranged in such a manner that factorial design statistical analysis could be applied to the data, and this analysis tested for synergism between flow waveform and duration of inspiration. The results of investigations of the physiological effects of changing the I:E ratio, within a four second respiratory cycle, were largely in agreement with previous studies. Inspiratory times of 2.2 to 0.5 seconds were investigated and the physiological dead space to tidal volume ratio (VD/VT ratio) increased, total dynamic compliance decreased, and mean airway and oesophageal pressure fell with reduction of the duration of inspiration. No change was found in cardiac output, pulmonary venous admixture, dynamic chest wall compliance, or alveolar to arterial oxygen partial pressure gradient (A - a PO2 gradient). There were suggestions that arterial partial pressure of oxygen (Pa,O2) fell, arterial partial pressure of carbon dioxide (Pa,CO2) rose, and the arterial to alveolar carbon dioxide partial pressure gradient (a - A PCO2) rose with reduction of the duration of inspiration. It might be expected that an increase in inspiratory time with increased mean airway and oesophageal pressures would reduce cardiac output. In the event, no doubt due to compensatory mechanisms in the healthy dog, this did not occur, but might be expected to occur in patients with cardiovascular disorders. For this reason, the shortest inspiratory time which did not seriously affect other variables, 1 to 1.5 seconds in a 4 second cycle, was considered to be the "best". Studies on flow waveforms indicated that for most of the physiological variables investigated (VD/VT ratio, total dynamic compliance, Pa,CO2, alveolar partial pressure of carbon dioxide (PA,CO2) , a - A PCO2, Pa,O2, A - a PO2 gradient and Q′s/Q′) the reversed ramp flow waveform was the least disruptive and ramp flow the most disruptive of the four waveforms examined. This did not apply, however, to studies on cardiac output or mean airway and oesophageal pressures where the effects were reversed, ramp flow being the least harmful. The results of the studies of the effects of flow waveforms are in disagreement with previously reported investigations. Synergism between the effects of flow waveform and I:E ratio was found when VD/VT ratio, Pa,CO2, PA,CO2, a - A PCO2 gradient, mean airway pressure and mean oesophageal pressure were studied. Previous reports have not considered this possibility. The controversy concerning the characteristics of the ideal ventilator is thus only partially resolved. Analogue studies suggested that sine, triangular or reversed ramp flow waveforms, at an inspiratory time of 1 to 1.5 seconds in a 4 second cycle, would be satisfactory if the effects of all factors were considered. In the physiological studies, the reversed ramp flow waveform proved least disruptive in most circumstances, but this flow waveform had the most disruptive effects on mean airway and oesophageal pressures and cardiac output. It may therefore be necessary to employ different waveforms in different cardiopulmonary disorders, and observations in such disorders should be made.