Ventilators used in anaesthesia were primarily designed to replace the minute ventilation of a patient with healthy lungs who is paralyzed or has a depressed respiratory drive as a result of anaesthetic agents and opiates.¹ In contrast intensive-care patients are normally encouraged to breathe alongside the mechanical ventilatory support which is gradually reduced during a slow weaning phase. Additionally at the end of anaesthesia, extubation is more precipitous with the rapid recovery of consciousness and return of spontaneous breathing. ICU ventilators, therefore, have to cope with the complexity of both machine- and patient-initiated respiration.

The pulmonary mechanics in terms of airways resistance and total lung compliance in ICU patients is rarely normal. ICU ventilators have, therefore, evolved to support respiration using such techniques as inverse ratio ventilation and high levels of PEEP, which are rarely required during routine anaesthesia. Ventilation for prolonged periods of time may require humidification techniques other than the simple heat and moisture exchanger commonly found in the operative setting. These alternative techniques such as hot water humidifiers are associated with increased accumulation of fluid in the respiratory circuits, which may compromise ventilator function. Large air leaks from the lungs or around airway tubing are more common in the ICU setting, and monitoring facilities on ICU ventilators have to be adapted to function in these unusual circumstances to prevent an excess of false alarms. Nonetheless, the flexibility of modern ICU ventilators is utilizable and sometimes indispensible in the operating theatre and modern anaesthesia ventilators are increasingly equipped with features of an ICU type ventilator.

Driving mechanisms

All ventilators require a driving force to deliver a gas flow into the patient. This force may either deliver the inspiratory gas directly to the patient or indirectly by compressing a bag or bellows containing the inspired gas mixture which in turn delivers the gasses to the patient.² Driving mechanisms which have been used in ventilators include the following:

1. Rotating electric motors. These may be used to:

a. drive a crank connected to a piston which in turn delivers the inspiratory gas. The disadvantage of these systems is that they have difficulty in altering inspiratory flow characteristics as the wheel and interlinked shaft drive will produce a fixed sinusoidal movement of the piston

b. power low pressure blowers. The advantage of this type of driving mechanism is that they can produce very high volumes of gas flow to enable the ability to cope with leaks from the patient circuit, in non-invasive ventilatory systems

c. drive gas compressors producing high pressure gas which may then be manipulated by pneumatic controls.

d. power a piston using a worm drive, this system is not seen in ITU ventilators (see Chapter 9, Dräger E anaesthetic ventilators).

2. Linear electric motors. These can be used to drive either pistons or diaphragms which are used in high frequency oscillators.

3. Tension springs which compress the gas in a storage bellows prior to being delivered to the patient. The disadvantage of this mechanism is that the pressure in the bellows is not constant but varies with the tension in the springs.

4. Weighted bellows. The gravitational force on a mass (the weight) will produce a constant driving pressure within the bellows; however, this mechanism is very susceptible to movement.

5. Pneumatic. These may be of two types:

a. Low-pressure systems using gas supplied from electric blowers or venturi devices.

b. High pressure with the gas supplied from a compressor via a high-pressure gas pipeline or cylinder which is connected to the ventilator.

Microprocessor electronic control

The use of microprocessor control has become virtually ubiquitous in all modern ICU ventilators. At the heart of these is the control microprocessor whose function is to receive data from the analogue to digital converters that measure pressure and flow within the ventilator system. With this information and under software control, the microprocessor provides the commands for the circuit board to precisely control the inspiratory and expiratory flow valves and secondary functions (e.g. nebulizer control or to flush pressure measurement lines). Data are also received from the oxygen and carbon dioxide sensors, if fitted. Information may also be received from external sources and interface boards that allow ventilators to talk to each other for synchrony in ventilator modes such as for independent lung ventilation. At least one current device can use electromyographic input to aid in ‘neural’ control of the ventilator synchronisation, see below NAVA.

Some ICU ventilators use dual microprocessors to control the ventilator function. The advantage of using two processors in parallel is that not only can the function of information display be separated from the control of the valves, but each microprocessor can check the output of the other against its own computations to ensure maximum patient safety and ventilator reliability (Fig. 10.1).

Figure 10.1 Microprocessor controls inside ventilator.

Information display

In the oldest ICU ventilators, the only information available to the operator was the analogue measurements of pressure or volumes (Fig. 10.2). With the same starting point of pressure measurement, microprocessor integration of the data now allows any number of derived variables to be monitored and displayed to inform the clinician of machine function and pulmonary mechanics. Alarm parameters can then be set for many of these values.

Figure 10.2 Analogue pressure display.

Digital information such as pressure values (peak, plateau, PEEP, etc.), calculations of tidal volume, machine or patient triggered breaths, inspired/expired volume differences and the quantity of gas leaking from the patient circuit were available to the user originally only in single-line numeric format (Fig. 10.3).³ With the availability of liquid crystal display (LCD) screens, more of this information, instead of being provided in single numeric displays, is available on a matrix screen, allowing the user not only to see the static numerical data, but also graphical information, such as flow, pressure and volume variations. These may be plotted in parallel against time or as loops against one another. The increasing use of graphical displays allows the user to more easily understand the effect that a change in ventilator controls has on the delivery of gas to the patient. Although LCD screens have the advantage of low-power consumption, they are only readable over a narrow viewing angle and are in turn now being replaced by TFT LCD (thin film transistor LCD) screens that can provide similar information in colour. These are readable over a larger viewing angle and in lower lighting conditions (Fig. 10.4).

Figure 10.3 LED digital display.

Figure 10.4 TFT panel display.

Inspiratory flow valve

Gas flow within ventilators may by controlled by one of two types of solenoid valves. In one type the valve has only two states: either being on or off; in the other type (proportional valves), the amount of opening when activated is directly related to the voltage applied to the solenoid.

High-speed proportional servo controlled (see below) flow valves are used in several manufacturers’ ventilators. They are capable of delivering flows from 20 to 3000 ml s⁻¹ and are adjusted by the ventilator’s microprocessor control (servo control) using an electrodynamic motor similar to that of a loudspeaker. A current flowing in the field coils of the solenoid generates the force to move the piston up and down; connected to the piston is the valve orifice which opens to allow gas to flow. This type of valve has a quick response time of typically 5 ms and with the small internal dead space of the ventilator, the flow rates change almost immediately in the patient circuit allowing precise control of the desired flow pattern and tidal volume. The microprocessor monitors the position of the valve and the pressure drop across it to enable it to continually adjust the flow to the required setting so that its performance characteristics are not affected by back pressure in the circuit (Fig. 10.5).

Figure 10.5 Inspiratory servo valve.

Some manufacturers use high-pressure, high-speed on-off gas solenoids, which control the flow of both oxygen and air. Under the microprocessor’s control these can be rapidly pulsed on and off to create the desired inspiratory flow pattern. Ventilators using these types of inspiratory solenoid no longer require a separate gas blender to create the oxygen air mixture as this is produced directly by the solenoids from the high-pressure gas pipeline. Not having a separate gas blender and mixing chamber allows the ventilator to rapidly change oxygen concentration within the patient circuit in response to altered settings by the operator, facilitating a swift 100% oxygen setting for use prior to tracheobronchial suctioning.

Flow sensors

Flows sensors (see Chapter 2, Measurement of pressure and gas flow) are often placed in both the inspiratory and expiratory limbs of ventilators. These not only allow the ventilator microprocessor control to sense and adjust gas flow, but also provide the user with measurements of inspiratory and expiratory tidal volumes, whether ventilator delivered or patient initiated.

Manufacturers have used a variety of flow sensors. These can be either a wire mesh pneumotachograph as incorporated into ventilators such as the Siemens 900C (Fig. 10.6) or a hot wire anemometer, as used in the expiratory limb of the Dräger Evita ventilators (Fig. 10.7) or the bidirectional variable orifice device that is incorporated into the patient end of the ventilator circuit in the Hamilton G5 ventilators (Fig. 10.8). The Maquet Servo-i uses an ultrasonic flow transducer (see Chapter 2). Flow sensors placed in the expiratory limb are susceptible to condensation hence some manufacturers incorporate a heating arrangement to ensure that these devices are kept free of water droplets.

Figure 10.6 Wire mesh pneumotachograph.

Figure 10.7 Hot wire anemometer. (Note that the very thin filaments of wire are not quite visible on their supports in this sensor photograph)

Figure 10.8 Bidirectional pneumotachograph.

Patient triggering

To facilitate the patient’s spontaneous ventilation alongside mechanically driven breaths in such modes as intermittent mandatory ventilation and pressure support (see below), ICU ventilators have to be able to sense the commencement of the patient’s own respiratory efforts and then provide a gas source from which they can breathe. This can be achieved by sensing a change in either flow or pressure within the circuit, triggering the ventilator to open the inspiratory valve. If the patient has to generate large changes of pressure within the circuit before obtaining any inspiratory gas supply this will add significantly to their work of breathing. Together with the increased pressure changes, a time delay from the commencement of the patient’s inspiratory effort to the start of gas flow will multiply the effect of any pressure drop; further increasing in the patient’s work of breathing.⁴

Pressure triggering

Early ventilators such as the Siemens Servo 900C provided only pressure triggering; the pressure-sensing device being placed upstream within the inspiratory limb inside the ventilator. The pressure drop resulting from the patient’s inspiratory effort activates the trigger whose sensitivity is usually set to −1–2 cm H₂O. When the threshold is reached, the electronic control within the ventilator opens the inspiratory valve allowing sufficient gas to reach the patient.⁵ The disadvantage of this method of sensing is that due to the remote placement of the sensor the patient may have to generate a considerably greater effort at the patient end of the circuit to activate the trigger threshold within the ventilator. This was particularly apparent when thin-walled, highly compliant, disposable plastic tubing was connected to the ventilator. In addition if a hot water humidifier is also connected in the inspiratory limb, the added compliance and resistance of this further increases both the pressure gradient and the patient’s work of breathing before obtaining any inspiratory gas supply.⁶

To overcome this problem, pressure sensing of the patient’s inspiratory effort can be achieved more reliably in the expiratory side of the ventilator, although this is still subject to the effects of condensation accumulating within the expiratory limb of the patient circuit. Other manufacturers, such as Hamilton in their series of ventilators, sense pressure directly at the patient end of the circuit to avoid these problems.

Flow triggering

To overcome the disadvantages of pressure triggering most ICU ventilators now provide a form of flow triggering.⁷ A flow trigger does not directly require change in pressure within the inspiratory circuit to enable a patient-initiated breath. The ventilators provide a continuous bias flow, frequently 10 l min⁻¹, which is introduced into the inspiratory circuit throughout all phases of the machine’s respiratory cycle. When the patient starts to initiate a breath from the circuit, the fall in bias flow in the expiratory limb of the ventilator is sensed and the ventilator opens the inspiratory valve to allow a patient-initiated breath. The change in flow required to trigger a patient breath can be adjusted by the operator and is often set between 1 l min⁻¹ and half the bias flow.⁸ The advantage of this mode is that it is not as affected by humidifiers in the inspiratory limb or condensation in the tubing so patient work of breathing is minimized.^6,^9,¹⁰ Most ventilators allow the user to adjust the sensitivity of the flow triggering to achieve the minimum work of breathing without the occurrence of falsely triggered breaths.¹¹