The operator interaction with the ventilator mainly happens through the ventilator display. The display or interface has evolved in parallel with the ventilators. The key to this evolution are the technological advances in the last three decades.2 The microprocessors, the digital displays, and the interactive screens have all permeated from other technological advances into the ventilator world. There are still remnants of the evolutionary process. In their initial ventilator generations, the interface had no or minimal manifestation of the interaction with the patient. The operator would enter the ventilator settings by using knobs or buttons that regulated simple functions (pressure, flow, or time). The results of these changes were evaluated in the patient clinical response, and occasionally through simple pressure analog displays. Some ventilators still use these type of displays (e.g., CareFusion 3100A high-frequency oscillator and Puritan Bennett LP-10, Fig. 3-8).
Most of the ventilators produced in the last decade have advanced displays, including liquid crystal displays and color touch screens with one or more multipurpose knobs or buttons. This allows the user to scroll through different menus and to select and activate the selections (e.g., Hamilton G5 ventilator, Fig. 3-9). The operator can customize the screen to the operator’s needs. Current ventilators allow graphical displays of alarms, settings, respiratory system calculations, and measurements. The ventilator display evolution has not necessarily resulted in easier management of the ventilator. These advances brought issues with the amount of information displayed, the actions taken with that information, and the ease of use of certain interfaces.6 As the level of sophistication has increased, we have been able to increase the number of ventilation parameters monitored. This requires a new level of training and understanding of human behavior. For example, a mode of ventilation may be preferentially chosen based on the amount of alarms it triggers,7 or its ease of use.6,8
The operator input refers to parameters or settings entered by the operator of the ventilator. Each mode of ventilation has particular features, some of which can be adjusted by the operator. We describe here the most common adjustable parameters. The effect of each parameter on the lung is better understood under the light of the equation of motion (see Chapter 2).9,10 A change of one parameter will lead to changes in others (i.e., in volume control, for the same respiratory characteristics changing the tidal volume will cause a change in peak airway pressure). Furthermore, knowing the basic construction and characteristics of a mode of ventilation (volume vs. pressure control breaths) or the breath sequence (mandatory vs. spontaneous) will help understand how the setting will affect the ventilator output (see Chapter 2).
The operator input is presented below in the order that follows the progression of a breath; starting with the gas inhaled, to triggering, targeting, cycling, and baseline variables.
Inspired Gas Concentration
A mechanical ventilator has the capacity of delivering different mixtures of gas. Most ventilators allow the administration of specific concentrations of oxygen. A few allow the administration of helium, nitric oxide, or anesthesia gases.
Oxygen is the most common gas administered to patients undergoing mechanical ventilation. The oxygen percentage in the inspired gas (FIO2) can be regulated in most ventilators by means of a direct adjustment of a specific control (21% to 100%). However, this is not true for all ventilators. For example, some home ventilators (e.g., LP-10 or the LTV 1150, Pulmonetic, CareFusion) use a connection to a low-pressure oxygen source to the ventilator or the patient circuit. The following formula can calculate the flow of oxygen to achieve a desired oxygen concentration:
where O2 required is 100% oxygen flow in L/min, f is the breathing frequency in breaths/min, VT is the tidal volume in liters and the FIO2 is the patient O2 concentration desired in decimal format (i.e., 30% = 0.3). An oxygen analyzer should be used to confirm the measurements. It must be recognized that changes in oxygen flow, breathing rate, or tidal volume will change the FIO2.
When transporting the critically ill patient, availability of oxygen supplies for the mechanically ventilated patient is crucial. Size and weight of cylinders makes transport difficult and presents an increased risk of fire. Branson et al. have described a solution using a portable oxygen concentrator (SeQual Eclipse II) paired with the Impact 754 and Pulmonetics LTV-1200 ventilators.11
For the rest of the current mechanical ventilators, the ventilator adjusts the mixture of air and oxygen to achieve the desired FIO2. The mixing of air is achieved by an internal or external blender. A blender may use proportioning valves that regulate the flow of air and oxygen to a mixing changer (Fig. 3-10). It is similar to the mechanism used to mix hot and cold water in a shower—the more oxygen needed, the larger the opening for oxygen and the smaller it is for air. To work properly, the blender requires a constant pressure within the working ranges of the device.
Schematic of a ventilator air–oxygen blending system using proportional valves.
Most current ventilators have oxygen sensors to monitor the FIO2. The oxygen sensor gives feedback to the operator to adjust the mixture, or alarms if there is a discrepancy between the set and delivered FIO2. The oxygen sensors detect changes in electrical current, which is proportional to the oxygen concentration. The most common techniques are: (a) paramagnetic, (b) polarographic, and (c) galvanic.12
Mixtures of helium and oxygen (heliox, HeO2) instead of air and oxygen are occasionally used to help patients on mechanical ventilation with obstructive airway diseases. Helium is less dense than air (Table 3-1).13 The decrease in density interferes with flow measurements, inspiratory and expiratory valve accuracy, and gas mixing.14 Several studies have evaluated the performance of mechanical ventilators delivering heliox14–16 and have shown that heliox does affect the performance of the ventilator. The interference of heliox is more evident in volume-control modes than in pressure-control modes.14,17 In pressure-control mode, the ventilator targets a set inspiratory pressure and the delivered tidal volume is dependent only on the mechanical properties of the respiratory system. The time constant may decrease but the delivered volume should be the same as for nonheliox gas delivery. In volume-control mode, delivered volume may be larger than, smaller than, or the same as expected depending on the design of the ventilator.14 Only a few ventilators (Maquet Servo i with heliox option, Hamilton G5 with heliox option, and the Viasys Avea with comprehensive model) are designed and calibrated for heliox delivery. Otherwise, the operator needs to be aware of the specific ventilator performance and correction formulas and factors14 such that potentially hazardous conditions do not develop.
Table 3-1: Properties of Pure Gases and Air |Favorite Table|Download (.pdf)
Table 3-1: Properties of Pure Gases and Air
|Gas||Thermal Conductivity (κ)(μcal · cm · s · °k)||viscosity (η) (Micropoises)||Density (ρ) (g/L)|
Inhaled nitric oxide (NO) is used as selective pulmonary vasodilator for patients with pulmonary hypertension, life-threatening hypoxia, or right-heart failure. Different devices to deliver NO have been described in the literature. Most of them were custom made and required the use of mixing chambers, stand-alone NO/nitric dioxide monitors, and manual titration of the gas flow. The large amount of custom-made devices led to inconsistent administration of NO.18 In 1998, the American Society for Testing Materials (ASTM) committee on anesthetic and respiratory equipment developed a standard to provide a minimum degree of safety of the devices used to deliver NO. The recommendation was to use a NO administration apparatus, and a NO/nitrogen dioxide analyzer. The Food and Drug Administration (FDA) enforces this recommendation, and so far, only one device is approved in the United States. The INOvent (Ikaria Inc, Clinton, NJ) delivery system uses a closed-loop scheme to measure and deliver NO in proportion to the inspiratory flow from the ventilator. NO is injected in the proximal limb of the inspiratory circuit, and measured close to the connection between the patient circuit and the endotracheal tube. Two portable systems are available—INO Max DS (Ikaria) and AeroNOx (PulmoNOx, Alberta, CA). As these devices are not universally available, the following formula19 can be used to calculate the NO flow rate required to achieve a desired concentration of NO when injected in the inspiratory limb at a constant gas flow,
where QNO is the flow rate of nitric oxide in L/min, CNOset is the desired NO concentration in parts per million (ppm), CNOcyl is the NO concentration in the cylinder in ppm (usually 800 ppm) and the QV is the ventilator gas flow.
The formula is accurate for constant flow systems. This presents a major problem when used with intermittent breaths (as most modes of ventilation) the patient will receive variable amounts of NO (a “bolus” with each mechanical breath).20 Furthermore, whenever the ventilator settings or the patient breathing pattern changes, the NO delivery will change. Finally, the use of NO will alter the gas delivery of the ventilator. For example, the INOvent system will add gas to and extract gas from the delivered breath. At 80 ppm it adds 10% more gas, although it also withdraws 230 mL/min through the gas-sampling port. Thus, the oxygen delivered will decrease, and the tidal volume may increase. The changes seem to be small (unless you see it in pediatric proportions), but it may affect the ventilator’s performance. Furthermore, as a flow of gas is introduced, the flow-triggering performance may be affected.
A ventilator-assisted breath can be started (triggered) by the machine or the patient. A machine-triggered breath is defined by the start of the inspiratory phase independent of any signal from the patient. The operator typically sets a breath frequency for machine-triggered breaths. A patient-triggered breath is one for which inspiration is started solely by a signal from the patient. The key operator set variable for patient triggering is sensitivity, or the magnitude of the patient signal required to initiate inspiratory flow. The patient signal can be obtained from measuring the airway pressure, flow, volume, electromyogram (EMG),21 abdominal motion (Graseby capsule22), thoracic impedance,23 or any other measurable signal of respiratory activity.24 Most intensive care ventilators measure pressure and flow (volume is integrated from flow) at the circuit. There are only a few ventilators that use other sources of signaling, diaphragmatic EMG (Servo i NAVA), thoracic impedance (Sechrist SAVI), and abdominal motion (Infant Star STAR SYNC, which is no longer commercially available).24,25
Ventilator triggering characteristics can be evaluated using different metrics.23,26–28 The most sophisticated device for evaluating ventilator performance is the ASL lung simulator (IngMar Medical Ltd., Pittsburgh, PA). This device can simulate both passive lung mechanics (e.g., resistance and compliance) as well as patient inspiratory and expiratory effort. It can display and record pressure, volume, and flow signals, and calculate a wide variety of performance metrics. Figure 3-11 shows an example of these waveforms with specific reference points for calculating performance metrics (from operator’s manual for software version 3.2). Using these reference points we can define the following key trigger metrics: Pmin (maximum pressure drop relative to PEEP during the trigger phase), pressure-time product (∫ Paw-PEEP dt from start of effort to return of airway pressure [Paw] to PEEP), patient trigger work (∫ Paw-PEEP dv from start of effort to return of Paw, to PEEP), and time to trigger (period from the start of effort to the return of Paw to PEEP).
Reference points on pressure, volume, and flow waveforms recorded by the ASL 5000 (IngMar Medical Ltd, Pittsburgh, PA). A. Start of inspiratory effort, B. beginning of inhalation as determined by the “breath start volume threshold,” C. lowest pressure during the trigger phase, Pmin, D. return of airway pressure to baseline during the trigger phase, E. end of inspiratory time, i.e., negative-going zero flow crossing, F. beginning of exhalation as determined by the “expiratory start volume threshold,” and G. end of expiratory time, i.e., positive-going zero flow crossing. (Reproduced, with permission, from Ingmar Medical. ASL 5000 v3.2 Operator’s Manual. Pittsburgh, PA: Author.)
Time is measured by the internal ventilator processor. The next breath is time triggered (in the absence of a patient trigger event) when the expiratory time has reached the threshold to maintain a set respiratory rate (e.g., if the set rate is 10 breaths per minute and the inspiratory time is set at 1 second, then the expiratory time is 5 seconds). Some modes allow the user to set the inspiratory and expiratory time [e.g., airway pressure release ventilation (APRV) and biphasic], thus fixing the inspiratory-to-expiratory timing (I:E) ratio and respiratory rate. In an effort to improve patient–ventilator interactions, the ventilator may synchronize the mandatory breath with the patient’s triggering signal if it falls within a threshold. The classic example is synchronized intermittent mandatory ventilation (SIMV). More recently APRV, as programmed in the Evita XL, delivers a machine breath if the patient trigger signal falls within 25% of the triggering time.29 Time triggering is also found as a safety mechanism. The operator or manufacturer enters a time after which the apnea alarm will trigger the delivery of a preset breath after a preset time is reached.
The patient inspiratory effort causes a drop in pressure in the airway and the circuit. Inspiration starts when pressure falls below the preset “sensitivity” threshold. The site of measurement will have an impact on the performance of the device. Pressure signals travel at the speed of sound, approximately 1 ft/ms.30 The farther the sensor is from the signal source, the longer the potential time delay. The closest measurements can be done in the trachea. Tracheal pressure measurements reflect actual airway pressure as the endotracheal tube resistance is bypassed. When used for ventilator triggering, tracheal pressure sensing results in decreased work of breathing.31–33 However, tracheal pressure measurements are not routinely done and require special equipment (endotracheal tube with monitoring port) and no current ventilator uses it to routinely trigger the ventilator.
The other sites of pressure measurement are the patient circuit Y or at the inspiratory or expiratory ports, each with its advantages and disadvantages (Table 3-2). Trigger performance will also be affected by the presence of humidifiers, filters, water condensation, patient circuit and exhalation valves. These will most often dampen, or rarely amplify, the pressure signal. Clinically, the presence of a dampened signal will require a larger pressure change (higher work of breathing) to reach the trigger threshold. On the contrary, presence of water in the pressure tubing may cause oscillation, which can falsely trigger mechanical breaths.
Table 3-2: Advantages and Disadvantages of the Different Circuit Pressure-Sensing Sites |Favorite Table|Download (.pdf)
Table 3-2: Advantages and Disadvantages of the Different Circuit Pressure-Sensing Sites
|A. Exhalation port: Well protected from mechanical abuse. During mechanical inhalation, accurately reads pressure at the Y. During inhalation, increases in inspiratory or expiratory circuit resistance do not compromise inspiratory flow output, except for manyfold increases.||Requires protection from moisture of exhaled gas. During spontaneous inspiration, underestimates pressure generated at the Y to trigger the ventilator. During exhalation, underestimates pressure at the Y. During exhalation, increases in expiratory circuit resistance compromise expiratory flow. Hence, system requires well-maintained expiratory filter to ensure that expiratory circuit resistance remains low.|
|B. Inhalation port: Well protected from mechanical abuse. Does not require protection from moisture or additional filters. During exhalation, accurately reads pressure at the Y as long as the inspiratory circuit remains patient. During inhalation, increases in expiratory circuit resistance do not compromise inspiratory-flow output.||During mechanical inhalation, overestimates pressure at the Y. During spontaneous inspiration, underestimates pressure generated at the Y to trigger the ventilator. During inhalation, increases in inspiratory circuit resistance compromise inspiratory flow output. For example, factors such as selection of humidifier and type of patient circuit yield varying patient inspiratory efforts for fixed ventilator settings.|
|C. Patient Y: During inhalation and exhalation, accurately reads both inspiratory and expiratory pressures. Pressure readings reflect relative condition of inspiratory and expiratory circuits.||Susceptible to mechanical abuse. Requires a separate pressure-sensing tube, which is prone to occlusion, blockage, and disconnection, all of which prevent sensing of patient effort.|
The trigger pressure sensitivity is usually set at 0.5 to 1.5 cm H2O below the baseline pressure. Common practice is to increase the sensitivity (i.e., decrease the pressure drop) until autotriggering occurs and then reduce sensitivity until the autotriggering just stops.30 Note that each ventilator comes with predetermined manufacturer set values and can be adjusted.
Flow triggering is based on the detection of a change in a constant, small, baseline (bias) flow through the patient circuit. The operator sets a flow sensitivity threshold. When the change in flow reaches the threshold, a breath is delivered. The changes in flow are detected at the expiratory valves or by a flow sensor in the patient circuit. The ventilator measures the flow from the ventilator and from the patient. In a closed circuit, the two flow values should remain equal in the absence of patient effort. When the patient makes an inspiratory effort, the expiratory flow drops, creating a difference between the inspiratory and expiratory flow values. When the difference in values reaches the preset sensitivity threshold, a breath is delivered. Some systems (Puritan Bennett, 7200) allow the operator to set both the bias flow and the trigger sensitivity. Newer devices set the bias flow according to the operator selected value for the triggering sensitivity. For example, the Puritan Bennett 840 sets the flow 1.5 L/min above the selected sensitivity, and the Hamilton G5 automatically sets the bias flow equal to two times the set sensitivity threshold. As a backup, if flow sensor is kinked or taken out of line, an internal pressure trigger of -2 cm H2O is used until the flow sensor is “online” again.
Flow change may be detected by placing a sensor just before the endotracheal tube. The close proximity to the patient may enhance triggering. It, however, exposes the sensor to secretions and moisture, which may affect its performance. Flow triggering seems more efficient than pressure triggering in terms of work of breathing.34 This, however, seems of no particular clinical relevance in the presence of appropriately set pressure triggering.35 Flow sensing may cause autotriggering secondary to noninspiratory flow changes. The flow change can happen in either the ventilator circuit (leak in the circuit or endotracheal tube) or the patient (cardiogenic oscillations or bronchopleural fistula).36,37
A novel approach to flow triggering is offered on the Dräger Infinity V500 ventilator in the APRV mode. Rather than setting a T-low time to determine the time triggering of each mandatory breath, the operator may set a percent of peak expiratory flow as the trigger threshold.
A breath may be triggered when a preset volume is detected as the result of a patient inspiratory effort. This is similar to flow triggering but using volume has the theoretical advantage of being less susceptible to signal noise (i.e., integrating flow to get volume cancels out some noise because of flow oscillations). Volume triggering is rare in ventilators but can be found on the Dräger Babylog VN500 infant ventilator.
The ideal approach to coordinate a mechanical ventilator with the patient inspiratory effort would be to use the neural output of the respiratory center. Direct measurement of the respiratory center output is currently not possible. The phrenic nerve has been used as a trigger signal in animal models,38,39 but not in humans. The only available clinical approach is measurement of the diaphragmatic electrical activity (Edi). Because the Edi is an electric signal, it easily becomes contaminated by the electrical activity of the heart, the esophagus, and other muscles.21 More importantly, the Edi requires an intact respiratory center, phrenic nerve, neuromuscular junction, and assumes that the diaphragm is the primary inspiratory muscle (e.g., rather than accessory muscles of ventilation).
The only clinically available system that uses diaphragmatic signal trigging is the neurally adjusted ventilatory assistance (NAVA) system. An esophageal catheter is used to measure the Edi. The sensitivity is set by entering a value above the background electrical noise. The trigger value is set in microvolts and represents the change in the electrical signal rather than an absolute value.40 The default setting is 0.5 microvolts, but it can be adjusted from 0 to 2 microvolts. As a backup trigger signal in the absence of a measurable Edi, NAVA uses flow or pressure triggering, whichever happens first.
The BiPAP Vision (Respironics Inc., Murrysville, PA) uses a triggering mechanism called shape-signal. The ventilator microprocessor generates a new flow signal, which is offset from the actual flow by 0.25 L/s and delays it for 300 milli seconds. The delay causes the flow shape signal to be slightly behind the patient’s flow rate. The mechanical breath is triggered when a sudden decrease in expiratory flow from an inspiratory effort crosses the shape signal.41
The Sechrist SAVI system (Sechrist Industries, Anaheim, CA) is the only mode available that uses transthoracic electrical impedance to trigger the ventilator.25 The thoracic impedance is obtained by placing two chest leads, one in the anterior axillary line on the right and the other in the posterior axillary line on the left. The sensors are placed high enough to avoid costal and subcostal retractions. The chest sensors measure the electrical impedance across the human body. As a breath occurs, the transthoracic impedance changes as a result of a different ratio of air-to-fluid in the thorax. The triggering threshold can be adjusted. The cardiac cycle may also cause interference with the signal.22,25
During inspiration, the variable limiting the magnitude of any parameter is called the target variable (previously known as the limit of the control variable, but the term limit is now reserved for alarm and safety conditions rather than control settings).42 A target is a predetermined goal of ventilator output. Targets can be viewed as the parameters of the targeting scheme (see Chapter 2). Within-breath targets are the parameters of the pressure, volume, or flow waveform. Examples of within-breath targets include inspiratory flow or pressure rise time (set-point targeting), inspiratory pressure and tidal volume (dual targeting), and constant of proportionality between inspiratory pressure and patient effort (servo targeting). Between-breath targets serve to modify the within-breath targets and/or the overall ventilatory pattern. Between-breath targets are used with more advanced targeting schemes, where targets act over multiple breaths. A simple example of a between-breath target is to compare actual exhaled volume to a preset between-breath tidal volume so as to automatically adjust the within-breath constant pressure or flow target for the next breath. Examples of between-breath targets and targeting schemes include average tidal volume (for adaptive targeting), percent minute ventilation (for optimal targeting), and combined partial pressure of carbon dioxide, volume, and frequency values describing a “zone of comfort” (for intelligent targeting).
The ventilator uses microprocessors to control the delivery of pressure. The pressure can be delivered with any pressure profile and in response to many signals. Currently, most modes of ventilation in which inspiratory pressure is targeted deliver the pressure rapidly and attempt to maintain the pressure constant throughout the inspiratory phase (square waveform). This means that the performance of the ventilator depends on the delivery of the pressure waveform and any departure from the ideal waveform leads to differences in performance between ventilators.43,44
The pressure rise during inspiration associated with volume and flow delivery is set by the operator (pressure control–continuous mandatory ventilation) or closed-loop algorithms (e.g., pressure-regulated volume control). Care should be exercised while setting the ventilator or reading the literature as there is significant variability between ventilator manufacturers and peer-reviewed literature in the definitions and nomenclature related to inspiratory pressures.43 The main problem stems from what historically has been used to define the inspiratory pressure. For example, in the same ventilator, for pressure control–continuous mandatory ventilation breaths the peak inspiratory pressure is stated in reference to the set end-expiratory pressure (PEEP), but for APRV the peak inspiratory pressure is stated in reference to the atmospheric pressure. To compound the confusion, on some ventilators the value of pressure support is set relative to PEEP (e.g., Drager Evita XL, Puritan Bennett 840), on others (LTV 950) pressure support is set relative to the atmospheric pressure (i.e., atmospheric pressure = zero airway pressure), and on at least one ventilator (BiVent in Servo i) pressure support may be set relative to inspiratory pressure (P-high). Figure 3-12 illustrates the two different ways used to define inspiratory pressure and the four different ways to define pressure support. Figure 3-13 illustrates the proposed solution to this problem.43 In this proposal, the term inspiratory pressure is defined as the set change in airway pressure during inspiration relative to set end-expiratory airway pressure during pressure-control modes.
Idealized airway pressure waveform showing various conventions used for pressure parameters. Note that there are two ways to define inspiratory pressure for mandatory breaths (green) and four ways to define inspiratory pressure (i.e., pressure support) for spontaneous breaths (red). CPAP, continuous positive airway pressure; PEEP, positive end-expiratory pressure; P-high, high pressure; P-low, low pressure. (Reproduced, with permission, from Chatburn RL, Volsko TA. Documentation issues for mechanical ventilation in pressure-control modes. Respir Care. 2010;55(12):1705–1716.)
Idealized pressure, volume, and flow waveforms for pressure control and volume control illustrating the use of proposed conventions for both set and measured airway pressures. IP, inspiratory pressure; PEEP, positive end-expiratory pressure; PIP, peak inspiratory pressure; Ppeak, peak pressure; Pplt, plateau pressure; PS-Patm, pressure support relative to atmospheric pressure; PS-PEEP, pressure support relative to positive end expiratory pressure; PS-PIP, pressure support relative to peak inspiratory pressure. (Reproduced, with permission, from Chatburn RL, Volsko TA. Documentation issues for mechanical ventilation in pressure-control modes. Respir Care. 2010;55(12):1705–1716.)
On some ventilators, inspiratory pressure rise is set relative to atmospheric pressure rather than set end-expiratory pressure. To distinguish this from inspiratory pressure as defined relative to PEEP, the term peak inspiratory pressure has been proposed.43 In contrast “peak airway pressure” is the measured peak airway pressure relative to atmospheric pressure. Often, for a good pressure-control system, there is seemingly no difference between set peak inspiratory pressure and measured peak airway pressure on the airway-pressure waveform during pressure-control modes. And even if the operator sees a transient small difference, this is not considered clinically important in most nonalarm cases. This leads clinicians to conceptually oversimplify what they see and make the mistake of assuming inspiratory pressure and peak airway pressure are synonymous. For example, measured peak airway pressure is often higher than set peak inspiratory pressure because of pressure transients from an underdamped pressure-control system or noise from patient movement. The introduction of the so-called active exhalation valve made possible unrestricted spontaneous breaths during the inspiratory phase of a mandatory pressure-control breath. New modes brought new terms. For example, P-high or PEEP high refers to the peak inspiratory pressure above atmospheric pressure in APRV (again, there is no standardization of either terminology or symbology in this mode).
The Drager Evita XL, when set in volume-control modes, allows the operator to set the maximum pressure (Pmax) that can be achieved during the delivery of a mandatory breath. The goal is to prevent pressure peaks while maintaining the set tidal volume. When the Pmax is reached during a given inspiration, the ventilator switches from volume control to pressure control (dual targeting) using the Pmax setting as the inspiratory pressure target. If the set tidal volume cannot be reached in the set inspiratory time, the ventilator will alarm.45
The speed with which the airway pressure reaches the set inspiratory pressure is called the rise time. (Rise time for flow can be set in the Maquet Servo i, but this feature is rare on ventilators.) The rise time may be set by the operator or automatically adjusted based on a computer algorithm (e500, Newport Medical Instruments Inc, Newport Beach, California). The name used to indicate pressure rise time varies by ventilator brand (e.g., inspiratory slope, P-ramp, plateau%, and slope rise time). Adjusting the rise time influences the synchronization between the patient and the ventilator secondary to changes in the initial inspiratory flow rate. The lower the rise time, the faster the pressurization rate46 and the higher the peak inspiratory flow.47 A higher initial inspiratory flow rate may decrease the work of breathing but can lead to patient discomfort and worse patient–ventilator synchrony. Conversely, too slow a rise time may result in increased work of breathing and longer mechanical inspiratory time, leading to a dissociation between patient breathing effort and the mechanical breath. That is, the relation between work of breathing, respiratory drive, and comfort with the duration of the rise time is not proportional.46,48 Because rules for setting an optimal rise time are lacking, based on these studies, both very rapid and slow rise time should be avoided. A more gradual rise may be needed in awake patients (for comfort) or patients with low compliance to prevent pressure overshoot and premature cycling of inspiration (Fig. 3-14).
Examples of different pressure rise times in three breaths in pressure-support mode. A. Rise time is set very low, resulting in a lower peak inspiratory flow. B. Rise time is set higher, resulting in a higher peak flow and shorter inspiratory time. C. Rise time is set very high, resulting in “ringing” of airway pressure signal and peak flow that is uncomfortable to the patient, who exerts an expiratory effort and prematurely terminates inspiration (indicated by the positive deflection of esophageal pressure). (Reproduced, with permission, from Macintyre NR. Patient-ventilator interactions: optimizing conventional modes. Respir Care. 2011;56(1):73–81.)
The operator is required to enter a tidal volume in any volume-control mode. This may be a direct setting or an indirect one by setting frequency or minute ventilation. The ventilator will control the tidal volume and the pressure will be the dependent variable. A tidal volume target, however, may also be set when the mode uses adaptive targeting in pressure control (e.g., pressure-regulated volume control [PRVC] on the Maquet ventilators).49 In such a case, inspiratory pressure is automatically adjusted between breaths by the ventilator to achieve an average measured tidal volume equal to the operator set target. There are four basic ways ventilators deliver a preset tidal volume (from least used to most commonly used):
By measuring the volume delivered and using the signal in a feedback control loop to manipulate the volume waveform.
By the displacement of a piston or bellows. An example of this is the Puritan Bennett LP10 home-care ventilator (piston) or some anesthesia ventilators (bellows).
By controlling the inspiratory pressure within a breath and automatically adjusting it between breaths to deliver a minimum set tidal volume. The volume delivered is targeted by a closed-loop algorithm, known as adaptive pressure control (see Chapter 2). This targeting scheme is available in most modern critical care ventilators under multiple names (e.g., PRVC, autoFlow, VC+, APV). A common confusion is that this is a volume-control mode, when, by the equation of motion, what is being controlled is pressure during a breath. A caveat with this targeting scheme is that in the presence of the patient’s inspiratory efforts, the tidal volume may be higher than set, and the support provided by the ventilator may be inappropriately low.50,51
By controlling flow, the volume delivered is indirectly controlled. Because flow and volume are inverse functions of time (i.e., volume is the integral of flow and flow is the derivative of volume), controlling one controls the other. In simple ventilators, there is no feedback signal for flow, just a known flow for an adjustable amount of inspiratory time. On more sophisticated ventilators, the operator can regulate the shape of the inspiratory flow waveform. A square waveform will create higher peak airway pressures and will require less time to deliver the set volume (which may result in lower mean airway pressures) than a descending ramp pattern.52–54 Some ventilators offer one waveform (e.g., the Dräger Evita XL offers only the square waveform) others have more (e.g., the Hamilton Veolar offers 50% or 100% descending ramps, sinusoidal, and square).55 Most current ventilators only provide the square waveform or a descending ramp profile.
Figure 3-15 compares volume delivery between standard volume and pressure control modes versus modes using adaptive pressure control.
Volume delivery in volume control (VC) and pressure control (PC) modes using set-point targeting versus pressure control using adaptive targeting. Notice how tidal volume (flow) remains constant in volume control with set-point targeting in the setting of increased patient effort. In adaptive pressure targeting, the inspiratory pressure is adjusted by an algorithm to keep the tidal volume at a target. The tidal volume, however, may be larger if the patient effort is large enough. In set-point pressure targeting, the pressure remains constant, and the tidal volume increases in response to patient effort.
In volume-control modes, the minimum minute ventilation is set by entering the tidal volume and respiratory rate. This assures that the patient will receive a minimum amount of ventilatory support. Some modes provide the option to enter a target minute ventilation (as a percent of the calculated minute ventilation for a given ideal body weight, adaptive-support ventilation [ASV]; e.g., Hamilton G5), while others will calculate it from the entered tidal volume and respiratory rate (mandatory minute volume [MMV]; e.g., Dräger Evita XL). The concept of automatically adjusting the ventilator settings to maintain a constant minute volume was first described by Hewlett and Plat in 1977.56 As implemented, for example, on the Dräger Evita XL ventilator, MMV is a form of volume control–intermittent mandatory ventilation. The operator presets the target minute ventilation by setting tidal volume and frequency. The ventilator then monitors the total minute ventilation as the sum of the minute ventilations generated by mandatory and spontaneous breaths. If the total minute ventilation is below the target value, the mandatory breath frequency will increase. As long, however, as the spontaneous minute ventilation is at least equal to the target value, mandatory breaths will be suppressed. In this way, the proportion of the total minute ventilation generated by spontaneous breaths can range from 0% to 100%. As a result, MMV may be considered a mode of automatic weaning.
Another version of MMV was used on the Hamilton Veolar ventilator (now obsolete); the target minute ventilation was maintained by automatic adjustment of inspiratory pressure (adaptive pressure support). That mode was replaced by ASV on newer Hamilton ventilators.49 ASV is the only commercially available mode to date that uses optimal targeting. It was first described by Tehrani in 1991.57 The operator inputs the patient’s height and percent of minute ventilation to be supported (25% to 350%). The ventilator then calculates the ideal body weight and estimates the required minute alveolar ventilation assuming a normal dead space fraction. Next, an optimum frequency is calculated based on work by Otis et al9 that predicts a frequency resulting in the least mechanical work rate. The target tidal volume is calculated as minute ventilation divided by respiratory frequency (MV/f). In ASV, there are two breath patterns based on the patient’s respiratory effort. If there is no patient effort, the ventilator delivers adaptive pressure-control ventilation; if there is patient effort, the patient receives adaptive pressure support. In both instances, the inspiratory pressure within a breath is controlled to achieve a target tidal volume.49
Table 3-3 summarizes the determinants of minimum and maximum minute ventilation for some common modes.
Table 3-3: Determinants of Minimum and Maximum Minute Ventilation for Some Common Modes |Favorite Table|Download (.pdf)
Table 3-3: Determinants of Minimum and Maximum Minute Ventilation for Some Common Modes
|Mode Name||A/C||SIMV||MMV||ASV||Smart Care|
|Operation||Operator enters a set rate and tidal volume. Patient may trigger breaths above set rate.||Operator enters a set rate and tidal volume. Patient may breath in between mandatory breaths with or without assistance.||Operator enters a set rate and tidal volume. Patient may breath with or without assistance. If his minute ventilation falls below minimum, then mandatory breaths initiate at a set rate.||Adaptive pressure control breaths target tidal volume and rate according to mathematical model.||Pressure support is titrated based on expert rules to achieve the range etPCO2.|
|Minimum minute ventilation||set VT × set f||set VT × set f||set VT × set f||Targeted by ventilator based on operator-entered body weight.||Targeted by ventilator to maintain “comfort zone” based on VT, f, and etPCO2.|
|Maximum minute ventilation||Variable: VT × total f||Variable: VT × total f||Variable: VT × total f||Variable but ventilator will reduce support if patient attempts to increase above estimated minute ventilation requirement.||Variable but ventilator will reduce support if patient attempts to increase above estimated minute ventilation requirement.|
The inspiratory flow can be adjusted by the operator on most ventilators that provide volume-control modes (see “Tidal Volume” above). In general, the ventilator operator will choose a peak flow and may have some waveform pattern options (e.g., square waveform or descending ramp). Although these settings appear simple, there are several points that may cause differences in performance and interpretation of data. First, the ventilator uses a microprocessor to control the delivery according to the preset tidal volume, inspiratory time, flow pattern, pressure limits, and ventilator-specific algorithms. During the breath, the flow delivery is adjusted according to a closed-loop feedback mechanism and proprietary software.2 The consequence is a difference in performance among ventilator brands, even in the same mode.27 Second, the interface may add confusion. For example, in the Dräger Evita XL, while on volume control, the operator will need to set the inspiratory flow, the inspiratory time, and tidal volume, whereas on the Hamilton G5, the options are customizable in three different ways! (Hopefully, all conducive to the same output.) The operator can enter (a) the I:E and the percent pause in inspiration, (b) the peak inspiratory flow and inspiratory time, or (c) the percent inspiratory time and plateau pause time. Underscoring that knowledge of the device used is essential. Finally, to add to the confusion, there are incorrect conclusions that sometimes permeate practice:
In pressure-control mode, the flow is controlled as a descending ramp. In a pressure-controlled breath, the volume and the flow are the manifestation of the respiratory system characteristics (resistance and compliance) and the patient’s respiratory effort. If the patient is passive (no respiratory effort), the flow will decay exponentially (see Fig. 3-7, A). If the patient has a respiratory effort, the flow pattern will be variable, according to the characteristics of the patient effort, the ventilator settings (inspiratory pressure, pressurization algorithm, triggering, etc.), and the respiratory system characteristic. The only way to have a standard descending ramp is to select that waveform and have the computer control the flow delivery in volume control.
The “autoflow” function adjusts the flow in a volume-controlled breath to the patient’s demand. Autoflow is available in Dräger Evita ventilators. It appears as an add-on for three modes of volume-control ventilation (controlled mechanical ventilation [CMV] or intermittent positive-pressure ventilation [IPPV], SIMV, and MMV). This “add on” is defined in the manual as automatic regulation of the inspiratory flow adjusted to the changes in lung conditions and to the spontaneous breathing demands.58,59 What this “add on” does is turn the mode from a volume-control mode to an adaptive pressure-control mode. This is the same as being on PRVC on the Maquet ventilators. They all automatically adjust the inspiratory pressure to achieve a target tidal volume and because this is a pressure-controlled breath, the flow will be variable (see “Tidal Volume” above).
The inspiratory flow setting has importance at different levels. The work of breathing is related to the peak flow and the pressurization rate. The balance between patient and ventilator work of breathing will be affected by the inspiratory flow setting. In regards to cycling, high flows can lead to high peak inspiratory pressures (peak inspiratory pressure [PIP] is directly proportional to resistance, the higher the flow, the higher the PIP), which may lead to reaching the pressure or flow-cycling threshold and ending the breath prematurely.59 But a more practical issue is this: does the flow-wave shape itself have any effect on patient outcome? Like most other questions about ventilator settings affecting patient outcome, after more than 30 years of research on this particular subject we still do not know the answer.
Studies from the early 1960s to early 1980s produced conflicting results, prompting Al-Saady and Bennett to design a better-controlled study, keeping tidal volume, minute ventilation, and I:E ratio constant.60 They discovered that compared to a constant inspiratory flow, a descending ramp flow (what they and many subsequent authors have called “decelerating flow”) resulted in a lower peak airway pressure, total respiratory resistance, work of inspiration, dead space-to-tidal volume ratio, and alveolar–arterial oxygen tension gradient. They also noted an increase in compliance and partial pressure of arterial oxygen (PaO2) with no changes in partial pressure of arterial carbon dioxide (PaCO2) or any hemodynamic variables. In 1991, Rau et al compared peak and mean airway pressure for seven different inspiratory flow waveforms (including square, ascending and descending ramps, and sinusoidal) under three different lung model conditions.54 For all models, the descending ramp flow waveform produced the lowest peak and the highest mean airway pressures, whereas the ascending ramp produced the opposite: the highest peak and lowest mean values. When compliance was low, mean airway pressure increased as peak airway pressure increased. When resistance was high, peak airway pressure was more affected by the peak flow setting than the waveform setting.
In 1996, Davis et al52 tested the hypothesis that a descending ramp flow waveform is responsible for improvements in gas exchange during pressure-control ventilation for acute lung injury. They compared volume control with a square or descending ramp waveform to pressure control with a square pressure waveform. Both pressure control and volume control with a ramp waveform provided better oxygenation at lower peak airway pressure and higher mean airway pressure compared to volume control with the square-flow waveform.
Polese et al61 compared square, sinusoidal, and descending ramp flow waveforms in patients after open heart surgery. They found that PaO2 and PaCO2 were not affected by changes in waveform. Peak airway pressure was highest with the sinusoidal waveform while mean airway pressure and total work of breathing were least with the square waveform. Yang et al53 applied square, sine, and descending ramp flow waveforms to patients with chronic obstructive pulmonary disease (COPD) and found that the descending ramp reduced inspiratory pressure, dead space-to-tidal volume ratio, and, PaCO2 but increased alveolar–arterial oxygen tension difference with no change in arterial oxygenation or hemodynamic variables.
Our own experience is that many clinicians prefer the descending ramp flow waveform when using volume control modes, with the observation that patients tend to be more comfortable, perhaps because of the higher flow earlier in the inspiratory phase.
Figure 3-16 illustrates an algorithm that can be used to adjust inspiratory flow to improve patient–ventilator synchrony.62
Algorithm for improving patient–ventilator synchrony. AI, asynchrony index, percent of inspiratory efforts that failed to trigger a breath; COPD, chronic obstructive pulmonary disease; PEEPi, intrinsic PEEP (aka auto-PEEP). (Modified from, with permission, Sassoon CSH. Triggering of the ventilator in patient-ventilator interactions. Respir Care. 2011;56(1):39–48.)
Proportional-assist ventilation (PAV)63 delivers pressure-control breaths with a servo targeting scheme (see Chapter 2).49 The pressure applied is a function of patient effort: the greater the inspiratory effort, the greater is the increase in applied pressure (Fig. 3-17). The form of PAV implemented on the Dräger Evita XL ventilator (called proportional pressure support) requires the operator to input desired assistance values for elastance and resistance. PAV implemented on the Puritan Bennett 840 ventilator (called PAV +) uses a different algorithm. It automatically calculates the resistance of the artificial airway, and combines resistance and elastance such that the operator enters only a single value representing the percentage work of breathing to be supported.64 The design differences between proportional pressure support and PAV + lead to significant performance differences.65
Pressure, volume, and flow waveforms for proportional assist ventilation.
Neurally Adjusted Ventilatory Support Level
NAVA is a mode that applies airway pressure proportionately to patient effort based on the voltage recorded from diaphragmatic activity. The “NAVA level” is the constant of proportionality (gain) between voltage and airway pressure. The operator enters the NAVA level, then the ventilator delivers pressure equal to the product of gain and the Edi. In simple terms, it states how much pressure the patient will receive for each microvolt of diaphragmatic activity:
where Paw(t) is the airway pressure (cm H2O) as a function of time (t), Edi(t) is the electrical activity of the diaphragm as a function of time (t), in microvolts (μV), and the NAVA level is the operator-set level of support in cm H2O/μV. The range is 0 to 30 cm H2O/μV.
The NAVA level is set according to the operator ventilation goals, level of inspiratory pressure support, tidal volume, apparent patient work of breathing, or respiratory rate. Recently, Roze et al66 proposed using the maximum Edi during a spontaneous breathing trial to help set the NAVA level (Fig. 3-18). By titrating the NAVA level to the a target Edi, the goal is to avoid excessive diaphragmatic unloading as well as respiratory muscle fatigue.
Airway pressure, flow, and electrical diaphragmatic activity curves in pressure support (left) and in neurally adjusted ventilatory assist (right). Edi, electrical activity of the diaphragm; PEEP, positive end-expiratory pressure, Td, trigger delay; Tiex, inspiratory time in excess; Tin, neural inspiratory time; Tiv, ventilator pressurization time. (Reproduced, with permission, from Piquilloud L, Vignaux L, Bialais E, et al. Neurally adjusted ventilatory assist improves patient–ventilator interaction. Intensive Care Med. 2011;37(2):263–271.)
Automatic Tube Compensation
Automatic tube compensation (ATC) is a mode that compensates for the flow-dependent pressure drop across an endotracheal tube during inspiration and expiration. It is thus intended to reduce or eliminate the resistive work of breathing imposed by the artificial airway. ATC is an add-on feature on several ventilators. When ATC is activated, the ventilator supplies airway pressure in proportion to the square of flow times, a gain factor that is determined by the size of the endotracheal tube. Because flow is positive during inspiration and negative during expiration, ATC pressure either adds to inspiratory pressure or subtracts from expiratory pressure (Fig. 3-19). Some ventilators calculate and display tracheal pressure as airway pressure minus ATC pressure. ATC can be used alone or added to the ventilating pressure in pressure-control modes. Interestingly, the way ATC was implemented in the intensive care unit ventilators is different from the original concept, where negative pressure could be applied during exhalation.67,68
Pressure waveforms illustrating automatic tube compensation (ATC). (Modified, with permission, from Dräger Medical AG & Co. KG. Infinity V500 Operator’s Manual. Luebeck, Germany.)
The inspiratory phase of a mechanical breath ends (cycles off) when a threshold value for a measured variable is reached. This variable is called the cycle variable, and it ends the inspiratory time. Cycling is characterized by the initiation of expiratory flow. The cycle variable may be preset (by the operator or the ventilator manufacturer), or automatically defined by the ventilator. Many different signals are used, for example, time, volume, pressure, flow, diaphragmatic signal, and thoracic impedance.
Inspiratory time is defined as the period from the start of inspiratory flow to the start of expiratory flow. Inspiratory time has two components; inspiratory flow time (period when inspiratory flow is above zero) and inspiratory pause time (period when flow is zero). In pressure-controlled or volume-controlled breaths, the inspiration is cycled (terminated) when the set inspiratory time elapses. In spontaneous modes of ventilation (NAVA, PAV, pressure support), the inspiratory time is dependent on the patient’s own neurally determined inspiratory time, level of support, cycling rule (flow, pressure, time, diaphragm activity), and safety rules (maximum set inspiratory time).
Inspiratory time is usually an operator-entered input but some modes of ventilation can automatically set it and change it based on expert rules and closed-loop feedback algorithms. Two notable algorithms are ASV and Adaptive I-Time. In ASV (Hamilton G5), the inspiratory time is automatically set at one expiratory time constant (of the measured respiratory system characteristics and it is never shorter than 0.5 second or longer than 2 seconds). In the Adaptive Flow and Adaptive I-Time in the Versamed iVent (GE Healthcare, Madison, WI), the ventilator automatically adjusts the inspiratory time and inspiratory flow to maintain a target I:E ratio of 1:2 and deliver the operator-set tidal volume.49
In volume-control modes, there are four possibilities for setting inspiratory time:
Operator sets tidal volume and inspiratory flow: inspiratory time is equal to the tidal volume divided by mean inspiratory flow.
Operator sets tidal volume and inspiratory time: mean inspiratory flow is equal to the tidal volume divided by the inspiratory time.
Operator sets tidal volume, inspiratory flow, and inspiratory time: if the inspiratory time is longer than the inspiratory flow time (set tidal volume divided by set flow), then an inspiratory hold is created and the pause time is equal to the inspiratory time minus the inspiratory flow time. For example, if the tidal volume is 600 mL (0.6 L) and the set inspiratory flow is 60 L/min (1L/s) then the inspiratory flow time is (0.6/1 = 0.6 s). Now, if the operator also sets the inspiratory time to 1 s, an inspiratory pause is created and it lasts 1.0 − 0.6 = 0.4 s.
On some ventilators, the operator sets pause time directly.
In pressure-control modes, the operator presets the inspiratory time directly for mandatory breaths. Thus, prolonging the inspiratory time causes the ventilator to decrease the expiratory time, possibly resulting in air trapping, larger tidal volumes, or cycle asynchrony. One must remember that the effect on tidal volume of the inspiratory time in a pressure-control breath will depend on the respiratory system characteristics (i.e., the time constant). Thus, a patient with a long time constant (high compliance and/or high resistance) will require a longer inspiratory time to achieve full pressure equilibration, cessation of flow, and complete tidal volume delivery.
Figure 3-16 illustrates an algorithm that can be used to adjust inspiratory time to improve patient–ventilator synchrony.62
The inspiratory pause is the period during which flow ceases but expiration has not begun (see inspiratory time). The expiratory valves are closed during this period. The inspiratory pause time is part of the inspiratory time. It is also named plateau time (PB 840, Covidien, Mansfield MA), Pause time (Servo i, Maquet,) or Pause (G5, Hamilton Medical). When set directly, pause time may be entered in seconds or as a percentage of the inspiratory time. When it is activated, most ventilators will display a plateau pressure (i.e., static inspiratory hold pressure). Increasing the inspiratory pause time will increase the mean airway pressure and thus the time the lung is exposed to volume and pressure. This may have a positive effect on oxygenation and ventilation by increasing mixing time and decreasing dead space.69,70
I:E is the ratio of inspiratory time to expiratory time (Fig. 3-20).
Divisions of the inspiratory and expiratory periods. A volume-controlled breath is depicted. A. End of inspiratory flow. B. Start of expiratory flow. C. End of expiratory flow. IFT, Inspiratory flow time; IPT, inspiratory pause time; IT, inspiratory time.
The I:E can also be described as the duty cycle or percent inspiration. In engineering, the duty cycle is defined as the time spent in active state as a fraction of the total time. In mechanical ventilation, the active state is the inspiratory time, and the total time is the sum of the inspiratory and expiratory times. It is expressed as a percentage. The larger the percentage, the longer the inspiratory time in relation to the total cycle time.
One can convert one to the other by the following formula:
Example: A duty cycle of 50% is an I:E of 1:1, a duty cycle of 33% is an I:E 1:2.
The relevance of I:E is highlighted in the context of the time constant. The time constant is a measure of how quickly the respiratory system can passively fill or empty in response to a step change in transrespiratory pressure.23 It is calculated as the product of resistance and compliance. The value obtained is the time that takes to achieve 63% of steady state. This percent change remains a constant, regardless of the combination of resistance and compliance. It follows that each time constant will lead to a 63% decrease or increase in volume. In Table 3-4, one can see the difference among time constants for different lung conditions. In COPD, the time constant is longer so the time required for exhalation is longer than for patients with acute respiratory distress syndrome. This table demonstrates the effect of the time constant during passive exhalation using previously published71 expiratory time constants for three conditions (normal lung was 0.78 seconds, for acute respiratory distress syndrome 0.51 seconds, and for COPD 1 second). In this example, expiration starts from a lung volume of 500 mL above functional resting capacity. When expiratory time equals one time constant, 63% of the tidal volume will be exhaled, leaving 37% of the tidal volume yet to be exhaled.
Table 3-4: Effect of Lung Condition on Time Constant and Expired Volume |Favorite Table|Download (.pdf)
Table 3-4: Effect of Lung Condition on Time Constant and Expired Volume
|Expiratory Time (s)||Expiration|
|Time Constant||Normal Lung||ARDS||COPD||Tidal Volume Remaining (mL)||Tidal Volume Exhaled||Tidal Volume Remaining|
The I:E ratio can be an operator-entered value, or just displayed as a calculated value based on common scenarios for mandatory breaths:
- Preset I:E ratio and frequency.
- Preset inspiration time (TI in seconds) and frequency (breaths/min). The frequency sets the ventilatory period (1/f) and the expiratory time is the period minus TI :
- Expiratory time and inspiratory time are fixed:
Note: some ventilators will synchronize inspiration and/or expiration of a mandatory breath if the patient effort is detected in a trigger/cycle window (e.g., SIMV or APRV), which may alter the I:E from the expected value based on settings.
Pressure cycling occurs when the ventilator reaches a preset peak airway pressure. Pressure cycling is most often a safety feature (i.e., an alarm setting) with current modes of ventilation. When a preset high-pressure alarm threshold is crossed, the ventilator will cycle the ventilator. The goal is to prevent the patient from exposure to hazardous pressures. Pressure cycling without an alarm is the normal operational state for some devices (e.g., VORTRAN automatic resuscitator).
Volume cycling occurs when a preset volume is reached. This occurs when the operator sets a tidal volume in volume-control modes. Volume cycling implies that inspired volume is monitored by the ventilator’s control system during inspiration and compared to a threshold value (the set tidal volume). But on some ventilators, despite the setting of a tidal volume, the actual cycle variable is time, that is, the time it takes to deliver the set tidal volume with the set inspiratory flow. Manufacturers seldom make this distinction clear in the operator’s manual.
Volume cycling can also be found as a default safety feature. In PAV + (Covidien PB 840 ventilator), one of the cycling criteria is volume. Once the operator-preset high inspired tidal volume limit is reached, the ventilator cycles the breath and alarms.
Flow cycling occurs when a preset flow or percentage of the peak flow is reached for pressure-control breaths. Flow cycling is most commonly found with the pressure-support mode but can be added as an “advanced setting” in other pressure-control modes on at least one ventilator (Avea, CareFusion). The flow-cycling threshold preset by the operator has been given many names: expiratory trigger sensitivity (Hamilton ventilators); trigger window (Engstrom Ohmeda); inspiratory termination peak inspiratory flow (Dräger Evita XL); expiratory threshold (Newport); flow termination (Pulmonetics LTV ventilators); PSV cycle (Avea, CareFusion); inspiratory cycle off (Servo i, Maquet); Ecycle (V200 respironics); and E sens (PB 840, Puritan Bennett).
During a breath in the pressure-support mode, the ventilator provides enough initial flow to achieve the set inspiratory pressure. The initial flow is high and then decays exponentially. Some ventilators have a preset default value for flow cycling (range: 5% to 30% of peak inspiratory flow); others allow the operator to adjust it (range: 1% to 80% of peak inspiratory flow). Only one device (e500, Newport Medical, Costa Mesa, CA) has automatic adjustment of the flow-cycling criteria. This device has a proprietary algorithm called FlexCycle. It will change the cycle criterion from 10% to 50% of peak flow based on measurements of airway pressure, the expiratory time constant, and expert-based rules applied through a closed-loop system.72
A default cycle criterion of 25% to 30% of the peak flow seems inappropriate as a “fit all” measure. The goal of adjusting the flow-cycling criterion is to avoid expiratory asynchrony.59 In expiratory asynchrony, the ventilator ends inspiration before or after the patient inspiratory effort. We must remember that flow is a manifestation of the respiratory system characteristics, respiratory muscle effort (inspiratory and expiratory) and the integrity of the lung-ventilator circuit. If the respiratory system has a prolonged time constant, a standard flow-termination criterion may be inappropriate as it will prolong inspiration. That may be the case for patients with COPD, where the standard criterion of 25% may be too low, and lead to expiratory asynchrony and increased work of breathing.73,74 Finally, a leak in the ventilator circuit (mask) or in the patient (endotracheal cuff or a bronchopleural fistula) may lead to lack of decay in the flow curve and thus asynchrony.72
Figure 3-16 illustrates an algorithm that can be used to adjust the flow-cycle threshold to improve patient–ventilator synchrony.62
One goal of mechanical ventilation is to improve the patient–ventilator synchrony. In a perfect setting, the beginning and end of an assisted breath would be correlated with the neural signal driving the inspiratory muscles. In conventional ventilation that is rarely the case.75 NAVA attempts to achieve this goal with the use of an electromyogram signal obtained from the diaphragm (Edi). As diaphragmatic activity decreases, so does the amplitude of the Edi curve. When it decreases below 70% of the peak signal (or 40% when the peak value is low), inspiration is cycled off. As a safety feature there is also a time-cycling mechanism. Piquilloud et al compared NAVA versus pressure support with the usual cycling criteria and found a significant improvement in expiratory synchrony (see Fig. 3-18).76
The baseline variables are the variables controlled during the expiratory time. Expiratory time is the period from the beginning of expiratory flow to the initiation of inspiratory flow. Flow and volume are not directly controlled during this period on any current ventilator. The most common value controlled is pressure relative to atmospheric pressure (zero-gauge pressure).
Positive End-Expiratory Pressure
The PEEP is established by the ventilator exhalation valve. A common source of confusion is the term continuous positive airway pressure versus PEEP. Continuous positive airway pressure is generally considered to be a mode on mechanical ventilators (or a mode of treatment for sleep apnea), whereas PEEP is the elevation of the baseline pressure during any mode of ventilation and is generally a setting for a mode. Until recently, the selection of PEEP has been a relatively arbitrary process and the meaning of “optimum PEEP” is debatable.77 Now, Hamilton Medical has developed the INTELLiVENT system for the G5 ventilator that uses an algorithm for automatic targeting of PEEP and FIO2. A closed-loop algorithm based on expert rules defines the response of the ventilator to measured ventilation variables, end-tidal carbon dioxide and pulse oximetry.
P-low is one of the settings entered for so-called “bilevel” modes like APRV (Fig. 3-21). P-low is just another name for PEEP. Similar to PEEP, the settings are dependent on the user. There is, however, a large discrepancy with the objective of PEEP. In APRV, P-low is set to zero.78 The goal is to maintain lung recruitment with the use of auto-PEEP induced by short T-low settings. P-low can also be set based on the biphasic model,79 where complete exhalation is allowed and P-low is then set with the same goals as PEEP.
Differences in P-low and T-low settings for airway pressure release ventilation and biphasic positive airway pressure. Notice the difference in I:E ratio. The operator enters P-high, P-low, T-high, and T-low. The patient may breath spontaneously. Green curves show flow and blue curves show inspiratory effort.
Expiratory time is defined as the period from the start of expiratory flow to the start of inspiratory flow. As stated above, the expiratory time is commonly dependent on the set inspiratory time, and set respiratory rate. It is rarely a fixed value. This occurs because making it a fixed value would produce, in most modes, changes in the inspiratory phase (inspiratory time, flow, and pressure). The most common exception to this is on ventilators that offer some form of APRV/biphasic pressure-control mode where expiratory time is set as “T-low.”
With exception of APRV/biphasic, in all the modern modes of ventilation the expiratory time is dependent on the inspiratory time and frequency; it is not an operator-set value. In APRV/biphasic, the operator sets the time spent at lower pressure, that is, exhalation (see Fig. 3-21). T-low can be set by the operator based on the peak expiratory flow,78 targeting exhaled tidal volume or allowing complete exhalation.80 Setting T-low sets the time trigger threshold for mandatory breaths. Among the methods described in setting T-low in APRV, targeting percent of peak expiratory flow (%PEF) is perhaps the most promoted method. The goal is to set the T-low short enough to avoid full exhalation, thereby generating air trapping.29 Adjusting T-low on the ventilator to manually maintain %PEF at 50% to 75% may be a tedious process, which may seem simple on paper, but in a spontaneously breathing patient can become a true challenge. Newer ventilators, like the Dräger Evita Infinity V500, have attempted to make the process easier by allowing the operator to set a trigger threshold based on a percentage of peak expiratory flow.
Ventilator alarms bring unsafe events to the attention of the clinician. Events are conditions that require clinician awareness or intervention. Events can be classified according to their level of priority.81
Immediately life-threatening events are classified as Level 1. They include conditions like insufficient or excessive gas delivery to the patient, exhalation valve failure, control circuit failure, or loss of power. Level 1 alarm indicators should be mandatory (cannot be turned off by the operator), redundant, and noncanceling.
Level 2 events range from mild irregularities in machine function to dangerous situations that could threaten patient safety if left unattended. Some examples are failure of the air–oxygen blending system, inadequate or excessive PEEP, autotriggering, circuit leak, circuit occlusion, inappropriate I:E ratio, and failure of the humidification system. Alarms in this category may be self-canceling (i.e., automatically turned off if the event ceases) and are not necessarily redundant.
Level 3 events indicate changes in the amount of ventilator support provided to the patient consequent to changes in the patient’s ventilatory drive or respiratory system mechanics and the presence of auto-PEEP. These events often trigger the same alarms as Levels 1 and 2.
Level 4 events are based entirely on patient condition. They may include events such as changes in gas exchange, dead space, oxygenation, and cardiovascular functions. Ventilators generally monitor these events and external monitors are required for alarms (the exception being exhaled carbon dioxide-level alarms built into the ventilator display).
Currently, ventilators do not display alarm settings as levels of priority. Instead, they tend to lump them all together on one screen that shows alarm limits and controls for changing them (Fig. 3-22). How to set alarm thresholds is a complicated topic that has been studied but for which little information is available regarding mechanical ventilation. The goal is to minimize false alarms and maximize true alarms. A high false alarm rate leads to clinician habituation and can also lead to inappropriate responses. In a recent study of an intensive care unit, 1214 alarms occurred and 2344 tasks were performed. On average, alarms occurred six times per hour; 23% were effective, 36% were ineffective, and 41% were ignored.82 In another intensive care unit study, alarms occurred at a rate of six per hour. Approximately 40% of the alarms did not correctly describe the patient condition and were classified as technically false; 68% of those were caused by manipulation. Shockingly, only 885 (15%) of all alarms were considered clinically relevant.83 Although these studies did not address mechanical ventilator alarms specifically, it is not hard to imagine similar results for such a study.
Alarm screen from the G5 ventilator. (Reproduced, with permission, from Hamilton Medical.)
Ventilator alarms are usually set by the operator as either an arbitrary absolute value or a percentage of the current value. Examples would be airway-pressure alarms (high and low) set at the current value plus or minus 5 cm H2O or low and tidal volume/minute ventilation set at plus or minus 25% of the current value.81 The problem is that the parameters for which alarms are important, and these three in particular, are highly variable, with significant portions at extreme values.84 Thus, limits set as absolute values or percentages may reduce safety for some extreme values while increasing nuisance events for other values. An alternative approach might be a type of “smart alarm,” whereby the alarm limits are automatically referenced to the current value of the parameter such that extreme values have tighter limits. Further research is needed to identify optimization algorithms (i.e., minimize both harmful and nuisance events).