TY - JOUR
AU - MD, David J. Dries, MSE,
AB - Abstract Controlled Mechanical Ventilation may be essential in the setting of severe respiratory failure but consequences to the patient including increased use of sedation and neuromuscular blockade may contribute to delirium, atelectasis, and diaphragm dysfunction. Assisted ventilation allows spontaneous breathing activity to restore physiological displacement of the diaphragm and recruit better perfused lung regions. Pressure Support Ventilation is the most frequently used mode of assisted mechanical ventilation. However, this mode continues to provide a monotonous pattern of support for respiration which is normally a dynamic process. Noisy Pressure Support Ventilation where tidal volume is varied randomly by the ventilator may improve ventilation and perfusion matching but the degree of support is still determined by the ventilator. Two more recent modes of ventilation, Proportional Assist Ventilation and Neurally Adjusted Ventilatory Assist (NAVA), allow patient determination of the pattern and depth of ventilation. Proposed advantages of Proportional Assist Ventilation and NAVA include decrease in patient ventilator asynchrony and improved adaptation of ventilator support to changing patient demand. Work of breathing can be normalized with these modes as well. To date, however, a clear pattern of clinical benefit has not been demonstrated. Existing challenges for both of the newer assist modes include monitoring patients with dynamic hyperinflation (auto-positive end expiratory pressure), obstructive lung disease, and air leaks in the ventilator system. NAVA is dependent on consistent transduction of diaphragm activity by an electrode system placed in the esophagus. Longevity of effective support with this technique is unclear. RANDOM VENTILATION Controlled Mechanical Ventilation frequently requires sedation and sometimes muscle relaxant administration. This clinical scenario is associated with cranial displacement of the diaphragm contributing to dependent lung collapse. During assisted mechanical ventilation, spontaneous breathing activity may restore the physiological displacement pattern of the diaphragm, recruiting dorsal, and better perfused regions; thus, improving regional ventilation in dependent lung zones.1,–3 Pressure Support Ventilation is the most frequently used mode of assisted mechanical ventilation in contemporary practice.4 A variety of investigators report that Pressure Support Ventilation improves lung function compared with Controlled Mechanical Ventilation in a variety of lung injury models. In appropriate patients, Pressure Support Ventilation may effectively increase arterial oxygenation in the setting of the Acute Respiratory Distress Syndrome. Pressure Support Ventilation effectiveness may be improved by modification of the technique to create random variation in tidal volume or “Noisy” Pressure Support. In Noisy Pressure Support, tidal volume varies randomly within preset parameters. Experimental models of Noisy Pressure Support demonstrate improved oxygenation and decrease in intrapulmonary shunt beyond that observed with conventional Pressure Support Ventilation.5 Improvement in oxygenation during assisted ventilation may be because of either alveolar recruitment and redistribution of ventilation to dependent (better perfused) zones or redistribution of perfusion toward nondependent areas.6,7 If assisted ventilation recruits the lungs, it may reduce tidal changes in ventilation which reflect cyclic opening and closing of small airways and alveoli. If recruitment does not occur, increase in tidal expansion in the dependent lung zones and deleterious tidal hyperinflation in the nondependent areas may be seen. Tidal reaeration and hyperinflation have been suggested to promote Ventilator Associated Lung Injury.8,9 A number of themes emerge from a series of trials on Noisy Pressure Support. First, Pressure Support Ventilation and Noisy Pressure Support improve oxygenation and reduce venous admixture compared with standard lung protective Pressure Assist Control Ventilation. Improvement in oxygenation during Pressure Support Ventilation and Noisy Pressure Support was not associated with increased aeration in dependent lung regions, rather, Noisy Pressure Support and Pressure Support reduced tidal reaeration and hyperinflation compared to Pressure Control Ventilation and redistributed pulmonary perfusion from dorsal to ventral lung regions. Compared with traditional Pressure Support Ventilation, Noisy Pressure Support Ventilation more effectively redistributes pulmonary perfusion from caudal to cranial zones.1,3,10,11 Noisy Pressure Support Ventilation has been compared to other modes of mechanical ventilation including airway pressure release ventilation and bilevel positive airway pressure (BiPAP).1 Again, Noisy Pressure Support appeared to increase variability of respiratory pattern and improve oxygenation by redistribution of perfusion toward ventilated nondependent lung regions with simultaneous lower mean airway pressure (Paw) and no increase in inspiratory effort or cardiac output. Despite incremental improvement in performance with Noisy Pressure Support compared to traditional Pressure Support Ventilation, it remains the ventilator, based on randomly determined pressure supplied, rather than the patient determining tidal volume and respiratory muscle deconditioning with relative disuse may be more prominent. ASSISTED VENTILATION: PROPORTIONAL ASSIST VENTILATION Proportional Assist Ventilation (PAV) is a synchronized ventilator support mode in which the ventilator generates pressure in proportion to instantaneous patient effort transduced by gas flow at the ventilator. The ventilator amplifies inspiratory efforts of the patient. Unlike other modes of partial support, there are no target flows, desired tidal volumes, minute ventilation standards, or Paw goals. Rather, the objective with PAV is to allow the patient to comfortably obtain whatever ventilation and breathing pattern the respiratory control system demands. The main operational advantages of PAV are synchrony with inspiratory efforts and adaptability of assistance to changes in ventilatory demand.12,13 To appreciate PAV, we see the alveoli and chest wall as an elastic compartment opposing expansion. Elastic recoil pressure is a function of how much lung volume deviates from passive functional residual capacity and the stiffness of the system: elastic recoil pressure (Pel) = V × E where V is lung volume above functional residual capacity and E is respiratory system elastance. In a passive system, elastic recoil pressure increases alveolar pressure as the lung is artificially inflated. During assisted ventilation, inspiratory muscles are active. These muscles decrease alveolar pressure by an amount corresponding to their pressure output (Pmus). At any instant, alveolar pressure (Palv) is the difference between Pel which tends to increase Palv and Pmus which tend to decrease it.12,14  The elastic compartment is connected to the external ventilator tubing via airways and the endotracheal tube. The ventilator controls pressure at the external airway (Paw). Air flows into the lungs when Paw exceeds alveolar pressure. Flow is a function of the difference between alveolar pressure and Paw (resistive pressure) and the resistance of the intervening tubing and airways. Thus,  Ultimately, distending forces are the sum of patient-generated respiratory muscle activity and ventilator-generated Paw and this distending force is opposed by the sum of resistive pressure drop across the tubing and airways and elastic recoil pressure. PAV gas delivery systems allow rapid and free-flow of gas in response to changes in Paw. Flow and volume leaving the ventilator are measured. Gains for the flow and volume signals are adjustable by separate amplifiers: flow-assist (FA) and volume-assist (VA). The summed output of the two amplifiers is the input to the ventilator pump. Thus, the ventilator pressure output is a function of instantaneous flow and volume leaving the ventilator after triggering based on assist settings (below) and detection of respiratory system mechanics (Figure 1). Figure 1. View largeDownload slide Simplified depiction of the way ventilator support is titrated in Proportional Assist Ventilation (PAV) through measurement of instantaneous flow and volume changes. The latter are induced by patient effort. Pmus, muscle pressure; PA, alveolar pressure; Paw, airway pressure; FA, flow assist; VA, volume assist. Printed with permission from: Eur Respir Monogr 2012;55:97–115. Figure 1. View largeDownload slide Simplified depiction of the way ventilator support is titrated in Proportional Assist Ventilation (PAV) through measurement of instantaneous flow and volume changes. The latter are induced by patient effort. Pmus, muscle pressure; PA, alveolar pressure; Paw, airway pressure; FA, flow assist; VA, volume assist. Printed with permission from: Eur Respir Monogr 2012;55:97–115. With amplifier support, a greater effort on the part of the patient will signal provision of more gas from the ventilator. Patient effort provides a positive relation between effort and machine assistance but does not per se cause the assist to be directly proportional to instantaneous effort. Proportionality in PAV is achieved through customized adjustment of the FA and VA gain. FA is the assist pressure per unit flow. If FA is 50% of airway resistance, the ventilator provides 50% of the resistive pressure. At 80% gain, the ventilator assumes 80% of resistive pressure. Likewise, setting the VA gain to 50% causes the ventilator to provide 50% of elastic pressure. The total assisted (Paw) is the sum of flow and volume assist (Figure 2). Figure 2. View largeDownload slide Tracings of flow, volume, airway pressure (Paw), and transdiaphragmatic pressure (Pdi) in a Proportional Assist Ventilation (PAV) breath including a brief end-inspiratory occlusion. Note that flow crosses zero (at cursor) during the declining phase of Pdi and that Paw during the occlusion rises in a manner that is the mirror image of the decline in Pdi (the scales of Paw and Pdi tracings are equivalent). Note also the progressive rise in volume while flow peaks earlier in inspiration before falling (see text). Printed with permission from: Younes et al. Am J Respir Crit Care Med 2001;164:50–60. Figure 2. View largeDownload slide Tracings of flow, volume, airway pressure (Paw), and transdiaphragmatic pressure (Pdi) in a Proportional Assist Ventilation (PAV) breath including a brief end-inspiratory occlusion. Note that flow crosses zero (at cursor) during the declining phase of Pdi and that Paw during the occlusion rises in a manner that is the mirror image of the decline in Pdi (the scales of Paw and Pdi tracings are equivalent). Note also the progressive rise in volume while flow peaks earlier in inspiration before falling (see text). Printed with permission from: Younes et al. Am J Respir Crit Care Med 2001;164:50–60. During the inspiratory phase, volume rises progressively peaking at end expiration. By contrast, flow peaks in early to middle inspiration and falls later. Thus, the relative contributions of resistive and elastic pressures vary considerably during the inspiratory phase. If the same percentage gain is used for both components, total assist represents the same percentage of total pressure regardless of the relative contribution of each. Percentage assist is then constant throughout. However, if different percentage values are used for FA and VA, total assist will represent a different percentage of total applied pressure at different times. To titrate support of the patient, PAV requires measurement of respiratory system mechanics which is achieved by randomly applied 300-millisecond pause maneuvers at end-inspiration at intervals of 4 to 10 breaths14 (Figure 2). Limitations of Proportional Assist Ventilation PAV is designed to respond to changes in flow and volume detected in the ventilator circuit. It is assumed that the pattern of these changes during evolution of inspiration expresses, ultimately, variations in diaphragmatic response to respiratory drive fluctuations. However, flow and volume are only the final response of respiratory input beginning with the brain and proceeding through the respiratory muscles. Thus, any factor which affects the path of neurologic control may impede or compromise coordinated delivery of ventilation assistance.12,15 Central nervous system diseases including cerebral vascular disorders, tumors, infections or injuries, drugs, or metabolic changes such as respiratory alkalosis may depress the respiratory center. Neuronal (Guillain–Barré Syndrome), neuromuscular (myasthenia gravis, muscle relaxants), or muscular disorders (muscular dystrophy, critical illness polymyopathy, fatigue, acid-base or electrolyte changes, shock) may diminish or prevent transformation of respiratory center impulses into mechanical output. Further, the force generated by the dominant inspiratory muscle, the diaphragm, depends greatly on its shape. If diaphragm fibers are shortened or the chest configuration is changed, as happens during hyperinflation, force generation as well as its transformation into negative alveolar pressure may be reduced.12,16 With PAV, signals for pressure delivery are inspiratory flow and volume. Thus, this mode is highly susceptible to dynamic hyperinflation (or auto-positive end expiratory pressure [PEEP]) which affects the triggering function. With volume assist or Pressure Support Ventilation, the ventilator once triggered, delivers a preset volume or pressure regardless of patient effort beyond the trigger point; the patient may relax the inspiratory muscles after triggering leaving the ventilator to deliver the volume, which depending on settings, may be substantial. On the contrary, in PAV, immediately after ventilator triggering, pressure delivery is driven by the patient. Because assistance will automatically terminate at the end of inspiratory effort, any delay in the onset of assistance reduces the fraction of neurologic inspiratory time which is receiving assistance. In addition, the level of effort required to trigger the ventilator is not assisted throughout the breath and the magnitude of pressure applied during the duration of neural inspiratory time is proportional only a fraction of patient effort. In critically injured patients, increasing dynamic hyperinflation (or auto-PEEP) to as little as 3.5 cm H2O decreases the portion of supported inspiratory effort from 86 to 66%. Therefore, a large portion of a breath may be unassisted even though a high gain is used in the PAV mode. Every effort should be made to reduce the magnitude of dynamic hyperinflation caused by increased airway resistance such as administration of bronchodilators or corticosteroids or to counterbalance dynamic hyperinflation or by adding external PEEP to compensate for auto-PEEP. Delayed triggering, because of a high threshold of triggering or slow ventilator response, also decreases the portion of assisted inspiratory efforts. Thus, events during the triggering phase may significantly affect function in PAV.17 Another major disadvantage of the original PAV mode was a complicated algorithm for quantification of respiratory system mechanics. A more recent version described as PAV-plus has improved noninvasive measurement of respiratory mechanics. However, this new measurement algorithm is also limited in patients with expiratory flow compromise and dynamic hyperinflation. These patients may not be good candidates for PAV.17 Because of operating principles of PAV, there is the potential for excessive pressure or volume delivery or “runaway.”12 Runaway occurs when pressure provided by the ventilator is greater than the sum of elastic and resistive pressures at some point during inflation. The ventilator continues to deliver volume despite the fact that the patient has terminated the inspiratory effort. Volume continues to increase until an alarm limit (pressure or volume) is activated or the compliance of the respiratory system is decreased, because the system approaches total lung capacity because of overinflation or when expiratory muscles are recruited by the patient as an attempt to exhale.17 Runaway occurs because flow or volume assist are set to values higher than respiratory system elastance or resistance. Nonlinearity of the pressure–volume and pressure–flow relationships may also cause volume assist and flow assist to be inappropriately set. Similarly, endotracheal tube resistance varies depending on flow rates. Therefore, flow assist that is appropriate more than a specified range of flow rates may not be appropriate for flows outside this range causing excessive assistance. With automatic measurement of respiratory system mechanics, runaway occurs infrequently and only when the preset percentage of assist approaches 90%. Senior investigators suggest that with PAV-plus, runaway was not observed if the gain was less than 80 to 85%. Similarly to over-assistance, air leaks encountered during noninvasive application of PAV dissociate Paw from muscle pressure, causing a runaway phenomenon.12 In the presence of leaks, the ventilator misinterprets flow and volume resulting in continuous patient support extending inspiratory assistance into patient exhalation. Therefore, leak compensation is important to apply with use of PAV both as a mode of invasive and noninvasive ventilation. A large bronchopleural fistula is also a deterrent for PAV application. Georgeopoulos in a recent review summarized key observations for optimal use of PAV.12,17,–21 Initial PEEP and oxygenation fraction are set using traditional criteria. Hypoxemia is managed by adjustment of PEEP and FiO2. Immediate response to initiation of PAV varies depending on whether the patient was over assisted or whether there was dysynchrony in the previous mode of ventilation. It is reasonable to start with a level of support sufficiently high to aid the patient but with a gain setting low enough to avoid the runaway phenomenon described above. An initial gain of 70% is proposed. Patient response to initial settings may range from very shallow breathing consistent with the presence of a significant number of ineffective efforts in the previous mode to episodes of apnea consistent with over-assistance on previous ventilator settings. Tidal volume may vary significantly on PAV. This reflects normal breath variability. The clinician should watch closely for signs of distress. A high respiratory rate does not indicate distress in and of itself. Other signs of distress such as changes in heart rate or blood pressure or the use of accessory muscles should be evaluated. PCO2 may rise after switching to PAV. Typically, this is because of over-ventilation before initiation of PAV. If this occurs, central apnea may be observed after initial transition to PAV. Clinical evidence of distress at 70% assist (gain) settings is relatively uncommon and usually occurs with delayed triggering because of severe dynamic hyperinflation or muscle weakness. Distress may also come with low compliance at low lung volumes as may be seen in patients with obesity, abdominal distention, or acute lung injury. These patients also are frequently hypoxemic. Increase in PEEP may be beneficial in this setting. Patients with continued distress at 70% assist even after PEEP adjustment may benefit from stepwise increase in assist up to 90%. Here, the runaway phenomenon is a risk. Some patients will continue to have distress even with assist levels of 85 to 90%. Frequently, these patients have a large portion of inspiratory effort which is not supported because of triggering delay. In some patients with severe obstructive lung disease and expiratory flow limitation during passive expiration, a low calculated value of expiratory resistance is obtained. For these patients, low expiratory resistance reflects endotracheal tube rather than airway resistance. PAV may not be adequate to support these patients even with a high gain setting. Finally, patients who deteriorate with decreasing percentage assist are not candidates for rapid weaning from mechanical ventilation. In severely injured patients, percentage of assist and PEEP are reduced slowly (possibly more than several days).17 ASSISTED VENTILATION: NEURALLY ADJUSTED VENTILATORY ASSIST (NAVA) As discussed above, the majority of patient-triggered or cycled modes of ventilation are controlled by Paw, flow, and/or volume measured in the respiratory circuit. Limitations of these signals to trigger and cycle off ventilator assist have been summarized above. Despite the goal of “patient-triggered ventilation,” patient ventilator dysynchrony occurs in at least 25% of patients ventilated with standard modes and is associated with prolonged duration of ventilation. Patients with frequent ineffective triggering also tend to receive excessive levels of ventilator support and/or sedation. Excessive assistance can cause muscle fiber injury and atrophy of the diaphragm. Conventional ventilation can induce loss of inspiratory muscle force of as much as 75%. Promoting spontaneous breathing and reducing sedation alone or together reduce the duration of mechanical ventilation.16,22,23 Respiratory neurons in the brainstem signal the diaphragm via the phrenic nerves. After neuromuscular transmission, diaphragmatic excitation occurs where action potentials propagate along the diaphragm muscle fibers. This is the source of measureable diaphragmatic electrical activity. The diaphragmatic electrical activity is generated by the neural respiratory output signal and is modulated by input from multiple respiratory reflexes feeding back to the respiratory centers. The diaphragmatic electrical signal is the primary input used to control NAVA.22,24 The latency time from stimulation of the phrenic nerve in the neck to the onset of diaphragmatic compound muscle action potential in healthy subjects is approximately 6 to 8 milliseconds. Diaphragmatic excitation stimulates contraction of muscle fibers and causes shortening. The result of diaphragmatic contraction is expansion of the respiratory compartment which causes lung distension and lowers pleural and alveolar pressures; thereby, lowering Paw and creating inspiratory flow. These signals (pressure, flow, and volume) are used in various ventilator systems to control conventional patient-triggered ventilation. The time between central respiratory output to the generation of inspiratory flow in a healthy subject is approximately 26 to 28 milliseconds. Factors including auto-PEEP, increased respiratory work load, impaired respiratory muscle function, and reduced respiratory drive secondary to sedation alone or in combination with analgesia will weaken the flow signal. A reduced flow signal is more difficult to detect by current ventilator flow transducers increasing the time delay to trigger ventilator-assist and in the worst case, failing to trigger the ventilator.25,–27 In the late 1950s, electromyography was employed to study diaphragmatic muscle activity. In this work, investigators obtained diaphragmatic electrical activity measurement with electrodes on a catheter passed down the esophagus. As many intubated adult patients in the intensive care unit are equipped with a nasogastric feeding tube, it seemed logical to follow up this work by refining esophageal electrode system measurements of diaphragmatic electrical activity for contemporary clinical use. Obstacles needing attention included development of a standardized and automated method to reduce artifact and filtering effects related to electrode configuration and positioning. The diaphragm electrical activity signal used for monitoring respiratory drive and for controlling the ventilator during NAVA is an electromyography (EMG) signal constituting a temporospatial summation of motor unit action potentials which in turn represent the summary of single fiber action potentials. The most significant advance in the development of esophageal leads for measurement of diaphragmatic electrical activity was the use of bipolar electrodes in a sequential order and an automated processing technique to track the displacement of the diaphragm. Surely, artifacts can influence the measurement of electrical activity of the diaphragm and must be controlled for to avoid misinterpretation of electrical signals from this muscle. In current NAVA circuits, artifact is screened out as part of signal processing with routine NAVA system function.22,28 Synchrony: Unique Opportunity with NAVA The time lag between neural inspiratory input and the occurrence of a ventilator breath affects all steps of the respiratory cycle (initiation, insufflation, cycling off). Among different forms of asynchrony, ineffective triggering from wasted effort is the most common during invasive mechanical ventilation. During noninvasive ventilation, leaks at the patient ventilator interface (mask or helmet) impair function of the pneumatic trigger and cycling system promoting asynchrony patterns such as auto-triggering or prolonged insufflation.29,30 Intrinsic PEEP or dynamic hyperinflation increases patient effort required to trigger the ventilator and increases the likelihood that inspiratory effort will fail to trigger a ventilator breath. Weak inspiratory effort which may occur during situations of low respiratory drive such as excessive ventilation is also a risk factor and is common in patients receiving high assist levels or sedation. For example, an excessive level of Pressure Support is associated with prolonged insufflation with dynamic hyperinflation or intrinsic PEEP.31,–33 During NAVA, ventilator assistance is triggered when the patient initiates an inspiratory effort, even during inspiration with auto-PEEP, and a decrease in the diaphragm signal terminates the assist. NAVA does not depend on measurements of Paw or flow and keeps ventilator assistance synchronous with inspiratory efforts independent of the presence of air leaks or auto-PEEP. NAVA, therefore, has two important features: 1) delivered pressure in theory is synchronous with diaphragmatic activity and 2) tidal volume is completely controlled by the output of the patient's respiratory control center.22,24 A frequent form of minor patient ventilator asynchrony is long inspiratory trigger delay (time lag between the onset of neural inspiration followed by detection of a breath initiated by the patient and finally the onset of ventilator pressurization). Several factors may increase the inspiratory trigger delay during Pressure Support Ventilation including the presence of auto-PEEP. The cycling off interval is the time difference between the end of the neural inspiratory signal and the end of ventilator pressurization. Compared with Pressure Support Ventilation, NAVA substantially improves patient ventilator synchrony by reducing the inspiratory trigger delay and reduces the total number of asynchrony events and by improving timeliness of expiratory cycling off.24 Increasing the level of ventilatory assist with standard modes of ventilation may expose the patient to potentially dangerous levels of volume and pressure and to uncoupling between the patient's neural output and ventilator assistance. In contrast to Pressure Support Ventilation, which is most frequently compared to NAVA, there is good evidence that NAVA offers protection against excessive Paw and tidal volume because there is down regulation of diaphragm signal in response to increasing assistance levels. The net result is a decrease in the amount of assistance provided. Unloading of the respiratory muscles with NAVA is always partial support as some level of spontaneous activity must be maintained to cause assist with NAVA24,31,34 (Figure 3). Figure 3. View largeDownload slide Respiratory muscle unloading with Neurally Adjusted Ventilatory Assist (NAVA). Tracings of electrical activity of the diaphragm (EAdi), volume, ventilator pressure (Paw), and esophageal (Pes) and transdiaphragmatic pressure (Pdi) obtained in a healthy subject performing maximal inspiration with three NAVA levels (zero, intermediate, and high). Increasing the NAVA level resulted in increased ventilator-delivered pressure and deactivation of the diaphragm. Printed with permission from: Sinderby et al. Chest 2007;131:711–7. Figure 3. View largeDownload slide Respiratory muscle unloading with Neurally Adjusted Ventilatory Assist (NAVA). Tracings of electrical activity of the diaphragm (EAdi), volume, ventilator pressure (Paw), and esophageal (Pes) and transdiaphragmatic pressure (Pdi) obtained in a healthy subject performing maximal inspiration with three NAVA levels (zero, intermediate, and high). Increasing the NAVA level resulted in increased ventilator-delivered pressure and deactivation of the diaphragm. Printed with permission from: Sinderby et al. Chest 2007;131:711–7. Several studies have evaluated the impact of increasing Pressure Support levels vs NAVA levels using similar methods of ventilator settings.24 Inspiratory Pressure Support titrated to obtain 6 to 8 mL/kg tidal volume was compared to NAVA. NAVA, in contrast to Pressure Support Ventilation, reduces the risk of over-assistance when the assist level is increased gradually. NAVA also improved patient-ventilator synchrony in contrast to Pressure Support Ventilation regardless of the underlying diagnosis. High levels of NAVA, however, may result in unstable periodic breathing patterns with delivery of high tidal volume followed by periods of apnea and signs of discomfort.35 NAVA has also been associated with improved gas exchange. Reasons for this are unclear. Possible explanations include the retention of a more natural breathing pattern in patients receiving this variable mode of support as opposed to the rigid pattern of support characterizing Pressure Support and other conventional modes of ventilation. Intact neurologic control is intended to keep the PCO2 value relatively constant. Thus, diaphragm activity and breathing pattern must adapt to a variety of conditions to maintain PCO2 within the normal range. Another regulatory goal is optimization of the work of breathing. The rate or depth of breathing can be adjusted to minimize the energy expenditure at a given respiratory effort or to regulate the stretch at the lungs.24,33,36 Ventilatory activity is nonlinear in nature and exhibits a chaotic mathematical complexity. Variability and complexity have been examined during Pressure Support Ventilation and NAVA. Compared with Pressure Support Ventilation, NAVA increased breathing pattern variability and flow complexity without changing diaphragm signal complexity. Accordingly, when the NAVA support level was increased from zero to a high level in healthy individuals, they adopted their inspiratory activity to the NAVA level in order to control tidal volume and regulate PCO2 over a wide range of NAVA settings. In contrast, during high levels of Pressure Support, tidal volume became almost entirely determined by the ventilator and hypocapnia developed. Thus, even at high levels of support with a NAVA system, tidal volume is not imposed by the ventilator but remains under the control of the patient's central respiratory centers. NAVA decreases the risk of over-assistance.24,37 Titration of the NAVA level may be performed by systematically increasing the degree of support to determine the optimal setting in order to unload the respiratory muscles. Transferring patients to NAVA indirectly includes interpretation of several interacting physiological parameters (listed above). While not directly measured, these parameters are accounted for in diaphragmatic signal activity during NAVA titration.22 Another approach considered for titration of NAVA is quantitative analysis of nerve signals to the diaphragm. Two recent studies suggest that the optimal NAVA level is 60 to 75% of the highest diaphragm signal recorded with minimal NAVA support.38,39 NAVA minimizes the risk of overinflation because the duration and level of pressurization remains under control of the respiratory centers. The risk of diaphragmatic inactivity is thus reduced24,40 (Figure 3). Commercially available NAVA systems trigger inspiration based on diaphragmatic nerve stimulation or from standard flow triggering on a first come, first served algorithm. NAVA has shown shorter inspiratory and expiratory trigger delays and fewer asynchrony episodes than Pressure Support. Eleven to forty-two percentage of respiratory cycles in NAVA were triggered by the flow trigger in one review. Shorter triggering delays observed in NAVA are invaluable. Some investigators, however, feel that better synchrony of inspiratory cycle-off is perhaps the most advantageous aspect of this mode of partial ventilatory support.44,45,52 When Pressure Support Ventilation has been compared to NAVA, ventilator inspiratory time increased compared to neural inspiratory time at increasing Pressure Support levels; whereas, the difference between neural and ventilatory inspiratory time remained constant during NAVA regardless of gain level. Ventilator inspiratory time was always longer than neural inspiratory time suggesting that delayed inspiratory cycle-off accounts for much of the difference between ventilator and neural time.41 In all physiological studies, NAVA has been associated with reduced frequency of asynchrony events. When asynchrony is seen, double triggering is the most common phenomenon. Double triggering is thought to occur because of two pneumatic cycles after biphasic neural signaling. Finally, physiological studies have shown higher variability in neural signal amplitude, tidal volume, and respiratory rate and a better match between variability at the neural level and variability of the output variables (tidal volume and respiratory rate) when NAVA is employed.42,43 NAVA must be evaluated in more patients over longer periods in order to understand whether this mode can replace the modes traditionally used during mechanical ventilation during the weaning such as Pressure Support Ventilation.24 One of the remaining questions, however, is situations when it is not desirable to let the respiratory centers drive ventilation. For example, situations of severe metabolic acidosis, of high respiratory drive and of high catecholamine levels may induce extreme hyperventilation which may be deleterious to the lungs. Thus, indications for sedation or muscle relaxant administration must be addressed with greater use of this mode as provision of these agents will preclude use of NAVA.44 Theoretical Benefits of NAVA Systems A number of clinical benefits can be anticipated with NAVA compared to other modes of ventilation. First, and most important, is improvement in patient ventilator interaction. Compared to all other assisted modes, NAVA is least vulnerable to changes in airway disease or airway resistance patterns. The use of neural input from respiratory centers to fully control the ventilator should significantly reduce inspiratory trigger delay especially in patients with dynamic hyperinflation and intrinsic PEEP. Neural input also offers the opportunity to reduce the incidence and severity of respiratory asynchrony and address ineffective triggering and wasted efforts which confound other modes of ventilation. Another important potential benefit is elimination of asynchrony because of air leaks because the neural trigger is independent of pneumatic signals. This advantage is particularly important in the setting of noninvasive ventilation. These studies have been confirmed in preliminary trials in animal models and humans.41,42,45 Another anticipated benefit is decrease in Ventilator-Induced Lung Injury. Preclinical studies demonstrate that NAVA has the potential to be a lung protective strategy because the delivery of excessive tidal volume and transpulmonary pressure is limited by neural feedback from pulmonary stretch receptors. Studies, including work in humans, have confirmed that increasing the level of NAVA support is not associated with increased tidal volume because increase in the assist level leads to corresponding decrease in neural input to the diaphragm which is tracked by NAVA.31,33,37 Diaphragm dysfunction may also be reduced by NAVA as this therapy minimizes the risk of ventilator-induced diaphragmatic atrophy because it is based on continuous coupling between ventilator assistance and patient neural output. During Pressure Support Ventilation, progressive increase in support ultimately leads to disappearance of neural stimuli to the diaphragm surely indicating over-assistance. In contrast, increasing the level of NAVA support can effectively unload the respiratory muscles without abolishing neural stimulation to the diaphragm. This was demonstrated in a study on healthy subjects performing maximal inspiratory maneuvers during NAVA at increasing levels of assistance.34,46 Finally, NAVA offers maintenance of physiologic variability in breathing. Normal breathing pattern in humans is highly variably and exhibits a chaos-like complexity. Unfortunately, almost all available conventional modes of ventilation tend to impose a monotonous pattern of support. It has been suggested that restoring a physiological variability in the breathing pattern during mechanical ventilation could be beneficial. Some workers suggest that reduced breathing variability is associated with an increased rate of weaning failure.42,47,48 Physiological studies comparing NAVA with Pressure Support Ventilation in the setting of critical illness found that while increasing assist levels were associated with comparable muscle unloading, the physiologic response in respiratory pattern during NAVA was different than that commonly observed during Pressure Support Ventilation. With Pressure Support Ventilation, increasing the level of assistance was associated with a decrease in respiratory rate and often an increase in tidal volume. The decrease in respiratory rate is particularly important in the clinical setting since it is interpreted as a sign of relief from respiratory distress. Other authors have observed that contrary to Pressure Support, during NAVA, increased levels of assistance are associated with little increase in tidal volume and no change in respiratory rate. Thus, during NAVA, evidence of muscle unloading and patient comfort may not be accompanied by a decrease in respiratory rate and respiratory rates higher than those commonly accepted in the clinical setting may be observed.31,33,37,42 Investigators favoring the use of NAVA suggest that high respiratory rate should be accepted as best mechanical response in the critically injured patients. These workers suggest that when Pressure Support is set to obtain a protective tidal volume, NAVA set to obtain equivalent peak inspiratory pressure produces respiratory rate similar to Pressure Support and, second, lower respiratory rates during Pressure Support Ventilation are associated with coincident increase in tidal volume and possibly with the occurrence of over-assistance. These investigators suggest that acceptance of respiratory rates above 30 breaths/minute may represent the result of best compromise between patient comfort, protective ventilation, and avoidance of over-assistance, not just during NAVA, but also with other modes of partial ventilatory support. Unfortunately, only a few long-term studies of NAVA are available in critical illness. Alternating use of NAVA and Pressure Support has been reported more than a 24-hour interval. Application of NAVA for 6 hours consecutively has also been reported. Obviously, the setting of critical illness requires longer intervals of support.31,41,42 NAVA: Remaining Issues Physiological studies have confirmed the value of partial ventilatory support with NAVA. Most of this work has been conducted in patients recovering from respiratory conditions or with stable problems. Long-term clinical studies and studies addressing the effectiveness of NAVA in acute respiratory distress are lacking. It is also unclear how to optimally set the NAVA support level. Contrary to Pressure Support, given the negligible effect of NAVA on classical respiratory pattern parameters (above), change in respiratory rate and tidal volume is less informative.39 Another crucial issue is maintenance of high quality signal and interpretation of neural stimulus values for the diaphragm. A variety of factors may affect signal quality such as catheter or patient position changes or poor filtering of the diaphragm signal. Other possible limitations arise in patients breathing predominantly through accessory inspiratory muscles. In these individuals, the amplitude of diaphragmatic neural signaling may underestimate the real amount of muscular effort and inspiration may be more easily triggered by flow than by the neural input to the diaphragm. Similar behavior could be observed in patients with high airway respiratory resistance and flow limitation who force expiration. In these patients, the first part of inspiratory flow may be generated by releasing expiratory muscle and the flow trigger may respond faster than the neural input to the diaphragm and the amplitude of the neural diaphragm trigger may underestimate the generated inspiratory pressure and effective work of breathing.42 ASSISTED VENTILATION: CONCLUSIONS Assisted ventilation offers a dimension in respiratory support not available with current technology. In most cases, this technology will not be relevant in the burn-injured population until acute resuscitation and/or management of an inhalation injury has taken place. However, new approaches to assisted ventilation highlight physiologic strategies necessary for successful transition from traditional mechanical ventilation to spontaneous breathing. Assisted modes may ultimately facilitate weaning in the patient with massive resuscitation associated with resuscitation-related lung injury, inhalation injury, or significant preexisting comorbidities. Thus, appreciation of the techniques described and the latest technology available to address these clinical concerns is appropriate. PAV and NAVA offer improved patient ventilator synchrony. PAV and NAVA manage ventilator assistance using different variables. PAV does not require additional hardware, while NAVA requires placement of a special nasogastric tube. Both modes can be used for invasive and noninvasive ventilation. NAVA is not affected by leaks or auto-PEEP (dynamic hyperinflation) as its function is based on the diaphragm EMG. PAV is unable to adjust to dynamic hyperinflation or auto-PEEP. PAV and NAVA are safe in patients with intact ventilatory drive. These modes may be best employed in patients with chronic respiratory failure where high levels of asynchrony are common. PAV experience has been reported with a variety of patient populations and this mode is available on a variety of ventilators.12,17 However, reported experience with NAVA is limited to small case study which emphasizes brief physiologic trials.31,49,–54 Finally, any patient requiring substantial sedation or having hemodynamic compromise would not be appropriate for PAV or NAVA.55 REFERENCES 1. M Gama de Abreu, PM Spieth, P Pelosi Noisy pressure support ventilation: a pilot study on a new assisted ventilation mode in experimental lung injury. Crit Care Med  2008; 36: 818– 27. Google Scholar CrossRef Search ADS PubMed  2. DJ Falk, KC Deruisseau, DL Van Gammeren, MA Deering, AN Kavazis, SK Powers Mechanical ventilation promotes redox status alterations in the diaphragm. J Appl Physiol (1985)  2006; 101: 1017– 24. Google Scholar CrossRef Search ADS PubMed  3. AR Carvalho, PM Spieth, A Güldner Distribution of regional lung aeration and perfusion during conventional and noisy pressure support ventilation in experimental lung injury. J Appl Physiol (1985)  2011; 110: 1083– 92. Google Scholar CrossRef Search ADS PubMed  4. A Esteban, ND Ferguson, MO Meade Evolution of mechanical ventilation in response to clinical research. Am J Respir Crit Care Med  2008; 177: 170– 7. Google Scholar CrossRef Search ADS PubMed  5. AR Carvalho, PM Spieth, P Pelosi Pressure support ventilation and biphasic positive airway pressure improve oxygenation by redistribution of pulmonary blood flow. Anesth Analg  2009; 109: 856– 65. Google Scholar CrossRef Search ADS PubMed  6. P Neumann, H Wrigge, J Zinserling Spontaneous breathing affects the spatial ventilation and perfusion distribution during mechanical ventilatory support. Crit Care Med  2005; 33: 1090– 5. Google Scholar CrossRef Search ADS PubMed  7. C Putensen, NJ Mutz, G Putensen-Himmer, J Zinserling Spontaneous breathing during ventilatory support improves ventilation–perfusion distributions in patients with acute respiratory distress syndrome. Am J Respir Crit Care Med  1999; 159: 1241– 8. Google Scholar CrossRef Search ADS PubMed  8. RD Hubmayr Perspective on lung injury and recruitment: a skeptical look at the opening and collapse story. Am J Respir Crit Care Med  2002; 165: 1647– 53. Google Scholar CrossRef Search ADS PubMed  9. RA Oeckler, RD Hubmayr Ventilator-associated lung injury: a search for better therapeutic targets. Eur Respir J  2007; 30: 1216– 26. Google Scholar CrossRef Search ADS PubMed  10. PM Spieth, AR Carvalho, P Pelosi Variable tidal volumes improve lung protective ventilation strategies in experimental lung injury. Am J Respir Crit Care Med  2009; 179: 684– 93. Google Scholar CrossRef Search ADS PubMed  11. PM Spieth, AR Carvalho, A Güldner Pressure support improves oxygenation and lung protection compared to pressure-controlled ventilation and is further improved by random variation of pressure support. Crit Care Med  2011; 39: 746– 55. Google Scholar CrossRef Search ADS PubMed  12. Younes MTobin M. Proportional-assist ventilation. Principles and practice of mechanical ventilation. 20133rd ed New York McGraw Hill:315–49 In:. 13. M Younes Proportional assist ventilation, a new approach to ventilatory support. Theory. Am Rev Respir Dis  1992; 145: 114– 20. Google Scholar CrossRef Search ADS PubMed  14. M Younes, J Kun, B Masiowski, K Webster, D Roberts A method for noninvasive determination of inspiratory resistance during proportional assist ventilation. Am J Respir Crit Care Med  2001; 163: 829– 39. Google Scholar CrossRef Search ADS PubMed  15. N Ambrosino, A Rossi Proportional assist ventilation (PAV): a significant advance or a futile struggle between logic and practice? Thorax  2002; 57: 272– 6. Google Scholar CrossRef Search ADS PubMed  16. D Gilstrap, N MacIntyre Patient-ventilator interactions. Implications for clinical management. Am J Respir Crit Care Med  2013; 188: 1058– 68. Google Scholar CrossRef Search ADS PubMed  17. Akoumianaki E, Kondili E, Georgopoulos D. Proportional assist ventilation. European Respiratory Monograph (European Respiratory Society Journals. 2012 United Kingdom: Maney Publishing:97–115 In:. 18. N Xirouchaki, E Kondili, K Vaporidi Proportional assist ventilation with load-adjustable gain factors in critically ill patients: comparison with pressure support. Intensive Care Med  2008; 34: 2026– 34. Google Scholar CrossRef Search ADS PubMed  19. S Marantz, W Patrick, K Webster, D Roberts, L Oppenheimer, M Younes Response of ventilator-dependent patients to different levels of proportional assist. J Appl Physiol (1985)  1996; 80: 397– 403. Google Scholar PubMed  20. E Giannouli, K Webster, D Roberts, M Younes Response of ventilator-dependent patients to different levels of pressure support and proportional assist. Am J Respir Crit Care Med  1999; 159: 1716– 25. Google Scholar CrossRef Search ADS PubMed  21. KL Yang, MJ Tobin A prospective study of indexes predicting the outcome of trials of weaning from mechanical ventilation. N Engl J Med  1991; 324: 1445– 50. Google Scholar CrossRef Search ADS PubMed  22. Sinderby C, Beck JCTobin M. Neurally adjusted ventilatory assist. Principles and practice of mechanical ventilation. 20133rd ed New York McGraw Hill:351–75 In:. 23. A Demoule, B Jung, H Prodanovic Diaphragm dysfunction on admission to the intensive care unit. Prevalence, risk factors, and prognostic impact—a prospective study. Am J Respir Crit Care Med  2013; 188: 213– 9. Google Scholar CrossRef Search ADS PubMed  24. N Terzi, L Piquilloud, H Rozé Clinical review: update on neurally adjusted ventilatory assist—report of a round-table conference. Crit Care  2012; 16: 225. Google Scholar CrossRef Search ADS PubMed  25. J Beck, C Sinderby, L Lindström, A Grassino Diaphragm interference pattern EMG and compound muscle action potentials: effects of chest wall configuration. J Appl Physiol (1985)  1997; 82: 520– 30. Google Scholar PubMed  26. DK McKenzie, SC Gandevia Phrenic nerve conduction times and twitch pressures of the human diaphragm. J Appl Physiol (1985)  1985; 58: 1496– 504. Google Scholar PubMed  27. A Mier, C Brophy, J Moxham, M Green Phrenic nerve stimulation in normal subjects and in patients with diaphragmatic weakness. Thorax  1987; 42: 885– 8. Google Scholar CrossRef Search ADS PubMed  28. J Beck, SB Gottfried, P Navalesi Electrical activity of the diaphragm during pressure support ventilation in acute respiratory failure. Am J Respir Crit Care Med  2001; 164: 419– 24. Google Scholar CrossRef Search ADS PubMed  29. S Parthasarathy, A Jubran, MJ Tobin Assessment of neural inspiratory time in ventilator-supported patients. Am J Respir Crit Care Med  2000; 162: 546– 52. Google Scholar CrossRef Search ADS PubMed  30. E Calderini, M Confalonieri, PG Puccio, N Francavilla, L Stella, C Gregoretti Patient-ventilator asynchrony during noninvasive ventilation: the role of expiratory trigger. Intensive Care Med  1999; 25: 662– 7. Google Scholar CrossRef Search ADS PubMed  31. D Colombo, G Cammarota, V Bergamaschi, M De Lucia, FD Corte, P Navalesi Physiologic response to varying levels of pressure support and neurally adjusted ventilatory assist in patients with acute respiratory failure. Intensive Care Med  2008; 34: 2010– 8. Google Scholar CrossRef Search ADS PubMed  32. J Spahija, M de Marchie, M Albert Patient-ventilator interaction during pressure support ventilation and neurally adjusted ventilatory assist. Crit Care Med  2010; 38: 518– 26. Google Scholar CrossRef Search ADS PubMed  33. N Terzi, I Pelieu, L Guittet Neurally adjusted ventilatory assist in patients recovering spontaneous breathing after acute respiratory distress syndrome: physiological evaluation. Crit Care Med  2010; 38: 1830– 7. Google Scholar CrossRef Search ADS PubMed  34. C Sinderby, J Beck, J Spahija Inspiratory muscle unloading by neurally adjusted ventilatory assist during maximal inspiratory efforts in healthy subjects. Chest  2007; 131: 711– 7. Google Scholar CrossRef Search ADS PubMed  35. N Patroniti, G Bellani, E Saccavino Respiratory pattern during neurally adjusted ventilatory assist in acute respiratory failure patients. Intensive Care Med  2012; 38: 230– 9. Google Scholar CrossRef Search ADS PubMed  36. Y Coisel, G Chanques, B Jung Neurally adjusted ventilatory assist in critically ill postoperative patients: a crossover randomized study. Anesthesiology  2010; 113: 925– 35. Google Scholar CrossRef Search ADS PubMed  37. M Schmidt, A Demoule, C Cracco Neurally adjusted ventilatory assist increases respiratory variability and complexity in acute respiratory failure. Anesthesiology  2010; 112: 670– 81. Google Scholar CrossRef Search ADS PubMed  38. L Brander, H Leong-Poi, J Beck Titration and implementation of neurally adjusted ventilatory assist in critically ill patients. Chest  2009; 135: 695– 703. Google Scholar CrossRef Search ADS PubMed  39. H Rozé, A Lafrikh, V Perrier Daily titration of neurally adjusted ventilatory assist using the diaphragm electrical activity. Intensive Care Med  2011; 37: 1087– 94. Google Scholar CrossRef Search ADS PubMed  40. S Jaber, BJ Petrof, B Jung Rapidly progressive diaphragmatic weakness and injury during mechanical ventilation in humans. Am J Respir Crit Care Med  2011; 183: 364– 71. Google Scholar CrossRef Search ADS PubMed  41. AW Thille, P Rodriguez, B Cabello, F Lellouche, L Brochard Patient-ventilator asynchrony during assisted mechanical ventilation. Intensive Care Med  2006; 32: 1515– 22. Google Scholar CrossRef Search ADS PubMed  42. Patroniti N, Grasselli G, Bellani GVincent JL. NAVA: applications and limitations. Annual update in intensive care and emergency medicine 2013. 2013 New York Springer:595–608 In:. 43. L Piquilloud, L Vignaux, E Bialais Neurally adjusted ventilatory assist improves patient-ventilator interaction. Intensive Care Med  2011; 37: 263– 71. Google Scholar CrossRef Search ADS PubMed  44. L Papazian, JM Forel, A Gacouin Neuromuscular blockers in early acute respiratory distress syndrome. N Engl J Med  2010; 363: 1107– 16. Google Scholar CrossRef Search ADS PubMed  45. M de Wit, KB Miller, DA Green, HE Ostman, C Gennings, SK Epstein Ineffective triggering predicts increased duration of mechanical ventilation. Crit Care Med  2009; 37: 2740– 5. Google Scholar PubMed  46. C Sinderby, J Beck Proportional assist ventilation and neurally adjusted ventilatory assist—better approaches to patient ventilator synchrony? Clin Chest Med  2008; 29: 329– 42, vii. Google Scholar CrossRef Search ADS PubMed  47. M Wysocki, MN Fiamma, C Straus, CS Poon, T Similowski Chaotic dynamics of resting ventilatory flow in humans assessed through noise titration. Respir Physiol Neurobiol  2006; 153: 54– 65. Google Scholar CrossRef Search ADS PubMed  48. MY Bien, SS Hseu, HW Yien Breathing pattern variability: a weaning predictor in postoperative patients recovering from systemic inflammatory response syndrome. Intensive Care Med  2004; 30: 241– 7. Google Scholar CrossRef Search ADS PubMed  49. C Passath, J Takala, D Tuchscherer, SM Jakob, C Sinderby, L Brander Physiologic response to changing positive end-expiratory pressure during neurally adjusted ventilatory assist in sedated, critically ill adults. Chest  2010; 138: 578– 87. Google Scholar CrossRef Search ADS PubMed  50. J Barwing, N Linden, M Ambold, M Quintel, O Moerer Neurally adjusted ventilatory assist vs. pressure support ventilation in critically ill patients: an observational study. Acta Anaesthesiol Scand  2011; 55: 1261– 71. Google Scholar CrossRef Search ADS PubMed  51. PM Bertrand, E Futier, Y Coisel, S Matecki, S Jaber, JM Constantin Neurally adjusted ventilatory assist vs pressure support ventilation for noninvasive ventilation during acute respiratory failure: a crossover physiologic study. Chest  2013; 143: 30– 6. Google Scholar CrossRef Search ADS PubMed  52. G Vagheggini, S Mazzoleni, E Vlad Panait, P Navalesi, N Ambrosino Physiologic response to various levels of pressure support and NAVA in prolonged weaning. Respir Med  2013; 107: 1748– 54. Google Scholar CrossRef Search ADS PubMed  53. D Berger, S Bloechlinger, J Takala, C Sinderby, L Brander Heart-lung interactions during neurally adjusted ventilatory assist. Crit Care  2014; 18: 499. Google Scholar CrossRef Search ADS PubMed  54. J Cecchini, M Schmidt, A Demoule, T Similowski Increased diaphragmatic contribution to inspiratory effort during neurally adjusted ventilatory assistance versus pressure support: an electromyographic study. Anesthesiology  2014; 121: 1028– 36. Google Scholar CrossRef Search ADS PubMed  55. RM Kacmarek Proportional assist ventilation and neurally adjusted ventilatory assist. Respir Care  2011; 56: 140– 8. Google Scholar CrossRef Search ADS PubMed  Copyright © 2014 by the American Burn Association
TI - Assisted Ventilation
JF - Journal of Burn Care & Research
DO - 10.1097/BCR.0000000000000231
DA - 2016-03-01
UR - https://www.deepdyve.com/lp/oxford-university-press/assisted-ventilation-JRrh5GEe02
SP - 75
EP - 85
VL - 37
IS - 2
DP - DeepDyve
ER -