| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Guest Access | Sign In via User Name/Password
|
|
|
|||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Department of Critical Care Medicine (Drs. Sinderby and Slutsky), St. Michael’s Hospital, Department of Medicine, University of Toronto, Toronto, ON, Canada; Department of Newborn and Developmental Pediatrics (Dr. Beck), Sunnybrook Health Sciences Centre, Department of Pediatrics, University of Toronto, Toronto, ON, Canada; Sacré-Coeur Hospital Research Center (Dr. Spahija), Department of Medicine, University of Montreal, Montreal, QC, Canada; Sir Mortimer B. Davis Jewish General Hospital (Dr. Marchie), McGill University, Montreal, QC, Canada; Pediatric Intensive Care Unit, Department of Pediatrics (Dr. Lacroix), Hôpital Sainte-Justine Research Center, Université de Montréal, Montreal, QC, Canada; and Pulmonary Rehabilitation and Respiratory Intensive Care Unit (Dr. Navalesi), Fondazione S. Maugeri, Pavia, Italy.
Correspondence to: Christer Sinderby, PhD, Department of Critical Care, St. Michael’s Hospital, University of Toronto, 30 Bond Street, Queen 4–072, Toronto, ON, Canada, M5B 1W8; e-mail: sinderby{at}rogers.com
Abstract
Background: Neurally adjusted ventilatory assist (NAVA) is a mode of mechanical ventilation in which the ventilator is controlled by the electrical activity of the diaphragm (EAdi). During maximal inspirations, the pressure delivered can theoretically reach extreme levels that may cause harm to the lungs. The aims of this study were to evaluate whether NAVA could efficiently unload the respiratory muscles during maximal inspiratory efforts, and if a high level of NAVA would suppress EAdi without increasing lung-distending pressures.
Method: In awake healthy subjects (n = 9), NAVA was applied at increasing levels in a stepwise fashion during quiet breathing and maximal inspirations. EAdi and airway pressure (Paw), esophageal pressure (Pes), and gastric pressure, flow, and volume were measured.
Results: During maximal inspirations with a high NAVA level, peak Paw was 37.1 ± 11.0 cm H2O (mean ± SD). This reduced Pes deflections from – 14.2 ± 2.7 to 2.3 ± 2.3 cm H2O (p < 0.001) and EAdi to 43 ± 7% (p < 0.001), compared to maximal inspirations with no assist. At high NAVA levels, inspiratory capacity showed a modest increase of 11 ± 11% (p = 0.024).
Conclusion: In healthy subjects, NAVA can safely and efficiently unload the respiratory muscles during maximal inspiratory maneuvers, without failing to cycle-off ventilatory assist and without causing excessive lung distention. Despite maximal unloading of the diaphragm at high levels of NAVA, EAdi is still present and able to control the ventilator.
Key Words: control of breathing • diaphragm electrical activity • inspiratory capacity • mechanical ventilation • patient ventilator interaction • respiratory muscle unloading
Neurally adjusted ventilatory assist (NAVA) is a mode of mechanical ventilation in which positive pressure is applied to the airway opening in proportion to the electrical activation of the diaphragm (EAdi).1 With NAVA, ventilator support is initiated when the neural drive to the diaphragm begins to increase. As the EAdi progressively increases, the assist increases proportionally and, most importantly, the pressure delivered by the ventilator is cycled-off when the EAdi is ended by the respiratory centers. The amount of assist delivered during NAVA depends on a proportionality factor, the so-called "NAVA level," which defines the magnitude of pressure delivered for a given EAdi. When the NAVA level is changed, the resulting pressure delivered by the ventilator depends on how the respiratory afferents modulate the neural output to the diaphragm. If the response to an increase in NAVA is not a reduction in the EAdi, the delivered pressure increases. However, a reflexive or involuntary reduction in EAdi would mitigate the effects of an increased NAVA level, and the delivered pressure may remain unchanged or be less than anticipated.
Thus, the response in EAdi to an increase in the NAVA level determines the resulting transpulmonary pressure (Ptp) and whether volume changes or not. Consequently, during maximal inspirations, when the EAdi is at its highest,2 the pressure delivered could reach extreme levels that may cause harm to the lungs.3 It is therefore important to determine with increasing NAVA levels whether or not the EAdi is suppressed thereby limiting the pressure delivered during maximal inspiration. Based on the above, the aims of the present study were to determine the following: (1) whether NAVA efficiently unloads respiratory muscles throughout a maximal inspiration in healthy subjects, and (2) if EAdi is suppressed during maximal inspirations with increasing levels of NAVA.
Materials and Methods
Subjects
Nine healthy subjects (one woman) were studied. Their mean (± SD) age, height, and weight were 37 ± 8 years, 172 ± 7 cm, and 71 ± 8 kg, respectively. Two subjects had prior knowledge of mechanical ventilation. The study was approved by the Scientific and Ethical Committees of Sainte-Justine’s Hospital, Montreal, Canada, and all subjects gave their informed consent.
Measurements
Electrical signals of the diaphragm were obtained using a multiple-array esophageal electrode (nine electrodes spaced 10-mm apart). Balloons were mounted on the same catheter for measurements of esophageal pressure (Pes), gastric pressure (Pga), and transdiaphragmatic pressure (Pdi). Flow was measured with a pneumotachograph (No. 2; Hewlett Packard; Palo Alto, CA) connected to a pressure transducer (± 3 cm H2O; Ohmega Engineering; Stanford, CT). Airway pressure (Paw) was measured with a pressure sensor (± 350 cm H2O; Sensym; Milpitas, CA) and was placed with the pneumotach between the mouthpiece and the ventilator (Servo 300; Maquet Critical Care; Solna, Sweden). Respiratory inductance plethysmography (Respitrace; Ambulatory Monitoring; Ardsley, NY) was used to evaluate rib cage and abdominal displacements. All signals were acquired simultaneously, displayed on-line to the investigators, and stored for off-line analysis.
EAdi Signal Processing
Signal processing of EAdi followed American Thoracic Society recommendations.4 Filters and algorithms giving the highest possible signal-to-disturbance ratio were applied.5 Changes in diaphragm position along the array were accounted for,56 yielding a signal not artifactually affected by changes in lung volume or chest wall configuration.78 The root-mean-square was used to quantify EAdi every 16 ms.910 Signal segments with residual disturbances were replaced by the previously accepted value, resulting in a processed EAdi signal.
Method for NAVA
The processed EAdi was used to control a Servo 300 ventilator according to Sinderby et al.1 NAVA is based on transforming the EAdi amplitude into a voltage every 16 ms and sending it to the Servo 300 ventilator, which responds by adjusting the pressure level according to a linear function. The EAdi can be multiplied by a number, which essentially is a proportionality factor determining the amount of pressure is delivered for a given EAdi. This factor is referred to as the NAVA level in the present work, but has also been referred to as NAVA gain in a previous publication.1 With an increase in the NAVA level, more pressure is delivered by the ventilator if the EAdi (ie, respiratory drive) does not decrease.
In the current application, NAVA was applied during inspiration, and the assist was cycled-off to zero positive end-expiratory pressure. For triggering on, ventilatory assist was initiated when the EAdi exceeded a threshold increment in EAdi. Given that the variability of the noise level was low, the trigger threshold was set to a fixed level that permitted early detection of increasing diaphragm activation without causing autotriggering when the diaphragm was inactive. For cycling-off, ventilatory assist was terminated when the EAdi fell below a percentage (default 80%) of peak inspiratory activity.
Experimental Protocol
Subjects were studied in sitting position, breathing at rest through a mouth piece connected to the ventilator. Subjects breathed at rest for 3 to 5 min and performed at least two maximal inspirations toward the end of the period. This was subsequently repeated with increasing NAVA levels, as long as increasing NAVA levels decreased the negative Pes deflection observed on the computer monitor. In the present study, no positive end-expiratory pressure was applied.
Analysis
The start and end of each maximal inspiration were determined using the flow signal. For each tidal or maximal inspiration, EAdi signal strength was calculated as the mean inspiratory EAdi with a baseline EAdi subtracted (mean electrical activity of the diaphragm [XEAdi]). The mean pressure swings for Paw (mean Paw [XPaw]), Pes (mean Pes [XPes]), Pga (mean Pga [XPga]), Pdi, and Ptp (mean Ptp [XPtp]) were also calculated. Volume was obtained by integration of the flow signal.
In order to compare the same conditions for the different subjects, the runs were classified into three NAVA levels. Zero NAVA level refers to the condition in which the subject was breathing on the ventilator circuit without assist. High NAVA was the highest level applied (abolishing or reversing the swing in Pes), and intermediate NAVA was the NAVA level when XPaw was approximately 50% of that observed during the highest NAVA level.
Statistical Analysis
Repeated-measures analysis of variance was used to compare variables between different levels of NAVA. Post hoc comparison was performed with a Tukey test. Correlation between the mean and peak pressures was performed with Pearson product moment correlation. The statistical analyses were performed using statistical software (Sigmastat, version 2.0; Jandel Scientific; San Rafael, CA). The level of significance for all statistical tests was p < 0.05. Data are presented as mean ± SD.
Results
During both breathing at rest and maximal inspirations, NAVA was well tolerated by the subjects at all NAVA levels. In all subjects, it was possible to increase the NAVA to a level where the negative Pes deflection generated during the inspiratory capacity (IC) maneuver was abolished or reversed to positive. The group mean data obtained during the quiet breathing periods (zero and high NAVA level) are shown in Table 1 .
|
|
|
|
|
The present study demonstrates that NAVA can unload the respiratory muscles of a healthy subject to a level at which it substitutes the inspiratory muscle contribution to distend the lungs at all lung volumes. At such a high NAVA level, the subject still maintains the breathing pattern and full voluntary control of the ventilatory assist. This study also demonstrates that maximal inspirations performed with a high level of NAVA suppress the diaphragm electrical activity, thereby limiting increases in lung-distending pressures and volumes. No problems to cycle-off the assist were observed with increasing NAVA levels.
Unloading With NAVA
Since NAVA is controlled by the EAdi, the subject can control the delivered pressure when NAVA is increased by adjusting the neural output from the respiratory centers. Therefore, a subject should be able to either do the following: (1) keep the same EAdi, allowing the increased NAVA level to directly translate into increased Paw; or (2) maintain Paw unchanged by counteracting the increased NAVA level by reducing the EAdi. In the present study, the subjects demonstrated an intermediate response, where increasing the NAVA level both increased Paw and reduced EAdi, which in turn resulted in decreased Pes and decreased Pdi. This agrees well with a previous study11 suggesting that respiratory assist relieves the inspiratory muscle workload, allowing respiratory drive to be reduced.
An interesting finding of the present study was that the changes in XPaw and XPes were equal, keeping the XPtp unchanged (Fig 2). In other words, with each increased level of NAVA, the ventilator pressure delivery increased to substitute the inspiratory pressure generated by the subject without altering the XPtp during tidal and maximal inspirations. The same relationship (maintained XPtp with increasing NAVA levels) has previously demonstrated in tracheotomized, anesthetized, and spontaneously breathing animals with acute lung injury.12 This strongly supports a close interaction between NAVA, the neural respiratory feedback mechanisms, and the EAdi output.
Another interesting finding was that the EAdi, although suppressed, ensured control of ventilatory assist during both quiet breathing and maximal inspirations, even at the highest NAVA levels. In some subjects (Fig 1, 4), high NAVA resulted in a positive Pes deflection large enough to eliminate Pdi, suggesting that the diaphragm was nearly 100% unloaded, while EAdi was still clearly present. In 1962, Agostoni13 showed that increasing "positive pressure breathing" produced a progressive reduction in Pdi despite small changes in diaphragm electrical activity, and that pressures of approximately 30 cm H2O could eliminate Pdi during quiet breathing while diaphragm electrical activity remained.
Termination of Maximal Inspiration
Despite 55 years having passed since Mills14 published the first article describing what limits the effort and depth of a voluntary maximal inspiration in a human, there is no consensus on this topic. Although this topic may be of academic relevance only in situations when maximal inspirations are performed without a ventilator, the introduction of a new generation of ventilators that deliver assist in proportion to the patient’s effort revives the question of whether diaphragm activity is downregulated with increasing levels of assist to prevent excessive lung-distending pressures during a maximal inspiration. More importantly, considering the current findings that lower tidal volumes may reduce ventilator-induced lung injury,15 knowledge about the possible presence/activity of lung-protective reflexes during neurally controlled mechanical ventilation is important.
Mills14 demonstrated an antagonistic abdominal muscle activation observed at the end of maximal inspiration that was accompanied by a closing of the glottis. Campbell and Green16 showed that voluntary and unassisted maximal inspirations are terminated in association with increased electrical activation of abdominal muscles and a sharp rise in abdominal pressure. Both studies suggest that reflex inhibition terminates maximal inspirations in healthy subjects. Later, Agostoni and Rahn17 demonstrated that active antagonistic abdominal muscle activation at maximal inspiration counterbalances the diaphragm activity and limits further expansion of the lung, which could protect the lung without involving reflexes. Mead et al18 demonstrated that with practice, both antagonistic abdominal muscle activation and closing of the glottis could be avoided at the end of maximal inspiration, and suggested that the maximum volume is determined by a balance arising from the elastic recoil of the respiratory system and the diminishing effectiveness of the inspiratory muscles at high lung volumes. In support of this finding, our group has previously demonstrated that the maximal voluntary activation of the diaphragm was similar during combined maximal inspiratory efforts against occluded airways at functional residual capacity and at total lung capacity in healthy subjects and in patients with diaphragm weakness,2 and that the pressure-generating capacity at total lung capacity decreases to approximately 40 cm H2O,8 which is comparable to the Paw required to inflate the lungs in the present study.
Regardless of the mechanism, the present study shows that application of a high NAVA level during maximal inspiration strongly suppresses XEAdi, limiting the XPtp, and increases IC modestly in healthy subjects. These findings along with our previous work12 in tracheotomized lung-injured rabbits support the possible role of lung-protective reflexes during neurally controlled mechanical ventilation. However, it should be noted that healthy subjects were evaluated in the present study, and therefore the reflex effects may be different from those of patients with respiratory failure.
Critique of the Study
The present study could be criticized for several reasons. First, the present study was performed in healthy, awake subjects, which today is not conventional for studies on control of breathing. However, awake subjects were a requirement since the main intervention was maximal voluntary inspirations.
Second, while breathing at rest, the tidal volume and minute ventilation were greater than anticipated. This was likely due to the increased dead space of the tubing, connectors, pneumotachograph, and the mouthpiece. However, since tidal volume and minute ventilation did not change between zero and high NAVA levels while breathing at rest, we do not expect that this offset of the respiratory drive would effect mechanical unloading with NAVA.
Third, for safety reasons, we used a nonsealing mouthpiece. Thus, we cannot exclude that the slight but significant increase in IC at high NAVA levels may have been caused by leaks around the mouthpiece or nose clip.
Conclusion
NAVA can efficiently unload the respiratory muscles at all lung volumes. Diaphragm electrical activity is downregulated with increasing NAVA level, limiting lung distension during maximal inspirations in healthy subjects. The findings of the present study suggest that NAVA is well integrated with respiratory control systems and provides assist in response to central respiratory output. Despite maximal unloading of the diaphragm at high levels of NAVA, EAdi is still present and able to control the ventilator.
Footnotes
Abbreviations: EAdi = electrical activity of the diaphragm; IC = inspiratory capacity; NAVA = neurally adjusted ventilatory assist; Paw = airway pressure; Pdi = transdiaphragmatic pressure; Pes = esophageal pressure; Pga = gastric pressure; Ptp = transpulmonary pressure; XEAdi = mean electrical activity of the diaphragm; XPaw = mean airway pressure; XPes = mean esophageal pressure; XPga = mean gastric pressure; XPtp = mean transpulmonary pressure
The work was performed at Hôpital Sainte-Justine in Montreal, Quebec, Canada.
This work was supported in part by the Canadian Intensive Care Foundation. Dr. Beck and Dr. Sinderby were supported by the Fonds de la Recherche en Santé du Québec.
This disclosure statement has been approved by St-Michael’s Hospital, Sunnybrook Health Sciences Centre, and University of Toronto. Dr. Beck and Dr. Sinderby have made inventions related to neural control of mechanical ventilation that are patented. The license for these patents belongs to Maquet Critical Care. Future commercial uses of this technology may provide financial benefit to Dr. Sinderby and Dr. Beck through royalties. Dr. Sinderby and Dr. Beck each own 50% of Neurovent Research Inc. Neurovent Research is a research and development company that builds the equipment and catheters for research studies. Neurovent Research has a consulting agreement with Maquet Critical Care. Dr. Slutsky consults for companies that make ventilators, specifically Maquet Critical Care and Hamilton Medical and is compensated for these consultations. Drs. Spahija, de Marchie, Lacroix, and Navalesi have no conflicts of interest to disclose.
Received for publication August 1, 2006. Accepted for publication September 8, 2006.
References
Related Editorial
This article has been cited by other articles:
|
|
 |
|
  S. H. Loring and A. Malhotra Inspiratory Efforts During Mechanical Ventilation: Is There Risk of Barotrauma? Chest, March 1, 2007; 131(3): 646 - 648. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
