help button home button
AJRCCM
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Published ahead of print on May 28, 2003, doi:10.1164/rccm.200203-241OC
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
200203-241OCv1
168/7/760    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Sharshar, T.
Right arrow Articles by Lofaso, F.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Sharshar, T.
Right arrow Articles by Lofaso, F.
American Journal of Respiratory and Critical Care Medicine Vol 168. pp. 760-769, (2003)
© 2003 American Thoracic Society

Original Article

Transdiaphragmatic Pressure Control of Airway Pressure Support in Healthy Subjects

Tarek Sharshar, Gilbert Desmarais, Bruno Louis, Gilles Macadou, Raphaël Porcher, Alain Harf, Jean-Claude Raphaël, Daniel Isabey and Frédéric Lofaso

Services d'Explorations Fonctionnelles, de Réanimation Médicale et Centre d'Innovations Technologiques, Hôpital Raymond Poincaré, Garches; Ecole Supérieure d'Ingénieurs en Electrotechnique et Electronique, Noisy le Grand; Département de Biostatistique et Informatique Médicale, Hôpital Saint-Louis, Paris; Service de Physiologie, Explorations Fonctionnelles, Hôpital Henri Mondor et Institut National de la Santé et de la Recherche Médicale, Faculté de Médecine, Créteil, France

Correspondence and requests for reprints should be addressed to Frédéric Lofaso, M.D., Ph.D., Service d'Explorations Fonctionnelles, Hôpital Raymond Poincaré, 104, boulevard Raymond Poincaré, 92380 Garches, France. E-mail: f.lofaso{at}rpc.ap-hop-paris.fr


   ABSTRACT
TOP
 ABSTRACT
METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
We designed a new servoventilator that proportionally adjusts airway pressure to transdiaphragmatic pressure (Pdi) generated by the subject during inspiration. Each cycle is triggered by either a preset Pdi increase or a preset inspiratory flow value (whichever is reached first), whereas cycling-off is flow-dependent. We evaluated the servoventilator in seven healthy subjects at normocapnia and three levels of hypercapnia (normocapnia + 3, + 6, and + 9 mm Hg) comparatively with spontaneous breathing. Triggering was by Pdi in six subjects and flow in one. At all end-tidal carbon dioxide pressure levels, time from onset of diaphragm electromyographic activity to inspiratory flow was similar with and without the servoventilator. Airway pressure increased proportionally to Pdi variation during servoventilator breathing. Flow, tidal volume, respiratory rate, intrinsic positive end-expiratory pressure, and esophageal and transdiaphragmatic pressure–time products increased significantly with hypercapnia with and without the servoventilator. Breathing pattern parameters were similar in the two breathing modes, and no differences were found for intrinsic positive end-expiratory pressure or gastric pressure variation during exhalation. Esophageal and transdiaphragmatic pressure–time products were lower with than without the servoventilator. The Pdi-driven servoventilator was well synchronized to the subjects effort, delivering a pressure proportional to Pdi and reducing respiratory effort at normocapnia and hypercapnia.

Key Words: triggering • servoventilator • respiratory drive

Interaction between the patient and ventilator is a major problem in the management of patients who require mechanical ventilation (1, 2). Patient–ventilator asynchrony can occur at various points in the respiratory cycle, causing discomfort, impairing gas exchange and cardiovascular function, and increasing the risk of barotrauma. To improve patient–ventilator synchrony, two approaches can be used. One consists of using sedation or muscle paralysis, but this increases the risk of complications related to prolonged mechanical ventilation, drug-related side effects, and/or respiratory muscle wasting. The second approach involves using assisted modes of mechanical ventilation, in which the patient triggers the machine with each breath. Various assisted modes have been developed. Modes using airway pressure, flow, and/or volume signals to trigger and control the inspiratory support have been extensively studied. However, these signals may fail to control the ventilator effectively, especially in patients with respiratory dysfunction, additional circuitry resistance, or air leaks (1). Alternatively, ventilatory support can be controlled by parameters reflecting either respiratory center neural output, such as phrenic nerve activity in animals (3) and diaphragmatic electrical activity in humans (1), or the diaphragmatic mechanical output, such as transdiaphragmatic pressure (Pdi).

We developed a new type of servoventilator that continuously adjusts the delivered airway pressure to a fixed proportion of the Pdi increase induced by inspiration. In the present study, we examined seven healthy individuals at three different levels of increased breathing demand induced by hypercapnia. Our goal was to assess the efficiency of our Pdi-driven servoventilator and its effect on diaphragmatic mechanical output comparatively with hypercapnia-matched spontaneous breathing.


   METHODS
TOP
 ABSTRACT
 METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
We have developed a device based on a new concept for controlling mechanical ventilatory assistance, in which both detection of inspiratory effort and pressurization rely upon online detection of variations in Pdi. Expiratory triggering is unchanged as compared to conventional flow-triggered pressure-support devices (4, 5).

Device Characteristics
We modified the ARM25 (Taema, Antony, France), a ventilator designed for use in intensive care units and effective in treating acute respiratory failure (4, 5). The modification ensured that pressure generation was driven by changes in Pdi. Because the ARM25 inspiratory circuit is completely open to the atmosphere (Figure 1) , the inspiratory load imposed on the patient before inspiratory triggering is negligible: a pressure of only 1/2 mm H2O generated by the patient at an inspiratory flow below 0.017 L/second (i.e., 1 L/minute inspiratory threshold) overcomes the 3.5 cm H2O/L/second resistance of the inspiratory line (including the one-way valve and all other elements present on this line). A confined jet of compressed air pressurizes the inspiratory airflow. Recirculation eddies in the immediate entrance zone of the inspiratory tube between the high-velocity jet and entrained flow (of air or CO2-enriched air in our study) (Figure 1) generate an adverse-to-flow longitudinal pressure gradient (6). The system works like a fluidic piston capable of pressurizing flow in the inspiratory line with minimal pressure loss because of the very low line resistance. Injection of compressed air (or CO2-enriched air) through an inspiratory circuit open to the atmosphere generates positive pressure in the circuit because of air (or CO2-enriched air) entrainment at ambient pressure. Because the patient's demand in terms of inspiratory flow modifies the ratio of jet to entrained flow and, therefore, the performance of the ventilator in terms of generated airway pressure, a computer-assisted servo-controlled system is used (Figure 1). A servo valve adjusts the jet flow to a value that minimizes the difference between airway pressure and the reference pressure, which is proportional to Pdi variations. Tidal volume is consequently dependent on both the subject's inspiratory activity and the ventilator's performance. Expiration occurs through a Bennett TT 11 372–00 expiratory valve (Tyco Healthcare/Mallinekrodt, St. Louis, MO), which is one of the lowest-resistance expiratory valves available (7).



View larger version (32K):
[in this window]
[in a new window]
  		Figure 1. Diagram of the ventilator driven by transdiaphragmatic pressure (Pdi). The ARM25 ventilator was modified to deliver an airway pressure proportional to the Pdi generated by the subject. The subject breathed in through an inspiratory circuit open to the atmosphere. Detection of inspiratory effort was based on two alternative modes of triggering: detection of inspiratory flow signal (flow-triggering) or detection of second-order filtered Pdi signal (filtered Pdi triggering). Sensitivity was 1 L/minute for flow triggering and ranged from 0.5 to 5.5 cm H2O for Pdi triggering. During inspiration, esophageal and gastric pressures were continuously computed, and Pdi was calculated by subtracting Pes from Pga. Airway pressure (Paw) generated by the servoventilator was calculated in real time using the following equation: Paw = P0 + K · {Delta}Pdi, where P0 is the Pdi value at triggering and K is a dimensionless factor ranging from 0.2 to 1.0. Paw was controlled and regulated by means of a proportional servovalve (PSV). Finally, the subjects exhaled through a Bennett TT 11 372–00 expiratory valve, which opened when inspiratory flow fell below 0.05 L/second.

 
Detection of the Subjects Inspiratory Effort
To improve servoventilator (ARM25) sensitivity, detection of the inspiratory effort was based on two criteria: a standard criterion and an improved criterion. Inspiration was triggered by whichever criterion was met first.

The standard detection criterion, or flow trigger, was based on the inspiratory flow signal. Flow was sensed by a hot wire–calibrated pneumotachograph placed on the inspiratory limb of the circuit. This flow detection system is extremely sensitive, especially at low-flow levels, because a heated wire responds to gas convection with an infinite slope and thus can ensure early detection of inspiratory effort in patients with chronic obstructive pulmonary disease (4, 5). The system was set so that pressurization started when inspiratory flow reached 0.017 L/second (i.e., 1 L/minute). The time lag for effective pressurization to occur was related to the performance of the device and was about 30 milliseconds.

The improved detection criterion, or Pdi trigger, was based on the second-order filtered Pdi signal. Pressurization was triggered when a predefined value of Pdi increase ({Delta}Pdi) was reached. This value could be adjusted between 0.5 and 5.5 cm H2O, depending on the noise-to-signal ratio associated with cardiac artifacts.

To adjust our program file, before building the servoventilator we conducted a study in spontaneously breathing subjects to predict whether the Pdi trigger or the flow trigger would be reached first. As shown in Figure 2 , online detection occurred earlier with Pdi triggering than with flow triggering when the Pdi detection threshold was set at 1 cm H2O. The time to effective pressurization was similar with Pdi triggering and with flow triggering, i.e., about 30 milliseconds. Consequently, the detection criterion was the only difference between the two triggering modes.



View larger version (36K):
[in this window]
[in a new window]
  		Figure 2. Difference in inspiratory onset detection using the filtered transdiaphragmatic pressure (Pdi) signal and the flow signal. The sensitivity of the filtered Pdi threshold and flow threshold were set at 1 cm H2O and 0.017 L/second, respectively. Detection was obtained online. In the figure, detection is shown as a transient increase in the squared signal for both detection modes. Detection of inspiration occurred earlier with the filtered Pdi threshold than with the flow threshold. Flow is expressed in L/minute and esophageal pressure, gastric pressure, and Pdi in cm H2O. Pdi = transdiaphragmatic pressure; Pes = esophageal pressure; Pga = gastric pressure.

 
Partial Compensation of Inspiratory Effort Based on Transdiaphragmatic Pressure Increase
Pdi was obtained by simultaneously measuring esophageal pressure (Pes) and gastric pressure (Pga) using two transducers mounted on a catheter and positioned on either side of the diaphragm. Pdi was calculated as the difference between Pga and Pes (Pdi = Pga - Pes). Note that Pdi is less sensitive than Pes to cardiac artifacts and does not comprise the pressure component related to thoracic compliance included in Pes. It is very important to note that airway pressure was regulated using the following Pdi function: Paw = P0 + K · {Delta}Pdi, where Paw is airway pressure, P0 is equal to 2 cm H2O, K is a dimensionless factor in the 0.2–1.0 range, and {Delta}Pdi is the Pdi increase.

To ensure a satisfactorily dynamic response for the entire system, we used loop regulation by state feedback, which included a model of the dynamic response of the servo valve evaluated in preliminary experiments.

Subjects
Seven healthy volunteers (six men and one woman; mean age [± SD], 39.7 ± 13.9 years; weight, 72.2 ± 4.8 kg; height, 176.0 ± 5.7 cm) were studied. All volunteers gave their informed consent.

Measurements
Each subject was seated comfortably, wearing a noseclip and breathing via a mouthpiece. Flow was measured by a Fleisch #2 pneumotachograph (Fleisch, Lausanne, Switzerland) connected to a differential pressure transducer (Validyne, Northridge, CA), airway pressure by a differential pressure transducer (Validyne), end-tidal carbon dioxide pressure (PETCO2) by an infrared gas analyzer (Gould), Pes and Pga by a catheter-mounted transducer (Gaeltec, Dunvegan, UK) (12), and diaphragm electromyographic activity by a Disa 13K63 bipolar esophageal electrode (Disa, Copenhagen, Denmark). The validity of Pes and Pga measurements was checked by analyzing the shape of the Pes curve after the patient drank water, by the occlusion technique, and by pressing on the stomach (8, 9). The electromyography catheter was glued to the catheter-mounted transducer used to measure Pes and Pga. The best position of the bipolar electrode in the esophagus was meticulously assessed based on the following criteria: (1) ECG artifacts displayed; (2) electromyographic signal displayed during relaxed breathing; (3) decrease or disappearance of electromyographic activity when the electrode was moved 1 and 2 cm upward and downward; and (4) increase in the electromyographic signal during an inspiratory effort. The esophageal electrodes were taped to the nose and their position marked on the skin and checked regularly throughout the study. The interelectrode distance was 2.2 cm.

All signals were digitized at 128 Hz using an analog–digital system (MP100, Biopac System). The diaphragm electromyographic activity was digitized at 2 000 Hz.

Experimental Setup
During spontaneous breathing, subjects breathed in a Hans Rudolph two-way nonrebreathing valve (#7200; Kansas City, MO). The inhalation circuit was connected to a 100-L bag containing CO2 and O2, which were mixed in a gas mixer. The exhalation circuit was open to the atmosphere. During servoventilator breathing, CO2 and O2 cylinders were connected to a gas mixer, which in turn was connected to the servoventilator (Arm 25; Taema, Antony, France).

Ventilator Adjustment
Inspiration could be triggered either by inspiratory flow, the threshold being set at the lowest flow detectable by our machine (i.e., 0.017 L/second), or by a Pdi increase, set at the lowest value not associated with autotriggering. The flow trigger and the Pdi trigger were both enabled and in competition with each other: the servoventilator was triggered by whichever trigger was activated first. Thus, whether Pdi or flow triggered the servoventilator depended on the subject's breathing pattern and on the sensitivity of the Pdi trigger.

K was adjusted based on the subjects comfort. In each subject, K was increased gradually until the subject reported discomfort. K was then set at the immediately preceding value. Discomfort was estimated subjectively by the subject, and hyperinflation was the main complaint.

Whatever the criterion used to detect inspiration, the end of the inspiratory period (and therefore the end of inspiratory pressure support) was detected based on the flow signal, with the threshold set at 0.05 L/second.

Experimental Protocol
Each subject was studied during spontaneous breathing and servoventilator breathing in random order. Each session comprised four test periods. The first period involved testing under normocapnic conditions (i.e., PETCO2 during spontaneous breathing). PETCO2 values from this period were used as the reference values. During servoventilator breathing, if hyperventilation occurred the inhaled CO2 concentration was increased to maintain normocapnia. In the second, third, and fourth periods the inhaled CO2 concentration was increased stepwise in order to increase the PETCO2 value by 3, 6, and 9 mm Hg from baseline, respectively. The time between two consecutive spontaneous breathing and servoventilator breathing sessions was at least 15 minutes of spontaneous breathing in room air.

Each period lasted at least 12 minutes. The first part of the period was used to document PETCO2 stability and lasted at least 5 minutes, as PETCO2 stability was defined as less than 1 mm Hg variation over 5 minutes. Data were recorded from the end of the fifth minute to the end of the period. Each spontaneous breathing or servoventilator breathing session lasted 1 hour.

Data Analysis
For each period, the variables were analyzed and averaged over 2 consecutive minutes after the tenth minute of each period. We considered that the PCO2 value during the previous period had only a negligible influence on respiratory motor output after the tenth minute.

As previously described (9), the following variables were measured breath-by-breath: time from EMG activity onset to flow onset; Pdi increase before the airway pressure nadir (Pdi triggering); inspiratory time (Ti), expiratory time (Te), and total time (Ttot = Ti + Te); VT; respiratory rate (RR = 1/Ttot); total ventilation (E = VT ·RR); PETCO2; peak airway pressure; {Delta}Pes and {Delta}Pga during inspiration; time from the Pdi peak to the airway pressure peak; Pga increase during exhalation (10); intrinsic positive end-expiratory pressure (iPEEP); and airway, esophageal, and transdiaphragmatic pressure–time products (PTPs) (cm H2O · second · minute-1). To determine whether this regulation was effective, we assessed the relationship between the measured airway PTP and the theoretical airway PTP, which was derived from the transdiaphragmatic PTP generated by the subject during the inspiratory time (theoretical airway PTP = 2 · Ti + K · transdiaphragmatic PTP). We used the airway PTP because this variable reflected the variation in airway pressure throughout the inspiratory time.

Statistical Analysis
Results are presented as means ± SEM. Measured parameters were evaluated using mixed-models analysis of variance. Main effects were the use of Pdi-driven servoventilator (thus comparing spontaneous breathing versus servoventilator breathing) and the level of PETCO2. Interactions between the two factors were added to the model and subsequently removed if not significant. In addition, the subject was added to the model as a random effect and different correlation structures for residuals were tested. The best structure was based on the modified Akaike criterion (11). Post hoc two-by-two comparisons were performed using Tukey's correction for multiple testing. The significance level was set at 5%, and all tests were two-sided. Agreement between two variables was assessed, according to the Bland and Altman test by plotting their mean against their difference.


   RESULTS
TOP
 ABSTRACT
 METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Efficiency of the Pdi-driven Servoventilator
The Pdi trigger was set above 2 cm H2O in one subject because of cardiac artifacts responsible for autotriggering at lower trigger values. In this subject, the servoventilator was always triggered by the inspiratory flow signal. In the other six subjects, the Pdi trigger was set at a value in the 0.7–1.5 cm H2O range (mean, 1.2 ± 0.5; median, 1.2) and the servoventilator was always triggered by the Pdi signal. Thus, the flow/Pdi trigger ratio was nil in these six subjects. When esophageal spasms were present, airway pressure regulation was abolished, both because Pdi did not increase and because the electric signal was saturated, but this affected only a few isolated cycles. In this rarely observed situation, the servoventilator was triggered by the flow signal and delivered only a constant minimal pressure, which was set at 2 cm H2O. However, esophageal spasms did not occur with sufficient frequency to affect Pdi triggering. In addition, in each CO2 condition, Pdi triggering occurred at a value similar to the preset value (Tables 1 and 2 and Figure 3) . The mean measured Pdi increase that triggered the ventilator was greater than the individual preset values in four subjects (1, 2, 3 and 6). This difference was due to the time needed for pressurization. During this time, Pdi continued to increase as a result of the subjects inspiratory effort. In two subjects (4 and 5), the mean measured {Delta}Pdi trigger was less than the individual preset values as a result of cardiac artifacts. Nevertheless, the agreement between preset and measured Pdi trigger values remained satisfactory. Increasing PETCO2 did not significantly change the measured Pdi trigger value (Table 2).


View this table:
[in this window]
[in a new window]
  		TABLE 1. Individual measured and preset values of the transdiaphragmatic pressure change used to trigger the servoventilator

 

View this table:
[in this window]
[in a new window]
  		TABLE 2. Servoventilator performances according to increase in pETco2 during servoventilator breathing

 


View larger version (12K):
[in this window]
[in a new window]
  		Figure 3. Representation of agreement between measured and preset change in transdiaphragmatic pressure ({Delta}Pdi). For each subject, the mean values for measured and preset {Delta}Pdi were plotted against the difference between measured and preset {Delta}Pdi. The mean, mean+1.96 SD, and mean -1.96 SD are also represented. Preset {Delta}Pdi trigger values were 0.7 in Subjects 1 and 2, 1.0 in Subject 3, 1.5 in Subjects 4 and 5, and 1.7 in Subject 6. Measured {Delta}Pdi trigger values correspond to the Pdi increase before the airway pressure nadir and are expressed in cm H2O. Mean measured {Delta}Pdi trigger values were calculated by averaging, for each PETCO2 level, the {Delta}Pdi trigger values of 10 consecutive cycles. The agreement between preset and measured {Delta}Pdi trigger values was satisfactory.

 
Increasing PETCO2 was associated with a statistically insignificant decrease in time from diaphragmatic EMG activity onset to inspiratory flow during servoventilator breathing (Table 2). Interestingly, in two subjects, inspiratory flow onset was initiated by the servoventilator and not by the subject (for example see Figure 4) . In the four other subjects in whom Pdi triggering occurred, the time lag from triggering to effective pressurization, which was about 30 milliseconds (see METHODS section) allowed the subjects to initiate the inspiratory flow themselves, despite earlier Pdi triggering.



View larger version (29K):
[in this window]
[in a new window]
  		Figure 4. Effect of intrinsic positive end-expiratory pressure (iPEEP). Flow, esophageal, gastric, transdiaphragmatic, and airway pressure signals measured during servoventilator breathing in one subject are displayed. In this subject, airway pressure was slightly positive at the end of expiration, indicating iPEEP. When iPEEP is present, the airway pressure nadir occurs before inspiratory flow onset. This was observed in two subjects. Flow is expressed in L/minute and esophageal, gastric, transdiaphragmatic, and airway pressure values in cm H2O. Servovalve power supply, expressed in arbitrary units, is the voltage used to feed the servo-valve mechanism (an electromagnet) located inside the servoventilator. The transient increase in this voltage signal was used to detect the beginning of pressure assistance, hence the sudden increase in airway pressure and the reversal of flow.

 
K ranged from 0.5 to 1.0 (mean, 0.7 ± 0.2). The inspiratory airway pressure peak during air breathing was between 7 and 10 cm H2O and increased significantly with PETCO2 during servoventilator breathing (Table 2). Airway pressure regulation by the servoventilator was effective, as suggested by the good agreement between measured airway PTP and theoretical airway PTP (Figure 5) . Interestingly, linear regression showed that the slope of measured versus theoretical airway PTP was close to the corresponding K value (data not shown).



View larger version (15K):
[in this window]
[in a new window]
  		Figure 5. Representation of agreement between measured and theoretical airway pressure–time product values. For each subject, the mean measured and theoretical airway pressure–time products were plotted against the difference between the measured and preset changes in transdiaphragmatic pressure ({Delta}Pdi). The mean, mean +1.96 SD, and mean -1.96 SD are also represented. The theoretical airway pressure–time product was calculated from Ti, K, and the diaphragmatic pressure–time product (PTPaw): theoretical PTPaw = 2 · Ti + K · PTPdi. K was the preset factor that coupled Paw with Pdi (Paw = P0 + K · Pdi). The value of K was 0.2 in Subject 5; 0.4 in Subjects 4 and 7; 0.5 in Subjects 2, 3, and 6; and 1 in Subject 1. For each subject, the slope of measured versus theoretical airway pressure–time products was close to the corresponding K value, supporting efficacy of the Pdi-driven servoventilator. Airway pressure–time product is expressed in cm H2O/second/minute.

 
During servoventilator breathing, the mean time from peak Pdi to peak airway pressure varied from 200 and 300 milliseconds and did not change significantly with PETCO2 (Table 2).

Effect of the Servoventilator on Breathing Pattern and Respiratory Effort Indices
Figure 6 shows examples of spontaneous breathing and servoventilator breathing recordings obtained during the third period. In three subjects, spontaneous breathing was the first session. Analysis of variance showed that E, VT, RR, Ti, Te, Ti/Ttot, VT/Ti, and iPEEP were not significantly different between spontaneous breathing and servoventilator breathing (Table 3) . Mean time from EMG activity onset to inspiratory flow onset was longer during servoventilator breathing than during spontaneous breathing, but the difference was not statistically significant (Table 3).



View larger version (33K):
[in this window]
[in a new window]
  		Figure 6. Trace showing PETCO2, flow, airway pressure, transdiaphragmatic pressure, and diaphragmatic electromyogram (EMG) activity during spontaneous breathing (SB) and servoventilator breathing (SVB; K = 0.4) in a representative subject exposed to hypercapnia (normocapnia + 6 mm Hg). Note that, as compared to the control situation, EMG activity and transdiaphragmatic swings decreased during SVB.

 

View this table:
[in this window]
[in a new window]
  		TABLE 3. Mean values of breathing variables obtained at each end-tidal carbon dioxide pressure level during spontaneous breathing and servoventilator breathing

 
{Delta}Pes and {Delta}Pga were not significantly different between spontaneous breathing and servoventilator breathing. The Pga increase during exhalation was not significantly different between spontaneous breathing and servoventilator breathing. The esophageal and transdiaphragmatic PTPs (Figure 7) were lower during servoventilator breathing than during spontaneous breathing.



View larger version (62K):
[in this window]
[in a new window]
  		Figure 7. Effect of servoventilator breathing on respiratory indices. Means ± SEM of transdiaphragmatic and esophageal pressure–time products (PTPdi and PTPes) during spontaneous breathing (SB, light gray bars) and servoventilator breathing (SVB, dark gray bars) at normocapnia and three levels of hypercapnia (normocapnia + 3, 6, and 9 mmHg). PTPdi, and PTPes increased significantly with PETCO2 (p < 0.0001). As compared to SB, PTPdi and PTPes were lower during SVB (p < 0.009; p < 0.005, respectively). Interaction was significant only with PTPes (p = 0.04). Post hoc tests showed that PTPdi and PTPes in the "normocapnic + 9 mm Hg" condition were significantly different from the values in the other condition.

 
Effect of PETCO2 on Breathing Pattern and on Respiratory Effort Indices
Table 3 shows the mean respiratory variable values measured at each PETCO2 level during spontaneous breathing and servoventilator breathing. Analysis of variance showed that E, VT, RR, VT/Ti, and iPEEP increased significantly with PETCO2 during both spontaneous breathing and servoventilator breathing. In contrast, Ti and mean time from EMG activity onset to inspiratory flow onset did not change significantly, and Te decreased significantly as PETCO2 increased.

During inspiration, {Delta}Pes and {Delta}Pga increased significantly with PETCO2, during both spontaneous breathing and servoventilator breathing (Table 3). The Pga increase during exhalation did not change significantly with PETCO2 during spontaneous breathing or servoventilator breathing. The esophageal and transdiaphragmatic PTPs increased significantly with PETCO2 (Figure 7) during both spontaneous breathing and servoventilator breathing.


   DISCUSSION
TOP
 ABSTRACT
 METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
The objective of the present study was to assess the efficiency of a new servoventilator in seven healthy volunteers at various levels of hypercapnia. Our servoventilator has two original features: pressurization can be triggered by a Pdi increase and, after triggering, airway pressure is delivered in proportion to the Pdi generated by the subject. Our main findings were as follows: (1) at all tested levels of hypercapnia, Pdi triggering avoided both untimely triggering and intense or prolonged triggering efforts; (2) airway pressure delivery was satisfactory and proportional to Pdi; and (3) the servoventilator decreased the work of breathing, as measured by the transdiaphragmatic PTP.

Methodological Issues
Our study raises a number of methodological issues. First, this study provides a technical description of our Pdi-driven servoventilator and a physiologic description of its effects on breathing pattern and diaphragmatic mechanical output at various levels of respiratory demand. We did not compare our Pdi-driven servoventilator to other ventilators. Therefore, our results cannot be extended to patients, and neither can our Pdi-driven servoventilator be considered an alternative to available mechanical ventilation modes. Additional studies would be needed to evaluate the effects of our servoventilator on patients and to compare them to the effects of other ventilators.

Second, it can be argued that PETCO2 is not an accurate measurement of arterial CO2 and that the observed differences in transdiaphragmatic PTP values between spontaneous breathing and servoventilator breathing were perhaps related to differences in arterial CO2 that escaped detection by PETCO2 measurement. For ethical reasons, we did not directly measure arterial CO2 and were consequently unable to assess possible differences between arterial CO2 and PETCO2. However, it has been reported that arterial CO2 and PETCO2 are correlated with each other during both spontaneous breathing and mechanical ventilation (12, 13) and that changes in PETCO2 consistently indicate changes in arterial CO2 (14).

Third, there is some evidence that arterial CO2 oscillations may contribute to respiratory center activation (15). Arterial CO2 oscillations were not taken into account in the present study. It should be noted, however, that removal of arterial CO2 oscillations at rest had no effect on central respiratory output in earlier studies (16, 17).

Trigger Characteristics
Our device was designed to let the patients inspire freely before triggering. This was possible because the inspiratory circuit was open to the atmosphere, so that the resistance of the inspiratory line was negligible, particularly below the flow trigger threshold (i.e., 0.1 L/second). Thus, the load imposed on the subject before triggering was negligible. Our Pdi-driven ventilator had two inspiratory triggering modes, one based on flow and the other on Pdi increase. Flow is one of the most efficient triggers (18) and has been widely used for the noninvasive treatment of acute respiratory failure (4, 5). By contrast, Pdi triggering has not been used previously. The present study does not provide a comprehensive statistical comparison of flow triggering and Pdi triggering. Therefore, our data do not constitute proof that one triggering mode is more efficient than the other. However, several arguments support the effectiveness of Pdi triggering. First, it has been shown that both Pes and Pdi are better indices than flow signal for detecting the onset of inspiratory effort (19). There is a time lag between the Pdi change and the onset of inspiratory flow, which corresponds to a period of isometric contraction. This prompted us to compare the sensitivity of flow detection to Pdi detection methods before building our servoventilator. We found that inspiratory effort onset was detected earlier by the Pdi signal than by the flow signal, as shown in Figure 2. Second, during servoventilator breathing, both flow triggering and Pdi triggering were enabled, yet cycles were triggered by the Pdi change in six of seven subjects. This was true at all tested PETCO2 levels, suggesting that Pdi triggering remained efficient when respiratory demand was high. Third, Pdi triggering was not associated with asynchronous cycles, ineffective triggering, or a need for strong triggering efforts. However, use of Pdi triggering may be limited by various factors, including esophageal spasms and cardiac artifacts. Esophageal spasms were sufficiently uncommon in our subjects that they did not affect Pdi triggering. In one subject, however, auto-triggering related to cardiac artifacts compelled us to set the Pdi trigger above 2 cm H2O, which resulted in flow triggering of the servoventilator. However, Pdi triggering may occur even with a Pdi trigger greater than 2 cm H2O because iPEEP is common in patients who need mechanical ventilation and decreases the efficiency of inspiratory flow in detecting the onset of inspiratory effort. In this situation, a high level of Pdi trigger might be better than a flow trigger or an airway pressure trigger. Nevertheless, cardiac artifacts may limit the use of Pdi as compared to other pressure-driving signals, most notably in bed-bound patients, because cardiac artifacts are more common in the supine position. However, assist modes are used chiefly in conscious patients who can be maintained in a seated or semiseated position. We suggest that Pdi may be less sensitive to cardiac artifacts than Pes. {Delta}Pdi/{Delta}Pes was greater than 1.0 in all seven subjects, and cardiac artifacts were similar in amplitude for both signals. This indicates that the cardiac noise-to-signal ratio can be reduced by using the higher amplitude signal (i.e., {Delta}Pdi), which may improve trigger sensitivity and prevent autotriggering. Pga is not influenced by cardiac activity and could, in theory, be used as the triggering and driving signal. However, we elected to use Pdi because of its considerably lower sensitivity to mechanical ventilation as compared to Pes or Pga alone. Mechanical inflation produces an increase in Pes, which is transmitted to the Pga independently from inspiratory activity. Accordingly, mechanical inflation does not change Pdi significantly. In our study, time from EMG activity onset to flow onset decreased with hypercapnia, albeit nonsignificantly, raising the possibility that detection of inspiratory effort by Pdi triggering may be less efficient during normocapnia. The decrease in time from EMG activity onset to flow onset as PETCO2 increased during servoventilator breathing was due to an increasingly steep Pdi rise as respiratory demand increased.

As with most pressure support devices, cycling-off with our Pdi-driven ventilator is based on flow. We voluntarily used flow rather than Pdi for cycling-off because it was not possible to anticipate the pattern of inspiratory activity that would occur with this new mode of mechanical ventilation. It has been shown that inspiratory activity is altered by respiratory muscle unloading (20). Thus, we were not sure that the Pdi signal peak would coincide with the onset of expiration during Pdi-driven ventilation as it does during spontaneous breathing. With an insensitive trigger set to a flow decrease to 50 ml/second, the time lag from the end of the Pdi peak to the end of ventilatory assistance was 200 to 300 milliseconds. Despite the limited efficiency of our expiratory triggering, expiratory muscle recruitment as reflected by the Pga increase during exhalation was not greater during servoventilator breathing than during spontaneous breathing. This finding established that the servoventilator did not increase expiratory loading. In addition, servoventilator breathing did not significantly increase iPEEP. As compared to spontaneous breathing, servoventilator breathing was not associated with an increase in Ti. On the contrary, this parameter showed a nonsignificant decrease. These findings indicate that expiration during servoventilator breathing occurred under relatively normal physiologic conditions, although we did not attempt to improve expiratory triggering. This result may be ascribable to the low resistance of the expiratory valve. Findings may be different in patients, particularly those with obstructive lung disease. The simplest way to assist these patients may be to increase the expiratory flow trigger value. Another possible solution may be to adjust the trigger on the Pdi signal decrease, although it remains unclear whether this method would improve cycling-off. Further studies are needed to devise better expiratory trigger strategies. Pdi deserves investigation as a potential expiratory trigger.

Pdi Signal Change as an Acceptable Index of Inspiratory Effort
The main characteristic of our servoventilator is that it generates an inspiratory airway pressure proportional to the Pdi increase produced by the subject during the inspiration time. We used the Pdi signal change rather than the absolute Pdi signal because our pressure probe did not give reliable values. This was expected, since piezoresistive transducer readings are highly temperature-dependent. In contrast to absolute Pdi, {Delta}Pdi has been shown to be unaffected by differences between room temperature and body temperature (21). As illustrated in Figures 4 and 5, respectively, airway pressure increased along the entire range of Pdi variation during servoventilator breathing in both Pdi-triggered and flow-triggered cycles. In addition, the measured airway PTP was satisfactory as compared to the theoretical airway PTP derived from the measured transdiaphragmatic PTP. These findings indicate that the Pdi-driven servoventilator was successful in achieving airway regulation at all tested PETCO2 levels.

Our Pdi-driven servoventilator can be roughly described as an external inspiratory muscle driven by the diaphragmatic mechanical output itself. Theoretically, the body can take advantage of this external inspiratory assistance in two ways. It can either maintain its central respiratory output, using the inspiratory assistance to increase ventilation, or it can keep ventilation unchanged by reducing its central output. We found that the transdiaphragmatic PTP was significantly lower during servoventilation than during PETCO2-matched spontaneous breathing. We acknowledge that Pdi may not reflect diaphragmatic energy expenditure: hyperinflation may reduce pressure generation, thereby contributing to decrease the Pdi, and Pdi changes during late inspiration are related both to active but decreasing diaphragmatic contraction and to passive lung recoil. Therefore, it is conceivable that the difference in transdiaphragmatic PTP between servoventilator breathing and PETCO2-matched spontaneous breathing was, to some extent, a mechanical effect of ventilation. However, the Pdi response to constant phrenic nerve stimulation has been shown to decrease by about 3% per 100 ml (22). In the present study, VT and PEEP differences were less than 400 ml and 1 cm H2O, respectively, between spontaneous and servoventilator breathing at all tested PETCO2 levels. Therefore, with Pdi-driven ventilation, the underestimation of inspiratory activity related to lung inflation is less than 20% at the end of lung inflation. In addition, Beck and colleagues (23) compared electrical diaphragmatic activity and Pdi signal during assisted mechanical ventilation. They found that varying the level of pressure support did not affect the Pdi modification for a given change in diaphragmatic EMG activity. Beck and coworkers concluded from their data that the neuromechanical coupling of the diaphragm was maintained and that Pdi change was an acceptable index of neural output by the respiratory center.

However, neuromechanical coupling of the diaphragm can be altered by changes in respiratory system mechanics. For instance, decreasing abdominal compliance by pressure applied to the abdomen may increase both the diaphragmatic load and the Pdi signal response to neural activity, thus inappropriately triggering the ventilator. We did not test the extent to which abdominal compliance may affect trigger sensitivity and flow delivery. When abdominal compliance decreases, neural output increases (24), but Pdi swing and ventilatory assistance increase more than neural activity. It would be of interest to test our Pdi-driven servoventilator in patients with reduced abdominal compliance related, for instance, to obesity, bloating, or ascites. Finally, we did not determine whether rib cage muscle contraction triggered the Pdi-driven servoventilator. Nevertheless, in patients with diaphragmatic paralysis, Pes can be used as the driving signal, despite limitations related to cardiac artifacts and esophageal spasms.

Effect of the Servoventilator on Breathing Pattern and Respiratory Motor Output
Our servoventilator was designed to provide Pdi-proportional airway pressure during the inspiratory time. To determine whether this was indeed the case, we assessed the relationship between measured airway PTP and theoretical airway PTP, which was derived from the transdiaphragmatic PTP generated by the subject during the inspiratory time (theoretical airway PTP = 2 · Ti + k · transdiaphragmatic PTP). We used the airway PTP because this parameter reflects airway pressure variations throughout the inspiratory time. We found a strong correlation between measured and predicted airway PTPs, indicating that airway pressure regulation was effective.

Breathing patterns were similar during servoventilator breathing and spontaneous breathing, as shown by the absence of statistically significant differences for E, VT, RR, Ti, Te, Ti/Ttot, or VT/Ti. These observations attest to the relative preservation of normal physiologic properties during use of our Pdi-driven servoventilator.

Respiratory centers are subjected to various behavioral influences, most notably during wakefulness. This may explain in part the decrease in diaphragmatic motor output. We were not able to evaluate this point. However, any behavioral response would probably decrease as PaCO2 increases. Indeed, the respiratory response to CO2 is usually considered to be chiefly automatic, although it may be altered by behavioral interferences (25).

The second possible explanation to the decrease in respiratory center output is neuromechanical inhibition of inspiratory activity. Whether a physiologically relevant mechanism of neuromechanical inhibition exists is a matter of debate (15, 26), although evidence from experimental and human studies supports this possibility. For instance, assisting inspiratory muscle activity by using conventional mechanical hyperventilation to increase VT (27, 28) or RR (28, 29) induces a marked decrease in respiratory motor output. This inhibitory effect has been observed at various PETCO2 levels ranging from hypocapnia to hypercapnia (9, 13, 30–34). It has been detected using a variety of indices of respiratory motor output, including peak inspiratory airway pressure (31), occlusion pressure after 0.1 second (35), rate of change in mask pressure (34), and surface or esophageal diaphragm EMG (9, 34). Neuromechanical inhibition has been demonstrated in awake (9, 28, 29, 35) and sleeping healthy humans (28, 33, 34, 36, 37). Whether it is preserved in double lung transplant recipients is currently under debate (37, 38). Neuromechanical inhibition has been found with various modes of conventional mechanical ventilation, including controlled or assisted ventilation (15, 29, 31, 33, 34, 38, 39) and pressure–support mechanical ventilation (9, 27, 35). It has also been suggested that neuromechanical feedback may account for a residual effect of assisted ventilation on the subsequent spontaneous breathing pattern (26). However, we were not able to evaluate this possibility because we did not record the transition between servoventilator breathing and spontaneous breathing. Recording started at least 1 minute after the onset of servoventilator breathing because of the time needed to find the proper setting. Finally, neuromechanical feedback is believed to be mediated by thoracic mechanoreceptors, particularly in the chest wall and diaphragm, whereas the role of upper and lower airway mechanoreceptors remains controversial (15, 26, 40).

The third possible explanation for the decay in inspiratory activity is neuromechanical dissociation, which may result from the contractile properties of the respiratory muscles (i.e., force–length relationships and force–velocity relationships). Muscle length, velocity of shortening during contraction, and curvature of the diaphragm can modify the relationships between neural and mechanical outputs (transdiaphragmatic and esophageal PTPs), especially during pressure ventilation (41). However, because neither iPEEP nor breathing pattern of our subjects were significantly modified by servoventilator breathing, we can conclude that the unloading produced by the servoventilator reduced respiratory muscle energy expenditure; whatever the underlying physiologic mechanisms, behavioral or otherwise, may have been, they were not elucidated in this study.

Comparison of the Pdi-driven Servoventilator to Other Modes of Mechanical Ventilation
Our concept of mechanical ventilation is very similar to that of Sinderby and colleagues (1), which uses a more direct estimation of respiratory center output, namely, diaphragmatic electrical activity recorded by a multiple-array electrode positioned in the esophagus at the level of and perpendicular to the diaphragmatic crura. The signals from each electrode pair in the array are differentially amplified, digitized into a personal computer, and filtered to minimize the influence of cardiac electrical activity and electrode motion. Unfortunately, we have no experience with the device used by Sinderby and colleagues, which is not available to us (1), and it is therefore difficult to compare our methods with theirs. Nevertheless, both methods seem to improve subject–ventilator interaction in terms of inspiratory triggering. In addition, they introduce a new approach to assisted ventilation, in which the patient can take full control of the mechanical support provided regardless of changes in respiratory drive, mechanics, and muscle function.

One of the main differences between our Pdi-driven servoventilator and conventional ventilators is that, during volumetric ventilation and pressure–support ventilation, a minimum VT is imposed independently from respiratory activity. This is obvious with volumetric ventilators but less so with pressure support. However, previous studies (23, 30, 42) have shown that pressure support can continue beyond the neural inspiratory time. In this situation, the inflation volume should be approximately equal to the product of inspiratory plateau pressure by static respiratory compliance (30, 35, 43). Behavioral alteration of the respiratory pattern is likely to occur with this type of device, which delivers a pressure only when inspiratory activity increases. Indeed, with conventional pressure–support devices, subjects and patients can voluntarily choose to produce just enough inspiratory activity to trigger the ventilator and then to stop their respiratory effort after triggering. This occurs during pressure support when the subjects and patients are asked to be as relaxed as possible and to reduce their respiratory activity as much as possible (30, 35, 43). Clearly, such behavioral influence cannot occur with our Pdi-driven servoventilator and/or servoventilators driven by airway pressure, flow, or volume.

The last few years have seen the development of so-called proportional-assist ventilation modes, in which airway pressure delivered by the machine is proportional to the airway pressure, flow, or volume generated by the subject. These new modes have been reported to decrease the work of breathing and to improve patient–ventilator interactions (44). We believe that our Pdi-driven servoventilator has several advantages over these new modes. First, Pdi more accurately reflects respiratory center output and respiratory needs than do airway pressure, flow, or volume, which are modified by airway resistance, lung and/or thoracic compliance, and air leaks. With our Pdi-driven servoventilator, airway pressure follows the pressure generated by the diaphragm independently from possible changes in the mechanical properties of the respiratory system. Second, ventilators directly influence flow and volume, which depend both on the patient's inspiratory activity and on the pressure delivered by the ventilator, and consequently there is a risk of runaway when flow or volume is used to drive the ventilator. Third, air leaks remain an unsolved problem with assist ventilation driven by flow or volume; Pdi-driven ventilation is not altered by air leaks. Fourth, the onset of inspiratory effort is detected sooner with Pdi than with flow, especially when iPEEP is present. Studies are needed to compare the efficiency of the Pdi-driven servoventilator to that of servoventilators driven by airway pressure, flow, or volume in healthy subjects and to evaluate the Pdi-driven servoventilator in patients with respiratory failure.

In conclusion, this study suggests a favorable interaction between healthy subjects and our innovative Pdi-driven servoventilator at various levels of respiratory demand. In addition, the Pdi-driven servoventilator unloaded the diaphragm, as indicated by the decreases in transdiaphragmatic and esophageal pressure–time products. This Pdi-driven ventilator deserves to be tested in patients and, in the future, may prove to be an alternative to proportional-assist ventilation or diaphragmatic EMG-driven ventilation.


   FOOTNOTES
 
Supported by the Institut National de la Santé et de la Recherche Médicale.

Conflict of Interest Statement: T.S. has no declared conflict of interest. G.D. has no declared conflict of interest. B.L. has no declared conflict of interest. G.M. has no declared conflict of interest. R.P. has no declared conflict of interest. A.H. has no declared conflict of interest. J-C.R. has no declared conflict of interest. D.I. has no declared conflict of interest. E.L. has no declared conflict of interest.

Received in original form March 25, 2002; accepted in final form May 15, 2003


   REFERENCES
TOP
 ABSTRACT
 METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Sinderby C, Navalesi P, Beck J, Skrobik Y, Comtois N, Friberg S, Gottfried SB, Lindstrom L. Neural control of mechanical ventilation in respiratory failure. Nat Med 1999;5:1433–1436.[CrossRef][Medline]
  2. Tobin MJ, Jubran A, Laghi F. Patient–ventilator interaction. Am J Respir Crit Care Med 2001;163:1059–1063.[Free Full Text]
  3. Remmers JE, Gautier H. Servo respirator constructed from a positive-pressure ventilator. J Appl Physiol 1976;41:252–255.[Abstract/Free Full Text]
  4. Brochard L, Isabey D, Piquet J, Amaro P, Mancebo J, Messadi AA, Brun-Buisson C, Rauss A, Lemaire F, Harf A. Reversal of acute exacerbations of chronic obstructive lung disease by inspiratory assistance with a face mask. N Engl J Med 1990;323:1523–1530.[Abstract]
  5. Brochard L, Mancebo J, Wysocki M, Lofaso F, Conti G, Rauss A, Simonneau G, Benito S, Gasparetto A, Lemaire F, et al. Noninvasive ventilation for acute exacerbations of chronic obstructive pulmonary disease. N Engl J Med 1995;333:817–822.[Abstract/Free Full Text]
  6. Isabey D, Boussignac G, Harf A. Effect of air entrainment on airway pressure during endotracheal gas injection. J Appl Physiol 1989;67:771–779.[Abstract/Free Full Text]
  7. Lofaso F, Aslanian P, Richard JC, Isabey D, Hang T, Corriger E, Harf A, Brochard L. Expiratory valves used for home devices: experimental and clinical comparison. Eur Respir J 1998;11:1382–1388.[Abstract]
  8. Baydur A, Behrakis PK, Zin WA, Jaeger M, Milic-Emili J. A simple method for assessing the validity of the esophageal balloon technique. Am Rev Respir Dis 1982;126:788–791.[Medline]
  9. Fauroux B, Isabey D, Desmarais G, Brochard L, Harf A, Lofaso F. Non-chemical influence of inspiratory pressure support on inspiratory activity in humans. J Appl Physiol 1998;85:2169–2175.[Abstract/Free Full Text]
  10. Lessard MR, Lofaso F, Brochard L. Expiratory muscle activity increases intrinsic positive end-expiratory pressure independently of dynamic hyperinflation in mechanically ventilated patients. Am J Respir Crit Care Med 1995;151:562–569.[Abstract]
  11. Hurvich C, Tsai C-L. regression and time series in model selection in small samples. Biometrika 1989;76:297–307.[Abstract/Free Full Text]
  12. Casati A, Gallioli G, Scandroglio M, Passaretta R, Borghi B, Torri G. Accuracy of end-tidal carbon dioxide monitoring using the NBP-75 microstream capnometer: a study in intubated ventilated and spontaneously breathing nonintubated patients. Eur J Anaesthesiol 2000;17:622–626.[CrossRef][Medline]
  13. Simon P, Skatrud J, Badr M, Griffin D, Iber C, Dempsey J. Role of airway mechanoreceptors in the inhibition of inspiration during mechanical ventilation in humans. Am Rev Respir Dis 1991;144:1033–1041.[Medline]
  14. Scheid P, Piiper J. Blood/gas equilibrium of carbon dioxide in lungs: a critical review. Respir Physiol 1980;39:1–31.[CrossRef][Medline]
  15. Younes M. Apnea following mechanical ventilation may not be caused by neuromechanical influences. Am J Respir Crit Care Med 2001;163:1298–1301.[Free Full Text]
  16. Takahashi E, Tateishi I, Yamamoto K, Mikami T. Effect of withdrawal of respiratory CO2 oscillations on respiratory control at rest. J Appl Physiol 1991;70:1601–1606.[Abstract/Free Full Text]
  17. Smith CA, Saupe KW, Henderson KS, Dempsey JA. Ventilatory effects of specific carotid body hypocapnia in dogs during wakefulness and sleep. J Appl Physiol 1995;79:689–699.[Abstract/Free Full Text]
  18. Lofaso F, Brochard L, Hang T, Lorino H, Harf A, Isabey D. Home versus intensive-care pressure support devices, experimental and clinical comparison. Am J Respir Crit Care Med 1996;153:1591–1599.[Abstract]
  19. Parthasarathy S, Jubran A, Tobin MJ. Assessment of neural inspiratory time in ventilator-supported patients. Am J Respir Crit Care Med 2000;162:546–552.[Abstract/Free Full Text]
  20. Gallagher CG, Younes M. Effect of pressure assist on ventilation and respiratory mechanics in heavy exercise. J Appl Physiol 1989;66:1824–1837.[Abstract/Free Full Text]
  21. Evans SA, Watson L, Cowley AJ, Johnston ID, Kinnear WJ. Normal range for transdiaphragmatic pressures during sniffs with catheter mounted transducers. Thorax 1993;48:753–758.
  22. ATS/ERS Statement on respiratory muscle testing. Am J Respir Crit Care Med 2002;166:518–624.[Free Full Text]
  23. Beck J, Gottrfried S, Navalesi P, Skrobik Y, Comtois N, Rossini M, Sinderby C. Electrical activity of the diaphragm during pressure support ventilation in acute respiratory failure. Am J Respir Crit Care Med 2001;164:419–424.[Abstract/Free Full Text]
  24. Lin K-H, Chuang C-C, Wu H-D, Chang C-W, Kou Y-R. Abdominal weight and inspiratory resistance: their immediate effects on inspiratory muscle functions during maximal voluntary breathing in chronic tetraplegic patients. Arch Phys Med Rehabil 1999;80:741–745.[CrossRef][Medline]
  25. Murphy K, Mier A, Adams L, Guz A. Putative cerebral cortical involvement in the ventilatory response to inhaled CO2 in conscious man. J Physiol 1990;420:1–18.[Abstract/Free Full Text]
  26. Dempsey JA, Skatrud JB. Apnea following mechanical ventilation may be caused by nonchemical neuromechanical influences. Am J Respir Crit Care Med 2001;163:1297–1298.[Free Full Text]
  27. Morrell MJ, Shea SA, Adams L, Guz A. Effects of inspiratory support upon breathing in humans during wakefulness and sleep. Respir Physiol 1993;93:57–70.[CrossRef][Medline]
  28. Manchanda S, Leevers AM, Wilson CR, Simon PM, Skatrud JB, Dempsey JA. Frequency and volume thresholds for inhibition of inspiratory motor output during mechanical ventilation. Respir Physiol 1996;105(1–2):1–16.[CrossRef][Medline]
  29. Lake FR, Finucane KE, Hillman DR. Diaphragm inhibition with positive pressure ventilation: quantification of mechanical effects. Respir Physiol 1999;118(2–3):149–161.[CrossRef][Medline]
  30. Azarian R, Lofaso F, Zerah F, Lorino H, Atlan G, Isabey D, Harf A. Assessment of respiratory compliance in awake subjects using pressure support. Eur Respir J 1993;6:552–558.[Abstract]
  31. Altose M, Castele R, Connors A, Dimarco A. Effects of volume and frequency of mechanical ventilation on respiratory activity in humans. Respir Physiol 1986;66:171–180.[CrossRef][Medline]
  32. Iber C, Simon P, Skatrud JB, Mahowald MW, Dempsey JA. The Breuer-Hering reflex in humans. Effects of pulmonary denervation and hypocapnia. Am J Respir Crit Care Med 1995;152:217–224.[Abstract]
  33. Henke K, Arias A, Skatrud J, Dempsey J. Inhibition of inspiratory muscle activity during sleep. Am Rev Respir Dis 1988;138:8–15.[Medline]
  34. Wilson C, Satoh M, Skatrud J, Dempsey J. Non-chemical inhibition of respiratory motor output during mechanical ventilation in sleeping humans. J Physiol 1999;518:605–618.[Abstract/Free Full Text]
  35. Scheid P, Lofaso F, Isabey D, Harf A. Respiratory response to inhaled CO2 during positive inspiratory pressure in humans. J Appl Physiol 1994;77:876–882.[Abstract/Free Full Text]
  36. Leevers A, Simon P, Xi L, Dempsey J. Apnoea following normocapnie mechanical ventilation in awake mammals: a demonstration of control system inertia. J Physiol (Lond) 1993;472:749–768.[Abstract/Free Full Text]
  37. Simon P, Dempsey J, Landry D, Skatrud J. Effect of sleep on respiratory muscle activity during mechanical ventilation. Am Rev Respir Dis 1993;147:32–37.[Medline]
  38. Lofaso F, Simmoneau G, Le Roy Ladurie F, Cerrina J, Chapelier A, Brenot F, Dartevelle P, Herve P. Frequency of mechanical ventilation and respiratory activity after double lung transplantation. Respir Physiol 1993;92:319–327.[CrossRef][Medline]
  39. Simon P, Griffin D, Landry D, Skatrud J. Inhibition of respiratory activity during passive ventilation: a role for intercostal afferents? Respir Physiol 1993;92:53–64.[CrossRef][Medline]
  40. Hering E. Die selbsteuerung der athmung durch den nervus vagus. Sitzber Akad Wiss Wien 1868;57:672–677.
  41. Corne S, Webster K, Younes M. Effects of inspiratory flow on diaphragmatic motor output in normal subjects. J Appl Physiol 2000;89:481–492.[Abstract/Free Full Text]
  42. Fernandez R, Mendez M, Younes M. Effect of ventilator flow rate on respiratory timing in normal humans. Am J Respir Crit Care Med 1999;159:710–719.[Abstract/Free Full Text]
  43. Stroetz RW, Hubmayr RD. Tidal volume maintenance during weaning with pressure support. Am J Respir Crit Care Med 1995;152:1034–1040.[Abstract]
  44. Younes M. Proportional assist ventilation, a new approach to ventilatory support. Theory. Am J Respir Crit Care Med 1992;145:114–120.



This article has been cited by other articles:


Home page
 Am. J. Respir. Crit. Care Med.Home page
A. Jubran, B. J. B. Grant, F. Laghi, S. Parthasarathy, and M. J. Tobin
Weaning Prediction: Esophageal Pressure Monitoring Complements Readiness Testing
Am. J. Respir. Crit. Care Med., June 1, 2005; 171(11): 1252 - 1259.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Respir. Crit. Care Med.Home page
J. Spahija, J. Beck, M. de Marchie, A. Comtois, and C. Sinderby
Closed-Loop Control of Respiratory Drive Using Pressure-Support Ventilation: Target Drive Ventilation
Am. J. Respir. Crit. Care Med., May 1, 2005; 171(9): 1009 - 1014.
[Abstract] [Full Text] [PDF]

Home page
 J. Appl. Physiol.Home page
T. Sharshar, E. T. Ross, N. S. Hopkinson, R. Porcher, A. H. Nickol, S. Jonville, M. J. Dayer, N. Hart, J. Moxham, F. Lofaso, et al.
Depression of diaphragm motor cortex excitability during mechanical ventilation
J Appl Physiol, July 1, 2004; 97(1): 3 - 10.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Respir. Crit. Care Med.Home page
M. J. Tobin
Critical Care Medicine in AJRCCM 2003
Am. J. Respir. Crit. Care Med., January 15, 2004; 169(2): 239 - 253.
[Full Text] [PDF]

Home page
 Am. J. Respir. Crit. Care Med.Home page
C. Sinderby
Ventilatory Assist Driven by Patient Demand
Am. J. Respir. Crit. Care Med., October 1, 2003; 168(7): 729 - 730.
[Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
200203-241OCv1
168/7/760    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow