INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Techniques for partial ventilatory support are intended for patients who have normal respiratory drive but who have difficulty sustaining adequate spontaneous ventilation. The most popular mode of such assistance is pressure-support ventilation (PSV), in which the ventilator generates a constant pressure that acts in addition to the patient's effort throughout the inspiratory period (1). With PSV, the pressure applied by the ventilator (P_appl) rises to a preset level that is held constant until a cycle-off criterion (a minimum inspiratory flow value) is reached. The inspiratory flow and tidal volume (VT) are thus related to the patient's inspiratory effort, the level of P_appl, and the respiratory system impedance; breathing frequency is determined by the patient's own respiratory drive (5).

Proportional-assist ventilation (PAV) is an alternative mode of partial ventilatory support in which the ventilator generates pressure in proportion to the patient's effort (5). Thus, during PAV, P_appl is a function of patient effort: the greater the inspiratory effort, the greater is the increase in P_appl. Flow and VT will therefore be determined by the level of proportionality between P_appl and the patient's effort, and the impedance of the respiratory system. Ventilator assistance terminates with the end of the inspiratory effort, and RR is determined by the patient's own respiratory drive (5).

Spontaneous variations in impedance of the respiratory system commonly occur in mechanically ventilated patients (8), and may impair the matching between the ventilator output and the patient's ventilatory demand during partial ventilatory support (12), possibly leading to the development of patient-ventilator asynchrony (11). The different approaches used in PSV and PAV to pressurize the lung could theoretically lead to marked differences in response to these variations in respiratory system impedance (5).

Respiratory loading has often been used to simulate changes in respiratory impedance (13, 14) and to evaluate the consequences of such changes on ventilatory patterns and respiratory muscle performance (15). The aim of this study was to assess ventilatory responses to added mechanical loads during PSV and PAV in patients during the weaning period. Our hypothesis was that during PSV, preservation of E consequent to respiratory loading would primarily be accomplished by an increase in RR caused by a large reduction in T. By contrast, we hypothesized that preservation of E after respiratory loading would occur through the relative preservation of VT, with a smaller effect on RR during ventilation with PAV. We anticipated that this compensatory strategy would result in greater patient comfort and less inspiratory muscle effort.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Patient Selection

Patients were enrolled in the study after entering the weaning process as prescribed by the attending physician. Entry criteria also included clinical and hemodynamic stability and a maximal inspiratory pressure (PI_max) of at least 20 cm H₂O. Exclusion criteria included unconsciousness and/or presence of a chest tube. Ten patients admitted to the intensive care units of the Policlinico and DiVenere Hospitals of the University of Bari were studied. They were intubated and mechanically ventilated for the management of acute respiratory failure. All patients were receiving PSV (Siemens Servo Ventilator 300; Siemens Elema AB, Berlin, Germany) with pressures ranging between 10 and 14 cm H₂O (11.9 ± 0.5 cm H₂O [mean ± SEM]). The local ethics committee approved the investigative protocol, and written informed consent was obtained from each patient. A physician not involved in the study was always present for patient care during the experimental phase of the study.

Study Protocol

A key element of the study was to ensure that PSV and PAV provided an equivalent level of support. To ensure this, we provided an equal degree of respiratory muscle unloading for both PSV and PAV, using the pressure-time product of the diaphragm per minute (PTP/min) as the target variable, since it has been shown to correlate with the O₂ cost of breathing (16). A similar decrease in this variable with each ventilatory mode would therefore indicate equivalent levels of ventilatory support. To accomplish this, we disconnected patients from the ventilator and allowed them to breathe spontaneously. Twenty to 30 consecutive breaths were allowed over a period of 2 to 3 min. VT, RR, the ratio of inspiratory time to total breathing cycle time r (TI/Ttot), and PTP/min of the diaphragm during spontaneous breathing (SB) were then measured.

Respiratory muscle inactivity was than achieved by injecting a short-acting hypnotic agent (propofol, 0.3 mg/kg/min for 5 min), and controlled mechanical ventilation was started (17). Ventilator parameters were set in such a way as to match the breathing pattern recorded during SB (18). Intrinsic positive end-expiratory pressure (PEEPi), static elastance (ERS_st) and total resistance (RRS_tot) of the respiratory system were measured by applying end-expiratory and end-inspiratory airway occlusions as previously described (9).

Thirty minutes after these measurements, when respiratory muscle activity returned toward normal (judged to have occurred when negative swings in esophageal pressure developed during inspiration) and the patients awakened, the Siemens ventilator was replaced by a Winnipeg ventilator (University of Manitoba, Winnipeg, MB, Canada). The design and operation of this unit are similar to those previously described (5, 6, 8, 12). The level of pressure (during PSV) and percentage of unloading (during PAV) were set to obtain a 60 to 70% decrease in PTP/min of the diaphragm relative to the values obtained during the spontaneous breathing (SB) trial. Values of ERS_st and RRS_tot obtained during the trial of controlled mechanical ventilation were used to set PAV. In patients with chronic obstructive pulmonary disease, PEEP was set at 80% of the PEEP_i measured during controlled mechanical ventilation (9). The set level of pressure (during PSV), the percentage of assistance (during PAV), the fractional inspired oxygen concentration (FI_O2), and the eventual PEEP level remained constant throughout the different experimental conditions.

The increase in respiratory workload was obtained by strapping the chest wall and the abdomen (19, 20). The chest wall was strapped with a nearly inelastic cloth corset (25 cm long in the craniocaudal axis) with adjustable straps, and a rectangular pneumatic cuff (20 × 30 cm) was inserted under the corset and placed over the anterior chest. A similar corset (15 cm long in the craniocaudal axis) was wrapped around the abdomen and a rectangular pneumatic cuff (20 × 30 cm) was placed between the corset and the ventral part of the abdomen (19, 20). The straps of the corsets were adjusted to that breathing was not hampered when the pneumatic cuffs were deflated. When all respiratory variables were stable, the pneumatic cuffs were inflated to a pressure of 20 mm Hg (19, 20).

Modes of ventilatory support were randomized through a concealed allocation approach, utilizing opaque, sealed envelopes containing the randomization schedule. Application of the respiratory load was randomized once the ventilatory mode was selected. Measurements were obtained from 2 to 3 min of data recorded before (load off) and after 5 to 8 min of chest and abdominal compression (load on) for each mode of ventilation.

Measurements

Flow was measured with a heated pneumotachograph (Fleisch No. 2; Fleisch, Lausanne, Switzerland) connected to a differential pressure transducer (Validyne MP 45 ± 2 cm H₂O; Validyne Co., Northridge, CA), which was inserted between the Y-piece of the ventilator circuit and the endotracheal tube. The pneumotachograph provided a linear recording over the experimental range of flow. Equipment dead space (not including the endotracheal tube) was 70 ml. Airway opening pressure (Pao) was measured proximal to the endotracheal tube, with a pressure transducer (Validyne MP 45 ± 100 cm H₂O). Changes in intrathoracic and abdominal pressures were evaluated by assessing esophageal (Pes) and gastric (Pga) pressures. Pes and Pga were measured with thin, latex balloon-tipped catheter systems connected by polyethylene catheters to separate differential pressure transducers (Validyne MP 45 ± 100 cm H₂O). The esophageal balloon was correctly positioned by means of an occlusion test (21). All of the variables described here were displayed on an eight-channel strip-chart recorder (Model 7718A; Hewlett-Packard Co., Cupertino, CA), and were collected on a personal computer through a 12-bit analog-to-digital converter at a sampling frequency of 100 Hz. Subsequent data analysis was done with the ANADAT software package (RHT-InfoDat, Montreal, PQ, Canada). Patients were studied while in the semirecumbent position.

Breathing pattern. VT was computed by digital integration of the flow signal. PI_max was measured as previously described (9). Inspiratory time (TI), expiratory time (TE), and total breathing cycle time (Ttot) were determined from the flow tracing.

Indexes of O₂ consumption of the diaphragm. Tidal excursions of Pes (Pes) and transdiaphragmatic pressure (Pdi) were determined. Pdi was calculated as Pga minus Pes. All pressure swings are reported as changes from the end-expiratory value rather than from absolute zero pressure. This permits exclusion of the passive increase introduced by the application of corsets (and particularly the increase in Pga), permitting comparison of the active pressures generated under the different experimental conditions (19, 20). PTP per breath (PTP/b) was obtained by measuring the area under the Pdi signal from the beginning of the inspiratory deflection to the end of inspiratory flow (16). PTP/min was calculated as PTP/b multiplied by RR. PTP/L was calculated as PTP/min divided by E.

Respiratory mechanics during PSV and PAV. Respiratory mechanics during the different experimental conditions were assessed with the Mead and Wittenberger technique (22). Briefly, inspiratory pulmonary resistance (RL) and elastance (EL) were calculated by fitting the equation of motion of a single-compartment model using multilinear regression, as follows:

(1)

where PL is inspiratory change in transpulmonary pressure (calculated by subtracting Pes from Pao), VT is tidal volume, and flow is peak inspiratory flow. The level of PEEP_i during the different experimental conditions (PEEPi,_dyn) was measured as the negative deflection in Pes from the onset of inspiratory effort to the point of zero flow. In the case of active recruitment of the abdominal muscles, this value was corrected by subtracting the decrease in Pga, when present, from the decrease in Pes during the interval when PEEPi,_dyn was measured (10, 23).

Intensity of dyspnea. The intensity of breathlessness was rated with a dyspnea visual analogue scale (VAS) at 10 to 15 min after the beginning of each experimental trial (24). Patients were asked to place a vertical mark on a printed 100-mm horizontal scale in response to the question: "How short of breath are you right now?" The line had descriptors below its extreme ends. On the left was the word "none," indicating no shortness of breath, and on the right was the opposite response, "extremely severe." For each experimental condition, patients placed a vertical mark on the line at the point that best represented the intensity of their dyspnea. Intensity was measured as the distance in millimeters from the left end of the horizontal line (corresponding to no dyspnea) to the mark placed by the patient. A fresh scale was presented on each occasion that these measurements of breathing comfort were made. Before the protocol began, directions for using the scale were read aloud, and all patients then practiced marking the scale.

Statistical Analysis

Results are expressed as mean ± SEM. Values obtained during the different experimental conditions were compared through repeated-measures two-way analysis of variance (ANOVA) and Bonferroni's test. Regression analysis was done with the least-squares method, using experimental data taken during the period from 2 to 3 min after data acquisition was begun (StatView, software package; Abacus Inc., Berkeley, CA).

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Causes of acute respiratory failure, gender, age, days of mechanical ventilation, static respiratory mechanical parameters, and blood gas values for the patients in the study are shown in Table 1. The PEEP level used in the study was 5.2 ± 0.7 cm H₂O.

Application of ventilatory support decreased inspiratory swings in Pdi (Figure 1), and PTP/b, PTP/min, and PTP/L (Figure 2) to a similar degree during PSV and PAV, relative to their magnitude during spontaneous breathing. This was obtained through the application of 12 ± 1 cm H₂O pressure during PSV; the percentages of elastic and resistive unloading during PAV were set to normalize patient resistive and elastic forces (9), and were 44 ± 3% and 58 ± 4%, respectively, which resulted in an applied pressure of 13 ± 2 cm H₂O. RR and levels of airway opening pressure (Pao) and VT were similar in both ventilatory modes prior to chest and abdominal binding (Table 2).

View larger version (17K): [in this window] [in a new window]   Figure 1.   Experimental records illustrating effects of chest and abdominal binding is a representative patient. From top to bottom: flow, Pao, volume change (V), and tidal excursion of transdiaphragmatic pressure (Pdi). SB = spontaneous breathing; PSV = pressure-support ventilation; PAV = proportional-assist ventilation.

View larger version (16K): [in this window] [in a new window]   Figure 2.   Effects of chest and abdominal binding on indexes of O2 of the diaphragm. Pdi = inspiratory tidal excursion of transdiaphragmatic pressure; PTP/b = pressure time product per breath; PTP/min = pressure time product per breath; PTP/L = pressure time product per liter; SB = spontaneous breathing; PSV = pressure-support ventilation; PAV = proportional-assist ventilation. Open bar: load off, closed bar: load on.

During SB, the dyspnea score was 46.0 ± 1.2 mm. When asked to indicate changes in the degree of their breathlessness during PSV and PAV with respect to the preceding SB condition, all patients reported similar reductions in dyspnea (38.9 ± 4.3, 23.5 ± 3.8, and 20.8 ± 5.2 mm during SB, PSV, and PAV, respectively; p < 0.001) (Figure 3).

View larger version (8K): [in this window] [in a new window]   Figure 3.   Visual analog scale (VAS) of patient-perceived intensity of breathlessness under different experimental conditions. SB = spontaneous breathing; PSV = pressure-support ventilation; PAV = proportional-assist ventilation.

Chest and abdominal compression caused a similar increase in Pdi and PTP/b with both modes of ventilatory support (Figure 2). During PAV, the increase in inspiratory muscle effort was followed by a concomitant increase in Pao, whereas during PSV, Pao remained constant after application of the respiratory load (Figure 1). Variables summarizing the patients' breathing patterns during the different experimental conditions are given in Table 2. During PSV and PAV, chest and abdominal compression caused similar increases in EL (28 ± 3% and 30 ± 2%; p = NS) and RL (49 ± 3% and 52 ± 4%; p = NS), respectively. A 2- to 3-cm H₂O increase in PEEPi was also observed in both modes of ventilatory support. E remained constant in both ventilatory modes after chest and abdominal compression. During PSV, this was achieved through a 58 ± 3% increase in RR that compensated for a 29 ± 2% reduction in VT. The magnitudes of the reduction in VT (10 ± 3%) and of the increase in RR (14 ± 2%) were significantly (p < 0.001) smaller during PAV.

During both PSV and PAV, chest and abdominal compression caused increases in PTP/min (142.9 ± 26.9 cm H₂O · s/ min, PSV, and 117.6 ± 16.4 cm H₂O · s/min, PAV) and PTP/L (13.4 ± 2.5 cm H₂O · s/L, PSV, and 9.6 ± 0.7 cm H₂O · s/L, PAV) (Figure 2), as well as in the dyspnea score (53.4 ± 4.0 mm, PSV, and 33.8 ± 5.9 mm, PAV) (Figure 3). During load application, values of PTP/min, PTP/L, and dyspnea score were significantly (p < 0.001) greater during PSV than during PAV (Figures 2 and 3).

Inspiratory muscle effort (estimated from PTP/b) was plotted against VT for 2 to 3 min of data acquisition during SB and during load off and load on conditions with PSV and PAV (Figure 4). During SB, a significant correlation (p < 0.001) was found between VT and PTP/b in all patients (slope = 0.07 ± 0.02). Relative to SB, application of PSV decreased and application of PAV increased the slope of this relationship (0.3 ± 0.1 and 0.02 ± 0.05, respectively; p < 0.001). During PSV, chest and abdominal compression displaced the correlation line downward without affecting its slope. During PAV, chest wall and abdominal compression decreased the slope of the relationship between PTP/b and VT in all patients to 0.1 ± 0.05 (p < 0.0001).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

This study demonstrates that in mechanically ventilated patients who are being weaned from ventilation, the strategy used to compensate for an acute increase in respiratory impedance differs markedly with PSV and PAV. During PSV, despite a substantial increase in inspiratory effort, VT decreased after chest and abdominal binding, whereas E was kept constant by an increase in RR. During PAV, E was maintained by increasing Pao such that VT and RR remained substantially unchanged. Larger values of PTP/min and PTP/L were required during PSV than during PAV to keep E constant in response to an increase in respiratory impedance. Furthermore, patients' sensation of dyspnea during chest and abdominal binding was less intense during PAV than during PSV.

Before discussing the results of the present study, we will address a number of general considerations concerning its experimental design, our assessment of inspiratory effort, and the validity of measuring dyspnea after administration of a hypnotic agent.

First, both PSV and PAV are capable of unloading the respiratory muscles (5); indeed with either mode, the work performed by the ventilator can be varied from zero to nearly the total work of breathing required of the patient. A critical feature of our experimental design was to provide equal support with both PSV and PAV before increasing respiratory system impedance. We chose to do this by setting the PSV and PAV support levels to provide a similar reduction in inspiratory effort, relative to a trial of SB (Figures 1 and 2) before the impedance load was applied. We did this because McGregor and Becklake (25) have shown that the O₂ cost of breathing is closely related to the mean pressure developed by the diaphragm, and assessed in the present study as the mean PTP/ min (26, 27). Independent evidence suggesting that this approach was successful was the observation that the dyspnea score decreased by an equivalent amount with PSV and PAV (Figure 3). In addition, the respiratory workload (i.e., PEEPi, RL, and EL for a given patient) was similar with the two modes of ventilatory support (Table 2). It is interesting to note that under these experimental conditions, PSV and PAV provided a similar breathing pattern (Table 2).

Second, the measurements in our study were obtained from 2 to 3 min of data recorded before and after 5 to 8 min of chest and abdominal compression. Given this, the results of our study may not reflect a steady-state condition.

Third, the time elapsed between the end of the the propofol infusion and the first assessment of dyspnea score was 68 ± 3 min, and was therefore longer than the awakening time (14 ± 13 min) reported for critically ill patients with a propofol infusion rate similar to the one used in our study (28).

Studies of the normal response to increases in chest and abdominal binding suggest that humans maintain nearly normal levels of alveolar ventilation unless such binding creates marked changes in impedance. The immediate response is a reduced VT with minimal change in RR (13, 19, 20). When strapping is maintained, VT progressively increases and E returns to control values, with little change in RR and arterial blood gas values (13, 19, 20). Inspiratory muscle effort is increased immediately after the chest wall and abdomen are strapped, and progressively increases during subsequent breaths as VT approaches control values (13, 19, 20). This increase in inspiratory effort returns VT and E to control values, with an increase in amplitude of respiratory muscle contraction and changes in shape of the phases of inspiratory muscle activity, without substantial changes in RR (13, 14). The steady-state response is due to the physiologic coupling between VT and inspiratory muscle effort (29). The magnitude of effort and VT are linearly correlated, and in the presence of an increase in respiratory impedance, the slope of the relationship between them will decrease; however, the greater output will restore VT to baseline levels (29). This physiologic ability to integrate respiratory drive, inspiratory muscle effort, and VT on the basis of different ventilatory requirements has been designated "neuroventilatory coupling" (29, 32).

Partial ventilatory support is principally indicated in patients whose respiratory drive is normal or high but who have difficulty in sustaining an adequate level of ventilation on their own, and in whom an abnormal relationship between effort and ventilation is present (33). The great majority of these patients have abnormal neuroventilatory coupling due to neuromuscular weakness (which necessitates a greater effort to produce a given pressure) and/or abnormal respiratory mechanics (which requires a greater pressure to generate a given level of ventilation). High ventilatory demand caused by sepsis, fever, or metabolic acidosis, for example, may further compound this situation (34). During partial ventilatory support, the patient- ventilator interface can be described by using the equation of motion (22). At any instant during a breath, the total pressure applied to the patient's respiratory system includes the pressure generated by the respiratory muscles (Pmus) and the pressure applied by the ventilator (P_appl). This pressure is dissipated against: (1) PEEPi; (2) the patient's resistance (Rtot); and (3) the patient's static elastance (EL_st). Under these circumstances, the act of breathing in a mechanically ventilated patient can be described at any instant:

(2)

where Pres represents the resistive pressure and is a function of inspiratory flow (Pres = inspiratory flow × Rtot) and Pel represents the elastic recoil pressure and is a function of VT (PEL = VT × Est). Assuming that Rtot and Est are linear, Equation 2 becomes:

(3)

By providing a constant P_appl, PSV unloads the respiratory muscles and improves the relation between patient effort and VT, in the sense that for a given inspiratory effort, the patient receives a greater volume than would have been received during SB (33). However, since the ventilator provides the same P_appl with every triggered breath, any increase in respiratory muscle activity will not produce an increase in ventilator applied pressure (5).

With PAV the ventilator provides a pressure assist that is proportional to the instantaneous inspiratory effort. The patient's ability to alter VT through changes in effort is hence preserved, and can be modulated by changing the level of proportionality (5-7, 9). With PAV, P_appl is a function of the flow and volume generated by patient effort. During PAV, P_appl = kR × flow + kE × volume (5, 9), where the coefficients kR and KE represent the proportionality between P_appl and flow and volume, respectively, generated by patient effort (5); in the case of linearity of RL and EL, and according to Equation 2, the use of PAV can be described by the following equation:

(4)

With PAV, what is preset is not the P_appl level (as in PSV), but the proportion between inspiratory muscle effort and ventilator applied pressure (regulated by setting the KR and KE values on the ventilator) (i.e., the extent to which P_appl will increase for a given increase in inspiratory muscle effort) (5, 6). Any increase in respiratory muscle activity will therefore be followed by a concomitant increase in ventilator applied pressure (5).

To evaluate the physiologic implications of these theoretical differences, we evaluated effects of PSV and PAV on neuroventilatory coupling at different levels of ventilatory workload (Figure 4) (26). The theoretical relation between inspiratory muscle effort (P_mus in Equation 1, quantified as PTP/b) and VT in a normal subject (slope = 0.5 in Figure 4) was estimated with data from a previously described model (5, 29). In our study, all patients had significant impairment of their neuroventilatory coupling during a trial of SB, as indicated by the values of the slope of the relationship between PTP/b and VT, which were below the normal predicted value (slope = 0.07 ± 0.02). By study design, application of PSV and PAV prior to chest and abdominal binding decreased PTP/b and increased VT by a similar amount (2.2 ± 0.8 cm H₂O · s and 2.1 ± 0.3 cm H₂O · s during PSV and PAV, respectively; and 0.71 ± 0.02 L and 0.68 ± 0.13 L during PSV and PAV, respectively). However, during PSV, the improvement in breathing efficacy occurred concomitantly with the further reduction in the slope of the relationship between PTP/b and VT (0.02 ± 0.05). On the other hand, application of PAV decreased effort and increased ventilation with an increase in the slope of this relationship that virtually returned it to the normal theoretical value (0.3 ± 0.1).

View larger version (18K): [in this window] [in a new window]   Figure 4.   Effects of increasing inspiratory effort on volume during PSV (top panel ) and PAV (bottom panel ). PTP/b: pressure time products of the diaphragm per breath; V: volume change. Open squares: spontaneous breathing; open circles: load off; closed circles: load on. Dotted line: theoretical relation between inspiratory muscle effort and volume in a normal subject. See text for further details.

Chest and abdominal binding increased PTP/b by a similar amount in both ventilatory modes (6.6 ± 0.8 cm H₂O · s and 6.4 ± 0.6 cm H₂O · s during PSV and PAV, respectively; Figure 2). However, during PSV, because of the parallel downward shift of the PTP/b-versus-VT relationship, VT fell, since the increase in PTP/b did not offset the increase in impedance. E remained constant only because of the increase in RR. On the other hand, during PAV, although chest wall and abdominal restriction decreased the slope of the PTP/b-versus-VT relationship (0.1 ± 0.05), the increase in PTP/b was sufficient to increase ventilator applied pressure to maintain T close to baseline values. Under these circumstances, E remained constant without substantial changes in RR. These results therefore explain why, despite a similar degree of inspiratory muscle effort, larger values of PTP/min and PTP/L during PSV (142.9 ± 26.9 cm H₂O · s/min and 13.4 ± 2.5 cm H₂O · s/L, respectively) than during PAV (117.6 ± 16.4 cm H₂O · s/min and 9.6 ± 0.7 cm H₂O · s/L, respectively) were required to keep E constant in response to an increase in respiratory impedance. The decreased sense of dyspnea during chest wall and abdominal binding observed during PAV is therefore consistent with the lower O₂ cost of breathing required to maintain a constant E with this ventilatory mode (27). However, before loading, RR tended to be lower during PSV than during PAV, although the difference was not significant; this could therefore have introduced a bias toward a larger increase in RR with load during PSV than during PAV.

A limitation of PSV may be the instability in VT of patients with marked changes in respiratory impedance (5, 33, 34). This may be clinically important, since recent studies have shown that spontaneous variations in total impedance of the respiratory system may occur in mechanically ventilated patients (8). This has led to the suggestion that in addition to intact respiratory drive, stable impedance is also an important requirement for the use of PSV as a sole means of ventilatory support (5, 33, 35). Information about the magnitude and the frequency of such changes in respiratory impedance in mechanically ventilated patients is scanty. Despite this lack of data, we would suggest that the increases in impedance induced by the banding in our study are of a magnitude sufficient to be clinically relevant. The obtained increases in elastance and resistance (~ 30% and 50%, respectively) were such that during PSV, they produced marked changes in both breathing pattern (58 ± 3% increase in RR and 29 ± 2 decrease in VT) and inspiratory muscle effort (224 ± 18% increase in PTP/b).

In a previous study, we showed that when ventilatory requirements were increased by acute hypercapnia, E increased mainly through changes in VT only during PAV; during PSV the increase in E was obtained through an increase in RR (26). This resulted in greater muscle effort and more pronounced dyspnea during PSV than during PAV (26). Results of the present study confirm that in the presence of variations in ventilatory requirements as a result of increases in mechanical load, the use of PSV may also be limited by the reduction in VT. However, in contrast to our previous observations (26), the present study suggests that impairment of the patient-ventilator interaction after an increase in mechanical load may also be observed during application of PAV. Although PAV was able to maintain a breathing pattern similar to that observed before loading conditions were applied, this occurred through a significant increase in patient workload. Although this preserved VT at a lower PTP/min and PTP/L than with PSV, the prolonged increase in inspiratory effort could lead to muscle fatigue. Since P_appl is proportional to P_mus during PAV, this may lead to an inadequate E, resulting in hypercapnia and acute respiratory acidosis (12). The full potential benefits of PAV can therefore be obtained only with continuous adaptation of the degree of ventilatory assistance provided by this modality to the changes in respiratory mechanics that should be continuously monitored in the definitive technologic implementation of PAV (9).

In conclusion, our data show that in mechanically ventilated patients in whom respiratory impedance is acutely increased by chest wall and abdominal binding, the physiologic capability of keeping VT and E constant through increases in inspiratory effort was preserved only during PAV. During PSV, despite a similar increase in inspiratory effort to that in PAV, VT decreased, and the increase in RR preserved E. The ventilator response to an added respiratory load during PSV required greater muscle effort and caused more pronounced patient discomfort than during PAV. These data confirm that with PSV, although the patient receives a mandatory degree of support, the patient's ability to modulate the ventilatory pattern through changes in motor output remains impaired. Although with PAV the ability of the patient to unload the work of breathing in proportion to inspiratory effort is enhanced, changes in respiratory impedance are followed by increases in inspiratory effort. This may eventually lead to the impairment of patient-ventilator interactions. This limitation of PAV may be related to the specific prototype used in our study, and may be overcome by implementing positive feedback to continuously adapt the level of assistance to changes in patient respiratory mechanics.