Comparison of Pressure Support Ventilation and Assist-Control Ventilation in the Treatment of Respiratory Failure

Comparison of Pressure Support Ventilation and Assist-Control Ventilation in the Treatment of Respiratory Failure* Manuel Tej eda, MD; Jesus Hector Boix, MD; Faustino Alvarez, MD; Reyes Balanza, MD; and Maria Morales, MD

Study objective: To assess whether pressure support ventilation (PSV) could be used as an alternative ventilatory mode to assist-control (A/C) ventilation in the treatment of respiratory failure. Design: A short-term (4-h) prospective study in which the beneficial effect of PSV on respiratory mechanics, gas exchange, arterial oxygenation, cardiovascular hemodynamics, and oxygen consumption was compared with A/C ventilation. Setting: ICU of a community hospital. Patients: Forty-five patients (mean age, 62.8 [11.8] years) with respiratory failure secondary to COPD, restrictive disorders, or neuromuscular disease requiring mechanical ventilatory support in the ICU were selected for study. Interventions: The mean duration of mechanical ventilation before the study was 7.16 (8.64) days. Patients were switched to the PSV mode of the mechanical ventilator for a period of 4 h after which conventional A/C ventilation was resumed. Results: Patients supported with PSV compared with A/C ventilation showed significantly higher tidal volume, minute ventilation, and inspiratory time in association with significantly lower pressure in the airway and I:E ratio. With regard to gas exchange data, an increase in dead space/tidal volume ratio (VDNT), decrease in Pa02 , and statistically but not clinically significant alteration of arterial oxygenation indexes were noted. However, when patients with COPD, restrictive disorders, and neuromuscular disease were compared, significant changes in arterial oxygenation parameters were found only in patients with restrictive disorders. There were significant deet·eases in heart rate, systolic pulmonary artery pressure, and pulmonary capillary wedge pressure when PSV was applied. Oxygen transport and oxygen consumption were unchanged. Conclusions: PSV could be a possible alternative to A/C ventilation in patients with respiratory failure. PSV caused an increase in VDNT in association with a significantly lower pressure in the airway and I:E ratio. Randomized studies are needed to defme the long-term benefits of both respiratory modes and the conditions in which PSV may be a valuable alternative to A/C ventilation. (CHEST 1997; 111:1322-25) Key words: assist-control ventilation; chronic obstructive pulmona1y disease; hemodynamics; neuromuscular disease; oxygen consumption; pressure suppmt ventilation; respiratory failure; respiratory mechanics; restJictive pulmona1y disease Abbreviations: A/C=assist-control; Flo2 = fraction of inspired oxygen; I: E= inspirationlexpiration; PEEP= positive endexpiratory pressure; PSV= pressure suppmt ventilation; T! = inspiratory time; VDIVT = dead space/tidal volume ratio; VE= minute ventilation; VI = inspiratory flow rate; VT= tidal volume

p ressure support ventilation (PSV), a r elatively new v entilatory mode, is a pressure assist form of mechanical ventilation designed to maintain a constant preset positive airway pressure during sponta*F ro m th e Intensive Care Unit, Hospital General de Requena (Drs. Tejeda and Afvarez), Requena, Valencia; Intensive Care Unit, Hospital Gran VIa (Dr. Boix) , Castellon; and Unidad d e Investigaci6n , H ospital Dr. Peset (Drs. Balanza and Morales), Valencia, Spain . Manuscript received March 1 , 1996; revision ac.-cepted October 22.

Reprint requests: Dr. Tejeda, Unidad de Cuidados Intensivos, Hospital General de Requena, Paraje Casablanca SIN 46340, Requena, Comunidad Valenciana, Spain 1322

neous inspiration.1-3 A key feature of this mode of ventilatory assistance is that it maintains and supports the patient's inspiratory effort. By definition, PSV is patient triggered and flow cycled. In addition to inspiratory time (TI), the patient controls the respiratory rate, inspiratory flow (VI), tidal volume (VT), and minute ventilation (VE ). Advantages proposed for PSV include decreased work of spontaneous breathing, more a dequate muscular training, better patient/ventilator synchrony, and improved subjective comfort when compared with conventional techniques of weaning. 4 However, to our Clinical Investigations in Critical Care

knowledge, the efficacy of PSV as an isolated ventilatory mode in patients with respiratory failure has not been established. Therefore, we assessed the short-term (4-h) hemodynamic and gas exchange effects of PSV to determine whether PSV would be an alternative ventilatory mode to assist-control (NC) ventilation in the treatment of respiratory failure.

MATERIALS AND METHODS Forty-five adult patients (32 men, 13 women) with respiratory failure requiring mechanical ventilatory support in the ICU were selected for study. Physiopathologic classification included respiratory failure caused by an exacerbation of previously documented COPD (n=5), respiratory failure due to neuromuscular disease without involvement of the pulmonary parenchyma (n = 13), and respiratory failure due to restrictive lung disease, including acute cardiogenic pulmonary edema (n=9), and acute lung injury in patients with pulmonary (n =6) and nonpulmonary causes (n= 12). All patients were receiving ventilatory support (Engstri:im-Erika; Gambro-Engstri:im AB; Sweden) in the AJC mode, had a Swan-Ganz catheter (Corodyn TD-E-N) in place, and evidenced hemodynamic stability. None of the patients could be weaned from mechanical ventilation. These patients had a mean (SD) age of 62.8 (11.8) years; a mean height of 166.6 (7.2) em; a mean weight of69.6 (11.9) kg; a mean APACHE II (acute physiology and chronic health evaluation) score of 15.1 (4.2); a mean length of ICU stay of 8.5 (10.9) days; a mean hemoglobin concentration of 10.8 (2.5) mgldL; a mean fraction of inspired oxygen (Flo 2 ) of 47 (14) %; a mean respiratory system compliance of 40.8 (11.9) Ucm H 2 0; a mean respiratory resistance of 10.9 (4.6) em H 2 0/Us; and a mean positive end-expiratory pressure (PEEP) of 1.1 (2.6) em H 2 0. The mean duration of mechanical ventilation before the study was 7.16 (8.64) days. The study protocol was approved by the institutional review board and informed consent was obtained from the families of all patients. The indication of A/C ventilation and the selection of ventilatmy settings were decided by the attending physician according to standard criteria accepted by consensus of the ICU staff prior to the study. Briefly, these included inspiratory pressure 60 Umin delivered in a square-wave pattern, (VT) 10 to 15 mUkg, respiratory frequency determined by the patient, and the minimal levels of Flo 2 and PEEP to obtain an arterial oxygen tension >65 mm Hg. Patients were switched to the PSV mode of the mechanical ventilator for a period of 4 h.During the first 30 min of PSV, the ventilator was set to obtained VT similar to that delive red in the AJC mode. The Flo 2 and the PEEP were constant for each patient during the study. PSV levels lower than 10 to 12 em H 2 0 were not used. An upper limit for the PSV level was not established, although levels >30 em H 2 0 were not delivered. Measurements

In the last 10 min of A/C ventilation and at the end of the PSV mode, the following were recorded: data on ventilatory parameters and respiratory mechanics; pulmonary gas exchange and arterial oxygenation; cardiovascular hemodynamics; and oxygen transport and oxygen consumption. Inspiratory flow rate (VI ), VT, respiratory frequency (f), VE, maximal airway pressure, and mean airway pressure were recorded from the expiratory readout of the ventilator. TI, inspiratory pause, and inspiration/expiration (I:E) ratio were measured at the proximal end of the orotracheal tube

using a differential pressure transducer (Combitrans; B Braun Medical AG; Melsungen, Germany). Expired C0 2 tension was also measured at the proximal end of the orotracheal tube using a C0 2 gas analyzer (Enstrom-Eliza). Radial arterial lines were placed and BP and heart rate were monitored using a disposable pressure transducer (Combitrans). Ide ntical transducers were used for measuring right atrium, pulmonary artety, and pulmonary capillary wedge pressures. Arterial and mixed venous blood specimens were analyzed in a blood gas analyzer (Corning 178; CIBA Corning Diagnostic Corp; Medfield, Mass). Cardiac output was determined in triplicate by thermodilution technique and measurements were obtained on a (Hewlett-Packard; Palo Alto, Calif) computer. Other parameters of gas exchange, arterial oxygenation, and oxygen transport were calculated using standard equations. Total body oxygen consumption was computed from the product of cardiac output and the difference in oxygen content between samples of arterial and mixed venous blood. All data are presented as mean (SD). Statistical analysis was performed using a paired t test. A p value of < 0.05 was considered significant.

RESULTS

The mean values of respiratory parameters are shown in Table l. Patients supported with PSV compared with NC ventilation showed significantly higher VT, VE, and TI in association with significantly lower pressure in the airway, VI, and I:E ratio. With regard to gas exchange data, an increase in dead space/tidal volume ratio (VDNT) (expressed in percent), decrease in Pa0 2 , and statistically but not clinically significant alteration of arterial oxygenation indexes were noted as patients were switched from NC ventilation to PSV (Table 2). However, when patients with COPD, restrictive disorders, and neuromuscular disease were compared, significant changes in arterial oxygenation parameters were noted only in patients with restrictive disorders (Table 3). Hemodynamic data are summarized in Table 4. There were significant decreases in heart rate, systolic pulmonary artery pressure, and pulmo-

Table !-Comparison of Ventilatory Parameters and Respiratory Mechanics Between A/C Ventilation and PSV* AIC, Mean

VT,mL f, breaths/min VE, mUrnin VI, Umin T!, s Pause, s Pmax, em H 2 0 Pm, em H 2 0 I:E ratio

(S D)

PSV, Mean (SD)

p Value

764.2 (101.5) 11.6 (2.7) 8,843 (2,145) 58.3 (3.9) 0.78 (0.08) 1.27 (0.09) 37.7 (7.7) 13.5 (4.0) 0.66 (0.16)

837.5 (235.9) 12.8 (5.5) (2,707) 9,978 30.8 (3.0) 1.63 (0.12) zero 22.8 (4.9) 8.9 (2.9) 0.53 (0.17)

0.034 NS 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 0.005

*Prnax =rnaximal ailway pressure; Pm =mean airway pressure; [=respiratory frequency; NS=not significant. CHEST I 111 I 51 MAY, 1997

1323

Table 2-Comparison of Pulmonary Gas Exchange and Arterial Oxygenation Between AIC Ventilation and PSV*

Arterial pH PaC0 2 , mm Hg PEC0 2 , mm Hg VDIVT ratio, % Pa02 , mm Hg Sa0 2 ,% Ca0 2 ,% Pa0 2/Flo2 P(A-a)0 2 , mm Hg Qs!Qt,% Bicarbonate, mmol!L Base excess, mmol!L

NC, Mean (SD)

PSV, Mean (SD )

p Valu e

7.39 (0.05) 40.6 (6.7) 4.61 (0.9) 0.18 (0.13) 120.8 (29.0) 97.7 (1.8) 14.7 (3.3) 276.9 (96.1 ) 164.4 (110.2) 0.18 (009) 24.8 (4.1) 0.5 (4.1 )

7.40 (006) 39.9 (7.3) 4.23 (0.8) 0.24 (0.17) 113.3 (28.5 ) 97.4 (2.2) 14.6 (3.3) 260.6 (93.5) 173.5 (107.7) 0.19 (0.10) 24.7 (4.2) 0.5 (4.2)

NS NS 0.0001 0.005 0.030 NS 0.023 0.042 0.016 0.048 NS NS

*PEC0 2 =expired C0 2 tension ; Sa0 2 =arterial oxygen saturation; Ca0 2 =concentration of oxygen in arterial blood; P(Aa)0 2 =alveolar-arterial oxygen difference; Qs/Qt =s hunt; NS=not significant.

nary capillary wedge pressure when PSV was applied. Oxygen transport and oxygen consumption were stable in the patient group (Table 5). Statistically significant differences among the three subgroups of patients with restrictive lung disease were not found.

DISCUSSION

PSV has been used as a mode of ventilation during a 4-h stable ventilatory support period in patients with respiratory failure of diffe rent etiology. The principal finding in this study was that PSV could maintain adequate ventilation and arterial oxygen-

Table 3-Comparison of Arterial Oxygenation in Patients With COPD, Restrictive Disorders, and Neuromuscular Disease Between A/C Ventilation and PSV* Pa0 2 , Sa02 , Qs/Qt, PaO/ P(A-a)0 2 , Flo 2 mm Hg mm Hg % % COPD

NC

PSV Neuromuscular disease

NC

PSV Restlictive disorders tVC PSV

81.8 83.6

94.8 94.0

0.26 0.29

163.8 168.0

242.8 238.2

131.7 132.0

98.6 98.6

0.11 0.11

355.4 363.5

83.8 85.9

123.1 t 110.6 1

97.9 97.5

0.18 0.20 1

263.5 1 233. 11

185.3 1 199.5 1

*See Table 2 footnote for explanatio n of' abbreviations. I tVC vs 1>SV, p < 0.05.

1324

Table 4-Comparison of Hemodynamic Data Between A/C Ventilation and PSV*

SBP, mm Hg DBP, mtTI Hg Mean BP, mm Hg Heart rate, beats/ min RAP, m111 Hg PAP, 111111 Hg Systolic Diastolic Mean PCWP, 111111 Hg

tVC , Mean (SD)

PSV, Mean (SD )

p Value

136.2 (26.7) 68.5 (173) 90.5 (209 ) 102.6 (20.1 ) 9.5 (2.5)

134.7 (281 ) 66.7 (13.0) 89.4 (18.3) 99.6 (18.5) 9.3 (2.9)

NS NS NS 0.014 NS

33.9 (6.7) 19.8 (2.7) 25.8 (3.4) 16.6 (2.8)

32.3 (7.9) 19.1 (3.2) 24.8 (4.1 ) 15.7 (3. 2)

0.018 NS NS 0.004

*SBP=systolic BP; DBP=diastolic BP; RAP = light atlium pressure; PAP = pulmonary artery pressure; PCWP = pulmonary capillary wedge pressure; NS=not significant.

ation at much lower levels of peak and mean airway pressure as compared with conventional A/C ventilation. With PSV, there is a greater VD/VT but maintaining an adequate ventilation with an increase in VE due to VT increases. In the PSV mode, the inspiratmy flow rate was smaller resulting in a better alveolar distribution of VT. 5 Moreover, while in the A/C mode, the mean duration of the inspiratory pause was 1.27 s (25% of the respiratory cycle); there is no inspiratory pause in the PSV mode, resulting in a decreased time available for redistribution of ventilation. 6 The increase in VDIVT ratio in the PSV mode, despite a lower VI, may be related to the absence of respiratory pause. As compared with A/C mode, there was a statistically significant reduction of arterial oxygenation in patients with restrictive disorders, presumably because of a greater ventilation/perfusion dysfunction. The presence of auto-PEEP was not assessed, although given the low

Table 5-Comparison of Oxygen Transport and Oxygen Consumption Data Between AIC Ventilation and PSV*

Cardiac output, Umin Cardiac index, Umin/111 2 D0 2 , mUmin/m2 Vo 2 , mUmin/m2 Pv0 2 , 111111 Hg Sv0 2 , 111111 Hg P(A-v)0 2 , mU% 0 2 extraction index, %

NC, Mean (SD )

PSV, Mean (SD)

p Value

6.74 (281) 3.65 (1.54) 554.8 (218 9) 132.5 (40.0 ) 42.6 (5.9) 74.5 (8.1) 3.64 (1.27) 0.25 (008)

6.66 (2.89) 3.64 (1.57) 542.0 (200) 126.0 (430) 42.0 (5.5) 74.2 (8.8) 3.60 (1.24) 0.25 (009)

NS NS NS NS NS NS NS NS

*D0 2 =oxygen delive1y; Vo2 = total body oxygen consumption; Sv0 2 = mixed venous oxygen saturation ; Pv0 2 = partial pressure of venous oxygen; P(A-v)0 2 =arterial-venous oxygen difference; NS = not significant. Clinical Investigations in Critical Care

respiratory frequencies and mean inspiratory/expiratmy relationships in both modes, it seems that auto-PEEP did not play a significant role in the variations of arterial oxygenation. Although in our study, the oxygen cost of breathing was not determined, total body oxygen consumption was equivalent in both modalities. PSV reduces the work of breathing roughly in proportion to the pressure delivered. 3 •7 Sufficient pressure level reduces completely the work of breathing. 8 ·9 In the study of Dries et al, 10 hemodynamic effects of PSV were assessed in cardiac surgery patients. Inclusion criteria included absence of documented preoperative pulmonary dysfunction. These authors described a tendency toward increase in heart rate, mean arterial pressure, central venous pressure, and pulmonary capillary wedge pressure at pressure support levels of 20 and 10 em H 2 0. Hemodynamic and oxygen transport parameters for PSV of 30 em H 2 0 were comparable to those recorded with intermittent mandatmy ventilation. In patients with preexisting pulmonary disease, we found statistically significant decreases in heart rate, systolic pulmonary artery pressure, and wedge pressure with PSV as compared with A/C ventilation. Cardiac output was unchanged, and these decreases were not clinically significant. These findings may be explained by complex pathophysiologic mechanisms during mechanical ventilatory support, including functional pressure-volume interrelationships with decreased right ventricular afterload 10 · 11 as well as by improved synchrony between the patient and the ventilatorY-14 In conclusion, PSV could be a possible alternative to A/C ventilation in patients with respiratory failure . PSV caused an increase in VDNT in association with a significantly lower airway pressure and I:E ratio. Neither PSV nor A/C ventilation may be appropriate as an initial mode in patients with abnormal central respiratory drive. Some of the stated advantages in circumstances similar to this study would be achieved by the newer generation of ventilators using pressure support type of flow, along with the advantages of still having backup support and guaranteed minimum VTs that would otherwise contraindicate the use of this modality in critically ill

patients. Much larger randomized studies are needed to define the long-term benefits of PSV and AIC modes and the conditions in which PSV may be a valuable alternative to AIC ventilation. ACKNOWLEDGMENT: We thank Marta Pulido, MD, for editing the manuscript and translating it into English. REFERENCES 1 Macintyre NR. Pressure support ventilation. Respir Care 1986; 31:189-90 2 Tokioka H, Saito S, Kosaka F. Comparison of pressure support ventilation and assist control ventilation in patients with acute respiratory failure. Intensive Care Med 1989; 15:364-67 3 Banner MJ, Kirby RR, Macintyre NR. Patient and ventilator work of breathing and ventilatory muscle loads at different levels of pressure support ventilation. Chest 1991; 100:531-33 4 Brochard L, Harf A, Lmino H, et al. Inspiratory pressure suppmt prevents diaphragmatic fatigue during weaning from mechanical ventilation. Am Rev Respir Dis 1989; 139:513-21 5 Connors AF Jr, McCaffree RD, Gray BA. Effect of inspiratory flow rate on gas exchange during mechanical ventilation. Am Rev Respir Dis 1981; 124:.537-43 6 Bergman NA. Effect of vaJ)'ing respiratory waveforms on distribution of inspired gas during artificial ventilation. Am Hev Hespir Dis 1969; 100:518-25 7 Tokioka H, Satito S, Takahashi T, et a!. Effectiveness of pressure support ventilation for mechanical ventilatory support in patients with status asthmaticus. Acta Anaesthesiol Scand 1992; 36:5-9 8 Macintyre NR, Leatherman NE. Ventilatory muscle loads and the frequency-tidal volume pattern during inspiratmy pressure-assisted (pressure-supported) ventilation. Am Rev Respir Dis 1990; 141:327-31 9 Macintyre NR. Respiratoq function during pressure support ventilation. Chest 1986; 89:677-83 10 Dries DJ, Kumar P, Mathru M, eta!. Hemodynamic effects of pressure support ventilation in cardiac surgeq patients. Am Surg 1991; 57:122-25 11 Hurst JM , Branson RD, Davis K Jr, eta!. Cardiopulmonaq effects of pressure support ventilation. Arch Surg 1989; 124:1067-70 12 Macintyre NR, Ho Ll. Effects of initial flow rate and breath termination criteria on pressure support ventilation. Chest 1991; 99:134-38 13 Cohen IL, Bilen Z, Krishnamurthy S. The effect of ventilator working pressure during pressure support ventilation. Chest 1993; 103:588-92 14 Tokioka H, Saito S, Kosaka F. Effect of pressure support ventilation on breathing pattern and respiratmy work. Intensive Care Med 1989; 15:491-94

CHEST/111/5/MAY,1997

1325