Cardiorespiratory Effects of Pressure Controlled Inverse Ratio Ventilation in Severe Respiratory Failure

clinical investigations in critical care Cardiorespiratory Effects of Pressure Controlled Inverse Ratio Ventilation in Severe Respiratory Failure* Edward Abraham, M.D ., F.C.C.P.; and Gary Yoshihara, R.R. T. Cardiorespiratory values were measured in nine patients with severe respiratory failure before and following initiation of pressure controlled inverse ratio ventilation (PCIRV) at an inspiratory to expiratory ratio of 2:1. All patients showed increases in PaO., with the mean Pa01 rising from 63 :t 4 (mean :t SEM) to 76 :t 8 mm Hg. Peak inspiratory pressure fell from 44 :t 4 to 39 :t 2 em U.O. There were no significant changes iii imy hemodynamic or oxygen metaboiism variable associated with the institution of PC-IRV. In particular, no significant alteration in cardiac index, pulmonary artery pressures, oxygen delivery, oxygen consumption, or oxygen extraction ratio occurred with the use

of PC-IRV. These results. suggest that PC-IRV may be a useful ventilatory modality in the treatment of severe respiratory failure since it results in improvement in arterial oxygenation without any deterioration in hemodynamic or tissue oxygen metabolism paratneters.

ressure controlled inverse ratio ventilation is a P recently describedl.2 ventilatory modality, in which

PEEP and increased mean airway pressures, its effects on hemodynamic variables have been studied in only a limited fashion. No changes were found 4 in mean arterial pressure, niean pulmonary artery pressure, or pulmonary capillary wedge pressure before and after institution of PC-iRV. The interaction between PC-IRV and cardiac output or parameters of tissue oxygen metabolism has not been reported. We initi· ated the present prospective study to better define the physiologic effects of PC-IRV on hemodynamic and cardiorespiratory variables.

the conventional inspiratory to expiratory (I:E) ratio is reversed, with the inspiratory phase becoming two to four times as long as the expiratory period. The PC-IRV has been reported3 to achieve improved oxygenation at lower peak airway pressures. Other advantages include lower minute volume and decreased levels of positive end expiratory pressure. 4 In PC-IRY, pressure control is used to change the inspiratory flow pattern so that each breath is initiated befbre expiratory flow from the previous breath reaches zero. A physiologic result of this ventilatory pattern is maintenance of end-expiratory pressures. 5 The prolonged inspiratory phase, coupled with positive end-expiratory pressure, usually results in de· creased peak inspiratory and increased mean airway pressures in patients receiving PC-IRV. 4 Respiratory modalities, such as PEEP, which result in increased peak and mean airway pressures have significant effects on hemodynamic and tissue oxygen metabolism parameters. 6 With high levels of PEEP, although Pa02 is improved, cardiac output may be severely compromised, resulting in marked decreases in tissue Do 2 , deleteriously affecting tissue oxygen metabolism. Although PC-IRV is associated with auto· *From the Division of Pulmonary and Critical Care Medicine, Department of Medicine, UCLA Medical Center, Los Angeles. Manuscript received March 31; revision accepted May 17.

1358

(Clam 1989; 96:1356-59) PCJ-RV =pressure controlled inverse ratio ventilation; CI=cardiaC index; Do.= oxygen delivery; Vo1 =oxygen consumption; O,En =oxygen eidraction ratios; I:E =inspiratory to expiratory; SVRI =systemic vascular resistance; PVRI::: pulmonary vascular resistance; Ca01 =arterial oxygen content; c-vO. = inixed venous oxygen cOntent

METHODS

lbtients Nine patients (Table 1) with severe ARDS, as manifested by diffuse pulmonary infiltrates on chest roentgenograms, arterial hypoxemia with widened A-a gradients despite supplemental oxygen, pulmonary capillary wedge pressures (WP) less than 20 mm Hg, and decreased. static and dynamic thoracic compliance were entered in the study. In each case, the patient was placed on PCIRV at the request of the attending physician, who judged the patient to be failing conventional volume controlled ventilation with conventional ratios (VC-CRV). In all cases, a flow-directed pulmonary artery (Swan-Ganz) catheter with a fiberoptic channel for continuous measurement of mixed venous oxygen saturation had been placed previously for hemodynllmic and cardiorespiratory monitoring. The mixed venous oxygen saturation determination from the pulmonary artery catheter was calibrated and verified by using a mixed venous blood sample in which the oxygen saturation was measured by COoximetry. All patients had indwelling arterial catheters and pulse oximeters. All patients previously had been placed on a Servo-controlled ventilator, operating in the volume 111\WM Ratio Ventilation In Sewn~ Reepiratory Fallunt (Abnlham, ro.hlhara)

Table 2- Ventilatory lbrameten

Table 1-fbtimt Claar~ Patient

Age/Sex

Major Clinical Problems

1

44/F

2

42/M

Liver failure, Gl bleed, hepatorenal syndrome Chronic myelogenous leukemia, post bone marrow transplantation, CMV pneumonia GI bleed, renal failure, aspiration pneumonia Myelofibrosis, sepsis, pulmonary hemorrhage SLE, Pneumocystis carlnU pneumonia, hypertension Acute lymphocytic leukemia, DIC, renal failure, sepsis Chronic myelogenous leukemia, post bone marrow transplantation, CMV pneumonia Acute myelogenous leukemia, post bone marrow transplantation,

3

61/M

4

67/F

5

3.liF

6

40IM

7

341M

8

46IF

PEEP (em H,O)

Flo1

15

1.0

15

1.0

12

1.0

10

1.0

12

0.8

6

0.7

14

1.0

7.5

0.6

52/F

sepsis Pneumoccocal sepsis, SLE

17

0.9

controlled ventilation mode. Prior to initiation of the PC-IRV trial, all patients were sedated with appropriate doses ofbenzodiazepines and paralyzed with vecuronium or atracurium by continuous intravenous infusion, after an initial bolus dose.

PC-IRV7Hal In each patient, immediately prior to initiation of the PC-IRV trial, a full set of hemodynamic and cardiorespiratory variables was measured. This included measurement of arterial systolic and diastolic pressure, heart rate, right atrial pressure, pulmonary artery systolic and diastolic pressure, and pulmonary capillary wedge pressure. Cardiac output was measured in triplicate by the thermodilution technique at end expiration. The timing of injection for cardiac output measurement was supervised by one of the authors (E.A.) to verify that the injection was initiated at the same point in the respiratory cycle. Determination of arterial blood gases (Pa01 , PaCO,, pH), arterial oxygen saturation {SaO.). calculated from the arterial blood gas and measured directly with the pulse oximeter, and mixed venous oxygen saturation (SVO.) were made. Inspired oxygen concentration (Flo.), PEEP, and peak airway pressure during inspiration were measured. The patient then was placed on PC-IRV, with an l :E ratio of2:1, using 67 percent inspiratory time and zero percent pause. The respiratory rate was adjusted to keep the end-expiratory pressure at the same level as the PEEP utilized during volume controlled ventilation. A strip recorder was interfaced with the ventilator to verify that the above goals had been achieved and that zero expiratory flow was not reached prior to the triggering of the next

breath.

Before PC-IRV 63±4 After PC-IRV 76±8*

46±5 39±6

pH 7.37±.05 7.42±.04

PIP, SaO,, % cmH,O 93±2 94±2

44±4 39±2

*p<0.05 vs value before PC-IRV. After a 30-minute stabilization period on PC-IRV with an l:E ratio of 2:1, another full set of cardiorespiratory parameters, as described above, was obtained. The Flo, and PIP again were measured. All of the PC-IRV trials were able to be continued at least for the 30-minute measurement period. In patients with improved oxygenation and without hemodynamic compromise, the PC-IRV was continued for periods as long as 96 hours.

Data Analysis Derived cardiorespiratory variables, systemic vascular resistance. pulmonary vascular resistance, arterial and mixed venous oxygen content, oxygen delivery, oxygen consumption, and oxygen extraction ratio were calculated using standard and previously described7 formulae. Where appropriate, cardiorespiratory and hemodynamic variables were normalized, using the calculated body surface area for each patient. Mean± standard error of the mean was calculated for each of the cardiorespiratory variables, both before and after institution of PCIRV. Comparison between values before and after initiation of PCIRV was performed by a paired Student's t-test. Differences were considered to be significant for p<0.05. RESULTS

Enterobacter cloaceae

9

PaO,, PaCO,, mmHg mmHg

Table 1 summarizes the characteristics of the patient population studied. Four men and five women were included. The average age was 46±4 years. Arterial blood gas values, arterial oxygen saturation and peak inspiratory pressures before and after initiation of PC-IRV are presented in Table 2. All patients showed an increase in Pa02 after the institution of PC-IRV, with the range of improvement in Pa02 being 1 to 18 mm Hg. The PaC02 decreased in seven patients and pH rose in six patients with initiation of PC-IRV. The PIP fell in eight of the nine patients after inverse ratio ventilation was begun. For the entire patient group, the changes in PaC02 and PIP after Table 3- Cardiornpiratorylbrameten

MAP(mm Hg) HR {beats/min) CI{UminiM.) SVRI (dynesosocm-•tm•) CVP(mm Hg) WP(mm Hg) PAS(mm Hg) PAD(mmHg) PVRI (dynesosocm-•tm•) Ca01 (ml) c-v01 (ml) Do1 (mVmin/m1) Vo1 (mVmin!m•) o.En(%)

Before PC-IRV

After PC-IRV

78±6 119±7 4.8±.7 1154± 181 12:!:2 13± 1 44:!:4 23±3 376:!:66 12.7:!:.5 8.7:!: .4 577± 102 159:!:22 30.6±3.8

75±4 111±5 4.9:!: .7 1114±175 14±2 14:!: 1 45±4 24±3 326±59 12.6± .6 9.1:!: .5 544±129 157:!:27 28.6±3.5

CHEST /96 /6 I DECEMBER, 1989

1357

beginning PC-IRV were not statistically significant. Minute ventilation did not significantly change after initiation of inverse ratio ventilation, 16.5 ± 4.1 Umin before PC-IRV and 18.2 ± 4.1 Umin after starting PCIRV. Cardiorespiratory parameters in the pre- and postPC-IRV periods are shown in Table 3. In the control period, when the patients were ventilated with conventional I:E ratios, they demonstrated a relative hyperdynamic state, with increased HR, CI, Do2 , 0 2Ext, and Vo 2 • Pulmonary artery pressures and PVRI were elevated. Following the initiation of PCIRV, there were no significant changes in the mean values for any of the cardiorespiratory parameters. In three patients, changes of more than 10 percent in MAP, CI, Do2 , or Vo 2 accompanied the institution of PC-IRV. In patient 4, the MAP decreased from 76 to 63 mm Hg. This fall in blood pressure was accompanied by greater than 30 percent decreases in cardiac output (from 4.0 to 2. 7Umin/m 2), CVP (from 16 to 10 mm Hg), WP (from 18 to 10 mm Hg), Do2 (from 396 to 273 mVminlm 2), 0 2Ext (from 28.3 to 21.3 percent), and Vo2 (from 84 to 59 mVmin/m 2). In this patient, SVRI was found to increase by 31 percent (from 1200 to 1570 dynes-seem - 5Im 2). The other two patients showing greater than 10 percent changes in cardiorespiratory values after initiation of PC-IRV had increases in CI and Do2 , accompanied by a decrease in SVRI, but only minimal alteration in MAP. In patient 7, CI increased by 20 percent (from 2.5 to 3.0 Umin!mJ, and Do2 by 25 percent (from 353 to 440 mVmin/m 2), while SVRI fell by 14 percent (from 789 to 676 dynes-seem - 5Im 2). In patient 8, greater than 20 percent increases were found for CI (3.6 to 4.6 Umin/m 2), Do2 (486 to 621 mVminlm 2), CVP (14 to 20 mm Hg), WP (12 to 15 mm Hg), and Vo2 (216 to 276 mVmin/m 2). In this patient, both SVRI (from 1111 to 852 dynes•seem- 5/m 2) and PVRI (from 356 to 273 dynes-seem - 5Im 2) fell with the institution of PC-IRV. DISCUSSION

In the present study, increases in Pa02 and decreases in PIP were found after the initiation ofPC-IRV. These results are similar to those found in previous investigations3 •4 of PC-IRV. The changes found in PaC02 and pH were minimal and probably could have been eliminated had the end-expiratory pressure been decreased, as was made possible by the improvement in oxygenation accompanying the use of PC-IRV. Because PC-IRV has been demonstrated to result in improved Pa0 2 and decreased PIP in patients with ARDS, it has been proposed as a useful ventilatory modality for severe respiratory failure. Although no clear improvement in patient outcome has yet been shown with the use of PC-IRV, several theoretic 1358

benefits are associated with its physiologic effects. Improved oxygenation as a result ofPC-IRV in patients with ARDS will permit use of lower inspired concentrations of oxygen, minimizing pulmonary toxicity associated with high Flo2 •8 High PIP is thought to contribute to barotrauma and pulmonary injury through the generation of elevated alveolar shear forces. 9 •10 Maintenance or improvement of oxygenation with PC-IRV at lower levels of PIP may decrease the incidence of these complications of mechanical ventilation. Although several studies, 3 •4 •11 including the present one, have found improvement in oxygenation associated with the use of PC-IRV, the physiologic mechanism(s) responsible for this effect is not well delineated. Increased functional residual capacity has been suggested 12 as one reason for the effects of PEEP on oxygenation. A similar increase in FRC may occur with PC-IRV, coincident with the increase in mean airway pressures that often accompanies use of this modality. In one study, 3 external end-expiratory volume was increased by an average of 1,200 ml with PC-IRV, indicating that increased FRC probably is seen with PC-IRV. An initially high inspiratory Bow rate, followed by a rapidly decelerating flow pattern, and a lengthened expiratory time constant accompany the use of PCIRV. This rapidly decelerating Bow pattern has been shown 13 to produce an improvement in oxygenation in patients with severe respiratory failure and may play a role in the changes in Pa02 found with PC-IRV. In addition, the long expiratory Bow constant may also improve oxygenation through achieving a more homogenous distribution of ventilation to previously underventilated lung units in patients with severe respiratory failure . 14 In the present study, there were no significant changes found in any cardiorespiratory parameters after the initiation ofPC-IRV at an I:E ratio of2:1. In particular, CI, Do2 , and Vo2 remained unaltered by the use ofPC-IRV. These results demonstrate that the improvement in oxygenation which accompanies use of PC-IRV at this I:E ratio is not associated with any deleterious effects on cardiac function or tissue oxygen delivery. Because higher I:E ratios were not utilized in this study, it is not possible to be sure that cardiorespiratory parameters would remain unchanged as the I:E ratio is further increased. Despite the improvement in Pa02 which occurred when PC-IRV was used in our patients, there was little change in either Ca02 or Do2 • This lack of alteration in Ca02 with PC-IRV was anticipated, since the patients studied had adequate oxygen saturation on conventional volume controlled ventilation, through the use of high Flo2 and PEEP, before being started on PC-IRV. Improvements in Sa02 and in Ca02 Inverse Ratio Ventilation in Sewre Respiratory Failure (Abraham, Yoshihara)

with initiation of PC-IRV, therefore, were expected to be minimal. Nevertheless, the increase in Pa02 associated with use of PC-IRV may be important in the management of the patients with severe respiratory failure since it permits further decreases in Fio2 and/ or PEEP, limiting risk of barotrauma and oxygeninduced pulmonary toxicity, while allowing maintenance of Ca02 and Do2 • Hemodynamic values that would affect pulmonary mechanics directly, through their effects on extravascular lung water or pulmonary vascular pressures, also showed no significant changes with use of PC-IRV. In particular, WP, PVR, and pulmonary artery pressures did not demonstrate any alteration with PC-IRV. Starling forces, associated with formation of extravascular lung water and pulmonary edema, therefore, were little changed with PC-IRV, and no directly detrimental effects on pulmonary hemodynamics or fluid transport patterns would be expected to accompany PC-IRV. Of the nine patients studied, only one showed significant hemodynamic deterioration associated with the initiation of PC-IRV. There were no preexistent physiologic factors which could have served to separate this patient from the others studied. In particular, while on conventional ventilation with a PEEP 10, this patient had a CI of 4 Uminlm 2 with elevated WP (18 mm Hg) and CVP (16 mm Hg). After PC-IRV was started, CI fell33 percent, accompanied by decreases in WP and CVP, suggesting that decreased cardiac filling with progression down a Frank-Starling filling curve occurred. The deterioration in this patient, although it did not cause termination of the PC-IRV trial, points out that PC-IRV cannot be used indiscriminately in acute respiratory failure , and probably should not be initiated without invasive cardiorespiratory monitoring that permits close observation of CI, Do2 , and other cardiorespiratory parameters. In addition, because sedation and paralysis, required to achieve the longer inspiratory times used in PC-IRV, increase the complexity of patient care, patients must be carefully selected and monitored for PC-IRV trials. In this study, PC-IRV was associated with improvement in Pa02 , decrease in peak airway pressures, and no significant overall changes in CI, tissue oxygen metabolism (Do2 , Vo 2 , 0 2Ext), or hemodynamic parameters. These results indicate that this respiratory

modality may be useful in patients with severe respiratory failure, since PC-IRV may permit decreased Flo2 and .Pif, with no significant cardiorespiratory effects. Further studies will be necessary to define the benefits, primarily in terms of survival, associated with the use ofPC-IRV. REFERENCES

1 Ravizza AG, Caroga D, Cerchiari EL, Ferrante M, Puppa TD, Villa L. In versed ratio and conventional ventilations: comparison of the respiratory effects. Anesthesiology 1983; 59:523 2 Cole AG, ~ller SF, Sykes MK. Inverse ratio ventilation compared with PEEP in adult respiratory failure . Intensive Care Med 1984; 10:227-32 3 Gurevitch MJ, Van Dyke J, Young ES, Jackson K. Improved oxygenation and lower peak airway pressure in severe adult respiratory distress syndrome. Chest 1986; 89:211-13 4 Tharratt RS, Allen RP, Albertson TE. Pressure controlled inverse ratio ventilation in severe adult respiratory failure. Chest 1988; 94:755-62 5 Lachmann B, Danzmann E, Haendly B, Johnson B. Ventilator settings and gas exchange in respiratory distress syndromes. In: Prakash 0 , ed. Applied physiology in clinical respiratory care. Boston: Martinus Nijhoff, 1982; 141-76 6 Tyler DC. Positive end-expiratory pressure: a review. Crit Care Med 1983; 11:300-08 7 Abraham E, Shoemaker WC. Sequential cardiorespiratory patterns associated with outcome in septic shock. Chest 1984;

85:75-80 8 Deneke SM , Fanburg BL. Normobaric oxygen toxicity of the lung. N Eng! J Med 1980; 303:76-86 9 Haake R, Schlichtig R, Ulstad DR, Henschen RR. Barotrauma. Chest 1987; 91:608-13 10 Kolobow T, Moretti MP, Fumagelli R, Mascheroni D, Prato P, Chen V, et al. Severe impairment in lung function induced by high peak airway pressure during mechanical ventilation. Am Rev Respir Dis 1987; 135:312-15 11 Lachmann B, Haendly B, Schultz H , Jonson B. Improved oxygenation, C01 elimination, compliance and decreased barotrauma following changes of volume-generated PEEP ventilation with inspiratory/expiratory (liE) ratio of 1:2 to pressure-generated ventilation with liE ratio of 4:1 in patients with severe adult respiratory distress syndrome (ARDS). Intensive Care Med 1980; 6:64 12 Kirby RR, Downs JB, Civetta JM, Modell JH, Cannemiller FJ, Klein EF, et al. High level positive end expiratory pressure (PEEP) in acute respiratory insufficiency. Chest 1975; 67:156-

63 13 AI Saady N, Bennett ED. Decelerating inspiratory Bow waveform improves lung mechanics and gas exchange in patients on intermittent positive-pressure ventilation. Intensive Care Med

1985; 11:68-75 14 Connors AF, McCaffree DR, Gray BA. Effect of inspiratory Bow rate on gas exchange during mechanical ventilation. Am Rev Respir Dis 1981; 124:537-43

CHEST /96 16 I DECEMBER,1989

1351