The effects of positive end-expiratory pressure on respiratory system mechanics and hemodynamics in postoperative cardiac surgery patients

We prospectively evaluated the effects of positive end-expiratory pressure (PEEP) on the respiratory mechanical properties and hemodynamics of 10 postoperative adult cardiac patients undergoing mechanical ventilation while still anesthetized and paralyzed. The respiratory mechanics was evaluated by the inflation inspiratory occlusion method and hemodynamics by conventional methods. Each patient was randomized to a different level of PEEP (5, 10 and 15 cmH2O), while zero end-expiratory pressure (ZEEP) was established as control. PEEP of 15-min duration was applied at 20-min intervals. The frequency dependence of resistance and the viscoelastic properties and elastance of the respiratory system were evaluated together with hemodynamic and respiratory indexes. We observed a significant decrease in total airway resistance (13.12 ± 0.79 cmH2O l-1 s-1 at ZEEP, 11.94 ± 0.55 cmH2O l-1 s-1 (P<0.0197) at 5 cmH2O of PEEP, 11.42 ± 0.71 cmH2O l-1 s-1 (P<0.0255) at 10 cmH2O of PEEP, and 10.32 ± 0.57 cmH2O l-1 s-1 (P<0.0002) at 15 cmH2O of PEEP). The elastance (Ers; cmH2O/l) was not significantly modified by PEEP from zero (23.49 ± 1.21) to 5 cmH2O (21.89 ± 0.70). However, a significant decrease (P<0.0003) at 10 cmH2O PEEP (18.86 ± 1.13), as well as (P<0.0001) at 15 cmH2O (18.41 ± 0.82) was observed after PEEP application. Volume dependence of viscoelastic properties showed a slight but not significant tendency to increase with PEEP. The significant decreases in cardiac index (l min-1 m-2) due to PEEP increments (3.90 ± 0.22 at ZEEP, 3.43 ± 0.17 (P<0.0260) at 5 cmH2O of PEEP, 3.31 ± 0.22 (P<0.0260) at 10 cmH2O of PEEP, and 3.10 ± 0.22 (P<0.0113) at 15 cmH2O of PEEP) were compensated for by an increase in arterial oxygen content owing to shunt fraction reduction (%) from 22.26 ± 2.28 at ZEEP to 11.66 ± 1.24 at PEEP of 15 cmH2O (P<0.0007). We conclude that increments in PEEP resulted in a reduction of both airway resistance and respiratory elastance. These results could reflect improvement in respiratory mechanics. However, due to possible hemodynamic instability, PEEP should be carefully applied to postoperative cardiac patients. Correspondence


Introduction
Pulmonary dysfunction is a significant cause of postoperative morbidity following open-heart surgery (1,2).The underlying causes seem to be multifactorial, including effects of anesthesia and muscle paralysis, sternotomy, inflammatory reactions due to extracorporeal circulation, increase in extravascular lung water, alveolar collapse and altered chest wall mechanics (3,4).Atelectasis and hypoxemia are the main clinical findings.Atelectasis seems to be caused by reduced lung volume and small airway collapse (5,6).Hypoxemia may reflect increased intrapulmonary shunt due to collapsed lung areas and/or altered ventilation-perfusion ratio (3).To reopen atelectatic lung units and improve arterial oxygenation, different levels of positive end-expiratory pressure (PEEP) have been proposed, most of them based on improvement of the oxygenation index (7,8).To our knowledge, few studies of respiratory mechanical properties are available about postoperative cardiac patients, submitted to PEEP to reopen collapsed lungs (9).Nevertheless, several investigators have shown that PEEP improves arterial oxygenation, although there is controversy regarding the proper level of PEEP to obtain alveolar opening and stabilization, causing minimal hemodynamic instability and barotrauma (10).It is also important to note that postoperative cardiac patients frequently may present severe hypoxemia, due to atelectasis that requires an alveolar recruitment strategy, employing high levels of PEEP.Simple bedside radiography is considered a poor method to detect postoperative atelectasis (6).Therefore, PEEP should be routinely employed in maneuvers for alveolar opening in the presence of hypoxemia.However, even though PEEP is recommended for alveolar reopening, the presence of hemodynamic instability can be further worsened by elevated inspiratory pressures and PEEP.In the present investigation, we studied the effects of dif-ferent levels of PEEP on respiratory system mechanical properties and oxygenation indexes, as well as its influence on the cardiovascular system in postoperative cardiac surgery patients.

Material and Methods
After approval by the Institutional Ethics Committee, 10 patients (7 males) gave informed written consent to participate in this study.Mean age was 52.3 years (37 to 59) and mean weight 66.7 kg (54 to 87).All patients underwent coronary bypass graft and were studied consecutively during the immediate postoperative period while still under the effects of anesthesia.Surgical procedure and anesthesia were standardized (midazolam, fentanyl and pancuronium bromide), and all patients were submitted to cardiopulmonary bypass with a membrane oxygenator.Inclusion criteria were: previously normal ejection fraction, absence of chronic lung disease, bypass time of 90 to 120 min, satisfactory circulation (absence of vasoactive drugs) and adequate gas exchange parameters upon admission to the intensive care unit.

Respiratory mechanical data
The patients were anesthetized and paralyzed and initially ventilated with a constant flow ventilator (Bear 5, Bear Medical Systems, Riverside, CA, USA) using an FIO 2 of 0.6, tidal volume (V t ) of 8 to 10 ml/kg, inspiratory square-wave flow of 1 l/s and respiratory frequency of 10 cycles/min.The end inflation occlusion method was used to measure the resistive and elastic properties of the respiratory system (11).Respiratory elastance (E rs ) was computed by dividing P el,rs by the tidal volume (12).Care was taken to avoid leaks in the system.Airflow (V) changes in lung volume (V l ) and tracheal pressure (P tr ) were obtained directly from the ventilator and stored on an IBM personal computer (IBM PC, IBM Computers, São Paulo, SP, Brazil) through a 12-bit analog to digital converter (DT 2801 A , Data Transition, Marlboro, MA, USA) at a sampling frequency of 200 Hz (13).V t was obtained by digital integration of the flow signal.The accuracy of the flow signal provided was tested by comparing different volume values from the ventilator (electronically integrated flow signal) with volumes simultaneously measured with a dry spirometer within the V t range.The transducer that provided P tr was tested by applying 5-s PEEP plateaux (5, 10, 15 and 20 cmH 2 O) during the ventilation of a rubber balloon.The values observed on the electronic display were compared with those measured with a calibrated pressure transducer (270, Hewlett-Packard, Waltham, MA, USA).The flow resistive properties of the equipment (endotracheal tube plus connectors) were experimentally calculated (14) and subtracted from the obtained resistance.A power function fitting this relationship was determined and used to calculate the resistive pressures dissipated along the equipment at any given flow during the tests (13).The equipment resistance was subtracted whenever necessary, so that the results reported here represent intrinsic resistance values.Expiratory flow became nil before the end of expiration and the onset of inspiratory flow was synchronous with the beginning of the positive pressure, showing that auto-PEEP was not present in any patient.
The end inflation occlusion method consists of inflating the relaxed respiratory system with a constant square wave flow provided by a ventilator, followed by a rapid airway occlusion at end inspiration, which is maintained until a plateau in tracheal pressure is achieved.In the present study, a rapid airway occlusion was performed during a constant inspiratory flow and was held for 2s of inspiratory pause.Initially, there is a fast drop in tracheal pressure (DP 1 ) from the peak airway pressure (P 0 or P trmax ) to a deflection in the pressure curve (P 1 ), followed by a slower decay (DP 2 ) until an apparent plateau is reached, which represents respiratory system static elastic recoil pressure (P 2 or P el,rs ) (Figure 1).DP 1 corresponds to the pressure loss across the airways, with some contribution of rapid resistive components of the chest wall, where DP 2 represents pressure dissipation due to lung and chest wall viscoelastic properties.(DP 1 + DP 2 ) divided by the previous inspiratory flow gives the total resistance of the respiratory system (R rsmax ).
DP 1 divided by the preceding flow gives minimal value of resistance (R init ), which is due mainly to airway resistive properties.
DP 2 divided by the inspiratory flow immedi- ately preceding airway occlusion indicates tissue initial resistance (R diff ) or (DP 2 ).Bates et al. (11) propose two values for respiratory system resistance: one that would be obtained in the absence of unequal time constants within the system and that is not affected by stress relaxation, corresponding to R init (DP 1 ); the other reflects the mechanical unevenness within the system and stress relaxation R diff (DP 2 ).The overall R rsmax corresponds to addition of R diff and R init (15).To avoid auto-PEEP each maneuver was performed at the bedside by allowing a complete expiration to zero end-expiratory pressure (ZEEP) at each step of increasing PEEP, and observing the straight part of the P-V curve.

Study design
After admission to the ICU, the patients had their volemic status adjusted, maintaining an end point of PCWP of 6 mmHg and RAP of 4 mmHg throughout the study.During data acquisition the patients were randomized and remained anesthetized and paralyzed.Additional doses of fentanyl, midazolam and pancuronium bromide were given whenever necessary.Respiratory mechanics and hemodynamics were measured after 15 min of application of three different PEEP levels: 5, 10 and 15 cmH 2 O including ZEEP.A pause of 20 min was allowed between each PEEP application period, when the ven-tilatory parameters returned to initial values.In order to avoid interference during the measurement period, neither end expiratory nor sustained inflation was used during or between measurement periods.In each set of measurements, including ZEEP, 7 to 10 breath cycles were pooled and averaged to provide one data point.In order to remove time from the end of surgery as an influencing variable, the sequence of PEEP level application was randomized.

Statistical method
Data were analyzed statistically by analysis of variance (ANOVA) for repeated measures followed by the Tukey test to determine the differences between the established study points within groups, and by the Student ttest to determine differences between groups.The level of significance was set at P<0.05.

Results
All patients had an uncomplicated clinical course and were discharged from the ICU within 48 h after admission.No patient required re-operation for bleeding or for any other cause.The results are shown in Tables 1 to 4.
A progressive reduction of the total resistance of respiratory system along with PEEP increment was observed.Values of R rsmax decreased significantly from 13.12 ± 0.79

Hemodynamic data
As demonstrated below, the augmentation of intra-thoracic pressure caused by PEEP application determined a depression of the cardiac function expressed by a decrease of cardiac output and derived indexes (Table 2).The elevation of atrial filling pressures as pulmonary arterial pressure also represents the effects of intra-thoracic distending pressure.Cardiac index (1 min -1 m -2 ) showed a significant and progressive decrease from 3.90 ± 0.

Respiratory data
As can be seen in Table 3

Discussion
The usefulness of PEEP throughout the respiratory cycle for the correction of hy-poxemia caused by acute respiratory failure has been clinically demonstrated since 1967 (16).Since then, several studies have been performed in order to establish the ideal PEEP that could restore the oxygenation with minimal impairment of oxygen delivery (4,5,17).In order to minimize pulmonary and circulatory negative influences, many authors recommend adjusting PEEP values according to respiratory mechanical properties (3,18,19).
Recently, several methods have been proposed to assess respiratory mechanical properties in artificially ventilated patients (18,20).Respiratory compliance or elastance and airway resistance can be readily calculated using the flow and pressure transducers incorporated into modern mechanical ventilators, if the raw signals can be connected to an external recording device (21).Among these, the constant flow inflation method has been employed by many authors (12,13,22,23).This technique was developed by Bates et al. (11), who reexamined an early analysis by Rattenborg and Holaday (24) of the behavior of the multicompartmental model of the respiratory system.Utilizing the end inflation occlusion method it is possible to measure the elasticity, resistance and its subcomponents of respiratory system.
In the present study a significant change was observed in pulmonary elastance or compliance at PEEP levels of 10 and 15 cmH 2 O.The decrease in respiratory elastance with PEEP can be explained by a supplementary alveolar recruitment.Normal airway resistance is of the order of 2.5 cmH 2 O l -1 s -1 and significant increases have been described in patients with chronic air flow limitation (26.4 cmH 2 O l -1 s -1 ) (21).The increase in airway resistance observed after cardiac surgery may reflect airway wall edema, presence of fluid or secretions within the airway lumen as well as losses of functional lung volume.
Previously we described variations of the overall R rsmax and its airway (DP 1 ) and viscoelastic (DP 2 ) components in patients immediately before and after cardiac surgery (13).There are few studies comparing R rsmax , DP 1 and DP 2 at different PEEP levels during the postoperative period after cardiac surgery.The significance of R init ((P trmax -P 1 )/ flow) or (DP 1 ) has only recently been clarified in human beings, as essentially representing airway resistance (24).As described above, the difference between P 1 and P 2 (i.e., DP 2 ) represents the slow postocclusion de- cay in tracheal pressure and may reflect stress relaxation due to the viscoelastic properties of the respiratory system and possibly the Pendelluft phenomenon that represents distribution of air among the different lung regions (11,(25)(26)(27).In normal subjects, Pendelluft probably has a relatively small role, however, this phenomenon may be more evident if there were an increase in time constant inhomogeneities of alveolar inflation and deflation within the lung.It is important to emphasize that R rsmax corresponds to the effective resistance at zero respiratory frequency, while DP 1 reflects the resistance at high frequency (11).Therefore, DP 2 , that reflects tissue viscance, is a measure of the frequency dependence of resistance, a considerably important clinical parameter (28).
It is important to emphasize that R rsmax is always greater than R init (DP 1 ).In chronic or acute lung diseases this difference tends to be higher owing to R diff (DP 2 ) elevation (20).
In the present study we observed a significant decrease in R rsmax due to a significant decrease in DP 1 , with increasing levels of PEEP.This effect can be explained by a probable increase in airway radius due to radial forces applied by the alveolar parenchyma on the airway wall.These data confirm a basic principle of respiratory mechanics, i.e., that flow resistance decreases with increasing lung volume, as determined by changes in DP 1 (29,30).The decrease in DP 1 and the nonsignificant effect of PEEP on DP 2 lead to a decrease in R rsmax .These findings in humans are in contrast with those observed in cats, since it was demonstrated in this species that R rsmax increases with lung volume (30) because of the increase in parenchyma viscoelastic losses.On the other hand, at fixed tidal volume, R rsmax could decrease with increased flow as demonstrated in adult respiratory distress syndrome patients (23).
Although DP 2 increased with increasing PEEP, this variation was not considered to be statistically significant.As previously mentioned, DP 2 reflects Pendelluft phe- nomena and stress relaxation; however, it is difficult to say which of the two was more prominent in the present study.The term stress relaxation used in this text refers to the phenomena of lung accommodation to positive intra-alveolar pressure, probably caused by the interstitial lung fibrillar matrix realignment and/or by a decrease of the forces generated by superficial tension.A stress relaxation increase is normally related to tissue edema and/or collapsed alveoli, a common situation after cardiac surgery, which is resolved by the alveolar recruitment effect of PEEP.On the other hand, DP In this study there was a significant increase in atrial filling pressure as well as in pulmonary vascular resistance, an effect which is probably related to increased intrathoracic and pleural pressure observed with increasing PEEP.As demonstrated by other studies at 10 cmH 2 O or higher, PCWP overestimates left ventricular filling pressure, even if the occluded catheter tip is positioned in the inferior regions of the lung (31,32).One of the critiques of our study is the lack of any correction for increased pleural pressure during PEEP application and making hemodynamic measurements with the patient connected to the ventilator.As expected, we observed a significant variation in right and left pressures mainly at 15 cmH 2 O of PEEP.On the right side of the heart, this fact is probably caused by an obstruction-like effect in venous return (31) on the left; the PCWP elevation is probably due to an artifact transmitted during lung distention which is more evident with increasing PEEP (33).Lozman et al. (34) observed a good correlation between PCWP and left atrium pressure at low levels of PEEP after cardiac surgery; however, at 10 and 15 cmH 2 O of PEEP, a discrepancy was observed between PCWP and left atrium pressure and no statistical correlation was found between these values.In their study, Pinsky et al. (31) found that in postoperative patients PCWP does not reflect or change with PEEP.However, PCWP reflects transmural pressure (left pressure) only during low levels of PEEP (<5 cmH 2 O) whereas nadir PEEP obtained by abrupt airway disconnection accurately reflects transmural pressure to at least 15 cmH 2 O.To minimize a possible interference with respiratory mechanical data acquisition, we did not disconnect our patients from the ventilator to obtain atrial filling pressures.In contrast, Van den Berg et al. (35) employed sustained inspiratory hold maneuver (24 s) in postoperative coronary bypass surgery patients to determine the dynamic changes in right and left ventricular output and found that sustained inspiratory pressure induces proportionally similar decreases in both right and left ventricle output.They concluded that the hemodynamic effects of positive inspiratory pressure ventilation will depend on the degree of lung inflation, on the inspiratory time, and on the time when measurements are made within the ventilatory cycle.In normal volunteers, Huemer et al. (36) employed Doppler hemodynamic indices to show that the fall in cardiac output during PEEP is caused by a reduction in ventricular filling due to decreased venous return.Our data, despite the fall in cardiac index, showed that DO 2 and VO 2 were preserved; in addition, CaO 2 increments due to the PEEPrelated increase in oxygen saturation were also able to partially compensate for the hemodynamic impairment.According to our results, although PEEP therapy above 10 cmH 2 O substantially increases oxygenation, it should be cautiously applied, especially in patients with marginal cardiac function due to contractility disturbances or hypovolemia.The positive results from PEEP higher than 10 cmH 2 O related to reduction in airway resistance and compliance restoration due to alveolar recruitment should be care-fully weighed against undesirable hemodynamic effects.

Validity and limitations of measurements of respiratory mechanics in patients
Among the several techniques available to study respiratory mechanics during mechanical ventilation, the single breath method (SBM) (37), the end-inflation occlusion method (EIOM) (11), and the interrupter technique (IT) (38) have been successfully applied to normal humans or patients with acute respiratory failure (1,(21)(22)(23)39).SBM allows a detailed account of respiratory system resistive properties throughout relaxed expiration.At a particular constant inspiratory flow, EIOM analyzes the frequencydependent behavior of respiratory system resistance (giving the infinite frequency and zero frequency resistances) and splits it into its homogeneous and uneven components, i.e., that corresponding to the combined summation of series/parallel elements, and that resulting from time constant inequalities within the system and/or stress relaxation, respectively.IT allows the study of volume and flow dependence of the resistive and elastic mechanical properties of the respiratory system by means of brief airway occlusions during relaxed expiration.In the present study, we utilized the EIOM because we are more confident in this procedure.Some conditions may modify the values obtained by EIOM.First, the magnitude of DP 2 can be influenced by the compliance and the resistance of the equipment.Unless equilibration of airway and alveolar pressure is complete, airway open pressure does not reflect the elastic recoil pressure of the respiratory system.An overestimation of DP 2 because of insufficient expiratory pause results in underestimation of resistive pressure and overestimation elastance.The ideal application of EIOM requires an instantaneous occlusion of the open airway pressure, which is possible to achieve since the occlusion valve of the ventilator has a latency in occlusion time.There could be an underestimation of the resistive pressures used to compute R rsmax and DP 1 because during the closure of the system some flow is still present and will cause an increase in lung volume and consequently in elastic recoil pressure and DP 1 .
At a constant flow R rsmax increases with lung volume, reflecting the volume dependence of stress relaxation.At a constant inflation volume R rsmax decreases with increasing flow, exhibiting a minimum value at a flow rate substantially higher than the eupneic flow range.In order to interpret measurements of respiratory system mechanics correctly, the investigator must be aware of the underlying assumptions and the clinical conditions under which they are violated (39).Throughout this study, we assumed that volume, pressure and flow data are derived from appropriately calibrated instruments and that inspired gas has been delivered at a constant flow (square wave flow on the ventilator), situations that, if unrecognized, will lead to erroneous results.Particular emphasis is placed on the need for absence of spontaneous respiratory activity, errors due to time-constant inequalities within the lung, the recognition of dynamic hyperinflation, and errors in determining volumes, flows and pressures due to leaks, gas compression, tubing compliance, ventilator malfunction and intrinsic resistance of tracheal tube and inspiratory circuit.The increased values of airway resistance and elastance in our patients obtained at ZEEP conditions suggest that some degree of injury is caused in the lung during the surgical procedure (40).
In conclusion, even though most of the investigations involving PEEP therapy have been carried out on patients with acute respiratory failure, data on its effects on respiratory mechanics in postoperative cardiac patients without respiratory failure are scarce (41).Along with the increase in PEEP we observed a significant decrease in total and airway resistance and in elastance.The dependence of volume on viscoelastic properties showed a slight but not significant tendency to increase with reduction in airway resistance.

DP 2
represents the slow decay phase of airway pressure after inspiratory airway occlusion and is represented by the interval between P 1 and P 2 (Figure1).DP 2 (cmH 2 O) did not change significantly with PEEP.A slight increase of this parameter was observed from 3.76 ± 0.49 (ZEEP) to 3.96 ± 0.55 (5 cmH 2 O of PEEP), 4.41 ± 0.66 (10 cmH 2 O of PEEP), and 4.51 ± 0.63 (15 cmH 2 O of PEEP) (P<0.4869).The variation of total resistance and its subcomponents data along with PEEP are represented in the Figure2.

Figure 1 -
Figure1-Flow and pressure curves used for the evaluation of respiratory mechanical properties by the end inspiratory occlusion method.P tr : Tracheal pressure; P 0 or P trmax : peak airway pressure; P 2 or P el,rs : static elastic recoil pressure; R init (DP 1 ): minimal value of resistance; R diff (DP 2 ): mechanical unevenness within the system and stress relaxation; R rsmax : total respiratory system resistance, corresponding to the sum of R diff and R init .
-2 , when we changed the PEEP level from 10 to 15 cmH 2 O the oxygenation improvement caused by the decrease in Qs/Qt maintained DO 2 at adequate levels.Therefore, considering a Qs/Qt reduction from 19.64 ± 2.06% at PEEP of 5 cmH 2 O to 15.71 ± 1.40% and 11.66 ± 1.24% at PEEP of 10 and 15 cmH 2 O, respectively, values from 10 to 15 cmH 2 O may be necessary to reduce the shunt fraction after cardiac surgery.