0
Clinical Investigations in Critical Care |

Determinants of Aortic Pressure Variation During Positive-Pressure Ventilation in Man* FREE TO VIEW

André Y. Denault, MD; Thomas A. Gasior, MD; John Gorcsan, III, MD; William A. Mandarino, MSME; Lee G. Deneault, MS; Michael R. Pinsky, MD, FCCP
Author and Funding Information

*From the Department of Anesthesiology and Critical Care Medicine (Drs. Denault, Gasior, and Pinsky, and Mr. Deneault) and the Division of Cardiology (Dr. Gorcsan and Mr. Mandarino), Department of Medicine, University of Pittsburgh Medical Center, Pittsburgh, PA.

Correspondence to: Michael R. Pinsky, MD, FCCP, 604 Scaife Hall, 3550 Terrace St, Pittsburgh, PA 15261; e-mail: pinsky@smtp.anes.upmc.edu



Chest. 1999;116(1):176-186. doi:10.1378/chest.116.1.176
Text Size: A A A
Published online

Study objectives: To define the relation between systolic arterial pressure (SAP) changes during ventilation and left ventricular (LV) performance in humans.

Design: Prospective repeat-measures series.

Setting: University of Pittsburgh Medical Center Operating Room.

Patients: Fifteen anesthetized cardiac surgery patients before and after cardiopulmonary bypass when the mediastinum was either closed or open.

Interventions: Positive-pressure ventilation.

Measurements and results: SAP and LV midaxis cross-sectional areas were measured during apnea and then were measured for three consecutive breaths. SAP increased during inspiration, this being the greatest during closed chest conditions (p < 0.05). Changes in SAP could not be correlated with changes in either LV end-diastolic areas (EDAs), end-systolic areas, or stroke areas (SAs). If SAP decreased relative to apnea, the decrease occurred during expiration and was often associated with increasing LV EDAs and SAs. SAP often decreased after a positive-pressure breath, but the decrease was unrelated to SA deficits during the breath. Increases in SAP were in phase with increases in airway pressure, whereas decreases in SAP, if present, followed inspiration. No consistent relation between SAP variation and LV area could be identified.

Conclusions: In this patient group, changes in SAP reflect changes in airway pressure and (by inference) intrathoracic pressure (as in a Valsalva maneuver) better than they reflect concomitant changes in LV hemodynamics.

Figures in this Article

Abbreviations: ABD = automated border detector; 2D = two dimensional; Δ down = maximum decrease in SAP after a positive-pressure breath, as compared to apneic SAP; EDA = end-diastolic area; ESA = end-systolic area; IABP = intra-aortic balloon pump; ITP = intrathoracic pressure; LV = left ventricular; Ppao = pulmonary artery occlusion pressure; SA = stroke area; SAP = systolic arterial pressure; TEE = transesophageal echocardiography; Δ up = maximum increase in SAP during the positive-pressure breath, as compared to apneic SAP; Vt = tidal volume

The hemodynamic effects of artificial ventilation are complex, interrelated, and dependent on the initial cardiovascular state of the subject.1It has been proposed that the aortic pressure variation in response to a positive-pressure breath can be used to define, in a nonambiguous fashion, the primary determinants of the existing hemodynamic state.2 If correct, this bedside test would be useful for evaluating the hemodynamic condition for all patients requiring mechanical ventilation. In the present study, we sought to define the determinants of aortic pressure variation relative to estimates of left ventricular (LV) area and function in humans.

Several investigators have defined ventilation-associated changes in arterial pressure as the variable that would define overall heart-lung interactions. Perel et al2observed that, when compared with an apneic baseline, systolic arterial pressure (SAP) decreased more in response to a positive-pressure breath in dogs that were hemorrhaged and presumably more preload-dependent, than in animals that were fluid resuscitated. This suggested to these investigators that in preload-dependent states, positive-pressure breathing decreased SAP by decreasing the LV preload. Massumi et al3observed that SAP may also increase after positive-pressure ventilation in patients with LV failure. They called this observation “reversed pulsus paradoxus.” Pizov et al4demonstrated a similar phenomenon in dogs with acute ventricular failure, but only following fluid resuscitation. These data led them to speculate that once LV preload was adequate (following fluid resuscitation), positive-pressure ventilation-induced changes in intrathoracic pressure (ITP) would augment the LV ejection by reducing the LV afterload.5 The increase in SAP in their studies was thought to represent the associated increase in LV stroke volume.

Whether or not LV stroke volume increases during positive-pressure inspiration in heart failure and the mechanisms by which LV stroke volume increases are not known. Potentially, LV stroke volume could increase by one of three mechanisms: (1) an increase in LV filling (increased end-diastolic volume), as the alveolar vessels are compressed during inspiration6; (2) a decrease in right ventricular residual end-diastolic volume, which increases LV diastolic compliance7; or (3) a decrease in afterload (decreased end-systolic volume), as the increase in ITP during positive-pressure inspiration decreases the transmural LV ejection pressure.1,5

Although it is tempting to conclude that changes in SAP during positive-pressure ventilation reflect perturbations in LV preload and afterload, the relation between changes in SAP and ventricular volumes during ventilation is unknown. Furthermore, the mechanism by which SAP varies during ventilation could involve processes independent of changes in the LV preload or afterload. SAP could vary during positive-pressure ventilation because of a direct transmission of the increased ITP to the aorta in a fashion analogous to phases 1 and 2 of a Valsalva maneuver. Inspiration would increase SAP similarly to the way it increases in the initial phase of a Valsalva maneuver.8If this were the case, stroke volume would eventually decrease because of the associated decrease in venous return,9 although SAP would remain elevated as long as ITP remained elevated. Most importantly, however, is that the increase in SAP would be in phase with inspiration, whereas any decrease in SAP would not. Furthermore, SAP could decrease during positive-pressure expiration because of the withdrawal of ITP-supported arterial pressure in the setting of a decreased aortic blood volume and in a fashion analogous to the release phase (phase 3) of a Valsalva maneuver. Here the decrease in SAP would invariably follow inspiration but need not reflect a decrease in stroke volume. Furthermore, these ventilation-associated changes in SAP need not be related to changes in LV volume, nor do they need to be influenced to a relative degree by the level of LV contractility, but they would reflect only the changes in ITP.

To separate these mechanisms, we studied the effect of positive-pressure ventilation on SAP and the LV midaxis cross-sectional area in patients undergoing coronary artery bypass surgery. We tested the hypothesis that changes in SAP during a positive-pressure breath are induced solely by in-phase changes in ITP and need not reflect changes in stroke volume, end-diastolic volume, or end-systolic volume. We reasoned that if the inspiration-associated increases in the pulmonary blood flow6 increased the LV preload, thus increasing SAP, then the increase in SAP would be similar whether the chest was open or closed, because changes in ITP would not primarily alter this interaction. However, if changes in ITP or ventricular interdependence were the primary factors altering LV SAP or preload, respectively, then closing the chest would accentuate the effect by inducing greater amounts of increased ITP and reduced systemic venous return. Finally, since Perel et al10 initially described this phenomena in patients following a cardiopulmonary bypass, we strove to assess the above interaction both before and after a bypass to ascertain whether the hemodynamic alterations known to occur following a bypass altered the subjects’ response.11

Patients

After approval of our protocol by the Institutional Review Board committee of the University of Pittsburgh, we studied 17 sequential patients undergoing elective coronary artery bypass surgery at the University of Pittsburgh Medical Center. The first 15 patients studied on the protocol are described below, whereas the last 2 patients were used to validate the accuracy of the LV area measurements. Informed consent was obtained. Profiles of the 15 initial patients are summarized in Table 1 . General anesthesia was induced by using high-dose narcotics and oxygen with no inhalation anesthetics. All patients were instrumented and monitored with a flow-directed balloon-tipped pulmonary artery catheter, central aortic catheter (femoral arterial line), endotracheal tube, and ECG. Airway pressure at the proximal end of the endotracheal tube was also monitored, and ventilation was provided with a volume-cycled ventilator (Servo Ventilator 900; Siemens-Elena; Solna, Sweden). All pressures were calibrated at zero at the midaxillary level, and they were transduced by using high-displacement transducers (TXX-R; Viggo-Spectramed; Oxnard, CA).

We excluded patients with a contraindication to the use of transesophageal echocardiography (TEE), simultaneous use of an intra-aortic balloon pump (IABP), lack of sinus rhythm, numerous premature ventricular complexes during the data acquisition periods, and the inability to obtain an adequate transgastric view from the midpapillary level. In the criteria put forward by Schnittger et al,12 an inadequate echocardiographic view was defined as< 75% of the perimeter of the ventricular image being defined by contiguous endocardial echoes. In practice, no patients were excluded because of a contraindication to the use of the TEE, and only 2 patients out of 17 were excluded because of the inability to obtain an appropriate echocardiographic image. The TEE probe was positioned at the beginning of the experiment and not moved until the data acquisition was completed. As a further form of quality control, the quality of the TEE images was analyzed on-line by a cardiac anesthesiologist. The images were recorded on videotape and reviewed by a cardiologist who specialized in echocardiography, using the same criteria. Because the purpose of our study was to examine the effect of positive-pressure on LV volume and SAP, subsequent analysis was performed using only the TEE images that both observers, who were blinded to the other’s opinions, agreed were acceptable in quality.

Transesophageal Echocardiographic Examination

Two-dimensional (2D) TEE signals were acquired by an echocardiographic automated border detector ([ABD]; Sonos 1500; Hewlett Packard Systems; Andover, CA). This methodology has been described and validated previously for this type of patient population.13 Briefly, an automated-edge border-detection algorithm is superimposed on the 2D echocardiographic image and displayed in real time on the video monitor as a red line that follows the endocardial contour (Fig 1 ). If the endocardial contours are still unclear, lateral or total gain control is adjusted to improve image resolution. The tracing of a region of interest is then drawn manually. Within this region of interest, integration of the edge-detected area occurs, which allows measurements of instantaneous area-time relation of the blood-pool area in square centimeters. From these data, stroke area (SA), which is the difference between the maximum end-diastolic area (EDA) and minimum end-systolic area (ESA), is calculated. SA, EDA, and ESA were taken to reflect stroke volume, end-diastolic volume, and end-systolic volume, respectively. The images analyzed from the TEE-ABD were obtained from a transgastric, midventricular, short-axis view by using the LV midpapillary level as an anatomic landmark. The area signal was calibrated at zero on both the ABD and the computer workstation with a predetermined area. These measurements and calculations, along with the pressure measures, were displayed on-line and stored on computer disk for subsequent analysis. An example of the recorded signal is shown in Figure 2 .

The TEE technique using this border-detection algorithm has been previously validated by us and others, and it has demonstrated a good correlation between LV stroke volume and LV SA in both animal and human studies when different modes were used to measure LV volume and when offline tracing was compared with manual planimetry tracing1318. To further validate the findings of the TEE, two additional patients were subsequently studied separately. In these patients (the validation group) an appropriately sized and calibrated electromagnetic flow probe (Cineflow II; Carolina Medical; King, NC) was placed around the ascending aorta immediately above the coronary artery. Estimations of apneic cardiac output, which were achieved by averaging three thermodilution curves by using 10 mL 5% dextrose solution at 4°C, were used for initial calibration of the electromagnetic flow signal. Integration of the aortic flow signal per beat was used to derive LV stroke volume. Stroke volume changes during ventilation were compared to SA changes to see if TEE-ABD signals accurately tracked LV volume changes.

Protocol

The protocol sequence consisted of observing the effects of a brief apneic interval (15 to 20 s), followed by performing standard positive-pressure ventilation (tidal volume[ Vt], 8 to 10 mL/kg; frequency, 15 breaths/min; fraction of inspired oxygen, 100%) on the dependent measured variables. This apnea-ventilation sequence was repeated three times at each step within the surgical procedure. Data were recorded before and after bypass and during both open and closed chest conditions, thereby yielding four sequential, separate steps. We assumed that the differences between the responses that occurred during closed conditions and the responses that occurred during open chest conditions would reflect the differences in ITP swings during ventilation, whereas the differences between responses that occurred before bypass and responses that occurred after bypass would reflect changes in the contractile state because a bypass induces a transient decrease in contractility.13 The validation group studies were only performed during open chest conditions. The pericardium was kept closed during the study before and after bypass.

Data Analysis

Mean apneic steady-state values for all variables were used to define baseline values for each step of the protocol. The maximum decrease in SAP after a positive-pressure breath, as compared to apneic SAP, was defined as Δ down. The maximum increase in SAP during the positive-pressure breath as compared to apneic SAP was defined as Δ up. Changes in SAP from the apneic baseline were analyzed in relation to the maximum variation in LV area before and after the positive-pressure breath during each of the four steps. SAP variation, mean LV area measurements, and their variations were compared within each condition by using analysis of variance. Statistical significance was defined as p < 0.05.

In an attempt to quantify any additional effect that the positive-pressure breath may have had on subsequent decreases in SAP, we estimated the cumulative SA deficit throughout the breath relative to mean apneic values and correlated this SA deficit with Δ down. The mean apneic SA was taken as the mean SA of the three cardiac cycles preceding the breath, whereas the SA deficit was taken as the sum of each SA minus the mean SA for each cardiac cycle from the start of the breath until the lowest systolic BP occurred. This volume is referred to in the text as the cumulative SA deficit. Data are expressed as mean (± SD).

Because the phase relation between positive-pressure ventilation and both SAP and LV area data would be different, depending on which process described above determined the response, we further analyzed the relations between airway pressure, arterial pressure, and LV area by Fourier analysis using the ventilatory cycle as the primary harmonic for three sequential breaths. Furthermore, we examined the effects of positive-pressure inspiration and expiration on the matching cardiac cycles to ascertain if single beat changes in EDA, ESA, or SA occurred relative to the ventilatory cycle.

In the validation group, TEE-ABD-derived SA and electromagnetic flow probe-derived stroke volume were compared over one complete ventilatory cycle by using simple correlation statistics and the method of least squares.

In total, 114 apnea/positive-pressure ventilation sequences in 15 patients were collected. The quality of the recorded images on videotapes using the criteria of Schnittger et al12 was assessed by two observers, each blinded to the interpretation of the other. Only data from sequences accepted by both observers were subjected to further analysis. Based on this screening criteria, 70 of the sequences (61%) were then analyzed. Thirty-six sequences were in open chest conditions, with 18 being recorded before bypass; whereas 34 sequences were in closed chest conditions, with 13 being recorded before bypass. Some subjects had more than one sequence acceptable for a specific condition. In these cases, the individual runs were compared in order to assess the reproducibility of the measurements. Intracondition variability was a < 10% maximal difference in LV EDA and a < 2.5 mm Hg maximal difference in SAP during apnea. The first run was chosen for subsequent analysis. Based on this algorithm, we compared the effect of positive-pressure ventilation relative to apnea in 11 subjects during the closed chest condition (5 before bypass) and 14 subjects during the open chest condition (10 before bypass). All subjects remained hemodynamically stable throughout the study protocol, and they were without the use of IABP counterpulsation or changes in inotropic therapy (Table 1).

Effects of Positive-Pressure Ventilation on SAP and LV Area

There was an increase or no change in SAP in 22 of the sequences (88%) during positive-pressure inspiration. When SAP increased, the increase in airway and arterial pressure occurred together. If SAP decreased, the decrease occurred during expiration. SAP never increased during expiration. The magnitude of increase or decrease in SAP was similar among subjects (Table 2 ). Overall, SAP increased by 2.3 ± 2.7 mm Hg and decreased by 3.7 ± 3.5 mm Hg with inspiration and expiration, respectively.

Although SAP tended to vary with the phase of ventilation, no consistent pattern of SA, EDA, or ESA with ventilation could be identified. During positive-pressure inspiration, SA increased in 11 sequences (44%), decreased in 13 sequences (52%), and stayed the same in 1 sequence (4%; mean change, −0.2 ± 0.9 cm2; Table 3 ). The EDA increased in 5 sequences (20%), decreased in 19 sequences (76%), and stayed the same in 1 sequence (4%; mean change,− 0.8 ± 1.1 cm2). The ESA increased in 3 sequences (12%), decreased in 19 sequences (76%), and stayed the same in 3 sequences (12%; mean change, −0.7 ± 0.8 cm2). Furthermore, there was no consistent relation between changes in SAP and changes in SA, EDA, or ESA (Fig 3 , left) or between Δ down (Table 2) and changes in SA, EDA, or ESA (Fig 3, right). In most subjects, however, both EDA and ESA decreased during the positive-pressure inspiration (Fig 4 ), but there was no difference between the degree of decrease in EDA compared with ESA.

In the validation group, changes in LV SA were associated with concomitant changes in the electromagnetic flow probe-derived LV stroke volumes during ventilation.

Effect of Cardiopulmonary Bypass on SAP and LV Area Relations

Before bypass, SAP increased during the positive-pressure inspiration in the majority of subjects (93%; Fig 5 ). During this period (15 sequences), positive-pressure ventilation increased the SA in 9 sequences (60%), decreased the SA in 5 sequences (33%) and caused no change in the SA in 1 sequence (7%; mean change, 0.1 ± 0.5 cm2). The Δ down tended to be greater after bypass, but this did not reach statistical significance. As for the group as a whole, positive-pressure inspiration had an inconsistent effect on LV areas: EDA increased in 4 sequences (27%), decreased in 10 sequences (67%), and stayed the same in 1 sequence (7%; mean change, −0.7 ± 0.9 cm2). ESA had a similar response during positive-pressure inspiration, increasing in 2 sequences (13%), decreasing in 12 sequences (80%), and staying the same in 1 sequence (7%; mean change, −0.8 ± 0.9 cm2). After cardiopulmonary bypass, the effects of positive-pressure inspiration on SAP were similar to prebypass responses, with SAP increasing in seven sequences (70%) and decreasing in three sequences (30%). In most sequences, however, after cardiopulmonary bypass, SA, EDA, and ESA decreased 80%, 90%, and 70%, respectively, during inspiration, compared with an apneic baseline (p < 0.05). Mean apneic EDA was smaller after bypass and demonstrated a greater decrease after a positive-pressure breath than before bypass (p < 0.05).

Effect of Open and Closed Chest Conditions

In both closed and open chest conditions (Fig 6 ), increases in SAP (Δ up) occurred in phase with the positive-pressure inspiration (86% open and 82% closed), whereas theΔ down occurred in expiration. The Δ down was greater in the closed than in the open chest condition (Table 2; p < 0.05). Changes in SA after positive-pressure ventilation were minimal (mean open chest change, −0.1 ± 1 cm2 and mean closed chest change, −0.2 ± 0.8 cm2), and they were not significantly different between open and closed chest conditions, with SA decreasing in 55% of the closed and in 50% of the open chest condition runs. Interestingly, in the closed chest condition, both EDA and ESA decreased during the positive-pressure inspiration in all the sequences, whereas in the open chest condition, this occurred in only 57% of the sequences (p < 0.05). Furthermore, EDA and ESA decreased more in the closed compared with the open chest condition (p < 0.05).

Cumulative SA Deficit and Δ Down

There was no measurable relation between the maximal decrease in SAP after a breath and the cumulative SA deficit (summed SA decrease relative to apneic SA; Fig 7 ). The cumulative SA deficit was higher in the postbypass period, and this was associated with a greater decrease in EDA (p < 0.05). The presence of an open or closed chest did not influence this relation.

Fourier Analysis

A Fourier analysis of the LV area and arterial pressure was done. In the closed chest condition, an SAP harmonic signal at the same frequency as the airway signal was found in 55% of the runs, whereas it was invariably seen in all of the LV area signals. In the open chest condition, this common harmonic signal was present in only 36% and 64% of the SAP and LV area signals, respectively. However, the phase angles of the SAP and LV area signals relative to the airway signal were different. The airway phase angle slightly preceded the arterial in > 80% of the runs by approximately one heartbeat in both the open and closed chest condition, and the two signals had indistinguishable phase angles in 15% of the runs. The airway pressure and the echo area were never in identical phase. In the closed chest condition, the change in the SAP signal preceded the change in the echo area in all cases, with the SAP tending to increase while the echo area was decreasing. In the open chest condition, this occurred 67% of the time.

We observed that changes in SAP during positive-pressure ventilation in humans cannot be explained by proportional changes in LV volume. The effects of positive-pressure ventilation on LV volume and SAP are complex and not interpretable by a single mechanism. If SAP increases relative to apneic values during ventilation, it does so during the inspiratory phase, whereas if SAP decreases, it does so during the expiratory phase. These changes in SAP are inconsistently associated with changes in SA, EDA, or ESA, suggesting that coupled changes in LV performance are not primary determinants of SAP variations during ventilation. Although the decrease in SAP after a positive-pressure breath appears to be accentuated during hypovolemic states,2 the mechanism for these changes does not appear to relate to coupled changes in LV preload. Because the changes in SAP during positive-pressure ventilation are in phase with the change in the airway pressure, we favor the hypothesis that positive-pressure, ventilation-induced changes in SAP may reflect concomitant changes in ITP similar to those in phase 3 of a Valsalva maneuver.

Consistent with a direct effect of ITP on SAP variations, the decrease in SAP was more pronounced in the closed (p < 0.05) than in the open chest condition for the same Vt. If the observed decrease in SAP was secondary to a decrease in venous return, then we would expect to see a decrease in SA and EDA. This was not the case. Although SA decreased in 56% of our subjects and EDA decreased in 80% of our subjects (Fig 3), this decrease occurred during the time that SAP was increasing. Furthermore, the relation between the degree of decrease in SAP and the corresponding decrease in SA and EDA (Fig 3) was variable. From the Fourier analysis, we saw that in the closed chest condition, the SAP signal always preceded the change in LV area. These data are consistent with the hypothesis that the increase in ITP from positive-pressure ventilation directly alters the SAP. Finally, if the decrease in SAP represents a deficit in LV output secondary to the positive-pressure breath, we would expect that SAP would decrease in proportion to the SA deficit preceding it during the breath. The data show that was also not the case (Fig 7), and it implies that a decrease in SAP during positive-pressure ventilation is not necessarily secondary to a decrease in LV stroke volume.

Our observations and those from Figure 6 in the study of Abel et al19 demonstrate less variation in SAP during the open chest condition rather than during the closed chest condition. Based on observations in three subjects, Perel et al10 and Pizov et al4 hypothesized that the absence of variation in SAP may indicate the presence of an underlying LV dysfunction during the open chest condition. Neither the data of Abel et al19 nor ours support this conclusion. Although we saw less SAP variation in our study group following cardiopulmonary bypass, when contractility is reduced,13 there is a considerable overlap in response between subjects before and after bypass. However, the average apneic EDA in our subjects with variation in SA during a positive-pressure breath of < 0.5 cm2 (n = 14) was 12.2 ± 5.5 cm2 compared with a mean apneic EDA of only 7.1 ± 3.7 cm2 in those subjects (n = 11) with > 0.5 cm2 variation. These data are consistent with results from subjects having a relatively fixed stroke volume, as is commonly seen in patients with heart failure.,13 According to these data, it appears that limiting ITP fluctuations (open chest conditions) minimized the SAP changes seen in response to positive-pressure ventilation, whereas impaired LV function minimizes the changes seen in LV area. This impaired LV function could be secondary to postbypass ischemia, stunning myocardium, or hibernating myocardium.

Massumi et al,3 Jardin et al,20 and Perel et al10 suggested that increases in SAP or reversed pulsus paradoxus during a positive-pressure breath would represent an increase in stroke volume either from an increase in preload6 or a decrease in afterload.1 A subset of our subjects appears to behave in a fashion similar to that described by these authors. In 10 of our subjects, SA increased during the positive-pressure inspiration, and in all but 1, this increase was in phase with the increase in SAP (Fig 3). In the nine subjects in whom both SAP and SA increased, the increase in SA was associated with an increase in EDA in only three subjects and a decrease in ESA in the remaining six subjects. These data imply that both increasing preload (increase in EDA) and decreasing afterload (decrease in ESA) may be the mechanisms inducing these changes. However, in this subset of our subjects, the reduction in afterload appears to be the more prevalent mechanism responsible for the changes in SAP and LV area.

Unfortunately, because many variables can affect the relation between SAP and LV volume, it is still difficult to infer cardiovascular status from changes in SAP. The multiple determinants of SAP variation operating in all patients could explain the variability of findings in both the literature and in our data on this relation. Scharf et al,21 in their demonstration in an animal model, used intramyocardial radiopaque markers to assess sequential volume changes and aortic flow probe data to assess stroke volume, and they found that changes in SAP resulted from a combined interaction between decreases in aortic flow and transmitted increases in ITP. They demonstrated that the aortic flow could decrease as the SAP increased and that this difference was even more pronounced with faster respiratory rates (as seen in their Figure 2).,21 We made similar observations in our validation group study.

The disparity in the interpretations of results obtained in these studies could be explained by the varied methods used for measuring LV volume and also by the effects of ventilation on LV area and SAP, which are not measured and recorded on-line.20,19In the study of Jardin et al,20 measurements were also done from a transthoracic approach, as opposed to a transesophageal view, from which image quality is significantly improved. Most of the studies on the effect of positive-pressure ventilation on LV volume and SAP were done in the closed chest condition, except for one in which four subjects had measurements made during the open chest condition.19 Interestingly, in those open chest subjects, as in our study, minimal variation in SAP occurred during positive-pressure ventilation. Finally, although these workers felt that SAP variations, specifically increases in SAP, reflected a fluid-resuscitated heart-failure state, our data did not support this hypothesis. We saw no differences in the SAP response to positive-pressure ventilation between subjects when referenced either to LV EDA, as a measure of preload, or to fractional area of contraction, as a measure of contractility.

Study Limitations

Although care was taken to study similar types of subjects undergoing comparable surgical stresses, marked variability in response among subjects occurred. This variation might be a reflection of differing intravascular volume status, as well as differing LV systolic and diastolic function. EDA as an estimate of LV filling volume varied from 3.4 to 26.4 cm2 in our patients, although most patients had an EDA around 9.8 cm2. Furthermore, we saw that subjects with a larger EDA had less variation in their LV area after a positive-pressure breath. Alteration in diastolic function can also occur after cardiac surgery.11 The average EDA and pulmonary artery occlusion pressure (Ppao) in our subjects before cardiopulmonary bypass were not significantly different before (EDA, 10.5 ± 5.9 cm2; Ppao, 14 ± 4 mm Hg) and after bypass (EDA, 8.6 ± 4.1 cm2; Ppao, 12 ± 5 mm Hg).

We studied similar patients with various levels of cardiac function. All of these patients had coronary artery disease and were undergoing the same type of surgery with open and closed chest conditions. Vasoactive medications were not changed during the protocol. IABP was used transiently in two patients, but it was turned off shortly prior to and during the positive-pressure breath because the arterial pressure waveform would not have been interpretable. Patients with severe left main disease have a prophylactic IABP inserted prior to the operation to prevent decompensation before the bypass, as in our patients. We deliberately chose to evaluate patients with nonhomogeneous cardiac functions because, as stated in the introduction, the goal of our study was to study the effect of positive-pressure ventilation on systolic arterial pressure over a spectrum of cardiac contractility.

Because our results seem to reflect a spectrum of responses in LV area to positive-pressure ventilation as opposed to a single mechanism, one could question the validity of our measurements, especially after considering that we accepted only 60% of the data for interpretation, based on our criteria. TEE has limited application during positive-pressure ventilation because the image can shift out of the region of interest. This is why we analyzed only high-quality images after two independent observers using accepted criteria reviewed them. The overall percentage of good-quality images in our study is nevertheless comparable to other studies in which the ABD technique has been used to assess the ventricular area.14,16,18 Still, issues of the rotational artifact of the LV cavity relative to the TEE, induced by positive-pressure breaths, need to be addressed. This rotational artifact can be either lateral or vertical to the plane of the 2D image. In both instances, the area measured by the ABD method may not reflect LV area at the same anatomic location seen during apnea. Lateral movement was clearly seen in several breaths in some of our subjects. In these examples, the ABD measurement moved off the region of interest borders, and these data were rejected. No data with lateral rotational artifacts were used in our analysis. Rotational artifacts were not the only cause for rejecting data. Despite the fact that the TEE probe was not moved during the data acquisition, upward and downward cardiac movement could have been missed because the esophageal position allows for no reference point within the thorax. However, Smith et al22 addressed this issue previously. They moved the TEE probe 2 cm in and out (4 cm total distance) in the transgastric position from a midpapillary short-axis vein. When they compared measures of EDA and ESA from these three vertical positions, they saw no difference. Because it is highly unlikely that vertical movement of the mediastinal block (including esophagus and heart) exceeded 2 cm during a positive-pressure breath, vertical rotational artifacts likely had little influence on our measured values. Furthermore, in two patients in open chest condition, we were able to validate our measurements further by using the electromagnetic flow probe on the aorta. Close correlation was observed between TEE SA and flow probe-derived stroke volume during both ventilation and inferior vena caval occlusion maneuvers.9

As a final note of validation, Leithner et al23 demonstrated in normal subjects that lung distention to levels similar to those used in our study reduced LV end-diastolic volume and induced an anterior rotational movement on the heart. They used MRI techniques and thus were limited to end-expiratory analysis as lung volume was progressively increased by the application of 15 cm H2O positive end-expiratory pressure. Because the movement they saw did not alter the midventricular short-axis orientation, if TEE had been used in their subjects, TEE would have not seen this movement as a volume artifact. Thus, it seems highly unlikely that our data in this selected population reflect rotational artifacts.

The variations that we measured in SAP were small, especially if they were expressed as a percentage.2 However, we did not limit our analysis to patients in which pulsus paradoxus, defined by a change in SAP ≥ 10 mm Hg, was present. Indeed, Rooke et al24 have shown in a human hemorrhage model that a SAP variation< 5 mm Hg or Δ down < 2 mm Hg indicates minimal intravascular depletion. Their observed changes in SAP following a 500-mL blood removal was 15.2 ± 7.5 mm Hg (their Table 1). We cannot exclude the possibility that, for larger variations in SAP, changes in ventricular volumes may play a more important role. Our data on closed chest conditions are similar to those obtained by Coriat et al,25 especially with regard to their Figure 4, which relates the percentage change in EDA with the Δ down. It illustrates how variable the response can be. Two patients had < 10% changes in EDA with a Δ down of > 5 mm Hg, and two others had EDA changes > 40% with a Δ down < 5 mm Hg. We both observed an average fall in EDA and an increase in Δ down, but there are large individual variations. However, Coriat et al,25 did not record the area measurement simultaneously with the arterial and airway pressure and did not demonstrate that changes in SAP reflect simultaneous changes in LV volume. It has been previously shown that for the same decrease in SAP or pulsus paradoxus occurring during normal inspiration, the aortic flow will decrease when associated with tamponade, but it will remain constant when associated with airway obstruction.24,26 Consequently, we observed that SAP variation with respiration in the range cannot be explained entirely by changes in LV volume.

We saw no difference between changes in EDA vs ESA within each subgroup. Because the absolute value of the end-systolic volume must be less than that of the end-diastolic volume, an equal decrease in both values might be interpreted differently if one used either the absolute volume or a percentage change. We used the absolute value of the area changes in our experiment, as based on previous studies20,27 to minimize this potentially confounding effect of proportionality.

We did not evaluate in our patients the effect of open or closed pericardium. We recorded the hemodynamic parameters when the chest was closed, open before and after bypass during closed pericardial conditions. We did not study the effect of a closed vs open pericardium because the clinical use of our observations, for the most part, is directed toward patients with a closed chest and pericardium. Pericardial constraint is a significant factor in relating positive-pressure ventilation and SAP. Pulsus paradoxus is a typical manifestation of an exaggerated increased pericardial pressure that would reduce venous return.28In addition, changes in pericardial pressure will alter the pressure-volume relationship of the left and right ventricles, and systolic performance is augmented by an intact pericardium.29Schertz and Pinsky30 observed, using an animal model, that LV ejection can enhance right ventricular stroke volume, but volume loading or the presence of an intact pericardium did not appreciably alter this interaction. Interestingly, we observed in our study that the presence of an open or closed chest condition did not significantly change SA during a positive-pressure breath (Table 3).

Mean airway pressure was not measured; we gave a Vt, and this raised the peak airway pressure. We did not readjust the Vt during the study in order to keep the same mean airway pressure for each patient. We used a volume-cycled ventilator, not a pressure-cycled ventilator. Because lung compliance can change after a bypass, it is possible that for an identical volume, a variable transpulmonary pressure could be generated. This could alter the change in SAP variation. We preferred to use a defined volume for the simplicity of the test and also because previous studies were done in this fashion. Furthermore, we have previously demonstrated in an intact canine model31 that if Vt is held constant, then ITP increases by similar amounts, despite markedly changing lung compliance induced by oleic acid infusion. Thus, our ventilatory protocol was more likely to sustain a constant ITP variation across conditions than a protocol in which Vt was varied to maintain a common peak airway pressure during positive-pressure ventilation.

In summary, the degree of variation in SAP in patients during cardiac surgery cannot be explained by matched changes in LV area, and these results suggest that changes in SAP during ventilation cannot be used solely to assess the determinants of cardiovascular instability.

Support for this study was provided by the Veterans Administration.

Table Graphic Jump Location
Table 1. Steady-State Apneic Hemodynamic Data*
* 

CO = cardiac output; FAC = fractional area of contraction; HR = heart rate; NA = not available; Pa = arterial pressure; Ppa = pulmonary artery pressure; Pra = right atrial pressure.

Figure Jump LinkFigure 1. A transgastric, midventricular, short-axis view obtained from TEE with ABD. The LV endocardial blood/tissue interface is shown in red. The region of interest encircles the LV and represents the area analyzed continuously. Shown below the echocardiographic image is the ECG and the LV area signal displayed over time.Grahic Jump Location
Figure Jump LinkFigure 2. ECG, arterial pressure (Pa), right atrial pressure (Pra), pulmonary artery pressure (Ppa), LV area (LVA), and airway pressure (Paw) recorded over time following apnea in a patient after cardiac bypass surgery during the closed chest condition.Grahic Jump Location
Table Graphic Jump Location
Table 2. Changes in SAP Following a Positive-Pressure Breath*
* 

Values are expressed as mean ± SD. % SAP up = percentage of maximum SAP during a positive-pressure breath minus SAP during apnea. % SAP down = % decrease in SAP during apnea minus minimal SAP during a positive-pressure breath. See Table 1 for expansion of abbreviation.

 

p < 0.05 in open vs closed chest.

Table Graphic Jump Location
Table 3. Changes in Sum SA and LV Area Following a Positive-Pressure Breath*
* 

Values are expressed as mean ± SD. Sum SA = sum of each SA minus apneic SA during the positive-pressure breath. See Table 1 for expansion of abbreviation.

 

p < 0.05 pre- vs postbypass.

 

p < 0.05 open vs closed chest.

§ 

p < 0.05 low vs high FAC.

Figure Jump LinkFigure 3. Relation between percentage of change in SAP and changes in LV area (left column), and the relation between percentage of absolute decrease in SAP and changes in LV area (right column) during a positive-pressure breath.▪ = SA; • = EDA; ▴ = ESA.Grahic Jump Location
Figure Jump LinkFigure 4. Relation between the change in EDA and ESA after a positive-pressure breath.Grahic Jump Location
Figure Jump LinkFigure 5. Relation between percentage change in SAP and changes in LV area during the prebypass (left column) and postbypass (right column) periods. ▪ = SA; • = EDA;▴ = ESA.Grahic Jump Location
Figure Jump LinkFigure 6. Relation between percent change in SAP and changes in LV area during open chest (left column) and closed chest (right column) conditions. ▪ = SA; • = EDA;▴ = ESA.Grahic Jump Location
Figure Jump LinkFigure 7. Relation between the fall in SAP expressed as Δ down and the sum of the SA difference (sum = SAa − SAx for all patients during the open and closed chest condition and for patients with a fractional area of contraction < 30% or > 30%).Grahic Jump Location
Pinsky, MR, Matuschak, GM, Klain, M (1985) Determinants of cardiac augmentation by elevations in intrathoracic pressure.J Appl Physiol58,1189-1198. [PubMed]
 
Perel, A, Pizov, R, Cotev, S Systolic blood pressure variation is a sensitive indicator of hypovolemia in ventilated dogs subjected to graded hemorrhage.Anesthesiology1987;67,498-502. [PubMed] [CrossRef]
 
Massumi, RA, Mason, DT, Vera, Z, et al Reversed pulsus paradoxus.N Engl J Med1973;289,1272-1275. [PubMed]
 
Pizov, R, Ya’ari, Y, Perel, A The arterial pressure waveform during acute ventricular failure and synchronized external chest compression.Anesth Analg1989;68,150-156. [PubMed]
 
Buda, AJ, Pinsky, MR, Ingles, NB, et al Effect of intrathoracic pressure on left ventricular performance.N Engl J Med1979;301,453-459. [PubMed]
 
Brower, R, Wise, RA, Hassapoyannes, C, et al Effect of lung inflation on lung blood volume and pulmonary venous flow.J Appl Physiol1985;58,954-963. [PubMed]
 
Taylor, RR, Corell, JW, Sonnenblick, EH, et al Dependence of ventricular distensibility on filling the opposite ventricle.Am J Physiol1967;213,711-718. [PubMed]
 
Nishimura, RA, Tajik, AJ The Valsalva maneuver and response revisited.Mayo Clin Proc1986;61,211-217. [PubMed]
 
Robertson, D, Stevens, RM, Friesinger, GC, et al The effect of Valsalva maneuver on echocardiographic dimensions in man.Circulation1977;55,596-602. [PubMed]
 
Perel, A, Koman, V, Geva, D, et al The systolic pressure variation may unmask acute left ventricular ischemic dysfunction upon weaning from cardiopulmonary bypass [abstract]. Anesthesiology. 1990;;73 ,.:A137
 
Humphrey, LS, Topol, EJ, Rosenfeld, GI Immediate enhancement of left ventricular relaxation by coronary artery bypass grafting: intraoperative assessment.Circulation1988;77,886-896. [PubMed]
 
Schnittger, I, Fitzgerald, PJ, Daughters, GT, et al Limitations of comparing left ventricular volumes by two-dimensional echocardiography, myocardial markers and cineangiography.Am J Cardiol1982;50,512-519. [PubMed]
 
Gorcsan, J, Gasior, T, Mandarino, WA, et al Assessment of the immediate effects of cardiopulmonary bypass on LV performance by on-line pressure-area relations.Circulation1994;89,180-190. [PubMed]
 
Pérez, JE, Waggoner, AD, Barzilai, B, et al On-line assessment of ventricular function by automatic boundary detection and ultrasonic backscatter imaging.J Am Coll Cardiol1992;19,313-320. [PubMed]
 
Gorcsan, J, Morita, S, Mandarino, WA, et al Two-dimensional echocardiographic automated border detection accurately reflects changes in left ventricular volume.J Am Soc Echocardiogr1993;6,482-489. [PubMed]
 
Cahalan, MK, Ionescu, P, Melton, HE, et al Automated real-time analysis of intraoperative transesophageal echocardiograms.Anesthesiology1993;78,477-485. [PubMed]
 
Gorcsan, J, Lazar, J, Romand, JA, et al On-line estimation of stroke volume by means of echocardiographic automated border detection in the canine left ventricle.Am Heart J1993;125,1316-1323. [PubMed]
 
Vandenberg, BF, Rath, LS, Stuhlmuller, P, et al Estimation of left ventricular cavity area with an on-line, semiautomated echocardiographic edge detection system.Circulation1992;86,159-166. [PubMed]
 
Abel, JG, Salerno, TA, Panos, A, et al Cardiovascular effects of positive pressure ventilation in humans.Ann Thorac Surg1987;43,198-206. [PubMed]
 
Jardin, F, Farcot, JC, Gueret, P, et al Cyclic changes in arterial pulse during respiratory support.Circulation1983;68,266-274. [PubMed]
 
Scharf, SM, Brown, R, Saunders, N, et al Hemodynamic effects of positive-pressure inflation.J Appl Physiol1980;49,124-131. [PubMed]
 
Smith, JS, Cahalan, MK, Benefiel, DJ, et al Intraoperative detection of myocardial ischemia in high-risk patients: electrocardiographic versus two-dimensional transesophageal echocardiography.Circulation1985;72,1015-1021. [PubMed]
 
Leithner, C, Podolsky, A, Globits, S, et al Magnetic resonance imaging of the heart during positive end-expiratory pressure ventilation in normal subjects.Crit Care Med1994;22,426-432. [PubMed]
 
Rooke, GA, Schwid, HA, Shapira, Y The effect of graded hemorrhage and intravascular volume replacement on systolic pressure variation in humans during mechanical and spontaneous ventilation.Anesth Analg1995;80,925-932. [PubMed]
 
Coriat, P, Vrillon, M, Perel, A, et al A comparison of systolic blood pressure variations and echocardiographic estimates of end-diastolic left ventricular size in patients after aortic surgery.Anesth Analg1994;78,46-53. [PubMed]
 
Shabetai, R, Fowler, NO, Gueron, M The effect of respiration on aortic pressure and flow.Am Heart J1963;65,525-533. [PubMed]
 
Jardin, F, Farcot, JC, Gueret, P, et al Echocardiographic evaluation of ventricles during continuous positive airway pressure.J Appl Physiol1984;56,619-627. [PubMed]
 
Shabetai, R, Fowler, NO, Fenton, JC, et al Pulsus paradoxus.J Clin Invest1965;44,1882-1898. [PubMed]
 
Janicki, JS, Weber, KT The pericardium and ventricular interaction, distensibility, and function.Am J Physiol1980;238,H494-H503. [PubMed]
 
Schertz, C, Pinsky, MR Effect of the pericardium on systolic ventricular interdependence in the dog.J Crit Care1993;8,17-23. [PubMed]
 
Romand, JA, Shi, W, Pinsky, MR Cardiopulmonary effects of positive pressure ventilation during acute lung injury.Chest1995;108,1041-1048. [PubMed]
 

Figures

Figure Jump LinkFigure 1. A transgastric, midventricular, short-axis view obtained from TEE with ABD. The LV endocardial blood/tissue interface is shown in red. The region of interest encircles the LV and represents the area analyzed continuously. Shown below the echocardiographic image is the ECG and the LV area signal displayed over time.Grahic Jump Location
Figure Jump LinkFigure 2. ECG, arterial pressure (Pa), right atrial pressure (Pra), pulmonary artery pressure (Ppa), LV area (LVA), and airway pressure (Paw) recorded over time following apnea in a patient after cardiac bypass surgery during the closed chest condition.Grahic Jump Location
Figure Jump LinkFigure 3. Relation between percentage of change in SAP and changes in LV area (left column), and the relation between percentage of absolute decrease in SAP and changes in LV area (right column) during a positive-pressure breath.▪ = SA; • = EDA; ▴ = ESA.Grahic Jump Location
Figure Jump LinkFigure 4. Relation between the change in EDA and ESA after a positive-pressure breath.Grahic Jump Location
Figure Jump LinkFigure 5. Relation between percentage change in SAP and changes in LV area during the prebypass (left column) and postbypass (right column) periods. ▪ = SA; • = EDA;▴ = ESA.Grahic Jump Location
Figure Jump LinkFigure 6. Relation between percent change in SAP and changes in LV area during open chest (left column) and closed chest (right column) conditions. ▪ = SA; • = EDA;▴ = ESA.Grahic Jump Location
Figure Jump LinkFigure 7. Relation between the fall in SAP expressed as Δ down and the sum of the SA difference (sum = SAa − SAx for all patients during the open and closed chest condition and for patients with a fractional area of contraction < 30% or > 30%).Grahic Jump Location

Tables

Table Graphic Jump Location
Table 1. Steady-State Apneic Hemodynamic Data*
* 

CO = cardiac output; FAC = fractional area of contraction; HR = heart rate; NA = not available; Pa = arterial pressure; Ppa = pulmonary artery pressure; Pra = right atrial pressure.

Table Graphic Jump Location
Table 2. Changes in SAP Following a Positive-Pressure Breath*
* 

Values are expressed as mean ± SD. % SAP up = percentage of maximum SAP during a positive-pressure breath minus SAP during apnea. % SAP down = % decrease in SAP during apnea minus minimal SAP during a positive-pressure breath. See Table 1 for expansion of abbreviation.

 

p < 0.05 in open vs closed chest.

Table Graphic Jump Location
Table 3. Changes in Sum SA and LV Area Following a Positive-Pressure Breath*
* 

Values are expressed as mean ± SD. Sum SA = sum of each SA minus apneic SA during the positive-pressure breath. See Table 1 for expansion of abbreviation.

 

p < 0.05 pre- vs postbypass.

 

p < 0.05 open vs closed chest.

§ 

p < 0.05 low vs high FAC.

References

Pinsky, MR, Matuschak, GM, Klain, M (1985) Determinants of cardiac augmentation by elevations in intrathoracic pressure.J Appl Physiol58,1189-1198. [PubMed]
 
Perel, A, Pizov, R, Cotev, S Systolic blood pressure variation is a sensitive indicator of hypovolemia in ventilated dogs subjected to graded hemorrhage.Anesthesiology1987;67,498-502. [PubMed] [CrossRef]
 
Massumi, RA, Mason, DT, Vera, Z, et al Reversed pulsus paradoxus.N Engl J Med1973;289,1272-1275. [PubMed]
 
Pizov, R, Ya’ari, Y, Perel, A The arterial pressure waveform during acute ventricular failure and synchronized external chest compression.Anesth Analg1989;68,150-156. [PubMed]
 
Buda, AJ, Pinsky, MR, Ingles, NB, et al Effect of intrathoracic pressure on left ventricular performance.N Engl J Med1979;301,453-459. [PubMed]
 
Brower, R, Wise, RA, Hassapoyannes, C, et al Effect of lung inflation on lung blood volume and pulmonary venous flow.J Appl Physiol1985;58,954-963. [PubMed]
 
Taylor, RR, Corell, JW, Sonnenblick, EH, et al Dependence of ventricular distensibility on filling the opposite ventricle.Am J Physiol1967;213,711-718. [PubMed]
 
Nishimura, RA, Tajik, AJ The Valsalva maneuver and response revisited.Mayo Clin Proc1986;61,211-217. [PubMed]
 
Robertson, D, Stevens, RM, Friesinger, GC, et al The effect of Valsalva maneuver on echocardiographic dimensions in man.Circulation1977;55,596-602. [PubMed]
 
Perel, A, Koman, V, Geva, D, et al The systolic pressure variation may unmask acute left ventricular ischemic dysfunction upon weaning from cardiopulmonary bypass [abstract]. Anesthesiology. 1990;;73 ,.:A137
 
Humphrey, LS, Topol, EJ, Rosenfeld, GI Immediate enhancement of left ventricular relaxation by coronary artery bypass grafting: intraoperative assessment.Circulation1988;77,886-896. [PubMed]
 
Schnittger, I, Fitzgerald, PJ, Daughters, GT, et al Limitations of comparing left ventricular volumes by two-dimensional echocardiography, myocardial markers and cineangiography.Am J Cardiol1982;50,512-519. [PubMed]
 
Gorcsan, J, Gasior, T, Mandarino, WA, et al Assessment of the immediate effects of cardiopulmonary bypass on LV performance by on-line pressure-area relations.Circulation1994;89,180-190. [PubMed]
 
Pérez, JE, Waggoner, AD, Barzilai, B, et al On-line assessment of ventricular function by automatic boundary detection and ultrasonic backscatter imaging.J Am Coll Cardiol1992;19,313-320. [PubMed]
 
Gorcsan, J, Morita, S, Mandarino, WA, et al Two-dimensional echocardiographic automated border detection accurately reflects changes in left ventricular volume.J Am Soc Echocardiogr1993;6,482-489. [PubMed]
 
Cahalan, MK, Ionescu, P, Melton, HE, et al Automated real-time analysis of intraoperative transesophageal echocardiograms.Anesthesiology1993;78,477-485. [PubMed]
 
Gorcsan, J, Lazar, J, Romand, JA, et al On-line estimation of stroke volume by means of echocardiographic automated border detection in the canine left ventricle.Am Heart J1993;125,1316-1323. [PubMed]
 
Vandenberg, BF, Rath, LS, Stuhlmuller, P, et al Estimation of left ventricular cavity area with an on-line, semiautomated echocardiographic edge detection system.Circulation1992;86,159-166. [PubMed]
 
Abel, JG, Salerno, TA, Panos, A, et al Cardiovascular effects of positive pressure ventilation in humans.Ann Thorac Surg1987;43,198-206. [PubMed]
 
Jardin, F, Farcot, JC, Gueret, P, et al Cyclic changes in arterial pulse during respiratory support.Circulation1983;68,266-274. [PubMed]
 
Scharf, SM, Brown, R, Saunders, N, et al Hemodynamic effects of positive-pressure inflation.J Appl Physiol1980;49,124-131. [PubMed]
 
Smith, JS, Cahalan, MK, Benefiel, DJ, et al Intraoperative detection of myocardial ischemia in high-risk patients: electrocardiographic versus two-dimensional transesophageal echocardiography.Circulation1985;72,1015-1021. [PubMed]
 
Leithner, C, Podolsky, A, Globits, S, et al Magnetic resonance imaging of the heart during positive end-expiratory pressure ventilation in normal subjects.Crit Care Med1994;22,426-432. [PubMed]
 
Rooke, GA, Schwid, HA, Shapira, Y The effect of graded hemorrhage and intravascular volume replacement on systolic pressure variation in humans during mechanical and spontaneous ventilation.Anesth Analg1995;80,925-932. [PubMed]
 
Coriat, P, Vrillon, M, Perel, A, et al A comparison of systolic blood pressure variations and echocardiographic estimates of end-diastolic left ventricular size in patients after aortic surgery.Anesth Analg1994;78,46-53. [PubMed]
 
Shabetai, R, Fowler, NO, Gueron, M The effect of respiration on aortic pressure and flow.Am Heart J1963;65,525-533. [PubMed]
 
Jardin, F, Farcot, JC, Gueret, P, et al Echocardiographic evaluation of ventricles during continuous positive airway pressure.J Appl Physiol1984;56,619-627. [PubMed]
 
Shabetai, R, Fowler, NO, Fenton, JC, et al Pulsus paradoxus.J Clin Invest1965;44,1882-1898. [PubMed]
 
Janicki, JS, Weber, KT The pericardium and ventricular interaction, distensibility, and function.Am J Physiol1980;238,H494-H503. [PubMed]
 
Schertz, C, Pinsky, MR Effect of the pericardium on systolic ventricular interdependence in the dog.J Crit Care1993;8,17-23. [PubMed]
 
Romand, JA, Shi, W, Pinsky, MR Cardiopulmonary effects of positive pressure ventilation during acute lung injury.Chest1995;108,1041-1048. [PubMed]
 
NOTE:
Citing articles are presented as examples only. In non-demo SCM6 implementation, integration with CrossRef’s "Cited By" API will populate this tab (http://www.crossref.org/citedby.html).

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging & repositioning the boxes below.

Find Similar Articles
CHEST Journal Articles
PubMed Articles
  • CHEST Journal
    Print ISSN: 0012-3692
    Online ISSN: 1931-3543