Canadian Health&Care Mall: Cardiac Augmentation by Phasic High Intrathoracic Pressure Support in Man

oxygen saturationThe hemodynamic data for all seven subjects are summarized in Table 2. The PHIPS was associated with an increase in mean Pes of 6.6±1.1 mm Hg (mean±SE) over the control states (p<0.001). There was no significant difference between the two control states in any of the hemodynamic variables, although the trend over time was to a lower cardiac output, mean arterial pressure, and Pes. This is consistent with the progressive hemodynamic deterioration that the subjects demonstrated both before and after the study. In five of the seven subjects, cardiac output increased with PHIPS; and for the entire group, there was a significant increase in cardiac output from initial control levels (3.6 ± 0.5 L/min) to PHIPS (4.2 ± 0.6 L/min) (Fig 1).

Arterial oxygen saturation was unaffected by PHIPS; thus the increase in cardiac output represents an increase in oxygen delivery. There was no significant change in heart rate over the course of the protocol, with all subjects demonstrating a sinus tachycardia (138 ± 12 beats per minute). The mean arterial pressure increased from control levels (43.0 ±6.1 mm Hg) with PHIPS (51.0±7.7 mm Hg) (p<0.01); however, there was no significant difference in mean arterial pressure relative to mean Pes (transmural mean arterial pressure) between the two states (control, 43.2 ± 5.6 mm Hg; PHIPS 42.5± 6.1 mm Hg) (Fig 2). In one subject (patient 3), we were unable to record transmural pulmonary arterial occlusion pressure due to failure to effectively occlude the pulmonary arterial catheter; however, in the remaining six subjects, there was no significant difference in transmural pulmonary arterial occlusion pressure between control conditions (16.6± 2.5 mm Hg) and PHIPS (19.0±3.5 mm Hg). In patient 3, transmural pulmonary arterial diastolic pressures were unchanged by PHIPS.

The gated cardiac blood pool scan for the single patient is shown in Figure 3. In this patient the PHIPS-associated increase in cardiac output from 2.8 to 3.3 L/min occurred with an improvement in ejection fraction from 36 to 46 percent despite a 10 percent fall in end-diastolic volume and a rise of 4 mm Hg in mean arterial pressure. Although mean arterial pressure increased, since mean Pes increased by 8 mm Hg in this patient, transmural mean arterial pressure fell by 4 mm Hg.


Left ventricular {unction can be described by the relationship between cardiac output and left ventricular filling pressure. Our study demonstrates that in the setting of severe left ventricular dysfunction (ie, with an expanded intravascular blood volume), increasing intrathoracic pressure improves left ventricular performance. These results are in agreement with recent reports. Buda and co-workers found that with sustained increases in intrathoracic pressure, left ventricular performance was maintained in the setting of a decreased preload. Calvin et al demonstrated that in some, but not all, patients with pulmonary edema, increasing intrathoracic pressure by the use of positive end-ventricular performanceexpiratory pressure increased left ventricular ejection and decreased end-diastolic volume. Mathru et al showed that in patients with depressed myocardial function, increasing intrathoracic pressure was associated with improved cardiac performance over that seen with decreasing intrathoracic pressure.

Since heart rate was unchanged, what could account for this observed increase in cardiac output and stroke volume at constant left ventricular filling pressure during PHIPS? Possibly transmural pulmonary arterial occlusion pressure does not accurately reflect left ventricular preload. If either left ventricular compliance was increased by PHIPS or juxtacardiac pleural pressure increased less than Pes with PHIPS, then left ventricular end-diastolic volume could increase during PHIPS at the same estimated left ventricular filling pressure as control. It is well known that right ventricular volume and function can alter left ventricular compliance. Since mean transmural pulmonary arterial pressures and transpulmonary pressure are unaltered by PHIPS, it is unlikely that right ventricular afterload or volume change significantly. Also, since our subjects had severely reduced left ventricular performance, not responsive to volume loading, it is unlikely that any small increase in left ventricular volume would significantly improve cardiac output. That Pes may not accurately reflect juxtacardiac pleural pressure during periods of increased intrathoracic pressure has been suggested by Craven and Wood; however, they found that the discrepancies in pressures were small if PEEP was not applied, and thus we believe that our measurement of mean Pes accurately reflects true increases in intrathoracic pressure.

Possibly our measurement of cardiac output is inaccurate. Measurements by thermodilution have been shown to vary throughout the respiratory cycle; however, we have previously shown that these variations are minimized in volume overload states and at respiratory frequencies of 20/min or greater (unpublished data). Since all of our subjects were in volume overloaded conditions and being ventilated at a frequency of 20/min, it is unlikely that significant variations in measurements of cardiac output by thermodilution occurred.A second possible mechanism which may account for our results is that PHIPS increases left ventricular contractility. This could occur if there were reflex increases in sympathetic tone or increased coronary blood flow. Since the high levels of vasopressor infusion were held constant and did not reverse the cardiac depression in our subjects prior to PHIPS, it is doubtful that either it or further increases in endogenous sympathetic tone could improve left ventricular contractility significantly. Increasing coronary blood flow increases cardiac performance. Since PHIPS was associated with an increase in mean arterial pressure of approximately 8 mm Hg, coronary blood flow could have increased; however, this increase in mean arterial pressure was offset by an associated increase in intrathoracic pressure; therefore, it is not clear whether PHIPS would be associated with any increase in coronary blood flow or contractility.

A third possibility, and one that we favor, is that PHIPS, by increasing intrathoracic pressure, augments left ventricular ejection by reducing left ventricular afterload. During conditions in which intrathoracic pressure changes widely, Permutt has suggested that transmural left ventricular pressure (left ventricular pressure minus intrathoracic pressure) more accurately reflects left ventricular afterload than does systolic aortic pressure alone. We used transmural mean arterial pressure to estimate this pressure load and found (Fig 2B) that PHIPS did not change transmural mean arterial pressure. Thus, the actual pressure load (transmural pressure) seen by the left ventricle during PHIPS is similar to the control steps pressure load. Since the pressure surrounding arterial pressurethe heart during PHIPS rises relative to the atmosphere, systolic ejection occurs against a rising mean arterial pressure but a constant transmural mean arterial pressure. It thus appears that in our subjects, PHIPS increased cardiac output and mean arterial pressure by decreasing left ventricular afterload, allowing the left ventricle to increase its output up to the given transmural mean arterial pressure to which it is limited.

The results of this study in patients are consistent with the results of our previous study in a canine model of acute ventricular failure, where increasing intrathoracic pressure increased left ventricular stroke work at a constant filling pressure; however, the exact mechanisms involved in this augmentation are still not clear. Not all of our subjects improved with PHIPS. In subject 1, this was probably due to binder-induced decreases in venous return, as manifested by a decrease in transmural pulmonary arterial occlusion pressure (Table 2); however, in another patient (patient

3), the reasons for the lack of improvement with PHIPS are not clear. Since PHIPS significantly increases mean airway pressure, there is a possibility that barotrauma could develop if this therapy was extended indefinitely. One patient was subsequently given PHIPS for three hours without untoward effects; however, this form of therapy is still experimental. It is not clear that this is the optimum method of improving left ventricular function with increased intrathoracic pressure.

In patients with pulmonary edema having decreased myocardial reserve and increased work of breathing, large negative swings in pleural pressure which accompany spontaneous respiration will aggravate left ventricular dysfunction by increasing left ventricular afterload. Placing such patients on intermittent posi-tive-pressure ventilation may improve cardiac efficiency and may increase cardiac output if venous return is not significantly decreased nor right ventricular function impaired by increased pulmonary vascular resistance. Perhaps there is a subgroup of critically ill patients being artificially ventilated who cannot be weaned from the ventilator because spontaneous breathing by decreasing intrathoracic pressure increases left ventricular afterload through negative swings in pleural pressure. This interaction may be important in the management of critically ill patients and suggests that spontaneous ventilatory efforts (ie, intermittent mandatory ventilation) could be deleterious to the patient with an unstable cardiovascular system until this instability is corrected.


Figure 1. Effect of PHIPS on cardiac output for all subjects. Circles are mean values for group (with SE bars). Cardiac output increases significantly with PHIPS from initial control values (p<0.05).


Figure 2. A (left), Effect of PHIPS on mean arterial pressure for all subjects. Circles are mean values for group (with SE bars). Mean arterial pressure increased significantly with PHIPS from initial control values (p<0.01). В (right), Effect of PHIPS on transmural mean arterial pressure (MAP-Pes) for all subjects. Circles are mean values for groups (with SE bars). Asterisk indicates estimated transmural mean arterial pressure using same Pes as initial control step. There is no significant difference in mean values among three steps.


Figure 3. Effect of PHIPS on cardiac performance. M UGA scan showing end-diastolic volumes and time-activity curve for patient 4. Control values were obtained prior to PHIPS. Note that PHIPS was associated with improved ejection fraction (EF) and decrease in end-diastolic volume while increasing cardiac output. End-systolic volume also decreases in the setting of an increase in mean arterial pressure, but a decrease in transmural arterial pressure. IPPV, Intermittent positive-pressure ventilation.

Table 2—Hemodynamic Data

Transmural Transmural Pulmonary Pulmonary Arterial Patient Cardiac Arterial Arterial Occlusion and Output, Pressure, Pressure, Pressure, Condition L/min mm Hg mm Hg mm HgPatient 1Control 4.6 76/42 25/17 16 End-inspiratory Airway Pressure, cm H2042 End-inspiratory/ End-expiratory Pes, mm Hg4/—6 Mean Pes, mm Hg0
PHIPS 4.4 74/48 23/12 10 75 23/10 12
Control 4.2 70/40 25/20 22 42 4/—6 0
ratient lControl 5.1 40/22 32/18 16 45 5/2 3
PHIPS 6.4 42/24 30/17 14 65 12/-1 6
Control 4.8 40/22 30/17 16 50 8/2 3
P&tient 3Control 2.9 42/20 32/17… 55 7/-14 -8
PHIPS 2.9 42/28 30/17… Control 2.9 40/20 30/15… 70 17/-10 0
Patient 4Control 2.7 70/30 25/16 16 45 12/ —3 2
PHIPS 3.3 70/35 31/20 17 68 25/0 8
Control 2.8 65/35 32/17 16 48 12/—6 0
Jr&tient оControl 2.2 63/50 42/22 20 55 12/ —3 2
PHIPS 3.0 80/60 45/22 22 72 20/0 8
Control 2.0 60/52 45/22 20 60 12/—4 1
ratient оControl 4.1 65/53 52/35 30 45 8/-1 1
PHIPS 5.3 82/65 47/35 32 70 16/0 7
Control 4.0 65/55 48/35 30 48 7/-3 -1
Patient 7Control 3.2 105/65 25/18 12 48 0
PHIPS 3.7 125/75 28/18 10 75 5
Control 3.3 110/60 25/16 12 50 0



Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

Blog at

Up ↑

%d bloggers like this: