Ashbaugh and colleagues reported the use of positive end expiratory pressure (PEEP) to treat hypoxia in patients with the adult respiratory distress syndrome in 1967. Despite its beneficial effects on the lung, PEEP decreased cardiac index, perhaps because of the increase in mean airway pressure described by Coumand and colleagues in 1948. However, the effects of PEEP on the heart remain controversial. Increased airway pressure shifted the central blood volume from the intrathoracic to the extrathoracic capacitance vessels. Increased pleural pressures compressed the heart, reduced left ventricular afterload and tended to reduce left ventricular volumes. Elevated alveolar pressures increased pulmonary vascular resistance and right ventricular afterload, and decreased filling of the left ventricle. Therefore, PEEP increased the afterload of the right ventricle and decreased the afterload of the left ventricle. PEEP may also depress ventricular function. Recent studies suggest that PEEP and volume loading increase right ventricular afterload, shift the interventricular septum to the left and compromise left ventricular function. Other studies found that PEEP released humoral agents which decreased cardiac contractility, altered coronary blood flow and induced myocardial ischemia.
This study was instituted to evaluate the effects of PEEP on ventricular function and myocardial metabolism in patients recovering uneventfully from elective coronary bypass surgery. In the early postoperative period, myocardial metabolism is vulnerable to hemodynamic stresses, and the application of PEEP in this setting may uncover subtle hemodynamic and myocardial metabolic abnormalities. Nuclear ventriculo-graphic examination was performed with a portable gamma camera to provide an estimate of ventricular volumes and ejection fraction. The coronary sinus was catheterized to evaluate coronary blood flow and myocardial metabolism. The protocol was designed to evaluate both the effect of increasing levels of PEEP from 5 to 10, and then to 15 cm H20, and the response to volume loading at 5 and 15 cm H20 PEEP.
Materials and Methods
Fifty patients undergoing elective coronary bypass surgery agreed to participate in an evaluation of postoperative myocardial function and metabolism. Additional instrumentation and the postoperative protocol were explained to each patient, who then signed a consent form approved by the Institutional Human Experimentation Committee. The patients were 53 ±7 years old, bad angina pectoris resistant to medical therapy, double or triple vessel coronary artery disease (2.7 ±0.5 diseased vessels) and normal ventricular function (ejection fraction 60 ±10 percent) at preoperative cardiac catheterization. Coronary artery bypass grafts (3.4 ±0.8 grafts per patient) were performed with cold crystalloid cardioplegia and a prolonged cross-clamp technique (cross-damp time 63 ±16 min), as recently reported. PEEP was evaluted in the intensive care unit 4 to 8 hours after cross-clamp removal, when rewarming was complete and after discontinuation of vasodilator medications.
All patients were ventilated with a volume respirator at a tidal volume of 12 mL/kg and 5 cm HtO PEEP. The initial measurements were made at 5 cm HsO PEEP. PEEP level was then increased to 10 cm H20; 10 min later, measurements were repeated. Fifteen minutes later, PEEP level was increased to 15 cm HtO, and measurements were repeated after 10 min. Arterial blood gas measurements revealed normal partial pressures of C02 and a partial pressure of oxygen greater than 100 mm Hg at each level of PEEP.
In 21 of the 50 patients, volume loading was performed to assess ventricular function at 5 and again at 15 cm HsO PEEP by infusing 250 ml (mL) of whole plasma to raise the left atrial pressure 2 to 4 mm Hg. The initial volume loading study was performed 30 min before the initial (5 cm H*0) PEEP measurements. The second volume loading study was performed 25 min after the application of 15 cm HaO PEEP. The remaining 29 patients were not volume loaded, and were studied only at 5,10 and 15 cm HsO PEEP.
The following catheters were inserted intraoperatively: radial arterial, left atrial, thermodilution pulmonary arterial and thermodilution coronary sinus. Hemodynamic measurements taken included: pulse, right atrial pressure, left atrial pressure, systolic and mean radial arterial pressure, and cardiac output Derived hemodynamic measurements included: cardiac index, stroke index and rate-pressure product, by standard formulae. Because pericardial pressures were not measured, transmural pressures such as pulmonary arterial pressure could not be reliably calculated. Arterial and coronary sinus blood samples were analyzed for Po* Pco2, and pH levels; oxygen saturation; (Co-oximeter Instrumentation Laboratories, Lexington, MA) and lactate (Rapid Lactate Stat Pack, Cal-biochemical-Behring, Lajolla, CA). Oxygen content (O^con) was derived from the hemoglobin (Hb) level, oxygen saturation (S) level and the Po* level by the formula:
Right and left ventricular myocardial performances (the relation between cardiac index or stroke index to the right and left end diastolic volume indices) were assessed by evaluating the response to volume loading at 5 and 15 cm H,0 PEEP by paired analysis of covariance. Left ventricular systolic function (the relation between the systolic blood pressure and the left ventricular end systolic volume index) was evaluated for each volume loading episode as an index of ventricular function, which was independent of preload and incorporated afterload.
Statistical analysis was performed using the Statistical Analysis Systems programs (SAS Institute Inc, Cary, NC). The hemodynamic, scintigraphic and myocardial metabolic parameters measured at 5,10 and 15 cm HsO PEEP were evaluated by a repeated-measures analysis of variance (ANOVA), and differences were specified by Duncans multiple range test1 when ANOVA was significant (p<0.05). The hemodynamic, scintigraphic and metabolic measurements before and after volume loading were compared by paired t-tests and by a two-way analysis of variance, with differences specified by Duncans multiple range f-test when ANOVA was significant (p<0.05). Differences in myocardial performance and systolic (unction between 5 and 15 cm H20 PEEP were assessed by analysis of covariance, employing the general linear models’ procedure,23 which permitted a simultaneous assessment of both slope and position. The differences between those patients who had improvement in myocardial metabolism with PEEP and those who had deterioration were assessed by unpaired t-tests and a two-way analysis of variance. Mean and standard error of the mean are depicted in the figures, and the mean and standard deviation are presented in the tables and text.