Warm heart surgery.

Hypothermia is widely acknowledged to be the fundamental component of myocardial protection during cardiac operations. Although it prolongs the period of ischemic arrest by reducing oxygen demands, hypothermia is associated with a number of major disadvantages, including its detrimental effects on enzymatic function, energy generation, and cellular integrity. We hypothesized that the ideal protected state of the heart would be electromechanically arrested and perfused with blood, that is, aerobic arrest. Under these conditions the fundamental need for hypothermia becomes questionable. We have developed a novel approach to myocardial protection during cardiac operations based on these concepts, in which the chemically arrested heart is perfused continuously with blood and maintained at 37 degrees C. In 121 consecutive coronary bypass procedures we have compared this approach with a historical cohort of 133 consecutive patients treated with hypothermic cardioplegia. Perioperative myocardial infarction was significantly less prevalent (1.7% versus 6.8%; p less than 0.05) in the warm cardioplegic group, as was the use of the intraaortic balloon pump (0.9% versus 9.0%; p less than 0.005) and the prevalence of low output syndrome (13.5% versus 3.3%; p less than 0.005). Cardiac output immediately after bypass was significantly higher than before bypass (3.1 +/- 0.9 versus 4.9 +/- 1.0 L/min; p less than 0.001) only in the warm cardioplegia group. Furthermore, the heartbeat in 99.2% of patients treated with continuous warm cardioplegia converted to normal sinus rhythm spontaneously after removal of the aortic crossclamp compared with only 10.5% of the hypothermic group. The time from removal of the aortic crossclamp to discontinuation of cardiopulmonary bypass (i.e., reperfusion time) was significantly shorter in the warm cardioplegia group (11 +/- 4.3 versus 27 +/- 5.6 minutes; p less than 0.001). Our results suggest that continuous normothermic blood cardioplegia is safe and effective. Conceptually, this represents a new approach to the problem of maintaining excellent myocardial preservation during cardiac operations.

Myocardial protection in the hypertrophied right ventricle.

Hypertrophied right ventricle presents a sensitive state that may not be adequately protected by modern cardioplegic methods. Cardiac metabolism, performance, and ultrastructure were measured in response to 1 hour of cardioplegic arrest in 15 pigs with right ventricular hypertrophy using intermittent hypothermic crystalloid, blood, and Flusol DA 20%-based cardioplegia. Reperfusion time was 1 hour. One hour after a 60-minute cross-clamp period, there were no differences in light microscopy. Total energy stores increased in 4 of 5 animals given blood cardioplegia compared with 1 of 5 for each of the other groups. Cardiac performance data also showed better results for animals treated with blood cardioplegia. After 30 minutes of reperfusion, animals receiving blood cardioplegia recovered 131% +/- 42% of preoperative systolic performance compared with 106% +/- 49% for Fluosol-treated animals and only 82% +/- 27% recovery for the crystalloid-treated group. After 60 minutes of reperfusion, the blood group showed 119% +/- 20% recovery compared with 89% +/- 23% and 85 +/- 50% recovery for Fluosol- and crystalloid-treated hearts, respectively. In conclusion, blood cardioplegia provided better protection than did crystalloid or Fluosol DA 20% cardioplegia when animals with right ventricular hypertrophy underwent 1 hour of cardioplegic arrest. It may have repaired damaged myocardium, leaving better hearts after cross-clamping than before.

A simple model of right ventricular hypertrophy.

Previous models of right ventricular hypertrophy (RVH) created by pulmonary artery (PA) banding in adult large animals have been associated with an unpredictable response of the right ventricle to the band and a high mortality due to the variable degree of acute stenosis. We studied the efficacy of PA banding in young pigs to produce RVH by progressive gradual stenosis during growth. Sixteen Yorkshire pigs at 6 weeks of age had nonconstricting 5-mm wide Dacron strips placed around the PA via a left minithoracotomy. The animals were returned for study in 2-3 months. There were no deaths during the growth period. Five sham-operated pigs acted as controls. Right ventricular free wall (RVFW) to total heart weight ratio was greater in the banded group (0.38 +/- 0.05 vs. 0.28 +/- 0.01, P less than .005) as was the RVFW to left ventricular free wall (LVFW) weight (1.09 +/- 0.25 versus 0.66 +/- 0.03, P less than .005). While the LVFW to total heart weight ratio decreased (0.36 +/- 0.04 vs 0.45 +/- 0.05, P less than .005), the septal ratio did not change (0.26 +/- 0.04 vs. 0.29 +/- 0.02, NS), indicating concomitant septal hypertrophy. This technique is simple, reliable, and reproducible in creating right ventricular and septal hypertrophy in the young pig with no mortality during maturation.

Potential deleterious effect of beta-adrenergic stimulation during warm-blood cardioplegia in rabbit hearts.

We hypothesized that beta-adrenergic stimulation with isoproterenol during continuous normothermic cardioplegic arrest would enhance the regenerative and regulatory function of the myocardium, resulting in improved cardiac function. We studied isolated rabbit hearts paced at approximately 200 beats per minute (bpm) and perfused by a support rabbit. We measured ventricular pressure over a range of ventricular volumes to determine maximal elastance (Emax) at baseline and 20 and 45 min after discontinuation of cardioplegia. Myocardial oxygen consumption (MVO2) measurements were performed simultaneously and during cardioplegic arrest. Hearts were prospectively randomized to receive either isoproterenol at 0.1 M or control in blinded fashion for 10 min during a 1-h continuous warm-blood cardioplegic arrest. Compared to control hearts, isoproterenol-treated hearts had trends toward longer time to first spontaneous heartbeat (control 141 +/- 43 vs. isoproterenol 200 +/- 74 s, p = .07), and longer time to capture of atrial pacing (control 214 +/- 52 vs. isoproterenol 288 +/- 91 s, p = .06). There was no difference observed in the MVO2 between isoproterenol-treated and control groups of hearts. MVO2 decreased during cardioplegia (p < .01), but there was no significant change in MVO2 during isoproterenol infusion during cardioplegic arrest. There was a significant reduction in Emax compared to baseline 20 min after discontinuation of cardioplegic arrest in both groups (control 7.3 +/- 1.7 mm Hg/microL vs. 9.0 +/- 1.7 mm Hg/microL, p = .02, isoproterenol-treated 6.8 +/- 2.8 mm Hg/microL vs. 8.2 +/- 2.6 mm Hg/microL, p = .01, respectively), with recovery of Emax by 45 min in control hearts only. We conclude that exposure of hearts to isoproterenol during warm cardioplegic arrest has a deleterious effect that may be mediated through mechanisms independent of increased myocardial oxygen consumption.

Role of carbon dioxide oscillation in the control of respiration in the anesthetized dog.

In order to examine the role of respiratory oscillation of PaCO2 (CO2 oscillation) in the control of respiration, we performed veno-venous bypass using a membrane lung in 10 anesthetized paralyzed dogs, where the dog was put on fixed mechanical ventilation so that we could keep average PaCO2 and PaO2 constant by adjusting FICO2 and FIO2 during CO2 loading/unloading. By venous CO2 loading/unloading we could widely change the CO2 output from the lung (11-440% of the control) resulting in large changes in the arterial CO2 oscillations (50-280% of the control for the maximum rate of rise of PaCO2 and 40-350% of the control for the maximum rate of fall of PaCO2), which was measured by a rapidly responding intra-arterial pH electrode. Despite such wide variations in CO2 oscillations we did not find any consistent change in the respiratory center output (minute phrenic activity). This held true after compensating the changes in the minute phrenic activity for the CO2 sensitivity of each individual dog. Thus, the present results may suggest that the CO2 oscillation plays little role in the control of respiration in mild increase in VCO2.