In vivo direct measurement of myocardial oxygen consumption in different animal models

Abstract

Myocardial oxygen consumption (MVO2) is an important indicator of cardiac performance reflecting oxygen supply and demand. It is a valuable physiological tool as MVO2 provides a quantifiable link between myocardial performance and total metabolism. Indirect measurement of MVO2 relies on extrapolation of hemodynamic parameters, as a reasonable correlation exists between MVO2 and measurements based on the left ventricular pressure-volume relationship. However, multiple factors influence oxygen consumption and must be taken into account when MVO2 is estimated. Major determinants of MVO2 include myocardial mass, heart rate, contractility, and wall stress, with minor influences from preload, basal oxygen requirement, and activation energy. MVO2 can be directly determined from a variety of methods, but requires measurement of myocardial blood flow by some process. Microsphere technology allows calculation of myocardial blood flow, but requires organ harvest. Direct catheterization of the coronary sinus allows determination of coronary sinus blood flow and MVO2 is then calculated as the product of coronary blood flow and coronary arterio-venous oxygen content difference, which is divided by heart rate to yield MVO2 per beat. Myocardial oxygen consumption was calculated through measurement of myocardial blood flow using colored microsphere technology to determine the effect of sevoflurane on cardiac performance in ferrets. A dose-dependent decrease in arterial blood pressure, left ventricular pressure, systemic vascular resistance, aortic flow, and dp/dt (an index of contractility) was detected as expired concentration of sevoflurane increased. Heart rate, central venous pressure, coronary vascular resistance, myocardial oxygen extraction ratio, and T (the time constant of relaxation) were unchanged. Cardiac external work decreased, as did myocardial oxygen consumption, causing increased cardiac efficiency at higher concentrations of sevoflurane. Sevoflurane caused minimal and predictable cardiovascular effects in ferrets without increasing myocardial metabolic demands. Data obtained from this study have not been previously reported for a species that is being commonly used in cardiovascular research. Colored microsphere technology was also used to determine myocardial oxygen consumption in a canine model of altered splanchnic blood flow. A model was developed for this study to examine the direct cardiovascular effects of two pathophysiological events (gastric ischemia and portal hypertension) that occur during gastric dilatation-volvulus (GDV), a severe clinical syndrome of the canine patient, independent of the effect of other events that overtly influence hemodynamics (caudal vena cava occlusion and thoracic impingement). The hypothesis of this study is that the alterations in splanchnic blood flow that occur early in the course of GDV cause early intrinsic cardiac damage due to interference with myocardial energy transfer. This damage may make the heart more susceptible to injury from further hemodynamic and hypoxic insults. It was found that portal hypertension and gastric ischemia caused an increase in MVO2, while cardiac external work was maintained. Therefore, a less efficient transfer of energy occurred in the myocardium following alteration of splanchnic blood flow. We speculate that reduction of cardiac efficiency could be an early event that occurs in GDV that may predispose these dogs to further cardiac dysfunction. The use of MVO2 is of considerable value in determining the effectiveness of medical and surgical interventions for the treatment of heart failure. Numerous heart failure models in a variety of species have been developed, all with their unique advantages and disadvantages. However, none of the models are the ideal reproduction of the human cardiac failure. Anthracycline-induced cardiac injury is most commonly performed using Adriamycin. This model not only replicates the clinical occurrence of doxorubicin-induced cardiotoxicity observed in cancer patients, but also causes severe dilated cardiomyopathy and heart failure that is seen in clinical dilated cardiomyopathy of various etiologies. A previously established protocol for Adriamycin-induced cardiomyopathy was altered in an effort to create a more consistent and less lethal model of heart failure. Adjustments that were enacted included a decrease in the cumulative dose of Adriamycin, reduction in number of intracoronary injections, and distribution of Adriamycin injection between the descending branch and the circumflex branch of the left coronary artery. The protocol of 22.5mg Adriamycin in 60mL saline administered intracoronary weekly for 2 weeks with concurrent administration of verapamil resulted in a predictable and severe model of dilated cardiomyopathy with reduced overall mortality compared to a previously published protocol. This protocol for induction of heart failure described here is relatively fast, technically simple, well tolerated by the animal, and has an acceptable mortality rate. Previous studies have demonstrated that in dogs with Adriamycin-induced cardiomyopathy, an unstimulated skeletal muscle wrap placed around the heart maintains cardiac efficiency and cardiac functional reserve while allowing reverse ventricular remodeling by preventing further ventricular dilatation, reducing afterload, and preserving diastolic function. As there is significant morbidity associated with latissimus dorsi muscle harvest, the objective of the last experiment was to determine whether a prosthetic cardiac wrap would have the same function as an unstimulated muscle wrap. The hypotheses were that in a canine model of cardiomyopathy, prosthetic cardiac binding would provide a myocardial sparing effect, preserve cardiac functional reserve, and would not interfere with diastolic function. Twelve normal adult large-breed dogs were used in this study and cardiomyopathy was induced in all dogs by the protocol described above. Three weeks after the last injection of Adriamycin, left and right heart catheterization was performed and echocardiography was repeated. Dogs were then randomly split into two groups with one group to undergo prosthetic cardiac binding (treatment group) and the other group to have no treatment (control group). Treatment dogs underwent surgery one week after baseline cardiac catheterization. Through a median sternotomy, the pericardium was resected at the level of the atrioventricular groove and replaced with polypropylene mesh. The mesh was sutured to the remaining pericardium and was in direct contact with the epicardium of the right and left ventricles. Echocardiography was performed three and six weeks after surgery and a terminal cardiac catheterization was performed at six weeks postoperatively. Left ventricular dimensions were unchanged in the treatment group from the time of surgery until 6 weeks postoperatively compared to control dogs whose parameters significantly increased during that same time frame (p = 0.005). However, no difference in cardiac function was found between treatment and control dogs. Therefore, while prosthetic cardiac binding may limit progressive cardiac dilatation, no improvement in cardiac function is attained with this procedure in dogs with Adriamycin-induced cardiomyopathy.