We studied the effects of cytomegalovirus (CMV) infection on 301 cardiac transplant recipients who were treated during the cyclosporine era of immunosuppression (1980 to the present). These patients received varying combinations of cyclosporine, azathioprine, prednisone, rabbit antithymocyte globulin, and OKT3 as their immunosuppressive therapy. Two hundred ten patients were free of CMV infection (non-CMV group). During the same period CMV infection developed in 91 patients, as manifested by a fourfold IgG serologic titer rise, demonstration of CMV inclusion bodies in tissue, or positive cultures for the virus (CMV group). The rate of graft rejection was significantly higher in the CMV group. Graft atherosclerosis was significantly more severe in the CMV group as judged by angiographic criteria or by pathologic study. Patient survival rates were significantly lower in the CMV group. Death caused by graft atherosclerosis was significantly more common among patients in the CMV group. Finally, the graft loss rate (from either death or retransplantation for atherosclerosis) was significantly greater in the CMV group. These data demonstrate that CMV infection in cardiac transplant recipients is associated with more frequent rejection, graft atherosclerosis, and death.

The shape of the coronary arterial pressure-flow relationship results from the interaction of a number of poorly understood physiological factors. Experiments in which coronary inflow and outflow pressures were coupled so that driving pressure was held constant showed that changes in inflow or outflow pressures altered coronary blood flow: coronary vascular resistance varied inversely with changes inflow pressure below 50 mm Hg and with changes in outflow pressure below 80 mm Hg. The magnitude of the influence of inflow pressure on resistance also depended on the fixed level of outflow pressure, the influence being large when the outflow pressure was low, and small when it was high. Inflow and outflow pressures, then, are two physiological factors which are determinants of the shape of the pressure-flow relationship, and their interaction contributes to the degree of curvature found in a particular relationship. These findings suggest that the use of linear regression in the interpretation of pressure-flow relationships results in poor estimation of resistance and zero-flow pressure. Other experiments measuring regional coronary blood flow using radionuclide-labeled microspheres resulted in the same inverse relationship between inflow pressure and resistance, regardless of mural depth, indicating that inflow pressure may influence resistance by distending vessels, rather than by causing sequential cessation of perfusion in successive transmural layers.

SUMMARY. When steady state pressure-flow relations are studied in the circumflex coronary artery, pressure gradients develop between it and other branches of the left coronary artery. To assess the effects of these pressure gradients, we compared the pressure axis intercept and shape of steady state circumflex pressure-flow relations in the presence and absence of gradients after autoregulation was abolished, both in the beating heart and during long diastoles in dogs. We used peripheral coronary pressures and radionuclide-labeled microspheres to assess arterial collateral flow. In the beating heart, interarterial pressure gradients reduced the curvature at low circumflex pressures, and overestimated the mean pressure axis intercept by 7.8 mm Hg (P < 0.05). The results were similar for the pressure-flow relations derived during long diastoles. This overesrimation exaggerates the difference between the pressure axis intercept and coronary sinus pressure. The peripheral coronary pressure and microsphere results indicate that these effects are mediated largely by arterial collateral flow. {Circ Res 56: 11-19, 1985) CORONARY arterial pressure-flow relations have been used to investigate the physical forces that regulate coronary blood flow (Cross et al., 1961;Downey and Kirk, 1975;Bellamy, 1978). The pressure axis intercept of these relations has been interpreted as the downstream pressure opposing forward flow in the coronary circulation; however, this pressure has greatly exceeded the coronary sinus pressure, which traditionally has been assumed to be the downstream pressure. The shape of these relations, found to be linear by some (Dole and Bishop, 1982;Eng et al., 1982) but curvilinear by others (L'Abbate et al., 1980;Klocke et al., 1981), is thought to convey information about the conductance of the coronary circulation. Because of this variability of the pressure axis intercept and the shape of coronary arterial pressure-flow relations, the physiological significance of both remains uncertain.One factor that might account for this variability is the specific coronary artery in which the pressureflow relation is obtained. In our laboratory, we have consistently found that the pressure axis intercept of a steady state circumflex pressure-flow relation (Verrier et al., 1980;Vlahakes et al., 1982) is much higher than that in a steady state left main coronary arterial pressure-flow relation (Rouleau et al., 1979;Uhlig et al., 1984). A possible explanation for the higher pressure axis intercept of the circumflex pressure-flow relation may be that when circumflex pressure is selectively lowered below that in the other branches of the left coronary artery, interarterial pressure gradients occur and are accompanied by collateral flow that is not measured by the upstream flow transducer. This error in measurement could underestimate total circumflex flow and thus overestimate the pressure axis intercept. Whether these pressure gradients also affect the shape of a circumflex pressure-flow relation is unknown.To assess...

The role of cardiac interstitial adenosine as an important metabolite in coronary autoregulation has not been established. We therefore measured steady-state cardiac interstitial adenosine concentration at a high and a low coronary inflow pressure using an epicardial diffusion well in anesthetized dogs. Although coronary resistance for the high and low pressure points showed highly significant differences (P less than 0.001), adenosine averaged 302 +/- 98 and 286 +/- 91 (SD) pmol/ml for the high and low pressure points, respectively (P greater than 0.20). Cardiac interstitial adenosine concentration was then measured with and without an intracoronary infusion of adenosine deaminase catalytic subunit. Adenosine averaged 28 +/- 21 (SD) pmol/ml during the infusion compared with 281 +/- 68 during control conditions (P less than 0.001). Finally, pressure-flow relations were obtained with and without the adenosine deaminase infusion, and there was no loss of autoregulation in the pressure of adenosine deaminase. These findings indicate that intracoronary adenosine deaminase markedly reduces interstitial adenosine concentration, that cardiac interstitial adenosine concentration remains constant during autoregulation, and that the coronary bed autoregulates normally when interstitial adenosine is reduced to levels close to zero. We conclude that cardiac interstitial adenosine concentration is not an important component in coronary autoregulation.