| ||||||||||||||
|
|
|||||||||||||
Department of Surgical and Perioperative Science, Anaesthesiology and Intensive Care, Umeå University Hospital, Umeå, Sweden
Address correspondence and reprint requests to Michael Broomé, Department of Cardiothoracic Surgery and Anesthesia, Huddinge University Hospital, S-141 86 Stockholm, Sweden. Address e-mail to michael.broome{at}thsurg.hs.sll.se
| Abstract |
|---|
|
|
|---|
IMPLICATIONS: The positive inotropic and chronotropic effects of angiotension II (Ang II) seen in previous studies seem to be mediated via extracardiac actions of Ang II. Coronary vascular tone is not affected by local Ang II infusion in anesthetized pigs.
| Introduction |
|---|
|
|
|---|
Whether the myocardial effects of Ang II are locally or centrally mediated (or both) is not clear. A facilitating effect on sympathetic transmission has been described both in the central nervous system (Ang II infusion in cerebral arteries without systemic effects) (9,10) and in peripheral tissues (1113). The effects seen in myocardial tissue preparations seem to be mediated, at least in part, via direct action on myocardial angiotensin receptors, because they can be blocked by specific AT1 receptor antagonists (5,14). The nonspecific positive inotropic effect of increased heart rate (HR), known also as the Treppe mechanism (15) or the force-frequency relationship, may also contribute to inotropic effects during HR increases (16). Intracoronary Ang II potentiates coronary sympathetic vasoconstriction in humans with coronary artery disease (17), facilitates norepinephrine release during sympathetic nerve stimulation in dogs (12), and increases coronary perfusion pressure in a Langendorff rabbit preparation with constant coronary flow (18).
The specific aim of our study was to compare the myocardial systolic and diastolic effects seen when infusing Ang II in the left coronary artery (LCA) with the findings from previous studies with systemic infusion, using a dose range that included smaller doses, presumed to be free of systemic effects, up to larger doses in which systemic circulatory responses could be provoked. We also aimed to measure the locally mediated effects of Ang II infusion on coronary vascular resistance and flow.
| Methods |
|---|
|
|
|---|
Anesthesia
Animals were premedicated with azaperone (Stresnil®; Janssen-Cilag, Beerse, Belgium) 2 mg/kg body weight and ketamine (Ketalar®; Parke-Davis, Somerville, NJ) 10 mg/kg body weight IM. Anesthesia was induced by using sodium pentobarbital (Pentobarbitalnatrium; Apoteksbolaget, Stockholm, Sweden) as a bolus of 15 mg/kg IV followed by an infusion at 1520 mg · kg-1 · h-1. No muscle relaxants were used. Tracheotomy was performed and the pigs were mechanically ventilated with 25%30% oxygen in air by using a volume-cycled ventilator (Siemens 900B; Siemens-Elema, Stockholm, Sweden). Ventilation was adjusted to normocapnia as judged by repeated arterial blood gas analyses (ABL 5; Radiometer, Copenhagen, Denmark).
The IV infusion protocol consisted of acetated Ringers solution at 510 mL · kg-1 · h-1 using a stable normal urine output and stable blood pressures as endpoints (19). Core body temperature was maintained between 37.5° and 38.5°C by using warm infusions and heating blankets.
Instrumentation
All intravascular catheters were inserted via cutdowns. An external carotid arterial catheter was used to measure mean arterial blood pressure (MAP) and to sample blood. A left external jugular venous three-lumen catheter (Arrow, Walpole, MA) was used to monitor central venous pressure (CVP) and to administer fluids and drugs. A 7F, four-lumen, flow-directed pulmonary artery thermodilution catheter (American-Edwards, Irvine, CA) was inserted via the left internal jugular vein and advanced to an occlusion position in the pulmonary arterial tree, where the balloon was deflated and the catheter secured. This catheter was used to measure mean pulmonary arterial pressure and to determine cardiac output (CO) as the average of three measurements within a 10% range (Wetenschappelijk Technische Instituut, Gravensand, The Netherlands).
A 7F 2-thermistor coronary sinus (CS) thermodilution catheter (CCS-7 U-90A; Webster Labs, Diamond Bar, CA) was placed in the CS with fluoroscopic guidance through the right internal jugular vein. The correct position was confirmed with a bolus of radiographic solution (Omnipaque; Nycomed, Asker, Norway). All pressures were monitored by using membrane pressure transducers (Ohmeda, Helsinki, Finland).
A combined tip manometer and conductance pigtail catheter (ANP:269; Sentron BV, Roden, The Netherlands) was placed in the left ventricle via a 7F sheath introducer in the right internal carotid artery using fluoroscopic guidance. The catheter was positioned along the left ventricular long axis with its tip in the apex, such that a stable sinus rhythm was maintained and the segmental signals were optimized.
A complete description of our use of conductance volumetry method has been published previously (20). In brief, it can be described as a nearly continuous (sampling frequency of 200 Hz in our setup) measurement of ventricular volume achieved through measurement of 45 segmental ventricular electrical conductances. The method is first calibrated through measurement of the conductivity of blood and subtraction of the electrical conductance of surrounding structures (parallel conductance) calculated after data sampling during an injection of hypertonic saline (21,22). After this calibration, CO was calculated as the sum of the averaged measured segmental volume changes during a single heartbeat multiplied by the HR. This value was compared with CO measured by thermodilution to obtain a gain correction factor,
(21).
Ventricular and blood conductivity measurements were made by using a signal processor (Leycom Sigma 5/DF; Cardiodynamics, Leiden, The Netherlands). Dual excitation was used with a field ratio of 0.3 (23). Blood conductivity, parallel conductance measurements, and calibration against thermodilution CO were performed before each dose-response series.
A 7F embolectomy catheter (Fogarty; Baxter Healthcare Corp., Deerfield, IL) was placed in the inferior vena cava via the right external jugular vein for the purpose of producing a rapid, controlled, and transient decrease in venous return through inflating the balloon for <6 s for measurement sequences. If changes in HR or ectopic beats were seen during inflation of the balloon, these data were discarded.
Finally, the main stem of the LCA was catheterized with a specially designed commercially available human coronary angiography catheter via a 7F introducer placed in the right internal carotid artery using fluoroscopic guidance. The position of the catheter was verified with a contrast injection and then secured.
Animals were killed at the end of experiments using potassium chloride and deepened anesthesia. Hemodynamic data were continuously recorded by using a 16-channel amplifier/polygraph (TA 5000; Gould, Valley View, OH) as well as a computer-based multichannel signal acquisition and analysis system (Biopac MP100 and Acknowledge III; Biopac Systems, Santa Barbara, CA). Conductance volumetry and tip manometry data were recorded and processed in digital form by using another commercial software (PC Conduct version 720.1; Cardiodynamics).
Study Protocol
Before starting this protocol, a separate protocol involving the administration of short-acting vasoactive drugs with repeated blood sampling and hemodynamic measurements was conducted. No additional instrumentation was involved in the previous protocol. A lengthy resting period was observed to assure stability and a resting circulation at the start of this protocol.
After a baseline control measurement (C1), Ang II (Apoteket AB, Stockholm, Sweden) was infused in the LCA in incremental doses (2.5 µg/h [A1], 10 µg/h [A2], and 40 µg/h [A3]). Measurements were repeated at each dose level. Hemodynamic steady state was reached within <5 min at each constant Ang II infusion rate, because the plasma half-life of Ang II is <1 min (24). Measurements were made 10 min after the start of each infusion rate. Doses were chosen to achieve pharmacologic local plasma concentrations with systemic effects only at the largest dose based on our previous experience (8). After finishing Ang II infusion, a 30-min recovery period ensued and a second control measurement (C2) was done.
HR was derived from the electrocardiogram. Systemic vascular resistance (SVR) was calculated as the difference between MAP and CVP divided by CO as measured with the thermodilution method and data from the pulmonary arterial catheter. Coronary vascular resistance (Rcs) was calculated as the difference between MAP and CVP divided by CS flow (Qcs). Arterial elastance (Ea) was derived from the conductance catheter as the ratio of ventricular end-systolic pressure to stroke volume (25). End-systolic elastance (Ees) was calculated as a linear regression of the end-systolic pressure volume points during controlled preload reduction (26). Preload recruitable stroke work (PRSW) was calculated as a linear regression of each beats stroke work and end-diastolic volume (EDV) during controlled preload reduction (27).
The time constant of isovolumetric relaxation,
(
1/2), was measured with a monoexponential fit of pressure data from the tip manometer in the left ventricle according to published algorithms, assuming a zero asymptote (28).
Coronary flow was calculated according to previously described algorithms using the thermodilution CS flow measurements (29,30). Minimal coronary plasma concentrations (CAngII) of Ang II were approximated by using measured CS flow (Qcs), measured arterial hemoglobin levels (Hb [g/L]), and the infused doses (D) of Ang II, and for the sake of simplicity, disregarding recirculation.
|
|
All results were analyzed by using repeated measures analysis of variance (SPSS® version 10.0 for Windows®; SPSS Inc., Chicago, IL). When significant main effects were found, planned simple contrasts relative to the preceding control were made. A simple Students t-test was used to compare control values before and after the administration of Ang II. A P value < 0.05 was considered significant.
| Results |
|---|
|
|
|---|
(
1/2), did not change significantly. No significant changes were seen in CS flow (Qcs) or coronary resistance (Rcs) (Fig. 2).
|
|
|
|
| Discussion |
|---|
|
|
|---|
We have also shown that intracoronary infusion of Ang II, which produced large normal to supranormal local plasma concentrations of Ang II, did not cause significant changes in coronary blood flow or coronary vascular resistance. Coronary vasoconstrictive effects have been reported in other species (18,31). Previous studies have suggested that Ang II effects may be mediated, at least in part, by increased sympathetic nerve system activity (12,17). In this pentobarbital anesthetized pig model, autonomic nervous system activity presumably was blunted. Therefore, if a coronary vasoconstrictive effect of Ang II is mediated through local effects on autonomic nervous structures, it is possible that such effects cannot be seen in our model. Also the response pattern in one earlier study in isolated rabbit hearts (31) showed a biphasic response over time, with an initial coronary vasoconstriction and then a slight coronary dilatory effect after a few minutes. The possibility of species-specific differences is always present. Finally, little is known about the Ang II receptor population in porcine hearts, although Ang II receptors exist (32). It is possible that mechanisms other than receptor-mediated vasoconstriction are also important for coronary resistance.
To distinguish direct myocardial and other local effects from remote effects, namely central nervous and autonomic nervous activity, one must either deliver the drug locally in the effector organ circulation in doses not leading to systemic circulatory effects or by other means isolate the myocardium from remote neural activity. If the drug is delivered locally in an intact animal, then both local effects mediated via sympathetic nerve endings and direct stimulation of myocardial Ang II receptors could be provoked and observed. However, if no myocardial effects were seen with local administration, that would suggest an extracardiac site of action.
In a previous study (8), we have shown a positive inotropic effect of Ang II when infused IV in doses 12.5200 µg/h. This achieved measured systemic plasma concentrations that were considerably smaller (802800 pg/mL) than the estimated minimal local intracoronary concentrations in this study (80012 000 pg/mL). Because Ang II concentrations in CS blood were not measured directly, this estimation was the best available means to compare local doses. The absolute dose ranges overlap and the increase in HR, MAP, and the tendency to increase CO at the largest dose in this study correspond to the changes seen in the previous study and may therefore be explained by systemic effects of Ang II.
Calculated minimal coronary concentrations at the largest dose 40 µg/h were several times larger than is seen in states of severe hypovolemic shock (33). We have attempted to explore the entire local myocardial physiologic dose-response range of Ang II.
Left ventricular pressure volume loop analysis during different combinations of pacing and Ang II infusion is a possible way to differentiate between the nonspecific inotropic effect of increased HR and rate-independent positive inotropic actions, although pacing rates need to be quite high, because the Ang II-induced HR in our previous study (8) exceeded 150 bpm in some animals.
Although the inotropic actions of Ang II seem to be mediated via extracardiac mechanisms, there still is a possibility that the positive chronotropic action of Ang II seen in our previous study (8) is mediated via local cardiac actions. Because HR is totally dependent on sinus node activity as long as the animal is in sinus rhythm, right coronary infusion of Ang II would be required to test this hypothesis, because the sinus node perfusion is most often supplied by the right coronary artery both in pig and in humans (34).
The anatomical presumption for this model is that the LCA supplies the majority of the left ventricle. During placement of the coronary catheter, the anatomy was evaluated with contrast injections. No major anatomical aberrations were seen during this procedure, implying that all pigs had a normal anatomy, with the left anterior descending branch supplying the apical region and the anterior wall, and the circumflex branch supplying the lateral wall and major parts of the posterior left ventricular wall.
Analysis of the pressure volume relationship may provide the most useful description of cardiac function available today (35). These measures are strongly, but not totally load independent. As an example, PRSW may be less influenced by changes in afterload than Ees (36). In our study, the changes in loading conditions have been small. It is therefore unlikely that our choice of methods has limited our possibilities to detect a local inotropic action of Ang II.
There are several other possible ways to differentiate between the suggested mechanisms for the inotropic action of Ang II, although they all have some inherent limitations. Infusion of Ang II in animals with pharmacologic blockade of
and ß receptors could clarify whether the effects are mediated via mechanisms involving these receptors, but cannot discriminate whether cardiac or extracardiac receptors are the most important.
A thoracic epidural or a pharmacologic ganglionic blockade possibly could discriminate between effects mediated via an increase in central sympathetic tone and direct cardiac actions, but cannot easily discriminate between intracardiac facilitation of norepinephrine release (12) and stimulation of angiotensin receptors on the cardiac myocytes (14).
Pentobarbital-based anesthesia probably has a consistent negative effect on myocardial performance, as well as blunting of autonomic nervous system activity. The model provided that this probable circulatory effect of pentobarbital was consistent throughout the protocol. Minimal changes in HR, blood pressures, and EDVs between controls suggested a constant anesthetic level and intravascular volume stability.
We conclude that, because no effect on systolic and diastolic myocardial function or coronary vascular tone was seen with infusion of Ang II in the LCA, the effects seen with systemic Ang II infusion in previous studies are likely mediated via Ang II action at extracardiac sites. One probable mechanism is central facilitation of sympathetic nervous system activity, but this issue needs to be specifically investigated in further studies.
| Acknowledgments |
|---|
The authors thank Anna-Maja Sundin and Tomas Ekman for excellent technical assistance throughout the study.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
D. Konrad, M. Haney, G. Johansson, M. Wanecek, E. Weitzberg, and A. Oldner Cardiac effects of endothelin receptor antagonism in endotoxemic pigs Am J Physiol Heart Circ Physiol, August 1, 2007; 293(2): H988 - H996. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. Konrad, A. Oldner, M. Wanecek, A. Rudehill, E. Weitzberg, B. Biber, G. Johansson, S. Haggmark, and M. Haney Positive inotropic and negative lusitropic effects of endothelin receptor agonism in vivo Am J Physiol Heart Circ Physiol, October 1, 2005; 289(4): H1702 - H1709. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|