Classic Papers Revisited: An Early Study of Cardioprotection by Volatile Anesthetics

Recovery of Contractile Function of Stunned Myocardium in Chronically Instrumented Dogs Is Enhanced by Halothane or Isoflurane. By Warltier DC, al-Wathiqui MH, Kampine JP, and Schmeling WT. Anesthesiology 1988; 69:552–65. Reprinted with permission. Following brief periods (5–15 min) of total coronary artery occlusion and subsequent reperfusion, despite an absence of tissue necrosis, a decrement in contractile function of the postischemic myocardium may nevertheless be present for prolonged periods. This has been termed “stunned” myocardium to differentiate the condition from ischemia or infarction. Because the influence of volatile anesthetics on the recovery of postischemic, reperfused myocardium has yet to be studied, the purpose of this investigation was to compare the effects of halothane and isoflurane on systemic and regional hemodynamics following a brief coronary artery occlusion and reperfusion. Nine groups comprising 79 experiments were completed in 42 chronically instrumented dogs. In awake, unsedated dogs a 15-min coronary artery occlusion resulted in paradoxical systolic lengthening in the ischemic zone. Following reperfusion active systolic shortening slowly returned toward control levels but remained approximately 50% depressed from control at 5 h. In contrast, dogs anesthetized with halothane or isoflurane (2% inspired concentration) demonstrated complete recovery of function 3–5 h following reperfusion. Because the anesthetics directly depressed contractile function, additional experiments were conducted in which a 15-minute coronary artery occlusion was produced during volatile anesthesia; however, each animal was allowed to emerge from the anesthetized state at the onset of reperfusion. Similar results were obtained in these experiments, demonstrating total recovery of contractile function within 3–5 h following reperfusion. Thus, despite comparable degrees of contractile dysfunction during coronary artery occlusion in awake and anesthetized dogs, the present results demonstrate that halothane and isoflurane produce marked improvement in the recovery of segment function following a transient ischemic episode. Therefore, volatile anesthetics may attenuate postischemic left ventricular dysfunction occurring intraoperatively and enhance recovery of regional wall motion abnormalities during reperfusion.

Muscle metaboreflex activation during dynamic exercise evokes epinephrine release resulting in β2-mediated vasodilation

Muscle metaboreflex-induced increases in mean arterial pressure (MAP) during submaximal dynamic exercise are mediated principally by increases in cardiac output. To what extent, if any, the peripheral vasculature contributes to this rise in MAP is debatable. In several studies, we observed that in response to muscle metaboreflex activation (MMA; induced by partial hindlimb ischemia) a small but significant increase in vascular conductance occurred within the nonischemic areas (calculated as cardiac output minus hindlimb blood flow and termed nonischemic vascular conductance; NIVC). We hypothesized that these increases in NIVC may stem from a metaboreflex-induced release of epinephrine, resulting in β2-mediated dilation. We measured NIVC and arterial plasma epinephrine levels in chronically instrumented dogs during rest, mild exercise (3.2 km/h), and MMA before and after β-blockade (propranolol; 2 mg/kg), α1-blockade (prazosin; 50 μg/kg), and α1 + β-blockade. Both epinephrine and NIVC increased significantly from exercise to MMA: 81.9 ± 18.6 to 141.3 ± 22.8 pg/ml and 33.8 ± 1.5 to 37.6 ± 1.6 ml·min−1·mmHg−1, respectively. These metaboreflex-induced increases in NIVC were abolished after β-blockade (27.6 ± 1.8 to 27.5 ± 1.7 ml·min−1·mmHg−1) and potentiated after α1-blockade (36.6 ± 2.0 to 49.7 ± 2.9 ml·min−1·mmHg−1), while α1 + β-blockade also abolished any vasodilation (33.7 ± 2.9 to 30.4 ± 1.9 ml·min−1·mmHg−1). We conclude that MMA during mild dynamic exercise induces epinephrine release causing β2-mediated vasodilation.

Renal effects induced by prolonged mPGES1 inhibition

The importance of membrane-bound PGE synthase 1 (mPGES1) in the regulation of renal function has been examined in mPGES1-deficient mice or by evaluating changes in its expression. However, it is unknown whether prolonged mPGES1 inhibition induces significant changes of renal function when Na+ intake is normal or low. This study examined the renal effects elicited by a selective mPGES1 inhibitor (PF-458) during 7 days in conscious chronically instrumented dogs with normal Na+ intake (NSI) or low Na+ intake (LSI). Results obtained in both in vitro and in vivo studies have strongly suggested that PF-458 is a selective mPGES1 inhibitor. The administration of 2.4 mg·kg−1·day−1 PF-458 to dogs with LSI did not induce significant changes in renal blood flow (RBF) and glomerular filtration rate (GFR). A larger dose of PF-458 (9.6 mg·kg−1·day−1) reduced RBF ( P < 0.05) but not GFR in dogs with LSI and did not induce changes of renal hemodynamic in dogs with NSI. Both doses of PF-458 elicited a decrease ( P < 0.05) in PGE2 and an increase ( P < 0.05) in 6-keto-PGF1α. The administration of PF-458 did not induce significant changes in renal excretory function, plasma renin activity, and plasma aldosterone and thromboxane B2 concentrations in dogs with LSI or NSI. The results obtained suggest that mPGES1 is involved in the regulation of RBF when Na+ intake is low and that the renal effects elicited by mPGES1 inhibition are modulated by a compensatory increment in PGI2. These results may have some therapeutical implications since it has been shown that prolonged mPGES1 inhibition has lower renal effects than those elicited by nonsteroidal anti-inflammatory drugs or selective cyclooxygenase-2 inhibitors.

Muscle metaboreflex-induced increases in cardiac sympathetic activity vasoconstrict the coronary vasculature

Ischemia of active skeletal muscle evokes a powerful blood pressure-raising reflex termed the muscle metaboreflex (MMR). MMR activation increases cardiac sympathetic nerve activity, which increases heart rate, ventricular contractility, and cardiac output (CO). However, despite the marked increase in ventricular work, no coronary vasodilation occurs. Using conscious, chronically instrumented dogs, we observed MMR-induced changes in arterial pressure, CO, left circumflex coronary blood flow (CBF), and coronary vascular conductance (CVC) before and after α1-receptor blockade (prazosin, 100 μg/kg iv). MMR was activated during mild treadmill exercise by partially reducing hindlimb blood flow. In control experiments, MMR activation caused a substantial pressor response-mediated via increases in CO. Although CBF increased (+28.1 ± 3.7 ml/min; P < 0.05), CVC did not change (0.45 ± 0.05 vs. 0.47 ± 0.06 ml·min−1·mmHg−1, exercise vs. exercise with MMR activation, respectively; P > 0.05). Thus all of the increase in CBF was due to the increase in arterial pressure. In contrast, after prazosin, MMR activation caused a greater increase in CBF (+55.9 ± 17.1 ml/min; P < 0.05 vs. control) and CVC rose significantly (0.59 ± 0.08 vs. 0.81 ± 0.17 ml·min−1·mmHg−1, exercise vs. exercise with MMR activation, respectively; P < 0.05). A greater increase in CO also occurred (+2.01 ± 0.1 vs. +3.27 ± 1.1 l/min, control vs. prazosin, respectively; P < 0.05). We conclude that the MMR-induced increases in sympathetic activity to the heart functionally restrain coronary vasodilation, which may limit increases in ventricular function.

Cardiovascular responses to exercise and muscle metaboreflex activation during the recovery from pacing-induced heart failure

Rapid recovery of resting hemodynamics from tachycardia- or arrhythmia-induced heart failure (HF) has been demonstrated in both humans and animals. However, little is known about cardiovascular responses to exercise in animals or about reflex control of the cardiovascular system during exercise while recovering from HF. Inasmuch as the reduced cardiac output (CO) during exercise in HF has been shown to lead to underperfusion of active skeletal muscle and tonic activation of the muscle metaboreflex, an improved CO during exercise in subjects recovering from HF may lead to higher skeletal muscle blood flows and to relief of this metabolic stimulus. We investigated cardiovascular responses to graded treadmill exercise and metaboreflex activation [evoked by imposed graded reductions in hindlimb blood flow (HLBF) during mild and moderate exercise] in chronically instrumented dogs during control, mild to moderate HF (induced by rapid ventricular pacing), and recovery from HF. Most hemodynamic responses to graded exercise returned to control within 24 h of disconnecting the pacemaker. After 2 wk of recovery, CO and HLBF at each workload were significantly higher than control. In addition, whereas the increase in CO that normally occurs with metaboreflex activation was markedly attenuated in HF, it completely returned in the recovery experiments. We conclude that cardiovascular responses to graded exercise during the recovery from pacing-induced HF return rapidly to near or above control and that the increased CO and HLBF in recovery likely relieved the metabolic stimulus and tonic metaboreflex activation that may have occurred during moderate exercise in HF.

Altered cardiac metabolic phenotype after prolonged inhibition of NO synthesis in chronically instrumented dogs

Acute inhibition of nitric oxide (NO) synthase causes a reversible alteration in myocardial substrate metabolism. We tested the hypothesis that prolonged NO synthase inhibition alters cardiac metabolic phenotype. Seven chronically instrumented dogs were treated with Nω-nitro-l-arginine methyl ester (l-NAME, 35 mg·kg−1·day−1 po) for 10 days to inhibit NO synthesis, and seven were used as controls. Cardiac free fatty acid, glucose, and lactate oxidation were measured by infusion of [3H]oleate, [14C]glucose, and [13C]lactate, respectively. After 10 days of l-NAME administration, despite no differences in left ventricular afterload, cardiac O2 consumption was significantly increased by 30%, consistent with a marked enhancement in baseline oxidation of glucose (6.9 ± 2.0 vs. 1.7 ± 0.5 μmol·min−1·100 g−1, P < 0.05 vs. control) and lactate (21.6 ± 5.6 vs. 11.8 ± 2.6 μmol·min−1·100 g−1, P < 0.05 vs. control). When left ventricular afterload was increased by ANG II infusion to stimulate myocardial metabolism, glucose oxidation was augmented further in the l-NAME than in the control group, whereas free fatty acid oxidation decreased. Exogenous NO (diethylamine nonoate, 0.01 μmol·kg−1·min−1 iv) could not reverse this metabolic alteration. Consistent with the accelerated rate of carbohydrate oxidation, total myocardial pyruvate dehydrogenase activity and protein expression were higher (38 and 34%, respectively) in the l-NAME than in the control group. Also, protein expression of the constitutively active glucose transporter GLUT-1 was significantly elevated (46%) vs. control. We conclude that prolonged NO deficiency causes a profound alteration in cardiac metabolic phenotype, characterized by selective potentiation of carbohydrate oxidation, that cannot be reversed by a short-term infusion of exogenous NO. This phenomenon may constitute an adaptive mechanism to counterbalance cardiac mechanical inefficiency.

Nitroglycerin reduces myocardial oxygen consumption during exercise despite vascular tolerance

The long-term benefits of nitroglycerin (NTG) therapy are limited by the development of vascular tolerance and endothelial dysfunction in conductance coronary arteries. We have determined whether nitrate tolerance extends to NTG effects on myocardial O2 consumption (MV̇o2) and the ability of endogenous nitric oxide (NO) to modulate MV̇o2 during exercise. In chronically instrumented dogs ( n = 8), hemodynamic and MV̇o2 responses to treadmill exercise were measured before, during tolerance (3 and 7 days of NTG delivery), and 7 days after NTG withdrawal. Acute NTG delivery caused a parallel downward shift of the MV̇o2-triple product (TP) relations and reversed the disproportionate increases in MV̇o2 caused by the blockade of NO formation. After 7 days of continuous transdermal NTG delivery, vascular tolerance was displayed as a >75% reduction of coronary blood flow (CBF) responses to NTG boluses. Despite vascular nitrate tolerance, MV̇o2-TP relations were shifted downward compared with pre-NTG exercise. Seven days after NTG withdrawal, vascular responses to boluses of NTG had recovered from tolerance, and MV̇o2-TP relations during exercise were back to pre-NTG level. At that time, blockade of NO formation failed to alter MV̇o2-TP relations. Thus NTG caused a sustained reduction of cardiac MV̇o2, independent of metabolic demand during exercise, despite tolerance of the coronary microcirculation. NTG-induced vascular tolerance and MV̇o2 reductions were reversible by NTG withdrawal, but endogenous NO-dependent modulation of O2 consumption was severely impaired.

Attenuated arterial baroreflex buffering of muscle metaboreflex in heart failure

Previous studies have shown that heart failure (HF) or sinoaortic denervation (SAD) alters the strength and mechanisms of the muscle metaboreflex during dynamic exercise. However, it is still unknown to what extent SAD may modify the muscle metaboreflex in HF. Therefore, we quantified the contribution of cardiac output (CO) and peripheral vasoconstriction to metaboreflex-mediated increases in mean arterial blood pressure (MAP) in conscious, chronically instrumented dogs before and after induction of HF in both barointact and SAD conditions during mild and moderate exercise. The muscle metaboreflex was activated via partial reductions in hindlimb blood flow. After SAD, the metaboreflex pressor responses were significantly higher with respect to the barointact condition despite lower CO responses. The pressor response was significantly lower in HF after SAD but still higher than that of HF in the barointact condition. During control experiments in the barointact condition, total vascular conductance summed from all beds except the hindlimbs did not change with muscle metaboreflex activation, whereas in the SAD condition both before and after induction of HF significant vasoconstriction occurred. We conclude that SAD substantially increased the contribution of peripheral vasoconstriction to metaboreflex-induced increases in MAP, whereas in HF SAD did not markedly alter the patterns of the reflex responses, likely reflecting that in HF the ability of the arterial baroreflex to buffer metaboreflex responses is impaired.