Desflurane and Isoflurane Exert Modest Beneficial Actions on Left Ventricular Diastolic Function during Myocardial Ischemia in Dogs

Methods: Eighteen experiments were performed in nine dogs chronically instrumented for measurement of aortic and left ventricular pressure, cardiac output, LAD coronary blood flow velocity, and LAD and left circumflex coronary artery sub-endocardial segment length. Regional myocardial contractility was evaluated with the slope of the preload recruitable stroke work relationship determined from a series of left ventricular pressure-segment length diagrams in the LAD and left circumflex coronary artery zones. Diastolic function was assessed with a time constant of isovolumic relaxation (tau), maximum segment lengthening velocity in LAD and left circumflex coronary artery regions, and regional chamber stiffness constants derived using monoexponential and three-element exponential curve fitting in each zone. On separate experimental days, hemodynamics and indices of regional functional were obtained in the conscious state and during 1.1 and 1.6 minimum alveolar concentration end-tidal desflurane or isoflurane before and during LAD occlusion.

Results: In conscious dogs, LAD occlusion abolished regional stroke work, increased chamber stiffness (monoexponential: 0.39 plus/minus 0.04 during control to 1.34 plus/minus 0.39 mm sup -1 during LAD occlusion), and decreased the rate of early ventricular filling in the ischemic zone. These changes were accompanied by increased contractility (slope: 103 plus/minus 8 during control to 112 plus/minus 7 mmHg during LAD occlusion), rapid filling rate (maximum segment lengthening velocity: 46 plus/minus 5 during control to 55 plus/minus 7 mm *symbol* s sup -1 during LAD occlusion), and chamber stiffness (monoexponential: 0.43 plus/minus 0.05 during control to 1.14 plus/minus 0.25 mm sup -1 during LAD occlusion) in the normal region. Increases in tau were also observed in the conscious state during the period of myocardial ischemia. Desflurane and isoflurane increased tau and decreased the slope and maximum segment lengthening velocity in a dose-related manner. Monoexponential and three-element exponential curve fitting were unchanged by the volatile anesthetics in the absence of ischemia. Myocardial contractility and rapid filling rate were enhanced in the nonischemic region during LAD occlusion in the presence of desflurane and isoflurane. In contrast to the findings in the conscious state, ischemia-induced increases in tau and chamber stiffness in the ischemic and normal zones were attenuated during anesthesia induced by desflurane and isoflurane.

Conclusions: The results indicate that increases in contractility of remote myocardium during brief regional ischemia were preserved in the presence of desflurane and isoflurane anesthesia. In addition, desflurane and isoflurane blunted ischemia-induced increases in tau and regional chamber stiffness in both the ischemic and nonischemic zones. These results demonstrate that the volatile anesthetics may exert important beneficial actions on left ventricular mechanics in the presence of severe abnormalities in systolic and diastolic function during ischemia. (Key words: Anesthetics, volatile: desflurane; isoflurane. Heart, diastole: chamber stiffness; diastolic left ventricular function; isovolumic relaxation; ventricular compliance. Heart, myocardial performance: left ventricular function; myocardial contractility; preload recruitable stroke work.)

ABRUPT occlusion of a coronary artery causes nearly immediate alterations in regional systolic and diastolic function in the ischemic zone. Systolic segment lengthening, postsystolic shortening, and loss of effective segmental stroke work occur concomitantly with diastolic creep (increases in unstressed segment length), declines in early diastolic filling, and increased regional chamber stiffness. These mechanical effects are accompanied by increases in total segment shortening, myocardial contractility and rapid ventricular filling rate and decreases in diastolic distensibility in the surrounding nonischemic myocardium. Alterations in global isovolumic relaxation, left ventricular filling, and chamber compliance may also occur in response to acute interruption of coronary blood flow depending on the extent, intensity, and duration of the ischemic stimulus. .

The actions of volatile anesthetics on the mechanical consequences of acute myocardial ischemia have been inadequately studied. Several previous investigations from this and other laboratories have demonstrated that volatile anesthetics exert cardioprotective effects during myocardial ischemia and reperfusion. A variety of mechanisms have been postulated to account for the antiischemic actions of potent inhalational agents including improvement of myocardial oxygen supply-demand relationships caused by negative inotropic and lusitropic effects and favorable alterations in ventricular loading conditions, reduction of intracellular Calcium²+ overload via declines in myocardial Calcium²+ availability and reserve resulting from partial inhibition of Calcium²+ channel activity, and attenuation of the detrimental effects of oxygen-derived free radicals. This investigation tested the hypothesis that desflurane and isoflurane do not exacerbate the functional responses to brief (2 min) occlusion of the left anterior descending coronary artery in the central ischemic zone and further, that these volatile anesthetics do not adversely alter systolic and diastolic mechanical compensation for this process in the remote nonischemic zone in chronically instrumented dogs.

All experimental procedures and protocols used in this investigation were reviewed and approved by the Animal Care Committee of the Medical College of Wisconsin. All conformed to the Guiding Principles in the Care and Use of Animals of the American Physiologic Society and were in accordance with the Guide for the Care and Use of Laboratory Animals (DHEW [DHHS] publication no. [NIH] 85–23, revised 1985).

The methods used to surgically implant instruments for systemic and coronary hemodynamic monitoring have been described in detail elsewhere. Under general anesthesia and aseptic conditions, a thoracotomy was performed in the left fifth intercostal space. Heparin-filled catheters were placed in the descending thoracic aorta for measurement of aortic blood pressure and the right atrium for fluid or drug administration. An ultrasonic flow probe (Transonics, Ithaca, NY) was positioned around the ascending thoracic aorta for measurement of relative cardiac output (minus coronary blood flow). Two pairs of miniature ultrasonic segment length transducers (5 MHz) for measurement of changes in regional contractile function (percent segment shortening) were implanted within the left ventricular subendocardium in the perfusion territories of the left anterior descending (LAD) and the left circumflex coronary arteries (LCCA). A high-fidelity, miniature micromanometer (Model P7, Konigsberg Instruments, Pasadena, CA) was inserted in the left ventricular apex for measurement of continuous left ventricular pressure, the maximum rate of increase and decrease of left ventricular pressure (+dP/dt^maxand -dP/dt^min, respectively), and the rate of increase of left ventricular pressure at 50 mmHg (+dP/dt⁵⁰). A heparin-filled catheter was inserted in the left atrial appendage. The left ventricular micromanometer was cross-calibrated in vivo against pressures measured via arterial and left atrial catheters (Model P50 pressure transducer, Gould, Oxnard, CA).

A single-port, 16-gauge heparin-filled catheter was placed in the apex of the left thoracic cavity between the lung and the chest wall through the thoracotomy incision for subsequent measurement of continuous intrathoracic pressure. A 1.5–2-cm segment of the proximal LAD was isolated and a precalibrated Doppler ultrasonic flow transducer was placed around this vessel for measurement of diastolic coronary blood flow velocity. A miniature hydraulic vascular occluder (In Vivo Metric, Healdsburg, CA) was secured around the LAD distal to the flow transducer for subsequent production of acute coronary artery occlusion. A hydraulic vascular occluder was also placed around the inferior vena cava for abrupt alteration of left ventricular preload. All instrumentation was secured, tunneled between the scapulae, and exteriorized via several small incisions. The pericardium was left open, the chest wall closed in layers, and the pneumothorax evacuated by a chest tube. Each dog was fitted with a jacket (Alice King Chatham, Los Angeles, CA) to prevent damage to the instruments and catheters, which were housed in an aluminum box within the jacket pocket.

After surgery, each dog was treated with analgesics as needed (Innovar-Vet [fentanyl and droperidol]; Pitman-Moore, Mundelein, IL). Antibiotic prophylaxis consisted of intramuscular cephalothin (40 mg *symbol* kg sup -1) and gentamicin (4.5 mg *symbol* kg sup -1). Dogs were allowed to recover for a minimum of 7 days before experimentation. Segment length signals in the LAD and LCCA regions and LAD coronary blood flow velocity were driven and monitored by ultrasonic amplifiers (Crystal Biotech, Hopkinton, MA). End-systolic segment length (ESL) was determined at 10 ms before maximum negative left ventricular dP/dt and end-diastolic segment length (EDL) was determined 10 ms before dP/dt first exceeded 140 mmHg *symbol* s sup -1 (immediately before the onset of left ventricular isovolumic contraction). The lengths were normalized according to the method described by Theroux et al. Percent segment shortening (% SS) was calculated using the formula:% SS =(EDL - ESL)*symbol* 100 *symbol* EDL sup -1. Percent segment shortening during isovolumic contraction (% ISS) and left ventricular ejection (% ESS) in the nonischemic region were calculated using the equations:% ISS =(EDL - EIL)*symbol* 100 *symbol* EDL sup -1 and % ESS =(EIL - ESL)*symbol* 100 *symbol* EDL sup -1, where EIL = end-isovolumic contraction segment length determined at the time when left ventricular pressure exceeded aortic pressure during early systole. Systolic lengthening (% SL) and postsystolic shortening (% PSS) in the ischemic zone were calculated using the equations:% SL =(L^max- EDL)*symbol* 100 *symbol* EDL sup -1, where L^max= maximum segment length, and % PSS =(L^min- end-systolic segment length)*symbol* 100; end-systolic segment length sup -1, where L^min= minimum segment length, respectively. An estimate of myocardial oxygen consumption, the pressure work index, was determined using the formula of Rooke and Feigl. Relative diastolic coronary vascular resistance was calculated as the ratio of diastolic arterial pressure to diastolic coronary blood flow velocity. All hemodynamic data were continuously monitored on a polygraph (model 7758A; Hewlett-Packard, San Francisco, CA) and digitized via a computer interfaced with an analog to digital converter.

Dogs (N = 9; weight = 25.7 plus/minus 0.3 kg; mean plus/minus SEM) were randomly assigned to receive desflurane or isoflurane on separate experimental days. Each dog was fasted overnight, and fluid deficits were replaced before experimentation with 500 ml normal saline, which was continued at 3 ml *symbol* kg sup -1 *symbol* h sup -1 for the duration of each experiment. After the instrumentation was calibrated, baseline systemic and coronary hemodynamics were recorded in the conscious state. Left ventricular pressure, intrathoracic pressure, and LAD and LCCA segment length waveforms were recorded continuously on the digital oscilloscope for later off-line analysis of diastolic function. Regional myocardial contractility was evaluated using two series of left ventricular pressure-segment length diagrams in the LAD and LCCA zones generated by abrupt constriction of the inferior vena cava, resulting in a reduction of left ventricular systolic pressure by approximately 30 mmHg over 10–15 cardiac cycles (Figure 1). Left ventricular pressure-segment length diagrams were rejected if heart rate increased more than 10% above baseline levels during the occlusion. In this case, pressure-length diagrams were repeated after steady-state hemodynamics had been re-established. Respiratory variation in left ventricular pressure in the conscious state was later reduced off-line by electronically subtracting the continuous intrathoracic pressure waveform from the left ventricular pressure waveform as previously described. The resultant left ventricular transmural pressure-segment length diagrams were used to assess contractile state in conscious dogs in two regions of the heart. The inferior vena caval occlusion was released immediately after recording pressure-length diagrams. During desflurane and isoflurane anesthesia, hemodynamic waveforms and left ventricular pressure-segment length diagrams were recorded in the LAD and LCCA regions at end expiration.

After obtaining data under control conditions, the LAD was acutely occluded using the chronically implanted miniature hydraulic vascular occluder. The left ventricular pressure-segment length diagram in the ischemic region rapidly developed characteristic loss of effective stroke work, dyskinetic systolic lengthening, postsystolic shortening, and diastolic creep (Figure 2). Systemic and coronary hemodynamics and pressure-length waveforms and diagrams were recorded after 2 min of LAD occlusion in the conscious state. Myocardial contractility in the LAD region was not evaluated during acute LAD occlusion because no stroke work was performed by this segment during ischemia. After data had been recorded, the LAD zone was reperfused by releasing the hydraulic occluder. Rapid recovery of regional systolic and diastolic function to preocclusion values occurred in all experiments.

All dogs underwent inhalation induction with desflurane or isoflurane in oxygen followed by tracheal intubation. Anesthesia was maintained with 1.1 and 1.6 minimum alveolar concentration (MAC) end-tidal desflurane or isoflurane in a random manner in a nitrogen (79%) and oxygen (21%) mixture. End-tidal anesthetic concentrations of desflurane and isoflurane were measured at the tip of the endotracheal tube by an infrared anesthetic analyzer (Datex Capnomac, Helsinki, Finland) calibrated for detection of desflurane. The infrared analyzer was calibrated with known standards before and during experimentation. The canine MAC values for desflurane and isoflurane used in this investigation were 7.2 and 1.28%, respectively. Each MAC level was maintained for 30 min. Systemic and coronary hemodynamics were then recorded and left ventricular pressure-segment length wave forms and diagrams were acquired in the manner described earlier before and during LAD occlusion. Arterial blood gases were maintained at conscious levels by adjusting respiratory rate and nitrogen and oxygen concentrations during each experiment. The anesthetic was discontinued and emergence allowed to occur after each experiment was completed. Dogs recovered for at least 2 days before subsequent experimentation. A total of 18 experiments in two separate groups (desflurane and isoflurane) were performed in which the same 9 dogs were studied.

The slope (M^w) of the regional preload recruitable stroke work relationship was used to determine myocardial contractility in the LAD and the LCCA regions as previously described. Briefly, in the conscious state and during anesthetic interventions, left ventricular pressure-segment length diagrams were obtained in the LAD and LCCA regions by transient constriction of the inferior vena cava. The area of each diagram, corresponding to segmental stroke work, was calculated by electronic integration and was plotted against the corresponding EDL for each loop. Linear regression analysis was used to describe the preload recruitable stroke work relationship slope (M^w) and length intercept (L^w): stroke work = M^w*symbol*(EDL - L^w). During acute LAD occlusion, no effective regional stroke work was performed in this region (F2-16) and the preload recruitable stroke work relationship was not calculated. The time constant of isovolumic relaxation (tau) was described assuming a non-zero asymptote of ventricular pressure decay using the method of Raff and Glantz. Left ventricular negative dP/dt was plotted against ventricular pressure in 2-ms intervals between peak negative dP/dt and 5 mmHg above end-diastolic pressure to yield tau as the negative inverse of the slope. Maximum segment lengthening velocity during rapid ventricular filling (dL/dt^max) was determined by differentiation of the continuous segment length waveform. Regional chamber stiffness constants were derived in each region by a simple monoexponential equation (K^p) and a three-element exponential relation (K^s) using left ventricular pressure-segment length data between minimum ventricular pressure and the beginning of atrial systole: P = D *symbol* e (sup Kp *symbol* L) and P = A *symbol* e (sup Ks *symbol* L)+ C, respectively, where P = left ventricular pressure, K^pand K sub s = regional diastolic chamber stiffness constants, L = segment length, and A, C, and D are curve fitting constants. Best-fit iterations were used to determine K^svia the Marquardt-Levenberg algorithm with commercially available software (SigmaPlot version 2.0, Jandel, San Rafael, CA).

Statistical analysis of data within and between groups in the conscious state, during anesthetic interventions, and during brief LAD occlusion was performed with multiple analysis of variance repeated measures followed by application of Student's t test with Bonferroni's correction for multiplicity. Changes within and between groups were considered statistically significant when the P value was < 0.05. All data are expressed as mean plus/minus SEM.

Brief, 2-min occlusion of the LAD caused significant (P < 0.05) increases in heart rate, rate-pressure product, pressure-work index, and left ventricular end-diastolic pressure in conscious dogs (Table 1). No change in mean arterial pressure, left ventricular systolic pressure, cardiac output, and systemic vascular resistance occurred. End-systolic and end-diastolic segment length increased and percent segment shortening was abolished in the ischemic zone during brief LAD occlusion (Table 2). Systolic lengthening and postsystolic shortening also increased in response to brief ischemia (systolic lengthening and postsystolic shortening = 0.1 plus/minus 0.1 and 0.2 plus/minus 0.2 during control to 4.8 plus/minus 0.6 and 11.9 plus/minus 0.7% during LAD occlusion, respectively). Significant increases in end-diastolic segment length and the length intercept of the preload recruitable stroke work relationship (L^w; 12.1 plus/minus 0.8 during control to 13.6 plus/minus 0.7 mm during LAD occlusion;Table 3) in the nonischemic region were also observed during LAD occlusion, suggesting that preload was enhanced in the LCCA segment. Brief LAD occlusion increased total percent segment shortening and percent segment shortening during isovolumic contraction in the LCCA perfusion territory, however, percent segment shortening during left ventricular ejection was unchanged, implying that the regional contribution to left ventricular ejection by the nonischemic zone was unaffected by coronary occlusion and that a fraction of augmented total segment shortening was wasted on expansion of the adjacent ischemic zone during isovolumic contraction. The slope of the preload recruitable stroke work relationship in the nonischemic region increased in response to LAD occlusion (M^w; 103 plus/minus 8 during control to 112 plus/minus 7 mmHg during LAD occlusion;T3-16), indicating enhanced contractility in the LCCA zone. No changes in left ventricular +dP/dt^maxand +dP/dt sub 50 were observed during coronary artery occlusion, suggesting that global myocardial contractility was unaffected by the brief LAD occlusion (T1-16). However, these isovolumic indices of myocardial contractility are relatively insensitive to regional changes in contractile function.

Alterations in diastolic function also occurred during LAD occlusion. An increase in the time constant of isovolumic relaxation was observed during coronary artery occlusion (tau; 35 plus/minus 1 during control to 38 plus/minus 1 ms during LAD occlusion), consistent with modest prolongation of this phase of diastole (T1-16). No change in -dP/dt^minwas observed, however. The rate of rapid ventricular filling was decreased in the ischemic zone (LAD dL/dt^max; 52 plus/minus 8 during control to 36 plus/minus 6 mm *symbol* s sup -1 during LAD occlusion;T2-16) and enhanced in remote normal myocardium (LCCA dL/dt^max; 42 plus/minus 4 during control to 51 plus/minus 7 mm *symbol* s sup -1 during LAD occlusion;T3-16). Significant increases in regional chamber diastolic stiffness constants (K^pand K^s) were observed in both the ischemic and the nonischemic zones during coronary artery occlusion (LAD and LCCA K^p: 0.39 plus/minus 0.04 and 0.43 plus/minus 0.05 during control to 1.34 plus/minus 0.39 and 1.14 plus/minus 0.25 mm sup -1 during LAD occlusion, respectively;T2-16and T3-16), consistent with declines in segmental compliance.

The effects of desflurane and isoflurane on systemic and coronary hemodynamics are summarized in T1-16and Table 4, respectively. The end-tidal anesthetic concentrations in the current investigation were 7.89 plus/minus 0.10 and 11.50 plus/minus 0.12% for desflurane and 1.41 plus/minus 0.02 and 2.03 plus/minus 0.03% for isoflurane. Desflurane caused significant increases in heart rate, rate-pressure product, pressure-work index, and left ventricular end-diastolic pressure and dose-related decreases in mean arterial pressure, left ventricular systolic pressure, and percent segment shortening in the LAD and LCCA regions before coronary artery occlusion. Decreases in stroke volume also occurred, but cardiac output and systemic vascular resistance were unchanged by desflurane. Diastolic coronary blood flow velocity remained constant and diastolic coronary vascular resistance was reduced at 1.6 MAC desflurane. Dose-dependent decreases in M^win the both the LAD and LCCA perfusion territories, +dP/dt^max, and +dP/dt⁵⁰occurred in desflurane-anesthetized dogs, and consistent with the known negative inotropic effects of this agent. Desflurane also produced dose-related negative lusitropic effects as indicated by increases in tau and decreases in dL/dt^maxin the LAD and LCCA zones before LAD occlusion. Desflurane caused no changes in K^por K^s.

Isoflurane caused systemic and coronary hemodynamics effects that were similar to those produced by desflurane (T4-16). Isoflurane increased heart rate and decreased mean arterial pressure, left ventricular systolic pressure, diastolic coronary vascular resistance and percent segment shortening in the LAD and LCCA regions. Isoflurane-induced reductions in mean arterial pressure and left ventricular systolic pressure were significantly greater than those observed with desflurane at 1.6 MAC, however. In contrast to the findings with desflurane, isoflurane decreased pressure-work index at 1.6 MAC. Unlike desflurane, isoflurane decreased cardiac output and systemic vascular resistance. Isoflurane depressed myocardial contractility (M^win the LAD and LCCA regions, +dP/dt^maxand +dP/dt⁵⁰), prolonged isovolumic relaxation (increased tau and decreased the magnitude of -dP/dt^min), and attenuated the early ventricular filling rate. Isoflurane caused no changes in K^por K^sin the LAD and LCCA zones.

In the presence of desflurane and isoflurane, LAD occlusion caused increases in left ventricular end-diastolic pressure, end-systolic and end-diastolic segment lengths in the ischemic region (T2-16and Table 5), and LCCA end-diastolic segment length (T3-16and Table 6). The degree of ischemia resulting from a 2-min occlusion of the LAD was functionally similar in conscious and anesthetized dogs as quantified by the degree of changes in percent segment shortening, systolic lengthening, and postsystolic shortening. Significant increases in total percent segment shortening and percent segment shortening during isovolumic contraction were observed in the LCCA perfusion territory concomitant with negative segment shortening (aneurysmal bulging) in the ischemic zone (T3-16and T6-16). However, percent segment shortening during left ventricular ejection in the nonischemic region remained unchanged by coronary occlusion in the presence of desflurane and isoflurane, findings that were similar to those observed in the conscious state. No changes in heart rate, mean arterial pressure, left ventricular systolic pressure, cardiac output, systemic vascular resistance, and stroke volume were observed during coronary artery occlusion in anesthetized dogs (T1-16and T4-16). In contrast to findings in conscious dogs, brief LAD occlusion during anesthesia did not alter the rate-pressure product or the pressure-work index. Increases in LCCA M^w(e.g., 64 plus/minus 4 before coronary occlusion compared to 80 plus/minus 6 mmHg during occlusion at 1.6 MAC desflurane) were also observed with brief ischemia in the presence of anesthesia, indicating that regional myocardial contractility was enhanced in the nonischemic zone. Relative increases in LCCA M^win response to regional ischemia were similar in anesthetized versus conscious dogs (Figure 3). No change in +dP/dt^maxand +dP/dt⁵⁰was observed with LAD occlusion, however, suggesting that global myocardial contractility was unaffected. In contrast to the findings in the conscious state, no change in tau was observed during LAD occlusion in dogs anesthetized with desflurane or isoflurane (F3-16). The early ventricular filling rate decreased in the LAD zone (36 plus/minus 5 before LAD occlusion compared to 22 plus/minus 3 mm *symbol* s sup -1 during LAD occlusion at 1.6 MAC desflurane) and increased in the LCCA zone (25 plus/minus 3 before LAD occlusion compared to 32 plus/minus 5 mm *symbol* s sup -1 during LAD occlusion at 1.6 MAC desflurane) in response to regional ischemia, findings similar to those observed in the conscious state (Figure 4). However, increases in K^p(Figure 5) and K^sin the normal and ischemic zones produced by brief LAD occlusion were attenuated in desflurane- and isoflurane-anesthetized dogs compared to conscious control subjects. No differences in alterations of left ventricular systolic and diastolic function by ischemia were observed between the desflurane and isoflurane groups.

Regional myocardial ischemia resulting from sudden coronary artery occlusion produces a series of characteristic systolic and diastolic mechanical effects in both the central ischemic zone and remote normal myocardium. Loss of effective stroke work, dyskinetic systolic lengthening, and postsystolic shortening occur within 60 s after the interruption of coronary flow. This contractile dysfunction is accompanied by diastolic creep (increased unstressed segment length), decreases in regional diastolic distensibility, and attenuated early diastolic filling rates in the ischemic zone. Rapid remodeling of nonischemic myocardium occurs as a compensatory response to the mechanical burden imposed by the ischemic area. Increases in total segment shortening are observed in the normal zone that are directly proportional to the size of the ischemic region. However, no change occurs in ejection segment shortening in the normal zone because a portion of the total shortening is expended in stretching the ischemic zone during isovolumic contraction. Myocardial contractility may also increase in the nonischemic region, presumably via a Frank-Starling mechanism. An increase in myocardial stiffness in normal myocardium also occurs in response to regional ischemia. The etiology of this decrease in compliance at a site remote from the ischemic locus remains unclear; however, this phenomenon may occur as a result of physiologic adaptation to altered global ventricular shape resulting from regional ischemia or tethering of the nonischemic myocardium to the adjacent ischemic zone. Enhanced early peak filling rates have also been observed in normal myocardium during regional ischemia, effects that have been proposed to result from the combined actions of declines in segmental compliance and increases in preload and contractility. Global left ventricular function may also be affected depending on the severity and duration of the ischemic stimulus and the quantity of affected myocardium. Transient but significant increases in tau and decreases in +dP/dt^maxand the magnitude of -dP/dt^minoccur within seconds after the onset of proximal coronary occlusion. These effects are probably related to marked regional contractile dyssynchrony produced by the acutely ischemic zone. These alterations in global systolic and diastolic function partially resolve within 60 s after the onset of regional ischemia consistent with rapid, compensatory remodeling of nonischemic myocardium. .

A substantial body of experimental evidence supports the contention that volatile anesthetics exert beneficial effects during myocardial ischemia and reperfusion injury. Halothane has been shown to attenuate ST segment changes caused by brief coronary artery occlusion. Decreased lactate production by isolated canine hearts with a severe LAD stenosis has been observed during enflurane anesthesia when perfusion pressure was artificially controlled. Halothane-induced reductions in regional contractile function were well tolerated and did not precipitate frank systolic dysfunction during a severe LAD constriction in open-chest, acutely instrumented dogs. Halothane also reduced myocardial infarct size after LAD occlusion in dogs. Potent inhalational anesthetics have been shown to enhance recovery of systolic function of stunned myocardium in the isolated heart and when these agents were administered before and during, but not after, brief periods of myocardial ischemia in vivo, effects that were accompanied by preservation of high energy phosphate levels. In addition, halothane preserved contractile function and ultrastructural integrity in isolated rat hearts during reperfusion after normothermic cardioplegic arrest. Because volatile anesthetics appear to produce protective effects during acute myocardial ischemia and reperfusion, the current investigation was designed to examine the hypothesis that desflurane and isoflurane do not adversely alter the systolic and diastolic mechanical responses to regional myocardial ischemia in the central ischemic and remote nonischemic zones.

The current findings indicate that a brief, 2-min LAD occlusion produces dyskinetic early systolic segment lengthening, postsystolic shortening, and loss of effective stroke work in the ischemic zone. The degree of ischemia produced by progressive coronary artery constriction and complete occlusion can be quantified by the degree of systolic lengthening and postsystolic shortening in the ischemic zone. These variables were similar in the LAD region in the conscious state and in the presence of desflurane or isoflurane anesthesia, suggesting that the relative intensity of the ischemia in this zone was similar in the conscious and anesthetized states. This conclusion requires qualification, however, because regional myocardial metabolism was not specifically measured in the ischemic zone. Left anterior descending coronary artery ischemia increased total segment shortening in the normal zone, however, ejection shortening was unaffected, indicating that a percentage of total LCCA shortening was wasted tethering the adjacent ischemic muscle. These findings are consistent with the studies of Lew et al. and Noma et al. Desflurane and isoflurane did not alter the percentage of normal zone shortening expended during isovolumic contraction during the 2-min LAD occlusion. Compensatory increases in LCCA myocardial contractility were observed in the conscious state and were preserved during desflurane and isoflurane anesthesia. The results confirm and extend the findings of Lowenstein et al. and indicate that like halothane, desflurane and isoflurane do not adversely affect the systolic compensatory responses of normal myocardium to distant ischemia. The increase in myocardial contractility observed in the nonischemic zone during regional ischemia was similar in the presence and absence of anesthetics, suggesting that the Frank-Starling mechanism, the presumed cause of increase in intrinsic inotropic state under these conditions, remains fully intact despite the inherent myocardial depressant properties of desflurane and isoflurane. .

Brief LAD occlusion caused diastolic creep (as indicated by increases in L^w), reduced the rate of early ventricular filling, and increased regional chamber stiffness in the ischemic zone, confirming the results of several previous investigations. These changes in ischemic zone diastolic behavior were accompanied by a modest but significant prolongation of isovolumic relaxation and increased the early ventricular filling rate and regional chamber stiffness in the nonischemic region in conscious dogs, verifying the observations of Kumada et al., Takahashi et al., and Marsch et al. Volatile anesthetics have been shown to prolong isovolumic relaxation and blunt early rapid ventricular filling, actions that occur in conjunction with direct negative inotropic effects. Despite these intrinsic negative lusitropic effects, isoflurane and desflurane abolished ischemia-induced increases in tau and partially attenuated increases in K sub p and K^sin response to coronary artery occlusion in both ischemic and nonischemic myocardium, suggesting that these agents may exert modest cardioprotective effects during diastole under these conditions. Decreases in indices of LCCA regional chamber stiffness during anesthesia probably resulted from concomitant declines in chamber stiffness within the ischemic zone during LAD occlusion. It is unlikely that differences in normal zone loading conditions between the conscious and anesthetized states contributed to the diminished values of K^pand K^sin the LCCA region, because similar increases in left ventricular end-diastolic pressure and end-diastolic segment length were observed in response to the LAD occlusion. The magnitude of changes in dL/dt^maxin the ischemic and normal zones was similar in conscious and anesthetized dogs, indicating that desflurane and isoflurane do not alter the responses of these regions to ischemia-induced changes in early ventricular filling rate despite producing differential actions on regional compliance. Thus, increases in dL/dt^maxin the LCCA zone during brief LAD occlusion probably resulted from the combined effects of enhanced preload and augmented contractility in normal myocardium.

The mechanisms responsible for volatile-anesthetic-induced cardioprotection during myocardial ischemia and reperfusion are incompletely understood. Because potent inhalational agents cause direct negative chronotropic, inotropic, and lusitropic effects and decrease left ventricular afterload to varying degrees, the beneficial effects of these anesthetics may be attributed to a favorable reduction in myocardial oxygen demand required for active contraction with concomitant relative preservation of energy-dependent vital cellular processes via decreases in basal metabolic rate. In the current investigation, significant increases in heart rate and calculated indices of myocardial oxygen consumption (rate-pressure product and pressure-work index) occurred during brief LAD occlusion in conscious dogs. In contrast, these variables remained unchanged in response to regional ischemia in desflurane- and isoflurane-anesthetized dogs. The findings suggest that desflurane and isoflurane may provide a degree of protection against ischemic injury by blunting increases in heart rate and myocardial oxygen consumption that occur with acute regional myocardial ischemia in the conscious state. However, rate-pressure product and pressure-work index before LAD occlusion were greater than conscious values in dogs anesthetized with 1.1 MAC desflurane, but not isoflurane, indicating that increases in baseline myocardial oxygen consumption had occurred with the administration of this anesthetic. Thus, preferential alterations in myocardial oxygen supply-demand relationships are probably not solely responsible for the antiischemic actions of volatile anesthetics. Recent evidence indicates that halothane also exerts protective effects during complete functional arrest induced by cardioplegia, protective effects that were attributed to a reduction in excessive intracellular Calcium²+ accumulation. .

Abnormal intracellular Calcium²+ accumulation has been identified as a critical feature of myocardial ischemia and reperfusion injury. Volatile agents may significantly lower excessive intracellular Calcium²+ during ischemia via a direct decline in the net transsarcolemmal Calcium²+ transient resulting from partially inhibited voltage-dependent Calcium²+ channel activity. Attenuation of Calcium²+ influx has been shown to decrease the availability of Calcium²+ for contractile activation, depress Calcium sup 2+-induced Calcium²+ release from the sarcoplasmic reticulum, and reduce Calcium²+ storage in the sarcoplasmic reticulum. This anesthetic-induced reduction in total intracellular Calcium²+ availability and reserve in the normal heart may be partially protective in ischemic myocardium. Potent inhalational agents have been shown to reduce the deleterious effect of oxygen-derived free radicals on left ventricular isovolumic pressure development in isolated rabbit hearts, suggesting that these agents may also protect the myocardium against injury by limiting intracellular Calcium²+ accumulation. Thus, partial attenuation of regional ischemia-induced diastolic dysfunction observed in the current investigation may also result from favorable alterations in intracellular Calcium²+ homeostasis produced by desflurane and isoflurane through a variety of potential mechanisms.

In summary, the current investigation has demonstrated that transient coronary occlusion modestly prolonged isovolumic relaxation and abolished regional stroke work, increased chamber stiffness, and decreased early ventricular filling rate in the ischemic zone coincident with enhanced regional contractility, increased early filling, and decreased chamber compliance in the remote normal region in the conscious state. Regional ischemia-induced increases in myocardial contractility and early filling rates were preserved in the normal zone during desflurane and isoflurane anesthesia. In contrast to the findings in the conscious state, however, increases in tau and chamber stiffness in the ischemic and normal zones, which occurred in response to brief LAD occlusion, were attenuated in the presence of desflurane and isoflurane. The current results indicate that these volatile anesthetics exert beneficial effects on diastolic mechanical activity in chronically instrumented dogs in the presence of myocardial ischemia.

The authors thank Dave Schwabe for excellent technical assistance.