circulation 1990 bolli 723 38

723

ElResearch Advances Series

Mechanism of Myocardial "Stunning"

Roberto Bolli, MD

P ostischemic ventricular dysfunction, or myo-cardial "stunning,"' has recently become thefocus of considerable attention among

researchers and clinicians. There are two fundamen-tal reasons for the explosive growth of interest in thisphenomenon. First, it is now recognized that sponta-neous reperfusion after coronary spasm or thrombo-sis is common in patients with coronary artery dis-ease. This implies that postischemic myocardialstunning is likely to be an important component ofthe natural history of this disorder. Second, with theadvent of interventional recanalization, an increas-ingly large number of patients are subjected to cor-onary reperfusion in an effort to preserve left ven-tricular function. The occurrence of myocardialstunning in these patients could significantly delaythe benefits of reperfusion therapy.

During the past decade, the efforts of experimentalinvestigators have produced considerable progress inour understanding of the mechanism responsible forpostischemic contractile dysfunction. This is clearlythe central problem because it will not be possible toeffectively prevent myocardial stunning unless itspathogenesis is elucidated. The purpose of this arti-cle is to critically review the various mechanismsproposed for postischemic dysfunction and theirpathophysiological and clinical implications. Themajor unresolved issues and areas for future researchwill be identified, and an attempt will be made tointegrate different theories into a unifying pathoge-netic paradigm.

Definition of Myocardial StunningOne cannot overemphasize the importance of a

clear definition of myocardial stunning because thisterm is sometimes loosely applied to situations inwhich the persistence of contractile abnormalitiesin postischemic tissue is due to other causes (e.g.,myocellular death). Postischemic dysfunction, ormyocardial stunning, is the mechanical dysfunctionthat persists after reperfusion despite the absence of

From the Section of Cardiology, Department of Medicine,Baylor College of Medicine, Houston, Tex.

Supported in part by National Institutes of Health (NIH) grantsHL-43151 and HL-36277 and NIH SCOR grant HL-42267. R.B.was a recipient of the Physician-Scientist Award, American Col-lege of Chest Physicians.Address for correspondence: Roberto Bolli, MD, Section of

Cardiology, Baylor College of Medicine, 6535 Fannin, MS F-905,Houston, TX 77030.

Received February 7, 1990; revision accepted May 8, 1990.

irreversible damage. The essential point of this def-inition is that postischemic dysfunction, no matterhow severe or prolonged, is a fully reversible abnor-mality. Diagnosis of myocardial stunning should notbe made unless reasonable assurance can be pro-vided that the tissue in question is still entirelyviable.

In accordance with this definition, myocardialstunning is a relatively mild, sublethal injury that mustbe kept quite distinct from myocardial infarction. It isunknown whether these two conditions share a com-mon mechanism. Consequently, this review will dis-cuss only results obtained in experimental prepara-tions in which mechanical dysfunction is likely to bedue to stunning rather than to cell death. Themechanism of lethal ischemia-reperfusion injury hasbeen reviewed previously2-7 and is beyond the scopeof this article.

Classification of Myocardial StunningExperimentally, the concept of myocardial stun-

ning encompasses a wide variety of settings withmajor pathophysiological differences. A classificationis in order because different mechanisms have beendescribed in different experimental conditions. Theavailable observations can be grouped into the fol-lowing categories (Table 1).

Myocardial Stunning After One Completely ReversibleIschemic Episode

In the dog, a coronary occlusion lasting less than 20minutes does not result in any myocardial necrosis2,8but nevertheless produces prolonged abnormalitiesof contractile performance.8-14 This is the classicmodel of myocardial stunning, the one in which thephenomenon was originally described,9 and the onemost commonly used in experimental investigations.

Myocardial Stunning After Multiple CompletelyReversible Ischemic Episodes

Repeated brief (5 or 10 minutes) coronary occlu-sions have a cumulative effect on systolic functionand result in prolonged contractile impairmentdespite absence of irreversible damage.15-19 The larg-est decrease in contractility occurs after the firstocclusion; with subsequent occlusions, the additionaldecrements in mechanical performance become pro-gressively smaller.20 (Whether repetitive ischemiaalso has a cumulative effect on the time to completerecovery, however, remains unclear.) This model of

by guest on March 27, 2015http://circ.ahajournals.org/Downloaded from

http://circ.ahajournals.org/

724 Circulation Vol 82, No 3, September 1990

TABLE 1. Classification of Myocardial Stunning and Evidence for Various Mechanisms Proposed in Experimental Animals

Evidence for pathogenetic role of:

Sarcoplasmic Reduced Damage ofOxygen reticulum Calcium calcium collagen

Experimental setting radicals dysfunction overload sensitivity matrix

Stunning due to decreased blood flowRegional ischemiaOne completely reversible ischemic

episode + + ? - -Multiple completely reversible

ischemic episodes + + ? - +One partly irreversible ischemic

episode (subendocardial infarction) ? ? ?Global ischemia

Isolated heart in vitro + ? + +

Cardioplegic arrest in vivo + ? ? ? ?Stunning due to increased 02 demands

Exercise-induced ischemia - ?

+, Published studies support this mechanism; + +, published studies from multiple laboratories consistently support this mechanism, andevidence is also available in conscious animal preparations; -, published studies do not support this mechanism; ±, published studies areconflicting; ?, no data are available.

myocardial stunning differs from the single 10- or15-minute occlusion model in that the mechanicaldysfunction develops gradually and is associated witha considerably greater total ischemic burden (50-60versus 10-15 minutes).

Myocardial Stunning After One Partly IrreversibleIschemic Episode (Subendocardial Infarction)

In the dog, when reperfusion is instituted after aperiod of coronary occlusion of more than 20 minutesbut less than 3 hours, the subendocardial portion ofthe region at risk is generally found to be infarcted,whereas variable quantities of subepicardial tissueremain viable.2 The recovery of function in thissubepicardial region salvaged by reperfusion is, how-ever, delayed for days or weeks.8,21-25 It is particularlydifficult to evaluate the effect of therapy on this formof postischemic dysfunction because the reperfusedregion contains a complex admixture of necroticsubendocardium and stunned subepicardium, andthe relative proportions of these two components arehighly variable. Other confounding factors includethe tethering of surviving myocytes by dead, nonfunc-tional tissue; the expansion of the infarcted region;and the progressive replacement of necrotic myocar-dium by scar.

Myocardial Stunning After Global Ischemia inIsolated HeartsThe occurrence of cell death in these preparations

depends on species, temperature, duration of isch-emia, and perfusate composition. Under selectedconditions, isolated hearts reperfused after transientischemia exhibit complete normalization of phos-phocreatine content and intracellular pH,26-34 indi-cating that viability is generally preserved. Althoughin these models the reversibility of the contractileabnormalities cannot be verified, the metabolic

recovery suggests that the injury is mostly nonlethal.Accordingly, despite the numerous obvious differ-ences from ischemia in vivo, under selected circum-stances isolated heart preparations can mimic myo-cardial stunning. In other cases, however, thesepreparations may be associated with significant celldeath,35-44 and their relevance to myocardial stun-ning is questionable.

Myocardial Stunning After Global Ischemia DuringCardioplegic Arrest In Vivo

Despite the use of hypothermic cardioplegia, globalischemia in intact animals is usually followed by pro-longed contractile abnormalities.45-48 The reversibilityof these derangements has not been documented, butunder carefully controlled conditions they are likely tobe due mostly to stunning.

Myocardial Stunning After Exercise-Induced IschemiaStresses such as exercise provoke myocardial isch-

emia and dysfunction in animals with a flow-limitingcoronary stenosis. With cessation of exercise, thesecontractile abnormalities persist.49 Thus, myocardialstunning can also occur after high-flow ischemia, inwhich the primary problem is an increase in oxygenrequirements rather than a decrease in supply.Because of the many significant pathophysiological

differences among these situations, one cannotassume that observations made in one setting neces-sarily apply to the others. An important, unresolvedissue is whether all forms of stunning share a com-mon pathogenesis.

Factors That Determine the Severity ofMyocardial Stunning

In conscious dogs undergoing a 15-minute coro-nary occlusion, there is a sensitive coupling betweenthe degree of myocardial dysfunction after reperfu-



Bolli Mechanism of Stunned Myocardium 725

TABLE 2. Mechanisms Proposed for Myocardial Stunning

Most likely mechanismsGeneration of oxygen-derived free radicalsExcitation-contraction uncoupling due to sarcoplasmicreticulum dysfunction

Calcium overloadOther proposed mechanisms

Insufficient energy production by mitochondriaImpaired energy use by myofibrilsImpairment of sympathetic neural responsivenessImpairment of myocardial perfusionDamage of the extracellular collagen matrixDecreased sensitivity of myofilaments to calcium

sion and the magnitude of blood flow reductionduring the preceding period of ischemia, wherebyeven small differences in ischemic perfusion areassociated with large differences in postischemicrecovery.12 (In open-chest dogs, this correlation isweaker,50 probably because of the confoundingeffects of anesthesia and trauma.) Furthermore, theseverity of stunning is greater in the inner layers ofthe left ventricular wall, which are the most severelyischemic, than in the outer layers.'4'51 Another im-portant factor is the duration of flow deprivation:The longer the ischemic period, the greater theensuing mechanical abnormalities.9'52

In conclusion, the severity of postischemic dysfunc-tion is determined primarily by the severity andduration of the antecedent ischemia. This concepthas two important implications. First, whatever maybe the precise mechanism responsible for stunning,such mechanism must be initiated and modulated byperturbations associated with ischemia. Althoughstunning appears to be a form of "reperfusion injury"(see below), it is ischemia that "primes" the myocar-dium for the development of such injury. Second, anyintervention that improves perfusion during ischemiawould be expected to attenuate stunning after reflow.Reducing the severity of ischemia is probably the mosteffective way to reduce postischemic dysfunction.

Mechanism of Myocardial StunningThus, in very general terms, the abnormalities of the

postischemic myocardium are governed by the abnor-malities occurring during ischemia. But what is thespecific sequence of events whereby transient isch-emia leads to prolonged depression of contractility?

Several different hypotheses, which are not neces-sarily exclusive, have been proposed (Table 2) andare reviewed below.

Insufficient Energy Production by MitochondriaATP levels in the stunned myocardium are

depressed and recover slowly,8,2453-55 with a timecourse similar to that of contractile function.8'24Thus, in the early 1980s, the hypothesis wasadvanced",54'55 that postischemic dysfunction mayresult from an inability of the myocardium to resyn-

thesize enough high-energy phosphates to sustaincontractile function, possibly because of loss of ade-nine nucleotide precursors. Numerous subsequentobservations, however, have refuted this theory.First, no correlation is observed between myocardialATP levels and recovery of contractility in severalmodels of postischemic dysfunction.26'27,56-59 Second,the content of phosphocreatine in the stunned myo-cardium is normal or supranormal ("phosphocre-atine overshoot"),8'26'27'3053'55'57'60 implying that thephosphorylation ability of the mitochondria is intact.Third, the stunned myocardium responds to inotro-pic stimuli with a marked and sustained increase incontractility, such that function exceeds the levelsmeasured at baseline.17,30,61-66 This striking inotropicresponse occurs without a decrease in ATP or phos-phocreatine stores,30,63 indicating that energy pro-duction is not inherently impaired and can keep pacewith high metabolic demands. Fourth, manipulationsthat increase ATP levels in the stunned myocardiumfail to increase mechanical function.33,67

In summary, the available evidence suggests that adefect in the rate of resynthesis of ATP by themitochondria is not the primary cause of myocardialstunning.

Impaired Energy Use by MyofibrilsMyocardial stunning is associated with a decrease

in myofibrillar creatine kinase activity68 and a reduc-tion of the substrate (free ADP) used by the myo-fibrillar creatine kinase to produce ATP at the con-traction site.30,68 Both of these abnormalities mightdisrupt normal energy use and thus have been pos-tulated to contribute to postischemic dysfunction.68This hypothesis, however, appears implausiblebecause it does not explain the considerable con-tractile and metabolic reserve of the stunned myo-cardium.17'30'61-66 Indeed, the reversal of postisch-emic dysfunction by inotropic stimuli implies that theresidual activity of myofibrillar creatine kinase is stillsufficient to run the myofibrillar ATPase reaction atnormal or supranormal rates.

Impairment of Sympathetic Neural ResponsivenessAfter 25 minutes of coronary occlusion, the reper-

fused myocardium loses its inotropic responsivenessto sympathetic nervous stimulation, suggesting injuryto the sympathetic-neural axis.69 However, no suchdefect is observed after 15 minutes of coronaryocclusion.70 Furthermore, as noted above, myocar-dial stunning occurs in isolated (and thereby dener-vated) hearts. Although functional sympathetic de-nervation may contribute to postischemic dysfunctionin certain situations in which contractility is stronglydependent on sympathetic drive, it seems unlikelythat this abnormality plays a major role in mostexperimental preparations of myocardial stunning.

Impairment of Myocardial PerfusionWhen myocardial stunning is produced by a 15-

minute coronary occlusion, postreperfusion subepi-




cardial blood flow is normal, whereas subendocardialflow is slightly decreased (-20%)."-14,71-73 Themechanism and significance of this phenomenon areunclear.72,73 Although it is possible, as speculated byHeyndrickx et al,1' that the reduced subendocardialperfusion may contribute to the impaired mechanicalperformance, two considerations suggest that it is nota major causative factor. First, there is no correlationbetween subendocardial blood flow and wall thicken-ing in the postischemic myocardium.72,73 Second, theloss of systolic function (dyskinesis or akinesis) is outof proportion to the slight (-20%) decrease insubendocardial perfusion.72'73When myocardial stunning is produced by

repeated 5-minute coronary occlusions, blood flow tothe postischemic myocardium, measured by radioac-tive microspheres, is generally normal.15-17,19 How-ever, microspectrophotometry reveals increased vein-to-vein heterogeneity of oxygen saturations, with anexcess of vessels demonstrating very low oxygensaturations, suggesting that the stunned myocardiumcontains discrete microregions of either increasedoxygen extraction or reduced flow.19 A nonuniformreduction of flow (microregions of ischemia inter-mixed with microregions of hyperemia) has also beeninvoked to explain the enhancement of function thatis observed in stunned myocardium when perfusion isaugmented.16 However, the findings that myocardiumstunned by repeated 5-minute coronary occlusionshas normal flow reserve'6 and intact microvascularpatency'9 exclude nonuniform ischemia secondary toanatomical obstruction. Furthermore, the hypothesisof persistent ischemia as a cause of dysfunction doesnot explain the normal or near-normal contractilereserve of the postischemic myocardium.17'65,66

In summary, impaired perfusion (either uniform ornonuniform) is unlikely to be the primary pathoge-netic mechanism in postischemic contractile failure.

Damage of Extracellular Collagen MatrixUsing a model of myocardial stunning produced by

12 5-minute coronary occlusions separated by 10minutes of reperfusion, Eng's group has demon-strated that postischemic dysfunction is associatedwith severe ultrastructural damage of the extracellu-lar collagen matrix74 and a modest (-10%) loss ofcollagen.75 These findings have led to the hypothesisthat myocardial stunning may be due to a structuraldefect, namely, disruption of the mechanical cou-pling function provided by the extracellular collagennetwork.74 To fully evaluate this hypothesis, it will beimportant to explore the time course of the collagenchanges and its relation to the time course of post-ischemic dysfunction. Because in conscious dogsundergoing one brief (c15 minutes) coronary occlu-sion most of the contractile recovery takes place inthe first 3-4 hours of reperfusion9'12,13,52 (an intervaltoo short for significant resynthesis or remodeling ofthe collagenous components), it seems improbablethat structural damage to collagen plays a pathoge-netic role in this model of myocardial stunning.

Indeed, a recent study76 demonstrated that collagenis disrupted by 12 5-minute coronary occlusions butnot by one 15-minute occlusion followed by reperfu-sion.

In summary, collagen damage may contribute tomyocardial stunning after multiple ischemic episodesbut is unlikely to be a causative factor after a singlecompletely reversible ischemic episode (Table 1).One possible explanation for this difference is that inthe former setting, repetitive ischemia results inrepetitive systolic bulging for a total duration of morethan 15 minutes; this may cause greater mechanicalstress and, consequently, greater damage of theextracellular collagen network.

Decreased Sensitivity of Myofilaments to CalciumUsing isolated ferret hearts subjected to 15 min-

utes of global ischemia at 370 C, Kusuoka et a127observed that the stunned myocardium exhibitsdecreased responsiveness to calcium, as manifestedby a decrease in the maximal calcium-activated forceand a decrease in the myocardial sensitivity to extra-cellular calcium. They speculated that the reducedsensitivity to extracellular calcium could in turn bedue to either a decrease in the intracellular free Ca21concentration ([Ca]i) transient or a decrease in thesensitivity of myofilaments to calcium.27 The formerpossibility seems improbable because two recentstudies with different techniques32'77 indicate that thecalcium transient is (paradoxically) increased in thestunned myocardium after 1532 or 20 minutes77 ofglobal ischemia at 370 C in isolated ferret hearts.Accordingly, it has been proposed32'77 that the fun-damental mechanism for postischemic dysfunction isnot insufficient availability of free cytosolic calciumduring systole but instead is reduced sensitivity of thecontractile apparatus to calcium. One problem withthis hypothesis derived from in vitro studies is that itdoes not explain two observations made in vivo. First,when the stunned myocardium is challenged withinotropic stimuli, it exhibits a normal or near-normalcontractile reserve.1765,66 Second, the apparent sen-sitivity of the stunned myocardium to intracoronarycalcium is not decreased.65 If the primary problemwas a reduced sensitivity of myofilaments to calcium,then the contractile response to exogenous calciumor to other inotropic agents (which act by raising[Ca]i) should be reduced.

In summary, studies in vitro suggest that myocar-dial stunning is a result of reduced calcium sensitivityrather than reduced calcium availability, but studiesin vivo are not readily reconcilable with this interpre-tation (Table 1).

Calcium OverloadSeveral lines of evidence point to a pathogenetic

role of calcium overload in myocardial stunning.First, when isolated ferret hearts subjected to 15minutes of global normothermic ischemia are reper-fused with solutions containing low concentrations ofcalcium, the postischemic contractile abnormalities




are significantly attenuated.27 The results of thisexperiment, in which no intervention is applied dur-ing ischemia, indicate that calcium entry upon reper-fusion is an important mechanism of myocardialstunning. A decrease in the severity of stunning isalso observed in hearts pretreated with ryanodine, aninhibitor of cellular calcium overload.29 Second,exposure of isolated ferret hearts to a transientcalcium overload in the absence of ischemia producesmechanical and metabolic abnormalities similar tomyocardial stunning.78 Third, transient intracellularacidosis during early reperfusion can prevent myo-cardial stunning in isolated ferret hearts subjected to15 minutes of global ischemia31 and open-chest dogsundergoing 15 minutes of coronary occlusion.79Because acidosis antagonizes not only the influx ofcalcium into cells (by inhibiting the Na4-Ca21exchange and the slow calcium channels) but also theintracellular binding of calcium,31 these results pro-vide indirect evidence that a transient calcium over-load contributes to postischemic dysfunction. Fourth,recent studies28'80'81 have shown that [Ca]i increasesbetween 10 and 20 minutes of global ischemia inisolated hearts. This is the same duration of ischemiathat results in myocardial stunning. In these models,[Ca]i returns to normal values within a few minutesafter reperfusion,28,7780 but appears to remain tran-siently elevated during very early reflow.77'82The mechanism for this rise in [Ca]i during isch-

emia remains uncertain. It may be secondary todecreased calcium uptake by the sarcoplasmicreticulum.83 It may also be mediated through theNa+-Ca2' exchange, secondary to a rise in intracel-lular Na+ during ischemia due to 1) metabolic inhi-bition of the Na+,K+-ATPase, and 2) acidosis andconsequent Na+-H+ exchange.84-86 Because the Na+-Ca2' exchange is inhibited by acidosis and intracel-lular pH in the stunned myocardium normalizesquickly after reflow 27,30'31'33'60 entry of calcium viathe Na4-Ca2' exchange may be greatly increasedupon reperfusion.84,86-88 Further research will benecessary to identify the precise circumstances lead-ing to calcium overload. The mechanisms by which atransient calcium overload induces prolonged con-tractile dysfunction are also unclear. Increased cyto-solic calcium can activate phospholipases and otherdegradative enzymes.6,34 In addition, elevation ofcytosolic calcium could trigger production of oxygenradicals via xanthine oxidase89 (see below).The calcium overload theory is not incompatible

with the fact that exogenous calcium amelioratesfunction in the stunned myocardium.27,65 The in-crease in [Ca]i is postulated to be a brief phenome-non immediately after reflow, following which theremay be either a normalization of [Ca]i transientsor a relative calcium deficiency (see "Excitation-Contraction Uncoupling Due to Sarcoplasmic Retic-ulum Dysfunction"). The early excessive [Ca]i levelscould damage intracellular organelles concernedwith contraction and thereby produce prolongedmechanical dysfunction.

In summary, considerable evidence suggests that atransient calcium overload during early reperfusioncontributes to the pathogenesis of myocardial stun-ning after global ischemia in vitro. The role ofcalcium overload in in vivo models of myocardialstunning remains to be defined (Table 1) and repre-sents an important area for future investigations.

Effect of calcium channel blockers. Calcium chan-nel blockers have been used to gain insights into therole of altered calcium homeostasis in vivo, butunfortunately the results are difficult to interpret.Verapamil,59 diltiazem,90 nifedipine,91 nitren-dipine,92 and amlodipine93 have been shown toimprove recovery of function in regionally stunnedmyocardium in intact animals. However, it isunclear whether these beneficial effects reflected adirect protective action of the drugs or were medi-ated by favorable modifications of afterload, pre-load, heart rate, and regional myocardial bloodflow, all of which could change the systolic proper-ties of the stunned myocardium.16,59 A direct pro-tective action was demonstrated by a recent study94in which nifedipine enhanced the functional recov-ery of the stunned myocardium independently ofany effect on systemic hemodynamics or regionalmyocardial blood flow. However, interpretation ofthese results is problematic because the beneficialeffects of nifedipine were noted even when treat-ment was started 30 minutes after reperfusion,94whereas myocardial stunning has been suggested toresult from a transient calcium overload immedi-ately after reflow.27 It is unlikely that nifedipineacted in a specific way to decrease [Ca]i becauseadministration of calcium or inotropic agents(which raise [Ca]i) improves contractile function inthe stunned myocardium.17,30,61-66 The mechanismof action of nifedipine in this intriguing study94remains to be elucidated.

Excitation-Contraction Uncoupling Due toSarcoplasmic Reticulum Dysfunction

Krause et a195 investigated this mechanism in acanine model of postischemic dysfunction producedby eight to 12 5-minute occlusions separated by10-minute reflow periods. Sarcoplasmic reticulumisolated from the stunned myocardium demonstrateda decrease in the ability to transport calcium, con-comitant with a reduction in the activity of theCa2+,Mg2+-ATPase.95 These researchers proposedthat a decrease in the amount of calcium stored inthe sarcoplasmic reticulum as a result of a reductionin the calcium pump activity could diminish contrac-tile protein activation through attenuated calciumrelease during systole.95The concept of inadequate delivery of calcium to

the contractile proteins, secondary to sarcoplasmicreticulum dysfunction, is an attractive hypothesis. Thepostulated relative calcium deficiency would be con-sistent with the observation that exogenous adminis-tration of calcium can return contractile function ofstunned myocardium to preischemic levels.65 It would




also be consistent with the notion that inotropic agents(which increase [Ca]i) can reverse myocardialstunning.17'30'61-66 However, since this hypothesisimplies that the amplitude of the [Ca]i transient isdecreased, it is not easily reconcilable with in vitrodata32'77 (see above). Considerably more research isnecessary to determine whether similar dysfunction ofthe sarcoplasmic reticulum occurs in other models ofmyocardial stunning (Table 1) and whether it is asso-ciated with decreased [Ca]i transients in vivo. Inaddition, the factors that cause injury to the sarcoplas-mic reticulum remain to be elucidated.

Generation of Oxygen-Derived Free RadicalsEffect of antioxidants on myocardial stunning after a

brief coronary occlusion. Approximately 7 years ago, anumber of investigators, including ourselves, postu-lated that myocardial stunning is caused in part bythe generation of reactive oxygen metabolites, suchas superoxide anion (02 ), hydrogen peroxide (H202),and hydroxyl radical (.OH), and began a series ofexperiments designed to test this hypothesis. We usedan open-chest canine preparation in which the leftanterior descending coronary artery is occluded for 15minutes and then reperfused.96 In the first study,97 wefound that administration of superoxide dismutase(SOD) (an enzyme that catalyzes the dismutation of*02 to 02 and H202) and catalase (an enzyme thatreduces H202 to 02 and H20) significantly enhancedrecovery of function after reperfusion. Attenuation ofmyocardial stunning by SOD and catalase was alsoobserved by other investigators50'57'98 using similarexperimental preparations. In a more recent study99we found that neither SOD nor catalase alone signifi-cantly improved recovery of function in the stunnedmyocardium; however, when the two enzymes werecombined, contractile recovery was significantly greaterthan that observed in controls or in dogs receivingeither enzyme alone. These results suggest that both*02- and H202 contribute to the cellular damageresponsible for myocardial stunning and that combinedadministration of SOD and catalase is more likely to beeffective against stunning than separate administration.The inability of SOD alone to mitigate postischemicdysfunction has also been observed in a pig model.100Although both *°2- and H202 appear to contrib-

ute to stunning, it remains uncertain whether theydo so by direct cytotoxicity or via formation of otherspecies. *°2 and H202 can interact through themetal-catalyzed Haber-Weiss reaction to generatethe highly reactive *OH radical.10' Consequently,accumulation of *°2 and H202 could producepostischemic dysfunction, at least in part, indirectlythrough generation of OH. To elucidate this prob-lem, we evaluated dimethylthiourea, an -OH scav-enger that is more effective than traditional *OHscavengers102 and does not react with *°2 or H202in vitro.'03 We observed that dimethylthiourea pro-duced a significant and sustained improvement in

the function of the stunned myocardium.103 Theseresults were further corroborated by studies withN-2-mercaptopropionyl glycine (MPG), a free rad-ical scavenger that readily enters the intracellularspace and is active orally. MPG was found to be apowerful scavenger of OH with no effect on *2 orH202 in vitro104 and to effectively attenuate myocar-dial stunning in vivo.104105 Taken together, theseresults suggest that the OH radical (or one of itsreactive products) is a mediator of postischemicdysfunction and that the beneficial effects of SODand catalase previously demonstrated5057'97-99 aredue in part to prevention of OH generation. Thisconclusion is also consistent with the finding thatthe iron chelator desferrioxamine attenuates post-ischemic dysfunction.106"07 Because iron catalyzesthe formation of OH (through the Haber-Weiss orFenton mechanism) as well as the propagation of.OH-initiated lipid peroxidation,10' the protectiveeffects of desferrioxamine are compatible with theview that OH is a mediator of myocardial stunning.The effectiveness of antioxidants that scavenge

.OH or prevent its generation should not be con-strued as evidence that all of the damage initiated byoxygen is mediated via OH. It is true that H202 is arelatively nonreactive species and its toxicity may bedue in large part to its reduction to OH.101 However,there is evidence that *°2 can directly cause cellulartoxicity in vitro (Reference 108; reviewed in Refer-ence 109). Furthermore, because catalase alone pre-vents formation of OH by removing H202, the factthat SOD has to be added to catalase to producesignificant attenuation of stunning in vivo99 impliesthe existence of a component of damage specificallydue to *02- It appears, therefore, that all threespecies (°2- , H202, and *OH) are important inpostischemic dysfunction (O2- and *OH as mediatorsof injury and H202 as a precursor of OH).99Although the studies discussed above50,57,97-99,103-107

consistently support the oxyradical hypothesis, theirsignificance is limited by the fact that they were allperformed in open-chest animals. Thus, artifacts dueto the combined effects of anesthesia, hypothermia,surgical trauma, volume and ionic imbalances, non-physiological conditions, and associated neuro-humoral perturbations, as well as other potentiallyconfounding variables, cannot be excluded. It istherefore essential that the oxyradical hypothesis betested in conscious animal preparations. In a recentexperiment,"10 we observed that the combinedadministration of SOD and catalase significantlyenhanced recovery of function after a 15-minutecoronary occlusion in conscious, unsedated dogs andthat this effect was sustained for 24 hours afterreflow. The accelerated recovery was not followed byany subsequent deterioration, indicating that post-ischemic depression of contractility is not a useful"protective" response to injury.

In summary, numerous investigations from severalindependent laboratories5057,97- 99,103-107, 10 uniformlysuggest that oxygen metabolites play a significant role




in the genesis of myocardial stunning after a 15-minute period of ischemia, in both open-chest andconscious animals. The relative contributions of thevarious metabolites have not been definitively estab-lished. It is clear that a portion of the damage ismediated by OH; however, the available evidencesuggests that myocardial stunning cannot be simplis-tically ascribed to a single oxygen species and that allof the three initial metabolites of oxygen (02 , H202and *OH) contribute to the cellular injury responsiblefor postischemic dysfunction. Iron also appears toplay a role in stunning, presumably by catalyzing theformation of OH.

Direct evidence for the oxyradical hypothesis. Amajor limitation of the studies reviewed hereto-fore50,57,97-99,103-107,110 is that all of the evidence pro-vided by these investigations is indirect and thereforeinconclusive. Clearly, to definitively confirm a caus-ative role of oxygen metabolites in postischemicdysfunction, it is necessary to directly demonstrateand quantitate free radical generation in the stunnedmyocardium in the presence and absence of antioxi-dant interventions.

Production of free radicals has been directlydemonstrated in isolated rabbit or rat hearts under-going global ischemia and reperfusion,42,4311-113but because of the numerous important differencesin experimental conditions, results obtained in vitrocannot necessarily be extrapolated to the intactanimal. We used the spin trap a-phenyl N-tert-butylnitrone (PBN) and electron paramagnetic reso-nance (EPR) spectroscopy to detect and measureproduction of free radicals in our in vivo model ofpostischemic dysfunction (15-minute coronaryocclusion in open-chest dogs).14 After infusion ofPBN, EPR signals characteristic of PBN radicaladducts were detected in the venous blood drainingfrom the ischemic-reperfused vascular bed.15 TheEPR signals were consistent with a mixture ofdifferent secondary lipid radicals, such as alkyl andalkoxy radicals. The myocardial production of rad-icals began during coronary occlusion but increaseddramatically after reperfusion, peaking at 2-4 min-utes. After this initial burst, production of radicalsabated but did not cease, persisting for as long as 3hours after reflow.1" There was a linear, positiverelation between the magnitude of adduct produc-tion and the magnitude of ischemic flow reduction,indicating that the intensity of free radical genera-tion after reflow is determined by the severity of theantecedent ischemia"15-the greater the degree ofhypoperfusion, the greater the subsequent produc-tion of free radicals and, by inference, the severityof reperfusion injury. These findings imply thatinterventions that improve perfusion during isch-emia will attenuate free radical reactions afterreflow. Generation of free radicals in in vivo modelsof regional myocardial stunning has also beenreported by Leiboff et al.1"6

In a subsequent study,17 we found that SOD pluscatalase suppressed the production of free radicals inthe stunned myocardium, indicating that these radi-

cals are derived from univalent reduction of oxygen.Furthermore, the inhibition of free radical produc-tion was associated with inhibition of myocardialstunning.17 More recently, we observed that MPG ordesferrioxamine administered just before reperfusionmarkedly attenuated myocardial stunning and theassociated production of PBN adducts04,114; how-ever, the same agents given 1 minute after reperfu-sion did not attenuate myocardial stunning or initialPBN adduct production. Thus, three different anti-oxidant interventions (SOD plus catalase, MPG, anddesferrioxamine) reduced postischemic dysfunctionat the same doses and under the same experimentalconditions in which they reduced formation of PBNadducts. The correlation between the two effectssuggests that the production of free radicals in thestunned myocardium plays a causal role in thedepression of contractility.

In summary, the results of the studies that havemeasured free radicals in the stunned myocar-dium104114-117 provide direct evidence supporting apathogenetic role of oxygen metabolites. Specifically,these studies indicate that 1) free radicals are pro-duced in the stunned myocardium in the intact dogafter reversible regional ischemia, 2) the univalentpathway of reduction of oxygen is the source of theradicals, and 3) inhibition of free radical reactionsresults in enhanced recovery of contractility (i.e., theradical reactions are necessary for postischemic dys-function to occur).

Effect of oxygen radicals on cardiac function. Theability of oxygen metabolites to depress myocardialfunction has been directly demonstrated in vitro andin vivo. Exposure of isolated rabbit interventricularsepta,18 isolated rat"19 or rabbit'08 papillary muscles,and isolated rat120-122 or rabbit'23"124 hearts to freeradical-generating solutions or pure H202 has uni-formly resulted in decreased mechanical functionand ATP levels (i.e., in changes similar to thoseobserved in the stunned myocardium). In most of thestudies in which an attempt was made to discern therelative roles of different oxygen species,118,119,122'123it was found that the deleterious effects could beprevented by catalase or by *OH scavengers but notby SOD, suggesting that H202 or its byproduct, *OH,was the oxygen metabolite responsible for theobserved depression of contractile function. In onestudy,108 however, SOD was protective, whereas cat-alase had no effect, implying that the major negativeinotropic agent was *02 . These in vitro observationshave been recently expanded by Przyklenk et al,'25who reported that infusion of xanthine oxidase pluspurine plus iron-loaded transferrin, administeredthrough a coronary vein in open-chest dogs, resultedin the development of significant wall motionabnormalities.

In summary, it is clear that reactive oxygen speciesdepress myocardial contractility both in vitro and invivo. The precise role of each species remains to bedefined. There is considerable evidence that thedetrimental effects of *°2- and H202 on cardiac




function are mediated in part by generation of *OH.This is in accordance with the results of in vivostudies,103-'07 which suggest an important role of *OHas a mediator of stunning. However, there is alsoevidence for a direct negative inotropic action of*02 , which is also congruent with observations invivo.99Mechanism of oxyradical-mediated contractile dys-

function. The precise mechanism whereby oxygenmetabolites depress contractile function remainsspeculative and represents one of the major unre-solved issues pertaining to the pathogenesis of myo-cardial stunning. Free radicals are reactive speciesthat have no specific target and can attack virtually allcellular components. In theory, every abnormalitythus far described in the stunned myocardium (seeabove) could be caused by oxyradicals. At least twokey cellular components, proteins and lipids, arelikely to be involved in free radical-initiated reac-tions in the postischemic myocardium. Activatedoxygen species can denaturate proteins and inacti-vate enzymes.126 Furthermore, they can produce per-oxidation of the polyunsaturated fatty acids con-tained in cellular membranes,127 which would impairselective membrane permeability and interfere withthe function of various cellular organelles.

Evidence for the occurrence of lipid peroxidationin the stunned myocardium is provided by two recentstudies. Romaschin et al128 observed increased myo-cardial concentration of hydroxy conjugated dienes(which are products of free fatty acid oxidation)during and after a 45-minute period of global nor-mothermic ischemia in open-chest dogs. The maxi-mal concentration of conjugated dienes was mea-sured at 5 minutes of reflow. Importantly, the tissueexamined after reperfusion was dysfunctional but notnecrotic, thus representing stunned myocardium.Weisel et al129 subsequently reported that in patientsundergoing cardioplegic arrest during coronaryartery bypass surgery, there was myocardial releaseof conjugated dienes in the coronary sinus blood at 3and 60 minutes of reperfusion, which was associatedwith a decrease in the myocardial concentration ofthe antioxidant a-tocopherol.

Dysfunction of the sarcoplasmic reticulum may bean important mechanism whereby oxyradicals medi-ate the contractile abnormalities observed afterreversible ischemia. As mentioned above, both cal-cium uptake and Ca2+,Mg2+-ATPase activity havebeen found to be significantly depressed in thestunned myocardium.95 A similar decrease in calciumuptake and Ca2',Mg2+-ATPase activity is observed inisolated sarcoplasmic reticulum after exposure tooxygen radicals,127,130 and oxyradical scavengers pre-serve sarcoplasmic reticulum function.'27,130Other studies suggest that the sarcolemma may

be a critical target of free radical-mediated dam-age. Oxyradicals have been shown to interfere withcalcium transport and calcium-stimulated ATPaseactivity in the sarcolemma.'31132 Oxygen radicalshave also been shown to interfere with the Na+-

Ca2+ exchange133 and to inhibit the Na+,K+-ATPaseactivity.134 Furthermore, the sarcolemmal Na+,K+-ATPase activity is impaired in reperfused myocar-dium, and this impairment is prevented byantioxidants.135 Impairment of the Na+,K+-ATPaseactivity would result in Na+ overload, with conse-quent activation of the Na+-Ca2+ exchange activ-ity.87'88 All of these observations imply that exces-sive production of oxyradicals could result inincreased transarcolemmal calcium influx and cel-lular calcium overload.

It is important to point out that the foregoingpostulated mechanisms involve alterations in calciumhomeostasis and thus would reconcile the oxyradicaland calcium-overload hypotheses of stunning intoone pathogenetic mechanism.

Sources of oxygen radicals in the stunned myocar-dium. Among the various potential mechanisms ofoxyradical generation in the stunned myocardium,only two have been explored thus far, namely, theenzyme xanthine oxidase and the activated neutro-phil. We found that the xanthine oxidase inhibitor,allopurinol, produced a marked improvement in thefunctional recovery of the stunned myocardium.136Furthermore, allopurinol inhibited the increased car-diac production of urate observed in control dogsduring ischemia and early reperfusion, indicatingeffective inhibition of xanthine oxidoreductase.136Attenuation of stunning has also been demonstratedwith oxypurinol after 15137 and 90 minutes138 ofischemia and reperfusion. These data suggest thatxanthine oxidase is one of the sources of the oxygenradicals that contribute to postischemic dysfunctionin the dog. Whether this concept applies to humansremains uncertain because the myocardial content ofxanthine oxidase is species dependent. Reportsregarding the myocardial content of the enzyme inthe human heart are conflicting.139-143

In contrast to these earlier reports,136-138 in arecent study,'44 oxypurinol and amflutizole, two xan-thine oxidase inhibitors, failed to mitigate stunningafter 15 minutes of ischemia in dogs. Whether thediscrepancy is due to different degrees of inhibitionof xanthine oxidoreductase or to other reasons isunclear.

Leukocytes are another potential source of oxygenmetabolites.145,146 However, several investigatorshave found that myocardial stunning is not attenu-ated by depletion of neutrophils with antiserum71'147or filtration,148 by administration of nafazatrom,which inhibits production of leukotrienes,149 byadministration of antibodies to the adhesion-promoting Mol glycoprotein,150 and by administra-tion of dextran, which inhibits leukocyte adherenceto the endothelium.151 Recent studies have shownthat the neutrophil content of the stunned myocar-dium after a 12-minute period of ischemia isdecreased152 and that myeloperoxidase activity (amarker of neutrophils) in myocardium stunned by a15-minute occlusion is not different from nonisch-emic myocardium.150 Furthermore, myocardial stun-




ning occurs in isolated heart preparations that aredevoid of circulating neutrophils. Taken together,these results71'147-'52 suggest that granulocytes do notplay a major role in the genesis of postischemicdysfunction after brief, reversible ischemia, althoughthey appear to be major mediators of irreversibletissue injury and vascular obstruction after prolongedperiods of ischemia.145"146153In contrast to the aforementioned findings, however,others have reported that depletion of granulocytesby Leukopak filtration prevents154 or alleviates155postischemic dysfunction after a 15-minute coronaryocclusion. The reasons for the differences betweenthese studies'54"155 and the aforementionedinvestigations71"147-151 are unclear and warrant addi-tional research.156 One possibility to be explored,raised by Engler and Covell,154 is that the relationbetween circulating neutrophil counts and myocar-dial damage may be nonlinear.In summary, in the canine model of myocardialstunning produced by one reversible ischemic epi-sode, xanthine oxidase may be a source of freeradicals (although the data are not entirely consis-tent), whereas the role of neutrophils remains con-troversial. Of course, many other processes couldlead to formation of °2- and H202 during myocar-dial ischemia and reperfusion, including activation ofthe arachidonate cascade, autoxidation of catechol-amines and other compounds, ischemia-induceddamage of the electron transport chain in the mito-chondria, and accumulation of reducing equiva-lents.127 It will be difficult to selectively assess therole of these potentially important mechanisms in acomplex system such as the intact animal. Identifica-tion of the sources of oxyradicals in the stunnedmyocardium represents another major unresolvedissue.

Time course of free radical-induced damage in thestunned myocardium. When exactly in the ischemia-reperfusion sequence that leads to stunning does thecrucial free radical-induced damage take place? In arecent study,104 we observed that infusion of theantioxidant MPG attenuated postischemic dysfunc-tion to a similar extent regardless of whether theinfusion was started before ischemia or 1 minutebefore reperfusion; however, infusion started 1minute after reflow was ineffective, suggesting thatthe critical radical-mediated injury occurs in the firstfew moments of reperfusion. We have subsequentlyobtained similar results with desferrioxamine.14 Fur-thermore, the spin trap PBN enhances contractilerecovery even when the infusion is commenced 20seconds before reflow; the magnitude of the protec-tive effect is similar to that observed when theinfusion is commenced before ischemia.1" Captopril,an angiotensin converting enzyme inhibitor withalleged *°2 scavenging properties, produces similarattenuation of postischemic dysfunction when theadministration is started before ischemia or 2 min-utes before reperfusion.157 The conclusion that asubstantial portion of the cellular damage responsi-

ble for stunning occurs immediately after reflow isfurther corroborated by direct measurements of freeradicals104"'15"17 and lipid peroxidation products128 inthe stunned myocardium, both of which have showna burst in the initial moments after restoration offlow. If free radical production is inhibited duringthis initial burst, postischemic dysfunction is miti-gated; however, if free radical production is inhibitedafter the first 5 minutes of reperfusion (i.e., after theinitial burst), no functional improvement isobserved.04 These observations suggest that the freeradicals important in causing myocardial stunningare those produced immediately after reflow.

In summary, myocardial stunning appears to be, atleast in part, a form of oxyradical-mediated "reper-fusion injury." This concept may have significanttherapeutic implications because it suggests thatantioxidant therapies begun after the onset of isch-emia could still be effective in preventing postisch-emic dysfunction; however, a delay in the implemen-tation of such therapies until after reperfusion wouldresult in loss of efficacy.

Role of oxygen radicals in other forms of myocardialstunning (Table 1). Studies reviewed thus far used onebrief (.15 minutes) coronary occlusion. In recentunpublished studies in our laboratory, we found thatthe *OH scavenger MPG markedly improves contrac-tile recovery after 10 5-minute coronary occlusionsseparated by 10-minute reflow periods in open-chestdogs. Thus, oxyradicals also appear to contribute tothe genesis of myocardial stunning after multiplebrief ischemic episodes. Furthermore, numerousstudies from the surgical literature suggest that oxy-gen radicals also contribute to postischemic dysfunc-tion after global ischemia in in vivo models of car-dioplegic arrest,45-48 and there is evidence thatoxidative stress contributes to postoperative dysfunc-tion in patients undergoing cardiac surgery.158Finally, antioxidants consistently alleviate mechani-cal dysfunction after global ischemia in isolatedhearts,35-4144 but the relevance of these in vitropreparations to myocardial stunning is uncertain, asdiscussed above.Data regarding the role of oxygen radicals in

myocardial stunning after a prolonged coronaryocclusion are conflicting. Three studies failed todetect improvement in functional recovery with SODand catalase after coronary occlusions lasting 1 hour,'5990 minutes,160 or 2 hours16' in open-chest'6' or con-scious dogs.'59 160Wel62 also observed that SOD failsto enhance recovery of contractility after a 2-hourcoronary occlusion in anesthetized dogs. Theseresults suggest that short-term administration ofantioxidant enzymes is not effective in mitigatingmyocardial stunning associated with subendocardialinfarction. It may be that the postischemic dys-function that follows a prolonged, partly irrevers-ible ischemic insult differs from the dysfunctionthat follows a short, completely reversible insult.However, other studies138"63 have shown that thecell-permeant antioxidants oxypurinol and N-




acetylcysteine attenuate myocardial stunning inde-pendently of infarct size limitation in closed-chestdogs subjected to 90 minutes of coronary occlusionand 24 hours of reflow.

Initial observations164 indicate that SOD and cat-alase do not alleviate exercise-induced stunning.

In summary, there is strong evidence that oxyrad-icals contribute to postischemic dysfunction afterglobal ischemia (in vitro as well as in vivo) and aftermultiple episodes of regional ischemia. They do notappear to contribute to exercise-induced postisch-emic dysfunction. The role of oxygen radicals inmyocardial stunning after a prolonged, partly irre-versible ischemic insult remains uncertain and repre-sents a major unresolved problem. Elucidation of thisissue will be difficult because the dysfunction is duein part to the presence of infarction and in part to thepresence of stunning- a situation that complicatesthe evaluation of therapy, as discussed above.

Integration of Different HypothesesMyocardial stunning is probably a multifactorial

process that involves complex sequences of cellularperturbations and the interaction of multiple patho-genetic mechanisms. Our understanding of this phe-nomenon is still fragmentary, and none of the theo-ries discussed herein (Table 2) explains the entirecascade of events that culminates in postischemiccontractile failure. For example, the oxyradicalhypothesis does not explain how reactive oxygenspecies cause mechanical dysfunction, and the sarco-plasmic reticulum theory does not explain why thisorganelle is damaged. Integration of the varioushypotheses is complicated by the fact that for themost part, each hypothesis has been developed in adifferent experimental preparation (Table 1).

Nevertheless, it is important to emphasize thatthese hypotheses are not mutually exclusive and infact may represent different parts of the same patho-physiological sequence. There is considerable evi-dence to suggest a link between generation of oxygenradicals and impaired calcium homeostasis in thesetting of reperfusion injury. For example, the dam-age associated with the "calcium paradox" resemblesthat associated with the "oxygen paradox" and prob-ably has a similar pathogenetic mechanism.165 Fur-thermore, as discussed above, oxyradicals generatedupon reperfusion can cause dysfunction of the sarco-plasmic reticulum127,130 and alter calcium flux acrossthe sarcolemma.131-135 The result of these actionswould be excitation-contraction uncoupling and cel-lular calcium overload.127131 In this regard, it hasrecently been demonstrated that reoxygenation ofcultured myocytes results in Ca'+ overload, which canbe greatly attenuated by antioxidant enzymes.85 Oxy-gen radicals could also damage the proteins of thecontractile machinery in a manner that reduces theirresponsiveness to calcium. On the other hand, cal-cium overload may exaggerate oxyradical productionby promoting the conversion of xanthine dehydroge-

nase to xanthine oxidase, which appears to be medi-ated by a calcium-dependent protease.89A unifying hypothesis for the pathogenesis of

myocardial stunning is proposed in Figure 1 (detaileddescription of the postulated mechanisms is providedin the figure legend). This paradigm is largely spec-ulative but nevertheless based on the evidence avail-able at this time and discussed in this review. Accord-ing to this conceptual scheme (Figure 1), oxyradicalgeneration, calcium overload, and sarcoplasmic retic-ulum dysfunction can be viewed as different aspectsof the same pathogenetic mechanism, thereby recon-ciling the three major current hypotheses of myo-cardial stunning.

In summary, the pathogenesis of myocardial stun-ning has not been definitively determined. Thus far,researchers have focused on the identification ofindividual mechanisms of injury. Three abnormalities(oxyradical generation, calcium overload, and excita-tion-contraction uncoupling) have emerged as likelycontributing factors (although in different experi-mental preparations), and in vitro evidence suggeststhat they are interrelated. It is now essential to clarifythe precise interactions among these three factors invivo and the role that each of them plays in thevarious experimental settings of myocardial stunning.

Clinical ImplicationsThere are numerous clinical settings in which the

myocardium is subjected to transient ischemia,including unstable or variant angina, acute myocar-dial infarction with early reperfusion (either sponta-neous or induced by thrombolytic therapy), open-heart surgery with cardioplegic arrest, and cardiactransplantation. Recent observations suggest thatthese situations may be associated with myocardialstunning, which may be an important factor precipi-tating left ventricular failure with its attendant mor-bidity and mortality.1,166 Furthermore, in patients inwhom silent episodes of ischemia recur at shortintervals in the same territory, the myocardium mightremain reversibly depressed for prolonged periods oftime or even chronically.1 Elucidation of the mecha-nism responsible for postischemic dysfunction isindispensable for developing effective clinical thera-pies. The evidence discussed herein implies thatmyocardial stunning could be effectively attenuatedby pharmacological manipulations targeted at spe-cific pathogenetic mechanisms, for example, byantioxidant agents given before reperfusion. Com-pared with standard treatments for cardiac failuresuch as inotropic stimulation, antioxidants representa novel approach; by interfering with the mechanismof cell injury, they might prevent rather than simplytemporarily correct postischemic dysfunction. Thisapproach also appears desirable because it is uncer-tain whether prolonged inotropic stimulation ofstunned myocardium has deleterious effects.




IN

over

REVERSIBLE ISCHEMIA / REPERFUSION

Xanthine oxidase ? , ? t Reducing equivalentseutrophil activation ? * *-? tMitochondrial "univalent leak"

Arachidonate cascade ? -- " ? Autoxidative processes_ SOD

Ja+ 2*0 H2 °2rload I FeFI1

*OH .

Enzyme inactivation KhhO.u Lipid peroxidation

Damage of:|<eofjj t ? OTHER S

/" ? SARCOPLASMIC *****K (e.g., CONTRACTII? SARCOLEMMA RETICULUM PROTEINS

i f+ + +'Ca+" overload Altered Cat+ homeostasis I Sensitivityf to Ca++

Excitation-contraction uncoupling-..~ IL.IMECHANICAL DYSFUNCTION

TRUCTURESILE or COLLAGENS MATRIX)

Loss ofmechanicalcoupling

FIGURE 1. Proposed pathogenesis of postischemic myocardial dysfunction. This proposal integrates and reconciles differentmechanisms into a unifyingpathogenetic hypothesis. Transient reversible ischemia followed by reperfusion could result in increasedproduction of superoxide radicals (°02 ) through several mechanisms, including 1) increased activity of xanthine oxidase, 2)activation ofneutrophils, 3) activation ofarachidonate cascade, 4) accumulation ofreducing equivalents during oxygen deprivation,5) derangements of intramitochondrial electron transport system resulting in increased univalent reduction of oxygen, and 6)autoxidation of catecholamines and other substances. Superoxide dismutase (SOD) dismutates '02- to hydrogen peroxide (H202);in the presence of catalytic iron, '02- and 11202 interreact in a Haber- Weiss reaction to generate the hydroxyl radical (.OH). H202can also generate *OH in the absence of '02 through a Fenton reaction provided that other substances (such as ascorbate) reduceFe (III) to Fe (II). '02- and OH attack proteins and polyunsaturated fatty acids, causing enzyme inactivation and lipidperoxidation, respectively. In reversible ischemia, the intensity of this damage is not sufficient to cause cell death but is sufficient toproduce dysfunction ofkey cellular organelles. Postulated targets offree radical damage include 1) the sarcolemma, with consequentloss of selective permeability, impairment of calcium-stimulated ATPase activity and calcium transport out of the cell, andimpairnent ofNa+K+-ATPase activity (the net result of these perturbations would be increased transarcolemmal calcium influxand cellular calcium overload); 2) the sarcoplasmic reticulum, with consequent impairment of calcium-stimulated A TPase activityand calcium transport (this would result in impaired calcium homeostasis- specifically, decreased calcium sequestration, whichwould contribute to increase free cytosolic calcium, and decreased calcium release during systole, which would causeexcitation-contraction uncoupling); and 3) possibly other structures, such as the extracellular collagen matrix (with consequent lossof mechanical coupling) or the contractile proteins (with consequent decreased sensitivity to calcium). At the same time, reversibleischemia-reperfusion could cause cellular Na+ overload due to 1) inhibition ofsarcolemmal Na+,K+-ATPase and 2) acidosis andNa+-H+ exchange. This could further exaggerate calcium overload by increased Na +-Ca2' exchange. An increase in free cytosoliccalcium would activate phospholipases and other degradative enzymes and further exacerbate the injury to the aforementioned keysubcellular structures (sarcolemma, sarcoplasmic reticulum, and contractile proteins). Thus, calcium overload could serve toamplify the damage initiated by oxygen radicals. In addition, calcium overload could in itself impair contractile performance andcontribute to mechanical dysfunction. It is also possible that the increase in free cytosolic calcium could increase oxyradicalproduction bypromoting conversion ofxanthine dehydrogenase to xanthine oxidase. The ultimate consequence ofthis complex seriesofperturbations is a reversible depression of contractility.

SummaryAmong the numerous mechanisms proposed for

myocardial stunning, three appear to be more plau-sible: 1) generation of oxygen radicals, 2) calciumoverload, and 3) excitation-contraction uncoupling.

First, the evidence for a pathogenetic role ofoxygen-derived free radicals in myocardial stunningis overwhelming. In the setting of a single 15-minute

coronary occlusion, mitigation of stunning by antiox-idants has been reproducibly observed by severalindependent laboratories. Similar protection hasbeen recently demonstrated in the conscious animal,that is, in the most physiological experimental prep-aration available. Furthermore, generation of freeradicals in the stunned myocardium has been directlydemonstrated by spin trapping techniques, and atten-

m




uation of free radical generation has been repeatedlyshown to result in attenuation of contractile dysfunc-tion. Numerous observations suggest that oxyradicalsalso contribute to stunning in other settings: afterglobal ischemia in vitro, after global ischemia duringcardioplegic arrest in vivo, and after multiple briefepisodes of regional ischemia in vivo. Compellingevidence indicates that the critical free radical dam-age occurs in the initial moments of reflow, so thatmyocardial stunning can be viewed as a sublethalform of oxyradical-mediated "reperfusion injury."

Second, there is also considerable evidence that atransient calcium overload during early reperfusioncontributes to postischemic dysfunction in vitro; how-ever, the importance of this mechanism in vivoremains to be defined.

Third, inadequate release of calcium by the sarco-plasmic reticulum, with consequent excitation-contraction uncoupling, may occur after multiplebrief episodes of regional ischemia, but its role inother forms of postischemic dysfunction has not beenexplored.

It is probable that multiple mechanisms contributeto the pathogenesis of myocardial stunning. Thethree hypotheses outlined above are not mutuallyexclusive and in fact may represent different steps ofthe same pathophysiological cascade. Thus, genera-tion of oxyradicals may cause sarcoplasmic reticulumdysfunction, and both of these processes may lead tocalcium overload, which in turn could exacerbate thedamage initiated by oxygen species. The conceptsdiscussed in this review should provide not only aconceptual framework for further investigation of thepathophysiology of reversible ischemia-reperfusioninjury but also a rationale for developing clinicallyapplicable interventions designed to prevent postisch-emic ventricular dysfunction.

AcknowledgmentThe excellent secretarial assistance of William

LaCour is gratefully acknowledged.References

1. Braunwald E, Kloner RA: The stunned myocardium: Pro-longed, postischemic ventricular dysfunction. Circulation1982;66:1146-1149

2. Jennings RB, Reimer KA: Factors involved in salvagingischemic myocardium: Effects of reperfusion of arterialblood. Circulation 1983;68(suppl I):I-25-I-36

3. Simpson PJ, Lucchesi BR: Free radicals and myocardialischemia and reperfusion injury. J Lab Clin Med 1987;110:13-30

4. Becker LC, Ambrosio G: Myocardial consequences of reper-fusion. Prog Cardiovasc Dis 1987;30:23-44

5. Kloner RA, Przyklenk K, Whittaker P: Deleterious effects ofoxygen radicals in ischemia/reperfusion: Resolved and unre-solved issues. Circulation 1989;80:1115-1127

6. Opie LH: Reperfusion injury and its pharmacologic modifi-cation. Circulation 1989;80:1049-1062

7. Engler R, Gilpin E: Can superoxide dismutase alter myocar-dial infarct size? Circulation 1989;79:1137-1142

8. Kloner RA, Ellis SG, Lange R, Braunwald E: Studies ofexperimental coronary artery reperfusion: Effects on infarct

size, myocardial function, biochemistry, ultrastructure andmicrovascular damage. Circulation 1983;68(suppl I):I-8-I-15

9. Heyndrickx GR, Millard RW, McRitchie RJ, Maroko PR,Vatner SF: Regional myocardial functional and electrophys-iological alterations after brief coronary artery occlusion inconscious dogs. J Clin Invest 1975;56:978-985

10. Weiner JM, Apstein CS, Arthur JH, Pirzada FA, Hood WBJr: Persistence of myocardial injury following brief periods ofcoronary occlusion. Cardiovasc Res 1976;10:678-686

11. Heyndrickx GR, Baig H, Nellens P, Leusen I, Fishbein MC,Vatner SF: Depression of regional blood flow and wallthickening after brief coronary occlusions. Am J Physiol1978;234:H653-H659

12. Bolli R, Zhu WX, Thornby JI, O'Neill PG, Roberts R:Time-course and determinants of recovery of function afterreversible ischemia in conscious dogs. Am J Physiol 1988;254:H102-H114

13. Charlat ML, O'Neill PG, Hartley CJ, Roberts R, Bolli R:Prolonged abnormalities of left ventricular diastolic wallthinning in the "stunned" myocardium in conscious dogs:Time-course and relation to systolic function. J Am CollCardiol 1989;13:185-194

14. Bolli R, Patel BS, Hartley CJ, Thornby JI, Jeroudi MO,Roberts R: Nonuniform transmural recovery of contractilefunction in the "stunned" myocardium. Am J Physiol 1989;257:H375-H385

15. Nicklas JM, Becker LC, Bulkley BH: Effects of repeated briefcoronary occlusion on regional left ventricular function anddimension in dogs. Am J Cardiol 1985;56:473-478

16. Stahl LD, Aversano TR, Becker LC: Selective enhancementof function of stunned myocardium by increased flow. Circu-lation 1986;74:843-851

17. Becker LC, Levine JH, DiPaula AF, Guarnieri T, AversanoTR: Reversal of dysfunction in postischemic stunned myo-cardium by epinephrine and postextrasystolic potentiation. JAm Coll Cardiol 1986;7:580-589

18. Schroder E, Kieso RA, Laughlin D, Schroder M, Meng R,Kerber RE: Altered response of reperfused myocardium torepeated coronary occlusion in dogs. J Am Coll Cardiol1987;10:898-905

19. Stahl LD, Weiss HR, Becker LC: Myocardial oxygen con-sumption, oxygen supply/demand heterogeneity, and micro-vascular patency in regionally stunned myocardium. Circula-tion 1988;77:865-872

20. Cohen MV, Downey JM: Myocardial stunning in dogs:Preconditioning effect and influence of coronary collateralflow. Am Heart J (in press)

21. Theroux P, Ross J Jr, Franklin D, Kemper WS, Sasayama S:Coronary arterial reperfusion: III. Early and late effects onregional myocardial function and dimensions in consciousdogs. Am J Cardiol 1976;38:599-606

22. Lavallee M, Cox D, Patrick TA, Vatner SF: Salvage ofmyocardial function by coronary artery reperfusion 1, 2, and3 hours after occlusion in conscious dogs. Circ Res 1983;53:235-247

23. Bush LR, Buja LM, Samowitz W, Rude RE, Wathen M,Tilton GD, Willerson JT: Recovery of left ventricular seg-mental function after long-term reperfusion following tem-porary coronary occlusion in conscious dogs: Comparison of2- and 4-hour occlusions. Circ Res 1983;53:248-263

24. Ellis SG, Henschke CI, Sandor T, Wynne J, Braunwald E,Kioner RA: Time course of functional and biochemicalrecovery of myocardium salvaged by reperfusion. JAm CollCardiol 1983;1:1047-1055

25. Matsuzaki M, Gallagher KP, Kemper WS, White F, Ross JJr: Sustained regional dysfunction produced by prolongedcoronary stenosis: Gradual recovery after reperfusion. Circu-lation 1983;68:170-182

26. Taegtmeyer H, Roberts AFC, Raine AEG: Energy metabo-lism in reperfused heart muscle: Metabolic correlates toreturn of function. JAm Coll Cardiol 1985;6:864-870

27. Kusuoka H, Porterfield JK, Weisman HF, Weisfeldt ML,Marban E: Pathophysiology and pathogenesis of stunnedmyocardium: Depressed Ca2' activation of contraction as a




consequence of reperfusion-induced cellular calcium over-load in ferret hearts. J Clin Invest 1987;79:950-961

28. Steenbergen C, Murphy E, Levy L, London RE: Elevation incytosolic free calcium concentration early in myocardialischemia in perfused rat heart. Circ Res 1987;60:700-707

29. Porterfield JK, Kusuoka H, Weisman HF, Weisfeldt ML,Marban E: Ryanodine prevents the changes in myocardialfunction and morphology induced by reperfusion after briefperiods of ischemia (abstract). Circ Res 1987;35:315A

30. Ambrosio G, Jacobus WE, Bergman CA, Weisman HF,Becker LC: Preserved high energy phosphate metabolicreserve in globally "stunned" hearts despite reduction ofbasal ATP content and contractility. J Mol Cell Cardiol1987;19:953-964

31. Kitakaze M, Weisfeldt ML, Marban E: Acidosis during earlyreperfusion prevents myocardial stunning in perfused ferrethearts. J Clin Invest 1988;82:920-927

32. Kusuoka H, Koretsune Y, Chacko VP, Weisfeldt ML, Mar-ban E: Excitation-contraction coupling in postischemic myo-cardium: Does failure of activator Ca21 transients underliestunning? Circ Res 1990;66:1268-1276

33. Ambrosio G, Jacobus WE, Mitchell MC, Litt MR, BeckerLC: Effects of ATP precursors on ATP and free ADP contentand functional recovery of postischemic hearts. Am J Physiol1989;256:H560-H566

34. Marban E, Koretsune Y, Corretti M, Chacko VP, Kusoka H:Calcium and its role in myocardial cell injury during ischemiaand reperfusion. Circulation 1989;80(suppl IV):IV-17-IV-22

35. Schlafer M, Kane PF, Kirsh MM: Superoxide dismutase pluscatalase enhances the efficacy of hypothermic cardioplegia toprotect the globally ischemic, reperfused heart. J ThoracCardiovasc Surg 1982;83:830-839

36. Shlafer M, Kane PF, Wiggins VY, Kirsh MM: Possible rolefor cytotoxic oxygen metabolites in the pathogenesis ofcardiac ischemic injury. Circulation 1982;66(suppl I):I-85-I-92

37. Casale AS, Bulkley GB, Bulkley BH, Flaherty JT, Gott VL,Gardner TJ: Oxygen free radical scavengers protect thearrested globally ischemic heart upon reperfusion. SurgForum 1983;34:313-316

38. Menasche P, Grousset C, Gauduel Y, Piwnica A: A compar-ative study of free radical scavengers in cardioplegic solu-tions: Improved protection with peroxidase. J Thorac Cardio-vasc Surg 1986;92:264-271

39. Myers CL, Weiss SJ, Kirsh MM, Shepard BM, Shlafer M:Effects of supplementing hypothermic crystalloid cardiople-gic solution with catalase, superoxide dismutase, allopurinol,or deferoxamine on functional recovery of globally ischemicand reperfused isolated hearts. J Thorac Cardiovasc Surg1986;91:281-289

40. Ambrosio G, Weisfeldt ML, Jacobus WE, et al: Evidence fora reversible oxygen radical-mediated component of reperfu-sion injury: Reduction by recombinant human superoxidedismutase administered at the time of reflow. Circulation1987;75:282-291

41. Ytrehus K, Gunnes S, Myklebust R, Mjos OD: Protection bysuperoxide dismutase and catalase in the isolated rat heartreperfused after prolonged cardioplegia: A combined studyof metabolic, functional and morphometric ultrastructuralvariables. Cardiovasc Res 1987;21:492-499

42. Garlick PB, Davies MJ, Hearse DJ, Slater TF: Direct detec-tion of free radicals in the reperfused rat heart using electronspin resonance spectroscopy. Circ Res 1987;61:757-760

43. Kramer JH, Arroyo CM, Dickens BF, Weglicki WB: Spin-trapping evidence that graded myocardial ischemia alterspostischemic superoxide production. Free Radic Biol Med1987;3:153-159

44. Ambrosio G, Zweier JL, Jacobus WE, Weisfeldt ML, Fla-herty JT: Improvement of postischemic myocardial functionand metabolism induced by administration of desferrioxam-ine at the time of reflow: The role of iron in the pathogenesisof reperfusion injury. Circulation 1987;76:906-915

45. Stewart JR, Blackwell WH, Crute SL, Loughlin V, Green-field LJ, Hess ML: Inhibition of surgically induced ischemia/

reperfusion injury by oxygen free radical scavengers. J ThoracCardiovasc Surg 1983;86:262-272

46. Johnson DL, Horneffer PJ, Dinatale JM Jr, Gott VL, Gard-ner TJ: Free radical scavengers improve functional recoveryof stunned myocardium in a model of surgical coronaryrevascularization. Surgery 1987;102:334-340

47. Gardner TJ: Oxygen radicals in cardiac surgery. Free RadicBiol Med 1988;4:45-50

48. Illes RW, Silverman NA, Krukenkamp IB, del Nido PJ,Levitsky S: Amelioration of postischemic stunning bydeferoxamine-blood cardioplegia. Circulation 1989;80(supplII1):III-30-III-35

49. Homans DC, Sublett E, Dai XZ, Bache RJ: Persistence ofregional left ventricular dysfunction after exercise-inducedmyocardial ischemia. J Clin Invest 1986;77:66-73

50. Gross GJ, Farber NE, Hardman HF, Warltier DC: Beneficialactions of superoxide dismutase and catalase in stunnedmyocardium of dogs. Am J Physiol 1986;250:H372-H377

51. O'Neill PG, Charlat ML, Hartley CJ, Roberts R, Bolli R:Nonuniform transmural response of the "stunned" myocar-dium to inotropic stimulation (abstract). JAm Coll Cardiol1987;9:145A

52. Preuss KC, Gross GJ, Brooks HL, Warltier DC: Time courseof recovery of "stunned" myocardium following variableperiods of ischemia in conscious and anesthetized dogs. AmHeart J 1987;114:696-703

53. DeBoer FWV, Ingwall JS, Kloner RA, Braunwald E: Pro-longed derangements of canine myocardial purine metabo-lism after a brief coronary artery occlusion not associatedwith anatomic evidence of necrosis. Proc Natl Acad Sci USA1980;77:5471-5475

54. Reimer KA, Hill ML, Jennings RB: Prolonged depletion ofATP and of the adenine nucleotide pool due to delayedresynthesis of adenine nucleotides following reversible myo-cardial ischemic injury in dogs. J Mol Cell Cardiol 1981;13:229-239

55. Swain JL, Sabina RL, McHale PA, Greenfield JC Jr, HolmesEW: Prolonged myocardial nucleotide depletion after briefischemia in the open-chest dog. Am J Physiol 1982;242:H818-H826

56. Neely JR, Grotyohann LW: Role of glycolytic products indamage to ischemic myocardium: Dissociation of adenosinetriphosphate levels and recovery of function of reperfusedischemic hearts. Circ Res 1984;55:816-824

57. Przyklenk K, Kloner RA: Superoxide dismutase plus catalaseimprove contractile function in the canine model of the"stunned" myocardium. Circ Res 1986;58:148-156

58. Glower DD, Spratt JA, Newton JR, Wolfe JA, Rankin JS,Swain JL: Dissociation between early recovery of regionalfunction and purine nucleotide content in postischaemicmyocardium in the conscious dog. Cardiovasc Res 1987;21:328-336

59. Przyklenk K, Kloner RA: Effect of verapamil on postischemic"stunned" myocardium: Importance of the timing of treat-ment. JAm Coll Cardiol 1988;11:614-623

60. Guth BD, Martin JF, Heusch G, Ross J Jr: Regional myo-cardial blood flow, function and metabolism usingphosphorus-31 nuclear magnetic resonance spectroscopyduring ischemia and reperfusion in dogs. J Am Coll Cardiol1987;10:673-681

61. Mercier JC, Lando U, Kanmatsuse K, Ninomiya K, Meer-baum S, Fishbein MC, Swan HJC, Ganz W: Divergent effectsof inotropic stimulation on the ischemic and severelydepressed reperfused myocardium. Circulation 1982;66:397-400

62. Ellis SE, Wynne J, Braunwald E, Henschke CI, Sandor T,Kloner RA: Response of reperfusion-salvaged, stunned myo-cardium to inotropic stimulation. Am Heart J 1984;107:9-13

63. Arnold JMO, Braunwald E, Sandor T, Kloner RA: Inotropicstimulation of reperfused myocardium with dopamine:Effects on infarct size and myocardial function. J Am CollCardiol 1985;6:1026-1034

64. Bolli R, Zhu WX, Myers ML, Hartley CJ, Roberts R:Beta-adrenergic stimulation reverses postischemic myocar-




dial dysfunction without producing subsequent functionaldeterioration. Am J Cardiol 1985;56:964-968

65. Ito BR, Tate H, Kobayashi M, Schaper W: Reversiblyinjured, postischemic canine myocardium retains normalcontractile reserve. Cire Res 1987;61:834-846

66. Heusch G, Schafer S, Kroger K: Recruitment of inotropicreserve in "stunned" myocardium by the cardiotonic agentAR-L 57. Basic Res Cardiol 1988;83:602-610

67. Hoffmeister HM, Mauser M, Schaper W: Effect of adenosineand AICAriboside on ATP content and regional contractilefunction in reperfused canine myocardium. Basic Res Cardiol1985;80:445-458

68. Greenfield RA, Swain JL: Disruption of myofibrillar energyuse: Dual mechanisms that may contribute to postischemicdysfunction in stunned myocardium. Circ Res 1987;60:283-289

69. Ciuffo AA, Ouyang P, Becker LC, Levin L, Weisfeldt ML:Reduction of sympathetic inotropic response after ischemiain dogs: Contributor to stunned myocardium. J Clin Invest1985;75:1504-1509

70. Heusch G, Frehen D, Kroger K, Schulz R, Thamer V:Integrity of sympathetic neurotransmission in stunned myo-cardium. JAppl Cardiol 1988;3:259-272

71. O'Neill PG, Charlat ML, Michael LH, Roberts R, Bolli R:Influence of neutrophil depletion on myocardial function andflow after reversible ischemia. Am J Physiol 1989;256:H341-H351

72. Bolli R, Triana JF, Jeroudi MO: Postischemic mechanicaland vascular dysfunction (myocardial "stunning" and micro-vascular "stunning") and the effects of calcium channelblockers on ischemia/reperfusion injury. Clin Cardiol 1989;12:III-16-III-25

73. Bolli R, Triana JF, Jeroudi MO: Prolonged impairment ofcoronary vasodilation after reversible ischemia: Evidence formicrovascular "stunning." Circ Res 1990;67:332-343

74. Zhao M, Zhang H, Robinson TF, Factor SM, SonnenblickEH, Eng C: Profound structural alterations of the extracellularcollagen matrix in postischemic dysfunctional ("stunned")but viable myocardium. JAm Coll Cardiol 1987;10:1322-1334

75. Charney RH, Takahashi S, Zhao M, Sonnenblick EH, FactorSM, Eng C: Collagen loss in the stunned myocardium(abstract). Circulation 1989;80(suppl II):II-99

76. Whittaker P, Przyklenk K, Boughner DR, Kloner RA: Col-lagen damage in two different models of stunned myocar-dium (abstract). J Mol Cell Cardiol 1989;21(suppl II):S163

77. Guarnieri T: Direct measurement of [Ca2"Ji in early andlate reperfused myocardium (abstract). Circulation 1989;80(suppl II):II-241

78. Kitakaze M, Weisman HF, Marban E: Contractile dysfunc-tion and ATP depletion after transient calcium overload inperfused ferret hearts. Circulation 1988;77:685-695

79. Hori M, Kitakaze M, Sato H, Iwakura K, Gotoh K, KusokaH, Kitabatake A: Transient acidosis by staged reperfusionprevents myocardial stunning (abstract). Circulation 1989;80(suppl II):II-600

80. Marban E, Kitakaze M, Kusoka H, Porterfield JK, Yue DT,Chacko VP: Intracellular free calcium concentration mea-sured with `9F NMR spectroscopy in intact ferret hearts. ProcNatlAcad Sci USA 1987;84:6005-6009

81. Lee H-C, Smith N, Mohabir R, Chisin WT: Cytosolic calciumtransients from the beating mammalian heart. Proc NatlAcadSci USA 1987;84:7793-7797

82. Marban E, Kitakaze M, Koretsune Y, Yue DT, Chacko VP,Pike MM: Quantification of [Ca2+]i in perfused hearts: Crit-ical evaluation of the SF-BAPTA and nuclear magneticresonance method as applied to the study of ischemia andreperfusion. Circ Res 1990;66:1255-1267

83. Krause SM, Hess ML: Characterization of cardiac sarcoplas-mic reticulum dysfunction during short-term normothermicglobal ischemia. Circ Res 1985;55:176-184

84. Lazdunski M, Frelin C, Vigue P: The sodium/hydrogenexchange system in cardiac cells: Its biochemical and phar-macological properties and its role in regulating internal

concentrations of sodium and internal pH. J Mol Cell Cardiol1985;17:1029-1042

85. Murphy JG, Smith TW, Marsh JD: Mechanisms ofreoxygenation-induced calcium overload in cultured chickembryo heart cells. Am J Physiol 1988;254:H1133-H1141

86. Tani M, Neely JR: Role of intracellular Na+ in Ca2' overloadand depressed recovery of ventricular function of reperfusedischemic rat hearts: Possible involvement of H+-Na+ andNa+-Ca2' exchange. Circ Res 1989;65:1045-1056

87. Grinwald PM: Calcium uptake during postischemic reperfu-sion in the isolated rat heart: Influence of extracellularsodium. J Mol Cell Cardiol 1982;14:359-365

88. Renlund DG, Gerstenblith G, Lakatta EG, Jacobus WE,Kallman CH, Weisfeldt ML: Perfusate sodium during isch-emia modifies postischemic functional and metabolic recov-ery in the rabbit heart. J Mol Cell Cardiol 1984;16:795-801

89. McCord JM: Oxygen-derived free radicals in postischemictissue injury. N Engl J Med 1985;312:159-163

90. Taylor AL, Golino P, Eckels R, Pastor P, Buja LM, WillersonJT: Differential enhancement of postischemic segmentalsystolic thickening by diltiazem. J Am Coll Cardiol 1990;15:737-747

91. Lamping KA, Gross GJ: Improved recovery of myocardialsegment function following a short coronary occlusion in dogsby nicorandil, a potential new antianginal agent, and nifedi-pine. J Cardiovasc Pharmacol 1985;7:158-166

92. Warltier DC, Gross GJ, Brooks HL, Preuss KC: Improve-ment of postischemic, contractile function by the calciumchannel blocking agent nitrendipine in conscious dogs. JCardiovasc Pharmacol 1988;12(suppl 4):S120-S124

93. Dunlap E, Millard RW: Amlodipine, a new long-actingcalcium channel blocking agent, improves recovery of"stunned" myocardium (abstract). Pharmacologist 1989;31:145

94. Przyklenk K, Ghafari GB, Eitzman DT, Kloner RA: Nifedi-pine administered after reperfusion ablates systolic contrac-tile dysfunction of postischemic "stunned" myocardium. JAm Coll Cardiol 1989;13:1176-1183

95. Krause SM, Jacobus WE, Becker LC: Alterations in cardiacsarcoplasmic reticulum calcium transport in the postischemic"stunned" myocardium. Circ Res 1989;65:526-530

96. Zhu WX, Myers ML, Hartley CJ, Roberts R, Bolli R:Validation of a single crystal for measurement of transmur-al and epicardial thickening. Am J Physiol 1986;251:H1045-H1055

97. Myers ML, Bolli R, Lekich RF, Hartley CJ, Roberts R:Enhancement of recovery of myocardial function by oxygenfree-radical scavengers after reversible regional ischemia.Circulation 1985;72:915-921

98. Murry CE, Richard VJ, Jennings RB, Reimer KA: Freeradicals do not cause myocardial stunning after four 5 minutecoronary occlusions (abstract). Circulation 1989;80(supplII):II-296

99. Jeroudi MO, Triana FJ, Patel BS, Bolli R: Effect of super-oxide dismutase and catalase, given separately, on myocardial"stunning." Am J Physiol (in press)

100. Buchwald A, Klein HH, Lindert S, Pich S, Nebendahl K,Wiegand V, Kreuzer H: Effect of intracoronary superoxidedismutase on regional function in stunned myocardium. JCardiovasc Pharmacol 1989;13:258-264

101. Halliwell PB, Gutteridge JMC: Oxygen toxicity, oxygen rad-icals, transition metals and disease. Biochem J 1984;219:1-14

102. Fox RB: Prevention of granulocyte-mediated oxidant lunginjury in rats by a hydroxyl radical scavenger, dimethylthio-urea. J Clin Invest 1984;74:1456

103. Bolli R, Zhu WX, Hartley CJ, Michael LH, Repine J, HessML, Kukreja RC, Roberts R: Attenuation of dysfunction inthe postischemic "stunned" myocardium by dimethylthiorea.Circulation 1987;76:458-468

104. Bolli R, Jeroudi MO, Patel BS, Aruoma 01, Halliwell B, LaiEK, McCay PB: Marked reduction of free radical generationand contractile dysfunction by antioxidant therapy begun atthe time of reperfusion: Evidence that myocardial "stunning"




is a manifestation of reperfusion injury. Circ Res 1989;65:607-622

105. Myers ML, Bolli R, Lekich RF, Hartley CJ, Michael LH,Roberts R: N-2-Mercaptopropionylglycine improves recov-ery of myocardial function following reversible regional isch-emia. JAm Coll Cardiol 1986;8:1161-1168

106. Bolli R, Patel BS, Zhu WX, O'Neill PG, Charlat ML,Roberts R: The iron chelator desferrioxamine attenuatespostischemic ventricular dysfunction. Am J Physiol 1987;253:H1372-H1380

107. Farber NE, Vercellotti GM, Jacob HS, Pieper GM, GrossGJ: Evidence for a role of iron-catalyzed oxidants in func-tional and metabolic stunning in the canine heart. Circ Res1988;63:351-360

108. Schrier GM, Hess ML: Quantitative identification of super-oxide anion as a negative inotropic species. Am J Physiol1988;24:H138-H143

109. DiGuiseppi J, Fridovich I: Oxygen toxicity in Streptococcussanguis: The relative importance of superoxide and hydroxylradicals. J Biol Chem 1982;257:4046-4051

110. Triana JF, Unisa A, Bolli R: Antioxidant enzymes attenuatemyocardial "stunning" in the conscious dog. FASEB J 1990;4:A622

111. Zweier JL, Flaherty JT, Weisfeldt ML: Direct measurementof free radical generation following reperfusion of ischemicmyocardium. Proc NatlAcad Sci USA 1987;84:1404-1407

112. Baker JE, Felix CC, Olinger GN, Kalyanaraman B: Myocar-dial ischemia and reperfusion: Direct evidence for freeradical generation by electron spin resonance spectroscopy.Proc NatlAcad Sci USA 1988;85:2786-2789

113. Zweier JL: Measurement of superoxide-derived free radicalsin the reperfused heart. J Biol Chem 1988;263:1353-1357

114. Bolli R, McCay PB: Use of spin traps in intact animalsundergoing myocardial ischemia/reperfusion: A newapproach to assessing the role of oxygen radicals in myocar-dial "stunning." Free Rad Res Comms 1990;9:169-180

115. Bolli R, Patel BS, Jeroudi MO, Lai EK, McCay PB: Demon-stration of free radical generation in "stunned" myocardiumof intact dogs with the use of the spin trap a-phenyl N-tert-butyl nitrone. J Clin Invest 1988;82:476-485

116. Leiboff RL, Arroyo CM, Schaer GL, Mergner GW, KramerJH, Miller DL, Visner MS, Weglicki WB: Free radicalgeneration in an in vivo model of regional myocardial stun-ning (abstract). FASEB J 1988;2:A818

117. Bolli R, Jeroudi MO, Patel BS, DuBose CM, Lai EK,Roberts R, McCay PB: Direct evidence that oxygen-derivedfree radicals contribute to postischemic myocardial dysfunc-tion in the intact dog. Proc Natl Acad Sci USA 1989;86:4695-4699

118. Burton KP, McCord JM, Ghai G: Myocardial alterations dueto free-radical generation.Am JPhysiol 1984;246:H776-H783

119. Blaustein AS, Schine L, Brooks WW, Fanburg BL, BingOHL: Influence of exogenously generated oxidant species onmyocardial function. Am J Physiol 1986;250:H595-H599

120. Shattock MJ, Manning AS, Hearse DJ: Effects of hydrogenperoxide on cardiac function and postischaemic functionalrecovery in the isolated "working" rat heart. Pharmacology1982;24:118-122

121. Ytrehus K, Myklebust R, Mjos OD: Influence of oxygenradicals generated by xanthine oxidase in the isolated per-fused rat heart. Cardiovasc Res 1986;20:597-603

122. Miki S, Ashraf M, Salka S, Sperelakis N: Myocardial dysfunc-tion and ultrastructural alterations mediated by oxygenmetabolites. J Mol Cell Cardiol 1988;20:1009-1024

123. Jackson CV, Mickelson JK, Pope TK, Rao PS, Lucchesi BR:02 free radical-mediated myocardial and vascular dysfunc-tion.Am J Physiol 1986;251:H1225-H1231

124. Goldhaber JI, Ji S, Lamp ST, Weiss JN: Effects of exogenousfree radicals on electromechanical function and metabolismin isolated rabbit and guinea pig ventricle: Implications forischemia and reperfusion injury. J Clin Invest 1988;83:1800-1809

125. Przyklenk K, Whittaker P, Kloner RA: In vivo infusion ofoxygen free radical substrates causes myocardial systolic, butnot diastolic dysfunction. Am Heart J 1990;119:807-815

126. Davies KJA: Protein damage and degradation by oxygenradicals: I. General aspects. JBiol Chem 1987;262:9895-9901

127. Thompson JA, Hess ML: The oxygen free radical system: Afundamental mechanism in the production of myocardialnecrosis. Prog Cardiovasc Dis 1986;28:449-462

128. Romaschin AD, Rebeyka I, Wilson GJ, Mickle DAG: Con-jugated dienes in ischemic and reperfused myocardium: Anin vivo chemical signature of oxygen free radical mediatedinjury. JMol Cell Cardiol 1987;19:289-302

129. Weisel RD, Mickle DAG, Finkle CD, Tumiati LC, MadonikMM, Ivanov J, Burton GW, Ingold KU: Myocardial free-radical injury after cardioplegia. Circulation 1989;80(supplIII):III-14-III-18

130. Rowe GT, Manson NH, Caplan M, Hess ML: Hydrogenperoxide and hydroxyl radical mediation of activated leuko-cyte depression of cardiac sarcoplasmic reticulum: Participa-tion of the cyclooxygenase pathway. Circ Res 1983;53:584-591

131. Kaneko M, Beamish RE, Dhalla NS: Depression of heartsarcolemmal Ca2' pump activity by oxygen free radicals. AmJ Physiol 1989;256:H368-H374

132. Kaneko M, Elimban V, Dhalla NS: Mechanism for depres-sion of heart sarcolemmal Ca2' pump by oxygen free radicals.Am J Physiol 1989;257:H804-H811

133. Reeves JP, Bailey CA, Hale CC: Redox modification ofsodium-calcium exchange activity in cardiac sarcolemmalvesicles. J Biol Chem 1986;261:4948-4955

134. Kramer JH, Mak IT, Weglicki WB: Differential sensitivity ofcanine cardiac sarcolemmal and microsomal enzymes toinhibition by free radical-induced lipid peroxidation. Circ Res1984;55:120-124

135. Kim M-S, Akera T: 02 free radicals: Cause of ischemia-reperfusion injury to cardiac Na+-K+-ATPase. Am J Physiol1987;252:H252-H257

136. Charlat ML, O'Neill PG, Egan JM, Abernethy DR, MichaelLH, Myers ML, Roberts R, Bolli R: Evidence for a pathoge-netic role of xanthine oxidase in the "stunned" myocardium.Am J Physiol 1987;252:H566-H577

137. Holzgrefe HH, Gibson JK: Enhanced function recovery inthe stunned canine myocardium by pretreatment with oxypu-rinol (abstract). JAm Coll Cardiol 1988;11:208A

138. Puett DW, Forman MB, Cates CU, Wilson BH, Hande KR,Friesinger GC, Virmani R: Oxypurinol limits myocardialstunning but does not reduce infarct size after reperfusion.Circulation 1987;76:678-686

139. Eddy LJ, Stewart JR, Jones HP, Engerson TD, McCord JM,Downey JM: Free radical-producing enzyme, xanthine oxi-dase, is undetectable in human hearts. Am J Physiol 1987;253:H709-H711

140. Jarasch ED, Bruder G, Heid HW: Significance of xanthineoxidase in capillary endothelial cells. Acta Physiol Scand1986;548(suppl):39-46

141. Muxfeldt M, Schaper W: The activity of xanthine oxidase inhearts of pigs, guinea pigs, rats, and humans. Basic ResCardiol 1987;82:486-492

142. Huizer T, de Jong JW, Nelson JA, Czarnecki W, Serruys PW,Bonnier JJRM, Troquay R: Urate production by humanheart. J Mol Cell Cardiol 1989;21:691-695

143. Grum CM, Gallagher KP, Kirsh MM, Shlafer M: Absence ofdetectable xanthine oxidase in human myocardium. J MolCell Cardiol 1989;21:263-267

144. Werns S, Ventura A, Li GC, Lucchesi BR: Amflutizole, a

xanthine oxidase inhibitor, does not attenuate myocardialstunning in the canine heart (abstract). Circulation 1989;80(suppl II):II-295

145. Lucchesi BR, Mullane KM: Leukocytes and ischemia-induced myocardial injury. Ann Rev Pharmacol Toxicol 1986;26:201-224

146. Engler R: Granulocytes and oxidative injury in myocardialischemia and reperfusion. Fed Proc 1987;46:2395-2396



738 Circulation Vol 82 No 3, September 1990

147. Shea MJ, Simpson PJ, Werns SW, Buda AJ, Alborzy-Khaghany AS, Hoff PT, Lucchesi BR: Effect of neutrophildepletion on recovery of "stunned" myocardium (abstract).Clin Res 1987;35:327A

148. Jeremy RW, Becker LC: Neutrophil depletion does notprevent myocardial dysfunction after brief coronary occlu-sion. JAm Coll Cardiol 1989;13:1155-1163

149. O'Neill PG, Charlat ML, Kim H-S, Pocius J, Michael LH,Hartley CJ, Roberts R, Bolli R: Lipoxygenase inhibitornafazatrom fails to attenuate postischemic ventricular dys-function. Cardiovasc Res 1987;21:755-760

150. Schott RJ, Nao BS, McClanahan TB, Simpson PJ, StirlingMC, Todd RF III, Gallagher KP: F(ab')2 fragments ofanti-mol (904) monoclonal antibodies do not prevent myo-cardial stunning. Circ Res 1989;65:1112-1124

151. Kerber RE, Shasby DM, Seabold J, Kieso R, Fox-Eastham K:Does reduction of leukocyte accumulation in reperfusedmyocardium affect stunning (abstract)? Circulation 1989;80(suppl II):II-401

152. Go LO, Murry CE, Richard VJ, Jennings RB, WeischedelGR, Reimer KA: Myocardial neutrophil accumulation duringreperfusion after reversible or irreversible ischemic injury.Am J Physiol 1988;255:H1188-H1198

153. Engler RL, Dahlgren MD, Morris DD, Peterson MA,Schmid-Schonbein GW: Role of leukocytes in response toacute myocardial ischemia and reflow in dogs. Am J Physiol1986;251:H314-H322

154. Engler R, Covell JW: Granulocytes cause reperfusion ven-tricular dysfunction after 15-minute ischemia in the dog. CircRes 1987;61:20-28

155. Westlin W, Mullane KM: Alleviation of myocardial stunningby leukocyte and platelet depletion. Circulation 1989;80:1828-1836

156. Kloner RA: Do neutrophils mediate the phenomenon ofstunned myocardium? JAm Coll Cardiol 1989;13:1164-1166

157. Westlin W, Mullane KM: Does captopril attenuatereperfusion-induced myocardial dysfunction by scavengingfree radicals? Circulation 1988;77(suppl I):I-30-I-39

158. Ferrari R, Alfieri 0, Curello S, Ceconi A, Cargnoni A,Marzollo P, Pardini A, Caradonna E, Visioli 0: Occurrenceof oxidative stress during reperfusion of the human heart.Circulation 1990;81:201-211

159. Asinger RW, Peterson DA, Elsperger KJ, Homans D: Long-term recovery of LV wall thickening after 1 hour of ischemiais not affected when superoxide dismutase and catalase areadministered during the first 45 minutes of reperfusion(abstract). JAm Coll Cardiol 1988;2:163A

160. Nejima J, Knight DR, Fallon JT, Uemura N, Manders WT,Canfield DR, Cohen MV, Vatner SF: Superoxide dismutasereduces reperfusion arrhythmias but fails to salvage regionalmyocardial function or myocardium at risk in conscious dogs.Circulation 1989;79:143-153

161. Przyklenk K, Kloner RA: "Reperfusion injury" by oxygen-derived free radicals? Effect of superoxide dismutase pluscatalase, given at the time of reperfusion, on myocardialinfarct size, contractile function, coronary microvasculature,and regional myocardial blood flow. Circ Res 1989;64:86-96

162. Patel BS, Jeroudi MO, O'Neill PG, Roberts R, Bolli R:Effect of human recombinant superoxide dismutase oncanine myocardial infarction. Am J Physiol 1990;258:H369-H380

163. Forman MB, Puett DW, Cates CU, McCroskey DE, Beck-man JK, Greene HL, Virmani R: Glutathione redox pathwayand reperfusion injury: Effect of N-acetylcysteine on infarctsize and ventricular function. Circulation 1988;78:202-213

164. Homans DC, Sublett E, Asinger R, Peterson D, Bache RJ:SOD+catalase does not attenuate regional left ventriculardysfunction following exercise induced ischemia (abstract). JAm Coll Cardiol 1988;78:II-76

165. Hearse DJ, Humphrey SM, Bullock GR: The oxygen paradoxand the calcium paradox: Two facets of the same problem? JMol Cell Cardiol 1978;10:641-668

166. Patel B, Kloner RA, Przyklenk K, Braunwald E: Postisch-emic myocardial "stunning": A clinically relevant phenome-non. Ann Intem Med 1988;108:626-628



R BolliMechanism of myocardial "stunning".

Print ISSN: 0009-7322. Online ISSN: 1524-4539 Copyright © 1990 American Heart Association, Inc. All rights reserved.

is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231Circulation doi: 10.1161/01.CIR.82.3.723

1990;82:723-738Circulation.

http://circ.ahajournals.org/content/82/3/723the World Wide Web at:

The online version of this article, along with updated information and services, is located on

http://circ.ahajournals.org//subscriptions/

is online at: Circulation Information about subscribing to Subscriptions:

http://www.lww.com/reprints Information about reprints can be found online at: Reprints:

document. Permissions and Rights Question and Answer information about this process is available in the

located, click Request Permissions in the middle column of the Web page under Services. FurtherEditorial Office. Once the online version of the published article for which permission is being requested is

can be obtained via RightsLink, a service of the Copyright Clearance Center, not theCirculationpublished in Requests for permissions to reproduce figures, tables, or portions of articles originallyPermissions:


http://circ.ahajournals.org/content/82/3/723

http://www.ahajournals.org/site/rights/

http://www.lww.com/reprints

http://circ.ahajournals.org//subscriptions/


circulation 1990 bolli 723 38

Documents

postischemic dysfunction

myocardial infarction

ormyocardial stunning

stunning r

myocardial stunning

baylor college of medicine

department of medicine

spontaneous reperfusion