- Email: [email protected]
Mechanisms of regional ischemia and antianginal drug action during exercise
Mechanisms
of Regional Ischemia and Antianginal During Exercise
Drug Action
John Ross, Jr
A
NGINA PECTORIS REPRESENTS a disorder of oxygen supply and demand and, of course, the most common type of angina pectoris is a consequence of regional ischemia produced by the increased oxygen demands of exercise or other stress. In addition, it is now clear that in some patients a coronary vasospastic component can contribute to ischemia by causing decreased oxygen supply during stress, or at rest. Ischemia unaccompanied by angina pectoris, so-called silent ischemia, also occurs frequently in patients with coronary heart disease. Many effective antianginal drugs having widely different properties are now available, in particular the long-acting nitrates, P-adrenergic and calcium channel blocking agents. In the clinical setting, understanding of the physiologic mechanisms involved in exercise-induced ischemia and the actions of these drugs has been limited by the lack of methods for accurately quantifying in man the degree of ischemia during exercise and its effect on the myocardium. The regional nature of myocardial ischemia, such as that produced by exercise in the presence of coronary artery narrowing, has certain unique features which contrast with the effects of global ischemia. lV2Thus, the aortic BP usually increases during exercise (due to augmented cardiac output and enhanced inotropic state of normally perfused regions of the left ventricle), thereby enhancing coronary perfusion pressure, whereas contractile performance in the ischemic region deteriorates’ and is affected in a complex way by the reduced contractility, altered loading conditions, and interactions with the remainder of the ventricle.2 Moreover, even though coronary perfusion pressure is higher, the initially increased regional oxygen demands in the ischemic zone during exercise (due to increased heart rate, BP, and regional myocardial contractility) shift the curve relating coronary flow to perfusion pressure upward,3 causing the lower limit of autoregulation to occur at a higher pressure.3T4 Moreover, tachycardia per se limits subendocardial perfusion,’ so that the inherent coronary vasodilator reserve in the subendocardium of the ischemic bed becomes quickly exhausted. Progmss
in Cardiovascular
Diseases,
Vol XXXI,
No 6 (May/June).
Inability to measure transmural coronary perfusion in patients with coronary heart disease has led to the development in our laboratory of a conscious canine model of single vessel coronary artery stenosis, in which regional myocardial blood flow quantifies the degree of regional ischemia and regional contraction provides a measure of its effects on the myocardium. There is no evidence that these dogs experience pain during exercise, so they may in fact represent a model of “silent ischemia.” THE EXPERIMENTAL
MODEL
The model of chronic coronary stenosis allows treadmill exercise to be performed reproducibly while regional myocardial function and regional perfusion are measured.6*7 The observed responses appear to resemble those of exerciseinduced ischemia in patients (with or without angina pectoris) in that severe regional contractile dysfunction rapidly develops, accompanied by ischemic changes on the regional electrogram and a subnormal BP response to exercise (Fig 1).6-8 Delayed recovery of regional contraction “stunning” also occurs after exercise (Fig 1),6-9 as may be observed clinically in postexercise studies on global and regional left ventricular wall motion.” In our experimental model, measurements of transmural myocardial blood flow distribution using tracer microspheres have permitted observations on the relations between regional blood flow and regional contraction which provide some insights into mechanisms responsible for the ischemia and how various From the Division of Cardiology, Department of Medicine, University of California San Diego School of Medicine, Lo Jolla. Supported by a Specialized Center of Research on Ischemic Heart Disease, HL-17682. awarded by the National Heart, Lung, and Blood Institute, National Institutes of Health, and a Chair in Cardiovascular Research awarded by the San Diego County Afiliate of the American Heart Association. Address reprint requests to John Ross. Jr, MD. Department of Medicine (M-013). University of California, San Diego, School of Medicine, La Jolla, CA 92093. 0 I989 by W.B. Saunders Company. 0033-0620/89/3106-0005%5.00/0 1969:
pp 455-466
455
JOHNROSS,JR
456
CONTROLRUN
I STANDING
LEFT
' 1 MIN
200
VENTRICULAR PRESSURE (mmH!)
dP/dt (mmH!/sed
DURING RUN (9.6 km/hr, 5% GRADE) MICROSPHE;ESlNJECTEO 10 SEC
0
+400iE
-4000
1I
18 POSTERIOR (ISCHEMIC) WALL THICKNESS 10 [[y* (mm) I8 ANTERIOR (CONTROij WALL THICKNESS (mm) 10 100 CORONARY BLOOO FLOW VELOCITY phasic (cm/w) 0
/ :
100 mean (cm/w)
‘#
*
/
0 SUBENOClCARlJ UNIPOLAR EKG (mV) ic (POSTERIOR WALL] HEART RATE (beats/mm)
+I 0
/--
VSTART RUN
J)’ k-k
kG
pharmacologic agents used to treat angina pectoris may act to ameliorate it. In this animal preparation, chronic coronary stenosis is produced by an ameroid constrictor positioned on the left circumflex coronary artery which, over the course of two to three weeks, narrows or in some instances occludes the coronary artery. However, the closure is sufficiently gradual so that coronary collateral channels develop, infarction does not occur, and a period of time exists (several days) during which antegrade plus collateral flow at rest are adequate to maintain the resting metabolic needs and normal regional myocardial function; however, during rapid atria1 pacing” or with treadmill exercise 5*6~*there is inadequate coronary blood flow to meet the increased metabolic demands, and ischemia with a severe regional wall motion abnormality develops (Fig 1). Such animals often exhibit critical coronary stenosis in which the mean coronary blood flow exhibits lack of vasodilator reserve (Fig l), and during exercise subendocardial myocardial blood flow falls whereas subepicardial flow rises slightly above the resting level.12 These changes in transmural
VEND RUN
Fig 1. Responses to treadmill exercise in a repreaentative dog with a chronically implanted ameroid constrictor. The initiil data are obtained when the animal is standing on the treadmill. The run is commencd. end at ten seconds a tracing at fast paper speed is shown documenting initiil augmentation of systolic thiikening of the posterior wall [in the ischemic zone), with another tracing during steady-state exercise at the time of determination of myocardiil Mood flow I-~)whm posteriorwanthlleningkImarkedlyreduced.Thefindtreckrgis obtained one minute after exercisewhenre#xmlfunctimisstill depressed below the control vakle.corCorarclrybloodflowvdocilyiSrlNBaWredWithaDopplar flowprobeandfellatorise~ ciably during exercise to the stwoticartery(pha&and~l. STsegmenteiewtkmduringthe exerciseiaevklentonthesubuw dwardll ektqpam obtained fromthel2demiiregion.Hefm rateisshownbythetadwmeter tracing.
blood flow distribution when normalized as a percentage of the simultaneous changes in blood flow in the control, normal area in a representative set of experiments are shown in Fig 2 (stippled bars). This maldistribution of flow marks the development of ischemia in the region subserved by the stenosed artery, while in the normal anterior wall transmural blood flow increases severalfold and myocardial contraction is enhanced6*12 as measured by systolic wall thickening using sonomicrometers.6~7 In this model it has been shown that two identical bouts of treadmill exercise cause the same hemodynamic responses and degree of regional ischemia when carried out three hours apart on the same day, with complete recovery between the two running periods.6 Therefore, the effects of an antianginal drug can be tested during the second run after a control exercise bout. REGIONAL 0, SUPPLY AND DEMAND DURING EXERCISE-INDUCED ISCHEMIA
In considering the effects of exercise-induced myocardial ischemia and the influence of an-
ANTIANGINAL
DRUGS
& EXERCISE-INDUCED
ISCHEMIA
LPI
CR AR
cI -CAAR-~ El
C CR AR ATEMOLOL RUN
Fig 2. Myocardiil blood Row in an ischemic control run (CR) and a run after atenold (AR) in dogs with a chronic coronary artery stenosis. Blood flow determinations in the subendacardisl (ENDO), midmyocerdium (MID) and subepicardii regions IEPIL as well as transmural myocardial blood flow (TMBF) are presented here in normalized form, expressed as a decimal fraction of the simultaneously measured blood flow in the normal anterior wall (control area). Control IC) blood flow determimtions are token while the animal is standing on the treadmill. Data reported previously by Matsuzaki et al.”
tianginal drugs, knowledge of myocardial 0, supply and demand relations would be desirable. Under controlled experimental conditions, overall 0, supply can be expressed in terms of myocardial 0, delivery, and 0, demand is reflected by the myocardial 0, consumption. Various indices of global left ventricular (LV) 0, demand have been used, such as the tension-time index, the heart rate-BP product, and the systolic pressure-time index; the latter has been used in a ratio together with an index of 0, supply (coronary blood flow) estimated from the diastolic pressure-time index, to estimate 0, supplydemand relations in the subendocardium of the left ventricle (LV) under conditions of maximum coronary vasodilation.’ 3 However, such global indices have limited applicability to local events, under conditions of regional ischemia. Other types of supply and demand indices have related measures of regional work to regional coronary blood flow.t4 In humans, or in the exercising animal, techniques are not available for directly measuring regional myocardial 0, consumption transmurally in different regions of the heart. Therefore, this discussion will emphasize regional myocardial blood flow as an index of 0, supply and regional systolic wall thickening as a marker of
457
0, utilization, while recognizing their limitations for this purpose. In this connection, it should be noted that in the conscious dog, when 0, supply is progressively limited by means of a hydraulic cuff to produce increasing levels of acute ischemia, subendocardial myocardial blood flow is the major determinant of regional wall contraction in the steady state.15 The relation between this flow and regional systolic wall thickening is nearly linear over a wide range of flow reductions in the conscious dog.16 The relation between the average transmural flow and function is also good but less close, and subepicardial flow correlates poorly. “*M This suggests that 0, supply and demand can remain in balance during acute periods of graded ischemia, provided there is some residual coronary blood flow. The Transient Period During Onset of Exercise
As treadmill exercise is begun in the chronic animal model of single vessel coronary artery disease, there is a sharp increase in the heart rate, systemic arterial pressure, and maximum (+) LV dP/dt, as well as in regional systolic wall thickening (expressed as a percent of enddiastolic wall thickness, or %WTh) (Fig 1). In the zone supplied by the stenosed artery, however, following this initial enhancement regional %WTh quickly falls to a level well below the normal resting value, accompanied by some reduction (but continued elevation above the resting value) of LV systolic pressure and LV dP/dt (Fig 1). Thus, during this transient period a condition of relative ischemia (ie, supply and demand imbalance) exists in the circumflex artery distribution, in which the rapidly changing hemodynamic and cardiac responses promptly augment the regional 0, demands, whereas the marked heart rate increase in the presence of a limited coronary vascular reserve rapidly impairs subendocardial blood flow. A hypoperfused subendocardium results from an increased number of systolic periods per minute, during each of which blood flow to the subendocardium is very low because of the increased extravascular tissue pressure caused by LV contraction.5 This effect of tachycardia causes a reduction in the time available for diastolic subendocardial perfusion per minute and per beat, a situation which reflects the previously described inverse relation between heart rate and subendocardial myocar-
458
dial blood flow in the presence of maximum coronary vasodilation (produced by adenosine infusion) in the absence of coronary stenosis.’ Regional Ischemia During Steady-State Exercise
After about one minute of strenuous exercise on the treadmill, hemodynamic conditions stabilize and the severely depressed regional wall contraction in the ischemic zone reaches a steady state (Fig l), accompanied by reduced subendocardial and midwall blood flows (compared to the resting values) and a higher level of subepicardial blood flow than at rest (but well below that in the normal zone during exercise, Fig 2). At that time, it may be postulated that the reduced regional contraction is matched to the reduced coronary blood flow and 0, supply. Determinants of 0, Supply and Demand During Steady-State Exercise
In other experiments, when exercise was carried out in the presence of different degrees of acute coronary stenosis to produce various steadystate levels of regional myocardial dysfunction, a close correlation relation was observed between subendocardial blood flow and contractile function.17 It is of interest that if the level of subendocardial myocardial blood flow in the ischemic region during such steady-state periods of exercise associated with varying degrees of coronary stenosis is expressed per beat and plotted against regional myocardial function (expressed as normalized %WTh), the relation lies very close to that observed under resting conditions (Fig 3). This observation is pertinent to the proposed concept ofperfusion-contraction matching between subendocardial myocardial blood flow and regional function during various degrees of acute ischemia. It also suggests that under the ischemic conditions of exercise, the heart rate is a key determinant of the amount of blood flow (and hence 0,) available for each active contraction which, in turn, will largely set the level of regional contractile dysfunction in the steady state. This notion derives from the fact that although the other determinants of myocardial Oz consumption including the arterial pressure and global myocardial contractility (expressed as LV dP/dt and velocity of regional contraction in
JOHN
ROSS,
JR
nonischemic areas) differ widely between the resting and exercise states, nevertheless the relationship between subendocardial myocardial blood flow per beat and regional contraction per beat in the ischemic zone is similar under these two conditions over a wide range of subendocardial blood flows (Fig 3). It may be concluded that decreased oxygen supply is a major factor associated with ischemia of the oxygen demand type produced by exercise. This decreased Oz supply is due primarily to the heart rate effect (tachycardia), but also may be enhanced by increased a-adrenergic coronary vasoconstrictor tone, as discussed subsequently, and elevated LV end-diastolic pressure during exercise may further contribute to impaired subendocardial perfusion. During the steady state of acute ischemia due to exercise, regional oxygen supply and demand per beat appear to be matched, with decreased contraction corresponding to the level of decreased coronary blood flow per beat. APPROACHES
TO ANTIANGINAL
THERAPY
Various forms of drug treatment for angina pectoris traditionally have been categorized by whether their predominant effect is through altering cardiovascular hemodynamics to reduce myocardial Oz demand, or by increasing 0, supply through augmenting myocardial blood flow. Modification of Regional 0, Demand
Based on the considerations outlined earlier, among the various determinants of ventricular performance that might be manipulated to reduce exercise-induced ischemia and improve regional contraction, slowing of the heart rate would seem likely to be of prime importance. Thus, lowering the aortic pressure alone would tend to decrease left ventricular afterload and 0, demands, but diminishing aortic pressure during exercise could also reduce distal coronary perfusion pressure and might impair flow in the bed beyond a coronary stenosis. A beneficial effect of reducing 0, demand in the ischemic zone by this approach, while not yet established, might occur with vasodilator drugs such as nitroglycerin which reduce the arterial pressure, LV size, and hence wall stress,” also diminishing afterload in the
ANTJANGlNAL
DRUGS
& EXERCISE-INDUCED
ISCHEMIA
459
All P0ltVT.S
1.5
a l l
5; 2? z
: .
.
l .
k--A,
l
AA
A
I
e
0 0
8
tA
0 EXERC/SE A AT ItEST
5 z00
0 A
.005
I
I
.OlO
.015
z l s -0.3 .g 1.5 - I CONtRULAM? I STEMOS~S cc POINTPERDOG . II -
1
.020
0
l
Fig 3. Relation between subendocardisl blood flow per beat and normalized systolic wall thickening (expressed as a decimal fraction of the wall thickening [WTH] at rest) in conscious dogs studied at rest during progrearive coronary stenosis, and during treadmill exercise with varying degrees of coronary stenosis produced by a hydraulic cuff. Under these conditions, the rest and exercise-flow function relations are superimposable. Triangles with bars indicate resting blood flows with standard deviations. Modiied and reproduced with permission.”
l 0
l
0
A /’ l
# 4 EXE/K/SE ---b AT REST
0'
c/)
I
I
.OlO
.015
1
-s
-0.3
.005 1
.
.020
SUBENDOCAROlALBLOOO FLOW PER BEAT (ml/min/g)
ischemic region, although other effects of nitroglycerin on the coronary circulation are likely to be of more importance, as discussed subsequently. Slowing of the heart rate can produce large changes in 0, demand per minute in the ischemic zone as well as permit a small increase in absolute subendocardial myocardial blood flow per minute by mechanisms discussed earlier. In addition, it will substantially increase the regional myocardial blood flow (and 0, availability) per beat. This effect, in turn, should lead to augmented regional contraction in the ischemic
zone, and consequently increased myocardial 0, consumption per beat. ,&Adrenergic blockade. Improvement of regional contraction in the ischemic zone during exercise after intravenous IV propranolol compared to control exercise without drug was demonstrated in a chronic coronary stenosis canine model.” Similarly, in patients with coronary heart disease and exercise-induced ischemia, improvement of the exercise ejection fraction after ,8-adrenergic blockade has been demonstrated.*’ More recently, the effects of P-blockade on both regional function and regional coronary blood
JOHN
460
flow distribution were examined after oral administration of atenolol in our animal model.7 Atenolol decreased the average exercise heart rate from 228 to 181 beats/min, the increase in systolic arterial pressure was less, and the increase in LV dP/dt was markedly blunted compared to control running. In the ischemic zone, normalized regional myocardial blood flow to the subendocardium increased (Fig 2). Absolute subendocardial flow increased slightly compared to control exercise, whereas the subepicardial blood flow was lower and the usual increase in transmural blood flow in normal zones was markedly attenuated. Despite reduction of the normal enhancement of contraction during exercise in nonischemic regions, regional wall dysfunction in the ischemic zone improved (%WTh from 4.0 to 8.5) after atenolol. The changes of regional blood flow per minute in the ischemic zone were associated with an even greater improvement of subendocardial blood flow per beat (average .0019 to .0030 mL/min/g, a 58% increase). In addition, atenolol caused partial amelioration of the transmural maldistribution of regional blood flow in the ischemic zone, the endocardial/epicardial flow ratio increasing significantly from 0.27 to 0.47 .7 These experimental findings implicate several mechanisms in the beneficial effects of pblockade. There is evidence that with exercise in the presence of critical coronary stenosis transmural steal of flow from subendocardial to subepicardial layers may occur,‘* and other findings suggest that even with severe dyskinesia of the wall during ischemia the outer myocardial fibers are actively contracting. 21 Therefore, it may be postulated that /?-blockade, by reducing the heart rate, arterial pressure, and myocardial contractility during exercise, reduces the 0, consumption of the outer wall layers in the ischemic region, reducing blood flow requirements to the epicardial zone which, in turn, should lead to increased coronary vascular resistance in that region, improve the distal coronary perfusion pressure, and lead to a reduction of transmural steal and the subendocardial blood flow deficit (Table 1). In support of this view are experiments at rest in the conscious dog with acute coronary stenosis, in which it was shown that propranonol improves regional contractile function in the ischemic
Table
1. Antianginal Ischemic
Mechanisms
Region
During
1. 1 Heart
rate
lo,
1 Arterial pressure 2. 1 Heart fate 3. 1 Heart rate 1 Arterial pressure 1 Contractility Abbreviations:
of &Blockade Steady
Cons,
JR
on
Exercise
Mechanismld
ACtiOtllS~
ROSS.
Effectls)
Con%“,
1 O,Avail/beat ontraction
r Diastolic Pet-f Time/min
ILIF
1 0, con%, 1 Transmural
1 MBF,,, t MBF,
7 t MBTLbeat
steal
1 endolepi consumption;
Perf,
perfusion;
Avail,
available.
zone.22 Studies in open-chest dogs with a noncircumferential coronary stenosis have confirmed this finding and further demonstrated improved subendocardial how and decreased subepicardial blood flow in the ischemic region after propranonol, without a change in transmural coronary blood flo~.‘~ The improvement in regional myocardial function was closely correlated with the distal diastolic coronary perfusion pressure per minute (DDPTI), and increased distal coronary perfusion pressure after propranonol was proposed to be due primarily to reduction of MVO, with restoration of autoregulation in the ischemic area, primarily in the subepicardium. This response also caused a reduction in the calculated stenosis resistance by increasing the distending pressure in the lesion.23-25 (The latter mechanism seems unlikely in our experiments with the chronic ameroid constrictor, which is circumferential. Moreover, in some experiments the artery was totally included). Gross et al later showed that prior drug-induced dilation of the coronary bed prevented the decrease in subepicardial blood flow and improvement in subendocardial blood flow by propranolol, although regional function was not measured.26 Changes in collateral blood flow seem unlikely to have played any role in those experiments, since the diastolic perfusion pressure difference from aorta to the distal coronary bed, a primary determinant of collateral blood flo~,‘~ was reduced.23 The mechanism involved during exercise in the ameroid constrictor model seems most likely to be that of vasoconstriction in the subepicardial region, since the aortic pressure was lower during exercise after atenolol than during control exercise, making increased collateral flow unlikely, and the increased subendocardial/subepicardial
ANTIANGINAL
DRUGS
& EXERCISE-INDUCED
ISCHEMIA
flow ratio was not accompanied by any change in mean transmural flow to the ischemic region after /3-blockade.7 As suggested earlier, the increased subendocardial blood flow per minute and per beat in the ischemic zone may also relate, in part, to a decreased number of systolic contractions per minute with more time for diastolic coronary perfusion,’ although this mechanism remains to be clearly established in the presence of ischemia. Thus, how much extravascular compression actually occurs in the subendocardium with systole during severe ischemia and contributes to the coronary vascular resistance in this setting is not known. In addition to the increased subendocardial blood flow per beat after ,&blockade, the increased time available for recovery of metabolism per beat and over several beats due to the diminished number of contractions per minute undoubtedly contributed to the increased contraction in the ischemic zone (Table 1). The singular importance of heart rate slowing to the effect of P-blockade on regional ischemia during exercise is evidenced by recent studies on exercise after B-blockade in which, by the use of atria1 pacing during exercise, the heart rate was returned to the high level which existed during control exercise before drug administration.28 Under pacing conditions, regional myocardial blood flow actually diminished slightly below the value observed in the control exercise period, and regional myocardial function failed to improve. Thus, elimination by pacing of the slowing of heart rate produced by atenolol completely counteracted any beneficial effect of P-blockade on the ischemic zone.28 It may be concluded from these observations that the conventional view of the mechanism of action of @-blockers, ie, relief of ischemia and angina pectoris by reducing myocardial 0, demands during exercise, should be modified to indicate that they also increase 0, supply to the subendocardium of the ischemic zone (Table 1). In fact, 0, consumption per beat in the ischemic zone must actually increase as regional contraction improves. Bradycardic drugs. In view of these observations on the importance of heart rate, the action of a new bradycardic agent, UL FS-49 (Thomae Pharmaceutical, Biberach, FRG, made available
461
by Dr Jurgen Dammgen) was tested in the chronic coronary stenosis model.i4 A considerably greater slowing of heart rate during exercise than that caused by P-blockade could be achieved with this drug (exercise heart rate 139 v 181 beats/min with atenolol) without any depressant effect on arterial pressure, or the normal increase of LV dP/dt and regional function in nonischemic zones. A marked improvement of regional myocardial function in the ischemic zone to a near normal level was observed after UL FS-49, compared to control exercise without drug (%WTh 9.3 to 21.5%). Regional subendocardial and mid-wall blood flow per minute improved slightly, while subendocardial flow per beat in the ischemic zone increased twofold.‘4 These observations further confirm the important role of heart rate in the regional ischemic response to exercise. Modification
of Regional 0, Supply
In addition to indirect effects on myocardial blood flow due to P-blockade and heart rate slowing, discussed above, the direct action of coronary vasodilator drugs can improve myocardial perfusion and 0, supply by actions on the large coronary arteries, the collateral circulation, and the small coronary resistance vessels. In this connection, there is increasing experimental evidence that additional vasodilator reserve can be recruited by pharmacologic agents, even in the presence of severe ischemia.29-32 Dilation of large coronary arteries. Vasodilation occurs in the epicardial coronary arteries after administration of agents such as nitroglycerin, calcium channel blocking drugs, or a-adrenergic blocking agents.33 There is also evidence in patients with coronary artery disease that such drug-induced increases in the diameter of large coronary arteries at the site of noncircumferential atherosclerotic lesions can reduce the degree of coronary artery stenosis, whereas aradrenergic agents cause vasoconstriction.34 Nonselective cy-adrenergic blockade has been shown to diminish stress-induced (ice water immersion) regional myocardial ischemia in patients with coronary heart disease, perhaps by this mechanism and/or by reducing a-adrenergic tone to coronary resistance vessels.35It should be pointed out that such effects of vasodilator drugs on the
JOHN
462
JR
/
caliber of the circumflex coronary artery stenosis should not be a factor in our chronic animal model, since fixed annular narrowing is produced by the ameroid constrictor. There is increasing experimental evidence that other mechanisms of vasodilation, including that due to endothelial derived relaxing factor, flowmediated dilation, and dilation of small vessels that are larger than the traditional arteriolar resistance vessels also are important in the regulation of coronary blood flo~.~~
100
Atenolol 0 Diltiazem A Atenolol and Dittiazem
‘-9iw /
l
I
80
Rest
, /
80 40 20
Dilation of coronary collateral and/or resistance vessels. In the chronic canine model,
increased regional myocardial function in the ischemic zone during severe exercise-induced ischemia at the same or reduced arterial pressure has been demonstrated after the calcium channel blocking drugs diltiazem, nifedipine, and verapamil, indicating that vasodilation with lowered vascular resistance occurred in the native coronary bed and/or coronary collateral channels.6*32,36 Diltiazem and nifedipine were shown to change conditions in the ischemic zone by increasing subendocardial and transmural blood flow in the ischemic and normal zones, and improving the operating position of the ischemic zone on the subendocardial myocardial blood flow-regional contraction relation (Fig 4) (regional blood flow was not measured in experiments with verapamil). With diltiazem, we observed a small reduction in arterial pressure and a modest reduction in heart rate (from 227 to 204 beats/min) during exercise, so that part of its effect was mediated by heart rate mechanisms discussed above (Table 2). However, its action to improve contraction and blood flow (subendocardial flow per beat increased 54%) were equal to those produced by atenolol (Fig 4), which caused a much larger reduction in exercise heart rate. Therefore, an additional vasodilator action of diltiazem was implicated (Table 2).6 In contrast to P-blockade,’ diltiazem caused no depression of contraction in normal regions during exercise.6 Nifedipine was studied at a low dosage sufficient to affect the coronary circulation without appreciable systemic hemodynamic actions.32 During exercise after nifedipine in the chronic coronary stenosis model, the systolic arterial pressure and heart rate were unchanged from control exercise. Nevertheless, a significant in-
ROSS,
0
0
20
40
60
80
100
SUBENDOCAROIALRMBF (post. as % ant.) Fig 4. Relations between subendocardial regional myocardial blood flow (RMBF) in the posterior ipost) irchemic wall (expressed as a percentage of that in the control anterior [ant] walll and normalized regional systolic wall thickening in the ischemic zone (expressed as a percentage of the resting value [%WT]). The reductions in flow and function during control running before drug administration are similar for atenolol, diltiazem. and the combination of these drugs. Significant improvement in both regional flow and regional function during exercise is evident with either drug alone, but when both drugs were given (open triangles) substantial further improvement during running is evident. Reproduced with permirsion.”
crease in subendocardial and transmural regional myocardial blood flow occurred, together with increased contraction in the ischemic re gion, suggesting that coronary vasodilator re serve was recruited. In contrast to P-blockade, nifedipine and diltiazem produced a lower LV end-diastolic pressure during exercise than control running without drug, which should favor subendocardial perfusion in the ischemic zone.32 Table
2.
Antianginal
u-Blockade
Mechanisms:
on Ischemic
Actions
Vasodilators
Region
During
Mechanism(s)
Effect
1.
Relax coronary smooth muscle
1 Vascular resistance of arterioles
2.
1 LVED
pressure
1 Heart
size
1 Subendocardial tissue pres-
and/or
3.
1 Arterial pressure 1 Heart size (1 Heart
and
Exercise
f MBF,-,,
collaterials r MBF,,,,
sure 1 0, Cons/minute
1 0, Avail/beat
rate, variable)
Abbreviations: LVED, footnotes to Table 1.
left ventricular
end
diastolic:
also
see
ANTIANGINAL
DRUGS
& EXERCISE-INDUCED
ISCHEMIA
Sizeable effects on regional myocardial contraction due to isosorbide dinitrate were also described in the canine model, although regional blood flow was not measured.18 This effect occurred at a lower systemic arterial pressure and smaller heart size, suggesting hemodynamic as well as coronary vasodilator effects, and it was suggested that the lowered LV end-diastolic pressure during exercise compared to control exercise might have contributed to improvement in subendocardial blood flow (Table 2). In addition, as discussed earlier, the decreased arterial pressure and heart size with nitroglycerin would be expected to reduce systolic wall stress, which should diminish 0, demand and also might contribute to enhancement of systolic wall thickening by reducing afterload in the ischemic zone.” In the chronic coronary stenosis model, it is not yet possible to differentiate clearly between an action of these vasodilator agents on vascular resistance in the poststenotic native coronary bed and that in the coronary collateral channels, even when the coronary artery is entirely occluded by the ameroid constrictor. Effects of the various agents tested do not appear different whether the coronary artery is partially open with poorly developed collaterals, or closed with a more fully developed collateral circulation. It will be necessary to develop a reliable method for measuring coronary perfusion pressure distal to the coronary stenosis during exercise before this question can be approached. As recently n-Adrenergic mechanisms. viewed by Feigl, the coronary vasculature of the dog contains both oi and aZ postjunctional receptors that induce vasoconstriction.37 Experiments in normal dog hearts not subjected to ischemia have indicated that increased a-adrenergic tone to the coronary vessels during exercise can reduce coronary blood IIow.~‘-~~ Investigations in anesthetized dogs have also shown that (Yadrenergic receptor-mediated coronary vasoconstriction can be produced by direct sympathetic nerve stimulation, even during severe ischemia.40 However, based on other experiments during milder ischemia using norepinephrine stimulation and a different blocking drug, Feigl has proposed that increased cr-adrenergic tone af-
463
fecting the subepicardial region may serve to maintain subendocardial perfusion.37 Recent studies in a conscious canine mode1 in which an cY,-adrenergic blocking drug (idazoxan) was infused into an acutely stenosed coronary artery during exercise, showed improvement in regional transmural and subendocardial blood flow as well as increased regional wall motion in the ischemic zone, suggesting that increased cY-adrenergic tone to the native coronary resistance vessels can limit subendocardial perfusion even during severe exercised-induced ischemia.41 Such studies are carried out after fi-adrenergic blockade to prevent direct and indirect vasodilator effects due to increased norepinephrine release consequent to blockade of the prejunctional a2 receptors.41 Some studies indicate that coronary collateral vessels in dogs do not contain a,-adrenergic receptors,42 whereas others suggest that they do43; whether or not a,-adrenergic blockade during exercise improved vasodilation in the native coronary resistance vessels or in coronary collaterals cannot be stated with certainty. There is also experimental evidence that nifedipine can prevent sympathetic coronary vasoconstriction in the vascular bed perfused by an artery with severe stenosis. 44 The effects in the exercising dog with regional ischemia of (pi blockade, and also the effects of IV LYEor a2 blockade in the presence and absence of P-blockade, remain to be investigated. It is of interest that investigations in patients with coronary stenosis have demonstrated exercise-induced vasoconstriction at the stenosis site, which was relieved by nitroglycerin4’ Also, a nonselective a-adrenergic blocking drug (phentolamine) given by the intracoronary route was shown to reduce evidence of ischemia during supine exercise in patients with coronary heart disease.46 It may be concluded that vasodilators can act through multiple mechanisms, including dilation of large coronary arteries (Table 2). They may also dilate native coronary resistance vessels, perhaps in part by relief of cY-adrenergic constrictor tone, and they may also dilate coronary collateral vessels, although the roles of these different vasodilator components cannot cur-
JOHN
464
rently be quantified in the chronic animal model of exercise-induced ischemia. Combined Antianginal
Therapy
It might be expected that combinations of antianginal drugs having different mechanisms of action, such as a p-blocking drug and a coronary vasodilator, would have additive effects, and beneficial effects of such combinations have been described in clinical studies.47 In the chronic coronary stenosis animal model, the combination of atenolol plus diltiazem was administered before exercise and compared to control exercise, as well as to experiments in other animals in which either diltiazem or atenolol was administered alone. The effect of the combined drug treatment was additive, both subendocardial myocardial blood flow and regional systolic wall thickening showing an improvement of approximately twofold over that observed with either agent alone (Fig 4).48 Also, this drug combination plus isosorbide dinitrate was shown to prevent regional dysfunction during milder ischemia.4g SUMMARY
Several mechanisms involved in the production of regional exercise-induced ischemia are described. Each offers the potential for modification using different types of antianginal drugs operating to alter regional 0, demands, improve regional perfusion, or both, leading to reduced ischemia and increased contractile function in the ischemic zone.
R066.J~
Evidence is presented for matching of regional subendocardial myocardial blood flow and flow per beat with regional myocardial contraction at various levels of ischemia at rest, during steadystate exercise, and after antianginal drugs, signifying a particularly important role for heart rate control. In addition to reducing myocardial 0, demand per minute, B-blockers and bradycardic drugs cause improvement of absolute subendocardial blood flow and particularly flow per beat by producing vasoconstriction in the epicardial region of the ischemic zone, with improvement of transmural blood flow distribution. Vasodilator drugs can act at several locations to increase regional blood flow and also to decrease 0, demands. A recruitable vasodilator reserve has been shown to exist during exerciseinduced ischemia either in native resistence vessels, collateral channels, or both, which appears to be due at least in part to reduction of increased a-adrenergic constrictor tone to the coronary vessels during exercise, even in the presence of severe ischemia. The potential for additive effects using combinations of bradycardic and vasodilating agents are described within a framework relating regional subendocardial blood flow to regional systolic contraction. The experimental findings described suggest some potential new directions for antianginal therapy and, along with recent clinical observations, support the use of combinations of antianginal agents that act by different mechanisms.
REFERENCES 1. Tomoike H, Franklin D, McKown D, et al: Regional 7. Matsuzaki M, Patritti J, Tajimi T, et al: Effects of fl myccardial dysfunction and hemodynamic abnormalities durblockade on regional myocardial flow and function during ing strenuous exercise in dogs with limited coronary flow. exercise. Am J Physioll6:H52-H60, 1984 Circ Res 42:487-496,1978 8. Grover-McKay M, Matsuzaki M, Ross J Jr: Dissocia2. Raw J Jr: Perspective: Assessment of ischemic regional tion between regional myocardial dysfunction and subendocarmyocardial dysfunction and its reversibility. Circulation dial ST segment elevation during and after exercise-induced 74:1186-1190.1986 ischemia in dogs. J Am Co11Cardiol 10:1105-l 112.1987 3. Mosher P, Ross J Jr, McFate PA, et al: Control of 9. Homans DC, Sublett E, Dai X-Z, et al: Persistence of coronary blood flow by an autoregulatory mechanism. Circ regional left ventricular dysfunction after exercise-induced Res l&250-259,1964 myocardial &hernia. J Clin Invest 77:66-73,1986 4. Hoffman JIE: Maximal coronary flow and the concept 10. Robertson WS, Feigenbaum H, Armstrong WF, et al: of coronary vascular reserve. Circulation 70: 153- 159,1984 Exercise echocardiography: A clinically practical addition in 5. Bathe RJ, Cobb FR: Effect of maximal coronary the evaluation of coronary artery disease. J Am Cot1 Cardiol vasodilation on transmural perfusion during tachycardia in 2:1085-1091,1983 the awake dog. Circ Res 41:648-653,1977 11. Kumada T, Gallagher KP, Battler A, et al: Compari6. Matsuzaki M, Guth B, Tajimi T, et al: Effect of the son of post-pacing and exercise-induced myocardial dysfunction during collateral development in conscious dogs. Circulacombination of diltiazem and atenolol on exercise-induced tion 65:1178-1185, I982 regional myocardial ischemia in conscious dogs. Circulation 72:233-243.1985 12. Gallagher KP, Osakada G, Matsuzaki M, et al:
ANTIANGINAL
DRUGS
& EXERCISE-INDUCED
ISCHEMIA
Myocardial blood flow and function with critical coronary stenosis in exercising dogs. Am J Physiol: Heart Circ Physiol 12:H698-H707.1982 13. Hoffman JIE, Buckberg GD: The myocardial supply: Demand ratiea critical review. Am J Cardiol 41327-332, 1978 14. Guth BD, Heusch G, Ross J Jr: Elimination of exercise-induced regional myocardial dysfunction by a bradycardiac agent in dogs with chronic coronary stenosis. Circulation 75:661-669, 1987 15. Gallagher KP, Kumada T, Koziol JA, et ai: Significance of regional wall thickening abnormalities relative to transmural myocardial perfusion in anesthetized dogs. Circulation 62:1266-1274, 1980 16. Gallagher KP, Matsuaaki M, Koziol JA, et al: Regional myocardial perfusion and wall thickening during ischemia in conscious dogs. Am J Physiol 16:H727-H738, 1984 17. Gallagher KP, Matsuxaki M, Osakada G, et al: Effect of exercise on the relationship between myocardial blood flow and systolic wall thickening in dogs with acute coronary stenosis. Circ Res 52:716-729, 1983 18. Kumada T, Gallagher KP, Miller M, et al: Improvement by isosorbide dinitrate of exercise-induced regional myocardial dysfunction in the dog. Am J Physiol 239: H399-H405,1980 19. Kumada T, Gallagher K, Shirato K, et al: Reduction of exercise-induced regional myocardial dysfunction by propranolol: Studies in a canine model of chronic coronary artery stenosis. Circ Res 46:190-200, 1980 20. Battler A, Ross J Jr, Slutsky R, et al: Improvement of exercise-induced left ventricular dysfunction with oral propranolol in patients with coronary heart disease. Am J Cardiol44:3 18-324, 1979 2 1. Gallagher KP, Osakada G, Hess OM, et al: Subepicardial segmental function during coronary stenosis and the role of myocardial fiber orientation. Circ Res 50~352-359, 1982 22. Tomoike H, Ross J Jr, Franklin D, et al: Improvement by propranolol of regional myocardial dysfunction and abnormal coronary flow pattern in conscious dogs with coronary narrowing. Am J Cardiol41:689-696, 1978 23. Buck JD, Hardman HF, Warltier DC, et al: Changes in ischemic blood flow distribution and dynamic severity of a coronary stenosis induced by fi blockade in the canine heart. Circulation 64:708-7 15, 198 1 24. Gould LK: Dynamic coronary stenosis. Am J Cardiol 45286-292, 1980 25. Santamore WP, Walinsky P: Altered coronary flow responses to vasoactive drugs in the presence of coronary arterial stenosis in the dog. Am J Cardiol45:276-285, 1980 26. Gross GJ, Buck JD, Warltier DC, et al: Role of autoregulation in the beneficial action of propranolol on ischemic blood flow distribution and stenosis severity in the canine myocardium. J Pharm Exp Ther 222:635-640, 1982 27. Brown BG, Gendel WD, Gott VL, et al: Coronary collateral blood flow following acute coronary occlusion: A diastolic phenomenon. Cardiovasc Res 8:621-631, 1974 28. Guth BD, Heusch G, Seitelberger R, Ross J Jr: Mechanism of beneficial effect of beta-adrenergic blockade on exercise-induced &hernia in conscious dogs. Circ Res 60:738-746. 1987
465 29. Canty JM, Klocke FJ: Reduced regional myocardial perfusion in the presence of pharmacologic vasodilator reserve. Circulation 71:370-377, 1985 30. Aversano T, Becker LC: Persistence of coronary vasodilator reserve despite functionally significant flow reduction. Am J Physiol248:H403-H411, 1985 31. Pantely GA, Bristow JD, Swenson LJ, et al: Incomplete coronary vasodilation during myocardial ischemia in swine. Am J Physiol249:H638-H647, 1985 32. Heusch G, Guth BS, Seitelberger R, Ross J Jr: Attenuation of exercise-induced myocardial ischemia in dogs with recruitment of coronary vasodilator reserve by nifedipine. Circulation 75:482-490, 1987 33. Young MA, Vatner SF: Regulation of large coronary arteries. Circ Res 59:579-596, 1986 34. Brown BG, Bolson EL, Dodge HT: Dynamic mechanisms in human coronary stenosis. Circulation 70:917-922, 1984 35. Feldman RL, Whittle JL, Marx JD, et al: Regional coronary hemodynamic responses to cold stimulation in patients without variant angina. Am J Cardiol 49:665-673, 1982 36. Osakada G, Kumada T, Gallagher KP, et al: Reduction of exercise-induced regional myocardial dysfunction by verapamil in conscious dogs. Am Heart J 101:707-712, 1981 37. Feigl EO: The paradox of adrenergic coronary vasoconstriction. Circulation 76:737-745, 1987 38. Murray PA, Vatner SF: a;adrenoceptor attenuation of the coronary vascular response to severe exercise in the conscious dog. Circ Res 45:654-660, 1979 39. Gwirtz PA, Gvern SP, Mass HJ, et al: cu,-adrenergic constriction limits coronary flow and cardiac function in running dogs. Am J Physiol 250:Heart Circ Physiol 19: H1117-H1126.1986 40. Heusch G, Deussen A: The effects of cardiac sympathetic nerve stimulation on the perfusion of stenotic coronary arteries in thedog. Circ Res 53:8-15, 1983 41. Seitelberger R, Guth BD, Heusch G, et al: Intracoronary Alpha,-adrenergic receptor blockage attenuates ischemia in conscious dogs during exercise. Circ Res 62: 435-442,1988 42. Harrison DG, Chilian WM, Marcus ML: Absence of functioning a-adrenergic receptors in mature canine coronary collaterals. Circ Res 59:133-142, 1986 43. Maruoka Y, McKirnan MD, Engler RL, et al: Functional significance of a-adrenergic receptors in mature coronary collateral circulation of dogs. Am J Physiol 253: H582-H590,1987 44. Heusch G, Deussen A: Nifedipine prevents sympathetic vasoconstriction distal to severe coronary stenosis. J Cardiovasc Pharmacol6:378-383, 1984 45. Gage JE, Hess OM, Murakami T, et al: Vasoconstriction of stenotic coronary arteries during dynamic exercise in patients with classic angina pectoris: Reversibility by nitroglycerin. Circulation 73:865-876, 1986 46. Berkenboom GM, Abramovicz M, Vandermoten, et al: Role of alpha-adrenergic coronary tone in exercisedinduced angina pectoris. Am J Cardio157: 195 198, 1986 47. Packer M, Meller J, Medina N, et al: Hemodynamic consequences of combined beta-adrenergic and slow calcium channel blockade in man. Circulation 65:660-668, 1982
466
48. Matsuxalci M, Guth B, Tajimi T, et al: Effect of the combination of diltiazem and atenolol on exercise-induced regional myocardial &hernia in conscious dogs. Circulation 72:233-243,1985
JOHN ROSS, JR
49. Guth BD, Tajimi T, Seitelberger R, et al: Experimental exercise-induced ischemia: Drug therapy can eliminate regional dysfunction and oxygen supply-demand imbalance. J Am Co11Cardiol7:1036-1046, 1986
of Regional Ischemia and Antianginal During Exercise
Drug Action
John Ross, Jr
A
NGINA PECTORIS REPRESENTS a disorder of oxygen supply and demand and, of course, the most common type of angina pectoris is a consequence of regional ischemia produced by the increased oxygen demands of exercise or other stress. In addition, it is now clear that in some patients a coronary vasospastic component can contribute to ischemia by causing decreased oxygen supply during stress, or at rest. Ischemia unaccompanied by angina pectoris, so-called silent ischemia, also occurs frequently in patients with coronary heart disease. Many effective antianginal drugs having widely different properties are now available, in particular the long-acting nitrates, P-adrenergic and calcium channel blocking agents. In the clinical setting, understanding of the physiologic mechanisms involved in exercise-induced ischemia and the actions of these drugs has been limited by the lack of methods for accurately quantifying in man the degree of ischemia during exercise and its effect on the myocardium. The regional nature of myocardial ischemia, such as that produced by exercise in the presence of coronary artery narrowing, has certain unique features which contrast with the effects of global ischemia. lV2Thus, the aortic BP usually increases during exercise (due to augmented cardiac output and enhanced inotropic state of normally perfused regions of the left ventricle), thereby enhancing coronary perfusion pressure, whereas contractile performance in the ischemic region deteriorates’ and is affected in a complex way by the reduced contractility, altered loading conditions, and interactions with the remainder of the ventricle.2 Moreover, even though coronary perfusion pressure is higher, the initially increased regional oxygen demands in the ischemic zone during exercise (due to increased heart rate, BP, and regional myocardial contractility) shift the curve relating coronary flow to perfusion pressure upward,3 causing the lower limit of autoregulation to occur at a higher pressure.3T4 Moreover, tachycardia per se limits subendocardial perfusion,’ so that the inherent coronary vasodilator reserve in the subendocardium of the ischemic bed becomes quickly exhausted. Progmss
in Cardiovascular
Diseases,
Vol XXXI,
No 6 (May/June).
Inability to measure transmural coronary perfusion in patients with coronary heart disease has led to the development in our laboratory of a conscious canine model of single vessel coronary artery stenosis, in which regional myocardial blood flow quantifies the degree of regional ischemia and regional contraction provides a measure of its effects on the myocardium. There is no evidence that these dogs experience pain during exercise, so they may in fact represent a model of “silent ischemia.” THE EXPERIMENTAL
MODEL
The model of chronic coronary stenosis allows treadmill exercise to be performed reproducibly while regional myocardial function and regional perfusion are measured.6*7 The observed responses appear to resemble those of exerciseinduced ischemia in patients (with or without angina pectoris) in that severe regional contractile dysfunction rapidly develops, accompanied by ischemic changes on the regional electrogram and a subnormal BP response to exercise (Fig 1).6-8 Delayed recovery of regional contraction “stunning” also occurs after exercise (Fig 1),6-9 as may be observed clinically in postexercise studies on global and regional left ventricular wall motion.” In our experimental model, measurements of transmural myocardial blood flow distribution using tracer microspheres have permitted observations on the relations between regional blood flow and regional contraction which provide some insights into mechanisms responsible for the ischemia and how various From the Division of Cardiology, Department of Medicine, University of California San Diego School of Medicine, Lo Jolla. Supported by a Specialized Center of Research on Ischemic Heart Disease, HL-17682. awarded by the National Heart, Lung, and Blood Institute, National Institutes of Health, and a Chair in Cardiovascular Research awarded by the San Diego County Afiliate of the American Heart Association. Address reprint requests to John Ross. Jr, MD. Department of Medicine (M-013). University of California, San Diego, School of Medicine, La Jolla, CA 92093. 0 I989 by W.B. Saunders Company. 0033-0620/89/3106-0005%5.00/0 1969:
pp 455-466
455
JOHNROSS,JR
456
CONTROLRUN
I STANDING
LEFT
' 1 MIN
200
VENTRICULAR PRESSURE (mmH!)
dP/dt (mmH!/sed
DURING RUN (9.6 km/hr, 5% GRADE) MICROSPHE;ESlNJECTEO 10 SEC
0
+400iE
-4000
1I
18 POSTERIOR (ISCHEMIC) WALL THICKNESS 10 [[y* (mm) I8 ANTERIOR (CONTROij WALL THICKNESS (mm) 10 100 CORONARY BLOOO FLOW VELOCITY phasic (cm/w) 0
/ :
100 mean (cm/w)
‘#
*
/
0 SUBENOClCARlJ UNIPOLAR EKG (mV) ic (POSTERIOR WALL] HEART RATE (beats/mm)
+I 0
/--
VSTART RUN
J)’ k-k
kG
pharmacologic agents used to treat angina pectoris may act to ameliorate it. In this animal preparation, chronic coronary stenosis is produced by an ameroid constrictor positioned on the left circumflex coronary artery which, over the course of two to three weeks, narrows or in some instances occludes the coronary artery. However, the closure is sufficiently gradual so that coronary collateral channels develop, infarction does not occur, and a period of time exists (several days) during which antegrade plus collateral flow at rest are adequate to maintain the resting metabolic needs and normal regional myocardial function; however, during rapid atria1 pacing” or with treadmill exercise 5*6~*there is inadequate coronary blood flow to meet the increased metabolic demands, and ischemia with a severe regional wall motion abnormality develops (Fig 1). Such animals often exhibit critical coronary stenosis in which the mean coronary blood flow exhibits lack of vasodilator reserve (Fig l), and during exercise subendocardial myocardial blood flow falls whereas subepicardial flow rises slightly above the resting level.12 These changes in transmural
VEND RUN
Fig 1. Responses to treadmill exercise in a repreaentative dog with a chronically implanted ameroid constrictor. The initiil data are obtained when the animal is standing on the treadmill. The run is commencd. end at ten seconds a tracing at fast paper speed is shown documenting initiil augmentation of systolic thiikening of the posterior wall [in the ischemic zone), with another tracing during steady-state exercise at the time of determination of myocardiil Mood flow I-~)whm posteriorwanthlleningkImarkedlyreduced.Thefindtreckrgis obtained one minute after exercisewhenre#xmlfunctimisstill depressed below the control vakle.corCorarclrybloodflowvdocilyiSrlNBaWredWithaDopplar flowprobeandfellatorise~ ciably during exercise to the stwoticartery(pha&and~l. STsegmenteiewtkmduringthe exerciseiaevklentonthesubuw dwardll ektqpam obtained fromthel2demiiregion.Hefm rateisshownbythetadwmeter tracing.
blood flow distribution when normalized as a percentage of the simultaneous changes in blood flow in the control, normal area in a representative set of experiments are shown in Fig 2 (stippled bars). This maldistribution of flow marks the development of ischemia in the region subserved by the stenosed artery, while in the normal anterior wall transmural blood flow increases severalfold and myocardial contraction is enhanced6*12 as measured by systolic wall thickening using sonomicrometers.6~7 In this model it has been shown that two identical bouts of treadmill exercise cause the same hemodynamic responses and degree of regional ischemia when carried out three hours apart on the same day, with complete recovery between the two running periods.6 Therefore, the effects of an antianginal drug can be tested during the second run after a control exercise bout. REGIONAL 0, SUPPLY AND DEMAND DURING EXERCISE-INDUCED ISCHEMIA
In considering the effects of exercise-induced myocardial ischemia and the influence of an-
ANTIANGINAL
DRUGS
& EXERCISE-INDUCED
ISCHEMIA
LPI
CR AR
cI -CAAR-~ El
C CR AR ATEMOLOL RUN
Fig 2. Myocardiil blood Row in an ischemic control run (CR) and a run after atenold (AR) in dogs with a chronic coronary artery stenosis. Blood flow determinations in the subendacardisl (ENDO), midmyocerdium (MID) and subepicardii regions IEPIL as well as transmural myocardial blood flow (TMBF) are presented here in normalized form, expressed as a decimal fraction of the simultaneously measured blood flow in the normal anterior wall (control area). Control IC) blood flow determimtions are token while the animal is standing on the treadmill. Data reported previously by Matsuzaki et al.”
tianginal drugs, knowledge of myocardial 0, supply and demand relations would be desirable. Under controlled experimental conditions, overall 0, supply can be expressed in terms of myocardial 0, delivery, and 0, demand is reflected by the myocardial 0, consumption. Various indices of global left ventricular (LV) 0, demand have been used, such as the tension-time index, the heart rate-BP product, and the systolic pressure-time index; the latter has been used in a ratio together with an index of 0, supply (coronary blood flow) estimated from the diastolic pressure-time index, to estimate 0, supplydemand relations in the subendocardium of the left ventricle (LV) under conditions of maximum coronary vasodilation.’ 3 However, such global indices have limited applicability to local events, under conditions of regional ischemia. Other types of supply and demand indices have related measures of regional work to regional coronary blood flow.t4 In humans, or in the exercising animal, techniques are not available for directly measuring regional myocardial 0, consumption transmurally in different regions of the heart. Therefore, this discussion will emphasize regional myocardial blood flow as an index of 0, supply and regional systolic wall thickening as a marker of
457
0, utilization, while recognizing their limitations for this purpose. In this connection, it should be noted that in the conscious dog, when 0, supply is progressively limited by means of a hydraulic cuff to produce increasing levels of acute ischemia, subendocardial myocardial blood flow is the major determinant of regional wall contraction in the steady state.15 The relation between this flow and regional systolic wall thickening is nearly linear over a wide range of flow reductions in the conscious dog.16 The relation between the average transmural flow and function is also good but less close, and subepicardial flow correlates poorly. “*M This suggests that 0, supply and demand can remain in balance during acute periods of graded ischemia, provided there is some residual coronary blood flow. The Transient Period During Onset of Exercise
As treadmill exercise is begun in the chronic animal model of single vessel coronary artery disease, there is a sharp increase in the heart rate, systemic arterial pressure, and maximum (+) LV dP/dt, as well as in regional systolic wall thickening (expressed as a percent of enddiastolic wall thickness, or %WTh) (Fig 1). In the zone supplied by the stenosed artery, however, following this initial enhancement regional %WTh quickly falls to a level well below the normal resting value, accompanied by some reduction (but continued elevation above the resting value) of LV systolic pressure and LV dP/dt (Fig 1). Thus, during this transient period a condition of relative ischemia (ie, supply and demand imbalance) exists in the circumflex artery distribution, in which the rapidly changing hemodynamic and cardiac responses promptly augment the regional 0, demands, whereas the marked heart rate increase in the presence of a limited coronary vascular reserve rapidly impairs subendocardial blood flow. A hypoperfused subendocardium results from an increased number of systolic periods per minute, during each of which blood flow to the subendocardium is very low because of the increased extravascular tissue pressure caused by LV contraction.5 This effect of tachycardia causes a reduction in the time available for diastolic subendocardial perfusion per minute and per beat, a situation which reflects the previously described inverse relation between heart rate and subendocardial myocar-
458
dial blood flow in the presence of maximum coronary vasodilation (produced by adenosine infusion) in the absence of coronary stenosis.’ Regional Ischemia During Steady-State Exercise
After about one minute of strenuous exercise on the treadmill, hemodynamic conditions stabilize and the severely depressed regional wall contraction in the ischemic zone reaches a steady state (Fig l), accompanied by reduced subendocardial and midwall blood flows (compared to the resting values) and a higher level of subepicardial blood flow than at rest (but well below that in the normal zone during exercise, Fig 2). At that time, it may be postulated that the reduced regional contraction is matched to the reduced coronary blood flow and 0, supply. Determinants of 0, Supply and Demand During Steady-State Exercise
In other experiments, when exercise was carried out in the presence of different degrees of acute coronary stenosis to produce various steadystate levels of regional myocardial dysfunction, a close correlation relation was observed between subendocardial blood flow and contractile function.17 It is of interest that if the level of subendocardial myocardial blood flow in the ischemic region during such steady-state periods of exercise associated with varying degrees of coronary stenosis is expressed per beat and plotted against regional myocardial function (expressed as normalized %WTh), the relation lies very close to that observed under resting conditions (Fig 3). This observation is pertinent to the proposed concept ofperfusion-contraction matching between subendocardial myocardial blood flow and regional function during various degrees of acute ischemia. It also suggests that under the ischemic conditions of exercise, the heart rate is a key determinant of the amount of blood flow (and hence 0,) available for each active contraction which, in turn, will largely set the level of regional contractile dysfunction in the steady state. This notion derives from the fact that although the other determinants of myocardial Oz consumption including the arterial pressure and global myocardial contractility (expressed as LV dP/dt and velocity of regional contraction in
JOHN
ROSS,
JR
nonischemic areas) differ widely between the resting and exercise states, nevertheless the relationship between subendocardial myocardial blood flow per beat and regional contraction per beat in the ischemic zone is similar under these two conditions over a wide range of subendocardial blood flows (Fig 3). It may be concluded that decreased oxygen supply is a major factor associated with ischemia of the oxygen demand type produced by exercise. This decreased Oz supply is due primarily to the heart rate effect (tachycardia), but also may be enhanced by increased a-adrenergic coronary vasoconstrictor tone, as discussed subsequently, and elevated LV end-diastolic pressure during exercise may further contribute to impaired subendocardial perfusion. During the steady state of acute ischemia due to exercise, regional oxygen supply and demand per beat appear to be matched, with decreased contraction corresponding to the level of decreased coronary blood flow per beat. APPROACHES
TO ANTIANGINAL
THERAPY
Various forms of drug treatment for angina pectoris traditionally have been categorized by whether their predominant effect is through altering cardiovascular hemodynamics to reduce myocardial Oz demand, or by increasing 0, supply through augmenting myocardial blood flow. Modification of Regional 0, Demand
Based on the considerations outlined earlier, among the various determinants of ventricular performance that might be manipulated to reduce exercise-induced ischemia and improve regional contraction, slowing of the heart rate would seem likely to be of prime importance. Thus, lowering the aortic pressure alone would tend to decrease left ventricular afterload and 0, demands, but diminishing aortic pressure during exercise could also reduce distal coronary perfusion pressure and might impair flow in the bed beyond a coronary stenosis. A beneficial effect of reducing 0, demand in the ischemic zone by this approach, while not yet established, might occur with vasodilator drugs such as nitroglycerin which reduce the arterial pressure, LV size, and hence wall stress,” also diminishing afterload in the
ANTJANGlNAL
DRUGS
& EXERCISE-INDUCED
ISCHEMIA
459
All P0ltVT.S
1.5
a l l
5; 2? z
: .
.
l .
k--A,
l
AA
A
I
e
0 0
8
tA
0 EXERC/SE A AT ItEST
5 z00
0 A
.005
I
I
.OlO
.015
z l s -0.3 .g 1.5 - I CONtRULAM? I STEMOS~S cc POINTPERDOG . II -
1
.020
0
l
Fig 3. Relation between subendocardisl blood flow per beat and normalized systolic wall thickening (expressed as a decimal fraction of the wall thickening [WTH] at rest) in conscious dogs studied at rest during progrearive coronary stenosis, and during treadmill exercise with varying degrees of coronary stenosis produced by a hydraulic cuff. Under these conditions, the rest and exercise-flow function relations are superimposable. Triangles with bars indicate resting blood flows with standard deviations. Modiied and reproduced with permission.”
l 0
l
0
A /’ l
# 4 EXE/K/SE ---b AT REST
0'
c/)
I
I
.OlO
.015
1
-s
-0.3
.005 1
.
.020
SUBENDOCAROlALBLOOO FLOW PER BEAT (ml/min/g)
ischemic region, although other effects of nitroglycerin on the coronary circulation are likely to be of more importance, as discussed subsequently. Slowing of the heart rate can produce large changes in 0, demand per minute in the ischemic zone as well as permit a small increase in absolute subendocardial myocardial blood flow per minute by mechanisms discussed earlier. In addition, it will substantially increase the regional myocardial blood flow (and 0, availability) per beat. This effect, in turn, should lead to augmented regional contraction in the ischemic
zone, and consequently increased myocardial 0, consumption per beat. ,&Adrenergic blockade. Improvement of regional contraction in the ischemic zone during exercise after intravenous IV propranolol compared to control exercise without drug was demonstrated in a chronic coronary stenosis canine model.” Similarly, in patients with coronary heart disease and exercise-induced ischemia, improvement of the exercise ejection fraction after ,8-adrenergic blockade has been demonstrated.*’ More recently, the effects of P-blockade on both regional function and regional coronary blood
JOHN
460
flow distribution were examined after oral administration of atenolol in our animal model.7 Atenolol decreased the average exercise heart rate from 228 to 181 beats/min, the increase in systolic arterial pressure was less, and the increase in LV dP/dt was markedly blunted compared to control running. In the ischemic zone, normalized regional myocardial blood flow to the subendocardium increased (Fig 2). Absolute subendocardial flow increased slightly compared to control exercise, whereas the subepicardial blood flow was lower and the usual increase in transmural blood flow in normal zones was markedly attenuated. Despite reduction of the normal enhancement of contraction during exercise in nonischemic regions, regional wall dysfunction in the ischemic zone improved (%WTh from 4.0 to 8.5) after atenolol. The changes of regional blood flow per minute in the ischemic zone were associated with an even greater improvement of subendocardial blood flow per beat (average .0019 to .0030 mL/min/g, a 58% increase). In addition, atenolol caused partial amelioration of the transmural maldistribution of regional blood flow in the ischemic zone, the endocardial/epicardial flow ratio increasing significantly from 0.27 to 0.47 .7 These experimental findings implicate several mechanisms in the beneficial effects of pblockade. There is evidence that with exercise in the presence of critical coronary stenosis transmural steal of flow from subendocardial to subepicardial layers may occur,‘* and other findings suggest that even with severe dyskinesia of the wall during ischemia the outer myocardial fibers are actively contracting. 21 Therefore, it may be postulated that /?-blockade, by reducing the heart rate, arterial pressure, and myocardial contractility during exercise, reduces the 0, consumption of the outer wall layers in the ischemic region, reducing blood flow requirements to the epicardial zone which, in turn, should lead to increased coronary vascular resistance in that region, improve the distal coronary perfusion pressure, and lead to a reduction of transmural steal and the subendocardial blood flow deficit (Table 1). In support of this view are experiments at rest in the conscious dog with acute coronary stenosis, in which it was shown that propranonol improves regional contractile function in the ischemic
Table
1. Antianginal Ischemic
Mechanisms
Region
During
1. 1 Heart
rate
lo,
1 Arterial pressure 2. 1 Heart fate 3. 1 Heart rate 1 Arterial pressure 1 Contractility Abbreviations:
of &Blockade Steady
Cons,
JR
on
Exercise
Mechanismld
ACtiOtllS~
ROSS.
Effectls)
Con%“,
1 O,Avail/beat ontraction
r Diastolic Pet-f Time/min
ILIF
1 0, con%, 1 Transmural
1 MBF,,, t MBF,
7 t MBTLbeat
steal
1 endolepi consumption;
Perf,
perfusion;
Avail,
available.
zone.22 Studies in open-chest dogs with a noncircumferential coronary stenosis have confirmed this finding and further demonstrated improved subendocardial how and decreased subepicardial blood flow in the ischemic region after propranonol, without a change in transmural coronary blood flo~.‘~ The improvement in regional myocardial function was closely correlated with the distal diastolic coronary perfusion pressure per minute (DDPTI), and increased distal coronary perfusion pressure after propranonol was proposed to be due primarily to reduction of MVO, with restoration of autoregulation in the ischemic area, primarily in the subepicardium. This response also caused a reduction in the calculated stenosis resistance by increasing the distending pressure in the lesion.23-25 (The latter mechanism seems unlikely in our experiments with the chronic ameroid constrictor, which is circumferential. Moreover, in some experiments the artery was totally included). Gross et al later showed that prior drug-induced dilation of the coronary bed prevented the decrease in subepicardial blood flow and improvement in subendocardial blood flow by propranolol, although regional function was not measured.26 Changes in collateral blood flow seem unlikely to have played any role in those experiments, since the diastolic perfusion pressure difference from aorta to the distal coronary bed, a primary determinant of collateral blood flo~,‘~ was reduced.23 The mechanism involved during exercise in the ameroid constrictor model seems most likely to be that of vasoconstriction in the subepicardial region, since the aortic pressure was lower during exercise after atenolol than during control exercise, making increased collateral flow unlikely, and the increased subendocardial/subepicardial
ANTIANGINAL
DRUGS
& EXERCISE-INDUCED
ISCHEMIA
flow ratio was not accompanied by any change in mean transmural flow to the ischemic region after /3-blockade.7 As suggested earlier, the increased subendocardial blood flow per minute and per beat in the ischemic zone may also relate, in part, to a decreased number of systolic contractions per minute with more time for diastolic coronary perfusion,’ although this mechanism remains to be clearly established in the presence of ischemia. Thus, how much extravascular compression actually occurs in the subendocardium with systole during severe ischemia and contributes to the coronary vascular resistance in this setting is not known. In addition to the increased subendocardial blood flow per beat after ,&blockade, the increased time available for recovery of metabolism per beat and over several beats due to the diminished number of contractions per minute undoubtedly contributed to the increased contraction in the ischemic zone (Table 1). The singular importance of heart rate slowing to the effect of P-blockade on regional ischemia during exercise is evidenced by recent studies on exercise after B-blockade in which, by the use of atria1 pacing during exercise, the heart rate was returned to the high level which existed during control exercise before drug administration.28 Under pacing conditions, regional myocardial blood flow actually diminished slightly below the value observed in the control exercise period, and regional myocardial function failed to improve. Thus, elimination by pacing of the slowing of heart rate produced by atenolol completely counteracted any beneficial effect of P-blockade on the ischemic zone.28 It may be concluded from these observations that the conventional view of the mechanism of action of @-blockers, ie, relief of ischemia and angina pectoris by reducing myocardial 0, demands during exercise, should be modified to indicate that they also increase 0, supply to the subendocardium of the ischemic zone (Table 1). In fact, 0, consumption per beat in the ischemic zone must actually increase as regional contraction improves. Bradycardic drugs. In view of these observations on the importance of heart rate, the action of a new bradycardic agent, UL FS-49 (Thomae Pharmaceutical, Biberach, FRG, made available
461
by Dr Jurgen Dammgen) was tested in the chronic coronary stenosis model.i4 A considerably greater slowing of heart rate during exercise than that caused by P-blockade could be achieved with this drug (exercise heart rate 139 v 181 beats/min with atenolol) without any depressant effect on arterial pressure, or the normal increase of LV dP/dt and regional function in nonischemic zones. A marked improvement of regional myocardial function in the ischemic zone to a near normal level was observed after UL FS-49, compared to control exercise without drug (%WTh 9.3 to 21.5%). Regional subendocardial and mid-wall blood flow per minute improved slightly, while subendocardial flow per beat in the ischemic zone increased twofold.‘4 These observations further confirm the important role of heart rate in the regional ischemic response to exercise. Modification
of Regional 0, Supply
In addition to indirect effects on myocardial blood flow due to P-blockade and heart rate slowing, discussed above, the direct action of coronary vasodilator drugs can improve myocardial perfusion and 0, supply by actions on the large coronary arteries, the collateral circulation, and the small coronary resistance vessels. In this connection, there is increasing experimental evidence that additional vasodilator reserve can be recruited by pharmacologic agents, even in the presence of severe ischemia.29-32 Dilation of large coronary arteries. Vasodilation occurs in the epicardial coronary arteries after administration of agents such as nitroglycerin, calcium channel blocking drugs, or a-adrenergic blocking agents.33 There is also evidence in patients with coronary artery disease that such drug-induced increases in the diameter of large coronary arteries at the site of noncircumferential atherosclerotic lesions can reduce the degree of coronary artery stenosis, whereas aradrenergic agents cause vasoconstriction.34 Nonselective cy-adrenergic blockade has been shown to diminish stress-induced (ice water immersion) regional myocardial ischemia in patients with coronary heart disease, perhaps by this mechanism and/or by reducing a-adrenergic tone to coronary resistance vessels.35It should be pointed out that such effects of vasodilator drugs on the
JOHN
462
JR
/
caliber of the circumflex coronary artery stenosis should not be a factor in our chronic animal model, since fixed annular narrowing is produced by the ameroid constrictor. There is increasing experimental evidence that other mechanisms of vasodilation, including that due to endothelial derived relaxing factor, flowmediated dilation, and dilation of small vessels that are larger than the traditional arteriolar resistance vessels also are important in the regulation of coronary blood flo~.~~
100
Atenolol 0 Diltiazem A Atenolol and Dittiazem
‘-9iw /
l
I
80
Rest
, /
80 40 20
Dilation of coronary collateral and/or resistance vessels. In the chronic canine model,
increased regional myocardial function in the ischemic zone during severe exercise-induced ischemia at the same or reduced arterial pressure has been demonstrated after the calcium channel blocking drugs diltiazem, nifedipine, and verapamil, indicating that vasodilation with lowered vascular resistance occurred in the native coronary bed and/or coronary collateral channels.6*32,36 Diltiazem and nifedipine were shown to change conditions in the ischemic zone by increasing subendocardial and transmural blood flow in the ischemic and normal zones, and improving the operating position of the ischemic zone on the subendocardial myocardial blood flow-regional contraction relation (Fig 4) (regional blood flow was not measured in experiments with verapamil). With diltiazem, we observed a small reduction in arterial pressure and a modest reduction in heart rate (from 227 to 204 beats/min) during exercise, so that part of its effect was mediated by heart rate mechanisms discussed above (Table 2). However, its action to improve contraction and blood flow (subendocardial flow per beat increased 54%) were equal to those produced by atenolol (Fig 4), which caused a much larger reduction in exercise heart rate. Therefore, an additional vasodilator action of diltiazem was implicated (Table 2).6 In contrast to P-blockade,’ diltiazem caused no depression of contraction in normal regions during exercise.6 Nifedipine was studied at a low dosage sufficient to affect the coronary circulation without appreciable systemic hemodynamic actions.32 During exercise after nifedipine in the chronic coronary stenosis model, the systolic arterial pressure and heart rate were unchanged from control exercise. Nevertheless, a significant in-
ROSS,
0
0
20
40
60
80
100
SUBENDOCAROIALRMBF (post. as % ant.) Fig 4. Relations between subendocardial regional myocardial blood flow (RMBF) in the posterior ipost) irchemic wall (expressed as a percentage of that in the control anterior [ant] walll and normalized regional systolic wall thickening in the ischemic zone (expressed as a percentage of the resting value [%WT]). The reductions in flow and function during control running before drug administration are similar for atenolol, diltiazem. and the combination of these drugs. Significant improvement in both regional flow and regional function during exercise is evident with either drug alone, but when both drugs were given (open triangles) substantial further improvement during running is evident. Reproduced with permirsion.”
crease in subendocardial and transmural regional myocardial blood flow occurred, together with increased contraction in the ischemic re gion, suggesting that coronary vasodilator re serve was recruited. In contrast to P-blockade, nifedipine and diltiazem produced a lower LV end-diastolic pressure during exercise than control running without drug, which should favor subendocardial perfusion in the ischemic zone.32 Table
2.
Antianginal
u-Blockade
Mechanisms:
on Ischemic
Actions
Vasodilators
Region
During
Mechanism(s)
Effect
1.
Relax coronary smooth muscle
1 Vascular resistance of arterioles
2.
1 LVED
pressure
1 Heart
size
1 Subendocardial tissue pres-
and/or
3.
1 Arterial pressure 1 Heart size (1 Heart
and
Exercise
f MBF,-,,
collaterials r MBF,,,,
sure 1 0, Cons/minute
1 0, Avail/beat
rate, variable)
Abbreviations: LVED, footnotes to Table 1.
left ventricular
end
diastolic:
also
see
ANTIANGINAL
DRUGS
& EXERCISE-INDUCED
ISCHEMIA
Sizeable effects on regional myocardial contraction due to isosorbide dinitrate were also described in the canine model, although regional blood flow was not measured.18 This effect occurred at a lower systemic arterial pressure and smaller heart size, suggesting hemodynamic as well as coronary vasodilator effects, and it was suggested that the lowered LV end-diastolic pressure during exercise compared to control exercise might have contributed to improvement in subendocardial blood flow (Table 2). In addition, as discussed earlier, the decreased arterial pressure and heart size with nitroglycerin would be expected to reduce systolic wall stress, which should diminish 0, demand and also might contribute to enhancement of systolic wall thickening by reducing afterload in the ischemic zone.” In the chronic coronary stenosis model, it is not yet possible to differentiate clearly between an action of these vasodilator agents on vascular resistance in the poststenotic native coronary bed and that in the coronary collateral channels, even when the coronary artery is entirely occluded by the ameroid constrictor. Effects of the various agents tested do not appear different whether the coronary artery is partially open with poorly developed collaterals, or closed with a more fully developed collateral circulation. It will be necessary to develop a reliable method for measuring coronary perfusion pressure distal to the coronary stenosis during exercise before this question can be approached. As recently n-Adrenergic mechanisms. viewed by Feigl, the coronary vasculature of the dog contains both oi and aZ postjunctional receptors that induce vasoconstriction.37 Experiments in normal dog hearts not subjected to ischemia have indicated that increased a-adrenergic tone to the coronary vessels during exercise can reduce coronary blood IIow.~‘-~~ Investigations in anesthetized dogs have also shown that (Yadrenergic receptor-mediated coronary vasoconstriction can be produced by direct sympathetic nerve stimulation, even during severe ischemia.40 However, based on other experiments during milder ischemia using norepinephrine stimulation and a different blocking drug, Feigl has proposed that increased cr-adrenergic tone af-
463
fecting the subepicardial region may serve to maintain subendocardial perfusion.37 Recent studies in a conscious canine mode1 in which an cY,-adrenergic blocking drug (idazoxan) was infused into an acutely stenosed coronary artery during exercise, showed improvement in regional transmural and subendocardial blood flow as well as increased regional wall motion in the ischemic zone, suggesting that increased cY-adrenergic tone to the native coronary resistance vessels can limit subendocardial perfusion even during severe exercised-induced ischemia.41 Such studies are carried out after fi-adrenergic blockade to prevent direct and indirect vasodilator effects due to increased norepinephrine release consequent to blockade of the prejunctional a2 receptors.41 Some studies indicate that coronary collateral vessels in dogs do not contain a,-adrenergic receptors,42 whereas others suggest that they do43; whether or not a,-adrenergic blockade during exercise improved vasodilation in the native coronary resistance vessels or in coronary collaterals cannot be stated with certainty. There is also experimental evidence that nifedipine can prevent sympathetic coronary vasoconstriction in the vascular bed perfused by an artery with severe stenosis. 44 The effects in the exercising dog with regional ischemia of (pi blockade, and also the effects of IV LYEor a2 blockade in the presence and absence of P-blockade, remain to be investigated. It is of interest that investigations in patients with coronary stenosis have demonstrated exercise-induced vasoconstriction at the stenosis site, which was relieved by nitroglycerin4’ Also, a nonselective a-adrenergic blocking drug (phentolamine) given by the intracoronary route was shown to reduce evidence of ischemia during supine exercise in patients with coronary heart disease.46 It may be concluded that vasodilators can act through multiple mechanisms, including dilation of large coronary arteries (Table 2). They may also dilate native coronary resistance vessels, perhaps in part by relief of cY-adrenergic constrictor tone, and they may also dilate coronary collateral vessels, although the roles of these different vasodilator components cannot cur-
JOHN
464
rently be quantified in the chronic animal model of exercise-induced ischemia. Combined Antianginal
Therapy
It might be expected that combinations of antianginal drugs having different mechanisms of action, such as a p-blocking drug and a coronary vasodilator, would have additive effects, and beneficial effects of such combinations have been described in clinical studies.47 In the chronic coronary stenosis animal model, the combination of atenolol plus diltiazem was administered before exercise and compared to control exercise, as well as to experiments in other animals in which either diltiazem or atenolol was administered alone. The effect of the combined drug treatment was additive, both subendocardial myocardial blood flow and regional systolic wall thickening showing an improvement of approximately twofold over that observed with either agent alone (Fig 4).48 Also, this drug combination plus isosorbide dinitrate was shown to prevent regional dysfunction during milder ischemia.4g SUMMARY
Several mechanisms involved in the production of regional exercise-induced ischemia are described. Each offers the potential for modification using different types of antianginal drugs operating to alter regional 0, demands, improve regional perfusion, or both, leading to reduced ischemia and increased contractile function in the ischemic zone.
R066.J~
Evidence is presented for matching of regional subendocardial myocardial blood flow and flow per beat with regional myocardial contraction at various levels of ischemia at rest, during steadystate exercise, and after antianginal drugs, signifying a particularly important role for heart rate control. In addition to reducing myocardial 0, demand per minute, B-blockers and bradycardic drugs cause improvement of absolute subendocardial blood flow and particularly flow per beat by producing vasoconstriction in the epicardial region of the ischemic zone, with improvement of transmural blood flow distribution. Vasodilator drugs can act at several locations to increase regional blood flow and also to decrease 0, demands. A recruitable vasodilator reserve has been shown to exist during exerciseinduced ischemia either in native resistence vessels, collateral channels, or both, which appears to be due at least in part to reduction of increased a-adrenergic constrictor tone to the coronary vessels during exercise, even in the presence of severe ischemia. The potential for additive effects using combinations of bradycardic and vasodilating agents are described within a framework relating regional subendocardial blood flow to regional systolic contraction. The experimental findings described suggest some potential new directions for antianginal therapy and, along with recent clinical observations, support the use of combinations of antianginal agents that act by different mechanisms.
REFERENCES 1. Tomoike H, Franklin D, McKown D, et al: Regional 7. Matsuzaki M, Patritti J, Tajimi T, et al: Effects of fl myccardial dysfunction and hemodynamic abnormalities durblockade on regional myocardial flow and function during ing strenuous exercise in dogs with limited coronary flow. exercise. Am J Physioll6:H52-H60, 1984 Circ Res 42:487-496,1978 8. Grover-McKay M, Matsuzaki M, Ross J Jr: Dissocia2. Raw J Jr: Perspective: Assessment of ischemic regional tion between regional myocardial dysfunction and subendocarmyocardial dysfunction and its reversibility. Circulation dial ST segment elevation during and after exercise-induced 74:1186-1190.1986 ischemia in dogs. J Am Co11Cardiol 10:1105-l 112.1987 3. Mosher P, Ross J Jr, McFate PA, et al: Control of 9. Homans DC, Sublett E, Dai X-Z, et al: Persistence of coronary blood flow by an autoregulatory mechanism. Circ regional left ventricular dysfunction after exercise-induced Res l&250-259,1964 myocardial &hernia. J Clin Invest 77:66-73,1986 4. Hoffman JIE: Maximal coronary flow and the concept 10. Robertson WS, Feigenbaum H, Armstrong WF, et al: of coronary vascular reserve. Circulation 70: 153- 159,1984 Exercise echocardiography: A clinically practical addition in 5. Bathe RJ, Cobb FR: Effect of maximal coronary the evaluation of coronary artery disease. J Am Cot1 Cardiol vasodilation on transmural perfusion during tachycardia in 2:1085-1091,1983 the awake dog. Circ Res 41:648-653,1977 11. Kumada T, Gallagher KP, Battler A, et al: Compari6. Matsuzaki M, Guth B, Tajimi T, et al: Effect of the son of post-pacing and exercise-induced myocardial dysfunction during collateral development in conscious dogs. Circulacombination of diltiazem and atenolol on exercise-induced tion 65:1178-1185, I982 regional myocardial ischemia in conscious dogs. Circulation 72:233-243.1985 12. Gallagher KP, Osakada G, Matsuzaki M, et al:
ANTIANGINAL
DRUGS
& EXERCISE-INDUCED
ISCHEMIA
Myocardial blood flow and function with critical coronary stenosis in exercising dogs. Am J Physiol: Heart Circ Physiol 12:H698-H707.1982 13. Hoffman JIE, Buckberg GD: The myocardial supply: Demand ratiea critical review. Am J Cardiol 41327-332, 1978 14. Guth BD, Heusch G, Ross J Jr: Elimination of exercise-induced regional myocardial dysfunction by a bradycardiac agent in dogs with chronic coronary stenosis. Circulation 75:661-669, 1987 15. Gallagher KP, Kumada T, Koziol JA, et ai: Significance of regional wall thickening abnormalities relative to transmural myocardial perfusion in anesthetized dogs. Circulation 62:1266-1274, 1980 16. Gallagher KP, Matsuaaki M, Koziol JA, et al: Regional myocardial perfusion and wall thickening during ischemia in conscious dogs. Am J Physiol 16:H727-H738, 1984 17. Gallagher KP, Matsuxaki M, Osakada G, et al: Effect of exercise on the relationship between myocardial blood flow and systolic wall thickening in dogs with acute coronary stenosis. Circ Res 52:716-729, 1983 18. Kumada T, Gallagher KP, Miller M, et al: Improvement by isosorbide dinitrate of exercise-induced regional myocardial dysfunction in the dog. Am J Physiol 239: H399-H405,1980 19. Kumada T, Gallagher K, Shirato K, et al: Reduction of exercise-induced regional myocardial dysfunction by propranolol: Studies in a canine model of chronic coronary artery stenosis. Circ Res 46:190-200, 1980 20. Battler A, Ross J Jr, Slutsky R, et al: Improvement of exercise-induced left ventricular dysfunction with oral propranolol in patients with coronary heart disease. Am J Cardiol44:3 18-324, 1979 2 1. Gallagher KP, Osakada G, Hess OM, et al: Subepicardial segmental function during coronary stenosis and the role of myocardial fiber orientation. Circ Res 50~352-359, 1982 22. Tomoike H, Ross J Jr, Franklin D, et al: Improvement by propranolol of regional myocardial dysfunction and abnormal coronary flow pattern in conscious dogs with coronary narrowing. Am J Cardiol41:689-696, 1978 23. Buck JD, Hardman HF, Warltier DC, et al: Changes in ischemic blood flow distribution and dynamic severity of a coronary stenosis induced by fi blockade in the canine heart. Circulation 64:708-7 15, 198 1 24. Gould LK: Dynamic coronary stenosis. Am J Cardiol 45286-292, 1980 25. Santamore WP, Walinsky P: Altered coronary flow responses to vasoactive drugs in the presence of coronary arterial stenosis in the dog. Am J Cardiol45:276-285, 1980 26. Gross GJ, Buck JD, Warltier DC, et al: Role of autoregulation in the beneficial action of propranolol on ischemic blood flow distribution and stenosis severity in the canine myocardium. J Pharm Exp Ther 222:635-640, 1982 27. Brown BG, Gendel WD, Gott VL, et al: Coronary collateral blood flow following acute coronary occlusion: A diastolic phenomenon. Cardiovasc Res 8:621-631, 1974 28. Guth BD, Heusch G, Seitelberger R, Ross J Jr: Mechanism of beneficial effect of beta-adrenergic blockade on exercise-induced &hernia in conscious dogs. Circ Res 60:738-746. 1987
465 29. Canty JM, Klocke FJ: Reduced regional myocardial perfusion in the presence of pharmacologic vasodilator reserve. Circulation 71:370-377, 1985 30. Aversano T, Becker LC: Persistence of coronary vasodilator reserve despite functionally significant flow reduction. Am J Physiol248:H403-H411, 1985 31. Pantely GA, Bristow JD, Swenson LJ, et al: Incomplete coronary vasodilation during myocardial ischemia in swine. Am J Physiol249:H638-H647, 1985 32. Heusch G, Guth BS, Seitelberger R, Ross J Jr: Attenuation of exercise-induced myocardial ischemia in dogs with recruitment of coronary vasodilator reserve by nifedipine. Circulation 75:482-490, 1987 33. Young MA, Vatner SF: Regulation of large coronary arteries. Circ Res 59:579-596, 1986 34. Brown BG, Bolson EL, Dodge HT: Dynamic mechanisms in human coronary stenosis. Circulation 70:917-922, 1984 35. Feldman RL, Whittle JL, Marx JD, et al: Regional coronary hemodynamic responses to cold stimulation in patients without variant angina. Am J Cardiol 49:665-673, 1982 36. Osakada G, Kumada T, Gallagher KP, et al: Reduction of exercise-induced regional myocardial dysfunction by verapamil in conscious dogs. Am Heart J 101:707-712, 1981 37. Feigl EO: The paradox of adrenergic coronary vasoconstriction. Circulation 76:737-745, 1987 38. Murray PA, Vatner SF: a;adrenoceptor attenuation of the coronary vascular response to severe exercise in the conscious dog. Circ Res 45:654-660, 1979 39. Gwirtz PA, Gvern SP, Mass HJ, et al: cu,-adrenergic constriction limits coronary flow and cardiac function in running dogs. Am J Physiol 250:Heart Circ Physiol 19: H1117-H1126.1986 40. Heusch G, Deussen A: The effects of cardiac sympathetic nerve stimulation on the perfusion of stenotic coronary arteries in thedog. Circ Res 53:8-15, 1983 41. Seitelberger R, Guth BD, Heusch G, et al: Intracoronary Alpha,-adrenergic receptor blockage attenuates ischemia in conscious dogs during exercise. Circ Res 62: 435-442,1988 42. Harrison DG, Chilian WM, Marcus ML: Absence of functioning a-adrenergic receptors in mature canine coronary collaterals. Circ Res 59:133-142, 1986 43. Maruoka Y, McKirnan MD, Engler RL, et al: Functional significance of a-adrenergic receptors in mature coronary collateral circulation of dogs. Am J Physiol 253: H582-H590,1987 44. Heusch G, Deussen A: Nifedipine prevents sympathetic vasoconstriction distal to severe coronary stenosis. J Cardiovasc Pharmacol6:378-383, 1984 45. Gage JE, Hess OM, Murakami T, et al: Vasoconstriction of stenotic coronary arteries during dynamic exercise in patients with classic angina pectoris: Reversibility by nitroglycerin. Circulation 73:865-876, 1986 46. Berkenboom GM, Abramovicz M, Vandermoten, et al: Role of alpha-adrenergic coronary tone in exercisedinduced angina pectoris. Am J Cardio157: 195 198, 1986 47. Packer M, Meller J, Medina N, et al: Hemodynamic consequences of combined beta-adrenergic and slow calcium channel blockade in man. Circulation 65:660-668, 1982
466
48. Matsuxalci M, Guth B, Tajimi T, et al: Effect of the combination of diltiazem and atenolol on exercise-induced regional myocardial &hernia in conscious dogs. Circulation 72:233-243,1985
JOHN ROSS, JR
49. Guth BD, Tajimi T, Seitelberger R, et al: Experimental exercise-induced ischemia: Drug therapy can eliminate regional dysfunction and oxygen supply-demand imbalance. J Am Co11Cardiol7:1036-1046, 1986