AMERICAN JOURNAL OF PHYSIOLBCY Vol. 230, No. 1, January 1976. Printed
in U.S.A.
Myocardial contractile force as a function of coronary blood flow JAMES M. DOWNEY Department of Physiology,
University
of South
DOWNEY, JAMES hf. MyocardiaZ contractde force as a function of coronary blood flow. Am. J, Physiol. 230(l): l-6. 1976.--The contractile force of the deep and superficial myocardial fibers was examined in the open-chest anesthetized dog as a function of coronary blood flow (CBF). When 1) dogs that failed to demonstrate coronary autoregulation were eliminated 5om the data base and 2) CBF and contractile force data were both normalized as a percent of their values when perfusion was from aortic pressure (autoperfusion), the relationship between them became very reproducible. Contractile force was highly dependent on the flow rate when the CI3F was below that chosen by autoregulation (the rate during autoperfusion). Conversely contractile force was relatively independent of flow at higher CBF. The contractile force-CBF curve thus was found to break precisely at the autoperfused CBF. When myocardial metabolism was elevated by paired electrical stimulation this relationship was unchanged. It was concluded that coronary blood flow is tightly regulated to match metabolic needs over a range of metabolic rates, paired
pacing;
garden
hose effect; autoregulation;
ischemia
iscontrolledbyanautoregulatory mechanism (4, 19). This system, which tends to maintain flow rate independent of perfusion pressure, is also sensitive to changes in cardiac metabolism (19). Evidence indicates that this control is the result of adenosine, a potent vasodilator, being released from hypoxic myocardial cells (5). Thus the heart is endowed with a control system that acts to match the delivery of oxygen and other metabolites to the heart’s needs. There is considerable evidence, however, that the blood flow rate this system provides is somewhat less than that required to optimize cardiac metabolism and thus performance. Gregg (11) reported that myocardial oxygen consumption (MVo,) could be increased by independently increasing CBF above the control level. Since then several investigators have found MVo, and/or ventricular performance to be limited by flow in what was thought to be a normal range of flows (1, 3, 8). This has not been a universal finding, however, in that others have not observed such a condition (21, 23, 28). After nearly a decade and a half of study the relationship between coronary perfusion and myocardial metabolism remains unclear. In many studies MVo, (8, 12, 21) or an index of contractility (1, 8) was plotted against flow rate per unit mass of myocardium. The resulting data show considerable variation probably because myocardial oxygen requirements vary greatly among individual experiCOR~NARYBLOODFLOW(CBF)
Florida
College of Medicine,
Tampa,
Florida
33620
ments, depending on the contractile or hemodynamic state. Furthermore, most of the reports do not indicate whether autoregulation was operative in the experimental preparations. It is well recognized that vascular tone and auturegulation often are absent in a canine heart preparation that is deteriorating. If such control is lost as a result of the conditions of the experiment, then the system will be unable to actively match flow to metabolism. Accordingly several workers do report that MVo, becomes less limited by flow in those hearts that exhibited some autoregulation (3, 29>, The present study attempts to correlate changes in myocardial performance with changes in CBF in hearts with intact autoregulatory control. To account for redistribution of blood flow across the ventricular wall, contractile force recordings from both deep and superficial myocardial fibers were used as an index of the heart’s performance. Contractile force was measured at flows both above and below the flow rate chosen by the autoregulatory mechanism and variations in myocardial metabolism among individual animals were then accounted for by expressing the independent variable, CBF, as a percent of the flow rate that autoregulation itself chose for the heart. METHODS
Mongrel dogs of either sex were anesthetized with sodium pentobarbital 5 mg/kg iv. The chests were opened in the fifth left interspace and the hearts exposed. The lungs were ventilated with 100% 0,. Blood pH was periodically monitored and ventilation rate was adjusted to maintain pH between 7.3 and 7.5. The left common coronary artery was exposed by a blunt dissection and a silk thread passed under it. A coronary cannula with the perfusion tubing attached was inserted into the subclavian artery and advanced into the left common coronary artery, where it was securely tied in place with the silk thread. The cannula was a double-lumen type (10) that withdrew arterial blood from the aorta via the outer lumen. Blood passed through the exteriorized circuit and into the coronary artery via the inner lumen. The tubing passed through the fingers of a Sigmamotor T8 peristaltic pump so that flow could be controlled when desired. Flow in the circuit was measured with a Statham K2000 electromagnetic flowmeter recording from an In Vivo Metrics cannulating-type transducer. Bypass tubing around the transducer allowed zeroing the flowmeter without interrupting coronary flow. Perfusion pressure was meas-
Downloaded from www.physiology.org/journal/ajplegacy by ${individualUser.givenNames} ${individualUser.surname} (130.070.008.131) on January 14, 2019.
2
J. M. DOWNEY
ured from a branch in the tubing distal to the pump have accompanied release. Second, when the coronary near the cannula. flow rate was abruptly changed with the pump a slow, Aortic pressure was measured from a Teflon catheter lo- to 20-s adjustment in perfusion pressure should have advanced through a femoral artery into the thoracic been seen as coronary tone changed in response to the aorta. Ventricular pressure was obtained from another new flow rate. Finally, pressure flow data collected Teflon catheter in the left carotid artery that was ad- during the experiment was later plotted. A region vanced across the aortic v ,alves. The left ventricul .ar where flow was nearly independent of perfusion prespressure was differentiated to obtain i ts derivative. A sure indicated good autoregulation. Preparations that sharp high-frequency cutoff was established by three failed to meet these criteria were terminated and no resistance-capacitance (RC) filters coupled to each other data were collected. in series, Each filter had a cutoff frequency (l/2 nRC) of The protocol was as follows: the bypass on the pump 75 Hz. was opened and the coronary artery was perfused with Separation of the contractile forces of the deep and the animal’s own aortic pressure (autoperfusion). The superficial fibers was accompli shed by orienting the pump was then engaged increasing or decreasing flow force gauge parallel to the fibers under study (10, 13). from the autoperfused rate. This was maintained until Contractile force of the superficial fibers was recorded the contractile force and perfusion pressure records from an isometric force gauge1 sewn in a base to apex reached a steady state (usually 30-60 seconds). The orientation (parallel to the superficial fibers) midway pump was then disengaged and autoperfusion was rebetween the left anterior descending and circumflex stored. The maneuver was repeated over a wide range of coronary arteries. Contractile force of the deeper fibers pump speeds. was recorded from an isometric force gauge modified by Two groups of dogs were studied. The first had gauges soldering a M-cm length of B-gauge hypodermic needle attached to record deep contractile force. After data perpendicular to each foot. This gauge was attached by were collected as described above the metabolic state of pressing the gauge onto the heart and impaling the these hearts was then augmented by paired electrical tissue with the prongs, which ensured good transmisstimulation and the procedure for data collection was sion of force from the deep fibers. The gauge was ori- repeated. Ten-volt pulses of 2 ms duration and sepaented perpendicular to the base-apex axis of the heart rated by 140-160 ms were delivered to bipolar electrodes between the left anterior descending and the circumflex sewn to the right ventricular free wall. Thus contractile arteries, with the gauge placed parallel to the deep force-CBF data were collected for both control and ones (25). paired-pacing fibers and normal to the more superficial conditions. In a second group of experiSuperficial sutures held the gauge in place. The gauges ments contractile force of the superficial fibers was studwere extended to about 130% of their initial length after ied as a function of CBF at a normal cardiac rhythm. attachment. Each gauge was tested by observing its response to 0.2 pg of intracoronary isoproterenol prior to RESULTS the co1lection of da ta. If contractile force did not increase Thirteen dogs were prepared for these experiments. by at least 100% the gauge was repositioned or the Two were eliminated because they failed to show a preparation was discarded. significant reactive hyperemia response. All of these Though it is not known precisely which fibers are contributing to the gauge’s response, it is possible to exhibiting a reactive hyperemia met the other two criteria for auturegulation as indicated in METHODS. The demonstrate clearly a differential sensitivity among pressure-flow relationship from one experiment can be events occurring at different depths in the myocardium seen in Fig. 1. The squares indicate flow with normal with this technique. In hearts with both gauges attached, Kirk et al. (13) demonstrated that topical cool- sinus rhythm and the circles indicate flow with paired ing of the surface of the heart with cold saline had no pacing. Though the curve is elevated with paired pacing myocardial metabolism both effect on the deep gauge but it typically doubled the as a result of increased response of the superficial gauge. Intravenous isoproterenol, on the other hand, caused similar increases in both gauges. Partial occlusion of the coronary artery, a maneuver that causes selective underperfusion of the subendocardium, depressed only the deep gauge. In previous studies deep gauges have been attached with deep sutures (10, 13) but this procedure is difficult and many preparations show attenuated responsiveness probably due to vascular occlusion by the sutures. The use of the rigid pins on the gauge rectifies these problems while preserving the selectivity of the gauge. In all experiments the presence of coronary autoregu50 100 150 lation was evaluated by the following criteria. First, m m/-/g CORONARY PERFUSION PRESSURE when the perfusion line was occluded for 5-10 s a reacFIG. 1. Pressure-flow curves from dog3, Table 1. CBF was altered tive hyperemia of at least 200% resting flow rate should I
l Metal-encased Box 412, Charleston,
type with movable SC. 29402).
foot
(John
A. Warren,
P. 0.
n
I
a
with a Sigmamotir pump while aortic pressure remained unchanged. Plateau region around 100 mmHg indicates autoregulation. Notice that paired pacing elevated CBF at all pressures,
Downloaded from www.physiology.org/journal/ajplegacy by ${individualUser.givenNames} ${individualUser.surname} (130.070.008.131) on January 14, 2019.
CONTRACTILE
FORCE-CORONARY
FLOW
3
RELATIONSHIP
show plateau regions where flow is nearly independent of perfusion pressure, indicating the presence of autoregulation. This graph is typical of the data. The records in Fig. 2 indicate the effects of changing the flow rate from the autoperfused level. Panels A and B reveal that increasing perfusion rate causes little change in deep contractile force. However, panek C and D reveal that contractile force is markedly compromised when flow is decreased. Quite often the pump was engaged with a setting equal to the autoperfused rate. Neither perfusion pressure nor contractile force changed when this occurred, indicating that the pulsations generated by the pump were not affecting the preparations. When deep contractile force is plotted against flow rate the graph in Fig. 3 results. The squares indicate data from one animal with a sinus rhythm. On the vertical axis deep contractile force is normalized by dividing the final value during pumping by the autoperfused or control value. The flow scale is similarly normalized by dividing the flow rate during pumping by the autoperfused rate. Thus flows are expressed as a percent of the flow rate that the heart’s regulatory mechanism establishes for itself. Because of the normalization procedure the curve must go through the index point 1, 1. At flows above 1, 1 the slope is very shallow, indicating that flow is adequate to maintain performante in this range. At flows below 1, 1, however, the curve falls off steeply, showing a region where performante is very flow dependent. The flow set by autoregulation was always on the knee of the curve. When cardiac metabolism was increased by paired pacing the points indicated by the circles in Fig. 3 resulted. Though paired pacing caused the regulatory mechanism to choose a new perfusion rate, as indicated by the pressure flow curve in Fig. 1 (circles), it did not change the relationship between contractile force and 8
A
c
CBF. The circles in Fig. 3 lie over the squares. Thus though autoperfused flow increased with paired pacing a flow was still provided that was at the knee of the contractile force-flow curve. This was a consistent finding in all five dogs in group Z. Figure 4 shows the data from all dogs plotted in the same manner as for Fig. 3. The open figures indicate normal cardiac rhythm and the solid figures indicate paired pacing. Though some variation is seen in the region below 1, 1 this seems to be the result of variability among animals since points taken during paired pacing tend to be very close to corresponding points with normal rhythm. The maximum positive derivative of the ventricular pressure (dP/&) is plotted against CBF in Fig. 5. Again, both variables are normalized as a percent of the autoperfused levels. These data were collected simultaneously with those in Fig. 3. Though changes in dP/&
l
l
SINUS
RHYTHM
PAIRED
PACE
m
d 1 5
1 IO CORONARY
1 15 BLOOD
1 20
FLOW
3. Deep contractile force plotted against coronary blood flow rate. Both were normalized by dividing steady-state value by value during autoperfusion. Curve therefore is forced through point 1, 1. Squares represent data taken while the heart beat at its own intrinsic rhythm and circles those taken during paired electrical stimulation. Data from dog 3, Table 1. FIG.
D
DCF
2oo
C8Fmf/
.__._--.
in
IT-l
o200,
AOP 100-B
mmkfg Illllllllwcz----.-_
k
45SECi
FIG. 2. Records in which CBF was either augmented (A and B) or reduced (C and D> from autoperfused level by engaging the pump. DCF, deep contractile force; AOP, aortic pressure; CP, mean coronary perfusion pressure; VP, ventricular pressure; and dP/&, derivative of ventricular pressure. Notice that DCF was affected more by reductions in CBF than increases. Slow adjustments in coronary perfusion pressure after abrupt changes in CBF are result of coronary autoregulation.
Downloaded from www.physiology.org/journal/ajplegacy by ${individualUser.givenNames} ${individualUser.surname} (130.070.008.131) on January 14, 2019.
4
J. M. DOWNEY
be maintained at the expense of the deeper fibers when tutal CBF is reduced (14). Thus the effects of underperfusion are first evidenced at the subendocardium (10, 13). The result is that the deep contractile force should be more sensitive to changes in total CBF. An additional group of dogs was studied and superficial contractile force was monitored to see if it really was less sensitive to changes in the CBF rate. In Fig. 6 contractile force-CBF data from six dogs with superficial gauges are plotted on the same axis as the control data from the five dogs in group I that had deep gauges. Though the data are very similar, three of the preparations failed to demonstrate an appreciable (5%) change in contractile force when the flow rate was reduced by 15% or greater from the control rate. Such a condition was not observed in group Z.
were usually in the same direction as the contractile force the changes were never as pronounced. Detectable changes in dP/dt often did not occur with small deviations in CBF away from the autoperfused level and thus a break point comparable to that seen in Fig. 3 is not easily demonstrated in the curves. Hemodynamic data for these five animals appear in Table 1. Deep contractile force was chosen for these experiments because flow in the subepicardial layers tends to
DISCUSSION
5
IO CORONARY
15
BLOOD
Unlike most of the previous investigations in which an index of cardiac metabolism is plotted against CBF the present study finds the relationship to be very reproducible* This is in part attributable to selecting only those animals with intact coronary regulation and then normalizing the data as a percent of that occurring with This normalization procedure for CBF autoperfusion. accounts for individual variations in cardiac metabolism among animals.
20
FLOW
FIG. 4. Composite of all deep contractile force data from 5 experiments. Data from individual dogs are identified by symbol shape. Open symbols correspond to data taken during sinus rhythm and solid symbols to those taken during paired pacing. Normalization is same as in Fig. 3.
n
l 0 @DEEP
A.SUPEHFICIAL b
I
I
I
5
IO
IS
CORONARY
BLOOD
I
?O
FLOW
6. Contractile force of deep fibers (circles) and superficial fibers (triangles) is plotted against coronary blood flow. Normalization is same as for Fig. 3. Curves are very similar; however, deep points tend to be lower than corresponding superficial points in range of flows below 1. FIG.
FIG. 5. Plot of maximum derivative of ventricular against coronary blood flow. Normalization again fused level. Data from dog 3, Table 1.
TABLE Dog No.
1. Hemodynamic Condition
--
pressure (dP/&) is against autoper-
data for ---5 group I dogs ----------Aortic
Pressure, mmHg ---p--p-p
dP/&,
-------mmHg/s
Relative DCF, mm
CBF,
mllmin
--
Heart Rate, beatslmin
LV wt, g
Dog Wt, kg
1
Control PP
90165 115180
1,200 2,250
8 13
64 126
160 124
98
18
2
Control PP
100172 1255180
NA NA
12 25
88 134
144 120
95
20
3
Control PP
125195 120192
1,750 2,250
12 20
84 105
164 152
109
20
4
Control PP
90165 110150
1,750 2,000
11 16
75 126
128 116
70
18
5
Control PP
10 22
85 108
120 112
82
20
135185 2,250 105180 3,000 ------PP --PP, paired pacing; dP/&, maximal positive derivative of ventricular deep contractile force during autoperfusion; values are in millimeters
pressure; excursion
-LV, left ventricle of recording pen,
-(including septum). Relative as gauge was not calibrated.
DCF
refers
to
Downloaded from www.physiology.org/journal/ajplegacy by ${individualUser.givenNames} ${individualUser.surname} (130.070.008.131) on January 14, 2019.
CONTRACTILE
FORCE-CORONARY
FLOW
RELATIONSHIP
When coronary flow is altered the distribution of that blood flow is also altered. At low perfusion pressures flow to the subendocardium is reduced to a greater extent than that at the subepicardium due to the high subendocardial tissue pressure (17). Therefore, depressed contractility due to hypoxia should appear first and be most pronounced at the subendocardium. For this reason it was thought that deep contractile force would probably be superior to other indices of contractility for this study. When CBF is reduced by lowering coronary perfusion pressure, redistributions cause deep contractile force to become depressed at consistently higher pressures than superficial force (10, 13). However, in the present experiment the two curves were very similar when plotted against CBF (Fig. 6). This is because the pressure-CBF curve experiences a plateau in the region of the autoperfused flow (Fig. 1). Thus large reductions in perfusion pressure result in a relatively small change in CBF, which tends to magnify the differences between the contractile force-pressure curves. In terms of actual flow values, then, superficial contractile force seems to have a relationship with CBF that is not as different from deep contractile force as was originally predicted. When the contractile force data were compared to measurements of dP/& the latter index was much less sensitive to changes in CBF. Changes in both preload and in afterload are known to affect dP/& (15, 27) and contractile force (ZO), yet at coronary flow values above about two-thirds of the autoperfused rate neither preload or afterload was appreciably altered in these experiments. Thus, both of these measurements should be valid. A possible explanation for this lack of sensitivity is that dP/& reflects the average contractile state across the ventricular wall and is not particularly sensitive to ischemia that is limited to the subendocardium. If the lack of sensitivity of dP/& to changes in CBF was the result of flow redistributions between layers, then the dP/& response should have been intermediate between contractile force at the subendocardium, where hypoxia first appears, and contractile force at the subepicardium, where hypoxia is last to appear. This was not the case, however. A more likely cause of the discrepancy between the two techniques is that velocity-related indices do not appear to be as sensitive to hypoxic depression as those related ti force (26). These data indicate that coronary blood flow is accurately regulated to match myocardial metabolism over a range of metabolic rates. Reductions in CBF from the autoperfused level proportionately decreased mechanical activity, whereas increases in CBF above this point had a minimal effect. Such a discontinuous relationship between MVo, (23, 28) or ventricular performance (8, 23) and CBF with a plateau region in the range of high CBF values has been seen by others and agrees with the present data. In all the experiments, however, contractile force was found to be slightly limited by blood flow at the flow rate provided by autoregulation. Doubling the flow rate over the autoperfused level typically raised contractile force by about 15%. Current evidence points toward adenosine as the mediator of coronary autorealation (4, 5).
5 The nucleoside is thought to be released from hypoxic myocardium. Its powerful vasodilatory action increases blood flow and restores oxygen balance to the tissue. Such a system requires that some adenosine must be released continuously since coronary tone is normally midway between a fully constricted and dilated state. The slight limitation to myocardial performance that coronary perfusion imposes on the heart at the autoperfused level may reflect a normal condition of slight hypoxia that is required to effect this continuous adenosine release. Alternatively this apparent limitation may be unrelated to the delivery of metabolites but rather purely a mechanical effect. Pressure in the coronary vessels, some have proposed, acts to stretch the myocardial fibers, increasing their preload through the socalled “garden hose” effect (2, 18). Such a mechanism may be operative here. Also an increased turgidity of the myocardium with elevated perfusion pressure could result in better mechanical coupling of surrounding tissue to the force gauge, thus slightly increasing its recorded force. Whatever the mechanism, the magnitude of this increase is indeed small. In terms of efficiency as a control system the match between blood flow requirements and supply in the heart appears to be optimal. The regulatory mechanism keeps perfusion at the knee of the contractile force-CBF curve. Though performance could be further increased slightly by a large increase in coronary flow the cost in terms of the additional increment in cardiac output required to supply this flow would be appreciable. On the other hand any savings in cardiac output afforded by reducing coronary flow over that determined by the regulatory mechanism would clearly be offset by greatly impaired performance. Since the contractile force-CBF relationship was essentially unchanged when metabolism was augmented by paired pacing, it would appear that this efficiency is maintained over a range of metabolic states. The difference between the slope of the contractile force-CBF curve above the autoperfused point as opposed to below it is so great that one is tempted to equate the region below that point with ischemia. Attenuated function has been used to indicate the onset of ischemia (7, 10, 24) but this index is only valid when all other factors that may decrease fu.nction can be ruled out. Lactate production (6) characterizes ischemic myocardium but the low venous lactate titer associated with mild ischemia or in cases where it is very localized is not always detectable. Enzymatic (14) and electrocardiographic (16) changes associated with ischemia are slow to develop and do not lend themselves to many experimental designs. We (9) as well as others (22) therefore have found it useful to interpret ischemia as a condition where function is significantly limited by the perfusion rate. Such a condition is easily detectid by simply noting if augmented function accompanies augmented CBF. The present data support that concept. The register between the autoperfused level and the break point in the contractile force curve is apparently the result of a regulatory mechanism matching flow to metabolism. An intervention that causes these two points to separate therefore would indicate that a direct
Downloaded from www.physiology.org/journal/ajplegacy by ${individualUser.givenNames} ${individualUser.surname} (130.070.008.131) on January 14, 2019.
6 vascular effect is present as opposed to vasomotion resulting from altered metabolism. Thus this technique should have practical value in differentiating primary from secondary vasoactivity. In summary, when co&ractile force and CBF are normalized as a percent of that occurring with autoperfusion the relationship between them becomes highly reproducible. Contractile force was found to be highly dependent on the flow rate in the range of flows below the CBF chosen by the au toregulatory mechanism. Conforce was relatively independent at versely, contractile
J. M. DOWNEY
higher flows. Finally, contractile force was found to be much more sensitive to changes in CBF than dPldt. I appreciate the contributed to this This work was Heart Association tion of Palm Beach Present address of South Alabama, Received
fine technical assistance that Mrs. Nancy Kramer study. supported by a Grant-In-Aid from the American and with *funds contributed by the Heart AssociaCounty. of J. M. Downey: Dept. of Physiology, University Mobile, Ala. 36688.
for publication
18 February
1975.
REFERENCES 1. ABEL, R. M., AND R. L. REM. Effects of coronary blood flow and perfusion pressure on left ventricular contractility in dogs. Circulation Res. 27: 961-971, 1970. 2. ARNOLD, G., F. KOSCHE, E. MEISSNER, A. NEITZERT, AND W. LOCHNER. The importance of the perfusion pressure in the coronary arteries for the contractility and the oxygen consumption of the heart. Pfhegers Arch. 299: 339-356, 1968. 3. BACHNER, M. D., F. LIOY, AND M. B. VISSCHER. Induced change in heart metabolism as a primary determinant of heart performance. Am. J. Physiol. 209: 519-531, 1965. 4. BERNE, R. M. Regulation of coronary blood flow. Physiol. Rev. 44: 1-29, 1964. 5. BERNE, R. M., R. RUBIO, J. G. DOBSON, AND R. R. CURNISH. Adenosine and adenine nucleotides as possible mediators of cardiac and skeletal muscle blood flow regulation. CircuZation. Res. 28: I-115-1-119, 1965. 6. BING, R. J. Cardiac metabolism. Physiol. Rev. 45: 171-213, 1965. 7. COHEN, M. V., J. M. DOWNEY, E. H. SONNENBLICK, AND E. S. KIRK. The effects of nitroglycerin on coronary collaterals and myocardial contractility, J, CZin. Invest. 52: 2836-2847, 1973. 8, DANIELL, H. B. Coronary flow alterations on myocardial contractility, oxygen extraction, and oxygen consumption. Am. J. Physiol. 225: 1020-1025, 1973. 9. DOWNEY, J. M., B. N. SINGH, D. F. COWAN, C. W. URSCHEL, AND E. S. KIRK. Adequacy of coronary perfusion during hemorrhagic hypotension (Abstract). Fe&r&Ott Proc. 31: 329, 1972. 10. FORMAN, R., E. S. KIRK, J. M. DOWNEY, AND E. H. SONNENBLICK, Nitroglycerin and heterogeneity of myocardial blood flow. Reduced subendocardial blood flow and ventricular contractile force. J. CZin. Invest. 52: 905-911, 1973. II. GREGG, D. E. Effect of coronary perfusion pressure on coronary flow and oxygen usage of the myocardium, CircuZation Res a 13: 497-501, 1963, 12. KAHLER, R. L., E. BRAUNWALD, L. L. KELMINSON, L. KEDES, C. A, CHIDSEY, AND S. SEGAL. Effect of alterations of coronary blood flow on the oxygen consumption of the non-working heart. CircuZutiott Res. 13: 501-510, 1963. 13. KIRK, E. S., M. E. TURBOW, H. BROOKS, AND E. H. SONNENBLICK. Myocardial performance relative to altered coronary blood flow. Myocu~diaZ BZood FLOW in. Man, edited by A. Maseri. Torino: Minerva Med., 1973. 14. KJEKSHUS, J. K., AND B. E. SOBEL. Depressed myocardial creatine phosphokinase activity following experimental myocardial infarction in rabbit. CircuZution Res. 27: 401-414, 1970, 15. MASON, P. T. Usefulness and limitations of the rate of rise of intraventricular pressure (dP/&) in the evaluation of myocardial contractility in man. Am. J. CurdioZ. 23: 516-527, 1969. 16. MAROKO, P. R., J. K. KJEKSHUS, B. E. SOBEL, T. WATANABE, J.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
W. COVELL, J. Ross, AND E. BRAUNWALD. Factors influencing infarct size following experimental coronary artery occlusions. Ci~culution 63: 67-82, 1971. MOIR, T. W. Subendocardial distribution of coronary blood flow and the effect of antianginal drugs. CircuZutiolz Res. 30: 621-627, 1972. MORGENSTERN, C,, U, H~LJES, G. ARNOLD, AND W. LOCHNER. The influence of coronary pressure and coronary flow on intracoronary blood volume and geometry of the left ventricle* Pfluegers Arch. 340: 101-111, 1973, MOSHER, P., J. Ross, P. A. MCFATE, AND R. F. SHAW. Control of coronary blood flow by an autoregulatory mechanism. CircuZution Res. 14: 250-259, 1964, ROBIE, N. W., AND W. H. NEWMAN. Effects of altered ventricular load on the Waltin-Brodie strain gauge arch. J. AppZ. Physiol. 36: 20-27, 1974. Ross, J,, JR., F. KLOCKE, G. KAISER, AND E. BRAUNWALD. Effect of alterations of coronary blood flow on the oxygen consumption of the working heart. Circulation Res. 13: X0-513, 1963. SARNOFF, S. J., R. B. CASE, P. E. WAITHE, AND J. P. ISAACS. Insufficient coronary flow and myocardial failure as a complicating factor in late hemorrhagic shock. Am. J. Physiol. 176: 439444, 1954, SARNOFF, S. J., J. P. GILMORE, N. S, SKINNER, JR., A. G. WAIT LACE, AND J. H, MITCHELL. Relation between coronary blood flow and myocardial oxygen consumption. Circulutiolz Res. 13: 522528, 1963. SCHELBERT, H. R., J. W. COVELL, J+ W. BURNS, P. R. MAROKO, AND J. ROSS, JR. Observations on factors affecting local forces in the left ventricular wall during acute myocardial ischemia. CircuZution Res. 29: 306-316, 1971. STREETER, D. D., JR., H. M. SPONTNITZ, D. P. PATEL, J. Ross, JR., AND E, H. SONNENBLICK. Fiber orientation in the canine left ventricle during diastole and systole. Circulation. Res. 24: 339347, 1969. TYBERG, J. V., L. A. YEATMAN, W. W. PARMLEY, C. W. URSCHEL, AND E. H. SONNENBLICK. Effects of hypoxia on mechanics of cardiac function, Am. J. Physiol. 218: 1780-1788, 1970, WALLACE, A. G., N. S. SKINNER, JR., AND J. H, MITCHELL. Hemodynamic determinants of the maximal rate of rise of left ventricular pressure. Am. J. PhysioZ. 205: 30-36, 1963. WEISBERG, H., L. N. KATZ, AND E, BOYD. Influence of coronary flow upon oxygen consumption and cardiac performance. Circulution Res. 13: 522-528, 1963. WEISFELDT, M. L., AND N. W. SHOCK. Effect of perfusion pressure on coronary flow and oxygen usage of non-working heart. Am. J, PhysioZ. 218: 95-101, 1970,
Downloaded from www.physiology.org/journal/ajplegacy by ${individualUser.givenNames} ${individualUser.surname} (130.070.008.131) on January 14, 2019.
in U.S.A.
Myocardial contractile force as a function of coronary blood flow JAMES M. DOWNEY Department of Physiology,
University
of South
DOWNEY, JAMES hf. MyocardiaZ contractde force as a function of coronary blood flow. Am. J, Physiol. 230(l): l-6. 1976.--The contractile force of the deep and superficial myocardial fibers was examined in the open-chest anesthetized dog as a function of coronary blood flow (CBF). When 1) dogs that failed to demonstrate coronary autoregulation were eliminated 5om the data base and 2) CBF and contractile force data were both normalized as a percent of their values when perfusion was from aortic pressure (autoperfusion), the relationship between them became very reproducible. Contractile force was highly dependent on the flow rate when the CI3F was below that chosen by autoregulation (the rate during autoperfusion). Conversely contractile force was relatively independent of flow at higher CBF. The contractile force-CBF curve thus was found to break precisely at the autoperfused CBF. When myocardial metabolism was elevated by paired electrical stimulation this relationship was unchanged. It was concluded that coronary blood flow is tightly regulated to match metabolic needs over a range of metabolic rates, paired
pacing;
garden
hose effect; autoregulation;
ischemia
iscontrolledbyanautoregulatory mechanism (4, 19). This system, which tends to maintain flow rate independent of perfusion pressure, is also sensitive to changes in cardiac metabolism (19). Evidence indicates that this control is the result of adenosine, a potent vasodilator, being released from hypoxic myocardial cells (5). Thus the heart is endowed with a control system that acts to match the delivery of oxygen and other metabolites to the heart’s needs. There is considerable evidence, however, that the blood flow rate this system provides is somewhat less than that required to optimize cardiac metabolism and thus performance. Gregg (11) reported that myocardial oxygen consumption (MVo,) could be increased by independently increasing CBF above the control level. Since then several investigators have found MVo, and/or ventricular performance to be limited by flow in what was thought to be a normal range of flows (1, 3, 8). This has not been a universal finding, however, in that others have not observed such a condition (21, 23, 28). After nearly a decade and a half of study the relationship between coronary perfusion and myocardial metabolism remains unclear. In many studies MVo, (8, 12, 21) or an index of contractility (1, 8) was plotted against flow rate per unit mass of myocardium. The resulting data show considerable variation probably because myocardial oxygen requirements vary greatly among individual experiCOR~NARYBLOODFLOW(CBF)
Florida
College of Medicine,
Tampa,
Florida
33620
ments, depending on the contractile or hemodynamic state. Furthermore, most of the reports do not indicate whether autoregulation was operative in the experimental preparations. It is well recognized that vascular tone and auturegulation often are absent in a canine heart preparation that is deteriorating. If such control is lost as a result of the conditions of the experiment, then the system will be unable to actively match flow to metabolism. Accordingly several workers do report that MVo, becomes less limited by flow in those hearts that exhibited some autoregulation (3, 29>, The present study attempts to correlate changes in myocardial performance with changes in CBF in hearts with intact autoregulatory control. To account for redistribution of blood flow across the ventricular wall, contractile force recordings from both deep and superficial myocardial fibers were used as an index of the heart’s performance. Contractile force was measured at flows both above and below the flow rate chosen by the autoregulatory mechanism and variations in myocardial metabolism among individual animals were then accounted for by expressing the independent variable, CBF, as a percent of the flow rate that autoregulation itself chose for the heart. METHODS
Mongrel dogs of either sex were anesthetized with sodium pentobarbital 5 mg/kg iv. The chests were opened in the fifth left interspace and the hearts exposed. The lungs were ventilated with 100% 0,. Blood pH was periodically monitored and ventilation rate was adjusted to maintain pH between 7.3 and 7.5. The left common coronary artery was exposed by a blunt dissection and a silk thread passed under it. A coronary cannula with the perfusion tubing attached was inserted into the subclavian artery and advanced into the left common coronary artery, where it was securely tied in place with the silk thread. The cannula was a double-lumen type (10) that withdrew arterial blood from the aorta via the outer lumen. Blood passed through the exteriorized circuit and into the coronary artery via the inner lumen. The tubing passed through the fingers of a Sigmamotor T8 peristaltic pump so that flow could be controlled when desired. Flow in the circuit was measured with a Statham K2000 electromagnetic flowmeter recording from an In Vivo Metrics cannulating-type transducer. Bypass tubing around the transducer allowed zeroing the flowmeter without interrupting coronary flow. Perfusion pressure was meas-
Downloaded from www.physiology.org/journal/ajplegacy by ${individualUser.givenNames} ${individualUser.surname} (130.070.008.131) on January 14, 2019.
2
J. M. DOWNEY
ured from a branch in the tubing distal to the pump have accompanied release. Second, when the coronary near the cannula. flow rate was abruptly changed with the pump a slow, Aortic pressure was measured from a Teflon catheter lo- to 20-s adjustment in perfusion pressure should have advanced through a femoral artery into the thoracic been seen as coronary tone changed in response to the aorta. Ventricular pressure was obtained from another new flow rate. Finally, pressure flow data collected Teflon catheter in the left carotid artery that was ad- during the experiment was later plotted. A region vanced across the aortic v ,alves. The left ventricul .ar where flow was nearly independent of perfusion prespressure was differentiated to obtain i ts derivative. A sure indicated good autoregulation. Preparations that sharp high-frequency cutoff was established by three failed to meet these criteria were terminated and no resistance-capacitance (RC) filters coupled to each other data were collected. in series, Each filter had a cutoff frequency (l/2 nRC) of The protocol was as follows: the bypass on the pump 75 Hz. was opened and the coronary artery was perfused with Separation of the contractile forces of the deep and the animal’s own aortic pressure (autoperfusion). The superficial fibers was accompli shed by orienting the pump was then engaged increasing or decreasing flow force gauge parallel to the fibers under study (10, 13). from the autoperfused rate. This was maintained until Contractile force of the superficial fibers was recorded the contractile force and perfusion pressure records from an isometric force gauge1 sewn in a base to apex reached a steady state (usually 30-60 seconds). The orientation (parallel to the superficial fibers) midway pump was then disengaged and autoperfusion was rebetween the left anterior descending and circumflex stored. The maneuver was repeated over a wide range of coronary arteries. Contractile force of the deeper fibers pump speeds. was recorded from an isometric force gauge modified by Two groups of dogs were studied. The first had gauges soldering a M-cm length of B-gauge hypodermic needle attached to record deep contractile force. After data perpendicular to each foot. This gauge was attached by were collected as described above the metabolic state of pressing the gauge onto the heart and impaling the these hearts was then augmented by paired electrical tissue with the prongs, which ensured good transmisstimulation and the procedure for data collection was sion of force from the deep fibers. The gauge was ori- repeated. Ten-volt pulses of 2 ms duration and sepaented perpendicular to the base-apex axis of the heart rated by 140-160 ms were delivered to bipolar electrodes between the left anterior descending and the circumflex sewn to the right ventricular free wall. Thus contractile arteries, with the gauge placed parallel to the deep force-CBF data were collected for both control and ones (25). paired-pacing fibers and normal to the more superficial conditions. In a second group of experiSuperficial sutures held the gauge in place. The gauges ments contractile force of the superficial fibers was studwere extended to about 130% of their initial length after ied as a function of CBF at a normal cardiac rhythm. attachment. Each gauge was tested by observing its response to 0.2 pg of intracoronary isoproterenol prior to RESULTS the co1lection of da ta. If contractile force did not increase Thirteen dogs were prepared for these experiments. by at least 100% the gauge was repositioned or the Two were eliminated because they failed to show a preparation was discarded. significant reactive hyperemia response. All of these Though it is not known precisely which fibers are contributing to the gauge’s response, it is possible to exhibiting a reactive hyperemia met the other two criteria for auturegulation as indicated in METHODS. The demonstrate clearly a differential sensitivity among pressure-flow relationship from one experiment can be events occurring at different depths in the myocardium seen in Fig. 1. The squares indicate flow with normal with this technique. In hearts with both gauges attached, Kirk et al. (13) demonstrated that topical cool- sinus rhythm and the circles indicate flow with paired ing of the surface of the heart with cold saline had no pacing. Though the curve is elevated with paired pacing myocardial metabolism both effect on the deep gauge but it typically doubled the as a result of increased response of the superficial gauge. Intravenous isoproterenol, on the other hand, caused similar increases in both gauges. Partial occlusion of the coronary artery, a maneuver that causes selective underperfusion of the subendocardium, depressed only the deep gauge. In previous studies deep gauges have been attached with deep sutures (10, 13) but this procedure is difficult and many preparations show attenuated responsiveness probably due to vascular occlusion by the sutures. The use of the rigid pins on the gauge rectifies these problems while preserving the selectivity of the gauge. In all experiments the presence of coronary autoregu50 100 150 lation was evaluated by the following criteria. First, m m/-/g CORONARY PERFUSION PRESSURE when the perfusion line was occluded for 5-10 s a reacFIG. 1. Pressure-flow curves from dog3, Table 1. CBF was altered tive hyperemia of at least 200% resting flow rate should I
l Metal-encased Box 412, Charleston,
type with movable SC. 29402).
foot
(John
A. Warren,
P. 0.
n
I
a
with a Sigmamotir pump while aortic pressure remained unchanged. Plateau region around 100 mmHg indicates autoregulation. Notice that paired pacing elevated CBF at all pressures,
Downloaded from www.physiology.org/journal/ajplegacy by ${individualUser.givenNames} ${individualUser.surname} (130.070.008.131) on January 14, 2019.
CONTRACTILE
FORCE-CORONARY
FLOW
3
RELATIONSHIP
show plateau regions where flow is nearly independent of perfusion pressure, indicating the presence of autoregulation. This graph is typical of the data. The records in Fig. 2 indicate the effects of changing the flow rate from the autoperfused level. Panels A and B reveal that increasing perfusion rate causes little change in deep contractile force. However, panek C and D reveal that contractile force is markedly compromised when flow is decreased. Quite often the pump was engaged with a setting equal to the autoperfused rate. Neither perfusion pressure nor contractile force changed when this occurred, indicating that the pulsations generated by the pump were not affecting the preparations. When deep contractile force is plotted against flow rate the graph in Fig. 3 results. The squares indicate data from one animal with a sinus rhythm. On the vertical axis deep contractile force is normalized by dividing the final value during pumping by the autoperfused or control value. The flow scale is similarly normalized by dividing the flow rate during pumping by the autoperfused rate. Thus flows are expressed as a percent of the flow rate that the heart’s regulatory mechanism establishes for itself. Because of the normalization procedure the curve must go through the index point 1, 1. At flows above 1, 1 the slope is very shallow, indicating that flow is adequate to maintain performante in this range. At flows below 1, 1, however, the curve falls off steeply, showing a region where performante is very flow dependent. The flow set by autoregulation was always on the knee of the curve. When cardiac metabolism was increased by paired pacing the points indicated by the circles in Fig. 3 resulted. Though paired pacing caused the regulatory mechanism to choose a new perfusion rate, as indicated by the pressure flow curve in Fig. 1 (circles), it did not change the relationship between contractile force and 8
A
c
CBF. The circles in Fig. 3 lie over the squares. Thus though autoperfused flow increased with paired pacing a flow was still provided that was at the knee of the contractile force-flow curve. This was a consistent finding in all five dogs in group Z. Figure 4 shows the data from all dogs plotted in the same manner as for Fig. 3. The open figures indicate normal cardiac rhythm and the solid figures indicate paired pacing. Though some variation is seen in the region below 1, 1 this seems to be the result of variability among animals since points taken during paired pacing tend to be very close to corresponding points with normal rhythm. The maximum positive derivative of the ventricular pressure (dP/&) is plotted against CBF in Fig. 5. Again, both variables are normalized as a percent of the autoperfused levels. These data were collected simultaneously with those in Fig. 3. Though changes in dP/&
l
l
SINUS
RHYTHM
PAIRED
PACE
m
d 1 5
1 IO CORONARY
1 15 BLOOD
1 20
FLOW
3. Deep contractile force plotted against coronary blood flow rate. Both were normalized by dividing steady-state value by value during autoperfusion. Curve therefore is forced through point 1, 1. Squares represent data taken while the heart beat at its own intrinsic rhythm and circles those taken during paired electrical stimulation. Data from dog 3, Table 1. FIG.
D
DCF
2oo
C8Fmf/
.__._--.
in
IT-l
o200,
AOP 100-B
mmkfg Illllllllwcz----.-_
k
45SECi
FIG. 2. Records in which CBF was either augmented (A and B) or reduced (C and D> from autoperfused level by engaging the pump. DCF, deep contractile force; AOP, aortic pressure; CP, mean coronary perfusion pressure; VP, ventricular pressure; and dP/&, derivative of ventricular pressure. Notice that DCF was affected more by reductions in CBF than increases. Slow adjustments in coronary perfusion pressure after abrupt changes in CBF are result of coronary autoregulation.
Downloaded from www.physiology.org/journal/ajplegacy by ${individualUser.givenNames} ${individualUser.surname} (130.070.008.131) on January 14, 2019.
4
J. M. DOWNEY
be maintained at the expense of the deeper fibers when tutal CBF is reduced (14). Thus the effects of underperfusion are first evidenced at the subendocardium (10, 13). The result is that the deep contractile force should be more sensitive to changes in total CBF. An additional group of dogs was studied and superficial contractile force was monitored to see if it really was less sensitive to changes in the CBF rate. In Fig. 6 contractile force-CBF data from six dogs with superficial gauges are plotted on the same axis as the control data from the five dogs in group I that had deep gauges. Though the data are very similar, three of the preparations failed to demonstrate an appreciable (5%) change in contractile force when the flow rate was reduced by 15% or greater from the control rate. Such a condition was not observed in group Z.
were usually in the same direction as the contractile force the changes were never as pronounced. Detectable changes in dP/dt often did not occur with small deviations in CBF away from the autoperfused level and thus a break point comparable to that seen in Fig. 3 is not easily demonstrated in the curves. Hemodynamic data for these five animals appear in Table 1. Deep contractile force was chosen for these experiments because flow in the subepicardial layers tends to
DISCUSSION
5
IO CORONARY
15
BLOOD
Unlike most of the previous investigations in which an index of cardiac metabolism is plotted against CBF the present study finds the relationship to be very reproducible* This is in part attributable to selecting only those animals with intact coronary regulation and then normalizing the data as a percent of that occurring with This normalization procedure for CBF autoperfusion. accounts for individual variations in cardiac metabolism among animals.
20
FLOW
FIG. 4. Composite of all deep contractile force data from 5 experiments. Data from individual dogs are identified by symbol shape. Open symbols correspond to data taken during sinus rhythm and solid symbols to those taken during paired pacing. Normalization is same as in Fig. 3.
n
l 0 @DEEP
A.SUPEHFICIAL b
I
I
I
5
IO
IS
CORONARY
BLOOD
I
?O
FLOW
6. Contractile force of deep fibers (circles) and superficial fibers (triangles) is plotted against coronary blood flow. Normalization is same as for Fig. 3. Curves are very similar; however, deep points tend to be lower than corresponding superficial points in range of flows below 1. FIG.
FIG. 5. Plot of maximum derivative of ventricular against coronary blood flow. Normalization again fused level. Data from dog 3, Table 1.
TABLE Dog No.
1. Hemodynamic Condition
--
pressure (dP/&) is against autoper-
data for ---5 group I dogs ----------Aortic
Pressure, mmHg ---p--p-p
dP/&,
-------mmHg/s
Relative DCF, mm
CBF,
mllmin
--
Heart Rate, beatslmin
LV wt, g
Dog Wt, kg
1
Control PP
90165 115180
1,200 2,250
8 13
64 126
160 124
98
18
2
Control PP
100172 1255180
NA NA
12 25
88 134
144 120
95
20
3
Control PP
125195 120192
1,750 2,250
12 20
84 105
164 152
109
20
4
Control PP
90165 110150
1,750 2,000
11 16
75 126
128 116
70
18
5
Control PP
10 22
85 108
120 112
82
20
135185 2,250 105180 3,000 ------PP --PP, paired pacing; dP/&, maximal positive derivative of ventricular deep contractile force during autoperfusion; values are in millimeters
pressure; excursion
-LV, left ventricle of recording pen,
-(including septum). Relative as gauge was not calibrated.
DCF
refers
to
Downloaded from www.physiology.org/journal/ajplegacy by ${individualUser.givenNames} ${individualUser.surname} (130.070.008.131) on January 14, 2019.
CONTRACTILE
FORCE-CORONARY
FLOW
RELATIONSHIP
When coronary flow is altered the distribution of that blood flow is also altered. At low perfusion pressures flow to the subendocardium is reduced to a greater extent than that at the subepicardium due to the high subendocardial tissue pressure (17). Therefore, depressed contractility due to hypoxia should appear first and be most pronounced at the subendocardium. For this reason it was thought that deep contractile force would probably be superior to other indices of contractility for this study. When CBF is reduced by lowering coronary perfusion pressure, redistributions cause deep contractile force to become depressed at consistently higher pressures than superficial force (10, 13). However, in the present experiment the two curves were very similar when plotted against CBF (Fig. 6). This is because the pressure-CBF curve experiences a plateau in the region of the autoperfused flow (Fig. 1). Thus large reductions in perfusion pressure result in a relatively small change in CBF, which tends to magnify the differences between the contractile force-pressure curves. In terms of actual flow values, then, superficial contractile force seems to have a relationship with CBF that is not as different from deep contractile force as was originally predicted. When the contractile force data were compared to measurements of dP/& the latter index was much less sensitive to changes in CBF. Changes in both preload and in afterload are known to affect dP/& (15, 27) and contractile force (ZO), yet at coronary flow values above about two-thirds of the autoperfused rate neither preload or afterload was appreciably altered in these experiments. Thus, both of these measurements should be valid. A possible explanation for this lack of sensitivity is that dP/& reflects the average contractile state across the ventricular wall and is not particularly sensitive to ischemia that is limited to the subendocardium. If the lack of sensitivity of dP/& to changes in CBF was the result of flow redistributions between layers, then the dP/& response should have been intermediate between contractile force at the subendocardium, where hypoxia first appears, and contractile force at the subepicardium, where hypoxia is last to appear. This was not the case, however. A more likely cause of the discrepancy between the two techniques is that velocity-related indices do not appear to be as sensitive to hypoxic depression as those related ti force (26). These data indicate that coronary blood flow is accurately regulated to match myocardial metabolism over a range of metabolic rates. Reductions in CBF from the autoperfused level proportionately decreased mechanical activity, whereas increases in CBF above this point had a minimal effect. Such a discontinuous relationship between MVo, (23, 28) or ventricular performance (8, 23) and CBF with a plateau region in the range of high CBF values has been seen by others and agrees with the present data. In all the experiments, however, contractile force was found to be slightly limited by blood flow at the flow rate provided by autoregulation. Doubling the flow rate over the autoperfused level typically raised contractile force by about 15%. Current evidence points toward adenosine as the mediator of coronary autorealation (4, 5).
5 The nucleoside is thought to be released from hypoxic myocardium. Its powerful vasodilatory action increases blood flow and restores oxygen balance to the tissue. Such a system requires that some adenosine must be released continuously since coronary tone is normally midway between a fully constricted and dilated state. The slight limitation to myocardial performance that coronary perfusion imposes on the heart at the autoperfused level may reflect a normal condition of slight hypoxia that is required to effect this continuous adenosine release. Alternatively this apparent limitation may be unrelated to the delivery of metabolites but rather purely a mechanical effect. Pressure in the coronary vessels, some have proposed, acts to stretch the myocardial fibers, increasing their preload through the socalled “garden hose” effect (2, 18). Such a mechanism may be operative here. Also an increased turgidity of the myocardium with elevated perfusion pressure could result in better mechanical coupling of surrounding tissue to the force gauge, thus slightly increasing its recorded force. Whatever the mechanism, the magnitude of this increase is indeed small. In terms of efficiency as a control system the match between blood flow requirements and supply in the heart appears to be optimal. The regulatory mechanism keeps perfusion at the knee of the contractile force-CBF curve. Though performance could be further increased slightly by a large increase in coronary flow the cost in terms of the additional increment in cardiac output required to supply this flow would be appreciable. On the other hand any savings in cardiac output afforded by reducing coronary flow over that determined by the regulatory mechanism would clearly be offset by greatly impaired performance. Since the contractile force-CBF relationship was essentially unchanged when metabolism was augmented by paired pacing, it would appear that this efficiency is maintained over a range of metabolic states. The difference between the slope of the contractile force-CBF curve above the autoperfused point as opposed to below it is so great that one is tempted to equate the region below that point with ischemia. Attenuated function has been used to indicate the onset of ischemia (7, 10, 24) but this index is only valid when all other factors that may decrease fu.nction can be ruled out. Lactate production (6) characterizes ischemic myocardium but the low venous lactate titer associated with mild ischemia or in cases where it is very localized is not always detectable. Enzymatic (14) and electrocardiographic (16) changes associated with ischemia are slow to develop and do not lend themselves to many experimental designs. We (9) as well as others (22) therefore have found it useful to interpret ischemia as a condition where function is significantly limited by the perfusion rate. Such a condition is easily detectid by simply noting if augmented function accompanies augmented CBF. The present data support that concept. The register between the autoperfused level and the break point in the contractile force curve is apparently the result of a regulatory mechanism matching flow to metabolism. An intervention that causes these two points to separate therefore would indicate that a direct
Downloaded from www.physiology.org/journal/ajplegacy by ${individualUser.givenNames} ${individualUser.surname} (130.070.008.131) on January 14, 2019.
6 vascular effect is present as opposed to vasomotion resulting from altered metabolism. Thus this technique should have practical value in differentiating primary from secondary vasoactivity. In summary, when co&ractile force and CBF are normalized as a percent of that occurring with autoperfusion the relationship between them becomes highly reproducible. Contractile force was found to be highly dependent on the flow rate in the range of flows below the CBF chosen by the au toregulatory mechanism. Conforce was relatively independent at versely, contractile
J. M. DOWNEY
higher flows. Finally, contractile force was found to be much more sensitive to changes in CBF than dPldt. I appreciate the contributed to this This work was Heart Association tion of Palm Beach Present address of South Alabama, Received
fine technical assistance that Mrs. Nancy Kramer study. supported by a Grant-In-Aid from the American and with *funds contributed by the Heart AssociaCounty. of J. M. Downey: Dept. of Physiology, University Mobile, Ala. 36688.
for publication
18 February
1975.
REFERENCES 1. ABEL, R. M., AND R. L. REM. Effects of coronary blood flow and perfusion pressure on left ventricular contractility in dogs. Circulation Res. 27: 961-971, 1970. 2. ARNOLD, G., F. KOSCHE, E. MEISSNER, A. NEITZERT, AND W. LOCHNER. The importance of the perfusion pressure in the coronary arteries for the contractility and the oxygen consumption of the heart. Pfhegers Arch. 299: 339-356, 1968. 3. BACHNER, M. D., F. LIOY, AND M. B. VISSCHER. Induced change in heart metabolism as a primary determinant of heart performance. Am. J. Physiol. 209: 519-531, 1965. 4. BERNE, R. M. Regulation of coronary blood flow. Physiol. Rev. 44: 1-29, 1964. 5. BERNE, R. M., R. RUBIO, J. G. DOBSON, AND R. R. CURNISH. Adenosine and adenine nucleotides as possible mediators of cardiac and skeletal muscle blood flow regulation. CircuZation. Res. 28: I-115-1-119, 1965. 6. BING, R. J. Cardiac metabolism. Physiol. Rev. 45: 171-213, 1965. 7. COHEN, M. V., J. M. DOWNEY, E. H. SONNENBLICK, AND E. S. KIRK. The effects of nitroglycerin on coronary collaterals and myocardial contractility, J, CZin. Invest. 52: 2836-2847, 1973. 8, DANIELL, H. B. Coronary flow alterations on myocardial contractility, oxygen extraction, and oxygen consumption. Am. J. Physiol. 225: 1020-1025, 1973. 9. DOWNEY, J. M., B. N. SINGH, D. F. COWAN, C. W. URSCHEL, AND E. S. KIRK. Adequacy of coronary perfusion during hemorrhagic hypotension (Abstract). Fe&r&Ott Proc. 31: 329, 1972. 10. FORMAN, R., E. S. KIRK, J. M. DOWNEY, AND E. H. SONNENBLICK, Nitroglycerin and heterogeneity of myocardial blood flow. Reduced subendocardial blood flow and ventricular contractile force. J. CZin. Invest. 52: 905-911, 1973. II. GREGG, D. E. Effect of coronary perfusion pressure on coronary flow and oxygen usage of the myocardium, CircuZation Res a 13: 497-501, 1963, 12. KAHLER, R. L., E. BRAUNWALD, L. L. KELMINSON, L. KEDES, C. A, CHIDSEY, AND S. SEGAL. Effect of alterations of coronary blood flow on the oxygen consumption of the non-working heart. CircuZutiott Res. 13: 501-510, 1963. 13. KIRK, E. S., M. E. TURBOW, H. BROOKS, AND E. H. SONNENBLICK. Myocardial performance relative to altered coronary blood flow. Myocu~diaZ BZood FLOW in. Man, edited by A. Maseri. Torino: Minerva Med., 1973. 14. KJEKSHUS, J. K., AND B. E. SOBEL. Depressed myocardial creatine phosphokinase activity following experimental myocardial infarction in rabbit. CircuZution Res. 27: 401-414, 1970, 15. MASON, P. T. Usefulness and limitations of the rate of rise of intraventricular pressure (dP/&) in the evaluation of myocardial contractility in man. Am. J. CurdioZ. 23: 516-527, 1969. 16. MAROKO, P. R., J. K. KJEKSHUS, B. E. SOBEL, T. WATANABE, J.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
W. COVELL, J. Ross, AND E. BRAUNWALD. Factors influencing infarct size following experimental coronary artery occlusions. Ci~culution 63: 67-82, 1971. MOIR, T. W. Subendocardial distribution of coronary blood flow and the effect of antianginal drugs. CircuZutiolz Res. 30: 621-627, 1972. MORGENSTERN, C,, U, H~LJES, G. ARNOLD, AND W. LOCHNER. The influence of coronary pressure and coronary flow on intracoronary blood volume and geometry of the left ventricle* Pfluegers Arch. 340: 101-111, 1973, MOSHER, P., J. Ross, P. A. MCFATE, AND R. F. SHAW. Control of coronary blood flow by an autoregulatory mechanism. CircuZution Res. 14: 250-259, 1964, ROBIE, N. W., AND W. H. NEWMAN. Effects of altered ventricular load on the Waltin-Brodie strain gauge arch. J. AppZ. Physiol. 36: 20-27, 1974. Ross, J,, JR., F. KLOCKE, G. KAISER, AND E. BRAUNWALD. Effect of alterations of coronary blood flow on the oxygen consumption of the working heart. Circulation Res. 13: X0-513, 1963. SARNOFF, S. J., R. B. CASE, P. E. WAITHE, AND J. P. ISAACS. Insufficient coronary flow and myocardial failure as a complicating factor in late hemorrhagic shock. Am. J. Physiol. 176: 439444, 1954, SARNOFF, S. J., J. P. GILMORE, N. S, SKINNER, JR., A. G. WAIT LACE, AND J. H, MITCHELL. Relation between coronary blood flow and myocardial oxygen consumption. Circulutiolz Res. 13: 522528, 1963. SCHELBERT, H. R., J. W. COVELL, J+ W. BURNS, P. R. MAROKO, AND J. ROSS, JR. Observations on factors affecting local forces in the left ventricular wall during acute myocardial ischemia. CircuZution Res. 29: 306-316, 1971. STREETER, D. D., JR., H. M. SPONTNITZ, D. P. PATEL, J. Ross, JR., AND E, H. SONNENBLICK. Fiber orientation in the canine left ventricle during diastole and systole. Circulation. Res. 24: 339347, 1969. TYBERG, J. V., L. A. YEATMAN, W. W. PARMLEY, C. W. URSCHEL, AND E. H. SONNENBLICK. Effects of hypoxia on mechanics of cardiac function, Am. J. Physiol. 218: 1780-1788, 1970, WALLACE, A. G., N. S. SKINNER, JR., AND J. H, MITCHELL. Hemodynamic determinants of the maximal rate of rise of left ventricular pressure. Am. J. PhysioZ. 205: 30-36, 1963. WEISBERG, H., L. N. KATZ, AND E, BOYD. Influence of coronary flow upon oxygen consumption and cardiac performance. Circulution Res. 13: 522-528, 1963. WEISFELDT, M. L., AND N. W. SHOCK. Effect of perfusion pressure on coronary flow and oxygen usage of non-working heart. Am. J, PhysioZ. 218: 95-101, 1970,
Downloaded from www.physiology.org/journal/ajplegacy by ${individualUser.givenNames} ${individualUser.surname} (130.070.008.131) on January 14, 2019.
