- Email: [email protected]
Comparative effects of nitroglycerin and nifedipine on myocardial blood flow and contraction during flow-limiting coronary stenosis in the dog
Comparative Effects of Nitroglycerin and Nifedipine on Myocardial Blood Flow and Contraction During Flow-Limiting Coronary Stenosis in the Dog WILLIAM
S. WEINTRAUB,
SATOSHI
AKIZUKI,
JAI
f3. AGARWAL,
MD,
MD,
FACC
MONTY M. BODENHEIMER, MD, VIDYA S. BANKA, MD, FACC RICHARD
H. HELFANT,
Philadelphia,
FACC
MD
MD,
FACC
FACC
Pennsylvania
From The Mid-Atlantic Heart and Vascular Institute, Presbyterian-University of Pennsylvania Medical Center and the School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania. This study was supported in part by Grant HL2654501 from The National Institutes of Health, Bethesda, Maryland. Manuscript received October 5, 1981; revised manuscript received February 1, 1982, accepted February 11, 1982. Address for reprints: William S. Weintraub, MD, Assistant Professor of Medicine, Division of Cardiology, Presbyterian-University of Pennsylvania Medical Center, 51 North 39th Street, Philadelphia, Pennsylvania 19 104.
Both nifedipine and nitroglycerin are used to treat angina pectoris. The comparative effects of these agents on myocardial blood flow and contraction in the setting of flow-limiting coronary stenosis are poorly understood. Thus 24 open chest dogs underwent carotid to left anterior descending coronary arterial perfusion with coronary flow probe and perfusion pressure monitoring. Segment length was measured with ultrasonic crystals in the subendocardial ischemic and nonischemic zones. Myocardial blood flow was measured with radioactive microspheres. Partial coronary occlusion was performed to attain a diastolic perfusion pressure of 40 mm Hg. Twelve dogs received intravenous nifedipine, 3 pg/kg per min, and 12 received intravenous nitroglycerin to reduce aortic pressure by 20 mm Hg. Partial occlusion resulted in a slight but significant decrease in segment shortening in the ischemic zone. Neither nitroglycerin nor nifedipine affected shortening in the ischemic zone. After occlusion, blood flow decreased in the subendocardial ischemic zone but was unchanged in the subepicardium. Nifedipine increased subendocardial blood flow in the nonischemic zone and decreased it in the ischemic zone but caused no change in subepicardial flow in the ischemic zone. In contrast, nitroglycerin decreased subendocardial and subepicardial blood flow in both the ischemic and nonischemic zones. In the setting of coronary stenosis, different classes of vasodilators may have varying effects on myocardial blood flow, suggesting different sites and mechanisms of action. In addition, segment function may not always reflect changes in myocardial blood flow.
Although nifedipine is a potentially important new agent for the treatment of myocardial ischemia,l nitroglycerin remains the most widely used drug for this purpose. However, there is little comparative information available on how these two agents affect the relation between myocardial blood flow and segment shortening distal to a flow-limiting coronary stenosis. This study was therefore designed to determine whether therapeutic doses of intravenous nifedipine and nitroglycerin increase or decrease myocardial blood flow distal to a fixed stenosis and to determine how the combination of effects on preload, afterload and blood flow to the ischemic zone affect myocardial shortening. Methods Experimental procedure: Twenty-four dogs were anesthetized with pentobarbital and placed on a Harvard respirator. Arterial blood gases were monitored with partial pressure of oxygen (POz) maintained between 80 and 100 mm Hg and hydrogen ion concentration (pH) between 7.36 and 7.44. A 7F catheter was passed into the right femoral artery and advanced to the thoracic aorta for aterial pressure monitoring. A stiff catheter, internal diameter 2 mm, was passed into the left femoral artery and advanced to the abdominal aorta for blood withdrawal. A catheter was also passed into the right femoral vein for intravenous infusions.
August 1982
The American
Journal of CARDIOLDGY
Volume 50
281
NITROGLYCERIN
AND NIFEDIPINE IN CORONARY
STENOSIS-WEINTRAUB
A thoracotomy was performed in the fifth left intercostal space and the heart supported in a pericardial cradle. A 7F catheter was passed into the left atrial appendage for injection of microspheres (Fig. 1). A 14 gauge stiff catheter, 10 cm long, was passed into the left ventricle through the apex to record left ventricular pressure and its first derivative (dP/dt). A 5 to 10 cm segment of the left common carotid artery was exposed. After administration of heparin, the left anterior descending coronary artery was isolated at the level of the first or second diagonal branch and ligated and an arteriotomy performed. A 14 gauge steel cannula was inserted into the left anterior descending artery, tied securely and continuously perfused from the left common carotid artery through plastic tubing with a minimal internal diameter of 2 mm. Coronary perfusion pressure was monitored with a strain gauge (Statham, P23 DB) and coronary flow with an electromagnetic flow probe with an internal diameter of 2 mm (Micron Instruments). The time from ligation of the left anterior descending artery to establishment of cannula perfusion averaged 50 seconds, and all dogs in which cannulation took more than 120 seconds were excluded from further study. During cannula perfusion of the left anterior descending coronary artery, pressures at the tip of the cannula and in the
tubing were identical. In addition, diastolic perfusion pressure in the tubing was identical to aortic perfusion pressure at flows of up to 100 ml/min. The cannula system was also tested in vitro with warm blood pumped through the system at rates
FIGURE 1. Experimental preparation showing the carotid to left anterior descending (LAD) arterial cannulation system, the electromagnetic flow probe and the screw clamp constrictor. Note that perfusion pressure (Perf. P) is measured distal to the constrictor. The potentially ischemic zone (hatched area) is colored by the Evans blue dye. The ring of myocardium from which tissue samples are taken for blood flow analysis is noted by the broken lines. Note that both nonischemic zone (NZ) and ischemic zone (IZ) crystals are within this ring. Nonischemic and ischemic zone crystals and samples of tissue for blood flow were at least 1 cm from the border between nonischemic and ischemic tissues. Ao = aorta; AoP = aortic pressure; CBF = coronary blood flow probe: CXL = left circumflex coronary artery; LAcath = left atrial catheter; LVP = left ventricular pressure; PA = pulmonary artery.
282
August 1882
The American
Journal of CARDIOLOGY
ET AL.
from 1.1 to 46 ml/min with pressures monitored in the proximal cannulation system and at the cannula tip. No pressure gradient was observed. The presence of coronary reserve was determined by using a 10 second period of total occlusion followed by reperfusion. All dogs with less than 100 percent reactive hyperemia were rejected as having either inadequate coronary reserve or stenosis in the cannula system. Placement of ultrasonic crystals: Two pairs of 2 mm ultrasonic crystals were placed in the subendocardium, one pair each in the nonischemic and ischemic zones (Fig. 1). The crystals in each pair were between 8 and 15 mm apart. The crystals were inserted into the inner third of the myocardium through small stab wounds perpendicular to the long axis of the left ventricle. The motion of the ultrasonic crystals was monitored with an ultrasonic imaging circuit (Schuessler and Associates). Microsphere technique: Microspheres (diameter 9 p) labeled with iodine-125 (lzsI), cerium-141 (i*lCe), strontium-85 (85Sr) or scandium-46 (4sS~) (3M Company) were used to measure myocardial blood flow. Microspheres were suspended in saline solution with a drop of Tween-SO@, agitated in an ultrasonic bath for at least 15 minutes and shaken in a vortex whirler before injection. The left atrium was injected with 2 to 3 million microspheres in 8 ml of saline solution over 15 to 20 seconds then given a 4 ml flush of microsphere-free saline solution. Starting before the injection of microspheres, blood was withdrawn from a femoral artery at 7.75 ml/min with a Harvard pump in order to obtain a reference blood flow. Blood withdrawal was continued for 1 minute after completion of the saline flush. Experimental protocol: The preparation was allowed to stabilize for at least 15 minutes after cannulation. Partial coronary occlusion was performed by obstructing the cannulation tubing with a screw clamp device to a minimal diastolic perfusion pressure of 40 mm Hg. The preparation was allowed to stabilize for 5 minutes and a first set of microspheres of one isotope was given. The dogs were then randomized into two groups. In group 1,12 dogs were given continuous intravenous nifedipine, 3 pg/kg per min, and in group 2,12 dogs were given continuous intravenous nitroglycerin to decrease aortic systolic blood pressure by 15 to 20 mm Hg but not to less than 90 mm Hg. Fifteen minutes after the infusion began, a second set of microspheres of a different isotope was given. By giving two sets of microspheres of different energy levels the blood flow before and after drug administration was determined. Tissue preparation: After the second microsphere injection, Evans blue dye was injected into the coronary cannula by hand-held syringe with sufficient pressure to stain the myocardium but insufficient to fill the intercoronary collateral vessels. The dog was killed and the position of the crystals noted and then removed. The heart was washed, dried, stuffed with gauze and wrapped with industrial strength aluminum foil and frozen. The heart was sectioned while still frozen to facilitate accurate cutting. A 2 to 3 cm wide transmural ring of myocardium was cut with the path of the cut perpendicular to the border between blue-stained ischemic tissue and unstained nonischemic tissue (Fig. 1). A section of myocardium was taken from the remote nonischemic zone. The ring was cut at the blue border with care taken to keep all blue-stained tissue on the ischemic side. A 1 cm sample of myocardium was taken on the ischemic (blue) side labeled border ischemic, followed by a 1 cm central ischemic sample. In all dogs the central ischemic sample contained the crystal tracts. All samples were divided into subendocardial, mid myocardial and subepicardial third.
Volume 50
NITROGLYCERIN
ECG
AND NIFEDIPINE IN CORONARY
STENOSIS-WEINTRAUB
ET AL.
-+--
Segment
length FIGURE 2. Sample recordings showing the method for calculation of shortening. The lefl panel, control period (c); right panel, after an intervention. The tracing from the top down are the electrocardiogram, the segment length representing the distance between the pair of crystals, the straight line of 0 mm crystal separation, and the first derivative of left ventricular pressure with respect to time (dP/dt) and left ventricular pressure (LVP). End-diastolic length (EDL) and end-systolic length (ESL) are measured from the 0 line to the segment length line, and these lengths are timed from dP/dt according to the method of Theroux et al.3 Percent shortening (% Ai_) is calculated as shown. Shortening (N% AL) and end-diastolic length (NEDL) may then be normalized to the fraction of the control period as shown.
,,,,,A LVP
J--I EDL,% A L c=
ESL,
EDL,
EDL x 100
%AL =
NEDL
The tissue samples were weighed and counted together with the blood samples, pure isotope standards and a background tube in a Beckman 8,000 well gamma counter for 10 minutes each. Myocardial blood flow was then determined by the method of Heymann et al2 A measure of resistance was determined from the ratio of diastolic perfusion pressure to coronary blood flow. Data recording and analysis: Hemodynamic data and myocardial shortening were obtained on an Electronics for Medicine VR-16 recorder. Myocardial shortening and enddiastolic length were calculated by the method of Theroux et al.” and then normalized t,o the control period (Fig. 2). All data are expressed as mean f standard deviation. Differences for any variable between subgroups of the two groups described earlier were analyzed by a two way analysis of variance, randomized block design.4 Differences between the two groups for any variable were analyzed by a one way analysis of variance.”
-
ESL
EDL
x 100 % A L
EDL = F,
N%AL
=%a~
Results Hemodynamics and coronary flow: Aortic, systolic and diastolic pressures were not different in the nifedipine and nitroglycerin treated groups before drug infusion (Table I). Systolic and diastolic pressures decreased after drug infusion in both groups. The decrease in pressure with nifedipine was greater than with nitroglycerin, but the difference between groups was significant only for diastolic pressure (p
TABLE I Hemodynamic Data Nifedipine
Control LVEDP (mm Hg) AoP (mm Hg) Systolic Diastolic Heart rate Rate-pressure product (mm Hg-beats/min)
7f3 11af22 a7 f 22 130 f 32 17,200 f7,500
p
Partial Occlusion a*4
Nitroglycerin
Nifedipine
12 16 123 f 30 13,600 f4,300 aof
Control 3f2
855
109 f
C
91 f 12’ 55 f 14’ 119f26 11,600’ f4.600
121 f 13 93f 13 136 f 25 16,500 f3,600
Partial Occlusion 3f2 119f 16 a7f 19 134 f 26 16,000 f3,aoo
Nitroglycerin 3f2 101 f 13’ 76 f 16’ 136 f 26 13,800* f3,300
l
end-diastolic pressure.
August 1982
The American Journal of CARDIOLOGY
Volume 50
283
NITROGLYCERIN AND NIFEDIPINE IN CORONARY STENOSIS-WEINTRAUB ET AL.
TABLE
II
Perfusion Pressure, Coronary Flow and Coronary Resistance Nifedipine Partial Occlusion
Nifedipine
87 f 40f
13‘ 1”
69 f 12’ 31 f 2’
25f9” 70 f 16’
20 f 8’ 56f ll+
1.9 f 0.9”
1.9 f 0.6
Control Pressure (mm Hg) Systolic Diastolic Blood flow (mbmin) Coronarv Normaked Resistance (mm Hg-minImI) l
115 f 25 86 f 26 33 f 9 100 2.8 f
1.0
Nitroglycerin Partial Occlusion
Control 120 f 92 f
13 13
27f 10 100 3.9 f
1.6
90f7’ 40fO’
77 f 9’ 35 f 4’
21 f 11” 76 f 21‘
16 f 8+ 61 f 18’
2.4 f
TABLE
2.6 f
1.2
nonischemic zone, nifedipine caused a significant increase in flow in the subepicardium (p
Ill
Regional Myocardlal
Blood Flow (ml/g per min) Nitroglycerin
Nifedipine Partial Occlusion lschemic zone Subendocardial blood flow Subepicardial blood flow Nonisdhemic zone Subendocardial blood flow Subepicardial blood flow
284
1.4’
p
nifedipine and by 2,200 f 1,200 units after nitroglycerin. Hemodynamic measurements in the cannulation system reflecting flow and perfusion pressure in the subserved vasculature (Table II) demonstrated no difference in pressure between the aorta and cannulation system before coronary occlusion. Partial occlusion decreased perfusion pressure to a minimal diastolic level of 40 mm Hg. Coronary flow decreased slightly but significantly in both groups. After drug infusion, diastolic and systolic perfusion pressures decreased in both groups. Coronary flow measured with the electromagnetic flow probe also decreased in both groups after drug infusion. Resistance in the perfused bed decreased significantly in both groups after partial occlusion, but neither nifedipine nor nitroglycerin significantly changed the total resistance in the perfused bed. Regional myocardial blood flow (Table III, Fig. 3 and 4): Before drug infusion, subendocardial blood flow in the ischemic zone was less than both subepicardial flow and less than subendocardial flow in the nonischemic zone in both drug treatment groups (p
l
Nitroglycerin
Nifedioine
Nitroolycerin
0.66 f 0.30 1.21 f 0.46
0.42 f 0.28’ 1.24 f 0.60
0.78 f 0.29 1.17 f 0.32
0.57 f 0.23‘ 0.92 f 0.23’
1.26 f 0.44+ 1.31 f 0.73
1.52 f 2.12 f
1.14 f 0.34+ 1.12 f 0.40
0.84 f 0.17’t 0.83 f 0.27’
l.Ol+ 1.02*+
p
August 1982
Partial Occlusion
The American Journal of CARDIOLOGY
Volume 50
NITROGLYCERIN AND NIFEDIPINE IN CORONARY STENOSIS
-
-
lschemic
-
Non-ischemic
--
WEINTRAUB ET AL.
lschemic -
Non-ischemic
2.2 2 .-5 a 2
2.0 t
/ /
1.8
/ I
1.6
/'
2
3 e
5
0.6
00
0.4
2
0.2
,e
1’ _~Y-c
1.2
0.8
z
/
1.4
;
/
Endo
’ Epi “0 0 z =
\
0.6
Y*
Endo
0.4
L
--,
$
kEndo
r’-
* pco.01
i
0.2
Postocclusion
p
Postocclusion
I
I
l
t
Postnifedipine
Postnitroglycerin
FIGURE 4. Influence of nitroglycerin on myocardial blood flow after coronary stenosis. Regional myocardial blood flow measured with microspheres is displayed before (left) and after (right) intravenous administration of nitroglycerin. Nitroglycerin resulted in a decrease in myocardial blood flow to the subendocardium and subepicardium in both the nonischemic and ischemic zones. Abbreviations as in Figure 3.
FIGURE 3. Influence of nifedipine on myocardial blood flow after coronary stenosis. Regional myocardial blood flow measured with microspheres is displayed before (left) and after (rlght) intravenous administration of nifedipine. Nifedipine increased flow to the nonischemic subepicardium (Epi) but did not affect nonischemic subendocardial (Endo) flow. In contrast, nifedipine decreased blood flow to the ischemic subendocardium blood flow but did not affect flow to the ischemic subepicardium.
glycerin have both similarities and some differences. Both nifedipine and nitroglycerin decreased systemic blood pressure as well as perfusion pressure distal to the fixed coronary stenosis. In addition, both agents reduced myocardial oxygen demand to a similar degree as measured by rate-pressure product. Despite these relatively similar effects on systemic and perfusion pressures these two agents had strikingly different effects on myocardial blood flow. Nifedipine increased flow to the nonischemic zone, although this was significant only in the subepicardium. It decreased flow in the ischemic zone of the subendocardium but had no effect on this zone in the subepicardium. In contrast, nitroglycerin decreased myocardial blood flow to the su-
emit zone (p CO.01). Nitroglycerin did not affect segment shortening in either the ischemic or the nonischemit zone. Nitroglycerin but not nifedipine decreased end-diastolic length slightly in both the ischemic and’ the nonischemic zones (p <0.05). Discussion Myocardial blood flow: This study casts new light on the nature of the coronary circulation and the influence of vasodilators. We have shown that mildly to moderately hypotensive doses of nifedipine and nitro-
TABLE IV End-Diastolic Length and Segment Shortening Nitroglycerin
Control lschemic EDL (mm) NEDL % AL N%AL Nonischemic EEbimm) % AL N%AL
8.9 f 16.7:
1.8 6.9
1 9.7 f 2.1 15.2:
1
5.2
Partial Occlusion
Nifedipine
Control
9.2 1.03 13.6 0.87
f f f f
2.0t 0.04 6.8+ O.ll+
9.2 1.04 14.0 0.88
f 1.9 f 0.02 f 7.4 f 0.15
11.9 f 2.5
9.8 1.01 15.4 1.03
f f f f
2.1 0.04 4.9 0.10
10.0 1.03 17.3 1.15
f f f f
11.2 f 2.4
2.3 0.05 5.V 0.19’
6.0
16.3: 1
5.3
16.5: 1
p CO.05 compared with value to left. t p CO.01 compared with value to left. % AL = percent segment shortening; N%L = fraction of control period shortening: EDL = enddiastolic enddiastolic length.
Partial Occlusion
Nitroglycerin
11.9 1.00 15.6 0.95
f f f f
2.5 0.02 5.8+ 0.05t
11.6 0.98 15.1 0.94
f f f f
2.5’ 0.05 5.0 0.06
11.2 1.00 17.0 1.02
f f f f
2.3 0.02 5.5 0.05
10.9 0.98 16.8 1.02
f f f f
2.3’ 0.04’ 5.5 0.05
l
August 1982
length; NEDL = fraction of control
The American Journal of CARDIOLOGY
Volume 50
peri&
285
NITROGLYCERIN
AND NIFEDIPINE IN CORONARY
STENOSIS-WEINTRAUB
bepicardium and subendocardium in both nonischemic and ischemic zones. Subendocardial flow in the ischemic zone remained less than flow in the subepicardial ischemic zone and less than nonischemic zone flow after administration of both nitroglycerin and nifedipine. Subepicardial flow in the ischemic zone was similar to subepicardial flow in the nonischemic zone both before and after nitroglycerin and before nifedipine. Nifedipine increased subepicardial flow greatly in the nonischemic zone so that, after nifedipine, subepicardial flow in the ischemic zone was much less than that in the nonischemic zone. The effects of nifedipine and nitroglycerin on myocardial blood flow can be interpreted in light of what we
know about the coronary circulation, the nature of coronary vascular reserve and myocardial oxygen demand. Our group5,6 as well as others7*s have previously shown that coronary reserve is exhausted in the subendocardium before it is exhausted in the subepicardium. At a minimal diastolic perfusion pressure of 55 mm Hg there is reserve in both the subendocardium and the subepicardium. At 40 mm Hg, diastolic perfusion pressure reserve is present in the subepicardium but exhausted in the subendocardium, and at 25 mm Hg reserve is exhausted across the myocardial wa11.6The results seen with nifedipine can be explained in relation to these concepts. It can be postulated that, after infusion of nifedipine, subendocardial flow in the ischemic zone decreased as diastolic perfusion pressure decreased from 40 to 31 mm Hg because subendocardial reserve was exhausted. However, subepicardial flow in the ischemic zone was unchanged, presumably because reserve was present to maintain flow unchanged despite a significant reduction in perfusion pressure. Blood flow in the nonischemic zone increased despite a decrease in systemic pressure, and coronary vascular reserve would be expected in this pressure range.6 Using the xenon technique in patients with coronary artery disease, Lichten et al.9 also noted that nifedipine increased global blood flow to the left ventricle. The results with nitroglycerin must be explained differently. Whereas the reduction in blood pressure
was smaller with nitroglycerin than with nifedipine (although only diastolic pressure was significantly different), blood flow actually decreased in the nonischemit zone. In addition, a systemic blood pressure of 101/76 mm Hg after nitroglycerin infusion would not exhaust coronary reserve in the absence of coronary stenosis..5ms Thus we cannot explain the reduction in myocardial blood flow in the nonischemic zone after nitroglycerin as a result of systemic hypotension. However, by decreasing afterload, nitroglycerin reduced myocardial oxygen demand and thus blood flow in the nonischemic zone decreased by autoregulation. This is consistent with findings by Gerry et al.‘O that nitroglycerin can reduce oxygen demand and reduce blood flow in the nonischemic zone and distal to a flow-limiting stenosis. Similarly, and in contrast to their data with nifedipine, Lichten et al.9 noted that nitroglycerin reduced global myocardial blood flow. Nitroglycerin has been noted to induce relaxation of in vitro coronary
288
August 1982
The American Journal of CARDIOLOGY
ET AL.
arterial strips only transiently presence in the medium.ll
despite its continuing
Thus in certain settings it is probable that autoregulation may overcome the direct vasodilatory effect of nitroglycerin, and with decreased oxygen demand
due to decreased afterload, blood flow in the nonischemit zone will decrease concomitantly. Subepicardial blood flow in the ischemic zone most probably also decreased by autoregulation in the setting of decreasing oxygen demand. This finding is also suggested by our data because the decrease in blood flow in the ischemic subepicardium is not significantly different from the decrease in flow to the nonischemic zone. Subendocardial reserve in the ischemic zone would be exhausted at 40 mm Hg before nitroglycerin infusion as noted, and it is consistent that flow in the ischemic subendocardium was less than flow in the nonischemic zone before drug infusion. Thus the reduction in ischemic subendocardial flow after nitroglycerin infusion was most probably due to the reduction in perfusion pressure, and not to decreased oxygen demand. Thus the effects of nitroglycerin and nifedipine on blood flow to the ischemic subendocardium were the same, but were different in all other zones. Recause autoregulation was not exhausted in the nonischemic zone in the nitroglycerin group, we must consider whether a larger dose of nitroglycerin would overcome autoregulation and lead to an increase in myocardial blood flow. This is a possibility, but it is not
likely to be of physiologic relevance unless blood pressure is maintained by a pressor or transfusion. In the group treated with nitroglycerin, systolic pressure decreased from 0 to 25 mm Hg and diastolic pressure from 5 to 35 mm Hg, yet both nonischemic subendocardial and subepicardial blood flow decreased in all 12 dogs. Thus, this physiologic effect was noted even though the actual reduction in systemic pressure induced by nitroglycerin was variable. It is thus likely that a dose of nitroglycerin sufficient to increase myocardial blood flow would cause significant hypotension. The effects of nifedipine and nitroglycerin on the coronary circulation are different, although both drugs are believed to act by interfering with calcium transport. Nifedipine is believed to act by decreasing the number of slow calcium channels.1”r1:3 The cell mechanism of action must be different to account for the different effects observed. Nifedipine appears to dilate the coronary resistance vessels and to increase flow in the nonischemic zone, whereas nitroglycerin does not appear to have a sustained direct effect on the resistance vessels, or the direct dilating effect may be masked by autoregulation. It has been hypothesized that nitroglycerin affects the capacitance vessels rather than the resistance vessels,iJ4 and this effect may be the mechanism of decrease in blood pressure. Thus the sites of action as well as the cell mechanism of action may be different for these agents. Myocardial shortening: The end results of the effect of both nifedipine and nitroglycerin on myocardial shortening are similar in this model of coronary stenosis. Partial occlusion caused only a slight reduction in
Volume 50
NtTROGLYCERINAND NlFEDlPlNE IN CORONARY STENOSIS-WEINTRAUB ET AL.
r
IV Nitroglycerin
IV Nifedipine
I
/ I
‘(
4 afterload
----l
\
t
4 preload
1 t%AL
1 4 %AL
c
4
/
-\
%ALJ
/
kAL--/ 0
FIGURE 5. Flowchart showing how intravenous (IV) nitroglycerin affected myocardial shortening distal to a fixed coronary stenosis. Nitroglycerin decreased systemic blood pressue (BP). The flowchart then branches. On the lefl side, decreased blood pressure led to decreased perfusion pressure (PP) and, when the stenosis was critical in the &endo&diurn, myocardial blood flow (MBF) decreased, which would adversely affect segment shortening (1% AL). On the right side, decreased blood pressure caused decreased afterload and decreased myocardial oxygen demand (MVO& which would preserve shortening (t % AL). Nitroglycerin also decreased preload slightly, which would also decrease myocardial oxygen demand. Segment shortening at the bottom is the end result of these complex processes.
subendocardial segment shortening in the ischemic zone despite a considerable decrease in subendocardial blood flow. This finding is consistent with previous data from our laboratory5 showing that shortening is initially maintained as flow decreases, but with further decreases in flow shortening rapidly decreases and is abolished at a small but finite blood flow level. Neither agent significantly affected ischemic zone shortening despite a decrease in subendocardial flow with both agents. We may hypothesize why shortening did not decrease further with declining blood flow. Nitroglycerin (Fig. 5) and nifedipine (Fig. 6) decreased systemic blood pressure and thus perfusion pressure distal to the coronary stenosis, leading to a significant decrease in myocardial blood flow in the ischemic subendocardium. Over this range of decrease in blood flow to the subendocardium our group5 as well as that of Vatner15 noted a significant decrease in myocardial shortening. However, the decreased afterload produced by these agents would decrease myocardial oxygen demand and thus lead to preservation of contractile function. Nitroglycerin has little direct effect on myocardial contractility,13J6 but the effects of nifedipine are profound. Although nifedipine has been shown to have negative inotropic effects on muscle strips in vitro,17 it has been shown to preserve both myocardial cell viability and function of ischemic myocardium in vivo (Fig. 6).ls21 In addition, if the stenosis was milder than in these experiments, nifedipine could have increased blood flow despite lowered perfusion pressure leading to preservation of shortening. Thus the effects of nifedipine on myocardial shortening in this setting are likely to be a complex inter-
FIGURE 8. Flowchart showing how intravenous (IV) nifedipine affected shortening distal to a fixed coronary stenosis. Nifedipine decreased systemic blood pressure. Tha flowchart then branches. On the lefl slde, decreased blood pressure led to decreased perfusion pressure and, because the stenosis was critical in the subendocardium, myocardial blood flow decreased and would be expected to adversely affect segment shortening. On the right side, decreased blood pressure caused decreased afterload and lowered myocardial oxygen demand, which would preserve shortening. Nifedipine has direct effects on the myocardium causing negative inotropy, but may also directly decrease oxygen demand. If the stenosis was not critical, nifedipine would have been expected to increase blood flow. Segment shortening at the bottom is the end result of these complex processes. Abbreviations as in Figure 5.
action of effects on afterload, blood flow and direct effects on the muscle. In addition, other factors such as heart rate and catecholamines may also affect segment shortening. In these experiments, left ventricular enddiastolic pressure was uniformly low so that little effect could be expected on preload, although nitroglycerin decreased end-diastolic length slightly. Decreased preload will reduce myocardial tension and oxygen demand, but also decrease inotropy through the Starling effect. Shortening is the end result of several processes and the relation to oxygen supply and blood flow is thus complex. Studies with intracoronary administration will be helpful in sorting out the importance of the various physiologic actions of these drugs. Nonetheless, with both agents it remains possible for segment function to be preserved in the setting of decreasing subendocardial blood flow. Comparison of nifedipine with other calcium antagonists: Verapamil and nifedipine are believed to have different cell mechanisms of action1 and, although they are both vasodilators, their relative effects on hemodynamics, coronary vasculature and contractility may vary. In a recent preliminary study in dogs, Urquhart et a1.22showed that at doses causing a similar decrease in systemic pressure, nifedipine resulted in an increase in ejection fraction but no change in left ventricular end-diastolic pressure, whereas verapamil resulted in increased left ventricular end-diastolic pressure and a decreased ejection fraction, thus suggesting that verapamil has greater negative inotropic action. Thus one cannot assume that, in a model such as the one
August 1982
The American Journal of CARDIOLOGY
Volume 50
287
NITROGLYCERINAND NIFEDIPINEIN CORONARY STENOSIS-WEINTRAUB ET AL.
used in this study, another calcium flux antagonist given at a dose causing a similar reduction in blood pressure would cause a similar change in myocardial blood flow or result in preserved shortening. Clinical implications: Given the limitations of the dog model, we have shown that intravenous administration of nifedipine and nitroglycerin can decrease blood flow distal to a flow-limiting coronary stenosis. Despite differences in effects on blood flow, the net effect on myocardial shortening was similar; despite a reduction in ischemic subendocardial blood flow,
myocardial shortening was preserved. Thus, even in the face of a vasodilator-induced decrease in blood flow distal to a fixed stenosis, shortening may be preserved if there is a concomitant decrease in myocardial oxygen demand. Conversely, preservation of contractile function does not necessarily mean that blood flow and oxygen supply are preserved. Acknowledgment We thank Janice Phillips, Scott Cluley and Robert Krumm for their expert technical assistance.
References 1. Stone PH, Antman ME, Muller JE, Braunwald E. Calcium channel blocking agents in the treatment of cardiovascular disorders. Part II: hemcdynamic effects and clinical applications. Ann intern Med 1980;93:888-904. 2. Heymann MA, Payne BD, Hoffman JIE, Rudolph AM. Blood flow measurements with radionuclide labeled particles. Prog Cardiovasc Disc 1977;20:55-79. 3. Theroux P, Ross J, Franklin D, Covell JW, Blood CM, Sasayanna S. Regional myocardial infarction in the unanesthetized dog. Circ Res 1977;40:158-85. 4. Sokal RR, Rohlf FJ. Biometry. San Francisco: JH Freeman, 1969;205-23.328-33. 5. Welntraub WS, Hattorl S, Agarwal JB, Bodenhelmer MM, Banka VS, Helfant RH. The relationship between myocardial blood flow and contraction by myocardial layer in the canine left ventricle during ischemia. Circ Res 1981;48:430-8. 6. Welntraub WS, Hattorl S, Agarwal JB, Bodenheimer MM, Banka VS, Helfant RH. Variable effect of nifedipine on myocardial blood flow at three grades on coronary occlusion in the dog. Circ Res 1981;48-937-42. 7. Rouleaux J, Boerboon LE, Surjahhana A, Hoffman JIE. The role of autoregulation and tissue diastolic pressure in the transmural distribution of left ventricular blood flow in anesthetized dogs. Circ Res 1979;45:804-15. 8. Guyton RA, McClenathan JH, Newman GE, Michaelis LL. Significance of subendocardial S-T segment elevation caused by coronary stenosis in the dog. Am J Cardiol 1977;40:373-80. 9. Llchten P, Engel HJ, Amende I, Raffenbeul W, Simon R. Mechanism of various antianginal drugs: relationship between regional flow behavior and contractility. In: Jatene AD, lichten PR, eds. The Third International Adalat Symposium. Amsterdam: Excerpta Medica, 1976; 14-29. 10. Gerry JL Jr, Sohaff HV, Kallman CH, Flaherty JT. Effects of nitroglycerin on regional myocardial ischemia induced by atrial pacing in dogs. Circ Res 1981; 48:569-76. 11. Fleckensteln A. Specific pharmacology of calcium in myocardium, cardiac pacemakers, and vascular smooth muscle. Annu Rev Pharmacol Toxicol 1977;17:149-66.
288
August 1982
The American
Journal of CARDIOLOGY
12. Bayer R, Rodenklrchen R. Kaufman R, Lee JH, Hennekes R. The effects of nifedipine on contraction and monophasic action potential of isolated cat myocardium. Naunyn Schmiedebergs Arch Pharmacol 1977;301:29-37. 13. Grun G, Fleckensteln A. Die Electromechanische Entkoppelung der glatten Gefafi-muskulatur als Grundprinzip der Coronardilatation druch nifedipine. Arzneim Forsch 1972;22:334-44. 14. Whlte SW, Parges WL, McRltchle RJ. Coronary haemodynamic effects of nifedipine (BAY a 1040) and glyceryl trinitrate in unanaesthetized dogs. Clin Exp Pharmacol Physiol 1974;1:77-86. 15. Vatner SF. Correlation between acute reductions in myocardial blood flow and function in conscious dogs. Circ Res 1980;47: 201-7. 16. Brodle BR, Chock L, Klausner S, Grossman W, Parmley W. Effects of sodium nitroprusside and nitroglycerin on tension prolongation of cat papillary muscle during recovery from hypoxia. Circ Res 1976;39:596-601. 17. Fleckensteln VA, Trltthaut H, Dorlng HJ, Byon KY. BAY a 1040 ein hockaktirer Ca+i=antagonisticher Inhibitor den elektronmechanischen Koppelungsprozesse im warmbluter Myocard. Arzneim Forsch 1972;22:22-33. 18. Welntraub WS, Hattorl S, Agarwal JB, Bodenhelmer MM, Banka VS, Helfant RH. The effect of nifedipine on ischemic myocardium (abstr). Clin Res 1980;28:22OA. 19. Welntraub WS, Hattorl S, Agarwal JB, Bodenhelmer MM, Banka VS, Helfant RH. The effect of nifedipine on myocardial contractility and blood flow. Circulation 1982;65:49-53. 20. Henry PD, Schuchlelb R, Davis J, Weiss ES, Sobel BE. Myocardial contracture and accumulation of mitochondrial calcium in ischemic rabbit heart. Am J Physiol 1977;1233:H677-84. 21. Selwyn AP, Wolman E, Fox K, Horlock P, Pratt J, Klein M. The effects of nifedipine on acute experimental myocardial ischemia and infarction in dogs. Circ Res 1979;44:16-23. 22. Urquhart J, Patterson RE, Bacharach S, Green MV, Epstein SE. Comparative effects of verapamil, diltiazem and nifedipine on hemodynamics and left ventricular function (abstr). Circulation 1981;64:Suppl IV:IV-230.
Volume 50
S. WEINTRAUB,
SATOSHI
AKIZUKI,
JAI
f3. AGARWAL,
MD,
MD,
FACC
MONTY M. BODENHEIMER, MD, VIDYA S. BANKA, MD, FACC RICHARD
H. HELFANT,
Philadelphia,
FACC
MD
MD,
FACC
FACC
Pennsylvania
From The Mid-Atlantic Heart and Vascular Institute, Presbyterian-University of Pennsylvania Medical Center and the School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania. This study was supported in part by Grant HL2654501 from The National Institutes of Health, Bethesda, Maryland. Manuscript received October 5, 1981; revised manuscript received February 1, 1982, accepted February 11, 1982. Address for reprints: William S. Weintraub, MD, Assistant Professor of Medicine, Division of Cardiology, Presbyterian-University of Pennsylvania Medical Center, 51 North 39th Street, Philadelphia, Pennsylvania 19 104.
Both nifedipine and nitroglycerin are used to treat angina pectoris. The comparative effects of these agents on myocardial blood flow and contraction in the setting of flow-limiting coronary stenosis are poorly understood. Thus 24 open chest dogs underwent carotid to left anterior descending coronary arterial perfusion with coronary flow probe and perfusion pressure monitoring. Segment length was measured with ultrasonic crystals in the subendocardial ischemic and nonischemic zones. Myocardial blood flow was measured with radioactive microspheres. Partial coronary occlusion was performed to attain a diastolic perfusion pressure of 40 mm Hg. Twelve dogs received intravenous nifedipine, 3 pg/kg per min, and 12 received intravenous nitroglycerin to reduce aortic pressure by 20 mm Hg. Partial occlusion resulted in a slight but significant decrease in segment shortening in the ischemic zone. Neither nitroglycerin nor nifedipine affected shortening in the ischemic zone. After occlusion, blood flow decreased in the subendocardial ischemic zone but was unchanged in the subepicardium. Nifedipine increased subendocardial blood flow in the nonischemic zone and decreased it in the ischemic zone but caused no change in subepicardial flow in the ischemic zone. In contrast, nitroglycerin decreased subendocardial and subepicardial blood flow in both the ischemic and nonischemic zones. In the setting of coronary stenosis, different classes of vasodilators may have varying effects on myocardial blood flow, suggesting different sites and mechanisms of action. In addition, segment function may not always reflect changes in myocardial blood flow.
Although nifedipine is a potentially important new agent for the treatment of myocardial ischemia,l nitroglycerin remains the most widely used drug for this purpose. However, there is little comparative information available on how these two agents affect the relation between myocardial blood flow and segment shortening distal to a flow-limiting coronary stenosis. This study was therefore designed to determine whether therapeutic doses of intravenous nifedipine and nitroglycerin increase or decrease myocardial blood flow distal to a fixed stenosis and to determine how the combination of effects on preload, afterload and blood flow to the ischemic zone affect myocardial shortening. Methods Experimental procedure: Twenty-four dogs were anesthetized with pentobarbital and placed on a Harvard respirator. Arterial blood gases were monitored with partial pressure of oxygen (POz) maintained between 80 and 100 mm Hg and hydrogen ion concentration (pH) between 7.36 and 7.44. A 7F catheter was passed into the right femoral artery and advanced to the thoracic aorta for aterial pressure monitoring. A stiff catheter, internal diameter 2 mm, was passed into the left femoral artery and advanced to the abdominal aorta for blood withdrawal. A catheter was also passed into the right femoral vein for intravenous infusions.
August 1982
The American
Journal of CARDIOLDGY
Volume 50
281
NITROGLYCERIN
AND NIFEDIPINE IN CORONARY
STENOSIS-WEINTRAUB
A thoracotomy was performed in the fifth left intercostal space and the heart supported in a pericardial cradle. A 7F catheter was passed into the left atrial appendage for injection of microspheres (Fig. 1). A 14 gauge stiff catheter, 10 cm long, was passed into the left ventricle through the apex to record left ventricular pressure and its first derivative (dP/dt). A 5 to 10 cm segment of the left common carotid artery was exposed. After administration of heparin, the left anterior descending coronary artery was isolated at the level of the first or second diagonal branch and ligated and an arteriotomy performed. A 14 gauge steel cannula was inserted into the left anterior descending artery, tied securely and continuously perfused from the left common carotid artery through plastic tubing with a minimal internal diameter of 2 mm. Coronary perfusion pressure was monitored with a strain gauge (Statham, P23 DB) and coronary flow with an electromagnetic flow probe with an internal diameter of 2 mm (Micron Instruments). The time from ligation of the left anterior descending artery to establishment of cannula perfusion averaged 50 seconds, and all dogs in which cannulation took more than 120 seconds were excluded from further study. During cannula perfusion of the left anterior descending coronary artery, pressures at the tip of the cannula and in the
tubing were identical. In addition, diastolic perfusion pressure in the tubing was identical to aortic perfusion pressure at flows of up to 100 ml/min. The cannula system was also tested in vitro with warm blood pumped through the system at rates
FIGURE 1. Experimental preparation showing the carotid to left anterior descending (LAD) arterial cannulation system, the electromagnetic flow probe and the screw clamp constrictor. Note that perfusion pressure (Perf. P) is measured distal to the constrictor. The potentially ischemic zone (hatched area) is colored by the Evans blue dye. The ring of myocardium from which tissue samples are taken for blood flow analysis is noted by the broken lines. Note that both nonischemic zone (NZ) and ischemic zone (IZ) crystals are within this ring. Nonischemic and ischemic zone crystals and samples of tissue for blood flow were at least 1 cm from the border between nonischemic and ischemic tissues. Ao = aorta; AoP = aortic pressure; CBF = coronary blood flow probe: CXL = left circumflex coronary artery; LAcath = left atrial catheter; LVP = left ventricular pressure; PA = pulmonary artery.
282
August 1882
The American
Journal of CARDIOLOGY
ET AL.
from 1.1 to 46 ml/min with pressures monitored in the proximal cannulation system and at the cannula tip. No pressure gradient was observed. The presence of coronary reserve was determined by using a 10 second period of total occlusion followed by reperfusion. All dogs with less than 100 percent reactive hyperemia were rejected as having either inadequate coronary reserve or stenosis in the cannula system. Placement of ultrasonic crystals: Two pairs of 2 mm ultrasonic crystals were placed in the subendocardium, one pair each in the nonischemic and ischemic zones (Fig. 1). The crystals in each pair were between 8 and 15 mm apart. The crystals were inserted into the inner third of the myocardium through small stab wounds perpendicular to the long axis of the left ventricle. The motion of the ultrasonic crystals was monitored with an ultrasonic imaging circuit (Schuessler and Associates). Microsphere technique: Microspheres (diameter 9 p) labeled with iodine-125 (lzsI), cerium-141 (i*lCe), strontium-85 (85Sr) or scandium-46 (4sS~) (3M Company) were used to measure myocardial blood flow. Microspheres were suspended in saline solution with a drop of Tween-SO@, agitated in an ultrasonic bath for at least 15 minutes and shaken in a vortex whirler before injection. The left atrium was injected with 2 to 3 million microspheres in 8 ml of saline solution over 15 to 20 seconds then given a 4 ml flush of microsphere-free saline solution. Starting before the injection of microspheres, blood was withdrawn from a femoral artery at 7.75 ml/min with a Harvard pump in order to obtain a reference blood flow. Blood withdrawal was continued for 1 minute after completion of the saline flush. Experimental protocol: The preparation was allowed to stabilize for at least 15 minutes after cannulation. Partial coronary occlusion was performed by obstructing the cannulation tubing with a screw clamp device to a minimal diastolic perfusion pressure of 40 mm Hg. The preparation was allowed to stabilize for 5 minutes and a first set of microspheres of one isotope was given. The dogs were then randomized into two groups. In group 1,12 dogs were given continuous intravenous nifedipine, 3 pg/kg per min, and in group 2,12 dogs were given continuous intravenous nitroglycerin to decrease aortic systolic blood pressure by 15 to 20 mm Hg but not to less than 90 mm Hg. Fifteen minutes after the infusion began, a second set of microspheres of a different isotope was given. By giving two sets of microspheres of different energy levels the blood flow before and after drug administration was determined. Tissue preparation: After the second microsphere injection, Evans blue dye was injected into the coronary cannula by hand-held syringe with sufficient pressure to stain the myocardium but insufficient to fill the intercoronary collateral vessels. The dog was killed and the position of the crystals noted and then removed. The heart was washed, dried, stuffed with gauze and wrapped with industrial strength aluminum foil and frozen. The heart was sectioned while still frozen to facilitate accurate cutting. A 2 to 3 cm wide transmural ring of myocardium was cut with the path of the cut perpendicular to the border between blue-stained ischemic tissue and unstained nonischemic tissue (Fig. 1). A section of myocardium was taken from the remote nonischemic zone. The ring was cut at the blue border with care taken to keep all blue-stained tissue on the ischemic side. A 1 cm sample of myocardium was taken on the ischemic (blue) side labeled border ischemic, followed by a 1 cm central ischemic sample. In all dogs the central ischemic sample contained the crystal tracts. All samples were divided into subendocardial, mid myocardial and subepicardial third.
Volume 50
NITROGLYCERIN
ECG
AND NIFEDIPINE IN CORONARY
STENOSIS-WEINTRAUB
ET AL.
-+--
Segment
length FIGURE 2. Sample recordings showing the method for calculation of shortening. The lefl panel, control period (c); right panel, after an intervention. The tracing from the top down are the electrocardiogram, the segment length representing the distance between the pair of crystals, the straight line of 0 mm crystal separation, and the first derivative of left ventricular pressure with respect to time (dP/dt) and left ventricular pressure (LVP). End-diastolic length (EDL) and end-systolic length (ESL) are measured from the 0 line to the segment length line, and these lengths are timed from dP/dt according to the method of Theroux et al.3 Percent shortening (% Ai_) is calculated as shown. Shortening (N% AL) and end-diastolic length (NEDL) may then be normalized to the fraction of the control period as shown.
,,,,,A LVP
J--I EDL,% A L c=
ESL,
EDL,
EDL x 100
%AL =
NEDL
The tissue samples were weighed and counted together with the blood samples, pure isotope standards and a background tube in a Beckman 8,000 well gamma counter for 10 minutes each. Myocardial blood flow was then determined by the method of Heymann et al2 A measure of resistance was determined from the ratio of diastolic perfusion pressure to coronary blood flow. Data recording and analysis: Hemodynamic data and myocardial shortening were obtained on an Electronics for Medicine VR-16 recorder. Myocardial shortening and enddiastolic length were calculated by the method of Theroux et al.” and then normalized t,o the control period (Fig. 2). All data are expressed as mean f standard deviation. Differences for any variable between subgroups of the two groups described earlier were analyzed by a two way analysis of variance, randomized block design.4 Differences between the two groups for any variable were analyzed by a one way analysis of variance.”
-
ESL
EDL
x 100 % A L
EDL = F,
N%AL
=%a~
Results Hemodynamics and coronary flow: Aortic, systolic and diastolic pressures were not different in the nifedipine and nitroglycerin treated groups before drug infusion (Table I). Systolic and diastolic pressures decreased after drug infusion in both groups. The decrease in pressure with nifedipine was greater than with nitroglycerin, but the difference between groups was significant only for diastolic pressure (p
TABLE I Hemodynamic Data Nifedipine
Control LVEDP (mm Hg) AoP (mm Hg) Systolic Diastolic Heart rate Rate-pressure product (mm Hg-beats/min)
7f3 11af22 a7 f 22 130 f 32 17,200 f7,500
p
Partial Occlusion a*4
Nitroglycerin
Nifedipine
12 16 123 f 30 13,600 f4,300 aof
Control 3f2
855
109 f
C
91 f 12’ 55 f 14’ 119f26 11,600’ f4.600
121 f 13 93f 13 136 f 25 16,500 f3,600
Partial Occlusion 3f2 119f 16 a7f 19 134 f 26 16,000 f3,aoo
Nitroglycerin 3f2 101 f 13’ 76 f 16’ 136 f 26 13,800* f3,300
l
end-diastolic pressure.
August 1982
The American Journal of CARDIOLOGY
Volume 50
283
NITROGLYCERIN AND NIFEDIPINE IN CORONARY STENOSIS-WEINTRAUB ET AL.
TABLE
II
Perfusion Pressure, Coronary Flow and Coronary Resistance Nifedipine Partial Occlusion
Nifedipine
87 f 40f
13‘ 1”
69 f 12’ 31 f 2’
25f9” 70 f 16’
20 f 8’ 56f ll+
1.9 f 0.9”
1.9 f 0.6
Control Pressure (mm Hg) Systolic Diastolic Blood flow (mbmin) Coronarv Normaked Resistance (mm Hg-minImI) l
115 f 25 86 f 26 33 f 9 100 2.8 f
1.0
Nitroglycerin Partial Occlusion
Control 120 f 92 f
13 13
27f 10 100 3.9 f
1.6
90f7’ 40fO’
77 f 9’ 35 f 4’
21 f 11” 76 f 21‘
16 f 8+ 61 f 18’
2.4 f
TABLE
2.6 f
1.2
nonischemic zone, nifedipine caused a significant increase in flow in the subepicardium (p
Ill
Regional Myocardlal
Blood Flow (ml/g per min) Nitroglycerin
Nifedipine Partial Occlusion lschemic zone Subendocardial blood flow Subepicardial blood flow Nonisdhemic zone Subendocardial blood flow Subepicardial blood flow
284
1.4’
p
nifedipine and by 2,200 f 1,200 units after nitroglycerin. Hemodynamic measurements in the cannulation system reflecting flow and perfusion pressure in the subserved vasculature (Table II) demonstrated no difference in pressure between the aorta and cannulation system before coronary occlusion. Partial occlusion decreased perfusion pressure to a minimal diastolic level of 40 mm Hg. Coronary flow decreased slightly but significantly in both groups. After drug infusion, diastolic and systolic perfusion pressures decreased in both groups. Coronary flow measured with the electromagnetic flow probe also decreased in both groups after drug infusion. Resistance in the perfused bed decreased significantly in both groups after partial occlusion, but neither nifedipine nor nitroglycerin significantly changed the total resistance in the perfused bed. Regional myocardial blood flow (Table III, Fig. 3 and 4): Before drug infusion, subendocardial blood flow in the ischemic zone was less than both subepicardial flow and less than subendocardial flow in the nonischemic zone in both drug treatment groups (p
l
Nitroglycerin
Nifedioine
Nitroolycerin
0.66 f 0.30 1.21 f 0.46
0.42 f 0.28’ 1.24 f 0.60
0.78 f 0.29 1.17 f 0.32
0.57 f 0.23‘ 0.92 f 0.23’
1.26 f 0.44+ 1.31 f 0.73
1.52 f 2.12 f
1.14 f 0.34+ 1.12 f 0.40
0.84 f 0.17’t 0.83 f 0.27’
l.Ol+ 1.02*+
p
August 1982
Partial Occlusion
The American Journal of CARDIOLOGY
Volume 50
NITROGLYCERIN AND NIFEDIPINE IN CORONARY STENOSIS
-
-
lschemic
-
Non-ischemic
--
WEINTRAUB ET AL.
lschemic -
Non-ischemic
2.2 2 .-5 a 2
2.0 t
/ /
1.8
/ I
1.6
/'
2
3 e
5
0.6
00
0.4
2
0.2
,e
1’ _~Y-c
1.2
0.8
z
/
1.4
;
/
Endo
’ Epi “0 0 z =
\
0.6
Y*
Endo
0.4
L
--,
$
kEndo
r’-
* pco.01
i
0.2
Postocclusion
p
Postocclusion
I
I
l
t
Postnifedipine
Postnitroglycerin
FIGURE 4. Influence of nitroglycerin on myocardial blood flow after coronary stenosis. Regional myocardial blood flow measured with microspheres is displayed before (left) and after (right) intravenous administration of nitroglycerin. Nitroglycerin resulted in a decrease in myocardial blood flow to the subendocardium and subepicardium in both the nonischemic and ischemic zones. Abbreviations as in Figure 3.
FIGURE 3. Influence of nifedipine on myocardial blood flow after coronary stenosis. Regional myocardial blood flow measured with microspheres is displayed before (left) and after (rlght) intravenous administration of nifedipine. Nifedipine increased flow to the nonischemic subepicardium (Epi) but did not affect nonischemic subendocardial (Endo) flow. In contrast, nifedipine decreased blood flow to the ischemic subendocardium blood flow but did not affect flow to the ischemic subepicardium.
glycerin have both similarities and some differences. Both nifedipine and nitroglycerin decreased systemic blood pressure as well as perfusion pressure distal to the fixed coronary stenosis. In addition, both agents reduced myocardial oxygen demand to a similar degree as measured by rate-pressure product. Despite these relatively similar effects on systemic and perfusion pressures these two agents had strikingly different effects on myocardial blood flow. Nifedipine increased flow to the nonischemic zone, although this was significant only in the subepicardium. It decreased flow in the ischemic zone of the subendocardium but had no effect on this zone in the subepicardium. In contrast, nitroglycerin decreased myocardial blood flow to the su-
emit zone (p CO.01). Nitroglycerin did not affect segment shortening in either the ischemic or the nonischemit zone. Nitroglycerin but not nifedipine decreased end-diastolic length slightly in both the ischemic and’ the nonischemic zones (p <0.05). Discussion Myocardial blood flow: This study casts new light on the nature of the coronary circulation and the influence of vasodilators. We have shown that mildly to moderately hypotensive doses of nifedipine and nitro-
TABLE IV End-Diastolic Length and Segment Shortening Nitroglycerin
Control lschemic EDL (mm) NEDL % AL N%AL Nonischemic EEbimm) % AL N%AL
8.9 f 16.7:
1.8 6.9
1 9.7 f 2.1 15.2:
1
5.2
Partial Occlusion
Nifedipine
Control
9.2 1.03 13.6 0.87
f f f f
2.0t 0.04 6.8+ O.ll+
9.2 1.04 14.0 0.88
f 1.9 f 0.02 f 7.4 f 0.15
11.9 f 2.5
9.8 1.01 15.4 1.03
f f f f
2.1 0.04 4.9 0.10
10.0 1.03 17.3 1.15
f f f f
11.2 f 2.4
2.3 0.05 5.V 0.19’
6.0
16.3: 1
5.3
16.5: 1
p CO.05 compared with value to left. t p CO.01 compared with value to left. % AL = percent segment shortening; N%L = fraction of control period shortening: EDL = enddiastolic enddiastolic length.
Partial Occlusion
Nitroglycerin
11.9 1.00 15.6 0.95
f f f f
2.5 0.02 5.8+ 0.05t
11.6 0.98 15.1 0.94
f f f f
2.5’ 0.05 5.0 0.06
11.2 1.00 17.0 1.02
f f f f
2.3 0.02 5.5 0.05
10.9 0.98 16.8 1.02
f f f f
2.3’ 0.04’ 5.5 0.05
l
August 1982
length; NEDL = fraction of control
The American Journal of CARDIOLOGY
Volume 50
peri&
285
NITROGLYCERIN
AND NIFEDIPINE IN CORONARY
STENOSIS-WEINTRAUB
bepicardium and subendocardium in both nonischemic and ischemic zones. Subendocardial flow in the ischemic zone remained less than flow in the subepicardial ischemic zone and less than nonischemic zone flow after administration of both nitroglycerin and nifedipine. Subepicardial flow in the ischemic zone was similar to subepicardial flow in the nonischemic zone both before and after nitroglycerin and before nifedipine. Nifedipine increased subepicardial flow greatly in the nonischemic zone so that, after nifedipine, subepicardial flow in the ischemic zone was much less than that in the nonischemic zone. The effects of nifedipine and nitroglycerin on myocardial blood flow can be interpreted in light of what we
know about the coronary circulation, the nature of coronary vascular reserve and myocardial oxygen demand. Our group5,6 as well as others7*s have previously shown that coronary reserve is exhausted in the subendocardium before it is exhausted in the subepicardium. At a minimal diastolic perfusion pressure of 55 mm Hg there is reserve in both the subendocardium and the subepicardium. At 40 mm Hg, diastolic perfusion pressure reserve is present in the subepicardium but exhausted in the subendocardium, and at 25 mm Hg reserve is exhausted across the myocardial wa11.6The results seen with nifedipine can be explained in relation to these concepts. It can be postulated that, after infusion of nifedipine, subendocardial flow in the ischemic zone decreased as diastolic perfusion pressure decreased from 40 to 31 mm Hg because subendocardial reserve was exhausted. However, subepicardial flow in the ischemic zone was unchanged, presumably because reserve was present to maintain flow unchanged despite a significant reduction in perfusion pressure. Blood flow in the nonischemic zone increased despite a decrease in systemic pressure, and coronary vascular reserve would be expected in this pressure range.6 Using the xenon technique in patients with coronary artery disease, Lichten et al.9 also noted that nifedipine increased global blood flow to the left ventricle. The results with nitroglycerin must be explained differently. Whereas the reduction in blood pressure
was smaller with nitroglycerin than with nifedipine (although only diastolic pressure was significantly different), blood flow actually decreased in the nonischemit zone. In addition, a systemic blood pressure of 101/76 mm Hg after nitroglycerin infusion would not exhaust coronary reserve in the absence of coronary stenosis..5ms Thus we cannot explain the reduction in myocardial blood flow in the nonischemic zone after nitroglycerin as a result of systemic hypotension. However, by decreasing afterload, nitroglycerin reduced myocardial oxygen demand and thus blood flow in the nonischemic zone decreased by autoregulation. This is consistent with findings by Gerry et al.‘O that nitroglycerin can reduce oxygen demand and reduce blood flow in the nonischemic zone and distal to a flow-limiting stenosis. Similarly, and in contrast to their data with nifedipine, Lichten et al.9 noted that nitroglycerin reduced global myocardial blood flow. Nitroglycerin has been noted to induce relaxation of in vitro coronary
288
August 1982
The American Journal of CARDIOLOGY
ET AL.
arterial strips only transiently presence in the medium.ll
despite its continuing
Thus in certain settings it is probable that autoregulation may overcome the direct vasodilatory effect of nitroglycerin, and with decreased oxygen demand
due to decreased afterload, blood flow in the nonischemit zone will decrease concomitantly. Subepicardial blood flow in the ischemic zone most probably also decreased by autoregulation in the setting of decreasing oxygen demand. This finding is also suggested by our data because the decrease in blood flow in the ischemic subepicardium is not significantly different from the decrease in flow to the nonischemic zone. Subendocardial reserve in the ischemic zone would be exhausted at 40 mm Hg before nitroglycerin infusion as noted, and it is consistent that flow in the ischemic subendocardium was less than flow in the nonischemic zone before drug infusion. Thus the reduction in ischemic subendocardial flow after nitroglycerin infusion was most probably due to the reduction in perfusion pressure, and not to decreased oxygen demand. Thus the effects of nitroglycerin and nifedipine on blood flow to the ischemic subendocardium were the same, but were different in all other zones. Recause autoregulation was not exhausted in the nonischemic zone in the nitroglycerin group, we must consider whether a larger dose of nitroglycerin would overcome autoregulation and lead to an increase in myocardial blood flow. This is a possibility, but it is not
likely to be of physiologic relevance unless blood pressure is maintained by a pressor or transfusion. In the group treated with nitroglycerin, systolic pressure decreased from 0 to 25 mm Hg and diastolic pressure from 5 to 35 mm Hg, yet both nonischemic subendocardial and subepicardial blood flow decreased in all 12 dogs. Thus, this physiologic effect was noted even though the actual reduction in systemic pressure induced by nitroglycerin was variable. It is thus likely that a dose of nitroglycerin sufficient to increase myocardial blood flow would cause significant hypotension. The effects of nifedipine and nitroglycerin on the coronary circulation are different, although both drugs are believed to act by interfering with calcium transport. Nifedipine is believed to act by decreasing the number of slow calcium channels.1”r1:3 The cell mechanism of action must be different to account for the different effects observed. Nifedipine appears to dilate the coronary resistance vessels and to increase flow in the nonischemic zone, whereas nitroglycerin does not appear to have a sustained direct effect on the resistance vessels, or the direct dilating effect may be masked by autoregulation. It has been hypothesized that nitroglycerin affects the capacitance vessels rather than the resistance vessels,iJ4 and this effect may be the mechanism of decrease in blood pressure. Thus the sites of action as well as the cell mechanism of action may be different for these agents. Myocardial shortening: The end results of the effect of both nifedipine and nitroglycerin on myocardial shortening are similar in this model of coronary stenosis. Partial occlusion caused only a slight reduction in
Volume 50
NtTROGLYCERINAND NlFEDlPlNE IN CORONARY STENOSIS-WEINTRAUB ET AL.
r
IV Nitroglycerin
IV Nifedipine
I
/ I
‘(
4 afterload
----l
\
t
4 preload
1 t%AL
1 4 %AL
c
4
/
-\
%ALJ
/
kAL--/ 0
FIGURE 5. Flowchart showing how intravenous (IV) nitroglycerin affected myocardial shortening distal to a fixed coronary stenosis. Nitroglycerin decreased systemic blood pressue (BP). The flowchart then branches. On the lefl side, decreased blood pressure led to decreased perfusion pressure (PP) and, when the stenosis was critical in the &endo&diurn, myocardial blood flow (MBF) decreased, which would adversely affect segment shortening (1% AL). On the right side, decreased blood pressure caused decreased afterload and decreased myocardial oxygen demand (MVO& which would preserve shortening (t % AL). Nitroglycerin also decreased preload slightly, which would also decrease myocardial oxygen demand. Segment shortening at the bottom is the end result of these complex processes.
subendocardial segment shortening in the ischemic zone despite a considerable decrease in subendocardial blood flow. This finding is consistent with previous data from our laboratory5 showing that shortening is initially maintained as flow decreases, but with further decreases in flow shortening rapidly decreases and is abolished at a small but finite blood flow level. Neither agent significantly affected ischemic zone shortening despite a decrease in subendocardial flow with both agents. We may hypothesize why shortening did not decrease further with declining blood flow. Nitroglycerin (Fig. 5) and nifedipine (Fig. 6) decreased systemic blood pressure and thus perfusion pressure distal to the coronary stenosis, leading to a significant decrease in myocardial blood flow in the ischemic subendocardium. Over this range of decrease in blood flow to the subendocardium our group5 as well as that of Vatner15 noted a significant decrease in myocardial shortening. However, the decreased afterload produced by these agents would decrease myocardial oxygen demand and thus lead to preservation of contractile function. Nitroglycerin has little direct effect on myocardial contractility,13J6 but the effects of nifedipine are profound. Although nifedipine has been shown to have negative inotropic effects on muscle strips in vitro,17 it has been shown to preserve both myocardial cell viability and function of ischemic myocardium in vivo (Fig. 6).ls21 In addition, if the stenosis was milder than in these experiments, nifedipine could have increased blood flow despite lowered perfusion pressure leading to preservation of shortening. Thus the effects of nifedipine on myocardial shortening in this setting are likely to be a complex inter-
FIGURE 8. Flowchart showing how intravenous (IV) nifedipine affected shortening distal to a fixed coronary stenosis. Nifedipine decreased systemic blood pressure. Tha flowchart then branches. On the lefl slde, decreased blood pressure led to decreased perfusion pressure and, because the stenosis was critical in the subendocardium, myocardial blood flow decreased and would be expected to adversely affect segment shortening. On the right side, decreased blood pressure caused decreased afterload and lowered myocardial oxygen demand, which would preserve shortening. Nifedipine has direct effects on the myocardium causing negative inotropy, but may also directly decrease oxygen demand. If the stenosis was not critical, nifedipine would have been expected to increase blood flow. Segment shortening at the bottom is the end result of these complex processes. Abbreviations as in Figure 5.
action of effects on afterload, blood flow and direct effects on the muscle. In addition, other factors such as heart rate and catecholamines may also affect segment shortening. In these experiments, left ventricular enddiastolic pressure was uniformly low so that little effect could be expected on preload, although nitroglycerin decreased end-diastolic length slightly. Decreased preload will reduce myocardial tension and oxygen demand, but also decrease inotropy through the Starling effect. Shortening is the end result of several processes and the relation to oxygen supply and blood flow is thus complex. Studies with intracoronary administration will be helpful in sorting out the importance of the various physiologic actions of these drugs. Nonetheless, with both agents it remains possible for segment function to be preserved in the setting of decreasing subendocardial blood flow. Comparison of nifedipine with other calcium antagonists: Verapamil and nifedipine are believed to have different cell mechanisms of action1 and, although they are both vasodilators, their relative effects on hemodynamics, coronary vasculature and contractility may vary. In a recent preliminary study in dogs, Urquhart et a1.22showed that at doses causing a similar decrease in systemic pressure, nifedipine resulted in an increase in ejection fraction but no change in left ventricular end-diastolic pressure, whereas verapamil resulted in increased left ventricular end-diastolic pressure and a decreased ejection fraction, thus suggesting that verapamil has greater negative inotropic action. Thus one cannot assume that, in a model such as the one
August 1982
The American Journal of CARDIOLOGY
Volume 50
287
NITROGLYCERINAND NIFEDIPINEIN CORONARY STENOSIS-WEINTRAUB ET AL.
used in this study, another calcium flux antagonist given at a dose causing a similar reduction in blood pressure would cause a similar change in myocardial blood flow or result in preserved shortening. Clinical implications: Given the limitations of the dog model, we have shown that intravenous administration of nifedipine and nitroglycerin can decrease blood flow distal to a flow-limiting coronary stenosis. Despite differences in effects on blood flow, the net effect on myocardial shortening was similar; despite a reduction in ischemic subendocardial blood flow,
myocardial shortening was preserved. Thus, even in the face of a vasodilator-induced decrease in blood flow distal to a fixed stenosis, shortening may be preserved if there is a concomitant decrease in myocardial oxygen demand. Conversely, preservation of contractile function does not necessarily mean that blood flow and oxygen supply are preserved. Acknowledgment We thank Janice Phillips, Scott Cluley and Robert Krumm for their expert technical assistance.
References 1. Stone PH, Antman ME, Muller JE, Braunwald E. Calcium channel blocking agents in the treatment of cardiovascular disorders. Part II: hemcdynamic effects and clinical applications. Ann intern Med 1980;93:888-904. 2. Heymann MA, Payne BD, Hoffman JIE, Rudolph AM. Blood flow measurements with radionuclide labeled particles. Prog Cardiovasc Disc 1977;20:55-79. 3. Theroux P, Ross J, Franklin D, Covell JW, Blood CM, Sasayanna S. Regional myocardial infarction in the unanesthetized dog. Circ Res 1977;40:158-85. 4. Sokal RR, Rohlf FJ. Biometry. San Francisco: JH Freeman, 1969;205-23.328-33. 5. Welntraub WS, Hattorl S, Agarwal JB, Bodenhelmer MM, Banka VS, Helfant RH. The relationship between myocardial blood flow and contraction by myocardial layer in the canine left ventricle during ischemia. Circ Res 1981;48:430-8. 6. Welntraub WS, Hattorl S, Agarwal JB, Bodenheimer MM, Banka VS, Helfant RH. Variable effect of nifedipine on myocardial blood flow at three grades on coronary occlusion in the dog. Circ Res 1981;48-937-42. 7. Rouleaux J, Boerboon LE, Surjahhana A, Hoffman JIE. The role of autoregulation and tissue diastolic pressure in the transmural distribution of left ventricular blood flow in anesthetized dogs. Circ Res 1979;45:804-15. 8. Guyton RA, McClenathan JH, Newman GE, Michaelis LL. Significance of subendocardial S-T segment elevation caused by coronary stenosis in the dog. Am J Cardiol 1977;40:373-80. 9. Llchten P, Engel HJ, Amende I, Raffenbeul W, Simon R. Mechanism of various antianginal drugs: relationship between regional flow behavior and contractility. In: Jatene AD, lichten PR, eds. The Third International Adalat Symposium. Amsterdam: Excerpta Medica, 1976; 14-29. 10. Gerry JL Jr, Sohaff HV, Kallman CH, Flaherty JT. Effects of nitroglycerin on regional myocardial ischemia induced by atrial pacing in dogs. Circ Res 1981; 48:569-76. 11. Fleckensteln A. Specific pharmacology of calcium in myocardium, cardiac pacemakers, and vascular smooth muscle. Annu Rev Pharmacol Toxicol 1977;17:149-66.
288
August 1982
The American
Journal of CARDIOLOGY
12. Bayer R, Rodenklrchen R. Kaufman R, Lee JH, Hennekes R. The effects of nifedipine on contraction and monophasic action potential of isolated cat myocardium. Naunyn Schmiedebergs Arch Pharmacol 1977;301:29-37. 13. Grun G, Fleckensteln A. Die Electromechanische Entkoppelung der glatten Gefafi-muskulatur als Grundprinzip der Coronardilatation druch nifedipine. Arzneim Forsch 1972;22:334-44. 14. Whlte SW, Parges WL, McRltchle RJ. Coronary haemodynamic effects of nifedipine (BAY a 1040) and glyceryl trinitrate in unanaesthetized dogs. Clin Exp Pharmacol Physiol 1974;1:77-86. 15. Vatner SF. Correlation between acute reductions in myocardial blood flow and function in conscious dogs. Circ Res 1980;47: 201-7. 16. Brodle BR, Chock L, Klausner S, Grossman W, Parmley W. Effects of sodium nitroprusside and nitroglycerin on tension prolongation of cat papillary muscle during recovery from hypoxia. Circ Res 1976;39:596-601. 17. Fleckensteln VA, Trltthaut H, Dorlng HJ, Byon KY. BAY a 1040 ein hockaktirer Ca+i=antagonisticher Inhibitor den elektronmechanischen Koppelungsprozesse im warmbluter Myocard. Arzneim Forsch 1972;22:22-33. 18. Welntraub WS, Hattorl S, Agarwal JB, Bodenhelmer MM, Banka VS, Helfant RH. The effect of nifedipine on ischemic myocardium (abstr). Clin Res 1980;28:22OA. 19. Welntraub WS, Hattorl S, Agarwal JB, Bodenhelmer MM, Banka VS, Helfant RH. The effect of nifedipine on myocardial contractility and blood flow. Circulation 1982;65:49-53. 20. Henry PD, Schuchlelb R, Davis J, Weiss ES, Sobel BE. Myocardial contracture and accumulation of mitochondrial calcium in ischemic rabbit heart. Am J Physiol 1977;1233:H677-84. 21. Selwyn AP, Wolman E, Fox K, Horlock P, Pratt J, Klein M. The effects of nifedipine on acute experimental myocardial ischemia and infarction in dogs. Circ Res 1979;44:16-23. 22. Urquhart J, Patterson RE, Bacharach S, Green MV, Epstein SE. Comparative effects of verapamil, diltiazem and nifedipine on hemodynamics and left ventricular function (abstr). Circulation 1981;64:Suppl IV:IV-230.
Volume 50