Effects of Adenosine on Human Coronary Circulation · usefulness of adenosine for studies of the...

1595

Clinical Investigation

Effects of Adenosine on Human CoronaryArterial Circulation

Robert F. Wilson, MD, Keith Wyche, BS, Betsy V. Christensen, BSN,

Steven Zimmer, MD, and David D. Laxson, MD

Adenosine is a potent vasodilator used extensively to study the coronary circulation of animals.Its use in humans, however, has been hampered by lack of knowledge about its effects on thehuman coronary circulation and by concern about its safety. We investigated in humans theeffects of adenosine, administered by intracoronary bolus (2-16 jig), intracoronary infusion(10-240 ,ug/min), or intravenous infusion (35-140 pg/kg/min) on coronary and systemichemodynamics and the electrocardiogram. Coronary blood flow velocity (CBFV) was measuredwith a 3F coronary Doppler catheter. The maximal CBFV was determined with intracoronarypapaverine (4.5±0.2. resting CBFV). In normal left coronary arteries (n=20), 16-pg boluses ofadenosine caused coronary hyperemia similar to that caused by papaverine (4.6±0.7 * restingCBFV). In the right coronary artery (n=5), 12-,ug boluses caused maximal hyperemia(4.4± 1.0 * resting CBFV). Intracoronary boluses caused a small, brief decrease in arterialpressure (similar to that caused by papaverine) and no changes in heart rate or in theelectrocardiogram. The duration of hyperemia was much shorter after adenosine than afterpapaverine administration. Intracoronary infusions of 80 ,g/min or more into the left coronaryartery (n=6) also caused maximal hyperemia (4.4±0.1. resting CBFV), and doses up to 240pg/min caused a minimal decrease in arterial pressure (-6±2 mm Hg) and no significantchange in heart rate or in electrocardiographic variables. Intravenous infusions in normalpatients (n=25) at 140 ,ug/kg/min caused coronary vasodilation similar to that caused bypapaverine in 84% of patients (4.4±0.9 -resting CBFV). At submaximal infusion rates,however, CBFV often fluctuated widely. During the 140-,ug/kg/min infusion, arterial pressure

decreased 6±7 mm Hg, and heart rate increased 24± 14 beats/min. One patient developed 1cycle of 2:1 atrioventricular block, but otherwise, the electrocardiogram did not change. Ineight patients with microvascular vasodilator dysfunction (ACBFV, <3.5 peak/resting velocityafter a maximally vasodilating dose of intracoronary papaverine), the dose-response charac-teristics to intracoronary boluses and intravenous infusions of adenosine were similar to thosefound in normal patients. These studies suggest that maximal coronary vasodilation can beachieved safely with intracoronary adenosine administration and that intravenous infusions ata rate of 140 ,ug/kg/min cause near-maximal coronary hyperemia in most patients. (Circulation1990;82:1595-1606)

Maany studies of the coronary circulationrequire the use of drugs that can safelyand reliably produce maximal coronary

hyperemia of brief duration (for example, for mea-

surement of coronary flow reserve, thallium-201 scin-tigraphy, and echocardiographic imaging).1-6 Anideal agent for these studies would cause maximal

From the Cardiovascular Division, Department of Medicine,University of Minnesota, Minneapolis.R.F.W. was supported by a Clinician-Scientist Award from the

American Heart Association with funds contributed, in part, bythe Minnesota Affiliate.Address for correspondence: Robert F. Wilson, MD, Box 508

UMHC, 420 Delaware Street, SE, Minneapolis, MN 55455.Received November 28, 1989; revision accepted June 26, 1990.

coronary vasodilation, would be effective when givensystemically or by the intracoronary route, would bequickly reversible, and would have no significanteffects on systemic hemodynamics or on electrocar-diographic variables.The two drugs currently available for producing

maximal coronary hyperemia in humans have unde-sirable characteristics.7-10 Papaverine, given by the

See p 1854intracoronary route, causes relatively brief (15-30seconds) maximal coronary hyperemia, but the totaldose that can be given is limited by its relatively slowsystemic excretion (half-life, 3-6 hours)." Conse-quently, intravenous or prolonged intracoronary

by guest on April 13, 2017

http://circ.ahajournals.org/D

ownloaded from

http://circ.ahajournals.org/

1596 Circulation Vol 82, No 5, November 1990

infusions cannot be given without incurring systemichypotension. In addition, intracoronary papaverineprolongs the QT interval and can cause polymor-phous ventricular tachycardia.10,'2 Intravenous dipyr-idamole also elicits maximal coronary hyperemia, butits duration of action is quite long (>30 minutes)and, therefore, precludes repeated measurementsduring the same study. Even though the vasodilatoreffects can be attenuated by methylxanthines, pro-longed ischemia, presumably due to coronary steal,has been reported.1314 Because dipyridamole causescoronary vasodilation by increasing interstitial aden-osine concentration15 and because adenosine is rap-idly metabolized, adenosine itself may be a safer,more practical agent for producing coronary hyper-emia. These studies were undertaken to evaluate theusefulness of adenosine for studies of the coronarycirculation of humans.

Since the identification of nucleoside adenosine inthe myocardium in 1929, many investigators haveshown that it is synthesized in the myocardium andthat interstitial adenosine concentration rises inresponse to increased metabolic oxygen requirementsand ischemia.15-17 Although the physiological role ofadenosine in regulating coronary blood flow is uncer-tain, exogenously administered adenosine is known tocause profound microvascular coronary dilation medi-ated by an adenosine receptor on the cell membraneof resistance vessel myocytes.18 Three properties ofadenosine have led to its extensive use in animalstudies designed to assess the effects of pathologicalstates (hypertrophy, infarction, and cardiomyopathy)on minimal total coronary resistance. First, intrave-nous or intracoronary adenosine can reliably increasecoronary conductance to maximal levels (that is, at orexceeding that produced by transient ischemia).10Second, the duration of action of adenosine is verybrief (5-30 seconds).17,19 Third, at high doses, aden-osine produces transmural vasodilation.20

Despite the widespread use of adenosine in animalstudies, concern over adenosine-induced hypoten-sion and heart block have hampered its use inhumans. In dogs, intravenous doses sufficient toproduce maximal coronary dilation also results in asignificant fall in systemic arterial blood pressure.20In addition, large doses of adenosine increase therefractory period of the sinoatrial and atrioventricu-lar nodes and can result in heart block.21 A prelimi-nary study using intracoronary adenosine in humansrevealed a strikingly high incidence of adenosine-induced conduction block in the atrioventricularnode.22The purpose of this study was to examine the

dose-response kinetics of intravenous and intracoro-nary adenosine administration in humans.

MethodsPatient Selection

Thirty-nine patients undergoing coronary angiogra-phy for the diagnosis of a chest pain syndrome were

studied. Two groups were studied. Thirty-one patientshad chest pain atypical for angina pectoris, normal ormildly stenotic (<50% diameter stenosis, visualinspection) epicardial coronary arteries, normal leftventricular function (contrast ventriculogram ejectionfraction, .50%), and normal coronary flow reserve(.3.5 peak/resting velocity ratio after a maximallyvasodilating dose of papaverine at a resting heart rateof .80 beats/min).2,78 Each of these patients alsounderwent M-mode and cross-sectional echocardiog-raphy and had normal left ventricular wall thickness(<11 mm septal and posterior wall diastolic thickness)and function. In addition, they had no other knowncondition that could have altered coronary microvas-cular tone or function, for example, collagen vasculardisease, anemia (hemoglobin, <11 g/dl), or priormyocardial infarction.23 Of these patients, eightreceived calcium channel antagonists, and 11 receivedaspirin within 24 hours of study.

Eight additional patients had microcirculatoryabnormalities (coronary flow reserve, .3.5 peak!resting blood flow velocity ratio after papaverineadministration) associated with ventricular hypertro-phy (n=5) or of uncertain etiology (n=3). Of thesepatients, one was taking ,B-adrenergic receptor antag-onists, two were taking calcium channel antagonists,and two received aspirin within 24 hours of study. Allpatients were studied after informed consent wasobtained, and all studies were approved by the Insti-tutional Review Board of the University of Minnesota.

Catheterization ProtocolCardioactive medications were continued until the

time of catheterization. Patients were brought to thecatheterization laboratory in a fasting state afterpremedication with diazepam (5-10 mg, orally).After vascular access was obtained, sodium heparinwas given intravenously in doses sufficient to prolongthe activated clotting time to more than 300 seconds.After diagnostic coronary and left ventricular angi-ography, a 20-MHz 3F coronary Doppler catheter(NuMed Inc., Hopkinton, N.Y.) was advancedthrough an 8F guiding catheter into the midportionof a coronary artery. The coronary artery studied wasrandomly assigned, but nondominant right coronaryvessels were not studied. The catheter position andrange gate were adjusted to obtain an adequatesignal of coronary blood flow velocity within thevessel. The technique has been described elsewhere.8Mean and phasic signals of coronary blood flowvelocity, arterial pressure obtained by the guidecatheter, and heart rate were continuously moni-tored. The arterial waveform obtained from theguide catheter was damped by the presence of thecoronary Doppler catheter, consequently, only meanarterial pressure could be accurately monitored.

Intracoronary papaverine. After measurement ofresting blood flow velocity, 6-12 mg papaverine (2mg/ml 0.9% saline) was injected through the guidecatheter into the coronary ostium, and the resultantincrease in coronary blood flow velocity wasrecorded. To confirm that any dose of papaverineproduced maximal coronary hyperemia, an addi-



ownloaded from


Wilson et al Adenosine and Coronary Circulation 1597

tional, larger dose was administered, and the result-ant hyperemic response was recorded. Blood flowvelocity was allowed to return to baseline levelsbetween doses of papaverine. In 31 patients, the leftcoronary artery was studied. In the remaining eightpatients, blood flow velocity was measured in theright coronary artery. The mean dose needed toproduce maximal hyperemia was 10±2 mg.

Intracoronary boluses of adenosine. After coronaryblood flow velocity had returned to basal levels,sequentially greater boluses of adenosine (2 ,gg/ml0.9% saline, MedCo Research, Los Angeles) wereinjected through the guide catheter into the coronaryostium. In the left coronary artery (n=25; 20 withnormal coronary flow reserve), boluses of 2, 4, 8, 12,and 16 gg were given. In the right coronary artery(n=8; five with normal coronary flow reserve), dosesof 2, 4, 8, and 12 ,jg were administered. Care wastaken to inject at a rate that would not cause reflux ofthe adenosine solution into the aorta. Immediatelybefore and at peak response after adenosine admin-istration, the electrocardiogram was recorded at afast paper speed (100 mm/sec) to assess adenosine-related changes in the PR, QRS, and QT intervals.

Intracoronary infusions of adenosine. The dose-response characteristics of intracoronary infusions ofadenosine were determined in the left coronaryartery of six patients with normal coronary flowreserve. Once coronary blood flow velocity hadreturned to normal after the papaverine dose-response study, the guiding catheter was filled withan adenosine solution, and adenosine was continuallyinfused into the left coronary at six doses: 10, 20, 40,80, 120, and 240 ,ug/kg/min. Infusions were per-formed with an infusion pump (Harvard Apparatus).Adenosine, prepared at 5.2, 20.8, 62.8, and 81.2gg/ml 0.9% saline, was infused at 1.91 or 3.82 ml/minto achieve the desired drug infusion rate. All infu-sions were continued for 2 minutes, and data wereobtained during the last 30 seconds of infusion. Tolimit the total study time, additional studies withintracoronary boluses or intravenous infusions werenot performed in these six patients.

Intravenous infusions of adenosine. After coronaryblood flow velocity returned to baseline, adenosine(1.0 mg/ml 0.9% saline) was infused into the femoralvein (n=5) or peripheral arm vein (n=25) at 70,ug/kg/min by a high-flow infusion pump. Twenty-fivepatients had normal coronary flow reserve; five hadreduced reserve. Two minutes later, the infusion wasincreased to 100 ,ug/kg/min, and 2 minutes later, itwas increased to 140 ,ug/kg/min. In addition, aninitial dose of 35 ,ug/kg/min was given to 17 patients.During the infusions, coronary blood flow velocity,mean arterial pressure, and heart rate were contin-uously recorded. Near the end of each infusion, anelectrocardiogram was recorded at 100 mm/sec paperspeed to assess changes in the electrocardiographicintervals. All infusions were continued for at least 2minutes, and measurements were obtained duringthe last 30 seconds of the infusion.

Data AnalysisIntracoronary bolus studies. The maximal change in

coronary blood flow velocity after a bolus of intra-coronary papaverine or adenosine was expressed asthe ratio of the maximal coronary blood flow velocity(after adenosine or papaverine administration) to theresting blood flow velocity (ACBFV). An index of thechange in total coronary resistance (ATCRI) wascalculated as the quotient of the peak hyperemiccoronary blood flow velocity (kHz) shift divided bymean arterial pressure at peak hyperemia (inmm Hg) and (coronary blood flow velocity at restdivided by arterial pressure at rest).The time course of the increase in coronary blood

flow velocity after intracoronary vasodilator adminis-tration was characterized by three parameters. T90%was defined as the time from the onset of injection tountil coronary blood flow velocity reached 90% of theeventual maximal increase in velocity. Tm. dur wasdefined as the time during which blood flow velocityremained 90% or more of the peak flow velocity. T10%was defined as the time from the onset of injectionuntil flow velocity returned to within 10% of basalflow velocity.

Intracoronary infusion studies. The change in coro-nary blood flow velocity during each infusion wascalculated as the quotient of the mean coronaryblood flow velocity (kHz shift) from 110-120 secondsafter the onset of the infusion (when a steady statehad been achieved) and the basal coronary bloodflow velocity. The changes in heart rate, arterialpressure, and an index of total coronary resistance(calculated as described above) were also assessedduring the last 10 seconds of the infusion.

Intravenous infusion studies. The change in coro-nary blood flow velocity during each intravenousadenosine infusion was calculated as the quotient ofthe peak mean blood flow velocity (kHz shift) duringthe infusion and the basal blood flow velocity. Incases where the blood flow velocity had markedcyclical variation during adenosine infusion (that is,.20% fluctuation in total coronary resistance indexat "steady state"), the existence of cyclic variationswas noted, but the average change in blood flowvelocity during the last 10 seconds of the infusion wasrecorded. The changes in heart rate, arterial pres-sure, and total coronary resistance were also assessedat the peak change in coronary blood flow velocity.

Statistical AnalysisDifferences between group means were tested by

analysis of variance. Paired differences were analyzedwith a paired t test. Linear correlation was assessedwith the least-squares method. Except where noted,all data are expressed as mean± SD. Statistical sig-nificance was defined as a p value of 0.05 or less.

ResultsIntracoronary Adenosine Boluses

Coronary blood flow velocity and resistance. Intra-coronary boluses of adenosine produced a dose-



ownloaded from



TABLE 1. Intracoronary Adenosine Boluses: Dose-Response Characteristics in Arteries With Normal Flow Reserve

ACBFV ATCRI AMAP AHR Percent T90% Tmax dur T10% APR AQRS AQT(x resting) (x resting) (mm Hg) (beats/min) maximal (sec) (sec) (sec) (msec) (msec) (msec)

Left coronary artery (n=20)Papaverine 4.8+1.0* 0.20±0.04* -8+5* 6±2* 18±5 30±11 111±17 4±12 -3±5 79±38*Adenosine (jig)

2 3.3±0.9*t 0.32±0.09*t -3±4*t -1±3t 0 11±3t 4±2t 22±6t -1±13 -1±4 -4±124 3.8±0.7*t 0.27±0.06*t -3+4*t 1±3t 35 12±3t 6±2t 26±6t -1±14 0±5 -7±18t8 4.2±0.7*t 0.23±0.03*t -5+4* 1±3t 50 14±3t 8±3t 32±5t 2±12 1±5 -2±20t12 4.5±0.8*t 0.22±0.04* -5÷5* 1±4t 80 14±5t 10±4t 36±2t -5±24 0±5 -3±19t16 4.6±0.7* 0.21±0.04* -7+5* 3±3t 90 13±3t 12±5t 37+7t 5±8 -1±3 -6±30t

Right coronary artery (n=5)Papaverine 4.5±0.9* 0.20±0.01* -15+11* 2±4 16±3 25±11 1,150+53 -11±11 -5±4 40±24*Adenosine (L£g)

2 3.4±1.3*t 0.29±0.11*t -10+10* 4±6 40 11±lt 5±3t 27±13t 10±8 0_3 -25_7*4 4.1_1.4* 0.24±0.09* -13±4** 3±5 40 11_lt 9-4t 33+41t -10±8 -5±4 -10+118 4.4±1.0* 0.21±0.04* -10±9*t 0±1 80 12±3t 13±5t 40±26t -7±13 -2±3 -7±1512 4.1±1.0* 0.20±0.01* -17±13*t 0±1 100 11±lt 16±11t 52±32t -3±18 -5±7 -10±6

Values are mean±SD.CBFV, coronary blood flow velocity; TCRI, total coronary resistance index; MAP, mean arterial pressure; HR, heart rate; percent

maximal, percent of patients with ATCRI within 10% of papaverine; T90%, time for CBFV to reach 90% of the maximum; Tm: dur, durationof >90% maximal CBFV; T10%, time for CBFV to return to within 10% of basal value.

*p<0.05 vs. basal condition; tp<0.05 vs. papaverine; *p<0.05 vs. left coronary.

dependent increase in coronary blood flow velocitysuch that a 16-,g bolus caused coronary hyperemiaequivalent to that of papaverine (Tables 1 and 2,Figures 1-3). After a 16-,ug bolus of adenosine intothe left coronary artery, 18 of 20 patients developedcoronary hyperemia to levels within 10% of thatproduced by intracoronary papaverine. In theremaining two patients, 12, 16, and 20 ,g adenosineequally caused coronary resistance to decrease to

0.20- and 0.25 * resting resistance, but not to within15% of the resistance after papaverine administra-tion (0.13- and 0.20. resting resistance, respective-ly). Hence, in 90% of patients, 16-,ug bolus into theleft coronary artery caused maximal vasodilation. Inthe right coronary artery, 12 ,g adenosine resulted inmaximal hyperemia (within 10% of that produced bypapaverine) in all patients. The dose-response rela-tion was similar in normal and abnormal arteries.

TABLE 2. Intracoronary Adenosine Boluses: Dose-Response Characteristics in Arteries With Reduced Flow Reserve

ACBFV ATCRI AMAP AHR Percent(x resting) (x resting) (mm Hg) (beats/min) maximal

Left coronary artery (n=5)Papaverine 2.8+0.6* 0.35+0.09* -9+3* 1+3Adenosine (tg)

2 2.1±0.8 0.52+0.19 -3±1 1±1 04 2.2±0.5 0.43±0.08 -9±3 1±3 08 2.6±0.7 0.38±0.13 -8+9 0±0 6012 2.8±0.6 0.34±0.07 -9±5 -2+2 10016 2.9±0.6 0.33±0.07 -9±5 -1+2 100

Right coronary artery (n=3)Papaverine 2.9±0.1* 0.33±0.05* -6±8 0±3Adenosine (,ug)

2 2.7±0.7* 0.37±0.08* -1+3 -2±1 04 3.0±0.1* 0.32+0.01* -3±1 -2±2 338 3.0±0.2* 0.30±0.01* -9+7* -3±4 10012 2.9±0.1* 0.30±0.01* -11±6* -3±4 100

Values are mean±SD.CBFV, coronary blood flow velocity; TCRI, total coronary resistance index; MAP, mean arterial pressure; HR, heart

rate; percent of patients with ATCRI within 10% of papaverine.*p<0.05 vs. basal condition.



ownloaded from



6

5 - Dose

ACBFV 38p(x resting) aLE

0 10 20 30 40 50Time (seconds)

FIGURE 1. Plot (from allpatients) of the change in coronaryblood flow velocity (ACBFV) after progressively greater dosesof intracoronary adenosine. Larger doses caused more pro-longed hyperemia.

The correlation of the maximal change in coronaryblood flow velocity after adenosine or papaverine wasr=0.90 (SEE, ±0.46) peak/resting velocity. A bolusdose of adenosine (2-16 gg) was repeated in 12patients (n= 15 measurements). The change in bloodflow velocity measured after the first bolus wasclosely correlated with the change in velocityobserved after the second injection (r=0.95; SEE,±0.07 peak/resting velocity).The onset of maximal hyperemia (Tg9%) was more

rapid after adenosine compared with that of papav-erine. The duration of maximal hyperemia (Tmax dur)and time required for blood flow velocity to return tobasal levels (T10%) increased with the dose of adeno-sine, but at any dose, both parameters were muchshorter than those observed after intracoronarypapaverine administration (Tables 1 and 2, Figures 1and 2).

Systemic hemodynamics. Intracoronary boluses ofadenosine produced small, brief, dose-dependentreductions in mean arterial pressure. The fall in

* adenosineo papavenne

ACBFV(Xresting) 3

0 20 40 60 80 100 120 140

time(sec)

FIGURE 3. Plot (from allpatients) ofthe change in coronaryblood flow velocity (ACBFV) after an intracoronary bolus ofadenosine (16 big) or papaverine (maximally vasodilatingdose, 10+2 mg). Both agents caused a marked increase incoronary blood flow velocity, but the response to adenosinewas much shorter than that elicited by papaverine.

arterial pressure after an 8-, 12-, or 16-,ug bolus wassimilar in magnitude to that found after intracoro-nary papaverine (mean dose, 10±2 mg). The fall inarterial pressure for an equal dose of adenosine wassignificantly greater when adenosine was injectedinto the right compared with the left coronary artery(Tables 1 and 2). The fall in arterial pressure after a12-,ug bolus into the right coronary artery was notdifferent from that after a 16-,g bolus into the leftcoronary artery, suggesting that doses with similarvasodilator effects caused similar changes in arterialpressure.There was no significant change in heart rate after

any dose of adenosine was injected into the right orleft coronary artery. Intracoronary papaverine, how-ever, caused a small, but significant, increase in heart

Phasic 5- mCBFV v -

5 peak CBFV 3.0 3.5 4.9 4.9Mer resti__CBF

~ shzOf 0- t2ug I t

Aorft 200- Adenosine 4pg B8g9 12pg

1 sec 30 sec

4.8

6pg

4.9

cpdpaverine 10mg

FiGURE 2. Record obtainedfrom a patient studied in the catheterization laboratory. Top two panels: Phasic and mean coronaryblood flow velocity (CBFV) in the left anterior descending coronary artery. Bottom three panels: Aortic blood pressure,intracoronary blood pressure (from the Doppler catheter), and electrocardiogram (ECG). Progressively greater intracoronaryboluses of adenosine caused stepwise increases in CBFVwithout significant changes in blood pressure or heart rate. Intracoronarypapaverine (far right) caused hyperemia similar in magnitude to that caused by 8-16 pg of intracoronary adenosine.

0-

200-

0-

Pre~e(mrnmHg)

CororryPressLre(tHg)

ECG



ownloaded from



TABLE 3. Intracoronary Adenosine Infusions: Dose-Response Characteristics

ACBFV ATCRI AMAP AHR APR AQRS AQT Percent Cyclic(x resting) (x resting) (mm Hg) (beats/min) (msec) (msec) (msec) maximal variation

Papaverine 4.0±0.2* 0.22±0.01* -13+2* 0+3 -3±4 1±2 55+8*Adenosine (gg/min)

10 1.9±0.6*t 0.66+0.14*t -5±3 -4±3 -5±2 -2+2 -2±3 0 020 2.8±0.4*t 0.34+0.05*t -8±3 -3+2 2+2 -2+2 -3+4- 17 040 3.8±0.4* 0.25±0.03* -2+3 -6+3 -2±2 -4±4 2±6 33 080 4.4±0.1* 0.22±0.01* -4±4 -7±4 -5±2 -2±3 -7±4 100 0120 4.1±0.3* 0.23±0.03* -9±3* -5±4t -6±2 2±2 -6±7 100 0240 4.2±0.2* 0.23±0.03* -6±2* -6±4t -2±2 -2±3 0±6 100 0

Values are mean±SD; n=6.CBFV, coronary blood flow velocity; TCRI, coronary vascular resistance index; MAP, mean arterial pressure; HR, heart rate; percent of

patients with ATCRI within 10% of papaverine; percent of patients exhibiting cyclic changes in CBFV (see text).*p<0.05 vs basal conditions; tp<0.05 vs. papaverine.

rate when injected into the left (but not the right)coronary artery.

Electrocardiographic changes. Intracoronary bolusesof adenosine did not significantly change the PR,QRS, or QT intervals on the electrocardiogram, evenwhen the injected dose was sufficient to cause maximalcoronary hyperemia or when the drug was injecteddirectly into the right coronary artery. Hence, therewas no evidence of significant sinoatrial or atrioven-tricular node dysfunction after administration of intra-coronary boluses of adenosine. Papaverine causedsignificant prolongation of the QT interval but did notchange the other electrocardiographic parameters.

Intracoronary Adenosine InfusionCoronary blood flow and resistance. There was a

dose-dependent increase in coronary blood flowvelocity during the intracoronary infusion of adeno-sine into the left coronary artery (Table 3). Allpatients had increased blood flow velocity maximally(equivalent to that achieved with intracoronarypapaverine) at infusion rates of 80 ,ug/min or less.Cyclic variations in blood flow velocity at submaximalinfusion rates (see below) were not observed.The time from the onset of a 240-,ug/kg/min

infusion until blood flow velocity was within 90% ofthe eventual peak was 47± 31 seconds. The time fromthe offset of the infusion until blood flow velocity waswithin 10% of basal values was 143+75 seconds.

Systemic hemodynamics. The heart rate was notsignificantly changed during the intracoronary infu-sions (Table 3), even at infusion rates fourfold higherthan those needed to cause maximal hyperemia (thatis, 240 ,ug/min). Mean arterial blood pressure fellminimally, but significantly, during the 120- and240-,gg/min infusions (mean change, -9±7 and-6±5 mm Hg, respectively).Electrocardiographic changes. None of the electro-

cardiographic intervals changed during the intracor-onary infusions of adenosine (Table 3).

Intravenous Adenosine InfusionCoronary blood flow velocity and resistance. Intrave-

nous adenosine infusions resulted in a dose-

dependent increase in coronary blood flow velocityand decrease in total coronary resistance such thatminimal total coronary resistance during the 140-jttg/kg/min infusion was not significantly different fromthat elicited by papaverine (Table 4, Figures 4 and 5).In 11 normal patients (44%), an infusion rate of 100,ug/kg/min was sufficient to decrease total coronaryresistance to within 10% of the minimal resistanceafter papaverine. In 10 of the remaining patients, aninfusion rate of 140 ,ug/kg/min was needed to reduceresistance to within 10% of the papaverine-inducedminimal resistance. Four patients, however, failed toachieve maximal vasodilation with an infusion rate of140 1g/kg/min. In two, the 140-,ug/kg/min infusionreduced coronary resistance to 0.22-0.27 * restingresistance, but that was at least 10-30% greater thanthe minimal resistance caused by papaverine. In theremaining two, the infusion failed even to decreasecoronary resistance to within 30% of the papaverine-induced minimal resistance. An increase in the infu-sion rate to 180-200 ,ug/kg/min failed to elicit anygreater vasodilation. Hence, 84% of normal patientsachieved maximal vasodilation at an infusion rate of140 j.g/kg/min or less, 92% achieved near-maximalhyperemia, but 8% failed to develop hyperemiawithin 30% of that produced by papaverine. Thedose-response relation was similar in a smaller groupof patients with microvascular dysfunction.At the lower infusion rates (70-100 yg/kg/min),

coronary blood flow velocity often rose and fell in acyclic pattern with a cycle length of about 30 seconds(Table 4, Figure 5). Two factors suggest that thecyclic variation in coronary conductance resultedfrom a cyclical variation in adenosine concentrationin the blood perfusing the myocardium. First, whenthe coronary infusion rate was increased, hyperemiabecame sustained at the maximal level. Second,intracoronary bolus injections of adenosine at thetime when coronary blood flow velocity had receded,resulted in a prompt increase in blood flow velocity.To further investigate the mechanism of this cyclicphenomena, we changed the infusion site from theperipheral vein to the femoral vein in three patients;the pattern did not change. There was no difference



ownloaded from



TABLE 4. Intravenous Adenosine Infusions: Dose-Response Characteristics

ACBFV ATCRI AMAP AHR APR AQRS AQT Percent Cyclic(x resting) (x resting) (mm Hg) (beats/min) (msec) (msec) (msec) maximal variation

Arteries with normal flow reserve (n =25)Papaverine 4.8+0.9* 0.20+0.03* -9+7* 5+5* 2±14 -3±5 74±13*Adenosine (gg/kg/min)

35 1.0±0.1t 0.97±0.08t 1±6t -1±5t -7±12 3+3 5±13t 0 070 3.1±1.2*t 0.40±0.26*t -5+5*t 8±7* 1±9 2±4 7±19t 16 48100 4.4±0.9*t 0.23±0.06* -7+6* 17±11*t -3±9 3±4 1±25t 44 52140 4.4±0.9* 0.24±0.01* -6±7* 24±14*t 1±6 1±4 -23±39t 84 8

Arteries with abnormal flow reserve (n=5)Papaverine 2.8±0.4 0.34±0.07 -8±5 0±3 1±7 0±5 52±27Adenosine (,ug/kg/min)

35 1.0±0.1 1.00±0.10 -7±3 -1±3 4±8 0+1 4±6 0 070 2.0±1.0 0.59±0.33 -5±4 2±3 5±4 4+3 -4±14 40 0100 2.4±1.2 0.51±0.38 -8±7 12±10 0±5 3±4 -19±21 60 40140 2.9±0.4 0.33+0.03 -3±6 12±10 2±8 -2±4 -27±28 80 20

Values are mean±SD.CBFV, coronary blood flow velocity; TCRI, coronary vascular resistance index; MAP, mean arterial pressure; HR, heart rate; percent of

patients with ATCRI within 10% of papaverine; percent of patients exhibiting cyclic changes in CBFV (see text).*p<0.05 vs. basal conditions; tp<0.05 vs. papaverine.

in the frequency of cyclic hyperemia in patients whodid and did not receive aspirin and in those withnormal and abnormal coronary vessels.The average time from the onset of infusion until

the maximal response was achieved was 84+±46 sec-onds (range, 23-125 seconds). The time from offsetof infusion (140 g.g/kg/min) until coronary bloodflow velocity returned to basal levels was 145+±67seconds (range, 54-310 seconds).

Systemic hemodynamics. Intravenous infusions ofadenosine produced a dose-dependent fall in meanarterial blood pressure and a rise in heart rate (Table2, Figure 6). At 140 ,tg/kg/min in normal patients,the heart rate rose 24+14 beats/min, and meanarterial pressure fell 6±7 mm Hg. In all patients, themean arterial pressure during the 140-,ug/kg/mininfusion remained more than 50 mm Hg.

Electrocardiographic changes. The PR, QRS, andQT intervals did not significantly change during theinfusions. Of importance, however, one normalpatient developed 2 cycles of 2:1 narrow-complexheart block during the 140-gg/kg/min infusion. Theinfusion was continued without further arrhythmia,and the PR interval did not change just before thetermination of the infusion. The patient had receiveddiltiazem and a ,B-adrenergic receptor antagonistcontinuously before catheterization.

SafetyOther than the one, brief episode of atrioventric-

ular block during intravenous adenosine infusion, nopatient developed arrhythmias, systemic hypotension(systolic blood pressure <90 mm Hg, diastolic bloodpressure <40 mm Hg), or electrocardiographicchanges suggestive of ischemia (>0.1 mV ST segmentdepression 80 msec after the "J' point). Many

patients developed a vague sensation of chest warmthor flushing during the intravenous infusion.

DiscussionThese data demonstrate in humans that adenosine,

given by the intracoronary or intravenous route, cancause near-maximal coronary vasodilation, equiva-lent in magnitude to that generated by intracoronarypapaverine, without causing clinically importantchanges in systemic hemodynamics or on the electro-cardiogram. In the left coronary artery, boluses of 16jig or more (12 gg in the right coronary artery) or acontinuous intracoronary infusion of 80 ,ug/kg/min ormore were needed to reliably cause maximal coronaryvasodilation. Most, but not all, patients developedmaximal vasodilation during intravenous infusions of140 ,ug/kg/min. The advantages of adenosine overpapaverine or dipyridamole are its very short duration

6

5

4

ACBFV(X resting)

3

2

0

- mean ± SD* p<.05 vs. papaverine

T

35 70 100 140

Adenosine (iglkgImin)

TT

Papaverine

FIGURE 4. Bar graph of change in coronary blood flowvelocity (ACBFV) during intravenous adenosine infusion andafter intracoronary papaverine.

T--7-- by guest on A

pril 13, 2017http://circ.ahajournals.org/

Dow

nloaded from



0-

5-

Ded, CBFV 12 4s 4.5

-A---1 i _ ;

0- - |l200- 35 pg I1 M~snm 1140 k mn

0-200-

3w

0-

30 sec

FIGURE 5. Record obtainedfrom the same patientpresented in Figure 2. Top two panels: Phasic and mean coronary bloodflowvelocity (CBFV) in the left anterior descending coronary artery. Bottom three panels: Aortic blood pressure, intracoronary bloodpressure (from the Doppler catheter), and electrocardiogram (ECG). Intravenous infusion of adenosine at 35 pg/kg/min failed tosignificantly alter CBFV When the infusion was increased to 70 pg/kg/min, CBFV rose and fell in a cyclical pattem. When theinfusion was further increased to 100 and 140 pg/kg/min, CBFV stabilized at a level similar to that achieved with a maximallyvasodilating dose of intracoronary papaverine.

of action, the absence of QT interval prolongation onthe electrocardiogram, and its efficacy when given bythe intravenous or intracoronary routes.

Potential Methodological ProblemsThere are several potential problems that should

be considered when interpreting these data. Thoserelated to measurement of coronary blood flow usinga coronary Doppler catheter have been discussed indetail elsewhere.8 Of importance, all patientsreceived intracoronary nitroglycerin to maximallydilate the vessel containing the Doppler catheter,and the guide catheter was withdrawn from thecoronary ostium at peak hyperemia to ensure thatmaximal blood flow into the ostium was not impededby the catheter.A second potential problem is that we purposely

did not control heart rate, arterial pressure, or leftventricular preload during these studies. In thisstudy, we administered two drugs that change botharterial pressure and heart rate. We have recentlyshown in humans that increases in heart rate aug-ment resting coronary blood flow without alteringmaximal hyperemic flow (reducing the peaklrestingvelocity ratio), whereas increases in arterial bloodpressure cause proportionately equal increases inresting and hyperemic blood flow (preserving thepeak/resting velocity ratio).24 We were careful, how-ever, to compare the changes in blood flow velocityelicited by adenosine and papaverine to a single basalcoronary blood flow velocity that was measuredbefore any drug was given. To compensate forchanges in arterial pressure, we expressed the vaso-dilation caused by each agent as the fractionalchange in resting coronary resistance (total coronaryresistance index). Hence, changes in systemic hemo-dynamics should not have affected comparisonsbetween different doses of adenosine or betweenadenosine and papaverine.A third potential limitation is that we primarily

studied patients with normal or moderately narrowedcoronary arteries and normal ventricular function.

Although not supported by prior studies in animals,the vasodilator response to adenosine may be selec-tively impaired by certain cardiomyopathies (forexample, viral or hypertrophic cardiomyopathy), bysevere coronary obstructions, or by other acquiredabnormalities.25 We observed a reduced vasodilatorresponse to adenosine (but not papaverine) in twoapparently normal humans. Other investigators havealso reported sporadic instances of reduced responseto adenosine in animals, the mechanism of which isunknown.26 Moreover, other patients have beenshown to have reduced sensitivity to dipyridamole,which causes vasodilation indirectly by increasinginterstitial adenosine concentrations.7,9 The lack of aresponse to adenosine with an intact response topapaverine may signify a specific defect in the micro-circulation that could have clinical significance. Inaddition, the intracoronary dose of adenosine

10 F

,Ameanarterialpressure(mmHg)

-5

-10

-15

-20

40

30

Aheartrate

(beats/min)

20

10

0

-10

'IT1Adenosine (pg&g/min) Papaverine

35 70 100 140

T

mean ± SD' pc05 vs. papaverine

FIGURE 6. Bargraphs ofchange in mean aortic bloodpressureand heart rate duping intravenous infusions of adenosine.

PhasicCBFV

(kHz shift)

MeanCBFV

(kHz shift)

AorticPressure

CoronaryPressure

EOG

..M InA 1

Nosm 9 1 r

T-- -5. -- .



ownloaded from



required to cause maximal hyperemia may be signifi-cantly higher in patients with increased ventricularmass. Our limited studies in patients with microvascu-lar dysfunction, however, failed to show any differencein the dose-response relation. Further studies will berequired to define the vasodilator properties andsafety of adenosine in humans with other cardiovas-cular disease (for example, sinoatrial or atrioventric-ular node disease, baroreceptor dysfunction, or ven-tricular hypertrophy).

Comparison With Previous StudiesAside from an assessment of the dose-response

characteristics of adenosine in humans, the twomajor findings of this study are an apparent speciesdifference in the response to intravenous adenosineand an observation in humans that coronary resis-tance fluctuates widely at submaximal adenosineinfusion rates. Rembert et a120 reported that anintravenous adenosine infusion rate of 1,000 ,ug/kg/min was required to elicit maximal transmural coro-nary hyperemia. Lesser infusion rates caused maxi-mal subendocardial vasodilation, but conductance ofblood to the subepicardial muscle was not maximal.In humans, however, we found that an infusion ofadenosine at only 140 gg/kg/min decreased coronaryresistance to within 14+8% of that induced by papav-erine, with a small concomitant increase in arterialblood pressure and a moderate increase in heart rate.These data suggest that humans are more sensitivethan are dogs to the vasodilator properties of aden-osine or that the rate of adenosine elimination isslower in humans.Although the species differences in response to

adenosine are clear, it should be emphasized that anadenosine infusion of 140 gg/kg/min in humans maynot have produced absolutely maximal vasodilationin the subepicardial myocardium in all patients. In16%, the minimal resistance during the 140-gg/kg/min infusion was more than 10% greater than thatproduced by papaverine. Similar intraspecies differ-ences in adenosine responsiveness have beenreported in dogs.26 Although higher infusion ratesmay possibly cause maximal hyperemia, the safety ofhigher doses needs to be established. Our experiencedemonstrated that an infusion rate of 140 gg/kg/minsignificantly increased the heart rate; higher infusionrates may not be well tolerated in a significantfraction of patients.We also found that intravenous infusion rates just

less than those required to produce a maximal fall incoronary resistance frequently cause a characteristicpattern of widely fluctuating coronary resistance(that is, >20%) that has not been reported in priorstudies in animals. The pattern was not affected bychanges in the site or concentration of infusion anddisappeared when the infusion rate was increased.We believe that this occurred in humans and not indogs because of the short half-life of adenosine andthe longer circulation time in humans that weigh70-100 kg compared with that in dogs that weigh

20-30 kg. Initially, the adenosine was infused into avein with relatively slow blood flow, resulting in ahigh concentration of adenosine. When the adeno-sine reached the systemic circulation, vasodilationand an increase in vein blood flow occurred. Theadenosine, continuously administered at the sameinfusion rate, was then diluted to a lower concentra-tion. Simultaneously, the adenosine administered atthe beginning of the infusion was metabolized. Thelower plasma concentration of adenosine that thenreached the coronary and systemic vasculature pro-duced less vasodilation, allowing blood flow todecrease. As venous blood flow returned to the basallevel, the plasma concentration of adenosine againrose and the cycle started again. If the circulationtime was faster (as in the smaller dog), then moreadenosine would survive more than one circulation(particularly if the elimination rate was slower), andcyclic variations would not occur. Similarly, if theinfusion rate was increased, the trough level ofadenosine would still be sufficient to cause maximalvasodilation.Two factors support this explanation. Additional

intracoronary adenosine, given when blood flowvelocity was decreasing, resulted in a prompt rise inflow velocity, suggesting that the fall in blood flowresulted from a lower arterial concentration of aden-osine rather than decreased local effect. Second,there is inferential evidence that a fraction of theadenosine was metabolized as it passed from thevenous infusion site to the myocardium. Under nor-mal circumstances, the left coronary artery receives2-3% of the total cardiac output. Unless there wassignificant metabolism, one would anticipate that theintracoronary dose needed to cause maximal hyper-emia would be 2-3% of the intravenous dose. Wefound, however, that the intracoronary infusion rateneeded to cause maximal hyperemia was less than1% of the intravenous dose, suggesting that someadenosine was metabolized before it arrived in thecoronary artery.An additional explanation for cyclic hyperemia is

that adenosine, at low doses, caused platelet activa-tion that secondarily reduced hyperemic blood flow.This is unlikely because in vitro studies have shownthat adenosine reduces platelet aggregation.27 Fur-thermore, animal models in which platelet-mediatedcyclic flow variation has been observed use severecoronary stenoses, cause cyclic reductions of restingcoronary blood flow (not hyperemic flow), and tendto have a much longer periodicity.28,29 In addition,pretreatment with aspirin, a drug that abolishes cyclicvariations in animal models, failed to alter the inci-dence of flow variations in humans.The clinical importance of the phenomenon is that,

in contrast to dogs, submaximal doses of adenosine inhumans may not cause sustained submaximal hyper-emia. Consequently, if adenosine is used as an adju-vant to 2Okl1 scintigraphy, injection of thallium at thelow ebb of coronary hyperemia may result in falselynormal findings.30 Fortunately, the vast majority of



ownloaded from



patients achieved sustained maximal coronary hyper-emia at an infusion rate of 140 ,g/kg/min. Thesestudies suggest that an infusion rate of 140 ,ug/kg/min should be used unless the patient developshypotension or marked tachycardia at a lower dose.

Prior investigators have reported that in humansmuch larger doses of intracoronary adenosine arerequired to cause maximal coronary vasodilation.Zijlstra et a122 found that intracoronary boluses of50-800 jig were required to produce maximal coro-nary hyperemia.22 The magnitude of maximal coro-nary blood flow obtained with papaverine or adeno-sine, however, was very low (1.6±0.3 peak/restingcoronary blood flow velocity), suggesting that manyof the patients studied had microvascular disease orother illnesses that may have altered the response toadenosine. They also reported a high incidence oftransient heart block that may have been related tothe large dose of adenosine administered concur-rently with agents that depress atrioventricular andsinoatrial node function (for example, diltiazem,,B-adrenergic receptor antagonists). Other investiga-tors31-34 have also studied the effects of adenosine onthe peripheral or coronary circulation of humans.The lack of dose-response data or problems in mea-surement techniques make the interpretation of theirfindings difficult.

Potential Uses ofAdenosine in HumansWe studied the effects of adenosine on the coro-

nary circulation of humans for two reasons: onerelated to clinical application and the other related tobasic study of the coronary circulation. It wasrecently suggested that intravenous adenosine infu-sions may supplant dipyridamole in providing coro-nary vasodilation in conjunction with 20`T1 scinti-graphic studies.30 The ultrashort half-life ofadenosine compared with that of dipyridamole maylessen the risk of prolonged coronary ischemia due tovasodilator-induced coronary "steal" and reduce thetime required for redistribution of the isotope. Foradenosine to replace dipyridamole, however, it mustsafely cause coronary hyperemia sufficient to differ-entiate myocardium perfused by a stenotic arteryfrom that supplied by a normal vessel. These studiessuggest that clinically tolerable doses of adenosinecause near-maximal coronary hyperemia in most, butnot all, patients. Although the importance of obtain-ing absolutely maximal (as opposed to near-maximal)coronary vasodilation may be insignificant becauseepicardial coronary stenoses first cause vasodilationof the subendocardium (the layer most sensitive tothe vasodilator properties of adenosine), further clin-ical studies will be needed to define the comparativesensitivity of 20lfl scintigraphy obtained using aden-osine or dipyridamole.35,36The second clinical use of adenosine is in measur-

ing coronary flow reserve in the catheterization lab-oratory. Intracoronary papaverine, the agent mostcommonly used to measure flow reserve, cannot begiven as a continuous infusion (making difficult its

use with measurement techniques with poor tempo-ral resolution such as digital subtraction angiographicmethods) and can cause ventricular tachycardia.7Intracoronary adenosine (bolus or infusion) may bemore advantageous than papaverine because it canbe given as a continuous infusion without resulting insystemic accumulation, it has no important effects onthe electrocardiogram at the doses tested, and it hasa short half-life that reduces the potential sequelae ofany toxicity (for example, heart block). Similarly, theability to cause prolonged coronary hyperemia usingcontinuous intracoronary infusions of adenosineshould enhance many physiological studies of thecoronary circulation.Although the precise role of adenosine in regulat-

ing coronary blood flow is not certain, several aspectsof its mechanism of action in causing microvascularvasodilation are known and should be consideredwhen the drug is used in diagnostic or therapeuticstudies. The myocytes of coronary resistance vesselshave an adenosine receptor (A2) on their cell mem-brane.37 When combined with adenosine, the plasmamembrane-bound receptor protein causes anincrease in adenylate cyclase activity and a subse-quent increase in intracellular cyclic AMP. Thereceptor activity is inhibited by theophylline-typecompounds (that is, methylxanthines), and theophyl-line has been shown in animals to reduce adenosine-induced hyperemia by 41-101%.38-41 Hence, patientsreceiving theophylline, caffeine, or other methylxan-thines may not develop maximal coronary hyperemiaduring adenosine infusion.A second effect of adenosine is presynaptic inhibi-

tion of norepinephrine release from sympathetic nerveterminals. Studies by Johannsen et a142 and othershave demonstrated that adenosine reduces, but doesnot fully override, coronary vasoconstriction duringneural sympathetic stimulation. Studies of the effect ofsympathetic neural stimulation on the coronary circu-lation (for example, cold pressor test) should take intoaccount the inhibitory effect of adenosine.

SafetyAdenosine has two important potential side

effects. Depression of sinoatrial or atrioventricularnode function occurs in a dose-dependent fashion inanimals and humans.21 The doses previously reportedto cause heart block in humans were far in excess ofthose given in this study, and many of the patientsconcomitantly had received other drugs that depressatrioventricular node function (for example, dilt-iazem and /3-adrenergic receptor antagonists). Weobserved one episode of 2:1 atrioventricular block(two cycles) during an intravenous infusion (140gg/kg/min) in one patient who had previouslyreceived diltiazem and atenolol. Although the epi-sode was clinically insignificant, more prolongedatrioventricular node block may occur during anintravenous infusion in patients with preexisting con-duction system disease or in patients taking dipyr-idamole. To avoid important heart block, the dose of



ownloaded from



intravenous infusions should be slowly titrated, andpatients should be instructed to discontinue dipyri-damole therapy. In addition, significant anemia maydecrease adenosine metabolism and intensify or pro-long its effect. Unlike papaverine, however, adeno-sine does not change the QT interval. Consequently,the 0.5% incidence of polymorphous ventriculartachycardia observed after intracoronary papaverinemay be avoided by the routine use of adenosine in themeasurement of flow reserve.

Intravenous infusions of adenosine can cause asignificant fall in arterial blood pressure and a reflex-mediated rise in heart rate. Although these minorhemodynamic changes were well tolerated in ourpatients, significant hypotension may result inpatients with an impaired baroreflex (and, hence,reduced reflex tachycardia) or with intravascularvolume depletion, or it may result in those treatedconcomitantly with dipyridamole.

Finally, although these data suggest that adenosinecan be safely administered to patients studied in thecatheterization laboratory, we studied only 39patients. Study in a much larger population will berequired to fully ascertain the incidence and predis-posing factors of any adverse effects. If more wide-spread use parallels our experience, adenosineshould greatly facilitate studies of the coronary cir-culation of humans.

AcknowledgmentsWe thank the fellows, faculty, and staff of the

University of Minnesota Hospital Cardiac Catheter-ization Laboratory for their patient assistance duringthe performance of these studies. We also thank KimBruce and Susan Meyer, BS, for their expert techni-cal assistance. Last, we thank Robert J. Bache, MD,for his thoughtful review of the manuscript.

References1. Zijlstra F, van Ommeren J, Reiber JHC, Serruys PW: Does the

quantitative assessment of coronary artery stenosis dimensionspredict the physiologic significance of a coronary stenosis?Circulation 1987;75:1154-1161

2. Wilson RF, Marcus ML, White CW: Prediction of the physi-ologic significance of coronary arterial lesions by quantitativelesion geometry in patients with limited coronary artery dis-ease. Circulation 1987;75:723-732

3. Legrand V, Hodgson JMcB, Bates ER, Aueron FM, ManciniGB, Smith JS, Gross MD, Vogel RA: Abnormal coronary flowreserve and abnormal radionuclide exercise test results inpatients with normal coronary arteries. J Am Coll Cardiol1985;6:1245-1253

4. Gould KL, Lipscomb K, Hamilton GW: Physiologic basis forassessing critical coronary stenosis: Instantaneous flowresponse and regional distribution during coronary hyperemiaas measures of coronary flow reserve. Am J Cardiol 1974;33:87-94

5. Picano E, Simonetti I, Carpeggiani C, Lattanzi F, Macerata A,Trivella MG, Marzilla M, L'Abbate A: Regional and globalbiventricular function during dipyridamole stress testing.Am JCardiol 1989;63:429-432

6. Marcus ML, Harrison DG, White CW, McPherson DD,Wilson RF, Kerber RE: Assessing the physiologic significanceof coronary obstructions in patients: Importance of diffuse

undetected atherosclerosis. Prog Cardiovasc Dis 1988;31:39-56

7. Wilson RF, White CW: Intracoronary papaverine: An idealvasodilator for studies of the coronary circulation in conscioushumans. Circulation 1986;73:444-452

8. Wilson RF, Laughlin DE, Ackell PH, Chilian WM, HolidaMD, Hartley CJ, Armstrong ML, Marcus ML, White CW:Transluminal, subselective measurement of coronary arteryblood flow velocity and vasodilator reserve in man. Circulation1985;72:82-92

9. Rosen JD, Simonetti I, Marcus ML, Winneford MD: Coronarydilation with standard dose dipyridamole and dipyridamolecombined with handgrip. Circulation 1989;79:566-572

10. Bookstein JJ, Higgins CB: Comparative efficacy of vasodila-tory methods. Invest Radiol 1977;12:121-127

11. Kramer WG, Romagnoli A: Papaverine disposition in cardiacsurgery patients and the effect of cardiopulmonary bypass. EurJ Clin Pharmacol 1984;27:127-130

12. Wilson RF, White CW: Serious ventricular dysrhythmias afterintracoronary papaverine. Am J Cardiol 1988;62:1301-1302

13. Homma S, Gilliland Y, Guiney TE, Strauss HW, Boucher CA:Safety of intravenous dipyridamole for stress testing withthallium imaging. Am J Cardiol 1987;59:152-154

14. Becker LC: Conditions for vasodilator-induced coronary stealin experimental myocardial ischemia. Circulation 1978;57:1103-1110

15. Klaubunde RE: Dipyridamole inhibition of adenosine metab-olism in human blood. Eur J Pharmacol 1983;93:21-26

16. Drury AN, Szent-Gyorgyi A: The physiological activity ofadenosine compounds with especial reference to their actionon the mammalian heart. J Physiol 1929;68:213-237

17. Berne RM: The role of adenosine in the regulation of coro-nary blood flow. Circ Res 1980;47:807-817

18. Olsson RA, Davis CJ, Khouri EM, Patterson RE: Evidence foran adenosine receptor on the surface of dog coronary myo-cytes. Circ Res 1976;39:93-98

19. Soderback U, Sollevi A, Fredholm BB: The disappearance ofadenosine from blood and platelet suspension in relation toplatelet cyclic AMP content. Acta Physiol Scand 1987;129:189-194

20. Rembert J, Boyd LM, Watkinson WP, Greenfield JC: Effect ofadenosine on transmural myocardial blood flow distribution inthe awake dog. Am J Physiol 1980;239(Heart Circ Physiol8):H7-H13

21. DiMarco JP, Sellers TD, Berne RM, West GA, Belardinelli L:Adenosine: Electrophysiologic effects and therapeutic use forterminating paroxysmal supraventricular tachycardia. Circula-tion 1983;68:1254-1263

22. Zijlstra F, Juilliere Y, Serruys PW, Roelandt JRTC: Value andlimitations of intracoronary adenosine for the assessment ofcoronary flow reserve. Cathet Cardvasc Diagn 1988;15:76-80

23. Taucheret M, Hilger HH: Application of the coronary reserveconcept to the study of myocardial perfusion, in Schaper W(ed): The Pathophysiology of Myocardial Perfusion. New York,Elsevier/North Holland, 1979, pp 141-167

24. McGinn AL, White CW, Wilson RF: Interstudy variability incoronary flow reserve: The importance of heart rate, arterialpressure, and ventricular preload. Circulation 1990 (in press)

25. Watt AH: Hypertropic Cardiomyopathy: A disease ofimpaired adenosine-mediated autoregulation of the heart.Lancet 1984;1:1271-1273

26. Olsson RA, Khouri EM, Bedynek JL, McLean J: Coronaryvasoactivity of adenosine in the conscious dog. Circ Res1979;45:468-478

27. Ohisalo JJ: Regulatory functions of adenosine. Med Biol1987;65:181-191

28. Folts JD, Crowell EB, Rowe GG: Platelet aggregation inpartially obstructed vessels and its elimination with aspirin.Circulation 1976;54:365-370

29. Ashton JH, Benedict CR, Fitzgerald C, Raheja S, Taylor A,Campbell WB, Buja LM, Willerson JT: Serotonin as a medi-ator of cyclic flow variations in stenosed canine coronaryarteries. Circulation 1986:73:572-578



ownloaded from



30. Staudacher RA, Mahmarian JJ, Hixon JB, Boyce TM, PacificoA, Kugiyama K, Verani MS: Adenosine thallium-201 scintig-raphy: Feasibility, safety and initial results in man (abstract). JAm Coll Cardiol 1989;13:161A

31. Sylven C, Jonzon B, Edlund A: Angina pectoris-like painprovoked by i.v. bolus of adenosine: Relationship to coronary

sinus blood flow, heart rate and blood pressure in healthyvolunteers. Eur Heart J 1989;10:48-54

32. Torssell L, Ekestrom S, Sollevi A: Adenosine-inducedincrease in graft flow during coronary bypass surgery. Scand JThorac Cardiovasc Surg 1989;23:235-239

33. Rowe GG, Afonso S, Gurtner HP, Chelius CJ, Lowe WC,Castillo CA, Crumpton CW: The systemic and coronaryhemodynamic effects of adenosine triphosphate and adeno-sine. Am Heart J 1962;64:228-234

34. Cox DA, Vita JA, Treasure CB, Fish RD, Selwyn AP, Ganz P:Reflex increase in blood pressure during the intracoronaryadministration of adenosine. J Clin Invest 1989;84:592-596

35. Canty JM, Klocke FJ: Reduced regional myocardial perfusionin the presence of pharmacologic vasodilator reserve. Circula-tion 1985;71:370-377

36. Aversano T. Becker LC: Persistence of coronary vasodilatorreserve despite functionally significant flow reduction. Am JPhysiol 1985;248(Heart Circ Physiol 17):H403-H411

37. Kusachi S, Thompson RD, Olsson RA: Ligand selectivity ofdog coronary adenosine receptor resembles that of adenylatecyclase stimulatory (Ra) receptors. J Pharmacol Exp Ther1983;227:316-321

38. Conradson T-BG, Dixon CMS, Clark B, Barnes PJ: Cardio-vascular effects of infused adenosine: Potentiation by dipyri-damole. Acta Physiol Scand 1987;387-391

39. Afonso S: Inhibition of coronary vasodilating action of dipyr-idamole and adenosine by aminophylline in the dog. Circ Res1970;26:743-752

40. Olsson RA, Bugni WJ: Coronary circulation, in Fozzard HA,Haber E, Jennings RB, Katz AM (eds): The Heart andCardiovascular System. New York, Raven Press, 1986, pp1018-1019

41. McKenzie JE, Steffen RP, Haddy FJ: Effect of theophylline onadenosine production in the canine myocardium.Am J Physiol1987;252(Heart Circ Physiol 21):H204-H210

42. Johannsen UJ, Mark AL, Marcus ML: Responsiveness tocardiac sympathetic nerve stimulation during maximal coro-nary dilation produced by adenosine. Circ Res 1982;50:510-517

KEY WORDS * adenosine * coronary circulation



ownloaded from


R F Wilson, K Wyche, B V Christensen, S Zimmer and D D LaxsonEffects of adenosine on human coronary arterial circulation.

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

is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231Circulation doi: 10.1161/01.CIR.82.5.1595

1990;82:1595-1606Circulation.

http://circ.ahajournals.org/content/82/5/1595the World Wide Web at:

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

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

is online at: Circulation Information about subscribing to Subscriptions:

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

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

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

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



ownloaded from

http://circ.ahajournals.org/content/82/5/1595

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

http://www.lww.com/reprints

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


Effects of Adenosine on Human Coronary Circulation · usefulness of adenosine for studies of the...

Documents

Transcript of Effects of Adenosine on Human Coronary Circulation · usefulness of adenosine for studies of the...