Prostacyclin infusion in patients with acute myocardial ... · Prostacyclin was well tolerated by most patients: neither systemic blood pressure nor heart rate differed betweenthetwogroups

Br HeartJ 1985; 53: 173-9

Prostacyclin infusion in patients with acute myocardialinfarction

P HENRIKSSON,* 0 EDHAG,* A WENNMALMtFrom the Deparmes of*Medicine and tClinical Physiology, Karolinska Institute, Huddinge University Hospital,Huddinge, Sweden

SUMMARY To investigate whether prostacyclin protects ischaemic myocardium in humans the effectof prostacyclin or placebo was studied in two groups of patients with acute myocardial infarctionwho presented within six and 16 hours of the onset of symptoms. Intravenous infusion of prostacyc-lin or placebo was started immediately after admission at a rate corresponding to 4-5 ng/kg/minute.The infusion was maintained for 72 hours. Clinical status, electrocardiograms, plasma enzymeactivity, infarct extension during the infusion, and reinfarction after the infusion were studied.Prostacyclin was well tolerated by most patients: neither systemic blood pressure nor heart ratediffered between the two groups. In the 11 patients who received treatment within six hours of theonset of symptoms prostacyclin significantly lowered the maximum plasma activities of creatinekinase MB and lactate dehydrogenase. In the 19 patients who received treatment 6-16 hours afterthe onset of symptoms prostacyclin had no such effect. None of the patients receiving prostacyclinhad an extension of the infarction during the infusion, whereas four patients receiving placebo did;this difference was significant.These data are the first to provide evidence that prostacyclin might limit myocardial injury in

patients with acute myocardial infarction.

Prostacyclin is a metabolite of arachidonic acid that isformed in vascular endothelium throughout thebody.I-3 Prostacyclin has vasodilative and plateletantiaggregatory properties, and in view of theseactions the compound may prevent platelet aggrega-tion and thrombosis in healthy vessels. Vascular for-mation of prostacyclin is reduced in athero-sclerosis,46 and this has been proposed as a factorresponsible for the increased incidence of thrombosisin this state.The actions mentioned above have also prompted

the proposal of prostacyclin as a suitable agent for thetreatment of acute myocardial infarction. Thefindings concerning endothelial production of pros-tacyclin in healthy and atherosclerotic vessels sug-gested that such treatment might restore lost pros-tacyclin activity in the cardiovascular system, espe-cially in the coronary circulation. In fact, prostacyclin

Requests for reprints to Dr P Henriksson, Department of Medicine,Karolinska Institute, Huddinge University Hospital,,S-141 86 Hud-dinge, Sweden.

Accepted for publication 9 October 1984

reduces the size of experimental infarction in variousanimal models.7-'0 On the other hand, the administ-ration of prostacyclin to humans with various signs ofischaemic heart disease has yielded inconclusiveresults. 11-13To evaluate the effect of prostacyclin in patients

with acute myocardial infarction, we performed adouble blind investigation in 30 patients.

Patients and methods

The study was performed during 1983-84 at this hos-pital. Patients admitted to the hospital coronary careunit and meeting the selection criteria of the studywere invited to participate. Informed consent wasobtained from all patients. The study protocol wasapproved by the local human investigations commit-tee.

SELECTION OF PATIENTSPatients admitted with chest pain typical of acuteinfarction (>30 min duration) and with confirmativeelectrocardiogram or plasma enzyme activity or bothwere included provided that treatment could be

173

on March 29, 2021 by guest. P

rotected by copyright.http://heart.bm

j.com/

Br H

eart J: first published as 10.1136/hrt.53.2.173 on 1 February 1985. D

ownloaded from

http://heart.bmj.com/

174

started within 16 hours of the onset of symptoms. Theage limits were 18-75 years; female patients wereincluded only if postmenopausal. Either (a) new orpresumably new Q waves of at least 0-04 s durationand amplitude amounting to >25% of the following Rwave in either (i) leads II, III, and aVF, (ii) leads Iand aVL, or (iii) two or more precordial leads, or (b)ST segment elevation of >2 mV in two leads were theconfirmative electrocardiographic criteria used. A risein plasma activity to at least twice the upper normallimit of the hospital laboratory's reference value fortwo of the enzymes-creatine kinase, aspartateaminotransferase, and lactate dehydrogenase-orappreciably increased myocardial specific creatinekinase MB activity was accepted as evidence of acutemyocardial infarction.

Patients with an earlier acute myocardial infarctionwithin the past two months, as well as those withpsychiatric, endocrinological or infectious or malig-nant diseases, were excluded.

RANDOMISATION AND STRATIFICATIONA randomisation protocol was prepared before thestudy and was kept by the pharmacist delivering thedrug to the coronary care unit during the entire inves-tigation. The randomisation protocol was adapted toyield 15 patients in each of the prostacyclin andplacebo groups. The investigation protocol also pre-scribed a separate study to compare the results ofpatients receiving prostacyclin with those receivingplacebo within six hours of the onset of symptoms.These patients were, however, not randomised sepa-rately.

ADMINISTRATION OF PROSTACYCLIN ORPLACEBOProstacyclin (synthetised by the Upjohn Companyand formulated by the Wellcome Foundation), storedas a freeze-dried powder, was dissolved in glycine buf-fer pH 10-5 immediately before infusion. A fresh solu-tion was prepared every 12 hours during the infusionperiod. Glycine buffer at pH 10-5 was infused as theplacebo. The infusion was given into a peripheral armvein using a volume controlled pump (IMED 922) at arate corresponding to 1 ng/k8/min and increasedstepwise over one hour to ammum of 5 ng/kg/min.If systolic blood pressure fell or heart rate increasedby > 10% before themaum infusion rate had beenreached the infusion was maintained at the currentrate. The overall infusion time was 72 hours. Duringthe last four hours the infusion rate was graduallytapered off in order to avoid rebound phenomena. Inaddition to the infusion of prostacyclin, the patientsreceived routine clinical and pharmacological treat-ment for acute myocardial infarction and were man-

Henriksson, Edhag, Wennmalm

aged by the ordinary staff of the coronary care unit.The investigators did not influence the routine care.

EVALUATION OF CLINICAL STATUS ANDLABORATORY FINDINGSAll patients underwent a baseline clinical and labora-tory evaluation before infusion of the drug was startedcomprising a physical examination, a standard 12 leadelectrocardiogram, chest x ray examination, routinechemical blood examination, and measurement ofplasma activity of creatine kinase MB, creatine kin-ase, aspartate aminotransferase, and lactate dehy-drogenase. During the infusion of prostacyclin a 12lead electrocardiogram was recorded daily. Plasmacreatine kinase MB activity was measured in samplestaken 8, 16, and 24 hours after the start of the infu-sion. Plasma creatine kinase, aspartate aminotransfer-ase, and lactate dehydrogenase activities were meas-ured daily until discharge. At the end of the study (30days after admission) a physical examination, includ-ing evaluation of the patient's functional status, andan electrocardiogram were performed. The 12 leadelectrocardiograms were analysed using a QRS scor-ing system'4 in which each recording was scored from0 to 29, with a value >2 strongly suggesting myocar-dial infarction. QRS scoring was performed on elec-trocardiograms recorded on days 3, 7, and 30 after thestart of the infusion. ST segment elevation was meas-ured as the sum of the vertical distances between theisoelectric line and the midpoint of the ST segment inthe two leads showing the most prominent elevationson day 1. ST elevation was assessed on days 1, 2, and3 after the start of the infusion. All electrocardio-graphic evaluations were performed by a trainedinterpreter who was not a member of the investigatingteam and was unaware of the type' of treatment thepatients had received.

Results

COMPARABILITY OF THE GROUPSTable 1 shows the result of the randomisation proce-dure. The number of patients, their sex distributionand age, as well as the time from the onset of symp-toms to the start of infusion did not differ between theprostacyclin and placebo groups. The number ofpatients receiving infusion within six hours of theonset of symptoms was slightly greater in the pros-tacyclin group. Some patient characteristics related toan increased incidence of ischaemic heart disease wereover-represented in the prostacyclin group, but thedifference is not significant. Table 2 shows thenumber of patients receiving beta blocking drugsbefore or during the infusion of prostacyclin orplacebo. The proportion of patients given beta block-ing drugs did not differ between the two groups. The



j.com/

Br H


ownloaded from


Prostacyclin infusion in patients with acute myocardial infarction

Table 1 Group characteristics in patients with acutemyocardial infarction randomised to receive prostacyclin orplacebo and the occurrence offactors related to increasedincidence ofmyocardial infarction in the two groups. Figures arenumbers ofpatients

Prostacyclin Placebo

Group characteristicsNo of patients 15 15Sex distribution (M/F) 10/5 12/3Age (yr)Mean (SEM) 62 (6) 60 (3)Range 53 (71) 40 (73)

Mean (SEM) time from onset ofpain to start of infusion (h) 8-1 (1-2) 9.6 (1-3)

Infarct localisation(anterior/posterior) 8/7 8/7

No of patients receiving infusionwithin six hours of onset ofsymptoms 7 4

Infarct localisation(anterior/posterior) 2/5 3/1

Related factorsSmoking 12 9Hypertension 8 6Hyperlipidaemia 2 3Earlier angina 6 3'Cardiac failure before infusion 7 3

proportion of patients receiving beta blocking drugswas in fact slightly higher in the placebo group, andthis tendency was reinforced during days 2 and 3 ofinfusion. The difference did not, however, reachstatistical significance.

DRUG.TOLERANCEOf the patients in the prostacyclin group, 12 could begiven the drug at a rate of 5 ng/kg/minute withoutexceeding the predetermined systolic blood pressurevalue or pulse rate. In one patient the infusion ratewas never increased above 4 ng/kg/minute because ofa tendency to hypotension, and in two other patientsthe rate had to be reduced to 4 ng/kg/minute for thesame reason. Figure 1 shows the values for the systolicand diastolic blood pressure and heart rate. Therewere no differences in these variables before, during,or after the infusion. Table 3 shows the cardiac stressindex (the product of the systolic blood pressure andthe heart rate) during the infusion in the two groups.Again there was no difference between the groups.

CLINICAL OBSERVATIONSTable 4 shows the occurrence of congestive heart fail-ure before and after the study. Two patients werereceiving treatment for heart failure when theyentered the study; both were in the prostacyclingroup. Another five patients in this group had clinicalor radiographic findings or both indicating heart fail-ure on entering the study. The number of patients inthe prostacyclin group requiring treatment for heartfailure did not increase during the course of the infu-sion. In the placebo group three patients had clinicalor radiographic evidence of heart failure on enteringthe study. During the course of the infusion, thisnumber increased to eight (Table 4).

Table 2 Administration ofbeta blocking drugs in the groups receivingprostacyclin orplacebo. Figures are numbers ofpatients(mean24 hour dosages in mg given in parentheses)

Drug Time (h)* Preinfarction Day I Day 2 Day 3

Prostacyclin groupMetoprolol 0-16 1 (200) 1 (200) 4 (125) 4 (125)

0-6 1 (200) 1 (200) 2 (175) 2 (175)Pindolol 0-16 1(10) 1(10) 1(10) 1(10)

0-6 0 0 0 0Atenolol 0-16 1(100) 1 (100) 1 (10) 1(10)

0-6 0 0 0 0None 0-16 12 12 9 9

0-6 6 6 5 5

Placebo groupMetoprolol 0-16 1(100) 2 (125) 7 (107) 7 (121)

0-6 0 0 4(75) 3(83)Alprenolol 0-16 2 (400) 2 (400) 2 (400) 2 (400)

0-6 0 0 0 0Timolol 0-16 0 0 0 1(20)

0-6 0 0 0 0Practololt 0-16 0 0 0 0

0-6 0 1(15) 0 0None 0-16 12 10 5 4

0-6 4 3 0 1

*From onset of chest pain to start of infusion.tGiven intravenously.

175



j.com/

Br H


ownloaded from


Henriksson, Edhag, Wennmalm

Table 5 Occurrence of infarct extension, reinfarction, anddeaths in paients wvith acute myocardial infarcion treated withprostacyclin or placebo. Figures are numbers ofpatients

Prostacyclin Placebo

Extension of infarction (duringthe infusion) 0* 4

Reinfarction (after the infusion) 2 0Deaths 0 1

*Significant difference from the placebo group: (p

Prostacyclin infusion in patients with acute myocardial infarction

74

64

5(

g 44

~c 34

178

Discussion

Patients receiving prostacyclin within six hours of theonset of symptoms had significantly lower maximumplasma activity of creatine kinase MB and lactatedehydrogenase than the placebo group. We regardthese lower values for plasma creatine kinase MB andlactate dehydrogenase activity in the prostacyclingroup as indicating smaller infarcts among thesepatients compared with the placebo group. As evidentfrom Fig. 3 the patients receiving early placebo infu-sion had higher maximum plasma creatine kinaseMB, creatine kinase, aspartate aminotransferase, andlactate dehydrogenase activities than the group as awhole. This is probably because patients with largerinfarctions-and hence higher maximum plasmaenzyme values-experience more pronounced symp-toms and therefore present earlier. The proportion ofpatients with anterior wall infarctions was greater inthe placebo group; whether or not this may haveinfluenced the data cannot be judged at present. Theproportion of patients taking beta blocking drugs wasnot higher in the prostacyclin group. Thus theobserved difference in plasma enzyme activity be-tween the groups can hardly be explained by a moreactive beta blocking regimen in the prostacycingroup. A direct cardioprotective action ofprostacycinin acute myocardial infarction therefore seems pos-sible.

In addition to the differences in maximum plasmacreatine kinase MB and lactate dehydrogenaseactivities, the time courses of the plasma activity ofthese enzymes are notable. Thus the curve forcreatine kinase MB activity in the prostacyclin groupwas flatter than in the placebo group. In addition, thepeak of the plasma lactate dehydrogenase curve inpatients infused with prostacyclin within six hourswas lower than in those given placebo. Interestingly,the plasma lactate dehydrogenase values in the pros-tacycin group were numerically lower during theentire infusion, approaching those in the placebogroup after the infusion.

Another important finding was that patients receiv-ing prostacycin appeared to be protected againstinfarct extension during the infusion. This was evi-denced by a significantly lower incidence of infarctextension in patients receiving prostacyclin than inthose receiving placebo. The lower incidence ofinfarct extension seems to accord well with theobserved lower maximum plasma creatine kinase MBand lactate dehydrogenase activities discussed above,in as much as they indicate that myocardium at risk atthe onset of the infusion is protected by the activedrug and, consequently, that enzyme release and apossible tendency to infarct extension are depressed.If this hypothesis is correct patients receiving pros-

Henriksson, Edhag, Wennmalmtacycin should run an increased risk of reinfarctionafter the infusion. The incidence of reinfarctiontended to increase in the present group of patientsreceiving prostacyclin; two developed reinfarctionafter the infusion compared with none of the patientswho had received placebo; the difference was, how-ever, not significant.A drug with vasodilative activity, like prostacyclin,

can be expected to induce a decrease in peripheralvascular resistance and a reflex increase in cardiacoutput. The latter is certainly not desirable in acutemyocardial infarction since the increase in cardiacoutput may require more dynamic work by the heart.In the present study prostacyclin did not affect bloodpressure and heart rate more than placebo-that is,with the infusion rate limits applied in the presentstudy the drug probably does not interfere appreci-ably with the central haemodynamics. Bradycardiawas not a general finding in the placebo group, prob-ably because the patients were not routinely givenbeta blocking drugs during the acute phase. Further-more, prostacyclin did not induce cardiac failure oreven aggravate signs of failure. In fact all patients inthe placebo group with signs of cardiac failure beforeinfusion also required treatment for this afterwards;in addition, five more patients in this group developedevidence of cardiac failure during the infusion. Incontrast, of the seven patients in the prostacycingroup who had signs of cardiac failure before the infu-sion, three did not require treatment afterwards.

Although the present study points to a beneficialeffect of prostacyclin in patients with acute myocar-dial infarction, the possible underlying mechanism isunknown. Since a myocardial saving effect of pros-tacycin is indicated in patients receiving treatmentwithin six hours of the onset of symptoms some con-ceptually related postinfusion treatment (that is,treatment aimed to counteract platelet aggregationand formation of thromboxane A2) should be giventhat can keep salvaged myocardium in a metaboliccondition that allows contractile work to be main,tained. In men with unstable angina aspirin has beenshown to protect against acute myocardial infarctionand death15; this drug should be effective also in thepresent situation, in which myocardium at risk maypersist after the infusion of prostacycin.

OE was supported by the Swedish National Associa-tion against Heart and Chest Diseases, andAW by theSwedish Medical Research Council. Prostacycin waskindly provided by the Upjohn Company.

References

1 Moncada S, Gryglewski R, Bunting S, Vane JR. Anenzyme isolated from artenes transforms prostaglandinendoperoxides to an unstable substance that inhibits



j.com/

Br H


ownloaded from


Prostacyclin infusion in patients with acute myocardial infarctionplatelet aggregation. Nature 1976; 263: 663-5.

2 Gryglewski RJ, Bunting S, Moncada S, Flower RJ, VaneJR. Arterial walls are protected against deposition ofplatelet thrombi by a substance (prostaglandin X) whichthey make from prostaglandin endoperoxides. Prostag-landins 1976; 12: 685-713.

3 Moncada S, Higgs EA, Vane JR. Human arterial andvenous tissues generate prostacyclin (prostaglandin X) apotent inhibitor of platelet aggregation. Lancet 1977; i:18-21.

4 Gryglewski RJ, Dembiniska-Kiec A, Zmuda A, Gryg-lewska T. Prostacyclin and thromboxane A2 biosynthesiscapacities of heart, arteries and platelets at various stagesof experimental atherosclerosis in rabbits. Atherosclerosis1978; 31: 385-94.

5 Sinzinger H, Feigl W, Silberbauer K, Oppolzer R,Winter M, Auerswald W. Prostacyclin (PGI2)-generationby different types of human atherosclerotic lesions. ExpPathol 1980; 18: 175-80.

6 Larrue J, Rigaud M, Daret D, Demond J, Durand J,Bricaud H. Prostacyclin production by cultured smoothmuscle cells from atherosclerotic rabbit aorta. Nature1980; 285: 480-2.

7 Ogletree ML, Lefer AM. Prostglandin-induced preser-vation of the ischemic myocardium. Circ Res 1978; 42:218-24.

8 Ogletree ML, Lefer AM, Smith JB, Nicolaou KC.

Studies on the protective effect of prostacyclin in acutemyocardial ischemia. EurJ Pharmacol 1979; 56: 95-103.

9 Araki H, Lefer AM. Role of prostacyclin in the preserva-tion of ischemic myocardial tissue in the perfused catheart. Circ Res 1980; 47: 757-63.

10 Jugdutt BI, Hutchins GM, Buildey BH, Becker LC.Dissimilar effects of prostacyclin, prostaglandin E1 andprostaglandin E2 on myocardial infarct size after coro-nary occlusion in conscious dogs. Circ Res 1981; 49:685-700.

11 Szczeklik A, Szczeklik J, Nizankowski R, Gluszko P.Prostacyclin for acute coronary insufficiency. Artery1980; 8: 7-11.

12 Szczeklik A, Szczeklik J, Nizankowski R. Prostacyclin,nitroglycerin and effort angina [Letter]. Lancet 1981; i:1006.

13 Chierchia S, Patrono C, Crea F, et al. Effects of intraven-ous prostacyclin in variant angina. Circulation 1982; 65:470-7.

14 Wagner GS, Freye CJ, Palmeri ST, et al. Evaluation of aQRS scoring system for estimating myocardial infarctsize. I. Specificity and observer agreement. Circulation1982; 65: 342-7.

15 Lewis HD Jr, Davis JW, Archibald DG, et al. Protectiveeffects of aspirin against acute myocardial infarction anddeath in men with unstable angina. N EnglJ Med 1983;309: 396-403.

179



j.com/

Br H


ownloaded from

Documents

Prostacyclin infusion in patients with acute myocardial ... · Prostacyclin was well tolerated by most patients: neither systemic blood pressure nor heart rate differed betweenthetwogroups