8
2 Accid Emerg Med 1997;14:2-9 REVIEW Thrombolysis after acute myocardial infarction Patrick A Nee The early treatment of acute myocardial infarction has become characterised by aggres- sive regimens aimed at limiting infarct size, the rapid treatment of arrhythmias and left ven- tricular dysfunction, and early evaluation for surgery or angioplasty. Intravenous thrombo- lytic (fibrinolytic) agents have become the mainstay of treatment for definite infarction and significant mortality and morbidity gains have been demonstrated with these drugs in a number of major studies. flow in infarct related vessels.6 The intravenous agent achieved the same end, haemorrhagic complications being reduced by employing a short duration infusion.7 Alternative agents with allegedly greater clot specificity were developed shortly afterwards. These new activators of plasminogen included anisoylated plasminogen-streptokinase activa- tor complex, anistreplase (APSAC),8 and recombinant tissue plasminogen activator, rt- PA.9 Whiston Hospital, Prescot, Merseyside L35 5DR, United Kingdom P A Nee Correspondence to: Dr Patrick A Nee, Consultant in Accident and Emergency Medicine and Critical Care Medicine Accepted for publication 16 October 1996 Keywords: thrombolysis; acute myocardial infarcation; streptokinase; rt-PA Background Thrombolysis began in 1933 when Tillet and Garner first isolated a fibrinolytic substance, which they called streptococcal fibrinolysin, from haemolytic streptococci.1 Christensen and Macleod renamed this substance strepto- kinase in 1945 and described its mechanism of action as converting plasminogen to plasmin which lyses fibrinogen and fibrin in coagulum.2 It was initially used clinically to debride infected tissue spaces such as chronic em- pyema and haemothorax.3 The purified strepto- kinase produced by Christensen after 1947 became used in early experiments which showed that prolonged intravenous infusion could dissolve intracoronary thrombus and 4 limit infarct size. Little progress was made for the next two decades because of fears over systemic side effects, especially bleeding and allergy, and because attention was focused on the establish- ment of coronary care units (CCU) within most major hospitals-an initiative which itself led to significant mortality gains. The current era began in 1980 when DeWood reported the use of coronary angio- graphy to demonstrate thrombus within the coronary arteries of patients soon after the onset of acute myocardial infarction symp- toms.5 Until that time it had not been appreci- ated that coronary thrombosis was the underly- ing pathological process in the majority of cases of myocardial infarction. The natural his- tory of myocardial infarction involves recanali- sation of the occluded artery so that intracoro- nary thrombus was often not seen at necropsy in patients who had died some days after the acute event. Intracoronary streptokinase was found to be effective in re-establishing blood Benefits of thrombolysis Early treatment with a thrombolytic agent lyses thrombus within coronary arteries and pro- motes recanalisation. Beneficial effects include limitation of infarct size, improved left ven- tricular function, and reduced mortality. Surprisingly, the first report of mortality reduction with intravenous streptokinase comes from as long ago as 1958.4 Small stud- ies carried out in the 1960s and 1970s supported these early findings, although con- cerns over side effects restricted widespread uptake. Several very large, placebo controlled trials reported survival benefits with the three principle thrombolytic agents in the 1980s.10-13 The most extensive description on mortality reduction appeared two years ago when the Fibrinolytic Therapy Trialists (FIT) Collaborative Group published an important review of all nine trials with more than a 1000 patients randomised per trial. Outcome data on almost 60 000 patients showed that throm- bolysis reduced 35 day mortality after acute myocardial infarction from 11.5% to 9.6%, saving 20 to 30 lives per thousand treatments, depending on delay to treatment.14 Limiting the extent of ischaemic necrosis reduces the risk from left ventricular dysfunc- tion, the principal cause of early and late non- arrhythmic mortality and morbidity. Left ventricular ejection fraction, albeit an inexact measure shortly after acute myocardial infarc- tion, is improved by thrombolysis. 15 The frequency of dangerous arrhythmia is also reduced, probably by improving the blood sup- ply to the highly arrhythmogenic "penumbra" of critically ischaemic myocardium surround- ing the infarcting area. Improved outcomes depend upon prompt reperfusion of ischaemic myocardium. Suc- 2 on November 13, 2020 by guest. Protected by copyright. http://emj.bmj.com/ J Accid Emerg Med: first published as 10.1136/emj.14.1.2 on 1 January 1997. Downloaded from

Thrombolysis after acute myocardial infarction · infarction (8.4% v 16.9% for anterior acute myocardial infarction) may account for these differences. Inferiorwall infarctions tendto

  • Upload
    others

  • View
    5

  • Download
    0

Embed Size (px)

Citation preview

Page 1: Thrombolysis after acute myocardial infarction · infarction (8.4% v 16.9% for anterior acute myocardial infarction) may account for these differences. Inferiorwall infarctions tendto

2 Accid Emerg Med 1997;14:2-9

REVIEW

Thrombolysis after acute myocardial infarction

Patrick A Nee

The early treatment of acute myocardialinfarction has become characterised by aggres-sive regimens aimed at limiting infarct size, therapid treatment of arrhythmias and left ven-tricular dysfunction, and early evaluation forsurgery or angioplasty. Intravenous thrombo-lytic (fibrinolytic) agents have become themainstay of treatment for definite infarctionand significant mortality and morbidity gainshave been demonstrated with these drugs in anumber of major studies.

flow in infarct related vessels.6 The intravenousagent achieved the same end, haemorrhagiccomplications being reduced by employing ashort duration infusion.7

Alternative agents with allegedly greater clotspecificity were developed shortly afterwards.These new activators of plasminogen includedanisoylated plasminogen-streptokinase activa-tor complex, anistreplase (APSAC),8 andrecombinant tissue plasminogen activator, rt-PA.9

Whiston Hospital,Prescot, MerseysideL35 5DR, UnitedKingdomP A Nee

Correspondence to:Dr Patrick A Nee,Consultant in Accident andEmergency Medicine andCritical Care Medicine

Accepted for publication16 October 1996

Keywords: thrombolysis;acute myocardial infarcation;streptokinase; rt-PA

BackgroundThrombolysis began in 1933 when Tillet andGarner first isolated a fibrinolytic substance,which they called streptococcal fibrinolysin,from haemolytic streptococci.1 Christensenand Macleod renamed this substance strepto-kinase in 1945 and described its mechanism ofaction as converting plasminogen to plasminwhich lyses fibrinogen and fibrin in coagulum.2It was initially used clinically to debrideinfected tissue spaces such as chronic em-

pyema and haemothorax.3 The purified strepto-kinase produced by Christensen after 1947became used in early experiments whichshowed that prolonged intravenous infusioncould dissolve intracoronary thrombus and

4limit infarct size.

Little progress was made for the next twodecades because of fears over systemic sideeffects, especially bleeding and allergy, andbecause attention was focused on the establish-ment of coronary care units (CCU) withinmost major hospitals-an initiative which itselfled to significant mortality gains.The current era began in 1980 when

DeWood reported the use of coronary angio-graphy to demonstrate thrombus within thecoronary arteries of patients soon after theonset of acute myocardial infarction symp-toms.5 Until that time it had not been appreci-ated that coronary thrombosis was the underly-ing pathological process in the majority ofcases of myocardial infarction. The natural his-tory of myocardial infarction involves recanali-sation of the occluded artery so that intracoro-nary thrombus was often not seen at necropsyin patients who had died some days after theacute event. Intracoronary streptokinase was

found to be effective in re-establishing blood

Benefits of thrombolysisEarly treatment with a thrombolytic agent lysesthrombus within coronary arteries and pro-motes recanalisation. Beneficial effects includelimitation of infarct size, improved left ven-tricular function, and reduced mortality.

Surprisingly, the first report of mortalityreduction with intravenous streptokinasecomes from as long ago as 1958.4 Small stud-ies carried out in the 1960s and 1970ssupported these early findings, although con-cerns over side effects restricted widespreaduptake. Several very large, placebo controlledtrials reported survival benefits with the threeprinciple thrombolytic agents in the1980s.10-13 The most extensive description onmortality reduction appeared two years agowhen the Fibrinolytic Therapy Trialists (FIT)Collaborative Group published an importantreview of all nine trials with more than a 1000patients randomised per trial. Outcome dataon almost 60 000 patients showed that throm-bolysis reduced 35 day mortality after acutemyocardial infarction from 11.5% to 9.6%,saving 20 to 30 lives per thousand treatments,depending on delay to treatment.14

Limiting the extent of ischaemic necrosisreduces the risk from left ventricular dysfunc-tion, the principal cause of early and late non-arrhythmic mortality and morbidity. Leftventricular ejection fraction, albeit an inexactmeasure shortly after acute myocardial infarc-tion, is improved by thrombolysis. 15 Thefrequency of dangerous arrhythmia is alsoreduced, probably by improving the blood sup-ply to the highly arrhythmogenic "penumbra"of critically ischaemic myocardium surround-ing the infarcting area.Improved outcomes depend upon prompt

reperfusion of ischaemic myocardium. Suc-

2

on Novem

ber 13, 2020 by guest. Protected by copyright.

http://emj.bm

j.com/

J Accid E

merg M

ed: first published as 10.1136/emj.14.1.2 on 1 January 1997. D

ownloaded from

Page 2: Thrombolysis after acute myocardial infarction · infarction (8.4% v 16.9% for anterior acute myocardial infarction) may account for these differences. Inferiorwall infarctions tendto

Thrombolysis after acute myocardial infarction

Table 1 Options for bleeding after thrombolytic therapy

Manual pressure on punctured vesselInterrupt thrombolytic and anticoagulant therapyIV fluid replacement or bloodProtamine sulphate, 1 mg per 100 units of heparin givenCryoprecipitate (15 bags) to increase fibrinogen levels to

lg/litreFresh frozen plasma (2-4 units)Platelets (4-12 bags)Epsilon-aminocaproic acid (5 g in 15-30 min)Tranexamic acid (lg slow IV injection repeated as necessary)

cessful reperfusion is clinically apparent as animprovement in chest pain accompanied byECG evidence of at least a 50% reduction inST segment elevation. There is also a rapid riseand fall in cardiac enzyme levels in the plasma.Failure to achieve reperfusion may indicate theneed to repeat the infusion or refer for urgentangiography.

Adverse effectsAdverse events are extremely frequent afterthrombolytic therapy, particularly when treat-ment is delayed or given inappropriately.Cardiac arrhythmias, especially self limitingruns of accelerated idioventricular rhythm,may reflect reperfusion. More serious rhythmdisturbances such as ventricular tachycardiaare more common when a thrombolytic agenthas not been given.

HYPOTENSIONHypotension accompanies one in 50 treat-ments. It is mediated by histamine release andvasodilatation, and is sometimes a marker ofpoor left ventricular function. It may respondto volume replacement but is occasionallysevere enough to cause the treatment to be dis-continued.

ALLERGYImmediate hypersensitivity reactions, morecommon after streptococcal sore throat, maycause hypotension as well as urticaria andwheezing. However the incidence of acute ana-phylaxis was less than 0. 1% in one of the earlymulticentre studies.10 There is no evidence

Table 2 Suggested contraindications to thrombolytic therapy

AbsoluteSuspected aortic dissectionActive bleedingPrevious subarachnoid or intracerebal haemorrhage

MajorMajor surgery; obstetric delivery; serious trauma within 2 weeksHead injury or cerebrovascular accident within 2 months*Severe gastrointestinal haemorrhage within 2 weeksHeavy vaginal bleeding

MinorHistory of cerebrovascular accident with residual disabilityPossible pregnancyUse of anticoagulants; known bleeding diathesisSuspected allergic reaction to agent being usedUncontrolled systolic blood pressure >180 mm HgHaemorrhagic diabetic retinopathyAcive peptic ulceration (current obtrusive symptoms in spite of treatment)Prolonged cardiopulmonary resuscitationAcute pancreatitisOesophageal varicesSevere liver disease

*Even with complete recoveryReproduced with permission from Vincent R. Myocardial infarction: thrombolysis afterinfarction. Prescribers Journal 1995;35:140-8.

3

that allergic reactions are prevented by prioradministration of corticosteroids or antihista-mines.

BLEEDINGHaemorrhage is a common complication in thefirst few hours after thrombolytic therapy.Minor episodes, such as ecchymoses or pro-longed bleeding from venepuncture sites,occur frequently. Major haemorrhage, definedas intracranial or gastrointestinal bleeding, a15% drop in packed cell volume, or haemor-rhage requiring transfusion, occurs in 1-2%despite the exclusion of high risk patients.Treatment options for severe bleeding duringor shortly after thrombolysis are listed intable 1.

STROKECerebrovascular events occur in up to 3% ofcases. Non-haemorrhagic strokes are usuallydue to embolisation of mural thrombi. Intra-cranial haemorrhage has a reported incidenceof 0.3 to 0.7 per cent. Cerebral bleeding occursmost frequently in the elderly, in hypertensives,and those with a history of previous stroke.Choice of agent is also relevant: in the ISIS 3study'6 there were 39 definite cerebral haem-orrhages among 12 848 patients randomisedto receive streptokinase and 92 such events in12 841 patients given rt-PA (P = 0.0001).Outcome data in controlled trials of throm-

bolytic agents are usually presented as 35 daysurvival without disabling stroke. Despite theincidence of three to seven catastrophic strokesper 1000 treatments there is still compellingevidence of absolute benefit when thrombo-lysis is carried out early in all eligible cases.The generally accepted contraindications to

thrombolytic therapy are listed in table 2.

Patient selectionAll patients presenting with a history sugges-tive of acute myocardial infarction are potentialcandidates for fibrinolytic therapy, but manydo not receive the treatment.'7 A major factorin treatment denials may be confusion over eli-gibility. The large international studies of the1980s were inconsistent in terms of exclusioncriteria, and up to two thirds of patients wereexcluded in some trials. For example GISSI'0included only patients with definite infarction,while ISIS 21l and ASSET13 included patientswith suspected acute myocardial infarction.The AIMS investigators'2 employed an upperage limit of 70 years and both AIMS andISAM18 excluded patients with chest pain ofmore than six hours duration. By contrast therewas no upper age limit with GISSI or ISIS 2,and patients were randomised up to 12 and 24hours respectively after symptom onset. Therewas also inconsistency between trials withregard to inclusion of patients with cardiogenicshock and those who had had previous acutemyocardial infarction.

Contradictory results between studies addedto the difficulties of interpretation. In GISSI nobenefit was demonstrated with streptokinaseafter six hours, whereas the ISIS 2 investigators

on Novem

ber 13, 2020 by guest. Protected by copyright.

http://emj.bm

j.com/

J Accid E

merg M

ed: first published as 10.1136/emj.14.1.2 on 1 January 1997. D

ownloaded from

Page 3: Thrombolysis after acute myocardial infarction · infarction (8.4% v 16.9% for anterior acute myocardial infarction) may account for these differences. Inferiorwall infarctions tendto

Nee

showed significant mortality gains up to 12hours after onset of chest pain. Results fromISIS 2 also showed improved survival withstreptokinase in inferior myocardial infarctionwhereas no such benefit was seen in ISAM orGISSI. Mortality rates in patients with sus-pected acute myocardial infarction presentingwith left bundle branch block (LBBB) werereduced from 28% to 14% in ISIS 2, whereasno significant difference was reported in thissubgroup by the GISSI investigators.The effect of thrombolysis allocation on

subgroups distinguished by presenting featureswas examined in the FTT review of trials.'4The odds ratios for death were significantlyreduced in patients less than 75 years old pre-senting within 12 hours of onset of chest painwho had either ST segment elevation or LBBBon the ECG. Other subgroups are consideredbelow.

ELDERLY PATIENTSAcute myocardial infarction is twice as com-mon in patients over the age of 75 years and ismore often fatal. The mortality of acutemyocardial infarction over the age of 75 is fivetimes that of the under 55 years group. There isno evidence to suggest that larger infarctionsare responsible for the excess mortality.Necropsy data from the GISSI 2 studyrevealed a higher incidence of cardiac rupturein fatal cases over the age of 70 compared tothose under 60 years of age.'9 Although thepercentage mortality reduction with throm-bolysis is less in the older age group, the abso-lute numbers of lives saved is greater. In theFTT review of trials, this mortality reductionreached statistical significance up to age 74,with a trend for improved survival in throm-bolysis allocated patients over 75 years of age.Another subgroup analysis suggested that onelife was saved for 33 thrombolytic treatments inthe over 75 years age group, compared withone life saved per 50 treatments in youngerpatients.20 The available evidence does not,therefore, support the adoption of an upper agelimit for thrombolysis although in practiceolder patients often have other contraindica-tions to treatment.

HYPERTENSIONThe data are inconsistent with regard to theutility of thrombolytics in hypertensive pa-tients. The FTT review of trials reported anon-significant trend in favour of treatment,but the confidence limits were wide. The majorrisk in this group is haemorrhagic stroke andthere is some agreement that the agents shouldbe avoided with systolic pressures over 180 mmHg. However, transient hypertension will oftenrespond to treatment with opiate analgesia,nitrates, or intravenous 3 blockers whereindicated.

HYPOTENSIONSystolic blood pressures of less than 100 mmHg occur most frequently after anterior acutemyocardial infarction. Pooled data presentedin the FTT review of trials suggest that throm-bolytic therapy is beneficial, reducing mortality

from 35.1% in controls to 28.9% in the treatedgroup (P < 0.01). Control group mortality ishigher than average in this subgroup so theabsolute number of lives saved is even greaterthan the percentage reduction suggests. Withsevere hypotension or cardiogenic shock (clini-cally apparent as systemic hypotension pluspulmonary oedema) the case for thrombolysisremains to be proved.

PREVIOUS ACUTE MYOCARDIAL INFARCTIONTwenty per cent of patients admitted withacute myocardial infarction will have had aprevious infarction and second and subsequentinfarctions carry a greater risk of death. Thereported mortality reduction from 14.1% to12.5% with fibrinolysis in patients with priormyocardial infarction therefore representsmore lives saved per treatment than in patientswith first acute myocardial infarction. Strep-tokinase is contraindicated in the first twoyears after a previous dose and there isevidence of persistence of neutralising antibod-ies for at least four years.2' rt-PA is preferred inthese circumstances.

INFERIOR INFARCTIONIn acute inferior wall infarction with STsegment elevation thrombolytic therapy re-duced 35 day mortality from 8.4% to 7.5% inthe FTT review of trials, indicating a trend forimproved survival with treatment that did notquite reach statistical significance.There were different interpretations between

the component trials: ISAM failed to showmortality gains with streptokinase after inferioracute myocardial infarction, whereas the ISIS 2investigators reported significant benefit. Rela-tively low control group mortality after inferiorinfarction (8.4% v 16.9% for anterior acutemyocardial infarction) may account for thesedifferences. Inferior wall infarctions tend to besmaller (by number of leads involved) thananterior infarctions. The number of lives savedby fibrinolysis after inferior infarctions is simi-lar to that observed with smaller anteriorinfarctions.

OTHER ECG FEATURESAcute ST elevation is not universal withmyocardial infarction. Acute myocardial inf-arction may present with ST segment depres-sion, for example, or the ECG may be normal.Broadening the criteria to include patients withsuspected acute myocardial infarction wouldenable more rapid thrombolysis in aproportion of patients but the data do not sup-port the use of thrombolytic agents in thesecircumstances. Subdivision of patients intoECG categories widens the confidence limits,however, and some have argued that noevidence of benefit is not evidence of nobenefit.22 Others have indicated that, whilecoronary occlusion mediates acute transmuralinfarction (presenting with ST elevation),intermittent platelet plugs may be the cause ofnon-Q-wave acute myocardial infarction. Thelatter are more rationally treated with heparinand aspirin.23 This view gains support fromthe findings of the LATE study in which late

4

on Novem

ber 13, 2020 by guest. Protected by copyright.

http://emj.bm

j.com/

J Accid E

merg M

ed: first published as 10.1136/emj.14.1.2 on 1 January 1997. D

ownloaded from

Page 4: Thrombolysis after acute myocardial infarction · infarction (8.4% v 16.9% for anterior acute myocardial infarction) may account for these differences. Inferiorwall infarctions tendto

Thrombolysis after acute myocardial infarction

presenters with non-diagnostic ECGs had apoorer outcome with thrombolysis than with-out.24

Chest pain of non-cardiac origin is morelikely in patients without the typical ECGchanges of acute myocardial infarction. Peri-carditis, peptic ulcer disease, and aortic dissec-tion may all present with cardiac-type pain. Inthe ASSET study the mortality rate amongpatients given rt-PA for non-ischaemic chestpain was 9.6% compared with 1.2% forplacebo. 13

CARDIOPULMONARY RESUSCITATIONProlonged cardiopulmonary resuscitation hasbeen regarded as a contraindication to throm-bolytic therapy, but what constitutes prolongedresuscitation has not been specified. Currentevidence suggests that there is no significantincrease in bleeding complications with throm-bolysis even after vigorous cardiopulmonaryresuscitation.25-27 Major bleeding is morecommon after prehospital resuscitation, how-ever, and thrombolysis in this situation is rela-tively contraindicated.28

LATE PRESENTERSThe benefits of thrombolysis are very muchtime dependent. Very early clot lysis amongpatients in the TEAHAT29 and GREAT30studies prevented the development of fullthickness changes in a proportion of patients,indicating a small window of opportunity dur-ing which a thrombolytic agent may reverse theprocess of infarction itself.

In the GISSI study the mortality gain in thefirst hour after onset of MI symptoms was 47lives saved per 1000 treatments, reducing to 17in the third to sixth hours. At more than sixhours no significant difference was observed.By contrast, in ISIS 2 benefit was observed upto 24 hours after randomisation but in thisstudy patients were not excluded if the exacttime of symptom onset was unknown.

Figure 1 shows the relation between absolutebenefit (survival without disabling stroke) andtime to treatment in 45 000 patients with STelevation or bundle branch block.'4 Loss ofbenefit per hour of delay was 1.6 (SD 0.6) per1000 patients treated. Benefit was still ob-served beyond the six hours interval, however.In 22 000 patients randomised 7-12 hoursafter onset of symptoms, mortality was re-duced by thrombolysis from 12.7 % to 11.1%(P < 0.01).Two other recent studies bear out these find-

ings. The EMERAS study published in 19933'examined 35 day mortality in 4534 patientsrandomised to streptokinase or placebo up to12 hours after onset of chest pain. Thromboly-sis reduced mortality from 14.7% to 12.7%. Inthe same year the findings of the LATE studywere published, rt-PA was found to reduce 35day mortality from 11.97% to 8.9%.24While it remains true that the maximum gain

is achieved by administration of a thrombolyticagent as soon as possible after symptom onset,some infarcts follow a "stuttering" course withmyocardial necrosis occurring gradually over aprolonged period of time. Often it may be dif-

0o Loss of benefitQ x=o 40 n 3000 per hour of delay8m to randomisation:

14 000 1.6 SD 0.6 per330 1000 patients

CL ~~~12 000

a) U) _

0H 1

0-(1) > 97000CD .>

0 6 12 18 24Hours from symptom onset to

randomisation

Figure 1 Absolute reduction in 35 day mortality versusdelay from symptom onset to randomization among 45 000patients with ST elevation or bundle branch block.Reproduced with permission from Lancet(1994;346:311-22; see reference 14).

ficult to determine the exact time of onset ofcontinuous pain in patients presenting withthis type of history. Furthermore, even after theprocess of infarction has begun there may stillbe some advantage to treating with a thrombo-lytic agent. Infarct limitation and ventricularsalvage are not relevant at this stage-benefitsare related to secondary events. Scar formationand remodelling of infarcted areas is morecomplete in thrombolytic treated patients,leading to fewer dysrrhythmias and less riskfrom cardiac failure and ventricular aneurysmformation.

Provided there are no other contraindica-tions, therefore, fibrinolysis may be attemptedup to 12 hours after the onset of chest pain,even later in cases with a good history anddiagnostic ECG.

Choice of agentIn UK practice streptokinase is the principaldrug for in-hospital thrombolysis. It is given asa solution of 1.5 MU in 100 ml of normalsaline over one hour. The drug costs approxi-mately £85 per dose (recommended NHSprice). Concomitant therapy is with oralaspirin 300 mg daily indefinitely and subcuta-neous heparin 5000 units three times daily or12 500 units twice daily until the patient isambulant. For prehospital use anistreplase(APSAC) is preferred. This agent is similar tostreptokinase but has been chemically modi-fied to permit bolus administration: complex-ing streptokinase with acylated human lyso-plasminogen allows a more rapid onset andprolonged half life. APSAC begins clot lysiswithin minutes and the effect lasts for four tosix hours. It is much more expensive thanstreptokinase, at £495 per dose. The dosageregime is 30 units reconstituted in 10 ml ofdiluent given by slow intravenous injectionover five minutes.Recombinant tissue plasminogen activator

(rt-PA), the costliest alternative at £750, is theagent of choice where there is known allergy toor previous use of streptokinase. rt-PA is anaturally occurring serine protease protein andis manufactured by recombinant DNA tech-niques. It is more clot specific than streptoki-

5

on Novem

ber 13, 2020 by guest. Protected by copyright.

http://emj.bm

j.com/

J Accid E

merg M

ed: first published as 10.1136/emj.14.1.2 on 1 January 1997. D

ownloaded from

Page 5: Thrombolysis after acute myocardial infarction · infarction (8.4% v 16.9% for anterior acute myocardial infarction) may account for these differences. Inferiorwall infarctions tendto

Nee

nase and binds to fresh intravascular thrombus.It was previously given over three hours (10 mgbolus followed by 50 mg over one hour then 40mg over two hours). More recently an acceler-ated or "front loaded" schedule has beenemployed: a 15 mg bolus dose followed by amaximum 50 mg (0.75 mg/kg) over 30minutes, then a maximum 35 mg (0.5 mg/kg)over one hour. Concomitant treatment in-cludes oral aspirin and intravenous heparin in a5000 unit bolus then 1000 units/h for aminimum 48 hours, aiming for an APTT of60-85 seconds.

Several important comparative studies werereported in the late 1980s, including GISSI 2which compared streptokinase and rt-PA withand without heparin32 and ISIS 3, in which allthree agents were compared.'6 At the timert-PA was the most popular fibrinolytic inNorth America, mainly because of the TIMI 1study of 1986 which showed far superiorpatency rates for infarct related arteries withrt-PA: 62% were patent on the 90 minute angio-gram versus 31% for streptokinase.33 How-ever, this study had been terminated early andEuropean cardiologists challenged its conclu-sions on the basis that early recanalisation,which may be transient, is an inadequateoutcome measure.

This transatlantic "Battle of the Clot-busters"34 seemed to have been settled whenISIS 3 reported in 1992. Among more than41 000 patients randomised between strepto-kinase, APSAC, and rt-PA in 914 hospitals,there was no significant difference in 35-daymortality. The findings are summarised in fig2. Further evidence in favour of the continueduse of streptokinase as agent of choice camefrom GISSI 2, which also failed to demonstratesurvival benefit with rt-PA in its study of20 891 patients. Mortality from acute myocar-dial infarction in these international megatrialswas approximately 10% regardless of choice ofthrombolytic agent. There was a much higherrate of cerebral haemorrhage with rt-PA in theISIS 3 study (0.7% v 0.3%, P = 0.0001)though the difference was not significant inGISSI 2 (0.4% v 0.3%).

Despite these data, many North Americancardiologists continued to favour rt-PA, citingmethodological problems with the megatrials,including outcome differences between Euro-pean hospitals and their US counterparts, thechoice oftwo different brands of rt-PA, and thedelayed subcutaneous heparin regime. Propo-nents of rt-PA regained some ground with thepublication of the GUSTO (global utilisationof streptokinase and TPA for occludedcoronary arteries) study. This internationalstudy involving 41 021patients randomised tofour treatment groups reported lower mortalityin the group receiving accelerated treatment withrt-PA and intravenous heparin than in the otherthree groups: streptokinase plus intravenousheparin, streptokinase plus subcutaneousheparin, and rt-PA plus streptokinase.35 Despitean increased rate ofhaemorrhagic stroke (0.72%v 0.54%) there was absolute benefit with rt-PAamounting to nine additional survivors withoutdisabling stroke per 1000 patients treated. The

12.5 r

10.0

0-0

.c7.5

>.)

0)

co.m 5.0

E

2.5 _

0.0 L0

Streptokinase 10.5% (1397/13 295)Tissue plasminogenactivator 10.3% (1368/13 290)Anistreplase 10.6% (1407/13 328)

1 2 3 4 5Weeks

Figure 2 Mortality of three agents compared in ISIS-3.Reproduce with permission from BMJ, 1991;302:1259-61(see reference 34).

greatest survival benefit was observed in the pre-determined subsets who presented within fourhours of chest pain onset, who were aged lessthan 75 years, and who had sustained anteriorinfarction. The mechanism of benefit was shownto relate to increased rates of early recanalisation.In an angiographic subgroup of 2431 GUSTOpatients, there was a higher rate of 90 minutepatency in the rt-PA treated patients, althoughinterestingly this was not sustained at 180

36minutes.In common with all the major studies,

GUSTO has been criticised on methodologicalgrounds. Although it was a prospective ran-domised controlled trial, GUSTO employedan open label study design. Also, in the dataanalysis, the streptokinase plus subcutaneousheparin arm was combined with the strepto-kinase plus intravenous heparin arm. Incon-sistencies between study groups have also beennoted, including the fact that there was moreintervention in the rt-PA group and that excesssurvivors with rt-PA were only reported in UStreated patients.37 In light of the high cost ofrt-PA, the additional stroke risk and the lack ofdemonstrable benefit from the agent in the sys-temic review of major trials, includingGUSTO, rt-PA has not become the agent ofchoice in the United Kingdom. Instead manyregional cardiology advisory groups offerguidelines for its use which are similar to thosein table 3.

Outside of these limited indications forrt-PA, streptokinase remains the thrombolyticagent of first choice.

Systems for timely thrombolysisSeveral elements contribute to treatment de-lays, such as the time taken for the patient toseek medical help, the response time of theemergency services, and transfer time to hospi-tal. Once in hospital the "door to needle time",

6

I

on Novem

ber 13, 2020 by guest. Protected by copyright.

http://emj.bm

j.com/

J Accid E

merg M

ed: first published as 10.1136/emj.14.1.2 on 1 January 1997. D

ownloaded from

Page 6: Thrombolysis after acute myocardial infarction · infarction (8.4% v 16.9% for anterior acute myocardial infarction) may account for these differences. Inferiorwall infarctions tendto

Thrombolysis after acute myocardial infarction

Table 3 Indications for tissue plasminogen activator (TPA)

Patients who fulfil ALL the following criteria:-Presentation equal to or less than 4 hours from onsetAge equal to or less than 75 yearsAnterior infarction

Significant persistent hypotension (less than 100 mm Hg systolic on several consecutivereadings)

Previous myocardial infarction treated with strepokinase or anistrepase (APSAC) AT ANYTIME in the past

Known allergy to strepokinaseRecent proven streptococcal infection

also known as the "door to drug interval",should be minimised by a robust triage systemwhich rapidly takes the patient with suspectedacute myocardial infarction to the clinical area

where thrombolysis may be initiated. Severalmethods of reducing delays have been evalu-ated in recent studies.

PREHOSPITAL THROMBOLYSISAPSAC and rt-PA are suitable for use in theprehospital arena. EMIP, a large Europeanmulticentre study involving 5469 patients,reported a median reduction in pain to needletimes of 55 minutes when APSAC was givenbefore transfer to hospital by doctors in mobilecoronary care units.38 The trial was terminatedearly when it was found that significant reduc-tions in deaths from cardiac causes were

observed, 18 lives being saved per 1000treatments. The North American MITI studyincluding 360 patients randomised to receivert-PA from paramedics or after transfer to hos-pital reported a median reduction of 33minutes in the prehospital group. This trans-lated into a reduction in hospital mortalityfrom 8.1% to 5.7%, equivalent to 24 livessaved per 1000 treatments.39

In the United Kingdom there has not beenoverwhelming enthusiasm for prehospitalthrombolysis. Some cardiologists have high-lighted the delays in the field recording of thecardiograph, lack of physician control, and theproblem of ECG interpretation by prehospitalpersonnel. One of the major problems withparamedic administration of these drugs is theneed for strict protocols which exclude largenumbers of patients. In a preliminary paper bythe MITI investigators, Seattle paramedics hadto screen 2472 patients to find 107 whosatisfied the inclusion criteria for prehospitaltreatment.40 Paramedic skill levels in theadministration of thrombolysis tend to declinebecause of the relatively few occasions where itis appropriate to initiate the treatment in thefield. Some studies have failed to showimproved survival, and even the two trials citedabove did not find a reduction in all cause

mortality. Consequently, in the urban setting,patients should be transferred to hospitalbefore thrombolysis, although prehospitalidentification of elligible patients may reducedelays to treatment. Self referral and admissionby emergency ambulance to the nearestaccident and emergency department saves

time when compared with calling a generalpractitioner.41

In rural communities, where there may beproblems of access to hospital, there is a case

for domiciliary fibrinolytic therapy. In the

GREAT study of 311 patients living 16 to 62miles from the nearest district general hospital,domicilliary use of APSAC by general practi-tioners led to a median time saving of 139 min-utes and a reduction in death rate from 11.5%to 6.7%, a mortality gain of 48 lives per thou-sand patients treated.30 There were no prob-lems with ECG interpretation among partici-pating general practitioners. At follow up, oneyear after the conclusion of the study, themortality rate among those patients treatedearly after symptom onset was less than halfthat of the late treated patients, at 10.4% v21 .6%.42

A&E DEPARTMENT OR ccu?A 1994 British Heart Foundation workingparty report recommended a standard for "callto needle time"-the interval between thepatient's first contact with the emergency serv-ices and the administration of the drug-of notmore than 90 minutes.43 The interval mostamenable to reduction by hospital practition-ers is the "door to needle" time. One NorthAmerican standard recommends that thisinterval should not exceed 30 minutes.44Systems aiming to minimise in-hospital delays,described as "fast tracking", are designed toeliminate multiple assessments by junior doc-tors before admission to the CCU. Thediagnosis of acute myocardial infarction ismade solely on the basis of the history and thepresenting ECG, repeated if necessary. OldECGs need not necessarily be obtained if notimmediately to hand and delays for cardiacenzyme levels or chest radiographs are avoided.This kind of system may lead to time savings ofthe order of 45 minutes.45More than one model exists for the timely

initiation of thrombolysis in hospital. Whererapid admission to a cardiac care unit is notpossible the working party recommends thatthrombolytic treatment be started in A&E. Inpractice this is often the most efficient method.Many studies have reported time savings withA&E initiated treatment. In one trial involving1934 patients presenting to six hospitals, emer-gency department thrombolysis was associatedwith a reduction in door to needle times from amedian 80 minutes to 31 minutes.46 Suppliesof streptokinase and rt-PA may be stored in theA&E department by arrangement with thehospital pharmacy. A&E budget holdersshould anticipate a significant impact on drugsexpenditure when rt-PA is widely prescribedwithin the emergency department (GrahamCS, personal communication).The working party report did not specify

which specialty (A&E or medicine/cardiology)should be responsible for initiating treatment.Ideally this is a decision for relatively seniorresident physicians, but A&E staff are morelikely to be immediately available. Providedthat agreement can be reached between localcardiologists and A&E consultants there is noreason why the latter may not take responsibil-ity for the use of thrombolytics in theemergency department. Tight protocols, in-cluding a list of contraindications, posted in the

7

on Novem

ber 13, 2020 by guest. Protected by copyright.

http://emj.bm

j.com/

J Accid E

merg M

ed: first published as 10.1136/emj.14.1.2 on 1 January 1997. D

ownloaded from

Page 7: Thrombolysis after acute myocardial infarction · infarction (8.4% v 16.9% for anterior acute myocardial infarction) may account for these differences. Inferiorwall infarctions tendto

8 Nee

A&E department should ensure that the drugsare given appropriately. One small studyreported no significant increase in adverseevents when thrombolysis was initiated in A&Eby A&E doctors.47 Regular clinical auditpermits examination of treatment delays anddenials, inappropriate prescriptions, and com-plications of treatment.

Summary and conclusionsAppropriate use of a thrombolytic agent maysave 20 to 30 lives per 1000 treatments.Thrombolysis should be considered in all

patients presenting with cardiac chest painlasting more than 30 minutes for up to 12hours after symptom onset. ECG criteriainclude ST elevation of at least 1 mm in limbleads and/or at least 2 mm in two or more adja-cent chest leads or left bundle branch block.There is no upper age limit. All patients shouldalso receive oral aspirin and subcutaneous(intravenous with rt-PA) heparin. Other adju-vant treatments have been reviewed previouslyin this journal.48

Streptokinase is the drug of choice exceptwhere there is persistent hypotension, previousstreptokinase or APSAC at any time, knownallergy to streptokinase, or a recent provenstreptococcal infection. In these circumstancesthe patient should receive rt-PA. Additionalindications for rt-PA, based on subset analysisby the GUSTO investigators, include patientswith ALL of the following: age less than 75years, presentation within four hours of symp-tom onset, and ECG evidence of anterior acutemyocardial infarction.Treatment should be initiated as soon as

possible. The greatest benefit is observed inpatients treated early, pain to treat intervals ofless than one hour make possible mortalityreductions of nearly 50%. " When" mattersmore than "where": fast tracking to the CCU isone option but A&E initiated thrombolysis isfeasible and timely. Prehospital thrombolysis isappropriate in certain geographical situations.The development of practical guidelines for

thrombolysis represents the most comprehen-sive example of evidence based medicine.Streptokinase was first shown to influence out-come in acute myocardial infarction nearly 40years ago. More recently alternative regimeshave been evaluated in several prospective ran-domised controlled trials yielding pooled dataon nearly 60 000 patients. However, systematicreview of cumulative data reveals a statisticallysignificantmortalitygainforintravenous strepto-kinase over placebo which could have beenidentified as early as 1971-at least 15 yearsbefore it became generally used in clinicalpractice.49

1 Tillett WS, Garner RL. The fibrinolytic activity ofhemolytic streptococci. J Exp Med 1933;58:485-502.

2 Christensen LR, Macleod CM. A proteolytic enzyme ofserum:characterization, activation and reaction with inhibi-tors. J Gen Physiol 1945-28:559-83.

3 Sherry S. The origin of thrombolytic therapy. J Am CoillCardiol 1989;14:1085- 92.

4 Fletcher AP, Alkjaersig N, Smyrniotis FE, Sherry S.Treatment of patients suffering from early myocardialinfarction with massive and prolonged streptokinasetherapy. Trans Assoc Am Physicians 1958;71:287-96.

5 DeWood MA, Spores J, Notske R, Mouser LT, BurroughsR, Golden MS, et al. Prevalence of total coronary occlusion

during the early hours of transmural myocardial infarction.N Engl J Med 1980;303:897- 902.

6 Rentrop P, Blanke H, Kostering K, Karsch KR. Acute myo-cardial infarction: intracoronary application of nitroglyc-erin and streptokinase in combination with transluminalrecanalization. Clin Cardiol 1979;2:354-63.

7 Schroder R, Biamino G, Von Leitner ER, Linderer T,Bruggemann T, Heitz J, et al. Intravenous short-term infu-sion of streptokinase in acute myocardial infarction. Circu-lation 1983;63:536-48.

8 Smith RAG, Dupe RJ, English PD, Green J. Fibrinolysiswith acyl enzymes: a new approach to thrombolytictherapy. Nature 1981;290:505-8.

9 Ryken DC, Collen D. Purification and characterization ofthe plasminogen activator secreted by human melanomacells in tissue culture. J Biol Chem 1981;256:7035-41.

10 GISSI (Gruppo Italiano per lo Studio della Sopravvivenzanell Infarto Miocardico). Effectiveness of intravenousthrombolytic treatment in acute myocardial infarction.Lancet 1986;i:397-401.

11 ISIS-2 (second international study of infarct survival)Collaberative Group. Randomized trial of intravenousstreptokinase, oral aspirin, both or neither among 17,187cases of suspected acute myocardial infarction. Lancet1988;ii:349-60.

12 AIMS (APSAC intervention mortality study) Trial StudyGroup. Effects of intravenous APSAC on mortality afteracute myocardial infarction: preliminary report of aplacebo-controlled clinical trial. Lancet 1988;i:545-9.

13 R Wilcox RG, Von Der Lippe G, Olsson CG, Jensen G,Skene AM, Hampton JR for the ASSET (Anglo-Scandinavian study of early thrombolysis) study group.Trial of tissue plasminogen activator for mortality reduc-tion in acute myocardial infarction. Lancet 1988;ii:525-30.

14 Fibrinolytic Therapy Trialists' (F-I) Collaborative Group.Indications for fibrinolytic therapy in suspected acute myo-cardial infarction: collaborative overview of early mortalityand major morbidity results from all randomised trials ofmore than 1000 patients. Lancet 1994;343:311-22.

15 Braunwald E. Myocardial reperfusion, limitation of infarctsize, reduction of left ventricular dysfunction and improvedsurvival:should the paradigm be expanded? Circulation1989;79:441.

16 ISIS-3 (Third International Study of Infarct Survival) Col-laberative Group. A randomized trial of streptokinase vsanistreplase and of aspirin plus heparin vs aspirin aloneamong 41,299 cases of suspected acute myocardial infarc-tion. Lancet 1992;339:753-70.

17 Muller DWM, Topol EJ. Selection of patients with acutemyocardial infarction for thrombolytic therapy. Ann InternMed 1990;1 13:949-60.

18 ISAM (Intravenous Streptokinase in Acute Myocardial Inf-arction) Study Group. A prospective trial of intravenousstreptokinase in acute myocardial infarction: mortality,morbidity and infarct size at 21 days. N Engl J Med 1986;314: 1465-71.

19 Maggioni AP, Maseri A, Fresco C, Franzosi MG, Mauri F,Santoro E, et al. Age related increase in mortality amongpatients with first myocardial infarctions treated withthrombolysis. The investigators of the Gruppo Italiano perlo Studio della Sopravvivenza nell Infarto miocardico(GISSI 2). N EnglJ Med 1993;329:1442-8.

20 Krumholz HM, Pasternak RC, Weinstein MC, FriesingerGC, Ridker PM, Tostenson AN, et al. Cost effectiveness ofthrombolytic therapy with streptokinase in elderly patientswith suspected acute myocardial infarction. N Engl J Med1992;327:7-13.

21 Elliott JM, Cross DB, Cederholm-Williams SA, White HD.Neutralizing antibodies to streptokinase four years afterintravenous thrombolytic therapy. Am J Cardiol 1993;71:640-5.

22 Rawles J. Deciding who should have thrombolysis (letter).BMJ 1993;307:799- 80.

23 Channer KS. Deciding who should have thrombolysis(letter). BMJ 1993;307:1146.

24 LATE (Late Assessment of Thrombolytic Efficacy) StudyGroup. Late assessment of thrombolytic efficacy withaltiplase 6-24 hours after onset of acute myocardial infarc-tion. Lancet 1993;342:759-66.

25 Cross SJ, Lee SH, Rawles JM, Jennings K. Safety of throm-bolysis in association with cardiopulmonary resuscitation.BMJ 1991;303:1241.

26 Scholz KH, Tebbe U, Herrmann C, Wojcik J, Lingen R,Chemnitius JM, et al. Frequency of complications of cardi-opulmonary resuscitation after thrombolysis during acutemyocardial infarction. Am J Cardiol 1992;69:724-8.

27 Weston CFM, Avery P. Thrombolysis following pre-hospitalcardiopulmonary resuscitation. Int J Cardiol 1992;37:195-8.

28 Tenaglia AN, Califf RM, Candela RJ, Kereiakes DJ, BerriosE, Young SY, et al. Thrombolytic therapy in patientsrequiring cardiopulmonary resuscitation. Am J Cardiol1991;68:1015- 9.

29 Risenfors M, Gustavsson G, Ekstrom L, Hartford M, Her-lintz J, Karlson BW, et al. Prehospital thrombolysis in sus-pected acute myocardial infarction:Results from the TEA-HAT study. J Intern Med 1991;734(suppl):3-10.

30 GREAT Group. Feasibility, safety and efficacy of domicili-ary thrombolysis by General Practitioners: GrampianRegion early anistreplase trial. BMJ 1992;305:548-58.

31 EMERAS (Estudio Multicentrico Estreptoquinasa Republi-cas de America del Sur) Collaberative Group. Randomizedtrial of late thrombolysis in patients with suspected acutemyocardial infarction. Lancet 1993;342:767-72.

on Novem

ber 13, 2020 by guest. Protected by copyright.

http://emj.bm

j.com/

J Accid E

merg M

ed: first published as 10.1136/emj.14.1.2 on 1 January 1997. D

ownloaded from

Page 8: Thrombolysis after acute myocardial infarction · infarction (8.4% v 16.9% for anterior acute myocardial infarction) may account for these differences. Inferiorwall infarctions tendto

Thrombolysis after acute myocardial infarction 9

32 GISSI (Gruppo Italiano per lo Studio della Streptochinasinell'Infarto miocardico) A factorial randomized trial ofalteplase versus streptokinase and heparin versus noheparin among 12,490 patients with acute myocardial inf-arction. Lancet 1990;336:65-71.

33 Chesbro JH, Knatterud G, Roberts R, Borer J, Cohen LS,Dalen J, et al. Thrombolysis in myocardial infarction(TIMI) trial, phase I: a comparison between intravenoustissue plasminogen activator and intravenous streptokinase.Clinical findings through hospital discharge. Circulation1987;76: 142-54.

34 O'Donnell M. Battle of the clotbusters. BMJ 1991;302:1259-61.

35 Global utilization of streptokinase and tissue plasminogenactivator for occluded coronary arteries (GUSTO) Investi-gators. An international randomized trial comparing fourthrombolytic strategies for acute myocardial infarction. NEngl J Med 1993;329:673-82.

36 Global utilisation of streptokinase and tissue plasminogenfor occluded coronary arteries (GUSTO) investigators.Theeffects of tisssue plasminogen activator, streptokinase orboth on coronary artery patency,ventricular function andsurvival after acute myocardial infarction. N Engl J Med1993;329:1615-22.

37 Lee KL, Califf RM, Simes J, Van der Werf F, Topol EJ.Holding GUSTO up to the light: global utilisation of strep-tokinase and tissue plasminogen activator for occludedcoronary arteries. Ann Intern Med 1994;120:876-81.

38 Leizorovicz A, Boissel JP, Julian D, Castaigne A, Haugh MC(The European Myocardial Infarction Project Group).Prehospital thrombolytic therapy in patients with sus-pected acute myocardial infarction. N Engl J Med1993;329:383-9.

39 Weaver WD, Cerqueira M, Hallstrom AP, Litwin PE, Mar-tin JS, Kudenchuk PJ, et al, for the Myocardial InfarctionTriage and Intervention (MITI) Project Group.

Prehospital-initiated versus hospital-initiated thrombolytictherapy. JAMA 1993;270:1211-6.

40 Weaver WD, Eisenberg MS, Martin JS, Litwin PE, ShaefferSM, Ho MT, et al. Myocardial Infarction Triage and Inter-vention Project-phase 1: patient characteristics and feasi-bility of prehospital initiation of thrombolytic therapy. J AmColl Cardiol 1990;15:925-31.

41 Pell ACH, Miller HC. Delays in admission of patients withacute myocardial infarction to coronary care:Implicationsfor thrombolysis. Health Bull 1990;48:225-31.

42 Rawles J. Halving of mortality at one year by domicilliarythrombolysis in the Grampion Region Early AnistreplaseTrial (GREAT). J Am Coll Cardiol 1994;23:1-5.

43 Weston CFM, Penny WJ, Julian DG, on behalf of the Brit-ish Heart Foundation working Group.Guidelines for theearly management of patients with myocardial infarction.BMJ 1994;308:767-71.

44 Eisenberg MS. Reducing the door to drug interval forthrombolytic therapy. Acad Emerg Med 1995;2:579-80.

45 Pell ACH, Miller HC, Robertson CE, Fox KAA. Effect of'fast track' admission for acute myocardial infarction ondelay to thrombolysis. BMJ 1992;304:83-7.

46 Birkhead JS. Time delays in provision of thrombolytic treat-ment in six district hospitals. Joint Audit Commitee of theBritish Cardiac Society and a cardiological commitee of theRoyal College of Physicians of London. BMJ 1992;305:445-8.

47 Nee PA, Gray AJG, Martin MAM. Audit of thrombolysisinitiated in an accident and emergency department. Qual-ity in Health Care 1994;2:29-33

48 Vincent R. Advances in the early diagnosis and managementof acute myocardial infarction. J Accid Emerg Med1996;13:74-9.

49 Mulrow CD. Rationale for systematic reviews. In: ChalmersI, Altman DG, eds. Systematic reviews. London: BMJ pub-lishing group, 1995.

THE FACULTY OF ACCIDENT ANDEMERGENCY MEDICINESPECIALTY EXAMINATION

The Specialty Examination of the Faculty of Accident andEmergency Medicine will be held on the following dates

21/22 May 1997 at the Royal College of Surgeons ofEdinburgh Closing date - 26 March 1997

Fee - £55012/13 November 1997 at the Royal College of Surgeons

of England Closing date - 17 September 1997Fee - £600

Regulations and application forms are available from:The Secretariat

Intercollegiate Specialty Boards3 Hill SquareEdinburghEH8 9DR

Tel: 0131 662 9222Facs: 0131 662 9444

on Novem

ber 13, 2020 by guest. Protected by copyright.

http://emj.bm

j.com/

J Accid E

merg M

ed: first published as 10.1136/emj.14.1.2 on 1 January 1997. D

ownloaded from