104 REVIEWS IN CARDIOVASCULAR MEDICINE FALL 2000 U nstable angina (UA), a major reason for emergency department visits, is responsible for more than 1 million hospital admissions annually in this country. 1 It is, therefore, imperative that health care providers be famil- iar with recent developments in the diagnosis and management of this condition. Acute coronary syndrome (ACS) refers to a constellation of clinical signs and symptoms produced by acute myocardial ischemia. It comprises UA and acute myo- cardial infarction (MI) associated with ST-segment elevation or depression. Patients with angina and no ST-segment elevation have either UA or non–ST-segment ele- vation MI (NSTEMI). The 2 conditions can be differentiated by the presence or absence of the circulating markers of myocardial necrosis. Braunwald and colleagues 2,3 have described 3 principal presentations of UA, which include rest angina, new-onset angina, and increasing angina. Rest angina is angina occurring at rest, usually lasting longer than 20 minutes and occurring within a week of presentation. New-onset angina is angina of at least Canadian Car- diovascular Society Classification III severity, with onset within 2 months of initial presentation. Increasing angina is previously diagnosed angina that is distinctly more frequent, is longer in duration, or occurs at a lower threshold. Great strides have been made in the understanding of unstable angina and its relationship to the acute coronary syndromes and myocardial infarction during the last decade of the 20th century. Detailed information about ECG changes and serum cardiac markers, as well as the conclusions drawn from numerous large, randomized interventional trials can now be integrated into the traditional clinical picture. Clinicians can now classify patients into diagnostic and prognostic categories and can perform risk stratification with unprecedented precision. With this information, the decision to hospitalize patients and the selection of noninvasive or invasive evaluation and management strategies can be individualized for optimal outcomes. [Rev Cardiovasc Med. 2000;1(2):104-119] Key words: Angina, unstable • Anticoagulation therapy • Antiplatelet therapy • Glycoprotein receptor antagonists • Myocardial ischemia • Revascularization TREATMENT UPDATE Management of Unstable Angina: Integrating the New Approaches Vatsal H. Mody, MD,* David P. Faxon, MD † *University of Southern California Keck School of Medicine, Los Angeles † University of Chicago Pritzker School of Medicine
Transcript
104 REVIEWS IN CARDIOVASCULAR MEDICINE FALL 2000
Unstable angina (UA), a major reason for emergency department visits, isresponsible for more than 1 million hospital admissions annually in thiscountry.1 It is, therefore, imperative that health care providers be famil-
iar with recent developments in the diagnosis and management of this condition.Acute coronary syndrome (ACS) refers to a constellation of clinical signs and
symptoms produced by acute myocardial ischemia. It comprises UA and acute myo-cardial infarction (MI) associated with ST-segment elevation or depression. Patientswith angina and no ST-segment elevation have either UA or non–ST-segment ele-vation MI (NSTEMI). The 2 conditions can be differentiated by the presence or absence of the circulating markers of myocardial necrosis.
Braunwald and colleagues2,3 have described 3 principal presentations of UA,which include rest angina, new-onset angina, and increasing angina. Rest anginais angina occurring at rest, usually lasting longer than 20 minutes and occurringwithin a week of presentation. New-onset angina is angina of at least Canadian Car-diovascular Society Classification III severity, with onset within 2 months of initialpresentation. Increasing angina is previously diagnosed angina that is distinctlymore frequent, is longer in duration, or occurs at a lower threshold.
Great strides have been made in the understanding of unstable angina and its relationship to the acute coronary syndromes and myocardial infarction during the last decade of the 20th century. Detailed information about ECG changes andserum cardiac markers, as well as the conclusions drawn from numerous large, randomized interventional trials can now be integrated into the traditional clinicalpicture. Clinicians can now classify patients into diagnostic and prognostic categoriesand can perform risk stratification with unprecedented precision. With this information, the decision to hospitalize patients and the selection of noninvasive orinvasive evaluation and management strategies can be individualized for optimal outcomes. [Rev Cardiovasc Med. 2000;1(2):104-119]
Management of Unstable Angina:Integrating the New Approaches
Vatsal H. Mody, MD,* David P. Faxon, MD†
*University of Southern California Keck School of Medicine, Los Angeles†University of Chicago Pritzker School of Medicine
FALL 2000 REVIEWS IN CARDIOVASCULAR MEDICINE 105
UNSTABLE ANGINA
PathogenesisACS is produced by an imbalance be-tween myocardial oxygen supply anddemand. The most common cause ofUA/NSTEMI is reduced myocardialblood flow from a nonocclusivethrombus that has formed over a dis-rupted atherosclerotic plaque. Factorsthat induce and promote inflamma-tion or atherogenesis, thus giving riseto an atherosclerotic lesion, includehypercholesterolemia and other lipid-related abnormalities; hypertension;high plasma homocysteine concentra-tions; endothelial insult secondary tosmoking; and, possibly, certain infec-tions, such as those caused by herpes-viruses or Chlamydia pneumoniae.4
In the advanced lesion, a fibrouscap separates the lesion from the arte-rial lumen. The cap covers a mixture ofleukocytes, macrophages, foam cells,lipids, and debris, which may form anecrotic core (Figure 1).4 A typical vul-nerable lesion consists of an eccentricplaque, rich in extracellular lipids with-in a large lipid pool, and has a thin, fi-brous cap. Degradation, followed bycap rupture, may result from expres-sion of matrix metalloproteinases (col-lagenases, elastases, and stromelysins)by macrophages. Cap rupture exposesthe lipid core, the most potent sub-strate for platelet-rich thrombus for-mation, to flowing blood.
Plaque disruption leads to plateletactivation, adhesion, and aggregation.Platelet activation causes changes inthe shape of platelets and conforma-tional alterations in glycoprotein (GP)IIb/IIIa receptors, allowing them tobind to fibrinogen. By forming bridgesbetween platelets, fibrinogen facili-tates platelet aggregation.5
Plaque disruption also leads to therelease of tissue factor, which is ex-pressed by smooth muscle and foam
cells in unstable plaques. Tissue factorinteracts with factor VIIa to initiate thecoagulation cascade. This causes localthrombin generation, fibrin deposi-tion and, ultimately, platelet-richthrombus formation.
The less common causes of UA/NSTEMI include severe spasm of a seg-ment of epicardial artery with onlymild coronary atherosclerosis (Prinz-metal angina), severe narrowing of anepicardial coronary artery withoutthrombus or spasm, and differenttypes of infections resulting in inflam-mation of the coronary arteries. Severeanemia, fever, tachycardia, and thyro-toxicosis can cause UA/NSTEMI by in-creasing oxygen demand when myo-cardial blood flow is compromised byatherosclerotic coronary stenosis.
Initial Evaluation andRisk StratificationWhen patients have symptoms thatsuggest ACS, the likelihood of coro-
nary artery disease (CAD) can be esti-mated from the medical history, phys-ical examination findings, ECG data,and serum marker measurements. Inparticular, the diagnostic and prognos-tic roles of the ECG and serum markershave become better characterized dur-ing the past several years. If the likeli-hood of CAD is intermediate or high,the clinical presentation is used to cat-egorize patients into a low, intermedi-ate, or high level of short-term risk ofischemic events, such as death or non-fatal MI.3
ECG. The 12-lead ECG is criticalnot only for adding support to theclinical suspicion of CAD but also forproviding prognostic information(Table 1). Transient ST-segment or T-wave changes that develop during asymptomatic episode at rest and re-solve when the patient becomesasymptomatic strongly suggest acuteischemia and a very high likelihood ofunderlying severe CAD.6 Inverted
Figure 1. Pathogenesis of the acute coronary syndrome. Vulnerable atherosclerotic plaques(those with a large lipid core, a thin fibrous cap, and inflammatory cells) can rupture. Shouldthe cap rupture, the highly thrombogenic lipid core may become the substrate for thrombusformation that can lead to partial or complete occlusion of the vessel.
Thinning and Foam celldecay of fibrous cap T-cell activation
Formation of Hemorrhage from necrotic core plaque microvessel
MacrophageMast cell
106 REVIEWS IN CARDIOVASCULAR MEDICINE FALL 2000
UNSTABLE ANGINA continued
T waves may indicate ischemia ornon–Q wave infarction, especiallywith T-wave inversion of at least 0.1mV in leads with dominant R waves.In patients with ACS, symmetric pre-cordial T-wave inversion of at least 0.3mV strongly suggests acute ischemia—caused, most likely, by a critical steno-sis of the left anterior descending coro-nary artery.7
Nonspecific ST-segment and T-wavechanges, usually defined as ST-seg-ment deviation no greater than 0.5mm or T-wave inversion of 0.1 mV orless, are less strongly diagnostic for UA.Established Q waves lasting at least0.04 second are also less helpful in thediagnosis of UA, although they indi-cate a high likelihood of significantCAD. A completely normal ECG in apatient with chest pain does not ex-
clude the possibility of ACS, since 1%to 6% of such patients eventuallyprove to have had an acute MI andsince 4% or more will be found to haveUA.8
The risk of death, based on admis-sion ECG data, is highest when there isa left bundle branch block (LBBB) pat-tern, paced rhythm, and left ventricu-lar hypertrophy. Risk of death is inter-mediate with ST-segment deviationsand lowest with isolated T-wavechanges. In the Thrombolysis in Myo-cardial Infarction (TIMI)-III Registry,patients with ST-segment depressionof 1 mm or greater had an 11% rate ofdeath or nonfatal MI at 1 year, andthose with LBBB had rates of 22.9%.9
In the majority of patients, there wereno ECG changes or only isolated T-wave changes; in these patients, the
rates of death or MI at 1 year were6.8% and 8.2%, respectively.
Serum cardiac markers. For pa-tients presenting without ST-segmentelevation, serum cardiac markers pro-vide valuable diagnostic and prognos-tic information. Until recently, CK-MB, the MB isoenzyme of creatine ki-nase (CK), has been the principalserum cardiac marker used to evaluatepatients with ACS. Despite its wide-spread acceptance, however, CK-MBmeasurement has several limitations.First, a low level of CK-MB in the bloodof healthy individuals limits sensitivi-ty and specificity for detecting myo-cardial necrosis. Second, CK-MB maybe elevated in severe damage of skele-tal muscle.10
Immunoassays have been devel-oped to detect cardiac-specific tro-ponin T (cTnT) and cardiac-specifictroponin I (cTnI). Both are sensitiveand specific markers of myocyte dam-age. For the clinical management ofpatients with ACS, cTnT and cTnI as-says are of comparable diagnostic andprognostic efficacy (Table 2). In pa-tients with renal insufficiency, howev-er, cTnI is superior to cTnT.11 Approxi-mately 30% of patients who presentwith rest pain but no ST-segment ele-vation and in whom a misdiagnosis ofUA would be made because they donot have elevated CK-MB levels are ac-tually found to have NSTEMI when as-sessed using cardiac-specific troponinassays.12
Elevated levels of cTnI or cTnT con-vey prognostic information beyondthat supplied by the clinical character-istics of the patient, the ECG at pre-sentation, and the predischarge exer-cise test.13,14 In patients without ST-seg-ment elevation and normal CK-MBlevels, elevated cTnI or cTnT concen-trations identify those at an increased
Table 1ECG Evidence of Unstable Angina
ECG pattern in a patient with possible ACS Interpretation
Established Q waves ≥ 0.04 s Suggest high probability of CAD but not strongly diagnostic for UA
Inverted T waves (especially Ischemia or non–Q waveof ≥ 0.1 mV in leads with infarctiondominant R waves)
Normal ECG with chest pain UA, acute MI, and CAD cannot beruled out
ST-segment deviation ≤ 0.5 mm Not strongly diagnostic for UA
risk for death.13,15 In addition, there isan almost linear relationship betweenthe cTnI or cTnT level and the risk ofmortality in patients presenting withACS (Figure 2).13,15,16 Moreover, patientspresenting without ST-segment eleva-tion but with elevated cardiac-specifictroponin levels may receive a greaterbenefit from treatment with GP IIb/IIIa inhibitors or low molecular weightheparin (LMWH).17,18
The clinical value of serial determi-nations of myoglobin levels for diag-nosing MI is limited by the brief dura-tion of myoglobin level elevation (lessthan 24 hours) and by lack of cardiacspecificity. A negative myoglobin testis extremely useful in ruling out myo-cardial necrosis, however.
Elevated plasma fibrinogen levelsindicate increased risk in patients withACS.19 Patients with elevated levels ofhigh-sensitivity C-reactive protein (amarker for acute phase of inflamma-tion) and with no elevation of tro-ponin levels on admission are at in-creased risk for an adverse outcome.20
Elevated levels of interleukin-6 andserum amyloid-A also imply an ad-verse outcome in patients with ACS.21,22
Early Hospital CareEarly inhospital care of patients withUA or ACS comprises therapy with an-tiplatelet agents, anticoagulants, GPIIb/IIIa receptor antagonists, and anti-ischemic drugs. Effective antiplateletagents include aspirin (acetylsalicylicacid [ASA]) and the thienopyridines,ticlopidine and clopidogrel. Sulfin-pyrazone, dipyridamole, prostacyclin,and prostacyclin analogs have notbeen associated with benefit in UA.Recommended anticoagulation thera-py includes LMWH and direct throm-bin inhibitors as well as unfractionated
heparin (UFH). Available GP IIb/IIIa re-ceptor antagonists are abciximab, epti-fibatide, and tirofiban. Anti-ischemicdrugs are nitrates, ß-blockers, and cal-cium channel blockers and are fre-quently used in conjunction with an-giotensin-converting enzyme inhib-itors (ACEIs).
Thrombolytic therapy does not im-prove clinical outcomes in the absenceof acute MI with ST-segment elevationor LBBB.23,24 A meta-analysis of throm-bolytic therapy in patients with UAdoes not support its use.2
Antiplatelet therapy. ASA preventsthe formation of thromboxane A2 by ir-reversibly inhibiting cyclooxygenase-1within platelets, thereby diminishingplatelet aggregation. Various studieshave shown that ASA reduces the riskof death or MI by 51% to 72% in pa-tients presenting with UA.25-27 In no tri-al, however, has the efficacy of differentdoses of ASA been compared. Usually, a
dosage of 160 or 325 mg is given to pa-tients with UA. The first dose should begiven as soon as the diagnosis of ACSis suspected, and it may be chewed sothat a high blood level of ASA is estab-lished rapidly.
The thienopyridine drugs inhibitadenosine diphosphate receptors andprevent transformation of GP IIb/IIIareceptors into the high-affinity stage.Two agents—ticlopidine and clopido-grel—are currently approved for an-tiplatelet therapy. Both are reasonableantiplatelet therapies for secondaryprevention and have an efficacy simi-lar to that of ASA.
Ticlopidine is effective for the sec-ondary prevention of stroke and MI. Ithas also been used for the preventionof stent closure and graft occlusion. Ina trial involving 652 patients with UA,ticlopidine reduced the rate of fataland nonfatal MI at 6 months by 46%(13.6% in patients not taking ticlopi-
Table 2Using Serum Cardiac Markers in UA/NSTEMI and ACS
• cTnT and cTnI assays have comparable diagnostic and prognostic valuewhen renal function is normal.
• cTnI is superior to cTnT in patients with renal insufficiency.
• Elevated cTnI or cTnT levels predict increased risk of death in patientswith normal CK-MB levels and no ST-segment elevation.
• Elevations in cTnI or cTnT levels may allow diagnosis of NSTEMI ratherthan UA in patients with rest angina, no ST-segment elevation, and normal CK-MB levels.
• Elevations in hs-CRP imply increased risk of poor outcome when troponin levels are normal.
• Elevations in levels of interleukin-6 and serum amyloid-A imply adverseoutcome in ACS.
• Elevations in plasma fibrinogen levels suggest increased risk in patientswith ACS.
dine versus 7.3% in patients taking it,P = .009).28 The benefit of ticlopidinewas apparent only after 2 weeks oftreatment.
Adverse effects of ticlopidine in-clude GI problems (such as diarrhea,abdominal pain, nausea, and vomit-ing), mild to moderate neutropenia (inapproximately 2.4% of patients), se-vere neutropenia (in 0.8% of patients),and thrombotic thrombocytopenicpurpura (TTP). Neutropenia usually resolves within 1 to 3 weeks after discontinuing therapy but may be fatal (although this is rare). TTP—an uncommon, life-threatening compli-cation—requires immediate plasma-pheresis. Patients taking ticlopidineneed to be monitored with a completeblood cell count, including a differen-tial count, every 2 weeks for the first 3months of therapy.
Clopidogrel is at least as effective asASA and may be slightly more so. Inthe Clopidogrel versus Aspirin in Pa-tients at Risk of Ischaemic Events (CAPRIE) trial, 19,185 patients with
recent ischemic stroke, recent MI, orsymptomatic atherosclerotic peripher-al vascular disease were randomized toreceive treatment with ASA, 325 mg/d,or clopidogrel, 75 mg/d.29 The relativeannual risk of ischemic stroke, MI, orvascular death was reduced by 8.7% in favor of clopidogrel, from 5.83% to 5.32% (P = .043).
A randomized, multicenter trialcomparing clopidogrel with ticlopi-dine, the Clopidogrel Aspirin Stent International Cooperative Study(CLASSICS), enrolled 1020 patientsundergoing coronary stenting.30 Thetrial showed better tolerance of clopid-ogrel, with or without a loading dose,than of ticlopidine. Stent thrombosisand other major complications oc-curred at the same frequency in all ofthe groups in the study.
Both clopidogrel and ticlopidineare indicated for patients who have UAbut who are unable to tolerate ASA be-cause of hypersensitivity or major GIcontraindications, especially gastritisor recent bleeding from a peptic ulcer.
Clopidogrel is preferred over ticlopi-dine because it is more potent, inhibitsplatelets more rapidly, possesses alonger half-life (making it suitable foronce-daily administration), and issafer. Clopidogrel was associated witha very low incidence of neutropenia inthe CAPRIE trial, but active surveil-lance of patients (not necessarily inrandomized trials) taking clopidogrelidentified 11 instances of TTP over a 2-year period.31
For patients with UA who are treat-ed with ticlopidine or clopidogrel, ini-tial treatment with UFH or LMWH isespecially important, because the on-set of antiplatelet activity is delayedrelative to ASA. With ticlopidine, careneeds to be taken during the acutephase of illness, because there is a 2-week delay before the full antiplateleteffect is reached.
Anticoagulation therapy. Heparinprevents thrombus propagation butdoes not lyse existing thrombi.32 In 2meta-analyses, 1 involving 3 random-ized trials33 and the other involving 6trials,27 the relative risk of death or MIwith the combination of ASA and hep-arin was reduced by 56% (P = .03) andby 33% (P = .06), respectively.
UFH is a glycosaminoglycan madeup of polysaccharide chains ranging inmolecular weight from 5000 to 30,000daltons. It exhibits both anti-Xa andanti-IIa activity, is sensitive to inactiva-tion by platelet factor 4, and bindsnonspecifically to plasma proteins andendothelial cells, which compromisesits ability to interact with antithrom-bin. It also has a variable dose-responsecurve and is relatively inefficient at in-hibiting the generation of thrombinbecause of the length of its polysac-charide chains. It exerts its anticoagu-lant effect by accelerating the action ofcirculating antithrombin III, a proteo-
Figure 2. Mortality at 42 days in patients with unstable angina according to baseline levels oftroponin I.15
lytic enzyme that inhibits thrombin(factor IIa), factor IXa, and factor Xa.
Most of the benefit of anticoagu-lants is short-term. Clinical trials haveindicated that a weight-adjusted dos-ing regimen could provide more pre-dictable anticoagulation than thefixed-dose regimen. The weight-ad-justed regimen is recommended; a bo-lus of 60 to 80 U/kg of heparin is given,followed by an initial infusion rate of12 to 18 U/kg/h, to achieve a target ac-tivated partial thromboplastin time(aPTT) in the range of 1.5 to 2.0 timescontrol.
The optimal duration of therapy re-mains undefined. In most of the trialsevaluating the use of UFH in UA, anti-coagulation therapy lasted for 2 to 5days. Mild thrombocytopenia may oc-cur in 10% to 20% of patients receiv-ing heparin and usually appears in thefirst days of therapy. Severe thrombo-cytopenia (platelet count lower than100,000/mL) occurs in 1% to 2% of pa-tients and typically appears after 3 to5 days of therapy.
A dangerous but rare complication,with an incidence of less than 0.2%, is autoimmune UFH-induced throm-bocytopenia with thrombosis (alsoknown as heparin-induced thrombo-cytopenia [HIT]). Serial platelet countsare necessary to monitor for HIT, and ahigh clinical suspicion of HIT man-dates immediate cessation of all hep-arin therapy (including that used toflush intravenous lines).
LMWHs are obtained by chemicalor enzymatic depolymerization of thepolysaccharide chains of heparin toprovide smaller chains, with molecu-lar weights ranging from 1000 to10,000 daltons. LMWHs are more se-lective in catalyzing the inhibition offactor Xa by antithrombin III and areless selective in inactivating thrombin.
The anti-Xa to anti-IIa ratios of differ-ent LMWHs vary, ranging from 1.9 to3.8.32
LMWH has distinct clinical advan-tages over UFH (Table 3). These in-clude less binding to plasma proteinsand endothelial cells; dose-indepen-dent clearance; and a longer half-life,which results in more reproducibleand sustained anticoagulation. As a re-sult, certain LMWHs can be adminis-tered subcutaneously twice daily.LMWHs stimulate platelets less thanUFH does and are less frequently asso-ciated with HIT. Furthermore, moni-toring of aPTT is not required withLMWH use.
A disadvantage of LMWH use isthat anticoagulation is less effectivelyreversed with protamine, compared
with UFH. As a result, planning is re-quired when coronary artery bypassgraft (CABG) surgery is anticipated. Inaddition, administration of LMWHduring percutaneous catheter inter-ventions (PCIs) precludes monitoringof activated clotting time for titratingthe dosage of anticoagulant.
In a pilot open-label study byGurfinkel and colleagues,34 the combi-nation of ASA and the LMWHnadroparin significantly reduced thetotal ischemic event rate, the rate of re-current angina, and the number of in-tervention procedures. The FragminDuring Instability in Coronary ArteryDisease (FRISC) study randomized1506 patients with UA or non–Q waveMI to receive either the LMWH dal-teparin subcutaneously or placebo.35
Table 3Advantages and Disadvantages of LMWHs
Advantages
Administration is subcutaneously, once or twice daily (some LMWHs)
No need for aPTT testing q6h
Incidence of HIT is lower than with UFH
Less binding to plasma proteins and endothelial cells than is UFH
Less stimulation of platelets than is produced by UFH
Longer half-life than has UFH
More reproducible and sustained anticoagulation than with UFH
Superior efficacy in reducing cardiac events and revascularization in UA/NSTEMI
Disadvantages
Anticoagulation more difficult to reverse with protamine, compared with UFH
Planning required when CABG surgery may be necessary
Dalteparin was associated with a 63%risk reduction in death or MI duringthe first 6 days (4.8% versus 1.8%, P < .001). At 40 days, a significant de-crease was observed in the compositeoutcome of death, MI, or revascular-ization (23.7% versus 18%, P = .005).There was no difference in the rates ofend points after 150 days.
The safety and efficacy of LMWHand UFH in patients with UA havebeen compared in 4 large randomizedtrials. These include the Fragmin inUnstable Coronary Artery Disease(FRIC) study,36 the Fraxiparine in Isch-
aemic Syndrome (FRAXIS) study,37 theEfficacy and Safety of SubcutaneousEnoxaparin in Non–Q-Wave CoronaryEvents (ESSENCE) study,38 and the TIMI-11B trial.34 Two trials document-ed a moderate benefit of the LMWHenoxaparin over UFH, and 2 trials, 1with nadroparin and 1 with dal-teparin, reported neutral or unfavor-able trends associated with LMWH use(Table 4).
The ESSENCE trial showed thatenoxaparin was more effective thanUFH in reducing rates of death, MI, orrecurrent angina. TIMI-11B confirmed
that outcomes with enoxaparin werebetter than with UFH but also showedthat prolonged enoxaparin therapy in-creased major bleeding without pro-viding additional risk reduction. Theconclusion of the FRIC study was thatdalteparin may be an alternative toUFH for patients with UA/NSTEMI butthat prolonged therapy with dal-teparin provided no benefit over ASAand heparin. In the FRAXIS study, out-comes with nadroparin were equiva-lent to outcomes with UFH, but no ad-ditional benefit accrued from the useof nadroparin for longer than 14 days.
The inconsistent outcomes may re-flect the fact that the studies used sev-eral LMWHs, with different molecularweights and anti-Xa/anti-IIa ratios(3.8, 2.7, and 3.6 for enoxaparin, dal-teparin, and nadroparin, respectively).The 2 trials with positive results(ESSENCE and TIMI-11B) used enoxa-parin, suggesting that differencesamong LMWHs may exist. A meta-analysis of these 2 trials, involving atotal of 7081 patients, showed a statis-tically significant reduction of approx-imately 20% in the rate of death, MI,or refractory ischemia at 2, 8, 14, and43 days and in the rate of death or MIat 8, 14, and 43 days.38
LMWHs are associated with signifi-cantly more frequent minor—but notmajor—bleeding, according to theESSENCE and TIMI-11B trials. Addi-tional experience with regard to safetyand efficacy of concomitant adminis-tration of LMWHs with GP IIb/IIIa an-tagonists and thrombolytic agents iscurrently being acquired. The FRISC,FRIC, TIMI-11B, and Fragmin and FastRevascularisation During Instability inCoronary Artery Disease (FRISC-II) tri-als evaluated the potential benefit ofprolonged administration of LMWHtherapy after hospital discharge. The
Table 4Conclusions From 4 Randomized Trials Comparing
Low Molecular Weight Heparin and Unfractionated Heparin in Patients With Unstable Angina
Trial Regimen Conclusion
FRIC Acute phase: dalteparin, Dalteparin may be an120 IU/kg SC bid; alternative to UFH for patientsprolonged phase: with UA/NSTEMI; prolongeddalteparin, 7500 IU qd therapy with dalteparin at a
lower dose qd did not confer any benefit over ASA + heparin
ESSENCE Enoxaparin, 1 mg/kg Combination enoxaparin + ASASC bid vs IV UFH was more effective than UFH + for 2 - 8 d ASA in reducing death, MI, or
recurrent angina
TIMI-11B IV UFH vs enoxaparin Enoxaparin produced better for 2 - 8 d; following 14-d outcome, compareddischarge, UFH group with UFH; prolonged therapy received placebo while with enoxaparin provided no enoxaparin group additional benefit and was continued therapy associated with more major for 43 d bleeding
FRAXIS IV UFH vs nadroparin Nadroparin for 14 d was asso-for 6 ± 2 d vs nadroparin ciated with worse outcome at for 14 d 90 d, compared with UFH
FRIC, Fragmin in Unstable Coronary Artery Disease; UFH, unfractionated heparin; UA, unstableangina; NSTEMI, non–ST-segment elevation myocardial infarction; ASA, acetylsalicylic acid (as-pirin); ESSENCE, Efficacy and Safety of Subcutaneous Enoxaparin in Non–Q-Wave Coronary Events;MI, myocardial infarction; TIMI-11B, Thrombolysis in Myocardial Infarction; FRAXIS, Fraxiparin inIschaemic Syndrome.
FALL 2000 REVIEWS IN CARDIOVASCULAR MEDICINE 111
UNSTABLE ANGINA
first 3 of these trials did not show abenefit of treatment beyond the acutephase. In FRISC-II, the composite pri-mary end point of death or MI, or ofrevascularization was reduced at 3months (P = .03), but the benefits werenot sustained at 6 months.39
Hirudin is a potent direct thrombininhibitor that binds the anion bindingsite and the catalytic site of thrombinwithout the need for a cofactor. Sever-al large trials comparing hirudin withUFH in patients with UA/NSTEMIhave demonstrated a modest reduc-tion in the composite end point ofdeath or nonfatal MI and a modest in-crease in the risk of bleeding. Hirudinpossesses no circulating inhibitor andproduces highly reproducible antico-agulation. A long-acting preparationof hirudin (PEG-hirudin) suitable foronce-daily subcutaneous administra-tion is currently being investigated.
Based on an analysis that includedthe Global Use of Strategies to OpenOccluded Arteries in Acute CoronarySyndromes (GUSTO-IIb), the TIMI-9B,the Organisation to Assess Strategiesfor Ischaemic Syndromes (OASIS)-1,and the OASIS-2 trials, further study ofthe use of hirudin in the managementof UA/NSTEMI seems warranted. Ac-cording to this analysis, the relativerisks of death or MI 35 days after ran-domization were 0.90 with hirudin,compared with UFH (P = .15), 0.88 forpatients receiving thrombolytics (P = .13), and 0.90 for patients not re-ceiving thrombolytics (P = .54).40
GP IIb/IIIa receptor antagonists.The GP IIb/IIIa receptor is specific toand abundant on the platelet surface.When platelets are activated, this re-ceptor undergoes a conformationalchange that increases its affinity forbinding fibrinogen and other ligands.Binding of 1 molecule of fibrinogen
to receptors on 2 platelets results inplatelet aggregation. This mechanismis independent of the stimulus forplatelet aggregation and represents thefinal and obligatory pathway to it.5
The platelet GP IIb/IIIa receptor antag-onists act by occupying the receptor,preventing fibrinogen binding and,thereby, platelet aggregation.
Three GP IIb/IIIa receptor antago-nists are currently available—abcix-imab, eptifibatide, and tirofiban. Ab-ciximab is a humanized murine anti-body that has a short plasma half-lifebut strong affinity for the receptor, re-sulting in some receptor occupancythat persists for weeks. Platelet aggre-gation gradually returns to normal 12to 24 hours following discontinuationof the drug. Abciximab is not specificfor GP IIb/IIIa, however, inhibiting thevitronectin receptor as well.
Eptifibatide is a cyclic heptapep-tide, and tirofiban is a nonpeptide.They have half-lives of 2 to 3 hoursand are highly specific for the GP IIb/IIIa integrin, with no effect on the vit-ronectin receptor. Platelet aggregationrapidly returns to normal in the hoursthat follow the discontinuation of thedrug.
These 3 GP IIb/IIIa receptor antag-onists have been compared with place-bo in 10 large trials involving morethan 32,000 patients who had ACSand who underwent PCI.41 In a meta-analysis incorporating all 10 trials, andfor which the end point was risk ofdeath or MI at 30 days, the GP IIb/IIIawas found to be uniformly better thanplacebo. For the GP IIb/IIIa receptorantagonists, the relative risk of deathor MI at 30 days was 0.79 (Figure 3),and this result was highly statisticallysignificant (P < 10-9).41 These studiessuggest that GP IIb/IIIa receptor antag-onists are useful in the treatment of
patients with UA, particularly those athigh risk.
Tirofiban as primary medical thera-py was studied in the Platelet ReceptorInhibition in Ischemic SyndromesManagement (PRISM)42 and PlateletReceptor Inhibition in Ischemic Syn-dromes Management in Patients Lim-ited by Unstable Signs and Symptoms(PRISM-PLUS)43 trials. The PRISM trialcompared tirofiban with heparin in3232 patients with UA/NSTEMI. Allpatients received ASA. Although theprimary composite outcome (rate ofdeath, MI, or refractory ischemia) wasreduced from 5.6% with UFH to 3.8%with tirofiban (relative risk, 0.67; P = .01) at the end of a 48-hour infu-sion period, the benefit did not persistat 30 days.
The PRISM-PLUS trial enrolled1915 patients with UA/NSTEMI. Pa-tients were randomized to receivetirofiban alone, UFH alone, or thecombination for 48 to 108 hours. Allpatients received ASA. The tirofiban-alone arm was dropped during the tri-al because of excess mortality. Theconclusion was that the clinical out-come was better for patients who re-ceived tirofiban and heparin than forpatients treated with heparin alone.
The incidence of the primary com-posite end point (death, MI, or refrac-tory ischemia at 7 days) was lower inpatients receiving tirofiban and hep-arin (12.9%) than in patients receivingheparin alone (17.9%; relative risk,0.68; P = .004). This benefit persisted at 30 days (P = .03) and at 6 months (P = .02). In addition, the end point reflecting incidence of death or non-fatal MI was reduced by 43% at 7 days (P = .006), 30% at 30 days (P = .03), and22% at 6 months (P = .06).
Eptifibatide as a primary medicaltherapy was studied in the Platelet
112 REVIEWS IN CARDIOVASCULAR MEDICINE FALL 2000
UNSTABLE ANGINA continued
Glycoprotein IIb/IIIa in Unstable Angi-na: Receptor Suppression Using Inte-grelin Therapy (PURSUIT)44 trial,which enrolled 10,948 patients withACS (without persistent ST-segment el-evation). All patients could receiveASA and heparin. The primary out-come of death or nonfatal MI at 30days was reduced from 15.7% to14.2% with eptifibatide (relative risk,0.91; P = .042). Within the first 96hours, a substantial treatment effectwas seen (9.1% versus 7.6%, P = .01),and this effect was sustained at 1 week(11.6% versus 10.1%, P = .02). Thehigh event rate reported in this trialwas due primarily to small elevationsin CK-MB concentration.
Abciximab has been studied mainlyin trials of percutaneous coronaryrevascularization. These are the Evalu-ation of 7E3 for the Prevention of Ischemic Complications (EPIC),45 Eval-
uation of PTCA to Improve Long-termOutcome with Abciximab GP IIb/IIIaReceptor Blockade (EPILOG),46 c7E3Fab Antiplatelet Therapy in UnstableRefractory Angina (CAPTURE),47 andEvaluation of Platelet IIb/IIIa Inhibitorfor Stenting (EPISTENT)48 trials. Its usehas consistently been associated withsignificant reductions in both the rateof MI and the need for urgent revascu-larization. PRISM, PRISM-PLUS, andthe Randomized Efficacy Study ofTirofiban for Outcomes and Restenosis(RESTORE)49 evaluated tirofiban. TheIntegrilin (eptifibatide) to MinimizePlatelet Aggregation and CoronaryThrombosis (IMPACT)50 and PURSUITtrials studied the use of eptifibatide.Overall, these trials showed a relativerisk of 0.79, compared with placebo(an incidence of death or MI at 30 daysof 9% with a GP IIb/IIIa receptor an-tagonist versus 11.1% for placebo,
P < 10-9).41
Unlike trials involving balloon an-gioplasty, the EPISTENT trial was de-signed to evaluate the efficacy of ab-ciximab as an adjunct to elective coro-nary stenting. The adjunctive use ofabciximab was associated with a sig-nificant reduction in the compositeclinical end point of death, MI, or ur-gent revascularization. The 30-day pri-mary end point occurred in 10.8% ofthe patients who received a stent andplacebo, in 5.3% of the patients whoreceived a stent and abciximab, and in 6.9% of patients who underwent balloon angioplasty (without stent placement) and received abciximab (P < .001).48 Most of the benefit fromabciximab was related to a reductionin the incidence of moderate to largeMI (defined by the presence of CK lev-els exceeding 5 times the upper limit ofnormal or a Q wave MI). At 1 year,
Figure 3. Meta-analysis of 10 randomized, placebo-controlled trials showing benefits of glycoprotein IIb/IIIa inhibitors in patients with unsta-ble angina and percutaneous catheter interventions.45 (PCI, percutaneous catheter intervention; ACS, acute coronary syndrome; MI, myocardialinfarction.)
IIb/IIIa Inhibitors in PCI and ACSDeath/MI at 30 days
stented patients who received abcix-imab had a lower mortality than didpatients who received stents withoutabciximab (0.8% versus 2.4%, repre-senting a 67% risk reduction; P = .01).
Treatment with a GP IIb/IIIa recep-tor antagonist increases the risk ofbleeding. Typically, such bleeding ismucocutaneous or involves the accesssite of vascular intervention. No trialhas shown an excess of intracranialbleeding with a GP IIb/IIIa inhibitor.ASA and heparin have been used withthe intravenous GP IIb/IIIa receptorantagonists in all trials.
Throughout the period of use of aGP IIb/IIIa receptor antagonist, hemo-globin concentration and plateletcount should be monitored and sur-veillance of patients for bleedingshould be carried out daily. Thrombo-cytopenia is an unusual complicationwith these agents. Severe thrombocy-topenia (platelet counts lower then50,000/mL) and profound thrombocy-topenia (platelet counts lower than20,000/mL) are seen in 0.2% and 0.5%of patients, respectively.
Long-term anticoagulation withwarfarin. The long-term administra-tion of warfarin has been evaluated inonly a few studies. Based on them, theevidence supporting the use of long-term anticoagulation combined withstandard ASA therapy in patients withACS is inconclusive.
The Antithrombotic Therapy inAcute Coronary Syndromes (ATACS)trial33 showed that there was a reduc-tion in the composite end point ofdeath, MI, and recurrent ischemiarates at 14 days with combinationtherapy that included ASA and war-farin compared with ASA alone (27%versus 10.5%, P = .04). The differencewas not significant after 12 weeks,however.
The OASIS pilot study51 showedthat low-intensity warfarin (fixed-dosetherapy at a dosage of 3 mg/d) had nobenefit, whereas the moderate-intensi-ty regimen (with warfarin dosage ad-justed to achieve an international nor-malized ratio of 2.0 to 2.5) reduced therisk of death, MI, or refractory anginaby 58% at 3 months (12.1% in thestandard-therapy group and 5.1% inthe warfarin-treated group, P = .08). Alarger OASIS trial,52 in which patientswere randomized either to a moderate-intensity regimen of warfarin or tostandard therapy (with ASA given toboth groups), did not show differencesin the rate of cardiovascular death, MI,or stroke after 5 months.
Use of anti-ischemic drugs. Rou-tine medical management should in-clude the use of nitrates, ß-blockers,and calcium channel blockers. Both
the Fourth International Study of In-farct Survival (ISIS-4)53 and the Grup-po Italiano per lo Studio della Soprav-vivenza nell’infarto Miocardico (GISSI-3)54 trials, involving patients with suspected acute MI, failed to show asignificant mortality benefit associatedwith nitrates. Use of nitroglycerin inUA, however, has been shown to re-duce incidence and severity of anginaand improve clinical stability. Anoverview of double-blinded, random-ized trials in which ß-blockers wereused in patients with threatening orevolving MI suggests that the risk ofprogression to acute MI is reduced byapproximately 13% with appropriateß-blockade.55
Calcium channel blockers may beused to control ongoing or recurringischemia-related symptoms in patientsalready receiving adequate doses of ni-
Main Points• Assays of cardiac-specific troponin T and cardiac-specific troponin I (cTnI)
provide equivalent diagnostic and prognostic information in most patientswith acute coronary syndrome. Measurement of cTnI levels is more reliablein patients with renal insufficiency.
• Patients with unstable angina (UA) or non–ST-segment elevation myocardialinfarction (NSTEMI) and high short-term risk of death or nonfatal MI arecandidates for aggressive therapy with ß-blockade, antithrombins, glyco-protein (GP) IIb/IIIa receptor antagonists, and aspirin.
• GP IIb/IIIa receptor antagonists are effective in reducing cardiac event ratesin the acute phase of medical management of UA/NSTEMI; this benefit ismagnified in patients who undergo percutaneous catheter interventions.
• Low molecular weight heparin may offer significant advantages over un-fractionated heparin (UFH); however, if immediate revascularization is an-ticipated, the use of UFH may be preferable.
• Following acute MI, angiotensin-converting enzyme inhibitors reduce mor-tality in patients with left ventricular dysfunction and are recommended inpatients with hypertension not controlled by ß-blockers and nitrates and inpatients with diabetes.
• Consider coronary artery bypass graft surgery specifically for the patientwith significant left main coronary artery disease (CAD) or for the patientwith multivessel CAD and depressed left ventricular function.
114 REVIEWS IN CARDIOVASCULAR MEDICINE FALL 2000
UNSTABLE ANGINA continued
trates and ß-blockers and in patientswith variant angina. Several meta-analyses incorporating data for all ofthe calcium channel blockers report-edly used in ACS, however, have indi-cated that there is no overall benefit inmortality reduction or MI prevention
associated with these agents.55,56 More-over, controlled trials suggest that ad-verse outcomes in patients with ACSare more frequent with the use ofrapid-release, short-acting dihydropy-rimidines when concurrent ß-block-ade is inadequate.57,58
ACEIs have been shown to reducemortality in patients with acute MIwith left ventricular dysfunction,59 indiabetic patients with left ventriculardysfunction following recent MI, andin a broad spectrum of patients withchronic CAD. Accordingly, ACEIsshould be used in such patients as wellas in patients with hypertension notcontrolled with ß-blockers and ni-trates. There is a trend toward the useof ACEIs and away from the use of cal-cium channel blockers in patients dis-charged from hospitals following MI.Postdischarge ASA use has increased to88%; ß-blocker use, to 65%; and ACEIuse, to 33%.60 Calcium channel block-er use has decreased to 18%.52
Risk StratificationAfter Initial StabilizationIt is important to differentiate betweenpatients who need prompt angiogra-phy without noninvasive testing andthose who can safely undergo furthernoninvasive testing for risk stratifica-tion. Prompt angiography withoutnoninvasive risk stratification is re-quired when medical treatment hasfailed to stabilize a patient’s condition.In general, noninvasive stress testingshould be done in patients who havebeen free of angina at rest, of anginaduring low-level activity, and of con-gestive heart failure for a minimum of24 to 48 hours and who are not other-wise considered to be at high risk.
Noninvasive test selection. Thechoice of stress test should be based onthe resting ECG, the ability to performexercise, and the local expertise andtechnologies available.61 Exercise stresstesting without imaging is appropriatefor the patient who is able to exerciseand whose resting ECG is normal. Forthe patient who is able to exercise andwhose resting ECG shows LBBB, left
Table 5Noninvasive Test Results Indicating
High-Risk Coronary Artery Disease62,63
Consider coronary angiography and possible revascularization when patients have any of these findings on noninvasive tests:
Exercise ECG testing
Angina during exercise
Duration of symptom-limited exercise < 6 METS
Exercise-induced ST-segment elevation except in lead aVR
Flat SBP response ≤ 130 mm Hg with progressive exercise
Horizontal or downsloping ST-segment depression (any of the followingsituations):
Involving multiple (≥ 5) leads
Lasting ≥ 6 min into recovery
Onset at heart rate ≤ 120 bpm
Onset at ≤ 6.5 METS
ST-segment depression ≥ 2 mm
Sustained decrease in SBP ≥ 10 mm Hg with progressive exercise
Sustained or symptomatic VT with exercise
Stress radionuclide myocardial perfusion imaging
Increased pulmonary radiotracer uptake after exercise
Large reversible anterior wall defect
Multiple reversible defects in 2 or more coronary artery regions
Transient LV dilatation immediately after exercise
Left ventricular imaging (stress radionuclide ventriculography)
Decrease in EF ≥ 10%
Exercise EF ≤ 50%
Resting EF ≤ 35%
METS, metabolic equivalents of tasks; SBP, systolic blood pressure; bpm, beats per minute; VT, ventriculartachycardia; LV, left ventricular; EF, ejection fraction.
FALL 2000 REVIEWS IN CARDIOVASCULAR MEDICINE 115
UNSTABLE ANGINA
ventricular hypertrophy, an intraven-tricular conduction delay, a preexcita-tion pattern, or a digoxin effect, con-sider exercise stress testing with imag-ing. Patients unable to exercise are candidates for pharmacologic stresstesting with imaging.
Patients with definite ACS shouldundergo echocardiography or radio-nuclide angiography to evaluate leftventricular function. Those whosenoninvasive test results predict a highrisk for adverse outcome should be re-ferred for coronary angiography andrevascularization (Table 5).62,63
It is important to perform stresstests at the appropriate time followingdiagnosis of UA/NSTEMI. Stress testingusing a standard protocol can be per-formed as soon as the patient’s clinicalcondition has stabilized and the pa-tient has been free of ischemic symp-toms for 24 to 48 hours. In a study of189 patients, symptom-limited exer-cise testing performed 3 to 7 days afteran episode of UA or non–Q wave MIcorrectly predicted those who woulddevelop adverse events during the firstmonth.64 This illustrates the impor-tance of early noninvasive testing forrisk stratification.
Selection of early managementstrategies. Several randomized trialsand observational studies have beencarried out to resolve the question ofwhether an early conservative ap-proach or an invasive approach is pre-ferred. The TIMI-IIIB,23 the Veterans Affairs Non–Q-Wave Infarction Strate-gies in Hospital (VANQWISH),65 andFRISC-II66 trials were randomized.
In the TIMI-IIIB trial, 15.5% of pa-tients who underwent an early inva-sive approach and 17.7% of thosetreated with an early conservative ap-proach reached the primary compositeend point of death, MI, or a positive
symptom-limited exercise test at 6weeks (P = .26). There was no differ-ence in the rate of death or MI be-tween the 2 treatment groups.
In the VANQWISH trial,65 the num-ber of patients who died or sustainednonfatal MI and the number who diedwere consistently higher in the groupmanaged with the invasive strategythan in the group managed conserva-tively. At hospital discharge, 36 pa-tients in the invasive-assessmentgroup died or sustained nonfatal MI,compared with 15 patients in the con-servative-care group (P = .004); 21 pa-tients in the invasively managed groupdied, compared with 6 in the conserv-atively managed group (P = .007). At 1month, 48 patients died or had non-fatal MI with invasive managementand 26 did so with conservative care (P = .012); there were 23 deaths in theinvasively managed group and 9 in the conservatively managed group (P = .021). At 1 year, 111 patients eval-uated invasively and 85 evaluated con-servatively died or had nonfatal MI (P = .05); 58 invasively managed and36 conservatively managed patientsdied (P = .025). These results could beexplained by the fact that the mortali-ty was higher (11.6%) among patientsundergoing CABG surgery in the earlyinvasive approach group than amongthe early conservative approach group(3.4%).
In the FRISC II trial,66 at 6 months,death or MI occurred in 9.4% of pa-tients assigned to the invasive strategyand in 12.1% of those assigned to thenoninvasive strategy (P = 0.03).58 Themortality associated with the invasivestrategy was 1.9%, compared with2.9% for the noninvasive strategy (P = .10). This showed that patientswho are not at high risk for revascular-ization and who receive 5 to 10 days of
treatment with dalteparin, ASA, ni-trates, and ß-blockers have a lowerlikelihood of MI or death with an ear-ly invasive approach than with thestandard conservative approach.
The TIMI-IIIB, VANQWISH, andFRISC-II trials were performed beforestents and GP IIb/IIIa receptor antago-nists were used. These data, therefore,need to be applied with caution today,when stents and GP IIb/IIIa receptorantagonists are available. In the OASISRegistry,67 there were no differences inthe rates of death or MI between coun-tries with the highest rates of invasiveprocedures (59%) and countries withlower rates (21%).
Among the patients randomized tothe conservative strategy in the TIMI-IIIB trial, factors independently pre-dicting failure of medical therapy werethe presence of reversible ST-segmentdeviation, a history of angina, prioruse of ASA or heparin, a family historyof premature CAD, and older age.68 Asthe number of risk factors increased, sodid the incidence of failure of medicaltherapy (from 32% in patients with 1risk factor to 88% in patients with 6risk factors). Such factors should beconsidered when selecting patients forexpedited angiography and revascular-ization.
To evaluate for revascularization, an early invasive strategy involvingprompt coronary angiography is rec-ommended for patients who have spe-cific signs and symptoms despite in-tensive anti-ischemic therapy. Suchsigns and symptoms include recurrentanginal ischemia, at rest or with low-level activities, and angina accompa-nied by evidence of congestive heartfailure, an S3 gallop, or new or worsen-ing mitral regurgitation.69 In addition,an early invasive strategy is advised inthe patient with high-risk findings on
116 REVIEWS IN CARDIOVASCULAR MEDICINE FALL 2000
UNSTABLE ANGINA continued
Initial risk stratification:Short-term risk of death or nonfatal cardiac ischemic events
Initial therapy
Hospital course
Management strategy decision point:Early invasive or noninvasive strategy
Course of action
Low risk
New-onset anginaNormal cardiac enzymesNormal or unchanged
noninvasive stress testing, depressedleft ventricular systolic function, he-modynamic instability, or sustainedventricular tachycardia. In the absenceof such findings, an early conservativeapproach appears acceptable.
Coronary RevascularizationPCI or CABG surgery is done to im-prove prognosis, relieve symptoms,prevent ischemic complications, andimprove functional capacity. Patientswith UA/NSTEMI and high-risk fea-tures will benefit from coronary revas-cularization. Such high-risk anatomicor disease-related features include leftmain CAD with greater than 50%stenosis, left main equivalent CAD, 3-vessel disease, CAD involving theproximal left anterior descending(LAD) coronary artery with 1- or 2-ves-sel disease, and 1- or 2-vessel diseasewithout proximal LAD disease butwith a large area of ischemia.
The indications for PCI and CABGsurgery in UA/NSTEMI are similar tothose in stable angina. In general,CABG surgery is clearly recommend-ed in patients with significant leftmain CAD and in patients with multi-vessel CAD and depressed left ventric-ular function.70
Postdischarge CareThe acute phase of UA/NSTEMI is usu-ally over within 2 months, after whichtime most patients resume a clinicalcourse similar to that of patients withchronic stable CAD. Medical therapyafter discharge from the hospitalshould include medications to prevent
future coronary events (ranging fromischemia to MI and death), controlischemic symptoms, and modify ma-jor risk factors to the extent possibleand practical.
Pharmacotherapy. Medications toprevent premature death and MI inthe patient who has had UA/NSTEMIare ASA, ß-blockers, ACEIs, and lipid-lowering agents, which should be usedin conjunction with dietary andlifestyle modifications. The ASA dos-age is 75 to 160 mg/d. ß-Blockade is in-dicated whether or not the patient hashad an MI. Lipid-lowering agents anddietary modification are appropriatefor any patient whose serum low-density lipoprotein (LDL) cholesterollevel exceeds 100 mg/dL. In particu-lar, 3-hydroxy-3-methylglutaryl coen-zyme A (HMG-CoA) reductase in-hibitors (statins) are recommended.ACEIs are especially important if thereis left ventricular dysfunction, the ejec-tion fraction in less than 40%, or thepatient is diabetic.
Patient education and support. Itis necessary to educate the patient re-garding the importance of lifestylechanges and awareness of general car-diac-related health issues. Such educa-tion is as much a part of comprehen-sive care as is pharmacotherapy.
Key risk-reduction measures aresmoking cessation, achievement ormaintenance of optimal weight (witha low-fat diet), aerobic exercise, andcontrol of hypertension and diabetesmellitus. Secondary prevention trialswith simvastatin and pravastatin haveamply demonstrated the profound im-pact of lowered LDL cholesterol levelson event-rate reduction, underscoringthe need to reach LDL cholesterol lev-els of less than 100 mg/dL.
The optimal time to initiate lipid-lowering measures after an episode of
ACS is unclear. A practical approachwould involve initiating lipid-lower-ing efforts as soon as possible, or theymay be forgotten completely.
UA/NSTEMI Management:Overview and SummaryThe history, physical examinationfindings, 12-lead ECG, and initialserum marker assays help differentiatepatients with UA into those with low,intermediate, or high short-term riskof death or nonfatal ischemic events(Figure 4).71 Patients with a low short-term risk may be observed in a facilitywhere cardiac monitoring is available,such as a chest pain unit. SubsequentECG and serum marker measurementsshould be obtained 4 to 8 hours afterthe first battery.
If the results of follow-up 12-leadECG and serum marker measurementsare normal, an exercise stress test orstress myocardial perfusion imagingmay be performed to provoke isch-emia. Patients with a negative stresstest result can be managed as outpa-tients. Patients with a high-risk stresstest result should be referred for coro-nary angiography and for revascular-ization.
Patients with UA/NSTEMI andhigh-risk features should be admittedto a coronary care unit and aggres-sively treated with standard medicaltherapy. This includes ASA, ß-blocker,antithrombin therapy, and a GP IIb/II-Ia inhibitor. After initial stabilization,such patients should undergo earlycoronary angiography and revascular-ization.
Patients with intermediate-risk fea-tures should be hospitalized and care-fully observed for complications, suchas sustained ventricular tachycardia orfibrillation, sinus tachycardia, atrialfibrillation or flutter, high-degree atrio-
Figure 4. Approach to the management ofpatients with unstable angina based onshort-term risk of death or nonfatal cardiacischemic events and on decision to pursueearly invasive or noninvasive strategy.
118 REVIEWS IN CARDIOVASCULAR MEDICINE FALL 2000
UNSTABLE ANGINA continued
ventricular block, sustained hypoten-sion, recurrent ischemia (documentedby symptoms or ST-segment change),any new mechanical defect (such as aventricular septal defect or mitral re-gurgitation), or pulmonary edema. Ifsuch complications develop, initialstabilization with aggressive medicaltherapy, followed by an early invasivestrategy, should be carried out.
Other intermediate-risk patientsmay undergo the initially conservativestrategy, which includes early, nonin-vasive risk stratification. Further deci-sion making must be based on the re-sults of noninvasive testing. The earlyuse of a reliable risk stratification strat-egy may help provide both medicallyoptimal and economically acceptablecare for patients. ■
References1. Detailed Diagnoses and Procedures: National Hospi-
tal Discharge Survey, 1996. Hyattsville, Md: Na-tional Center for Health Statistics; 1998. Ad-vance Data From Vital and Health Statistics, No.138.
2. Braunwald E. Unstable angina. A classification.Circulation. 1989;80:410-414.
3. Braunwald E, Mark DB, Jones RH, et al. UnstableAngina: Diagnosis and Management. Rockville,Md: Agency for Health Care Policy and Researchand the National Heart, Lung, and Blood Insti-tute, Public Health Service, US Dept of Healthand Human Services; 1994. AHCPR publication94-0602.
4. Ross R. Atherosclerosis—an inflammatory dis-ease. N Engl J Med. 1999;340:115-126.
6. Savonitto S, Ardissino D, Granger CB, et al. Prog-nostic value of the admission electrocardiogramin acute coronary syndromes. JAMA. 1999;281:707-713.
7. de Zwaan C, Bar FW, Janssen JH, et al. Angio-graphic and clinical characteristics of patientswith unstable angina showing an ECG patternindicating critical narrowing of the proximalLAD coronary artery. Am Heart J. 1989;117:657-665.
8. Slater DK, Hlatky MA, Mark DB, et al. Outcomein suspected acute myocardial infarction withnormal or minimally abnormal admission elec-trocardiographic findings. Am J Cardiol. 1987;60:766-770.
9. Cannon CP, McCabe CH, Stone PH, et al. Theelectrocardiogram predicts one-year outcome of
patients with unstable angina and non-Q wavemyocardial infarction: results of the TIMI III Reg-istry ECG Ancillary Study. Thrombolysis in Myo-cardial Ischemia. J Am Coll Cardiol. 1997;30: 133-140.
10. Adams JE III, Abendschein DR, Jaffe AS. Bio-chemical markers of myocardial injury. Is MBcreatine kinase the choice for the 1990s? Circu-lation. 1993;88:750-763.
11. Maynard SJ, Menown IB, Adgey AA. Troponin Tor troponin I as cardiac markers in ischaemicheart disease. Heart. 2000;83:371-373.
12. Ravkilde J, Horder M, Gerhardt W, et al. Diag-nostic performance and prognostic value ofserum troponin T in suspected acute myocardialinfarction. Scand J Clin Lab Invest. 1993;53:677-685.
13. Ohman EM, Armstrong PW, Christenson RH, etal. Cardiac troponin T levels for risk stratificationin acute myocardial ischemia. GUSTO IIA Inves-tigators. N Engl J Med. 1996;335:1333-1341.
14. Lindahl B, Andren B, Ohlsson J, et al. Risk strati-fication in unstable coronary artery disease. Ad-ditive value of troponin T determinations andpre-discharge exercise tests. FRISK Study Group.Eur Heart J. 1997;18:762-770.
15. Antman EM, Tanasijevic MJ, Thompson B, et al.Cardiac-specific troponin I levels to predict therisk of mortality in patients with acute coronarysyndromes. N Engl J Med. 1996;335:1342-1349.
16. Lindahl B, Venge P, Wallentin L. Relation be-tween troponin T and the risk of subsequent car-diac events in unstable coronary artery disease.The FRISC Study Group. Circulation. 1996;93:1651-1657.
17. Hamm CW, Heeschen C, Goldmann B, et al.Benefit of abciximab in patients with refractoryunstable angina in relation to serum troponin Tlevels. c7E3 Fab Antiplatelet Therapy in UnstableRefractory Angina (CAPTURE) Study Investiga-tors. N Engl J Med. 1999;340:1623-1629.
18. Lindahl B, Venge P, Wallentin L. Troponin Tidentifies patients with unstable coronary arterydisease who benefit from long-term antithrom-botic protection. Fragmin in Unstable CoronaryArtery Disease (FRISC) Study Group. J Am CollCardiol. 1997;29:43-48.
19. Becker RC, Cannon CP, Bovill EG, et al. Prognos-tic value of plasma fibrinogen concentration inpatients with unstable angina and non-Q-wavemyocardial infarction. Am J Cardiol. 1996;78:142-147.
20. Haverkate F, Thompson SG, Pyke SD, et al. Pro-duction of C-reactive protein and risk of coro-nary events in stable and unstable angina. Euro-pean Concerted Action on Thrombosis and Dis-abilities Angina Pectoris Study Group. Lancet.1997;349:462-466.
21. Biasucci LM, Vitelli A, Liuzzo G, et al. Elevatedlevels of interleukin-6 in unstable angina. Circu-lation. 1996;94:874-877.
22. Liuzzo G, Biasucci LM, Gallimore JR, et al. Theprognostic value of C-reactive protein and serumamyloid A protein in severe unstable angina. NEngl J Med. 1994;331:417-424.
23. Effects of tissue plasminogen activator and acomparison of early invasive and conservativestrategies in unstable angina and non-Q-wavemyocardial infarction. Results of the TIMI IIIBTrial. Thrombolysis in Myocardial Ischemia. Cir-
culation. 1994;89:1545-1556.24. Baigent C, Collins R. Appleby P, et al. ISIS-2: 10
year survival among patients with suspectedacute myocardial infarction in randomised com-parison of intravenous streptokinase, oral as-pirin, both, or neither. The ISIS-2 (Second Inter-national Study of Infarct Survival) CollaborativeGroup. BMJ. 1998;316:1337-1343.
25. Lewis HD Jr, Davis JW, Archibald DG, et al. Pro-tective effects of aspirin against acute myocardialinfarction and death in men with unstable angi-na. Results of a Veterans Administration Cooper-ative Study. N Engl J Med. 1983;309:396-403.
26. Cairns JA, Gent M, Singer J, et al. Aspirin, sulfin-pyrazone, or both in unstable angina. Results ofa Canadian multicenter trial. N Engl J Med. 1985;313:1369-1375.
27. Oler A, Whooley MA, Oler J, Grady D. Addingheparin to aspirin reduces the incidence of myo-cardial infarction and death in patients with un-stable angina. A meta-analysis. JAMA. 1996;276:811-815.
28. Balsano F, Rizzon P, Violi F, et al. Antiplatelettreatment with ticlopidine in unstable angina. Acontrolled multicenter clinical trial. The Studiodella Ticlopidina nell’Angina Instabile Group.Circulation. 1990;82:17-26.
29. CAPRIE Steering Committee. A randomised,blinded, trial of clopidogrel versus aspirin in pa-tients at risk of ischaemic events Lancet. 1996;348:1329-1339.
30. Bertrand ME, Rupprecht HJ, Urban P, et al. Dou-ble-blind study of the safety of clopidogrel withand without a loading dose in combination withaspirin compared with ticlopidine in combina-tion with aspirin after coronary stenting: theClopidogrel Aspirin Stent International Cooper-ative Study (CLASSICS). Circulation. 2000;102:625-629.
31. Wood AJ. Thrombotic thrombocytopenic purpu-ra associated with clopidogrel—a need for newapproaches to drug safety. N Engl J Med. 2000;342:1824-1826.
33. Cohen M, Adams PC, Parry G, et al. Combina-tion antithrombotic therapy in unstable restangina and non-Q-wave infarction in nonprioraspirin users. Primary end points analysis fromthe ATACS trial. Antithrombotic Therapy inAcute Coronary Syndromes Research Group. Cir-culation. 1994;89:81-88.
34. Gurfinkel EP, Manos EJ, Mejail RI, et al. Lowmolecular weight heparin versus regular heparinor aspirin in the treatment of unstable anginaand silent ischemia. J Am Coll Cardiol. 1995;26:313-318.
35. Fragmin During Instability in Coronary ArteryDisease (FRISC) Study Group. Low-molecular-weight heparin during instability in coronaryartery disease. Lancet. 1996;347:561-568.
36. Klein W, Buchwald A, Hillis SE, et al. Compari-son of low-molecular-weight heparin with un-fractionated heparin acutely and with placebofor 6 weeks in the management of unstablecoronary artery disease. Fragmin in UnstableCoronary Artery Disease study. Circulation.
FALL 2000 REVIEWS IN CARDIOVASCULAR MEDICINE 119
UNSTABLE ANGINA
1997;96:61-68.37. Comparison of two treatment durations (6 days
and 14 days) of a low molecular weight heparinwith a 6-day treatment of unfractionated hep-arin in the initial management of unstable angi-na or non-Q wave myocardial infarction:FRAX.I.S. (FRAXiparine in Ischaemic Syndrome).Eur Heart J. 1999;20:1553-1562.
38. Antman EM, Cohen M, Radley D, et al. Assess-ment of the treatment effect of enoxaparin forunstable angina/non-Q-wave myocardial infarc-tion: TIMI 11B-ESSENCE meta-analysis. Circula-tion. 1999;100:1602-1608.
39. Fragmin and Fast Revascularisation during In-Stability in Coronary Artery Disease Investiga-tors. Long-term low-molecular-mass heparin inunstable coronary-artery disease: FRISC IIprospective randomised multicentre study.Lancet. 1999;354:701-707.
40. Effects of recombinant hirudin (lepirudin) com-pared with heparin on death, myocardial infarc-tion, refractory angina, and revascularisationprocedures in patients with acute myocardial ischaemia without ST elevation: a randomisedtrial. Organisation to Assess Strategies for Isch-emic Syndromes (OASIS-2) Investigators. Lancet.1999; 353:429-438.
42. Platelet Receptor Inhibition in Ischemic Syn-drome Management (PRISM) Study Investiga-tors. A comparison of aspirin plus tirofiban withaspirin plus heparin for unstable angina. N EnglJ Med. 1998;338:1498-1505.
43. Platelet Receptor Inhibition in Ischemic Syn-drome Management in Patients Limited by Un-stable Signs and Symptoms (PRISM-PLUS) StudyInvestigators. Inhibition of the platelet glyco-protein IIb/IIIa receptor with tirofiban in unsta-ble angina and non-Q-wave myocardial infarc-tion. N Engl J Med. 1998;338:1488-1497.
44. The PURSUIT Investigators. Inhibition of plateletglycoprotein IIb/IIIa with eptifibatide in patientswith acute coronary syndromes. N Engl J Med.1998;339:436-443.
45. Topol EJ, Calif RM, Weisman HF, et al, on behalfof the EPIC Investigators. Randomised trial ofcoronary intervention with antibody againstplatelet IIb/IIIa integrin for reduction of clinicalrestenosis: results at six months. Lancet. 1994;343:881-886.
46. Lincoff AM, Tcheng JE, Califf RM, et al, for theEPILOG Investigators. Sustained suppression ofischemic complications of coronary interven-tion by platelet GP IIb/IIIa blockade with abcix-imab: one year outcome in the EPILOG trial. Cir-culation. 1999;99:1951-1958.
47. The CAPTURE Investigators. Randomised place-bo-controlled trial of abciximab before and dur-ing coronary intervention in refractory unstableangina: the CAPTURE study. Lancet. 1997;349:1429-1435.
48. The EPISTENT Investigators. Randomised place-bo-controlled and balloon-angioplasty-con-trolled trial to assess safety of coronary stentingwith use of platelet glycoprotein-IIb/IIIa block-ade. Evaluation of Platelet IIb/IIIa Inhibitor forStenting. Lancet. 1998;352:87-92.
49. The RESTORE Investigators. Effects of plateletglycoprotein IIb/IIIa blockade with tirofiban on
adverse cardiac events in patients with unstableangina or acute myocardial infarction undergo-ing coronary angioplasty. Circulation. 1997;96:1445-1453.
50. The IMPACT-II Investigators. Randomised place-bo-controlled trial of effect of eptifibatide oncomplications of percutaneous coronary inter-vention: IMPACT-II. Lancet. 1997;349-14221428.
51. Anand SS, Yusuf S, Pogue J, et al. Long-term oralanticoagulant therapy in patients with unstableangina or suspected non-Q-wave myocardial in-farction: organization to assess strategies for ischemic syndromes (OASIS) pilot study results.Circulation. 1998;98:1064-1070.
52. The Medical Research Council’s General PracticeResearch Framework. Thrombosis preventiontrial: randomised trial of low-intensity oral anti-coagulation with warfarin and low-dose aspirinin the primary prevention of ischaemic heart dis-ease in men at increased risk. Lancet. 1998;351:233-241.
53. ISIS-4 (Fourth International Study of Infarct Sur-vival) Collaborative Group. ISIS-4: a randomisedfactorial trial assessing early oral captopril, oralmononitrate, and intravenous magnesium sul-phate in 58,050 patients with suspected acutemyocardial infarction. Lancet. 1995;345:669-685.
54. Gruppo Italiano per lo Studio della Sopravviven-za nell’infarto Miocardico. GISSI-3: effects oflisinopril and transdermal glyceryl trinitratesingly and together on 6-week mortality andventricular function after acute myocardial in-farction. Lancet. 1994;343:1115-1122.
55. Yusuf S, Wittes J, Friedman L. Overview of resultsof randomized clinical trials in heart disease. II.Unstable angina, heart failure, primary preven-tion with aspirin, and risk factor modification.JAMA. 1988;260:2259-2263.
56. Held PH, Yusuf S, Furberg CD. Calcium channelblockers in acute myocardial infarction and un-stable angina: an overview. Br Med J. 1989;299:1187-1192.
57. Furberg CD, Psaty BM, Meyer JV. Nifedipine.Dose-related increase in mortality in patientswith coronary heart disease. Circulation. 1995;92:1326-1331.
58. Pepine CJ, Faich G, Makuch R. Verapamil use inpatients with cardiovascular disease: an overviewof randomized trials. Clin Cardiol. 1998;21:633-641.
59. ACE Inhibitor Myocardial Infarction Collabora-tive Group. Indications for ACE inhibitors in theearly treatment of acute myocardial infarction:systematic overview of individual data from100,000 patients in randomized trials. Circula-tion. 1998;97:2202-2212.
60. Kizer JR, Cannon CP, McCabe CH, et al. Trendsin the use of pharmacotherapies for acute my-ocardial infarction among physicians who de-sign and/or implement randomized trials versusphysicians in routine clinical practice: the MILIS-TIMI experience. Multicenter Investigation onLimitation of Infarct Size. Thrombolysis in Myo-cardial Infarction. Am Heart J. 1999;137:79-92.
61. Gibbons RJ, Balady GJ, Beasley JW, et al.ACC/AHA guidelines for exercise testing. A re-port of the American College of Cardiolo-gy/American Heart Association Task Force on
Practice Guidelines (Committee on Exercise Test-ing). J Am Coll Cardiol. 1997;30:260-311.
62. Chaitman BR. Exercise stress testing In: Braun-wald E, ed. Heart Disease: A Textbook of Cardiovas-cular Medicine. 5th ed. Philadelphia: WB Saun-ders Company; 1997:153-176.
64. Larsson H, Areskog M, Areskog NH, et al. Shouldthe exercise test (ET) be performed at dischargeor one month later after an episode of unstableangina or non-Q-wave myocardial infarction?Int J Card Imaging. 1991;7:7-14.
65. Boden WE, O’Rourke RA, Crawford MH, et al.Outcomes in patients with acute non-Q-wavemyocardial infarction randomly assigned to aninvasive as compared with a conservative man-agement strategy. Veterans Affairs Non-Q-WaveInfarction Strategies in Hospital (VANQWISH)Trial Investigators. N Engl J Med. 1998;338:1785-1792.
66. FRagmin and Fast Revascularisation During In-Stability in Coronary Artery Disease Investiga-tors. Invasive compared with non-invasive treat-ment in unstable coronary-artery disease: FRISCII prospective randomised multicentre study.Lancet. 1999;354:708-715.
67. Yusuf S, Flather M, Pogue J, et al. Variations be-tween countries in invasive cardiac proceduresand outcomes in patients with suspected unsta-ble angina or myocardial infarction without ini-tial ST elevation. OASIS (Organisation to AssessStrategies for Ischaemic Syndromes) Registry In-vestigators. Lancet. 1998;352:507-514.
68. Stone PH, Thompson B, Zaret BL, et al. Factor as-sociated with failure of medical therapy in pa-tients with unstable angina and non-Q wavemyocardial infarction. A TIMI-IIIB databasestudy. Eur Heart J. 1999;20:1084-1093.
69. Scanlon PJ, Faxon DP, Audet AM, et al. ACC/AHAguidelines for coronary angiography. A report ofthe American College of Cardiology/AmericanHeart Association Task Force on practice guide-lines (Committee on Coronary Angiography).Developed in collaboration with the Society forCardiac Angiography and Interventions. J AmColl Cardiol. 1999;33:1756-1824.
70. Eagle KA, Guyton RA, Davidoff R, et al. ACC/AHA guidelines for coronary artery bypass graftsurgery: a report of the American College of Car-diology/American Heart Association Task Forceon Practice Guidelines (Committee on CoronaryArtery Bypass Graft Surgery). J Am Coll Cardiol.1999;34:1262-1347.
71. Yeghiazarians Y, Braunstein JB, Askari A, StonePH. Unstable angina pectoris. N Engl J Med. 2000;342:101-114.
