Early Invasive Strategy in Unstable Coronary Artery Disease: Outcome in Relation to Risk Stratification

Full text

(1)

(2)

(3)

(4)

(5)

(6)

(7) . !

(8) "

(9) # $

(10)

(11)

(12)

(13)

(14) %&. '!( ! ')*+. $,$ -!. '!,$,! /$+! -! //$+$ 0110.

(15) Dissertation for the Degree of Doctor of Philosophy (Faculty of Medicine) in Cardiology presented at Uppsala University in 2002. ABSTRACT Diderholm, E. 2002. Early Invasive Strategy in Unstable Coronary Artery Disease-Outcome in Relation to Risk Stratification. Acta Universitatis Uppsaliensis. Comprehensive Summaries of Uppsala Dissertations from the Faculty of Medicine 1178. 72 pp. Uppsala ISBN 91-554-5387-2. In unstable coronary artery disease (CAD) it still is a matter of debate which patients should undergo early revascularisation. In the FRISC II study (n=2457) an early invasive strategy was, compared to a primarily non-invasive strategy, associated with reduced mortality and myocardial infarction (MI) rates. However, in this heterogeneous group of patients, tools for an appropriate selection to revascularisation are needed. From the FRISC II study we evaluated the prognosis, the angiographic extent of CAD and the effects of an early invasive strategy in relation to risk variables on admission. The occurrence of ST depression and/or elevated levels of Troponin T were associated with a higher risk for death and MI, more severe CAD and also with a reduction of death or MI by the early invasive strategy. Elevated levels of the inflammatory markers C-reactive protein (CRP) and interleukin-6 (Il-6) were associated with a higher mortality but an unchanged MI rate. Elevated levels of Il-6, but not CRP, identified patients with a large reduction of mortality by the invasive strategy. Age ≥ 70 years, male gender, diabetes, previous MI, ST depression and elevated levels of troponin and markers of inflammation were independently associated with an adverse outcome. The FRISCscore was constructed using these 7 variables. At FRISC-score ≥ 5 an early invasive strategy markedly reduced mortality and MI, at FRISC–score 3-4 death/MI was reduced, whereas in patients with a FRISC-score 0-2 neither mortality nor death/MI was influenced. In unstable CAD, a non-invasive strategy seems justified only for patients at low risk, i.e. FRISC score ≤ 2. In patients with intermediate and high risk, i.e. FRISC-score ≥ 3, an early invasive strategy is recommended. Key words: Unstable angina, non-ST elevation myocardial infarction, risk stratification, coronary angiography, revascularisation. Erik Diderholm, Department of Medical Sciences, Cardiology, University Hospital, SE-751 85 Uppsala, Sweden © Erik Diderholm 2002 ISSN 0282-7476 ISBN 91-554-5387-2 Printed in Sweden by Tryck & Medier, Uppsala 2002 -2-.

(16) Papers This thesis is based on the following original papers, which will be referred to by their Roman numerals.. I.. ST depression in ECG at entry indicates severe coronary lesions and large benefits of an early invasive treatment strategy in unstable coronary artery disease-the FRISC II ECG substudy Erik Diderholm, Bertil Andrén, Gunnar Frostfeldt, Margareta Genberg, Thomas Jernberg, Bo Lagerqvist, Bertil Lindahl, Lars Wallentin Eur Heart J 2002, Jan; 23: 41-9. II.. The prognostic and therapeutic implications of increased Troponin T levels and ST depression in unstable coronary artery disease-the FRISC II invasive Troponin TECG substudy Erik Diderholm, Bertil Andrén, Gunnar Frostfeldt, Margareta Genberg, Tomas Jernberg, Bo Lagerqvist, Bertil Lindahl, Per Venge, Lars Wallentin Am Heart J 2002, May; 143 (5): 760-7. III.. Inflammatory activity and the effect of an early invasive strategy in unstable coronary artery disease-the FRISC II invasive CRP-fibrinogen substudy Erik Diderholm, Bo Lagerqvist, Bertil Lindahl, Henrik Toss, Agneta Siegbahn, Lars Wallentin Submitted. IV.. Relationship between interleukin-6 and mortality in patients with unstable coronary artery disease: effects of an early invasive or noninvasive strategy Eva Lindmark, Erik Diderholm, Lars Wallentin , Agneta Siegbahn JAMA 2001 Nov; 286:2107-13. V.. The FRISC score for selection of patients for an early invasive treatment strategy in unstable coronary artery disease. Bo Lagerqvist, Erik Diderholm, Bertil Lindahl, Sten Husted, Frederic Kontny, Elisabeth Ståhle, Eva Swahn, Per Venge, Agneta Siegbahn, Lars Wallentin Submitted. Reprints were made with permissions from the publishers. -3-.

(17) Contents Abbreviations Introduction Background Pathogenesis of coronary artery disease (CAD) Clinical manifestations of the acute coronary syndrome (ACS) Prognostic implications of ECG Prognostic implications of biochemical markers of myocardial damage Prognostic implications of biochemical markers of inflammation Risk score in UCAD Coronary angiography Medical treatment of UCAD Early revascularisation in UCAD Aims of the study Material and methods Study population Definition of endpoints ECG Laboratory analyses Coronary angiography Construction of the FRISC-score Statistical methods Results Main FRISC II study Prognosis in relation to risk variables Coronary artery lesions in relation to risk variables Effects of an early invasive strategy in relation to risk variables Discussion Early invasive strategy-randomised studies Prognosis in relation to risk variables Coronary artery lesions in relation to risk variables Mechanisms for adverse prognosis Effects of an early invasive strategy in relation to risk variables Limitations of the study Summary and conclusion Acknowledgements References. -4-. 5 6 6 6 7 9 10 11 13 13 14 14 16 17 17 19 20 20 21 22 22 23 23 24 30 32 37 37 41 45 47 48 53 55 56 58.

(18) Abbreviations ACS AHCPR CABG CAD CI CRP CV death/MI ECG FRISC FRISC II (invasive) FRISC II (medical) GP GUSTO IHD Il-6 LAD LDL LMWH LV MI NNT NSTEMI PTCA RR ST STEMI TACTICS 3-VD/LMD TIMI TnI TnT TPA UA UFH UCAD VANQWISH VINO. acute coronary syndrome Agency for Health Care Policy and Research coronary artery bypass grafting coronary artery disease confidence interval C-reactive protein coefficient of variation the composite of death and MI electrocardiography, electrocardiographic Fragmin during InStability in Coronary artery disease Fast Revascularisation during InStability in Coronary artery disease Fragmin during InStability in Coronary artery disease glycoprotein The Global Use of Strategies to Open Occluded Coronary Arteries ischemic heart disease Interleukin-6 left anterior descending artery low density lipoprotein low molecular weight heparin left ventricular myocardial infartion numbers needed to treat (to prevent one event) non ST elevation myocardial infarction percutaneous transluminal coronary angioplasty risk ratio ST segment ST elevation myocardial infarction Treat Angina with Aggrastat and Determine Cost of Therapy with an Invasive or Conservative Strategy three vessel or left main disease Thrombolysis In Myocardial Infarction Troponin I Troponin T tissue plasminogen activator unstable angina unfractionated heparin unstable coronary artery disease (UA or NSTEMI) Veterans Affairs Non-Q-Wave Infarction Strategies in Hospital V_alue of First Day Angiography/Angioplasty I_n Evolving N_on-ST Segment Elevation Myocardial Infarction: An O_pen Multicenter Randomized Trial.. -5-.

(19) Introduction Ischemic heart disease (IHD) is still. thrombolysis. 4. or. percutaneous. one of the major causes of mortality in the. transluminal coronary angioplasty. Western world 1. Approximately 20% of. (PTCA) 5. If no ST elevation is present the. medical emergencies are patients. recommended course of therapy is anti-. presenting with chest pain or other. thrombotic and anti-ischemic.. symptoms of IHD 2. Most of them are. One of the major issues of the 90s has. admitted to the coronary care unit. been whether or not early revascularisation. presumed to have an acute coronary. should be performed on this latter group. syndrome (ACS), as later confirmed in. of patients, considering the benefit of. about 40% of the cases 3. The ECG on. improved myocardial blood supply versus. admission determines the initial course of. the risks and costs of the procedure. Hence. ACS therapy. Patients presenting ST. the need to identify readily available risk. segment (ST) elevation immediately. indicators as a means to select the patients. receive reperfusion treatment i.e.. who benefit most from such procedures.. Background Pathogenesis of coronary artery disease. flow and factors like smoking, raised lowdensity lipoprotein (LDL) levels,. Atherosclerosis, the dominant cause. hypertension, diabetes or genetic. of coronary artery disease (CAD), is. mechanisms. The endothelium produces. primary located in the intima of the vessel. endocrine substances, which actively. wall. A continuous accumulation of lipids. participate in the prevention of thrombosis. and an associated inflammatory process. and regulation of vascular muscle tone 7.. form the atherosclerotic plaque, which. Along with diminished or lost endothelial. eventually progresses to luminal narrowing. function, inflammatory cells and lipids can. 6. A healthy endothelium functions as a. migrate into the intima forming foam cells.. selective lipoprotein permeable barrier, but. These mainly consist of LDL particles. dysfunction of the endothelium can be. phagocytosed by monocyte derived. prompted by shear-stress, turbulent blood. macrophages, which secrete cytokines and -6-.

(20) growth regulating molecules 6. Smooth. 16. Plaque rupture exposes prothrombotic. muscle cells proliferate and produce. material, e.g. collagen and tissue factor, to. extracellular matrix proteins, mainly. circulating platelets and coagulation. collagen, which form the fibrous cap. factors, which initiate thrombus formation.. separating the atheroma from the lumen.. The clinical consequences vary with a. Thus, the plaque is made up of two main. number of factors, notably such as size of. components, a lipid-rich core and the. the involved coronary artery, size of. fibrous cap 8-10.. jeopardised myocardium, availability of. Atherosclerotic plaque formation is a. preformed collaterals, size and depth of. slow and insidious process, which can. coronary lesion, local prothrombotic-. proceed many years without yielding. fibrinolytic. symptoms. In the first stages of the. embolisation and vasoconstriction 12,13,17.. atherosclerotic process, the lumen of the. The degree of local or systemic. vessel will be maintained due to expansion. inflammation also seems to play a key role. of the arterial wall – remodelling 11. The. in the clinical development 18,19.. balance,. downstream. growing plaque eventually protrudes into the vessel lumen, limiting flow and. Clinical manifestations of the acute. disrupting the balance between myocardial. coronary syndrome. perfusion supply and demand 9, with possible onset of stable angina pectoris,. Symptomatic IHD can be divided in. typically noticed first on exertion. An ACS. stable angina pectoris and ACS. The latter. may occur at any time during the. is further divided in unstable angina (UA),. atherosclerotic process, regardless of. non-ST elevation myocardial infarction. stable angina. It usually is caused by rupture. (NSTEMI), ST elevation myocardial. or erosion of the plaque, causing thrombus. infarction (STEMI) and sudden death.. formation with abrupt narrowing or occlusion of the vessel 12,13.. Unstable angina:. The morphologic characteristics of. UA, first described in 1923 by Waern. the plaque are of vital importance regarding. as a precursor to myocardial infarction. the risk for plaque rupture. Thus, plaques. (MI) 20. Other names used are preinfarction. with a lipid rich soft core, a thin capsule. angina, status anginosus or crescendo. and signs of inflammatory activity, as. angina. Unstable angina is defined as new. indicated by increased macrophage. onset of severe angina, impairment of. infiltration, are more likely to rupture 14-. stable angina, angina at rest or angina after -7-.

(21) MI. The most common underlying cause is. is limited because the occlusion is. platelet aggregation and coagulation with. transient and/or the myocardium protected. thrombus formation at a disrupted plaque.. by pre-formed collaterals 23,24.. The thrombotic process leads to severe. The only observed distinction between. stenosis, transient occlusion, downstream. NSTEMI and UA is based on elevated levels. embolisation and vasoconstriction and. of biochemical markers. The timing of. thereby episodes of sudden decrease of. blood samples, analytical methods used and. blood flow, causing ischemia in the. their cut-off limits vary between different. myocardium 9,12.. centres making the separation between. The initial clinical diagnosis of UA is. these two entities diffuse. Previously the. usually based on a history of chest pain at. term minimal myocardial damage, MMD,. rest or light exertion accompanied by ECG. was often used to indicate a condition in. changes suggesting ischaemia. However,. the border zone between UA and NSTEMI. the ECG findings are unspecific since as. 25. However, based on an ESC/AHA. well T wave abnormalities as to some extent. consensus document from 2000, all. ST deviation can be caused by other reasons. conditions with myocardial ischemia. than myocardial ischemia 21. Even non-. associated with elevation of biochemical. cardiac reasons as tachycardia and anaemia. markers of myocardial necrosis now fulfil. might contribute to unstable angina, as. the definition of MI 26.. noted in the Braunwald UA classification. UA and NSTEMI are similar with. where the condition is divided in different. respect to pathogenesis, morphological. groups depending on the occurrence of. findings and prognosis and accordingly are. chest pain at rest, the pathogenesis and the. treated alike 27. In clinical trials these two. need of treatment 22.. diagnoses often have been merged into one, unstable coronary artery disease. Non ST elevation myocardial infarction. (UCAD) or UA/NSTEMI.. Prolonged myocardial ischaemia, ST elevation MI. subsequent permanent myocardial damage and elevated levels of biochemical markers. The underlying reason for STEMI is. of myocardial necrosis mainly distinguish. typically an abruptly occurring occlusive. NSTEMI from UA. In most cases an. thrombus without adequate collateral. occlusive thrombus or downstream. supply to the area at risk 28. The symptoms. embolisation in a coronary artery occurs.. are often more severe compared to UCAD.. However, the damage of the myocardium. In many cases, the patient presents with -8-.

(22) sudden onset of chest pain, cold sweat,. have been a prerequisite for inclusion. In. nausea and circulatory instability. The. such trials the 6 months rate of death has. myocardial damage will be greater and thus,. varied between 5-7% whereas death/MI has. the left ventricular (LV) function more. occurred in 13-18% 30-32. Even within. reduced. A large MI, as indicated by. these more defined patient groups, the. pathological Q-waves on the ECG, will. prognosis will still vary with respect to. develop if the occlusion persists over the. different clinical variables. Over the years. next few hours. In some cases, the ST. a large number of individual and. elevation is neither followed by Q-waves. combination of risk indicators have been. nor extensive myocardial damage, due to. identified and are presented below.. rapidly re-established blood flow, either Prognostic implications of ECG. spontaneously or by reperfusion treatment.. In UCAD, ST segment deviation is the. Sudden death Electrical instability with the. most specific ECG finding indicating. development of ventricular arrhythmia is. myocardial ischemia. Traditionally, 0.1 mV. the most common cause of circulatory. of ST depression has been the standard cut-. arrest causing sudden death in the ACS 29.. off level for detection of significant. The diagnosis of UA/MI still must be based. ischemia. This has consistently been. on patient history, biochemical markers of. associated with an increased risk of. myocardial damage and ECG readings. If. subsequent death/MI in previous trials 33-. the patient dies suddenly because of. 38. In addition, several trials have found a. ischemia induced arrhythmia based on. relation between the magnitude of ST. plaque rupture, the above diagnoses can not. depression and an increased risk of future. be made. Sudden death without any other. complications 38-40. Even a slighter ST. explanation usually is presumed to be an. segment depression (0.05-0.1 mV) seems. ACS.. to indicate worse outcome 40-42. A minor ST elevation, not fulfilling criteria for. Prognosis in unstable coronary artery. reperfusion therapy but indicative of. disease. UCAD, is also associated with an elevated. The. UCAD. population. risk of future death/MI 35,42.. is. heterogeneous and the prognosis varies considerably in different studies. In most. Isolated T-wave inversion may also be. clinical trials objective signs of ischemia. an indicator of myocardial ischemia. This -9-.

(23) finding is less specific and associated with. diagnostic accuracy of measurements of. considerably lower future risk compared. levels of the biochemical markers of. to ST depression 38,40-42. In fact, there is. myocardial ischemia has increased. controversial information whether isolated. substantially due to the development of new. T-wave inversion is associated with more. markers with improved sensitivity and. future events when compared to no STT. specificity.. changes. In some studies patients with isolated T-wave inversion run a higher risk. CK / CK-MB. 35,40 whereas no difference was seen in. Creatine kinase (CK) is an enzyme. other trials 38,41. Although the risk with. involved in the formation of ATP. The. isolated T-wave inversion might be. introduction of this marker, compared to. unimpressive in large patient materials, if. the previously used ASAT, markedly. located in anterior leads, it might indicate. improved the ability to accurately confirm. the presence of a severe proximal stenosis. or exclude MI. However, CK consists of. of the left anterior descending artery. three isoforms with different tissue. (LAD) 43,44.. distributions, mainly in skeletal and heart muscle. The most heart specific of these. Prognostic implications of biochemical. isoforms is CK-MB 47.. markers of myocardial damage. CK-MB mass measurement has been shown to be more stable than enzyme. Severe myocardial ischemia lasting. activity and also appears to be more. more than 15-20 minutes leads to. sensitive 48. Even in patients without. disruption of the cell membrane and cell. standard criteria for MI, slightly elevated. death. The intracellular constituents then. levels of CK-MB mass has been associated. diffuse into the interstitium and, from. with an increased risk for subsequent. there, to the circulating blood where they. cardiac death 49. However, CK-MB might. can be measured 45. Some of these markers. also be elevated at damage of skeletal. are rapidly released due to a high. muscle and strenuous exercise 50.. concentration in the cytoplasm whereas others are more structurally bound and. The troponins. released more slowly 46. The molecular. The troponin complex is the main. size of the marker and its half-life in plasma. regulatory protein of the thin filament of. are other factors affecting the time-window. the cardiac myofibril and regulates the. for its detection. Over the years the. actin-myosin interaction that causes -10-.

(24) contraction. It consists of three subunits:. new events after ACS 55. An advantage with. troponin T (TnT), which binds to. TnI compared to TnT might be a lower rate. tropomyosin; troponin I (TnI), which. of false positive results in patients with. inhibits the actin-myosin interaction; and. renal failure 56.. troponin C, which binds calcium. TnT and. Because of the improved sensitivity. TnI exist in three separate isoforms located. and specificity in the troponins, a joint ESC/. in cardiac and slow and fast skeletal muscle. ACC committee proposed elevated levels. 51. The current assays only measure the. of TnT or TnI as the principal marker for. cardiac isoform of TnT and TnI.. the diagnosis of MI. An increased level for troponin was defined as a measurement. Troponin T exists both unbound in a. exceeding the 99th percentile of a reference. cytosolic pool and is structurally bound to. control group provided that the coefficient. the myofibrils. This explains both the early. of variation (CV) at this level is ≤ 10% 26.. rise and the wide time window for detection of troponin elevation after myocardial. Prognostic implications of biochemical. damage 52, although the plasma half-life is. markers of inflammation. only two hours. Thus, TnT can be used for early (2-3 hours) as well as late (10-14. Increased inflammatory activity,. days) diagnosis of MI. In accordance, it is. caused by e.g. trauma, infection or. seldom used for the diagnosis of early re-. inflammatory diseases, induces the release. infarction where CK-MB is more useful.. of cytokines and growth factors from. The first assays of TnT suggested a cut-. several cell types e.g. monocytes,. off level of 0.1 µg/L for detection of. lymphocytes and smooth muscle cells 57.. myocardial damage, which has been used. The cytokine interleukin-6 (Il-6) is the. in the majority of clinical trials. Among. major stimulator of acute phase proteins. patients without MI, based on CK-MB. synthesised in the liver, such as C-reactive. criteria, those with elevated TnT levels had. protein (CRP), fibrinogen, plasminogen. a higher rate of future death/MI 53,54. This. and plasminogen activator inhibitor -1 (PAI-. reflects a higher sensitivity for TnT than. 1) 58. Inflammatory processes are also. CK-MB to detect myocardial damage.. involved in atherosclerosis. In fact, Troponin I has similar properties to. atherosclerosis has pathogenetical and. TnT. It seems to be as useful as TnT to. histological findings similar to classical. identify patients with increased risk for. inflammatory -11-. disorders. such. as.

(25) rheumatoid arthritis, pulmonary fibrosis,. understood although it seems to be involved. chronic. in the activation of the early components. pancreatitis. and. glomerulosclerosis 10.. of the complement cascade leading to. Large follow-up studies of healthy. phagocytos 68. CRP has also been shown. subjects indicate a higher risk for future. to induce tissue factor expression in. cardiac events if there are signs of a raised. monocytes 69. Therefore CRP might be. systemic inflammatory activity, such as. directly. increased CRP, fibrinogen or Il-6 levels 59-. pathophysiology of the thrombotic. 61. There is also evidence that plaques with. component of atherosclerosis.. involved. also. in. the. a higher amount of inflammatory cells are more likely to rupture and cause an ACS. Fibrinogen. 15. In UCAD, patients presenting with. Fibrinogen is also an acute phase. increased levels of the inflammatory. reactant produced in the liver. There is a. markers CRP or fibrinogen have a poorer. strong correlation between increased. prognosis. This might partly be a result of. fibrinogen levels and well-established risk. concomitant myocardial damage, which. factors for atherosclerosis e.g. increased. will affect both the prognosis and the. age,. inflammatory response 62,63. However,. cholesterolemia, diabetes and hypertension. even in selected UCAD patients without. 70-73. Still, fibrinogen is an independent. myocardial damage, raised levels of. cardiovascular risk factor both in healthy. inflammatory markers are associated with. subjects 60 and in patients with ACS 74. This. more subsequent events 63-66.. has. smoking,. supported. obesity,. the. hyper-. theory. that. arteriosclerosis partly is an inflammatory CRP. disease. CRP was discovered in patients with. In addition, fibrinogen is involved in. acute pneumococcal pneumonia. As it. the thrombotic process, both in platelet. reacted with the C-polysaccharide of the. aggregation and the coagulation cascade.. pneumococcal cell wall, it was named C-. Thus, in ACS, an increased level of. reactive protein. CRP is an acute phase. fibrinogen is not only an indicator of. reactant and is widely used for measuring. inflammation but also directly involved in. the activity of inflammatory and infectious. the thrombotic process.. disorders 67. The major stimulus to CRP secretion from the liver is Il-6. The. Il-6. biological function of CRP is not fully. The cytokine Il-6 is mainly produced -12-.

(26) by activated macrophages. Il-6 is the only. increasing interest in devising different. substance known to induce synthesis of all. scoring systems based on several of these. the acute phase reactants by the liver.. risk indicators. These scores contribute to. However,. direct. further improve risk stratification 77,78 and. proinflammatory and anti-inflammatory. can become instrumental in the choice of. effects 58. Il-6 has been suggested to. treatment 79,80. Yet, prior to the present. influence the atherosclerotic process by. study there was limited knowledge. decreasing lipoprotein lipase activity,. concerning relations between both. which increases macrophage uptake of. individual risk factors and a risk score and. lipids. In addition, Il-6 enhances the activity. the effect of an early invasive approach in. of the hypothalamic-pituitary-adrenal axis,. UCAD.. Il-6. also. has. which is associated with obesity, hypertension and insulin resistance 75.. Coronary angiography. In healthy men, elevated Il-6 levels have been associated with increased risk for. Coronary angiography still remains the. MI 61. In ACS raised Il-6 levels indicate. golden standard for establishing presence,. higher risk of death and MI 76. As Il-6, CRP. site and severity of CAD. Furthermore, all. and fibrinogen all reflect inflammatory. coronary revascularisation procedures are. activity, their individual importance for the. based on the results of the coronary. development of the atherosclerotic process. angiography. The severity of the coronary. and its complications is difficult to assess.. lesions can be evaluated by the number of vessels diseased, the amount of. Risk score in unstable coronary artery. myocardium in jeopardy, the occurrences. disease. of complex lesions and thrombus and estimation of the coronary blood flow.. A large number of studies have, by using univariable analyses, identified. These parameters correlate to an adverse prognosis 81,82.. numerous risk indicators for future events. In UCAD the lesion morphology. in UCAD. Most of these studies have also. seems to be the most important risk. confirmed the independent contribution of. indicator 83,84. The culprit lesion is in. these risk indicators in multivariable. typical cases characterised by ulceration,. analysis. However, the risk in patients with. haziness, irregular borders, overhanging. different combinations of risk variables is. edges and/or multiple irregularities. hard to assess. This has lead to an. (complex morphology) 85-88. A thrombus -13-.

(27) Addition of the platelet inhibitor. might be observed, but is better visualised with angioscopy 89.. clopidogrel further reduces the risk of. Despite a fundamental similarity in. subsequent events 98. In patients treated. the angiographic findings between UA and. with PTCA, platelet inhibition by blocking. NSTEMI there are differences. The. the GP IIb/IIIa receptor reduces procedure. occurrence of total occlusion is around 20-. related event rates 99,100. However, in. 30% in NSTEMI and around 10% in. patients managed without early PTCA, the. unstable angina 23 . Furthermore, if. additional effects of GP IIb/IIIa inhibition. occlusion is present, the occurrence of. are limited 32,101,102.. collaterals is more common in UA 90.. Anti-ischemic treatment with beta-. Nevertheless, coronary angiography. blockers 103 or calcium-antagonists 104,105. has limitations. Although considerably. are associated with a reduction of cardiac. decreased, a potential periprocedural risk. events after MI. The beneficial effect of. still exists, especially if left main disease. the lipid lowering statins has been. is present. From a registry study of almost. confirmed in both primary prevention 106. a quarter of a million patients 1984-87,. and in patients with confirmed IHD 107,108.. death occurred in 0.1%, MI in 0.06% and. Also in the setting of ACS statins seem to. stroke in 0.07% 91.. reduce event rates 109,110.. Medical treatment of unstable coronary. Early revascularisation in unstable. artery disease. coronary artery disease. Anti-thrombotic and anti-ischemic. When medical treatment has failed to. drugs are the cornerstones in the treatment. alleviate symptoms of ischemia in chronic. of UCAD. Studies in the nineteen-eighties. stable angina, revascularisation with either. showed that platelet inhibition by aspirin. coronary artery bypass grafting (CABG) or. and. by. PTCA is recommended. This treatment. unfractionated heparin (UFH) reduces the. gives excellent symptom relief 111,112 and. risk of death/MI and episodes of recurrent. improves survival in high-risk patients. The. ischemia 92-94 . Over the last years. absolute survival benefits with CABG is. subcutaneous low molecular weight. most evident in patients with left main. heparins (LMWH) have been found more. disease, three vessel disease, two vessel. effective and convenient than UFH, which. disease including stenosis of the proximal. therefore has been replaced 95-97.. part of the LAD, reduced LV function or. coagulation. inhibition. -14-.

(28) signs of severe ischemia 113. In terms of symptom relief, the same strategy also applies in UCAD 114 although there is a greater complication risk, especially at CABG 115. Previously only one large-scale study, TIMI IIIB, has compared early invasive with non-invasive strategy in UCAD and found no reduction of death or MI by means of invasive strategy. Despite the lack of evidence for any benefits in survival or reduction of other major cardiac events by early revascularisation, the number of patients managed with such procedures gradually has increased over the years 116.. -15-.

(29) Aims of the study In patients with unstable coronary artery disease to evaluate. 1. the prognosis in primarily non-invasively managed patients 2. the angiographic extent of coronary artery disease 3. the effects of an early invasive strategy. in relation to. a. signs of ischemia in ECG on admission (paper I) b. signs of myocardial damage (paper II) c. signs of inflammation (paper III-IV) d. a multivariable risk score (paper V). -16-.

(30) Material and methods Study population: patients with unstable coronary artery disease. primarily non-invasive with an early. (paper I-V). invasive strategy in patients with UCAD117. Fifty-eight hospitals in Sweden, Norway. The FRISC II invasive study (Fast. and Denmark took part in the trial. Between. Revascularisation during InStability in. May 1996 and May 1998 2457 patients. Coronary artery disease) was an open,. were included in the study. An overview of. randomised, controlled trial comparing a. the study is shown in fig 1.. Main FRISC II invasive study. 0 days. 0-2 days. Inclusion criteria: ST depression or T-wave inversion or elevated enzymes/troponins. Admission ECG. Concomitant treatment Aspirin. Beta-blockade Dalteparin 120 IU/kg/12h 5-7 days Blood samples: Troponin T Troponin I CRP Fibrinogen Interleukin-6. Randomisation. 0-3 days. Noninvasive* n = 1235. 5-7 days. Paper I-V. Invasive* n = 1222. Exercise test. Coronary angiography/ PTCA-CABG. Severe ischemia, recurrent angina or myocardial infarction. Coronary angiography/ PTCA-CABG. 6 & 12 months. Events. Events. Figure 1. Design of the FRISC II invasive study. * Dalteparin or placebo 3 months.. -17-.

(31) Inclusion and exclusion criteria Postmenopausal women and men. females < 80 kg and males < 70 kg and 7500 U bid at higher body weight.. above 40 years admitted to hospital because. In the FRISC II medical trial 118 three. of increasing or rest symptoms of cardiac. months treatment with dalteparin was. ischemia were eligible for the trial. An. compared to placebo. The study included. objective sign of ischemia had to be. 1032 additional patients who were not. verified either as 0.1 mV of ST depression. randomised in the FRISC II invasive study .. or T wave inversion or raised levels of. These patients are included in paper IV but. biochemical markers of myocardial. not in the present report.. damage. Informed consent was mandatory. Major exclusion criteria were. Concomitant treatment. increased risk of bleeding (cerebrovascular. As soon as possible after admission. event within 12 months, trombocytopenia. the patients were treated with aspirin, beta-. < 100x 109, uncontrolled hypertension,. blockade and as needed organic nitrates and. ongoing oral anticoagulant treatment,. calcium-antagonist. The use of ACE-. anemia), Creatinin > 150µmol/L, INR >. inhibitors at symptoms or signs of LV. 1.4, previous open heart surgery, PTCA last. dysfunction and statins at cholesterol levels. six months, osteoporosis and other. above 5.0 mmol/L or LDL-cholesterol. concomitant significant heart disease.. above 3.5 mmol/L was encouraged. In patients with diabetes, dietary advice, oral. Randomised medical treatment From admission, or at the latest from randomisation, all patients were treated. sylfonylurea or subcutaneous insulin was recommended, aiming for a HbA1c level ≤ 6.5 mmol/L.. with the LMWH dalteparin (120 U/kg bid, maximal 10000 U bid) for at least five days. Randomised invasive and non-invasive. and always until the revascularisation. strategies. procedure or the predischarge exercise. The direct invasive strategy required. test. Within 48 hours after admission they. coronary angiography within a few days of. were randomised by a two by two factorial. enrolment aiming for revascularisation. design to either an early non-invasive or an. within 7 days from start of open label. invasive strategy and to 3 months continued. dalteparin. Revascularisation was. treatment with subcutaneous dalteparin or. recommended in all patients with a ≥ 70%. placebo. The long term dalteparin. diameter obstruction in any artery. (Fragmin®) dosage was 5000 U bid for. supplying a significant proportion of the -18-.

(32) myocardium. PTCA was recommended if. telephone interview and if appropriate,. there were 1 or 2 target lesions while. complemented with evaluation of the. CABG was to be preferred in patients with. patients’ record.. three-vessel or left main disease (3-VD/ LMD).. Myocardial infarction was based on the presence of two out of the conventional three criteria, i.e. typical chest pain,. The non-invasive strategy included. diagnostic ECG recording (mainly new Q-. coronary angiography in patients with. wave) or elevation of biochemical markers. refractory or recurrent symptoms despite. of myocardial damage. Concerning non-. maximal medical treatment or severe. procedural myocardial infarction the. ischemia at a pre-discharge symptom-. decision levels for biochemical markers. limited exercise test. The exercise test. were: i) concentration of CK-MB-mass. criteria for performing angiography and. above the local hospital’s decision level for. revascularisation were i) ST-depression ≥. myocardial infarction at one measurement. 0.3 mV, ii) limiting chest pain associated. or ii) catalytic activity of CK, CK-B, or. with either a low Wmax (< 90 W in men or. CK-MB above the local decision level at. < 70 W in women) or a fall in blood. two subsequent determinations or iii). pressure or iii) ST-elevation without. above the double local decision level at one. preceding concomitant Q-waves or T-wave. measurement. Myocardial infarction in. inversion on exercise testing. During long-. relation to PTCA was defined by i) CK-MB-. term follow-up invasive procedures should. mass 1.5 times above the local hospital’s. be considered, regardless of randomised. decision level of myocardial infarction at. strategy, for all patients with incapacitating. one measurement or ii) catalytic activity. symptoms, recurrence of instability or. of CK, CK-B, or CK-MB at one. myocardial infarction.. measurement three times above or iii) at two determinations 1.5 times above the. Definition of endpoints. local decision level. Only new Q-waves were used for the diagnosis of myocardial. The primary endpoint of the study was. infarction in association with CABG.. death or MI at 6 months. During this time. Autopsy was recommended to establish. period data on all events, procedures,. cause of death.. symptoms and medication were collected. During the first 6 months, all reported. at regular outpatient visits. The 12 months. deaths, myocardial infarctions, elevation of. follow-up was performed by a standardised. biochemical markers in relation to PTCA -19-.

(33) procedures and new Q-waves reported by. ≥ -0.1 mV in leads V2 and aVF (if QRS. the ECG core laboratory were adjudicated. positive). In V1, aVR and III T waves were. by an independent clinical event. not evaluated. ST elevation was not. committee. After this period, information. evaluated.. on further events was evaluated by the local investigator based either on outpatient. Laboratory analyses. visits or on telephone contacts with all surviving patients and, in the case of readmission to hospital, on hospital records. The cause of death during this period was based either on hospital records or on death certificates.. At the time for randomisation venous blood was collected in tubes containing EDTA or citrate and centrifuged at each site. Plasma was frozen in aliquots and stored at -70°C until the analysis. All the analyses were performed at the Department of Clinical Chemistry, Uppsala University. ECG. Hospital, without knowledge of the ECG was collected in all patients on admission. (inclusion. randomisation,. before. ECG), and. at. patients’ randomisation and clinical outcome.. after. revascularisation and at three and six. Troponin T. months. The ECGs were evaluated at a core. Third generation TnT assay (Elecsys,. laboratory by one out of six experienced. Roche Diagnostics) was used. This assay. cardiologists or clinical physiologists, who. uses a human recombinant cardiac TnT for. were. calibration, which substantially has. unaware. of. the. patients’. randomisation and clinical outcome.. improved the precision and sensitivity of. Rhythm, presence and location of. the test. The lower limit of detection is 0.01. pathologic Q-waves, ST elevation, ST. µg/L and the upper reference level of. depression and T wave inversion were. healthy individuals is at the same level,. recorded. ST depression was considered. according to the manufacturer. At this level. present if it was horizontal or down sloping. however, the CV is unacceptably high. The. and at least 0,05 mV below the isoelectric. functional sensitivity, defined as the level. (P-R segment) level at 60 milliseconds. with a CV < 20%, of the assay is 0.03. after the J point. Inverted T wave(s) was. µg/L, which therefore was used as cut-off. considered present if the T wave was. level.. isoelectric, negative or biphasic in leads V3-V6, aVL (if R>0.5 mV) , I and II or -20-.

(34) Troponin I. which also was used as cut-off level.. TnI was analysed with Access Accu TnI, (Beckman Coulter Inc.). Also for this. Coronary angiography. analysis the functional sensitivity of the assay is 0.03 µg/L, which accordingly was used as cut-off level.. All coronary angiograms performed within six months were locally evaluated and recorded in a standardised format. The coronary vessels were divided in segments,. CRP The high sensitivity CRP test, Immulite Automated Analyser and assay kit (Diagnostic Products Corporation), was used for analysis of CRP levels 119. This is a chemiluminescent enzyme-labelled immunometric assay based on the ligandlabelled monoclonal antibody and separation by anti-ligand-coated solid phase. Fibrinogen Fibrinogen was analysed by rate nephelometry with a Beckman Array protein system (Beckman Instruments). The assay was performed according to the recommendation of the manufacturer except that goat anti-human fibrinogen (Atlantic Antibodies) was used. Human plasma standard (Behring Diagnostics GmbH) was used for calibration.. 4 from the right coronary artery, 5 from the LAD-diagonal branches, 4-6 from the left circumflex artery-obtuse marginal branches and one from the left main coronary artery, in accordance with the AHA committee report on coronary artery disease from 1975 120. Degree of stenosis (%) in each of these segments was evaluated into one of five degrees (0, <50, 50-70, >70 or 100= occlusion). A ≥ 50% stenosis in at least one of the segments in the respective vessel area (right coronary artery, LAD-diagonal and circumflexobtuse marginal) was considered significant in respect of the traditional stratification into 1-, 2- and 3-vessel disease. Also for the left main artery, a ≥ 50% stenosis was considered significant. The occurrence of visible thrombus and the TIMI flow were evaluated in each segment. If possible, the culprit lesion was identified and recorded.. Il-6 Levels of Interleukin-6 (Il-6) were analysed with Immulite (Diagnostic Products Corporation). The lower detection limit of this system is 5 ng/L,. -21-.

(35) Construction of the FRISC-score. placebo and dalteparin long-term treatment were combined in the analyses.. Variables at baseline, which could. The efficacy analyses concerning. predict a worse prognosis, were selected. death, MI and need for revascularisation. on the basis of previous experiences from. were point estimates including only. univariable risk stratification in stable and. patients with an adjudicated event or with. unstable CAD. All variables were. recorded absence of the evaluated event. dichotomised and cut-off levels were. until at least 335 days of follow-up. Kaplan-. chosen based on previous experiences. The. Meier 1-survival plots were used to. variables achieving a significance level. illustrate the timing of events without any. < 0.05 were inserted in a forward stepwise. analysis of significance. Differences. logistic regression model. Variables. between categorical variables were. independently (p<0.05) contributing to. analysed for significance by the Pearson. either the rate of death or death/MI were. (linear-by-linear association or Fisher´s. included as components of the FRISC-. exact test if appropriate) chi-square test. score.. with results presented as risk ratios with. Each factor was given 1 point. 95% confidence intervals. For continuous. regardless of the magnitude of the odds. variables, the analysis of variance. ratio or whether the factor was present as. (ANOVA) model was used to assess. independent risk indicator for either or. significance of differences.. both of death or death/MI. Thus, the FRISC-. Multivariable logistic regression. score was for each patient simply calculated. analyses were used to identify independent. by counting the number of factors. risk indicators for future cardiac events.. contributing to the score. Only patients. We used the Hosmer-Lemeshow test to. with available data from all the factors in. assess the goodness of fit of the forward. the FRISC-score were included in the. logistic regression models used for the. analyses.. construction of the FRISC-score. The predictive value of the FRISC-score was. Statistical methods. also evaluated with the area under a receiver operating curve, ROC. For all tests, p<0.05. The statistical evaluation was. was considered significant. The data. performed on an intention-to-treat basis.. processing and statistical analyses were. As the dalteparin treatment did not. performed using the SPSS 10.0 or 10.1. significantly influence the results 117 the. statistical program. -22-.

(36) Results invasive strategy 117. In the invasive cohort. Main FRISC II study In the main FRISC II study the primary. more than 70% of the patients had a. endpoint, death or myocardial infarction at. revascularisation procedure within 10 days. 6 months, was reduced from 12.1% in the. with a slight increase during 12 months.. non-invasive to 9.4% in the invasive cohort,. The corresponding number in the non-. p=0.031. The corresponding numbers for. invasive cohort was 9.0%, with a. mortality were 2.9 and 1.9% respectively,. continuous increase during follow-up, fig 4.. p=0.10. After the initial two weeks, survival. 0.16. Non-invasive n=1235. analyses of both mortality and death/MI. 0.14. p = 0.005 Probability of death or myocardial infarction. indicated a continuously increasing separation of event rates between the two treatment strategies favouring the early invasive strategy, fig 2-3. Thus, at 12. 0.04. 30. 60. 90. 120 150 180 210 240 270 300 330 360. Figure 2. Probability of death or myocardial infarction in invasive and non-invasive groups. Reprint with permission from Elsevier Science (The Lancet, 2000, 356, 12).. symptoms of angina pectoris, cardiac for. 0.06. days. Also the need for readmission, need. Invasive n=1222. 0.08. 0.00 0. reduced, from 3.9 to 2.2%, p=0.016 121.. and. 0.10. 0.02. months also mortality was significantly. medication. 0.12. late. revascularisation were reduced by the 1.00. 0.05. 0.04. Probability of death. Probability of revascularisation. 0.90. Non-invasive n=1235. p = 0.016. 0.03. Invasive n=1222 0.02. 0.01. Invasive strategy n=1222. 0.80 0.70 0.60 0.50. Non-invasive strategy n=1235. 0.40 0.30 0.20 0.10. 0.00 0. 0.00. 30. 60. 90. 0. 120 150 180 210 240 270 300 330 360. Figure 3. Probability of death in invasive and non-invasive groups. Reprint with permission from Elsevier Science (The Lancet, 2000, 356, 12).. 30. 60. 90. 120 150. 180 210 240 270 300 330. 360 days. days. Figure 4. Timing of first revascularisation in invasive and non-invasive groups. Reprint with permission from Elsevier Science (The Lancet, 2000, 356, 11).. -23-.

(37) The blood samples for analyses of. 39 h (10th to 90th percentile, 18 to 71 h). markers of myocardial damage and. from the last episode of chest pain. An. inflammation. overview of the substudies analysed in the. were. obtained. at. randomisation which occurred in median. papers is shown in fig 5. Paper I (n=2408) ECG possible to evaluate. Paper II (n=2286) ECG and available analysis of Troponin T. Paper III (n=2208) Available analyses of CRP and fibrinogen. FRISC II invasive study n=2457. Paper IV (n=2306) Available analysis of Interleukin-6. Paper V (n=2287) All independent risk variables available. Figure 5. Overview of the papers and criteria for recruitment in the studies.. (i.e. ≥ 0.03 µg/L).. Prognosis in relation to risk variables, non-invasive group. In patients with ST depression the 12 months rate of death/MI was 18.2% as. ST depression and Troponin T (paper I-. compared to 10.4% in the rest of the study. II). population, table 1 and fig 6. In patients ST depression occurred in 1114. without STT changes and in those with. (45.5%) patients, isolated T wave inversion. isolated T wave inversion, at 12 months the. was found in 871 (35.6%) and 423 (17.3%). rates of death/MI were similar, 10.3 and. had no STT changes. Approximately two. 10.5% respectively. Also the death rate was. thirds of patients had TnT elevation. higher in patients with ST depression, 5.8 -24-.

(38) Table 1. Death and death or myocardial infarction within 12 months in the invasive and noninvasive strategy in relation to occurrence of ST depression in admission ECG. patients Inv/noninv n/n. Invasive n (%). Events Noninvasive n (%). ST depression Death Death or MI. 542/571 540/571. 18 (3.3%) 65 (12.0%). 33 (5.8%) # 104 (18.2%). No ST depression Death Death or MI. 651/643 650/643. 8 (1.2%) 58 (8.9%). 13 (2.0%) # 67 (10.4%). RR (95% CI)*. p. †. 0.58 (0.33-1.01) 0.66 (0.50-0.88). 0.050 0.004. †. 0.61 (0.25-1.46) 0.86 (0.61-1.20). 0.26 0.36. †. Comparison between patients with and without ST depression: RR 2.86 (CI 95 1.52-5.38), p=0.001 Comparison between patients with and without ST depression: RR 1.75 (CI 95 1.31-2.33), p<0.001 Abbreviations, see table 2.. #. Table 2. Death and death or myocardial infarction within 12 months in relation to troponin T and its combination with the occurrence of ST depression, in the invasive and non-invasive groups. patients Inv/noninv n/n TnT < 0.03 µg/L Death Death/MI TnT ≥ 0.03 µg/L Death Death/MI TnT<0.03µg/L&no ST↓ Death Death/MI TnT<0.03µg/L& ST↓ Death Death/MI TnT≥0.03µg/L&no ST↓ Death Death/MI TnT ≥ 0.03 µg/L & ST↓ Death Death/MI. Events Invasive Noninvasive n (%) n (%). RR (95% CI)*. p. †. 0.47 (0.14-1.54) 0.20 0.87 (0.53-1.43) 0.59. 33(4.2%) # 132(16.6%). †. 0.63 (0.37-1.09) 0.098 0.70 (0.54-0.90) 0.005. 0(0%) 14(6.2%). 3(1.5%) 18(8.8%). 0.11 0.70 (0.36-1.38) 0.31. 152/149 152/149. 4(2.6%) 14(9.2%). 5(3.4%) 12(8.1%). 0.78 (0.22-2.86) 0.75 1.14 (0.55-2.39) 0.72. 394/406 393/406. 7(1.8%) 40(10.2%). 10(2.5%) 46(11.3%). 0.72 (0.28-1.88) 0.50 0.90 (0.60-1.34) 0.60. 367/389 365/389. 13(3.5%) 48(13.2%). 23(5.9%) 86(22.1%). 0.60 (0.31-1.16) 0.13 0.60 (0.43-0.82) 0.001. 377/353 377/353. 4(1.1%) 28(7.4%). 8(2.3%) # 30(8.5%). 761/795 758/795. 20(2.6%) 88(11.6%). 225/204 225/204. †. Comparison between patients with and without elevated Troponin T: RR : RR 1.83 (CI 95 0.85-3.92), p=0.11 # Comparison between patients with and without elevated Troponin T: RR 1.95 (CI 95 1.34-2.85), p < 0.001 ＊ Risk ratio and 95% confidence interval refer to effect by invasive strategy compared to non-invasive strategy. inv/noninv = invasive / noninvasive TnT = Troponin T MI = myocardial infarction ST↓ = ST depression. -25-.

(39) compared to 2.0%, table 1. A raised TnT. depression on admission and increased. level (≥ 0.03 µg/L) was associated with. levels of markers of myocardial damage. increased risk for death/MI at 12 months,. (TnT ≥ 0.03 µg/L) were independently. 16.6 vs 8.5%, p<0.001, and a weak trend. related to the composite of death and MI,. to a higher mortality, 4.2 vs 2.3%, p=0.11,. table 3. Age ≥ 70 years, diabetes, ST. table 2. If both ST depression and TnT ≥. depression on admission and increased. 0.03 µg/L were present death/MI at 12. levels of Il-6/CRP were independently. months occurred in 22.1% compared to. associated with an increased mortality, table. 10.0% in the rest of the study population,. 4. Diabetes was the strongest independent. RR 2.21 (95% CI 1.66-2.93) p<0.001.. risk factor both for death and death/MI.. Also for mortality these patients were at. Increased inflammatory activity was after. higher risk, 5.9 and 2.4% respectively, RR. diabetes the strongest predictor for. 2.49 (95% CI 1.36-4.57) p=0.002.. mortality. Variables associated with an. Inflammatory markers (paper III-IV). independent risk for death and/or death/MI. A CRP level ≥ 10 mg/L was observed. were included in a risk score for evaluation. in 35% of the patients and 26% had an Il-6. of prognosis and effect of the invasive. level ≥ 5 ng/L. Increased levels of CRP,. strategy. Thus, age ≥ 70 years, male sex,. fibrinogen and Il-6 did not correlate to a. diabetes, previous MI, ST depression and. raised rate of the composite of death and. increased levels of markers of myocardial. MI. However, raised levels of both CRP and. damage and inflammation were included in. Il-6, but not fibrinogen, were significantly. the score. In order to obtain strata with. associated with increased mortality. For. adequate patient numbers those with 0 or 1. tertiles of CRP the 12 months death rates. risk variable and those with ≥ 5 risk. were 2.5, 3.2 and 6.0% respectively,. variables were merged.. p=0.014. In patients with Il-6 ≥ 5 ng/L. There was a linear relation between the. mortality was 7.9% compared to 2.3% in. numbers of risk indicators and the. patients with Il-6 < 5 ng/L, RR 3.47 (95%. occurrence of death and the composite of. CI 1.94-6.21) p<0.001.. death and MI, p < 0.001 for both. In patients with 5-7 of these variables death rate was. Multivariable analyses and the FRISC-. 15.4% and death/MI occurred in 36.9%.. score (paper V). The corresponding numbers in patients with. In the multivariable analyses, age ≥ 70 years, male sex, diabetes, previous MI, ST. 0-1 variables were 0.5 and 4.7% respectively, fig 7-8. -26-.

(40) Probability of death or myocardial infarction. .20 ST depression n = 571 .18 .16 p < 0.001. .14 .12. No STT changes n = 204. .10 Isolated T wave inversion n = 439. .08 .06 .04 .02 0.00 0. 60. 120. 180. 240. 300. 360 Days. Figure 6. Probability of death or myocardial infarction in relation to STT changes on admission in the non-invasive cohort. The p value is evaluated by chi2 analysis as a point estimate at twelve months and refers to difference in outcome between the patients with and without ST depression. % mortality 18. 15.4. 16 14 12 10 8. 6.2. 6 4 2. 2.0. 1.3. 0-1. 2. 3. 4. 5-7. 212. 299. 302. 209. 130. 0.5. 0 FRISC score. n. Figure 7. One-year mortality in relation to FRISC-score, non-invasive cohort, p<0.001 (linearby-linear association chi-square test). % death or myocardial infarction 40. 36.9. 35 30 25. 19.6. 20. 12.9. 15. 8.7. 10 5. 4.7. 0 FRISC score. n. 0-1. 2. 3. 4. 5-7. 212. 299. 302. 209. 130. Figure 8. One-year death or myocardial infarction in relation to FRISC-score, non-invasive cohort, p<0.001 (linear-by-linear association chi-square test).. -27-.

(41) -28326/1176 27.7. Il-6 ≥ 5 ng/L / CRP ≥ 10 mg/L 16.9. 16.0. 16.7. 18.2. 19.4. 23.5. 29.9. 18.0. 12.4. 13.4. 15.8. 21.7. present. median value. Interleukin-6 ≥ 5 ng/L or if not available, C-reactive protein ≥ 10 mg/L. †. 13.4. 12.1. 8.3. 10.4. 11.3. 11.5. 12.0. 12.4. 15.3. 14.6. 10.5. 10.7. absent. risk variable. Death/MI (%) risk variable. Troponin T ≥ 0.03 µg/L or if not available, Troponin I ≥ 0.03 µg/L. §. ‡. *Hosmer-Lemeshow test, chi-square=4.2, p=0.84. 617/1169 52.8. † Fibrinogen ≥ 3.8 g/L §. 572/1215 47.1. ST depression at entry 828/1188 69.7. 428/1235 34.7. Angina pectoris >3 months ‡. 268/1235 21.7. Previous myocardial infarction. Increased troponin. 377/1235 30.5 144/1235 11.7. 588/1194 49.2. Diabetes. †. Hypertension. Cholesterol ≥ 5.8mmol/L. 834/1235 67.5 671/1218 54.2. Body mass index ≥ 26kg/m 2†. Male sex. %. 382/1235 30.9. n. Age ≥ 70 years. Risk variable. variable criteria. Fulfilling risk. Table 3. Death or myocardial infarction in relation to risk variables, univariable and multivariable analyses*, non-invasive cohort.. 1.26 (0.94-1.69). 1.32 (0.99-1.76). 2.00 (1.38-2.91). 1.75 (1.31-2.32). 1.72 (1.31-2.62). 2.05 (1.55-2.70). 2.49 (1.85-3.34). 1.46 (1.10-1.93). 0.81 (0.61-1.08). 0.92 (0.69-1.21). 1.51 (1.09-2.09). 2.04 (1.55-2.67). Risk ratio (95% CI). Odds Ratio (95% CI) p. n=1168. multivariable analyses. 1.50 (1.00-2.25). 0.31. 0.049. 0.13. 0.056. <0.001 1.84 (1.18-2.86). <0.001 1.72 (1.21-2.45). <0.001. 0.007. 0.003. 0.07. <0.001 2.00 (1.38-2.89) <0.001. <0.001 3.14 (2.05-4.81) <0.001. 0.008. 0.14. 0.54. 0.011. <0.001 1.95 (1.37-2.78) <0.001. p. univariable analyses.

(42) -29§. 326/1176 27.7. 617/1169 52.8 8.3. 4.9. 4.5. 5.8. 5.6. 6.7. 12.5. 6.4. 3.1. 3.7. 4.2. 6.8. present. risk variable. median value. Interleukin-6 ≥ 5 ng/L or if not available, C-reactive protein ≥ 10 mg/L. †. 2.4. 2.7. 2.5. 2.0. 3.0. 3.1. 2.7. 2.8. 4.5. 3.8. 3.2. 2.6. absent. risk variable. Mortality (%). Troponin T ≥ 0.03 µg/L or if not available, Troponin I ≥ 0.03 µg/L. §. ‡. *Hosmer-Lemeshow test, chi-square = 2.8, p = 0.73. Il-6 ≥ 5 ng/L / CRP ≥ 10 mg/L. Fibrinogen ≥ 3.8 g/L. 828/1188 69.7. ST depression at entry †. 572/1215 47.1. Angina pectoris >3 months ‡. 428/1235 34.7. Previous myocardial infarction. Increased troponin. 144/1235 11.7 268/1235 21.7. Diabetes. 588/1194 49.2 377/1235 30.5. Hypertension. 671/1218 54.2. Body mass index ≥ 26kg/m. † Cholesterol ≥ 5.8mmol/L. 834/1235 67.5 2†. 382/1235 30.9. %. Male sex. n. variable criteria. Age ≥ 70 years. Risk variable. Fulfilling risk. 3.52 (2.00-6.17). 1.79 (0.97-3.29). 1.79 (0.87-3.66). 2.86 (1.52-5.38). 1.89 (1.08-3.28). 2.16 (1.23-3.82). 4.55 (2.60-7.94). 2.28 (1.31-3.95). 0.69 (0.38-1.23). 0.97 (0.55-1.72). 1.30 (0.69-2.42). 2.64 (1.52-4.59). Risk ratio (95% CI). 0.15. 2.54 (1.29-5.00). 0.007. 0.18. 0.055. <0.001 3.26 (1.75-6.09) <0.001. 0.057. 0.11. 0.001. 0.023. 0.007. <0.001 4.18 (2.17-8.04) <0.001. 0.003. 0.21. 0.92. 0.42. 0.014. Odds Ratio (95% CI) p. n=1157. multivariable analyses. <0.001 2.18 (1.17-4.07). p. univariable analyses. Table 4. Mortality in relation to risk variables, univariable and multivariable analyses*, noninvasive cohort..

(43) Coronary artery lesions in relation to. Inflammatory markers (paper III). risk variables. No significant correlation was observed between the occurrence of 3-VD/. ST depression and Troponin T (paper I-. LMD and high levels of CRP, fibrinogen. II). or Il-6. However, low levels were more In patients with ST depression 8.5%. often associated with freedom from. had no significant stenoses compared to 17. significant coronary stenosis. There was no. and 19% in patients without STT changes. correlation between tertiles of CRP,. and with isolated T wave inversion. fibrinogen or Il-6 level and the occurrence. respectively. Three-VD/LMD occurred in. of visible thrombus. However, the number. 18, 24 and 45% in the groups with no STT. of patients with TIMI III flow in all vessels. changes, isolated T wave inversion and ST. decreased with higher levels, for CRP 60,. depression respectively, fig 9.. 54 and 42% (p<0.001) and for fibrinogen. In patients with TnT level ≥ 0.03 µg/L 36% had 3-VD/LMD. Increasing levels of. 57, 51 and 48% (p=0.009), whereas for Il6 no correlation was observed.. TnT were associated with more visible thrombus and fewer patients with TIMI III. Combination of risk variables, FRISC-. flow in all major coronary vessels 122.. score (paper V). In patients with both ST depression and. The FRISC-score correlated to the. TnT ≥ 0.03 µg/L, 3-VD/LMD occurred in. severity of CAD. Thus, in the high risk. 49% compared to 17% if neither TnT ≥. group with 5-7 factors from the score, no. 0.03 µg/L nor ST depression were present,. patient was without significant stenoses and. fig 10.. 62% had 3-VD/LMD compared to 11% in % 40 35. 0-VD. 30 1-VD. 25 20. 2-VD. 15 3-VD. 10 5. LMD. 0 No STT changes Isolated T inv n = 218 n = 427. ST depression n = 529. Figure 9. Extent of coronary artery disease at angiography in relation to STT changes on admission, invasive cohort only. 0 VD = no significant stenosis, 1-, 2- and 3 -VD = 1-, 2- and 3- vessel disease, LMD = left main disease, inv = inversion.. -30-.

(44) patients with 0-1 factors, fig 11. Also the. 12, whereas no correlation to the. number of patients with TIMI III flow in all. occurrence of visible thrombus was. vessels correlated to the FRISC-score, fig. observed.. % 50 45 40. 0-VD. 35 30. 1-VD. 25 20. 2-VD. 15. 3-VD/ LMD. 10 5 0. No ST dep ST dep n = 223 n = 149 Troponin T < 0.03 µg/L. No ST dep ST dep n = 391 n = 357 Troponin T ≥ 0.03 µg/L. Figure 10. Extent of coronary artery disease at angiography in relation to Troponin T level and occurrence of ST depression (ST dep) in the invasive cohort. 0-VD = no significant stenosis, 1- and 2VD = 1- and 2- vessel disease, 3-VD/LMD = 3-vessel or left main disease. %. 70 60 50. 0-VD. 40. 1-VD. 30. 2-VD. 20. 3-VD/ LMD. 10 0. 0 FRISC score. n. 0-1 179. 2 279. 3 326. 4 204. 5-7 129. Figure 11. Extent of coronary artery disease at angiography in relation to FRISC-score, invasive cohort. Abbreviations, see fig 10. % TIMI III flow in all vessels. 90 80. 76.8. 70. 58.3. 60. 52.6. 50. 39.2. 40. 26.4. 30 20 10 0 FRISC score. n. 0-1. 2. 3. 4. 5-7. 177. 278. 325. 204. 129. Figure 12. Number of patients (%) with TIMI III flow in all major vessels in relation to FRISC-score, invasive cohort, p<0.001 (linear-by-linear association chi-square test).. -31-.

(45) Effects of the invasive strategy in. death/MI was reduced, from 16.6 to 11.6%,. relation to risk variables. table 2. If both ST depression and TnT ≥ 0.03 µg/L were present death/MI was. ECG and Troponin T (paper I-II). reduced from 22.1 to 13.1%, whereas a. In patients with ST depression the. numeric but not significant reduction of. invasive strategy reduced death/MI at 12. mortality was observed, table 2 and fig 15-. months from 18.2 to 12.0% and death from. 16. In patients with neither of these risk. 5.8 to 3.3% (p=0.050), table 1 and fig 13-. indicators no significant effect on death or. 14. Also in patients with TnT ≥ 0.03 µg/L. death/MI was observed, table 2.. Probability of death or myocardial infarction. .20. Noninvasive (n=571). .18 .16. ST dep. P < 0.01 .14. Invasive (n=540) .12. Noninvasive (n=643). .10 .08. No ST dep. Invasive (n=650). .06 .04 .02 0.00 0. 60. 120. 180. 240. 300. 360. Days. Figure 13. Probability of death or myocardial infarction in relation to strategy and occurrence of ST depression on admission. The p value is evaluated by chi2 analysis as a point estimate at twelve months and refers to difference in outcome between invasive and non-invasive strategy. ST dep = ST depression. 0.06. Noninvasive (n=571). Probability of death. 0.05. ST dep. p = 0.050 0.04. Invasive (n=542) 0.03. 0.02. No ST dep. Noninvasive (n=643). 0.01. Invasive (n=651). 0.00 0. 60. 120. 180. 240. 300. 360. Days. Figure 14. Probability of death in relation to strategy and occurrence of ST depression on admission. The p value is evaluated by chi2 analysis as a point estimate at twelve months and refers to difference in outcome between invasive and non-invasive strategy. ST dep = ST depression.. -32-.

(46) Probability of death or myocardial infarction. 0.25. TnT ≥ 0.03 µg/L and ST dep Noninvasive n = 389. 0.20. TnT ≥ 0.03 µg/L and ST dep Invasive n = 365. 0.15. TnT < 0.03 µg/L and/or no ST dep Noninvasive n = 759. 0.10. TnT < 0.03 µg/L and/or no ST dep Invasive n = 770. 0.05. 0.00 0. 60. 120. 180. 240. 300. 360. Days. Figure 15. Probability of death or myocardial infarction during twelve months in relation to invasive or non-invasive treatment strategies and combination of Troponin T < or ≥ 0.03 µg/L and occurrence of ST depression. TnT = Troponin T. ST dep = ST depression.. TnT ≥ 0.03 µg/L and ST dep Noninvasive n = 389. 0.06. Probability of death. 0.05. TnT ≥ 0.03 µg/L and ST dep Invasive n = 367. 0.04. 0.03 TnT < 0.03 µg/L and/or no ST dep Noninvasive n = 759. 0.02. TnT < 0.03 µg/L and/or no ST dep Invasive n = 771. 0.01. 0.00 0. 60. 120. 180. 240. 300. 360. Days. Figure 16. Probability of death during twelve months in relation to invasive or non-invasive treatment strategies and combination of Troponin T < or ≥ 0.03 µg/L and occurrence of ST depression. TnT = Troponin T. ST dep = ST depression.. -33-.

(47) Inflammatory markers (paper III-IV) The invasive strategy reduced death/. total population by the invasive strategy without any indication of heterogeneity. MI mainly in the two lower tertiles of CRP and fibrinogen, whereas in the upper tertile. between the tertiles of CRP or fibrinogen, fig 19.. no significant effect was observed, fig 17. In addition, in female patients in the highest. The reduction of death/MI was similar in patients with Il-6 levels < and ≥ 5 ng/L.. tertile of fibrinogen, death MI occurred in 19.8% in the invasive compared to 10.7%. However, for mortality the beneficial. in the non-invasive group, p= 0.049, fig 18. However, the mortality was reduced in the. observed in patients with Il-6 ≥ 5 ng/L, table. n/n inv/noninv. effect of the invasive strategy was mainly 5 and fig 20.. Death or MI (%) inv. noninv. C-reactive protein Low tertile. 373/365. 7.5. 14.5. Middle tertile. 359/373. 10.3. 16.1. Upper tertile. 368/367. 12.2. 12.8. Fibrinogen Low tertile. 379/383. 7.9. 11.5. Middle tertile. 344/317. 8.4. 17.0. Upper tertile. 377/405. 13.5. 15.3. 0.25. 0.5. 0.75. Invasive strategy better. 1. 1.5. 2. Non-invasive strategy better. Figure 17. Effect of the early invasive strategy on death or myocardial infarction at 12 months in relation to tertiles of C-reactive protein and fibrinogen (risk ratio and 95% confidence interval). Inv = invasive. Noninv = non-invasive. MI = myocardial infarction. n/n Death or MI (%) inv/noninv inv noninv Male C-reactive protein Low tertile 265/258 Middle tertile 256/241 Upper tertile 271/253. 7.2 10.9 10.0. 17.1 19.1 13.0. Fibrinogen Low tertile Middle tertile Upper tertile. 272/269 234/228 286/255. 8.5 7.7 11.5. 13.0 18.4 18.0. Female C-reactive protein Low tertile 108/107 Middle tertile 103/132 Upper tertile 97/114. 8.3 8.7 18.6. 8.4 10.6 12.3. Fibrinogen Low tertile Middle tertile Upper tertile. 6.5 10.0 19.8. 7.9 13.5 10.7. 107/114 110/89 91/150. 0.25 0.5 0.75 1 1.5 2 4 Invasive strategy better Non-invasive strategy better. Figure 18. Effect of the early invasive strategy on death or myocardial infarction at 12 months in relation to gender and tertiles of C-reactive protein and fibrinogen. Abbreviations, see fig 17.. -34-.

(48) Non-invasive. Invasive. 7 % mortality at one year. 6 5 4 3 2 1 0 Tertile n=2208. 1 2 3 C-reactive protein. 1. 2 fibrinogen. 3. Figure 19. One-year mortality in relation to strategy and tertiles of C-reactive protein and fibrinogen. Interleukin-6 ≥ 5 ng/L Noninvasive n = 318. ,08 ,07. Probability of death. ,06 ,05 ,04. Interleukin-6 ≥ 5 ng/L Invasive n = 324. ,03 Interleukin-6 < 5 ng/L Noninvasive n = 838 ,02 Interleukin-6 < 5 ng/L Noninvasive n = 826. ,01 0,00 0. 60. 120. 180. 240. 300. 360. Days. Figure 20. Probability of death in relation to strategy and Interleukin-6 levels (< or ≥ 5ng/L).. Table 5. Outcome in relation to treatment strategy and Interleukin-6 levels. patients Inv/noninv n/n. Events Invasive Noninvasive n (%) n (%). RR (95% CI)*. p. Interleukin-6 ≥5 ng/L Death Death or MI. 324 / 318 324 / 318. 9 (2.8%) 37 (11.4%). 25 (7.9%) 0.35 (0.17-0.74) 0.004 53 (16.7%) 0.68 (0.46-1.02) 0.056. Interleukin-6 < 5 ng/L Death Death or MI. 826 / 838 826 / 838. 16 (1.9%) 81 (9.8%). 19 (2.3%) 0.85 (0.44-1.65) 0.64 112 (13.4%) 0.73 (0.56-0.96) 0.023. inv/noninv = invasive / non-invasive * Risk ratio and 95% confidence interval refer to effect of the invasive strategy.. -35-.

(49) FRISC-score (paper V). non-invasive strategy regarding death or. For evaluation of treatment effect the. death/MI. If 3-4 factors were present death/. patients were categorised into low,. MI was reduced from 15.7 to 10.8% by the. intermediate or high risk based on the. early invasive strategy. In the high risk. FRISC-score, 0-2, 3-4 or 5-7. In patients. group with 5-7 factors, the invasive strategy. with 0-2 factors there was no significant. reduced mortality from 15.4 to 5.2% and. difference between the early invasive and. death/MI from 36.9 to 16.5%, fig 21-22.. non-invasive. % 18 16 14 12 10 8 6 4 2. 15.4. 5.2 3.3. 3.0. 1.4 0.4. 0 Score n RR p. invasive. 0-2 972. 3-4 1051. 0.32 (0.07-1.52) 0.18. 0.89 (0.46-1.74) 0.74. 5-7 264 0.34 (0.15-0.78) 0.006. Figure 21. One-year mortality in relation to treatment strategy and the FRISC-score. Risk ratios, confidence intervals and p values refer to the comparison between the invasive and non-invasive strategies. % non-invasive invasive. 40. 36.9. 35 30 25 20. 10. 16.5. 15.7. 15 7.0. 10.8. 8.0. 5 0. Score n RR p. 0-2 972 1.14 (0.73-1.77) 0.56. 3-4 1049 0.69 (0.50-0.94) 0.020. 5-7 263 0.45 (0.29-0.70) < 0.001. Figure 22. One-year death or myocardial infarction in relation to treatment strategy and the FRISCscore. Risk ratios, confidence intervals and p values refer to the comparison between the invasive and non-invasive strategies.. -36-.

(50) Discussion Despite the introduction of new efficient anti-thrombotic drugs, patients. while CABG is associated with an improved survival in high-risk patients 113.. with UCAD still have a high risk for subsequent death or MI. An even more. The FRISC II study was the first and. aggressive treatment including early. so far the only large scale study to. coronary angiography and revascularisation. demonstrate both an improved survival and. has been suggested to further decrease event. a reduction in MI by an early invasive. rates. There have been considerations for. strategy in UCAD. Concerning the. hazards associated with revascularisation. reduction in MI these results were. procedures in the early phase of the ACS 123. However, the continuous development. confirmed in the similarly large TACTICS. in new technologies both at PTCA and. the invasive strategy was associated with. CABG have reduced the peri-procedural. reduced mortality and MI rates 128 .. and long-term risk making early. Recently the reduction in mortality by an. revascularisation more appealing. Still,. early invasive strategy after MI has also. until the present study, no major advantages. been supported in a large scale observation. with an early invasive strategy had been. study 129.. trial 127. Also in the smaller VINO study. demonstrated in UCAD 115,124. However, in patients with STEMI treated with. The design and the magnitude of the. thrombolysis and with residual ischemia at. FRISC II study also allowed identification. a. test. of risk indicators and a risk score for. revascularisation with PTCA or CABG was. identification of patients with the largest. associated with lower rates of death/MI 125.. benefit of an early invasive strategy.. predischarge. exercise. Furthermore, in patients with STEMI, in meta-analyses from randomised studies,. Early invasive strategy-randomised. primary PTCA was compared to. studies. thrombolysis associated with a reduced risk. Except for the FRISC II study, three. of future MI and mortality 5. In stable. other large-scale studies have compared an. angina pectoris however, compared to. early invasive with a non-invasive strategy,. medical treatment, PTCA does not seem to. table 6. Before the FRISC II study was. be associated with lower event rates 126. started only the neutral results of the TIMI -37-.

(51) Table 6. Background data and results from the four large randomised studies comparing an early invasive with a non-invasive strategy.. n Inclusion period Diagnosis Medication (LMW)heparin GP IIb/IIIa-inh Stents. TIMI IIIb. VANQWISH. FRISC II. TACTICS. 1473 1989-1992 NSTEMI/UA. 920 1993-1995 NSTEMI. 2457 1996-1998 NSTEMI/UA. 2220 1997-1999 NSTEMI/UA. Yes No No. No (if needed) No No. Yes No (10%) Yes. Yes Yes Yes. Noninv Angiography Discharge / 7d Revascularisation Discharge / 10d 6 - 12 months 6-12 mon cabg 6-12 mon ptca 30d mort cabg 30d mort ptca Prim endpoint Death 30 days 6 months 12 months Death or MI 6 months 12 months. Inv. Noninv. Inv. Noninv. Inv. Noninv. Inv. 57. 98. 24. 94. 10. 96. 51. 97. 40 58 30 28. 60 64 30 34. 33 21 12 3.4 3.6. 44 23 21 10.4 0. 9 37 19 18 1.7 0.4. 71 77 36 41 2.1 0.2. 36 44 16 28 3.6 1.9. 60 61 22 40 3.6 1.9. 18.1*. 16.2*. 26.9. #. 12.1. †. 19.4. ‡. ‡. #. 29.9. †. 9.4. §. 15.9. 2.0 4.4. 2.3 4.1. 2.0 7.9. 5.0 12.6. 1.4 2.9 3.9. 1.1 1.9 2.2. 1.6 3.5 -. 2.2 3.3 -. 12.2. 10.8. 18.6. 24. 12.1 14.1. 9.4 10.4. 9.5 -. 7.3 -. Data presented are percent of patients NSTEMI = non ST elevation MI UA = unstable angina - = not available noninv = non-invasive inv = invasive mon = months ‡ 42 days mort = mortality MI = myocardial infarction LMW = low molecular weight inh = inhibitors d = days cabg = coronary artery bypass grafting prim = primary ptca = percutaneous transluminal coronary angioplasty * Death, MI or a failed symptom-limited exercise test at 6 weeks # Death or MI at the end of the study, average 23 months † Death or MI at six months § Death, MI or rehospitalisation for an acute coronary syndrome at six months. -38-. §.

(52) IIIB trial were available. Shortly after the. study. The follow-up period was in average. inclusion period of the FRISC II study the. 23 months (range 12-44).. VANQWISH study was published. Finally. There was no difference between the. the TACTICS-TIMI 18 trial was published. invasive and non-invasive strategy. about two years after the FRISC II study.. concerning the primary endpoint. However, at discharge, one and twelve months both. TIMI IIIB 124,130 From October 1989 to June 1992,. mortality and death/MI were significantly higher in the invasive cohort.. 1473 patients with unstable angina or nonq-wave MI were randomly assigned in a 2x2. TACTICS 127. factorial design to 1/ an early invasive or a. Between December 1997 and. primarily non-invasive strategy and to 2/. December 1999, the study included 2220. tissue plasminogen activator (TPA) or. patients with unstable angina or non-ST. placebo. The primary endpoint in the. elevation MI. The primary endpoint was a. invasive/non-invasive arm of the study was. composite. the composite of death, MI and a failed. rehospitalisation for an ACS at six months,. symptom-limited exercise test at 6 weeks.. which was reduced by the early invasive. In the medical study treatment with. strategy. Also the rate of death/MI was. TPA was, compared to placebo, associated. reduced at six months, 7.3 versus 9.5%, RR. with an increased risk for death/MI. There. 0.74 (95 CI 0.54-1.00), p<0.05.. of. death,. MI. and. was no difference between the invasive and non-invasive strategy concerning the. Comparison of the four studies (table 6). primary endpoint. However, secondary. A probable explanation to the. endpoints such as length of initial. beneficial effects observed in the two later. hospitalisation and the need for. studies, FRISC II and TACTICS, is the. rehospitalisation were reduced by the. reduced complication rate associated with. invasive strategy.. revascularisation procedures over the last years. At PTCA the introduction of. VANQWISH 115. coronary stenting has provided an excellent. From April 1993 to December 1995,. tool for reduction of the occurrence and. 920 patients (> 97 % men) with non-q-wave. the consequences of an abrupt vessel. MI were randomised to either invasive or. closure 131,132. The use of GPIIb/IIIa. conservative management. The primary. inhibitors has reduced the risk of. endpoint was death/MI at the end of the. procedure-related MI 99,133. Both of these -39-.

No results found