Silent Ischemia Predicts Infarction and Death During 2 Year Follow-Up of Unstable Angina

Transcription

1 756 JACC Vol. 10, No.4 Silent Ischemia Predicts Infarction and Death During 2 Year Follow-Up of Unstable Angina SIDNEY O. GOTTLIEB, MD, FACC, MYRON L. WEISFELDT, MD, FACC, PAMELA OUYANG, MD, FACC, E; ~AVID MELLITS, ScD, GARY GERSTENBLIT, MD, FACC Baltimore, Maryland Silent myocardial ischemia as detected on olter electrocardiographic (ECG) monitoring is present in >50% of patients with unstable angina despite intensive medical therapy. The presence and the extent of silent ischemia have been correlated with an increased risk of early (1 month) unfavorable outcome including myocardial infarction and need for coronaryrevascularizationfor persistent symptoms. Seventy patients with unstable angina who had undergone continuous ECG monitoring for silent ischemia were followed up for 2 years; 37 patients (Group I) had olter ECG evidence of silent ischemia at bed rest in the coronary care unit during medical treatment with nitrates, beta-receptor blockers and calcium channel antagonists; the other 33 patients (Group II) had no ischemic ST segment changes (symptomatic or silent) on olter monitoring. Over a 2 year follow-up period, myocardial infarction occurred in 10 patients in Group I (in 2 it was fatal) compared with one nonfatal infarction in Group II (p < 0.01 by Kaplan-Meier analysis); revascularization with either coronary bypass surgery or angioplasty for symptomatic ischemia was performed in 11 Group I and 5 Group II patients (p < 0.05). Multivariate Cox's hazard analysis demonstrated that the presence of silent ischemia was the best predictor of 2 year outcome. Therefore, persistent silent myocardial ischemia despite medical therapy in patients with unstable angina carries adverse prognostic implications that persist over a 2 year period. () Am Coll CardioI1987;lO:756-60) Asymptomatic ischemic ST segment changes on continuous olter electrocardiographic (ECO) monitoring are common in patients with coronary artery disease with stable (1-3) and unstable (4-6) angina pectoris and are uncommon in normal subjects (7). The presence of asymptomatic ST segment changes on olter monitoring was previously demonstrated (4) to identify patients at high risk for the early (I month) development of myocardial infarction or need for coronary revascularization for recurrent symptoms. We recently completed a 2 year follow-up evaluation of a group of 70 patients with unstable angina who underwent contin- uous olter monitoring during the first 2 days in the coronary care unit after hospital admission, while they were being treated with combination medical therapy in a protocol that included nitrates, beta-receptor blockers and calcium channel antagonists. The goal of the present study was to determine in these patients the long-term prognostic significance of silent ischemia detected on olter monitoring with respect to the 2 year clinical outcome of death and myocardial infarction. The need for revascularization with either bypass surgery or coronary angioplasty for recurrent symptoms was also assessed. From the Division of Cardiology, Departmentof Medicine, The Francis Scott Key Medical Center and The Johns opkins ospital, Baltimore, Maryland. This study was supported in part by Ischemic eart Disease Specialized Center of Research Grant Ll from the National eart, Lung, and Blood Institute, National Institutes of ealth, Bethesda, Maryland. Manuscript received February 2, 1987; revised manuscript received April 22, 1987, accepted May 6, Address for reprints: Sidney O. Gottlieb, MD, Director, Cardiac Catheterization Laboratory, The Francis Scott Key Medical Center, The Johns opkins Medical Institutions, 4940 Eastern Avenue, Baltimore, Maryland by the American Collegeof Cardiology Methods Study patients. The study group comprised 70 patients with a clinical diagnosis of unstable angina who were treated with combination triple drug medical therapy in two doubleblind, randomized, placebo-controlled trials (8,9). The eligibility criteria for the study were as follows: I) All patients had unstable angina, defined as at least one episode of chest pain at rest with documented ischemic (ECG) changes con /87/$3.50

2 lacc Vol. 10, No 4 GOTTLIEB ET AL. 757 sisting of ST segment depression or elevation 2: I mm or T wave inversion of pseudonormalization, 2) All patients were hospitalized in the coronary care unit at The Johns opkins ospital or Francis Scott Key Medical Center, were placed at bed rest and treated with long-acting nitrates, propranolol (160 mg/day) and nifedipine (80 mg/day). There was no upper age exclusion from the study. Patients with symptomatic cerebrovascular disease, congenital heart disease, symptomatic congestive heart failure, a prior history of cardiac surgery or second degree or higher atrioventricular (AV) block were excluded, as were patients with left bundle branch block, left ventricular hypertrophy or other significant severe rest abnormalities of the ST segment that would interfere with the interpretation of the transient ST changes on olter monitoring. More than 95% of eligible patients were approached for participation in the study, and of these more than 80% agreed to participate and provided informed consent. The decision for revascularization with bypass surgery or coronary angioplasty was based on demonstrated failure of medical therapy, consisting of at least two episodes of recurrent rest angina with ischemic ECG changes, by prior agreement of the full-time staff responsible for patient care at both hospitals, All clinical decisions forthese patients were made without knowledge of the results of olter monitoring. The study group was thus representative of patients with rest angina who could be treated with combination medical therapy in the coronary care unit setting. Study design. Continuous ECG monitoring was performed for a 48 hour period in the coronary care unit after initiation of triple drug therapy, and the patients were followed up for a 2 year period for I) the occurrence of death, 2) myocardial infarction defined by prolonged chest pain with ECG changes and elevation of the creatine kinase level to at least twice normal, and 3) the need for revascularization with bypass surgery or coronary angioplasty for recurrent symptoms, The results of continuous olter monitoring were not available to the physicians making the clinical decisions, The olter recordings were performed using the two leads that had shown the most pronounced reversible ischemic changes on the initial 12 lead ECG obtained during pain, They were analyzed by two investigators who had no knowledge of other data for the number and duration of ischemic episodes defined by 2: I mm ST segment depression or elevation from baseline occurring 0,08 second after the J point, lasting for at least I minute. Each episode was correlated with symptoms in the coronary care unit, and thereby determined to be either symptomatic or silent. A normal interval of at least 2 minutes was required after the resolution of each episode before another discrete episode was counted. The system utilized for olter monitoring was an amplitudemodulated reel to reel two channel monitor system. Coronary angiography was performed in most (70%) but not all patients, as clinically indicated, and the coronary angiograms were analyzed without knowledge of other data by two observers using visual estimations of the number of major coronary arteries with 70 or 90% diameter stenoses. Ergonovine provocation testing was not performed in the presence of a stenosis of 2:70% diameter. Left ventricular ejection fraction was determined by either contrast left ventriculography or radionuclide ventriculography using the gated blood pool scan technique. Data analysis. Continuous variables were analyzed using unpaired t tests of significance, and categorical data were analyzed with chi-square tests as appropriate. Kaplan Meier analyses (10) were used to analyze the time-dependent cumulative probability of clinical outcomes, using the methods of Breslow (11). A multivariate Cox's hazard function analysis (12) was utilized to examine the predictive value of 15 variables for 2 year clinical outcome; these included age, sex, diabetes, hypertension, current cigarette smoking, prior myocardial infarction, chest pain during the first 2 study days, silent ischemia detected by olter monitoring, ST segment elevation during the index episode of pain, global ST changes during the index episode of pain, number of coronary arteries with 2:70% or 2:90% diameter obstruction, left anterior descending coronary disease with 2':90% diameter obstruction and left ventricular ejection fraction obtained by contrast ventriculography or gated blood pool scanning. Continuous independent variables such as age and ejection fraction were included in the multivariate analysis as continuous measures. The multivariate analysis examined the relative risk afforded by each variable for the occurrence Table 1. Two Year Adverse Clinical Outcomesfor Patients With and Without Silent ST Segment Changes on Initial 48 our olter Monitoring Adverse Clinical Outcome Cardiac death Nonfatal MI CABG or PTCA for symptoms Total Group I Silent Ischemia (n = 37) ~ }(27%) II (30%) 21 (57%) Group II No Silent Ischemia (n = 33) n(3%) 5 (15%) 6 (18%) p Value* <0.01 <0.05 <0.00\ *Derived from Kaplan-Meier analysis. Breslow test. CABG = coronary bypass surgery; MI = myocardial infarction; PTCA = percutaneous transluminal coronary angioplasty.

3 758 GOTTLIEB ET AL. lacc Vol. 10. No NO SILENT ISCEMIA 80 >- f- SO -J III «III 0 a: 40 a. 20 I,,'------, I ----I I ~ I 1L., IL. SILENT ISCEMIA Figure 1. Kaplan-Meier curves illustrating the probabilities of not experiencing death or myocardial in _ farction over the 2 year follow-up period for the 37 patients with (Group I) and the 33 patients without (GroupII) silent ischemic ST changes on initial olter monitoring. The difference between the two groups is significant at the p < 0.01 level f / t----j o S0 DAYS 4BO SOD 720 of the defined outcome. In a stepwise fashion it presented the order in which the variables added information that improved the ability to classify patients into predefined groups. The relative risks of an unfavorable outcome over the 2 year period were calculated. Data are presented as mean ± SO, or as frequency, when appropriate. Results The baseline clinical characteristics of the study group have been previously described (4). The comparison groups were as follows: 37 patients (Group I) had silent ischemic ST changes on olter monitoring and the other 33 (Group II) had no ischemic ST changes on olter monitoring. The Group I patients had an average of 5.5 episodes each over the 48 hour recording period; the median duration per episode was 20 minutes. Of the 37 patients in Group I, 8 had symptoms associated with at least one episode, and episodes of silent ST changes as well, whereas the other 29 patients in this group had only silent episodes of ST changes on olter monitoring. During the first 2 days, nine Group II patients had chest pain without accompanying ECG changes on olter monitoring. There were no significant baseline differences between the two groups with regard to cardiovascular risk factors, medical therapy and extent of fixed coronary artery disease at angiography; however, mean left ventricular ejection fraction at rest was 56 ± 14% in Group I patients versus 64 ± 13% for Group II (p < 0.05). Medical therapy was continued over the 2 year period in the patients who remained clinically stable and did not experience any of the adverse study outcomes. Two year clinical outcomes. The adverse clinical outcomes over the 2 year follow-up period are presented in Table I. Over the 2 year period, 10 patients in Group I (27%) had a myocardial infarction that was fatal in 2. In contrast, only one patient in Group II experienced a nonfatal myocardial infarction over the 2 year period (p < 0.01 for differences between the groups using Kaplan-Meier anal o BO ysis) (Fig. I). When the 8 patients who had both symptomatic and silent ischemia in Group I are excluded, there was a significantly higher risk of death or infarction over 2 years for the 29 Group I patients (p = 0.01) with only silent olter-detected ischemia than in the 33 Group II patients without olter-detected ST changes. Over the 2 year period, II patients in Group I and 5 in Group II required coronary bypass surgery or angioplasty for recurrent, medically refractory symptoms (p < 0.05 by Kaplan-Meier analysis). Thus, for the combined clinical outcomes of death, infarction or revascularization for recurrent symptoms as well, Figure 2. Relativerisk curvesderivedfrom Cox's hazard function analysis demonstrating the risk for death, myocardial infarctionor revascularization for recurrent symptoms and illustrating the effects of the silent ischemia (SI) and recurrent chest pain (CP) variables on adverse outcome. There is a fivefold relative risk for outcomes associated with silent ischemia and a threefold relative risk associated with chest pain during the initial 2 days. When both variables are present, the relative risk for experiencing an adverse outcome is increasedby a factor of O.B >- ~ J ld < ld ~ 0.4 a SI. +SI. L----l- ~+:;SlI. -cp L- '- ~-SI. +CP -CP +CP DAYS

4 JACC Vol. 10, No.4 October 1987: GOTTLIEB ET AL. 759 there was a significantly greater incidence in Group I (21 of 37 patients, 57%) as compared with Group II (6 of 33 patients, 18%) (p < 0,001 by Kaplan-Meier analysis using the Breslow test), The Cox's hazardfunction analysis (Fig. 2) revealed that silent olter-detected ischemia was the most important predictor of an unfavorable outcome (death, infarction or need for revascularization) over the 2 year follow-up period (p < 0.001); the next most important variable was chest pain during the initial 2 study days (p < 0.0 I). No other variable was found to be a significant predictor of adverse 2 year outcome in the presence of these two variables. The relative risk curves for an unfavorable outcome was 5 in the presence of silent ischemia and 3 for chest pain; it increased to 15 when both variables are present (Fig. 2). Discussion Patients with unstable angina pectoris have a high incidence of silent myocardial ischemia as detected by continuous olter monitoring, even in the presence ofcombination medical therapy that is sufficient in most cases to control symptoms (4-6,13). Our previous report (4) demonstrated that the presence and duration of silent ischemia are significantly correlated with early (I month) unfavorable outcome of infarction or recurrent symptoms requiring revascularization. In the present report, 2 year follow-up of the same study patients reveals that silent olter-detected ischemia is significantly associated with an increased risk of death or recurrent infarction over a 2 year period. In the majority of patients, the adverse clinical outcome occurred within the first several months. These observations are consistent with the hypothesis that the development of unstable angina is related to endothelial injury or disruption due to a ruptured atherosclerotic plaque with superimposed vasoconstriction or thrombus, or both, because the period of highest risk during the first few months approximates the time required for endothelial healing (14-16). Recent studies by Ludmer et al. (17) demonstrated an abnormal vasoconstrictive response to acetylcholine (a potent vasodilator in normal coronary arteries) in atherosclerotic coronary arteries, suggesting that endothelial dysfunction may be a mediator of inappropriate vasoconstriction. Other studies (18,19) also identified muscarinic cholinergic defects in experimental and human atherosclerosis. Clinical implications. This study suggests that the use of olter ECG monitoring for ST segment changes can be used with other currently utilized data-including the initial symptomatic response to therapy, the extent of myocardium threatened by high grade coronary obstructions, left ventricular function and results of low level exercise treadmill testing-to provide additional risk stratification information (20,21). It is difficult to directly compare the contributions of ischemic ECG changes detected by olter monitoring and stress testing in this group, because many patients with unstable angina experience an early adverse outcome before stress testing can be performed. One advantage of olter monitoring is that it can be performed at any time in the course of hospitalization. In addition, most of the episodes of ischemic ST changes detected in these patients occur at a low heart rate (9), suggesting that a primary reduction in coronary blood flow rather than a response to increased oxygen demands may play an important pathophysiologic role in olter-detected ischemia. Thus, olter monitoring and exercise stress testing may complement each other by providing insight into both the amount of dynamic disease activity (that is, vasoconstriction and platelet aggregation and thrombosis) and the amount of fixed atherosclerotic coronary disease as indicated by demand-related ischemia. These findings suggest that the prognostic information provided by olter monitoring for ischemic ST changes should be incorporated into clinical management of patients with unstable angina. Cardiac catheterization and early coronary revascularization (22,23) are clearly indicated for patients who do not respond to intensive medical therapy and continue to experience chest pain after treatment with nitrates (24), beta-receptor blockers (9), calcium channel antagonists (8), anticoagulants (25) and antiplatelet agents (26,27). owever, when symptoms of chest pain are controlled with medical treatment it is difficult to identify which patients may exhibit continued stability without further interventional management. Use of the information obtained from both olter monitoring and low level exercise testing (21) together with the anatomic information obtained from cardiac catheterization may allow for a rational and tailored management approach for both higher and lower risk patient subgroups. References I. Deanfield J, Maseri A, Selwyn AP, et al. Myocardial ischaemia during daily life in patients with stable angina: its relation to symptoms and heart rate changes. Lancet 1983;2: Deanfield le, Shea M, Rihiero P, et al. Transient ST-depression as a marker of myocardial ischemia during daily life. Am J Cardiol 1984:54: Stem S, Tsivoni D. Early detection of silent ischaemic heart disease by 24-hour electrocardiographic monitoring of active subjects. Br eart J 1974;36: Gottlieb SO, Weisfeldt ML, Ouyang P, Mellits ED, Gerstenblith G. Silent ischemia as a marker for early unfavorable outcomes in patients with unstable angina. N Engl J Med 1986;314: Johnson SM, Mauritson DR, Winniford MD, et al. Continuous electrocardiographic monitoring in patients with unstable angina pectoris: identification of high-risk subgroup with severe coronary disease, variant angina and/or impaired early prognosis. Am eart J 1982;103: Biagini A, Mazzei MG, Carpeggiani C. et ai. Vasospastic ischemic mechanism of frequent asymptomatic transient ST-T changes during

5 760 GOTTLIEB ET AL JACC Vol. 10. No.4 continuous electrocardiographic monitoring in selected unstable angina patients. Am eart J 1982;103: Deanfield JE, Ribiero P, Oakley CM, Krikler S, Selwyn AP. Critical analysis of ST segment changes in normal subjects: implications for ambulatory monitoring in angina pectoris. Am 1 Cardiel 1984:54: Gerstenblith G, Ouyang P, Achuff SC, et al. Nifedipine in unstable angina: a double-blind, randomized trial. N Engl J Med 1982:306: Gottlieb SO, Weisfeldt ML, Ouyang P. et al. Effect of the addition of propranolol to therapy with nifedipine for unstable angina pectoris: a randomized, double-blind, placebo-controlled trial. Circulation 1986:73: Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc 1958;53: II. Breslow N. A generalized Kruskal- Wallis test for comparing k samples subject to unequal patterns of censorship. Biometrika 1970;57: Kalbfleisch JD, Prentice RL. The Statistical Analysis of Failure Time Data. New York: John Wiley, 1980:70-117; Nadernanee K, Intarachot V, Singh PN, Josephson MA, Singh BN. Characteristics and clinical significance of silent myocardial ischemia in unstable angina. Am J Cardiol 1986;58:26B-33B. 14. Maseri A. Patbogenetic classification of unstable angina as a guideline to individual patient management and prognosis. Am J Med 1986;80 (suppl 4C): Sherman T, Litvack F, Grundfest W. et al. Coronary angioscopy in patients with unstable angina pectoris, N Engl J Med 1986;315: Ross R, Glomset JA. The pathogenesis of atherosclerosis. N Engl J Med 1976:295:369-77, Ludmer PL, Selwyn AP. Shook TL, et al. Paradoxical vasoconstriction induced by acetylcholine in atherosclerotic coronary arteries. N Engl J Med 1986:315: orio Y, Yasue, Rokutanda M, et al. Effects of intracoronary injection of acetylcholine on coronary artery diameter. Am 1 Cardiel 1986;57: Bossaller C, abib GB, Yamamoto, Williams C, Wells S, enry PD. Impaired muscarinic endothelium-dependent relaxation and cyclic guanosine 5' -monophosphate formation in atherosclerotic human coronary artery and rabbit aorta. J Clin Invest 1987;79: Ouyang P, Brinker JA, Mellits ED, Weisfeldt ML, Gerstenblith G. Variables predictive of successful medical therapy in patients with unstable angina: selection by multivariate analysis from clinical, electrocardiographic and angiographic evaluations. Circulation 1984;70: Butman SM, Olson G. Gardin 1M. Piters KM. ullett M, Butman LK. Submaximal exercise testing after stabilization of unstable angina pectoris. J Am Coli Cardiol 1984;4: Rahimtoola S, Nunley D. Grunkerneier G, Tepley 1. Lambert L, Starr A. Ten-year survival after coronary bypass surgery for unstable angina. N Engl J Med 1983;69: DeFeyter PJ, Serruys PW, Van Den Brand M, et al. Emergency coronary angioplasty in refractory unstable angina. N Engl J Med 1985:313: Lin S-G, Flaherty J'T. Crossover from intravenous to trans-dermal nitroglycerin therapy in unstable angina pectoris. Am J Cardiol 1985;56: Telford AM, Wilson C. Trial of heparin versus atenolol in prevention of myocardial infarction in intermediate syndrome. Lancet 1981: I: Lewis D Jr, Davis JW, Archibald DG, et al. Protective effects of aspirin against acute myocardial infarction and death in men with unstable angina. Results of a Veterans Administration Cooperative Study. N Engl J Med 1983;309: Cairns JA, Gent M, Singer J, et al. Aspirin, sulfinpyrazone, or both, in unstable angina: results of a Canadian Multicenter Trial. N Engl J Med 1985:313: