Stroke in Patients With Acute Coronary Syndromes
Incidence and Outcomes in the Platelet Glycoprotein IIb/IIIa in Unstable Angina: Receptor Suppression Using Integrilin Therapy (PURSUIT) Trial
Background—The incidence of stroke in patients with acute coronary syndromes has not been clearly defined because few trials in this patient population have been large enough to provide stable estimates of stroke rates.
Methods and Results—We studied the 10 948 patients with acute coronary syndromes without persistent ST-segment elevation who were randomly assigned to placebo or the platelet glycoprotein IIb/IIIa receptor inhibitor eptifibatide in the Platelet Glycoprotein IIb/IIIa in Unstable Angina: Receptor Suppression Using Integrilin Therapy (PURSUIT) trial to determine stroke rates, stroke types, clinical outcomes in patients with stroke, and independent baseline clinical predictors for nonhemorrhagic stroke. Stroke occurred in 79 (0.7%) patients, with 66 (0.6%) nonhemorrhagic, 6 intracranial hemorrhages, 3 cerebral infarctions with hemorrhagic conversion, and 4 of uncertain cause. There were no differences in stroke rates between patients who received placebo and those assigned high-dose eptifibatide (odds ratios and 95% confidence intervals 0.82 [0.59, 1.14] and 0.70 [0.49, 0.99], respectively). Of the 79 patients with stroke, 17 (22%) died within 30 days, and another 26 (32%) were disabled by hospital discharge or 30 days, whichever came first. Higher heart rate was the most important baseline clinical predictor of nonhemorrhagic stroke, followed by older age, prior anterior myocardial infarction, prior stroke or transient ischemic attack, and diabetes mellitus. These factors were used to develop a simple scoring nomogram that can predict the risk of nonhemorrhagic stroke.
Conclusions—Stroke was an uncommon event in patients with acute coronary syndromes in the PURSUIT trial. These strokes are, however, associated with substantial morbidity and mortality rates. The majority of strokes were of nonhemorrhagic causes. Eptifibatide was not associated with an increase in intracranial hemorrhage, and no significant effect on nonhemorrhagic stroke was observed. We developed a useful nomogram for assigning baseline nonhemorrhagic stroke risk in this patient population.
The incidence of stroke in patients with acute myocardial infarction treated with and those without thrombolytic therapy has been well established.1 2 3 4 5 6 7 8 9 10 11 12 13 The incidence of stroke in patients with acute coronary syndromes without persistent ST-segment elevation is less clearly defined, as few trials in this patient population have been large enough to provide a stable estimate of stroke rates given the relative infrequency of this adverse event. In the 2 largest trials of this population to date, stroke occurred in 0.8% of patients with acute coronary syndromes not treated with thrombolysis.14 15 The majority of these strokes were nonhemorrhagic. Risk factors for stroke in this patient population have not been previously analyzed.
The recently completed Platelet Glycoprotein IIb/IIIa in Unstable Angina: Receptor Suppression Using Integrilin Therapy (PURSUIT) trial is the largest trial to date of patients with acute coronary syndromes without persistent ST-segment elevation.16 We prospectively collected information about patients with suspected stroke to determine the incidence of stroke, stroke types, outcomes in patients with stroke, and the independent baseline clinical and demographic risk factors for nonhemorrhagic stroke in this patient population.
The PURSUIT trial enrolled 10 948 patients in 726 hospitals in 27 countries. Enrollment criteria have been published16 : In brief, patients with symptoms of ischemic chest pain at rest lasting 10 minutes or longer within the previous 24 hours that was accompanied by either transient ST-segment elevation >0.5 mm, transient or persistent ST-segment depression >0.5 mm, or T-wave inversion >1 mm within 12 hours of the chest pain; or had a creatine kinase myocardial enzymes fraction above the upper limit of normal for that hospital. Exclusion criteria included persistent ST-segment elevation >1 mm; evidence of bleeding; systolic blood pressure >200 mm Hg or diastolic blood pressure >110 mm Hg; major surgery within 6 weeks; a nonhemorrhagic stroke within 30 days or any history of hemorrhagic stroke; renal failure; or treatment with thrombolytic therapy within 24 hours.
Patients were randomly assigned in double-blind fashion to an intravenous bolus and infusion of placebo, 180 mm/kg bolus plus infusion of 1.3 μg/kg per minute of eptifibatide, or 180 μg/kg bolus plus infusion of 2.0 μg/kg per minute of eptifibatide. Per prespecified plan, after 3218 patients were randomized, an independent Data and Safety Monitoring Committee reviewed the safety data. The committee recommended dropping the lower dose because the high dose had an acceptable safety profile.
A study drug was to be infused until hospital discharge or up to 72 hours, whichever came first (or up to 96 hours in patients undergoing percutaneous intervention at 72 hours). All patients were to receive daily aspirin, and heparin was recommended but not required. Intravenous heparin was to be given as a 5000 U bolus and 1000 U/h infusion with adjustment to maintain activated partial thromboplastin time in the 50 to 70 seconds range. Lower doses were recommended for patients who weighed <70 kg. Other concomitant medications, diagnostic cardiac procedures, or percutaneous or surgical interventions were at the discretion of the treating physician.
Stroke was defined as an acute new neurological deficit resulting in death or lasting for >24 hours. A Stroke Adjudication Committee (see Appendix) was established to independently adjudicate and categorize all suspected strokes. Patients with suspected strokes were identified from the Case Report Form.
The evaluation of patients with stroke was prospectively planned. The protocol specified that all patients with a new neurological deficit undergo complete evaluation including brain imaging. Clinical notes, discharge summaries, neurological or neurosurgical consultation notes, autopsy reports, and results of computed tomographic or magnetic resonance imaging (MRI) studies were collected on all patients with suspected stroke. The committee reviewed all available medical records and determined by consensus if a stroke did or did not occur and if so, categorized the type of stroke. The Stroke Adjudication Committee members were blinded to treatment assignment.
Strokes were divided into 4 main categories: primary hemorrhagic, nonhemorrhagic, hemorrhagic conversion of infarct, and uncertain. Primary hemorrhagic was diagnosed if a focal collection of intracranial blood was seen on brain imaging or at autopsy and was not believed to represent hemorrhagic conversion. Nonhemorrhagic stroke was categorized if there was a low-density lesion on computed tomography, high-intensity lesion on MRI, or clinical evidence of a stroke and no focal collection of intracranial hemorrhage on brain imaging studies or at autopsy. Hemorrhagic conversion was diagnosed if blood was present within an area of cerebral infarction, but the event was not thought to represent a primary hemorrhagic stroke. The uncertain classification was used if there was clinical evidence of a stroke and no brain imaging or autopsy data were available to determine the type of stroke.
Patient Functional Assessment
Functional status was determined for all patients with stroke at hospital discharge or 30 days, whichever came first. Patients were classified by site investigators as not disabled if they had no deficit (no sequelae) or minor deficits (functional status unchanged) and as disabled if they had moderate deficits (significant limitations of activity) or severe deficits (unable to live independently or return to work). The correlation of this classification of disability and direct patient interviews in stroke survivors has been shown to be excellent.9
Continuous variables are shown as medians with 25th and 75th percentiles. Discrete variables are shown as frequencies and percentages. Stroke data for all 10 948 patients are included. Analyses that include comparisons between treatment groups only include the 9461 patients assigned placebo or high-dose eptifibatide because the low-dose eptifibatide group was not contemporaneous with the other 2 groups as the result of discontinuation of the low-dose arm after the interim analysis by the Data Safety and Monitoring Board (see Methods). Logistic regression modeling was used to determine the univariable predictors of nonhemorrhagic stroke and the multivariable baseline risk factors for nonhemorrhagic stroke. Predictors were tested with the use of the Wald or likelihood χ2 test. Results are presented as odds ratios and 95% confidence intervals.
A scoring nomogram was created from the coefficients from the baseline multivariable regression modeling. Each independent predictor was assigned a score according to its predictive value. The sum of the scores indicates the probability of a nonhemorrhagic stroke, based on baseline predictors for individual patients.17
A total of 79 (0.7%) strokes occurred within 30 days in the 10 948 patients enrolled in the PURSUIT trial. Table 1⇓ shows the type of strokes by treatment assignment. Nonhemorrhagic stroke was the most common stroke type. Specifically, nonhemorrhagic stroke was observed in 28 (0.6%) high-dose eptifibatide patients and in 33 (0.7%) placebo patients. The median (25th, 75th) time to onset of nonhemorrhagic stroke symptoms was 6.5 (3, 12) days (Figure 1⇓). The clinical outcomes for patients with stroke are shown in Table 2⇓. Overall, 54% of patients with stroke died or were disabled at the time of hospital discharge or 30-day follow-up, whichever came first.
Table 3⇓ shows the baseline clinical characteristics for patients with and those without stroke and for patients with nonhemorrhagic stroke. Patients with stroke were older and more often female, had lower body weight and higher heart rate, and more frequently had prior myocardial infarction, diabetes mellitus, history of hypertension, hypercholesterolemia, coronary artery bypass surgery, and history of stroke or transient ischemic attack. Patients with stroke more commonly experienced adverse in-hospital events including hypotension, atrial fibrillation, congestive heart failure, cardiogenic shock, and coronary artery bypass surgery (Table 4⇓).
Table 5⇓ shows the univariable baseline factors associated with nonhemorrhagic stroke. Five independent baseline clinical and demographic predictors of stroke occurring by 30-day follow-up were identified (Table 6⇓). Higher heart rate was the most important independent baseline clinical predictor; older age, previous anterior myocardial infarction, prior stroke or transient ischemic attack, and history of diabetes mellitus were the other statistically significant predictors. Other possible but not statistically significant predictors were prior percutaneous intervention (P=0.09), history of aspirin use (P=0.06), and history of hypercholesterolemia (P=0.09). The predicted probability in individual patients for nonhemorrhagic stroke within 30-day follow-up can be calculated by using the nomogram in Figure 2⇓.
In the PURSUIT trial of almost 11 000 patients, the overall stroke rate through 30-day follow-up was 0.7%. The majority of these strokes were not hemorrhagic. We identified 5 easily determined independent baseline clinical and demographic risk factors for nonhemorrhagic stroke in the PURSUIT population (higher heart rate, older age, prior anterior myocardial infarction, prior stroke or transient ischemic attack, and diabetes mellitus).
In patients with ST-segment elevation, acute myocardial infarction, the incidence of stroke in the era before the routine use of thrombolytic therapy was 1.7% to 3.2%.1 2 3 4 5 In the thrombolytic era, the incidence of stroke has decreased, but there is a higher proportion of intracranial hemorrhage (0.07% to 1.5%) and a lower rate of nonhemorrhagic stroke (0.1% to 1.3%).6 7 8 9 10 11 12 The rates observed for nonhemorrhagic stroke in the current study were similar to those for patients with acute myocardial infarction treated with thrombolysis and, as would be expected, the incidence of intracranial hemorrhage was lower. The observed rate of intracranial hemorrhage was 0.05% (6/10948) without evidence of an increase in the eptifibatide groups compared with placebo (4 patients vs 2 patients). These data support the safety of eptifibatide regarding intracranial hemorrhage.
The Global Use of Strategies to Open Occluded Arteries in acute coronary syndromes (GUSTO IIb) trial14 was the largest trial of patients with acute coronary syndromes prior to PURSUIT. In GUSTO IIb, 8011 patients with acute coronary syndromes without persistent ST-segment elevation were randomly assigned to intravenous heparin or the novel antithrombin, hirudin. All patients were to be treated with aspirin. The overall incidence of stroke was 0.8%. Nonhemorrhagic stroke occurred in 0.5% of patients and 0.09% of patients had a primary hemorrhagic stroke.14 No difference in stroke rates was observed by treatment assignment. These data are similar to the stroke rates in PURSUIT, in which patients were to be treated with heparin and aspirin and randomly assigned to the glycoprotein IIb/IIIa receptor inhibitor eptifibatide or to placebo. The patients in PURSUIT were more often female (35% vs 33%), younger (64 vs 66 years), heavier (78 vs 76 kg), and had lower enrollment systolic blood pressure (130 vs 139 mm Hg), lower heart rate (72 vs 74 bpm) and were more likely to have a history of diabetes (23% vs 19%), hypertension (55% vs 48), or previous coronary intervention (13% vs 12%) compared with the GUSTO IIb population. However, the impact of these factors on stroke rates in the 2 trials is unknown.
No previous studies have determined the risk factors for stroke in the unstable angina population to use for comparison with our results. In the GUSTO I trial of >41 000 patients treated with thrombolysis for acute myocardial infarction, the baseline independent risk factors for nonhemorrhagic stroke were similar to those identified for the PURSUIT population.13 In GUSTO I, as in PURSUIT, older age, higher heart rate, history of stroke or transient ischemic attack, and diabetes mellitus were independent baseline predictors of stroke.18 Prior anterior myocardial infarction was not a risk factor in the GUSTO I stroke prediction model, but prior angina was an independent predictor. A history of hypertension was also an independent predictor in the GUSTO I population, but it was not in the PURSUIT nonhemorrhagic stroke model, although it did achieve borderline statistical significance in univariable analysis. The reason higher heart rate is an independent predictor for nonhemorrhagic stroke is not clear from these data, but heart rate may correlate with larger infarctions that predispose patients to a higher likelihood of atrial arrhythmia and left ventricular thrombi.
The clinical outcomes of stroke patients with unstable angina/non–Q-wave myocardial infarction have not been previously reported. In PURSUIT, stroke was associated with substantial morbidity and mortality rates. In GUSTO I, the percentage of nonhemorrhagic stroke patients who died or were disabled was comparable with that in the PURSUIT population (57% vs 54%).9
The incidence of nonhemorrhagic stroke in patients assigned high-dose eptifibatide compared with placebo was similar (0.6% vs 0.7%, respectively; P=0.53). Treatment assignment was not a statistically significant predictor of nonhemorrhagic stroke in the univariable or multivariable analyses. The overall number of strokes was small; therefore definitive conclusions cannot be made about the potential risks or benefits of glycoprotein IIb/IIIa receptor blockade on the incidence of nonhemorrhagic stroke from these data. A lower incidence of nonhemorrhagic stroke in the eptifibatide group would have been consistent with the known benefits of antiplatelet therapy in decreasing the incidence of stroke in patients with acute myocardial infarction or prior stroke as well as in other high-risk patient groups including unstable angina.19 Ongoing studies of prolonged oral glycoprotein IIb/IIIa receptor antagonists in patients with acute coronary syndromes and cerebrovascular disease will provide more definitive data about treatment benefits with such agents in these patient populations.
Our study has several limitations. The baseline predictors of stroke may not account for all the risk for stroke in this patient population. Fifty percent of nonhemorrhagic strokes occurred >6.5 days after enrollment, and we were not able to account for in-hospital procedures or events that may be associated with stroke, primarily because of the relatively small sample size and limited regression modeling that could be performed. There was no systematic assessment of the cardiac rhythm at the time of study enrollment; therefore atrial fibrillation was not included in the prediction model. However, only 159 (1.5%) patients had atrial fibrillation reported on the Case Report Form before or within 4 hours of study enrollment, and only 2 of these patients had nonhemorrhagic stroke.
The mechanism of the strokes, particularly the nonhemorrhagic strokes, was not assessed. However, systematic review of all suspected strokes by the Stroke Adjudication Committee, which included analysis of computed axial tomographic data, MRI data, or autopsy results in 80% of patients, allowed determination of stroke types in a majority of cases, and the number of “unknowns” was small.
The logistic regression model and associated scoring nomogram were developed with the use of only 64 nonhemorrhagic strokes and therefore require validation in a larger patient population. However, the similarity to the prediction model from the GUSTO-I trial is supportive. Finally, these results are only applicable to patients with acute coronary syndromes without persistent ST-segment elevation treated with heparin and aspirin and should not be generalized to all patients with acute coronary syndromes.
Stroke was an uncommon event in patients with acute coronary syndromes without persistent ST-segment elevation treated with antiplatelet and antithrombin therapy in the PURSUIT trial. The majority of strokes were of nonhemorrhagic causes. Despite the low incidence of stroke, its prevalence is substantial because more than 1 million patients present each year with non–ST-segment elevation acute coronary syndromes to hospitals in the United States, with similar numbers in Western Europe. Therefore >7000 patients with acute coronary syndromes in the United States alone have a stroke, and these strokes are associated with significant morbidity and mortality rates. The use of eptifibatide, a glycoprotein IIb/IIIa receptor antagonist, was not associated with an increased incidence of intracranial hemorrhage, and we observed no significant effect on the occurrence of nonhemorrhagic stroke. Higher heart rate, older age, prior stroke or transient ischemic attack, prior anterior myocardial infarction, and diabetes mellitus were independent baseline predictors of nonhemorrhagic stroke. Physicians can use these factors to determine the probability of nonhemorrhagic stroke with a simple scoring nomogram (Figure 2⇑). The impact of stroke risk assessment on patient management requires further study.
Stroke Adjudication Committee
Cardiologists were Brian S. Crenshaw, Christopher B. Granger, Robert A. Harrington, Kenneth W. Mahaffey, and Julie M. Miller. Neurologists were Mark J. Alberts, Carmelo Graffagnino, and Daniel Laskowitz.
The authors thank John Daniel for his help with manuscript preparation and editorial support.
- Received September 14, 1998.
- Revision received February 3, 1999.
- Accepted February 16, 1999.
- Copyright © 1999 by American Heart Association
Thompson PL, Robinson JS. Stroke after acute myocardial infarction: relation to infarct size. BMJ. 1978;2:457–459.
Komrad MS, Coffey CE, Coffey KS, McKinnis R, Massey EW, Califf RM. Myocardial infarction and stroke. Neurology. 1984;34:1403–1409.
Report of the Working Party on Anticoagulant Therapy in Coronary Thrombosis to the Medical Research Council. Assessment of short-anticoagulant administration after cardiac infarction. BMJ. 1969;1:335–342.
Fibrinolytic Therapy Trialists’ (FTT) Collaborative Group. Indications for fibrinolytic therapy in suspected acute myocardial infarction: collaborative overview of early mortality and major morbidity results from all randomised trials of more than 1000 patients. Lancet. 1994;343:311–322.
Maggioni AP, Franzosi MG, Santoro E, White H, Van de Werf F, Tognoni G, for the Gruppo Italiano per lo Studio della Sopravvivenza nell’Infarto Miocardico (GISSI-2) and the International Study Group. The risk of stroke in patients with acute myocardial infarction after thrombolytic and antithrombotic treatment. N Engl J Med. 1992;327:1–6.
Gore JM, Sloan M, Price TR, Randall AM, Bovill E, Collen D, Forman S, Knatterud GL, Sopko G, Terrin ML. Intracerebral hemorrhage, cerebral infarction, and subdural hematoma after acute myocardial infarction and thrombolytic therapy in the Thrombolysis in Myocardial Infarction Study: Thrombolysis in Myocardial Infarction, Phase II, pilot and clinical trial. Circulation. 1991;83:448–459.
Gore JM, Granger CB, Simoons ML, Sloan MA, Weaver WD, White HD, Barbash GI, Van de Werf F, Aylward PE, Topol EJ, for the GUSTO-I Investigators. Stroke after thrombolysis: mortality and functional outcomes in the GUSTO-I trial. Circulation. 1995;92:2811–2818.
Maggioni AP, Franzosi MG, Farina ML, Santoro E, Celani MG, Ricci S, Tognoni G. Cerebrovascular events after myocardial infarction: analysis of the GISSI trial. Gruppo Italiano per lo Studio della Streptochinasi nell’Infarto Miocardico (GISSI). BMJ. 1991;302:1428–1431.
Longstreth WT Jr, Litwin PE, Weaver WD, for the MITI Project Group. Myocardial infarction, thrombolytic therapy, and stroke: a community-based study. Stroke. 1993;24:587–590.
The TIMI IIIB Investigators. Effects of tissue plasminogen activator and a comparison of early invasive and conservative strategies in unstable angina and non–Q-wave myocardial infarction: results of the TIMI IIIB trial. Circulation. 1994;89:1545–1556.
The Platelet Glycoprotein IIb/IIIa in Unstable Angina. Receptor Suppression Using Integrilin Therapy (PURSUIT) Trial Investigators. Inhibition of platelet glycoprotein IIb/IIIa with eptifibatide in patients with acute coronary syndromes without persistent ST-segment elevation: a randomized, placebo-controlled, clinical trial. N Engl J Med.. 1998;339:436–443.
Harrell FE. Design: S plus functions for biostatistical/epidemiologic modeling, testing, estimation, validation, graphics, prediction, and typesetting by storing enhanced model design attributes in the fit. UNIX version available from email@example.com; 1996.
Mahaffey KW, Granger CB, Sloan MA, Thompson T, Gore JM, Weaver WD, White HD, Simoons ML, Barbash GI, Topol EJ, Califf RM, for the GUSTO-I Investigators. Risk factors for in-hospital nonhemorrhagic stroke in patients with acute myocardial infarction treated with thrombolysis: results from GUSTO-I. Circulation. 1998;97:757–764.
The Antiplatelet Trialists’ Collaboration. Collaborative overview of randomised trials of antiplatelet therapy, I: prevention of death, myocardial infarction, and stroke by prolonged antiplatelet therapy in various categories of patients. BMJ. 1994;308:81–106.