Original Paper Received: September 16, 2013 Accepted: December 9, 2013 Published online: January 30, 2014
Cerebrovasc Dis 2014;37:147–152 DOI: 10.1159/000357799
Myocardial Infarction as a Complication in Acute Stroke: Results from the Austrian Stroke Unit Registry Thomas Gattringer a Kurt Niederkorn a Leonhard Seyfang b Thomas Seifert-Held a Nicole Simmet a Julia Ferrari c Wilfried Lang c Michael Brainin b, d Johann Willeit e Franz Fazekas a Christian Enzinger a
a
Department of Neurology, Medical University of Graz, Graz, b Center of Clinical Neurosciences, Danube University Krems, Krems, c Department of Neurology, Hospital Barmherzige Brueder Vienna, Vienna, d Department of Neurology, Danube Clinic Tulln, Tulln, and e Department of Neurology, Medical University of Innsbruck, Innsbruck, Austria
Key Words Ischemic stroke · Hemorrhagic stroke · Transient ischemic attack · Stroke unit · Myocardial infarction · Complication
Abstract Background: Patients with transient ischemic attack (TIA) and stroke have an increased risk for subsequent cardiac events including myocardial infarction (MI), which might be associated with a worse clinical outcome. Rapid identification of stroke patients at higher risk for MI might foster intensified cardiac monitoring or certain therapeutic strategies. However, information regarding acute MI as a complication of stroke in the very acute phase is limited. Moreover, there are no systematic data on the occurrence of MI following intracerebral hematoma. We thus aimed to assess the frequency, clinical characteristics and short-term outcome of patients suffering from acute MI in the stroke unit setting. Methods: We analyzed 46,603 patients from 32 Austrian stroke units enrolled in the prospective Austrian Stroke Unit Registry because of TIA/ acute stroke over a 6-year period (January 1, 2007 to January 13, 2013). A total of 41,619 patients (89.3%) had been treated for TIA/ischemic stroke and 4,984 (10.7%) for primary intracerebral hemorrhage (ICH). Acute MI was defined according to clinical evaluation, ECG findings and laboratory assessments.
© 2014 S. Karger AG, Basel 1015–9770/14/0372–0147$39.50/0 E-Mail [email protected] www.karger.com/ced
Patients with evidence for MI preceding the cerebrovascular event were not considered. Results: Overall, 421 patients (1%) with TIA/ischemic stroke and 17 patients (0.3%) with ICH suffered from MI during stroke unit treatment for a median duration of 3 days. Patients with TIA/ischemic stroke and MI were significantly older, clinically more severely affected and had more frequently vascular risk factors, atrial fibrillation and previous MI. Total anterior circulation and left hemispheric stroke syndromes were more often observed in MI patients. Patients with MI not only suffered from worse short-term outcome including a higher mortality (14.5 vs. 2%; p < 0.001) at stroke unit discharge, but also acquired more stroke complications like progressive stroke and pneumonia. Multivariate analyses identified previous MI and stroke severity at admission (according to the National Institutes of Health and Stroke Scale score) as factors independently associated with the occurrence of MI on the stroke unit. Conclusions: While quite rare in the acute phase after stroke, MI is associated with a poor short-term outcome including a higher mortality. Patients with previous MI and severe stroke syndromes appear to be at particular risk for MI as an early complication in the stroke unit setting. Further studies are needed to determine whether increased vigilance and prolonged (cardiac) monitoring or certain therapeutic approaches could improve the outcome in these high-risk patients. © 2014 S. Karger AG, Basel
Dr. Thomas Gattringer Department of Neurology, Medical University of Graz Auenbruggerplatz 22, AT–8036 Graz (Austria) E-Mail thomas.gattringer @ medunigraz.at
Downloaded by: St. Andrews University 138.251.14.35 - 12/3/2014 11:55:42 PM
Cardiac events including myocardial infarction (MI) are important complications in stroke, being associated with higher mortality and worse functional outcome [1, 2]. While stroke and MI share some common vascular risk factors, acute vascular damage to specific brain areas has also been implicated in cardiac dysfunction and myocardial damage [3, 4]. There is strong evidence for an increased risk of subsequent MI and cardiac death in the subacute period after a cerebrovascular event (CVE) [1, 2], but information regarding MI as an early complication of acute stroke has been limited to small patient series so far [5, 6]. However, such knowledge could have important clinical implications as rapid identification of stroke patients at higher risk for a subsequent MI might foster intensified cardiac monitoring or certain therapeutic strategies. Moreover, to the best of our knowledge there are no existing data on the occurrence of MI subsequent to an intracerebral hemorrhage (ICH), a clinical situation that poses a special therapeutic dilemma. We thus sought to assess the frequency of and risk factors for MI following acute stroke within the large nationwide Austrian Stroke Unit registry.
Methods Since 2003, 32 of the 34 Austrian stroke units have been prospectively collecting data on the demographic and clinical characteristics of all admitted patients above the age of 18 years, including information on their acute management, complications and outcome. The registry is part of a governmental quality assessment program for stroke care in Austria. Anonymized data are centrally administered and scientific analyses have to be approved and supervised by an academic review board. Details of the registry have been reported previously [7]. Formal approval for each data analysis by a local ethics committee is not needed [8]. To ensure high data quality, immediate data entry is obligatory, and the web-based database includes online plausibility checks and help icons for documentation issues. Biannual educational meetings of all stroke unit neurologists also serve to guarantee uniform documentation. In 2006, the registry was enriched by documentation of acute stroke complications including MI. In line with previous evaluations in the acute stroke setting, standard diagnostic criteria were recommended for assessment of clinical complications [7, 9]. Acute MI was defined according to clinical evaluation, ECG findings (according to Minnesota coding [10]) and laboratory assessments (cardiac enzymes including troponins). Patients with evidence of MI preceding the cerebrovascular index event were not considered. The registry did not consider a distinction between ST- and non-ST-elevation MI, nor was it designed to capture clin-
148
Cerebrovasc Dis 2014;37:147–152 DOI: 10.1159/000357799
ical data concerning special aspects of MI treatment (especially newer generation antiplatelets, beta blockers, coronary angioplasty or bypass surgery). The present analysis was based on all patients with a clinically and/or neuroradiologically established diagnosis of ischemic stroke, transient ischemic attack (TIA, transient focal neurological deficit that is fully resolving and lasts no longer than 24 h) and ICH who were registered from January 1, 2007 to January 13, 2013. This comprised a total of 46,603 patients with acute stroke or TIA. Statistics All data were processed using the statistical environment R (version 2.15.2; R Foundation for Statistical Computing, Vienna, Austria) [11]. Comparisons between categorical variables were made with the χ2 test or Fisher’s exact test, where appropriate. Group comparisons of quantitative or ordinal variables were performed with the Wilcoxon rank sum test. Regression coefficients were tested using the t test. The significance level was set at p < 0.05. A multivariate logistic regression model (including demographic and clinical characteristics as well as level of preexisting disability) with the occurrence of MI as target variable was fitted. The model specification was done using the bidirectional stepwise feature selection procedure, optimizing the Bayesian Information Criterion. All variables considered for the model are listed in table 2. Stroke severity was categorized according to the National Institutes of Health and Stroke Scale (NIHSS) scores (0–3 minor stroke, 4–7 reference category, 8–11 moderate to severe stroke, 12–42 stroke highly suggestive of large artery occlusion). The multivariate model which was fitted using all cases with a complete data set (n = 41,198) is shown in table 3. It contains only those variables that have been identified as significantly and independently associated with the complication of MI. The McKelvey and Zavoina pseudo R2 of the model was 0.21.
Results
Over the 6-year study period, 17 (0.3%) of the 4,984 patients with primary ICH and 421 (1%) of the 41,619 patients with ischemic CVE (ischemic stroke: 383/32,233, 1.2%; TIA: 38/9,386, 0.4%) suffered from acute MI during stroke unit treatment. To assure sufficient patient numbers for statistical testing, predictors of MI as complication were calculated for ischemic CVE only. Demographics, clinical characteristics and risk factors for individuals with ischemic CVE dichotomized by the presence or absence of the complication of MI are shown in table 1. In univariate analyses, patients who experienced MI were older and had more severe NIHSS scores at admission. The occurrence of MI was also associated with unfavorable vascular risk factor profiles and more prevalent cardiac disease such as atrial fibrillation and previous MI. Hypercholesterolemia was more often noted in non-MI Gattringer et al.
Downloaded by: St. Andrews University 138.251.14.35 - 12/3/2014 11:55:42 PM
Introduction
Table 1. Demographics and clinical characteristics of MI versus non-MI patients
Table 2. The following variables were included in the stepwise variable selection procedure as potential explanatory variables
Variable
MI (n = 421)
Non-MI (n = 41,198)
p value
Variable
Levels
Median age, years IQR Female Preexisting disabilitya Median NIHSS at admission IQR TIA Hypertension Diabetes mellitus Hypercholesterolemia Peripheral artery disease Smoking Atrial fibrillation Previous MI Previous stroke Location of clinical symptoms Left hemispheric Right hemispheric Brainstem Cerebellar Stroke syndromeb LACS TACS PACS POCS Thrombolysis Stroke etiologyc Microangiopathy Cardiac embolism Macroangiopathy Unknown Complications Intracranial hemorrhage Progressive stroke Brain edema Recurrent stroke Pneumonia Prognosis at discharge Median NIHSS IQR Death mRS 3–5
79.5 71.1–84.6 217 (51.5) 75 (17.8) 9 4–16 38 (9) 363 (86.2) 120 (28.5) 217 (51.5) 43 (10.2) 79 (18.7) 181 (43) 240 (57) 104 (24.7)
73.6 63.6–81.9 19,365 (47) 5,362 (13) 4 1–8 9,348 (22.7) 32,961 (80) 10,358 (25.1) 22,811 (55.4) 3,003 (7.3) 7,289 (17.6) 11,078 (26.8) 3,744 (9.1) 9,679 (23.4)
Cerebrovasc Dis 2014;37:147–152 DOI: 10.1159/000357799
Myocardial Infarction as a Complication in Acute Stroke: Results from the Austrian Stroke Unit Registry Thomas Gattringer a Kurt Niederkorn a Leonhard Seyfang b Thomas Seifert-Held a Nicole Simmet a Julia Ferrari c Wilfried Lang c Michael Brainin b, d Johann Willeit e Franz Fazekas a Christian Enzinger a
a
Department of Neurology, Medical University of Graz, Graz, b Center of Clinical Neurosciences, Danube University Krems, Krems, c Department of Neurology, Hospital Barmherzige Brueder Vienna, Vienna, d Department of Neurology, Danube Clinic Tulln, Tulln, and e Department of Neurology, Medical University of Innsbruck, Innsbruck, Austria
Key Words Ischemic stroke · Hemorrhagic stroke · Transient ischemic attack · Stroke unit · Myocardial infarction · Complication
Abstract Background: Patients with transient ischemic attack (TIA) and stroke have an increased risk for subsequent cardiac events including myocardial infarction (MI), which might be associated with a worse clinical outcome. Rapid identification of stroke patients at higher risk for MI might foster intensified cardiac monitoring or certain therapeutic strategies. However, information regarding acute MI as a complication of stroke in the very acute phase is limited. Moreover, there are no systematic data on the occurrence of MI following intracerebral hematoma. We thus aimed to assess the frequency, clinical characteristics and short-term outcome of patients suffering from acute MI in the stroke unit setting. Methods: We analyzed 46,603 patients from 32 Austrian stroke units enrolled in the prospective Austrian Stroke Unit Registry because of TIA/ acute stroke over a 6-year period (January 1, 2007 to January 13, 2013). A total of 41,619 patients (89.3%) had been treated for TIA/ischemic stroke and 4,984 (10.7%) for primary intracerebral hemorrhage (ICH). Acute MI was defined according to clinical evaluation, ECG findings and laboratory assessments.
© 2014 S. Karger AG, Basel 1015–9770/14/0372–0147$39.50/0 E-Mail [email protected] www.karger.com/ced
Patients with evidence for MI preceding the cerebrovascular event were not considered. Results: Overall, 421 patients (1%) with TIA/ischemic stroke and 17 patients (0.3%) with ICH suffered from MI during stroke unit treatment for a median duration of 3 days. Patients with TIA/ischemic stroke and MI were significantly older, clinically more severely affected and had more frequently vascular risk factors, atrial fibrillation and previous MI. Total anterior circulation and left hemispheric stroke syndromes were more often observed in MI patients. Patients with MI not only suffered from worse short-term outcome including a higher mortality (14.5 vs. 2%; p < 0.001) at stroke unit discharge, but also acquired more stroke complications like progressive stroke and pneumonia. Multivariate analyses identified previous MI and stroke severity at admission (according to the National Institutes of Health and Stroke Scale score) as factors independently associated with the occurrence of MI on the stroke unit. Conclusions: While quite rare in the acute phase after stroke, MI is associated with a poor short-term outcome including a higher mortality. Patients with previous MI and severe stroke syndromes appear to be at particular risk for MI as an early complication in the stroke unit setting. Further studies are needed to determine whether increased vigilance and prolonged (cardiac) monitoring or certain therapeutic approaches could improve the outcome in these high-risk patients. © 2014 S. Karger AG, Basel
Dr. Thomas Gattringer Department of Neurology, Medical University of Graz Auenbruggerplatz 22, AT–8036 Graz (Austria) E-Mail thomas.gattringer @ medunigraz.at
Downloaded by: St. Andrews University 138.251.14.35 - 12/3/2014 11:55:42 PM
Cardiac events including myocardial infarction (MI) are important complications in stroke, being associated with higher mortality and worse functional outcome [1, 2]. While stroke and MI share some common vascular risk factors, acute vascular damage to specific brain areas has also been implicated in cardiac dysfunction and myocardial damage [3, 4]. There is strong evidence for an increased risk of subsequent MI and cardiac death in the subacute period after a cerebrovascular event (CVE) [1, 2], but information regarding MI as an early complication of acute stroke has been limited to small patient series so far [5, 6]. However, such knowledge could have important clinical implications as rapid identification of stroke patients at higher risk for a subsequent MI might foster intensified cardiac monitoring or certain therapeutic strategies. Moreover, to the best of our knowledge there are no existing data on the occurrence of MI subsequent to an intracerebral hemorrhage (ICH), a clinical situation that poses a special therapeutic dilemma. We thus sought to assess the frequency of and risk factors for MI following acute stroke within the large nationwide Austrian Stroke Unit registry.
Methods Since 2003, 32 of the 34 Austrian stroke units have been prospectively collecting data on the demographic and clinical characteristics of all admitted patients above the age of 18 years, including information on their acute management, complications and outcome. The registry is part of a governmental quality assessment program for stroke care in Austria. Anonymized data are centrally administered and scientific analyses have to be approved and supervised by an academic review board. Details of the registry have been reported previously [7]. Formal approval for each data analysis by a local ethics committee is not needed [8]. To ensure high data quality, immediate data entry is obligatory, and the web-based database includes online plausibility checks and help icons for documentation issues. Biannual educational meetings of all stroke unit neurologists also serve to guarantee uniform documentation. In 2006, the registry was enriched by documentation of acute stroke complications including MI. In line with previous evaluations in the acute stroke setting, standard diagnostic criteria were recommended for assessment of clinical complications [7, 9]. Acute MI was defined according to clinical evaluation, ECG findings (according to Minnesota coding [10]) and laboratory assessments (cardiac enzymes including troponins). Patients with evidence of MI preceding the cerebrovascular index event were not considered. The registry did not consider a distinction between ST- and non-ST-elevation MI, nor was it designed to capture clin-
148
Cerebrovasc Dis 2014;37:147–152 DOI: 10.1159/000357799
ical data concerning special aspects of MI treatment (especially newer generation antiplatelets, beta blockers, coronary angioplasty or bypass surgery). The present analysis was based on all patients with a clinically and/or neuroradiologically established diagnosis of ischemic stroke, transient ischemic attack (TIA, transient focal neurological deficit that is fully resolving and lasts no longer than 24 h) and ICH who were registered from January 1, 2007 to January 13, 2013. This comprised a total of 46,603 patients with acute stroke or TIA. Statistics All data were processed using the statistical environment R (version 2.15.2; R Foundation for Statistical Computing, Vienna, Austria) [11]. Comparisons between categorical variables were made with the χ2 test or Fisher’s exact test, where appropriate. Group comparisons of quantitative or ordinal variables were performed with the Wilcoxon rank sum test. Regression coefficients were tested using the t test. The significance level was set at p < 0.05. A multivariate logistic regression model (including demographic and clinical characteristics as well as level of preexisting disability) with the occurrence of MI as target variable was fitted. The model specification was done using the bidirectional stepwise feature selection procedure, optimizing the Bayesian Information Criterion. All variables considered for the model are listed in table 2. Stroke severity was categorized according to the National Institutes of Health and Stroke Scale (NIHSS) scores (0–3 minor stroke, 4–7 reference category, 8–11 moderate to severe stroke, 12–42 stroke highly suggestive of large artery occlusion). The multivariate model which was fitted using all cases with a complete data set (n = 41,198) is shown in table 3. It contains only those variables that have been identified as significantly and independently associated with the complication of MI. The McKelvey and Zavoina pseudo R2 of the model was 0.21.
Results
Over the 6-year study period, 17 (0.3%) of the 4,984 patients with primary ICH and 421 (1%) of the 41,619 patients with ischemic CVE (ischemic stroke: 383/32,233, 1.2%; TIA: 38/9,386, 0.4%) suffered from acute MI during stroke unit treatment. To assure sufficient patient numbers for statistical testing, predictors of MI as complication were calculated for ischemic CVE only. Demographics, clinical characteristics and risk factors for individuals with ischemic CVE dichotomized by the presence or absence of the complication of MI are shown in table 1. In univariate analyses, patients who experienced MI were older and had more severe NIHSS scores at admission. The occurrence of MI was also associated with unfavorable vascular risk factor profiles and more prevalent cardiac disease such as atrial fibrillation and previous MI. Hypercholesterolemia was more often noted in non-MI Gattringer et al.
Downloaded by: St. Andrews University 138.251.14.35 - 12/3/2014 11:55:42 PM
Introduction
Table 1. Demographics and clinical characteristics of MI versus non-MI patients
Table 2. The following variables were included in the stepwise variable selection procedure as potential explanatory variables
Variable
MI (n = 421)
Non-MI (n = 41,198)
p value
Variable
Levels
Median age, years IQR Female Preexisting disabilitya Median NIHSS at admission IQR TIA Hypertension Diabetes mellitus Hypercholesterolemia Peripheral artery disease Smoking Atrial fibrillation Previous MI Previous stroke Location of clinical symptoms Left hemispheric Right hemispheric Brainstem Cerebellar Stroke syndromeb LACS TACS PACS POCS Thrombolysis Stroke etiologyc Microangiopathy Cardiac embolism Macroangiopathy Unknown Complications Intracranial hemorrhage Progressive stroke Brain edema Recurrent stroke Pneumonia Prognosis at discharge Median NIHSS IQR Death mRS 3–5
79.5 71.1–84.6 217 (51.5) 75 (17.8) 9 4–16 38 (9) 363 (86.2) 120 (28.5) 217 (51.5) 43 (10.2) 79 (18.7) 181 (43) 240 (57) 104 (24.7)
73.6 63.6–81.9 19,365 (47) 5,362 (13) 4 1–8 9,348 (22.7) 32,961 (80) 10,358 (25.1) 22,811 (55.4) 3,003 (7.3) 7,289 (17.6) 11,078 (26.8) 3,744 (9.1) 9,679 (23.4)
