Early Stroke following Acute Myocardial Infarction: Incidence, Predictors and Outcome in Six Middle-Eastern CountriesAlbaker O.a · Zubaid M.a · Alsheikh-Ali A.A.b, g · Rashed W.c · Alanbaei M.a · Almahmeed W.b · Al-Shereiqi S.Z.d · Sulaiman K.e · Al Qahtani A.f · Al Suwaidi J.f · for the Gulf RACE Investigators
aDepartment of Medicine, Faculty of Medicine, Kuwait University, Kuwait, Kuwait; bInstitute of Cardiac Sciences, Sheikh Khalifa Medical City, Abu Dhabi, United Arab Emirates; cDepartment of Medicine, Mubarak Al-Kabeer Hospital, Ministry of Health, Kuwait, Kuwait; dPriority Diseases Control Section, Department of Non-Communicable Diseases Surveillance and Control, Ministry of Health, and eRoyal Hospital, Muscat, Oman; fDepartment of Cardiology and Cardiovascular Surgery, Hamad Medical Corporation, Doha, Qatar; gInstitute for Clinical Research and Health Policy Studies and Department of Medicine, Tufts Medical Center and Tufts University School of Medicine, Boston, Mass., USA Corresponding Author
Background and Objectives: Stroke is a potential complication of acute myocardial infarction (AMI). The aim of this study was to identify the incidence, risk factors predisposing to stroke and in-hospital outcome during the index admission with AMI among patients in the Middle East. Methods: For a period of 6 months in 2006 and 2007, 5,833 consecutive AMI patients were enrolled from 64 hospitals in 6 Middle East countries. Results: The incidence of in-hospital stroke following AMI was 0.85%. Most cases were ST segment elevation AMI-related and ischemic in nature. Patients with in-hospital stroke were older than patients without stroke and were more likely to be female (36 vs. 18.6%, p = 0.0033). They were also more likely to have diabetes mellitus, dyslipidemia, prior AMI, or percutaneous/surgical coronary revascularization. Patients with stroke were more likely to present with advanced Killip class II–IV, higher mean heart rate and higher serum creatinine. Independent predictors of stroke were age, prior stroke, prior coronary artery bypass surgery, anterior AMI and systolic blood pressure >190 mm Hg on presentation. Early administration of statins was independently associated with reduced stroke risk (odds ratio, OR, 0.4, 95% confidence interval, CI, 0.19–0.90, p = 0.025). Stroke was fatal in 44% of the cases and was independently associated with in-hospital mortality (adjusted OR 12.5, 95% CI 5.7–27.4, p < 0.01). Conclusion: There is a low incidence of in-hospital stroke in Middle-Eastern patients presenting with AMI but with very high fatality rates. Early statin therapy was associated with a significant reduction in stroke risk. Future work should be focused on reducing the risk and improving the outcome of this devastating complication.
Copyright © 2011 S. Karger AG, Basel
Although stroke is the second leading cause of death worldwide, stroke complicating acute myocardial infarction (AMI) is relatively uncommon. The risk of developing stroke is highest in the first 4 weeks after AMI , especially the first few days. The risk decreases gradually thereafter . Although the incidence of AMI-associated stroke has declined, mortality from this devastating complication remains high . Several studies explored risk factors for stroke after AMI and these studies were almost exclusively conducted in developed countries, and data from other ethnicities are lacking. Middle Eastern patients presenting with AMI are relatively younger and are more likely to have diabetes mellitus when compared to their Western counterparts [4,5,6]. Furthermore, although percutaneous revascularization is the primary reperfusion therapy in the Western world, thrombolytic therapy remains the primary reperfusion therapy worldwide including the Middle East [4,5,6].
The INTERSTROKE study  that enrolled 3,000 patients with first stroke in 22 countries has recently reported risk factors for stroke which include hypertension, smoking, abdominal obesity, diabetes, stress and depression; however, there is little agreement in the literature about predictors of stroke after AMI with the exception of older age and prior stroke [4,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40]. For example, hypertension [14,15,22,33,34] and diabetes [22,27,32,33,36,37] were shown in some, but not all, studies to be predictors of stroke. Hence, studying stroke in AMI patients of different ethnicities, especially in regions where diabetes is highly prevalent, may contribute to our understanding of this devastating complication. The aim of our study was to define the incidence and determinants of in-hospital stroke after AMI in Middle Eastern patients.
The Gulf Registry of Acute Coronary Events (Gulf RACE) enrolled 8,176 patients admitted with the final diagnosis of acute coronary syndrome (ACS) in 64 hospitals from 6 Middle Eastern Gulf countries over a 6-month period in 2006 and 2007. Details of study organization have been previously described [5,6]. For the purpose of this analysis, we included all documented AMI patients (n = 5,833), 12 AMI patients had insufficient stroke data and were, therefore, excluded from this analysis. Patients with ST segment elevation AMI and left bundle branch block (LBBB) AMI were grouped as the STEMI group. Patients with unstable angina were excluded from the current analysis. We analyzed patients with early stroke after AMI and compared them with those who did not have stroke. Follow-up was made from admission to death or discharge.
Per the Gulf RACE protocol, stroke was defined as a neurologic deficit persisting more than 24 h. Stroke types: hemorrhagic stroke – a stroke with documentation on imaging of hemorrhage in the cerebral parenchyma, or subarachnoid hemorrhage; ischemic stroke – a focal neurologic deficit that results from a thrombus or embolus (and is not due to hemorrhage), and unknown/CT scan not done – if the type of stroke could not be determined by imaging or other means (e.g. from lumbar puncture) or if imaging was not performed. Disability from stroke was classified as none to minor if the patient is left with no or mild disability and is completely self-dependent and functional, severe if the patient has severe limitation of functional ability, or is bedridden or in a wheel chair and needs considerable help or is fully dependent on others for self-care and daily activities at home or work. If the patient died from complications of stroke, the case was classified as death . Early statin use was defined as statin administration during the first 24 h of hospitalization.
Univariate comparisons of patients with and without strokes were made. Continuous variables are presented as medians (interquartile limits) and compared using the Wilcoxon rank sum test. Categorical variables are shown as percentages and were compared using the χ2 test. Using binary logistic regression, the odds ratio was calculated and factors significantly associated with stroke were identified. χ2 tests with corrections for multiple comparisons (Bonferroni) were performed on each of the presenting complaints. p < 0.05 was considered statistically significant. All analyses were performed using SAS 9.1 for analysis (Cary, N.C., USA).
Of 5,821 patients with AMI in the present analysis, 3,196 had STEMI while 2,625 patients did not have STEMI (NSTEMI).
Fifty patients (0.85%) sustained stroke during their index hospitalization. The rate of in-hospital stroke among STEMI patients was 1.1 % (36 out of 3,196) while it was 0.5% (14 out of 2625) among patients with NSTEMI. In-hospital stroke was classified as ischemic in 35 patients (70%). Hemorrhagic stroke occurred in 7 patients (14%), while in the remaining 8 patients (16%) stroke type was not identified.
Baseline Characteristics of Patients with Stroke (table 1)
Patients who sustained in-hospital stroke were about 7 years older than patients without stroke (62 vs. 55, p < 0.0001) and were more likely to be female (36 vs. 18.6%, p = 0.0033). Stroke patients were also more likely to have diabetes mellitus (58 vs. 38.9%, p = 0.0082), dyslipidemia (44 vs. 26.6 %, p = 0.0094) and a previous history of AMI (40 vs. 21.5%, p = 0.0029). Similarly, they were more likely to have a history of prior coronary artery bypass surgery (CABG) (18 vs. 4.1%) and more likely to be on aspirin prior to admission when compared to nonstroke patients (50 vs. 32.6%, p = 0.0144). Stroke patients were also more likely to have peripheral vascular disease and a prior history of stroke than nonstroke patients. There was a trend towards an increased prevalence of hypertension among stroke patients (58 vs. 44.3%, p = 0.06). Although there were no differences between the two groups in the frequency of atrial fibrillation at presentation; stroke patients were more likely to develop atrial fibrillation during the hospitalization period when compared to nonstroke patients (p = 0.0002).
|Table 1. Baseline characteristics of patients with or without stroke complicating AMI|
Stroke patients compared to nonstroke patients were more likely to present with atypical chest pain (12 vs. 6%), dyspnea (14 vs. 9%) and late presentation (symptom onset to presentation time >12 h in the STEMI group; 50 vs. 30%). They also presented with advanced Killip class II–IV (p < 0.001), higher median heart rate (96 vs. 84 beats per minute, p = 0.0002) and systolic blood pressure >190 mm Hg (16 vs. 5%, p = 0.0046). Interestingly, diastolic blood pressure was lower among stroke patients (70 vs. 80 mm Hg; p = 0.0046). Stroke patients were more likely to have left ventricular dysfunction (55 vs. 33%; p = 0.01). Stroke patients had higher serum creatinine levels on admission when compared to nonstroke patients [106 (80–168) vs. 88 (70–106), p = 0.0005].
In-Hospital Treatment Patterns of Patients with Stroke (table 2)
The treatment patterns for patients with stroke are presented in table 2. There were no differences between the two groups in the use of aspirin, clopidogrel, unfractionated heparin, low molecular-weight heparin, glycoprotein IIb/IIIa inhibitors or thrombolytic therapy. Nor were there any differences in the use of β-blockers, angiotensin-converting enzyme or angiotensin-receptor inhibitors. Stroke patients were less likely to receive statins. The overall use of coronary angiography was low in our registry and its use was even lower in patients who had stroke (8 vs. 22%).
|Table 2. In-hospital treatment of patients with or without stroke complicating AMI|
In-Hospital Outcomes of Patients with Stroke (table 3)
Patients who suffered in-hospital stroke were more likely to have their hospital course complicated with recurrent myocardial ischemia (22 vs. 9.9%), cardiogenic shock (40 vs. 6.3%) and major bleeding (8.2 vs. 0.8%) compared to those without stroke. Death occurred in 44% of patients who sustained in-hospital strokes compared to 4% of those who did not sustain in-hospital strokes. Furthermore, 25% of stroke patients had severe disability and 49% had no or minor disabilities. Length of stay was significantly longer among stroke compared to nonstroke patients (7 vs. 5 days; p = 0.0007).
|Table 3. In-hospital outcome of patients with or without stroke complicating AMI|
Multivariate Predictors of Stroke and of Outcome (table 4)
The following variables were independently associated with a significantly increased risk of in-hospital stroke: older age, STEMI/LBBB AMI, prior stroke, prior CABG and systolic blood pressure >190 mm Hg on presentation (table 4). Early administration of a statin (within 24 h of presentation) was independently associated with a significantly lower risk of in-hospital stroke. There was no significant interaction between AMI type (STEMI/LBBB vs. NSTEMI) and any of the variables tested in the model.
|Table 4. Presentation characteristics associated with in-hospital stroke in patients with AMI|
In-hospital stroke was independently associated with a significantly higher risk of in-hospital death. The odds of mortality are 12.5 higher (5.7–27.4) in patients with stroke compared to those without stroke after adjusting for gender, Global Registry of Acute Coronary Events (GRACE) risk score, prior history of diabetes, hypertension, dyslipidemia, smoking, peripheral vascular disease, prior stroke, prior AMI/percutaneous coronary intervention (PCI)/CABG and prior aspirin use.
The current study reports a low prevalence of stroke among Middle Eastern patients presenting with AMI. Seventy percent of these strokes were ischemic in origin and 14% were hemorrhagic. Predictors of stroke were older age, prior CABG, prior stroke, elevated systolic pressure on admission and STEMI presentation. Although the prevalence of stroke was low and comparable to that of reported registries in Western countries, the consequences in stroke patients were dismal, with a 44% in-hospital mortality rate and 25% severe disability. The current study suggests early use of statin therapy is protective against stroke occurrence in patients with AMI.
The current study reports 0.85% stroke prevalence in Middle Eastern patients presenting with AMI in the current era. The reported prevalence of stroke from studies performed mainly in the Western world varied between as low as 0.31 and 1.9% (table 5). Our findings are consistent with those reported in the Euro Heart Survey (0.8% in STEMI and 0.7% in NSTEMI)  and that of the GRACE study  which reported 0.9% in-hospital stroke rates in ACS patients. Using data from the National Registry of Myocardial Infarction (NRMI) 3 and 4, Van de Graaf et al.  reported 0.54 and 0.31% in-hospital stroke risk in patients undergoing thrombolytic therapy or primary PCI consecutively. More recently, Saczynski et al.  reported a 1.4% overall stroke risk in patients hospitalized with AMI using the Worcester Heart Attack Study for the period between 1986 and 2005. They noticed an increased prevalence of stroke through the 1980s and 1990s, but the prevalence declined slightly thereafter. This decline in the prevalence of stroke over the recent years was also observed in the North Sweden MONICA (Monitoring of Trends and Determinants in Cardiovascular Disease Study) . When compared to the current study, stroke patients in the NRMI 3 and 4, Euro Heart ACS Survey and the Wercester Heart Attack Study were significantly older (>10 years) and less likely to have diabetes mellitus. Also, while thrombolysis is the main modality of reperfusion therapy in the current registry, significantly more patients received primary PCI in the other studies.
|Table 5. Prevalence and impact of stroke after ACS in reported studies|
In a population-based study in Olmsted County (Minnesota) between 1979 and 1998, Witt et al.  reported a 44-fold increase risk of stroke during the first 30 days after AMI and this risk remained 2–3 times higher than expected during the first 3 years after AMI. Ng et al.  reported a 7.17% 1-year risk of stroke in Chinese patients after AMI, which is much higher than in other reports. However, a firm conclusion cannot be drawn from the study Ng et al. because of the small number of patients enrolled (279 patients). The current study extends these observations reporting a low stroke prevalence in Middle Eastern patients primarily receiving thrombolytic therapy as a reperfusion therapy. Furthermore, the proportion of ischemic to hemorrhagic stroke (2/3) is also consistent with previous reports.
Several investigators identified characteristics that would increase the risk of stroke in patients with AMI. These risk factors include age, female gender, hypertension, prior or in-hospital CABG, prior AMI, anterior AMI, renal impairment, atrial fibrillation on admission or developing during hospitalization, low body weight, black race, congestive heart failure, elevated admission heart rate and systolic or diastolic blood pressure. While some of these factors, such as age and prior history of stroke, were consistently shown to increase the risk of developing stroke data are conflicting regarding other variables, such as female gender and history of hypertension. In the current study, we also observed older age, history of prior stroke, history of prior CABG and anterior AMI to be independent predictors of increased risk of stroke.
History of hypertension was not a predictor of stroke in the current study; however, admission systolic blood pressure >190 mm Hg was a powerful predictor of increased risk. Elevated systolic [4,15,34] and or diastolic blood pressure [23,32,34] on admission as predictors of stroke were also reported by some previous investigators. On the other hand, Wienbergen et al.  reported lack of increased risk of stroke with elevated systolic >180 mm Hg or diastolic blood pressure >110 mm Hg. Although diabetes mellitus was highly prevalent in our study group, diabetes was not an independent predictor of stroke early after AMI, and this is consistent with some but not all previous reports. Although stroke patients in our registry were more likely to develop atrial fibrillation during hospitalization, atrial fibrillation was not an independent predictor of stroke risk. This observation is consistent with observations from the Olmsted County, Minnesota, study  and contradictory to those of other reports [4,18,32,35,43]. The variability in findings among the various registries may be attributed to several factors including patient population, ethnicities and the use of thrombolytic therapy versus primary percutaneous revascularization therapy, and underscores the need for further international studies that include adequate representations of female gender and various ethnicities.
The advent of thrombolysis and primary PCI undoubtedly resulted in significant reduction in stroke risk from 5% in the 1970s to 1% in the current era. More recently, Van de Graaf et al.  reported early thrombolytic therapy within 15 min of presentation to be associated with a reduction in stroke risk (odds ratio, OR, 0.58; 95% confidence interval, CI, 0.36–0.94) while primary PCI within 90 min of arrival was associated with a trend of reduced risk (OR 0.68, CI 0.41–1.12). This reperfusion benefit appeared not to be related to improvement in cardiac function. The overall performance of PCI was very low in the current registry; however, it is unlikely that increased PCI use would have decreased the incidence of stroke. Cronin et al.  analyzing data from the OASIS (Organization to Assess Strategies for Ischemic Syndromes) and OASIS-2 trial reported an increased risk of stroke on NSTEMI patients in relation to early CABG (OR 3.95, p = 0.003) and not PCI at 6 months’ follow-up.
We observed a significant reduction in stroke risk in patients receiving early statin therapy (within 24 h of admission). In the GRACE registry, prior statin use was associated with reduced stroke risk in ACS patients . More recently, Kinlay et al.  using the MIRACL (Myocardial Ischemia Reduction with Aggressive Cholesterol Lowering) data reported elevated inflammatory markers in those who developed stroke after ACS when compared to nonstroke patients. These inflammatory markers included C-reactive protein, serum amyloid A and interleukin-6. Furthermore, high-dose atorvastatin significantly reduced the elevated risk of stroke in the 16 weeks after ACS when compared to placebo. The investigators hypothesized that statins reduce the high early risk of stroke in ACS by an anti-inflammatory effect in carotid, intracerebral and aortic atheromata. In addition to its early benefit, lipid-lowering therapy was also shown to be associated with a reduction in stroke risk at long-term follow-up after AMI [26,32,45,46]. These findings are further supported by the observations from the SPARCL trial (Stroke Prevention by Aggressive Reduction in Cholesterol Levels) of significant reduction in recurrent stroke risk in patients with recent stroke who received statin therapy .
Mortality from ischemic stroke can be up to 10–40% and even higher with hemorrhagic stroke. We found a 44% mortality rate in patients who sustained stroke. It appears that although therapeutic advancements for AMI have reduced the incidence of post-AMI stroke, it did not change the rate of poststroke mortality.
Our data were collected from an observational study. The fundamental limitations of observational studies cannot be eliminated because of the nonrandomized nature and unmeasured confounding factors. However, well-designed observational studies provide valid results and do not systemically overestimate the results compared with the results of randomized controlled trials. Our study was focused on in-hospital risk of stroke. We did not look for risk factors by day of onset. Stroke occurring within the first few days may be different from those occurring a week or two later. Detailed causes of death among the patient population, notably whether death was related to stroke, are lacking. Finally, long-term follow-up for post-AMI stroke will reveal further risk factors for late stroke. Investigations focused on such risk factors were beyond the scope of our study but are needed for our population.
The current study reports a low incidence of stroke among Middle Eastern AMI patients with a very high in-hospital mortality rate. While older age, prior stroke, prior CABG, anterior AMI and elevated systolic blood pressure on admission were associated with an increased risk, statin therapy within 24 h of admission was associated with a significant reduction in risk. Future work should be focused on reducing the risk and outcome of this devastating complication.
Bahrain: Haitham Amin (National Coordinator), Husam Noor, Rashid Albannai;
Kuwait: Bassam Bulbanat (National Coordinator), Shahid Zubair, Mustafa M. Ridha, Marwan Abu Rezq, Hesham Saad, Mousa Akbar, Rashed Al-Hamdan, Fahad Al-Enizi;
Oman: Jawad Al-Lawati (National Coordinator), Kadhim Sulaiman (National Coordinator), B. Kamath, P.P. Singh, Marei Aysha, Ali Yousif, Faisal Tamimi, Kurain Mathew, Ahmed Mosad, A. Narayan, Adel Badr Eldin, Qassim Shimal, S.K. Samantray, Kamiran Dabagh, Zakaria Boghdady, K.J. Sulaiman, Sulieman Al Sheraiqi;
Qatar: Nidal A. Asaad (National Coordinator), Amar Mohammad Salam;
United Arab Emirates: Nazar Albustani (National Coordinator), Srinath Kidambi, Vipin Thomas, Afzal Y. Ali, Nayan Rao, V.J. Sebastian, Wael El-Abbassi, Mohamed Ibrahim, M. Taiseer El-Massri, Adel A.S. Wassef, Amrish Agrawal, Joseph Kurein, Anthony Thomas, Ismail Jalian;
Yemen: Ahmed Al-Motarreb (National Coordinator), Nora Al-Sagheer, Abdu Hamood, Mohammed Abu Ghanem, Abd Al-Jalil Al Wazeer, Abdullah Al-Shameeri, Majed Al-Showbaki, Elham Al-Qudari, Araf Ahamed Saleh, Hanan Mojamal, Abd Al-Raheem ba Khashwen, Faiz Khoba, Abd Al-Hakim Al-Hammadi, Munir Al-Absi, Mohammad Tantawi, Khalad Al-Zanan, Ahmed Al-Rubaidi, Faud Ali, Abd ARakeeb Al-GHobari, Madian Shehab, Motee Al-Awlagi.
Gulf RACE is a Gulf Heart Association project and was financially supported by Sanofi Aventis, Paris, France, and Qatar Telecommunications Company, Doha, Qatar.
The authors have no conflict of interest to disclose. The sponsors had no role in study design, data collection or data analysis. The sponsors had no role in the writing of the report and submission of the manuscript.
Jassim Al Suwaidi, MB, ChB, FACC, FSCAI, FESC
Department of Cardiology and Cardiovascular Surgery
Hamad General Hospital, PO Box 3050, Doha (Qatar)
Tel. +974 439 2464, E-Mail Jha01@hmc.org.qa
Received: December 20, 2010
Accepted: June 6, 2011
Published online: November 1, 2011
Number of Print Pages : 12
Number of Figures : 0, Number of Tables : 5, Number of References : 46
Vol. 32, No. 5, Year 2011 (Cover Date: November 2011)
Journal Editor: Hennerici M.G. (Mannheim)
ISSN: 1015-9770 (Print), eISSN: 1421-9786 (Online)
For additional information: http://www.karger.com/CED