Roquer J, Ois A, Rodríguez-Campello A, Gomis M, Munteis E, Jiménez-Conde J, Cuadrado-Godia E, Martínez-Rodríguez JE. Atherosclerotic Burden and Early Mortality in Acute Ischemic Stroke. Arch Neurol. 2007;64(5):699-704. doi:10.1001/archneur.64.5.699
The influence that previous clinical expressions of systemic atherosclerosis may have on evolution and early mortality in patients with acute ischemic stroke is not known.
To evaluate the influence that atherosclerotic burden (ATB), assessed by a simple clinical scale, has on the 30-day mortality in patients with first-ever ischemic stroke.
Retrospective review of case series from a prospective stroke record. An ATB score ranging from 0 to 2 was created using the history of ischemic heart disease and peripheral arterial disease. The impact of this score on the 30-day mortality was analyzed by multivariate regression analysis.
Tertiary university hospital.
A total of 1527 patients with first-ever ischemic stroke.
Main Outcome Measure
The 30-day mortality rate was 13.8%. Multivariate regression analysis showed an association between the ATB score and the 30-day mortality (P<.001). Comparing patients having no previous ATB with those with an ATB score of 1 or 2, the odds ratio (OR) for 30-day mortality increased from 1.71 (95% confidence interval [CI], 1.06-2.75) for patients with an ATB score of 1 to 5.90 (95% CI, 2.48-14.04) for those with an ATB score of 2. Age (OR, 1.05; 95% CI, 1.03-1.08), National Institutes of Health Stroke Scale score at admission (OR, 1.22; 95% CI, 1.18-1.25), atrial fibrillation (OR, 1.61; 95% CI, 1.10-2.35), hyperlipidemia as protector (OR, 0.39; 95% CI, 0.25-0.60), and glycemia at admission (OR, 1.07; 95% CI, 1.02-1.12) were also predictors of 30-day mortality.
Previous symptomatic atherosclerotic disease evaluated by a simple clinical score is an independent predictor of early mortality in patients with first-ever ischemic stroke.
Atherothrombotic ischemic stroke (IS), ischemic heart disease (IHD), and peripheral arterial disease (PAD) are different clinical expressions of atherosclerosis that share similar risk factors and preventive treatments and often appear concomitantly. However, they have been scarcely studied as a whole. Atherosclerotic burden (ATB) is a term used to describe the global extension of arteriosclerosis that has been related to poor outcome after IHD or PAD.1,2 To our knowledge, the influence that previous clinical expressions of systemic atherosclerosis may have on evolution and early mortality in patients with IS has not been previously evaluated. We hypothesize that early mortality in patients with first-ever acute IS may be related to their ATB evaluated by a simple clinical score based on their history of IHD and PAD.
Between January 1, 1997, and July 31, 2005, 2466 consecutive patients with acute IS were admitted with a diagnosis based on the World Health Organization definition of stroke.3 Patients with intracerebral hemorrhage (n = 309), previous stroke (n = 467), and incomplete data (n = 163) were excluded. All of the patients were treated following the stroke unit's therapeutic and diagnostic protocols. Since September 1, 2002, patients with National Institute of Neurological Disorders and Stroke criteria during the first 3 hours after stroke onset have received intravenous tissue plasminogen activator treatment.
All of the patients were assessed within 24 hours after stroke onset and fulfilled a clinical protocol that included demographic data and the following vascular risk factors based on their presence during admission, a prior physician diagnosis, or need for medical treatment: arterial hypertension (evidence of ≥2 blood pressure measurements >140/90 mm Hg recorded on different days before stroke onset); diabetes (fasting serum glucose level >7.0 mmol/L [>126 mg/dL]); hyperlipidemia (serum cholesterol levels >5.69 mmol/L [>220 mg/dL] or triglyceride levels >2.26 mmol/L [>200 mg/dL]); atrial fibrillation (AF) confirmed by an electrocardiogram performed during admission or previously; IHD (history of angina pectoris or myocardial infarction); heart failure (HF); PAD (intermittent claudication or a resting ankle-brachial index score <0.90 in any leg); and current smoking. Peripheral arterial studies were performed during hospitalization only in patients with symptomatic arterial disease and no previous vascular study. The diagnosis of cerebral infarction was confirmed in all of the patients by computed tomography or magnetic resonance imaging. Stroke severity was assessed by the National Institutes of Health Stroke Scale4 (NIHSS) performed at admission. Renal function evaluated by the serum creatinine level at hospital admission was also recorded. Patients were classified according to the Trial of Org 10172 in Acute Stroke Treatment (TOAST) criteria.5 Medical in-hospital complications were defined following the criteria by Davenport et al6 and included neurological, nonneurological, and iatrogenic complications. The primary outcome of the study was 30-day mortality.
To evaluate the impact of previous symptomatic atherosclerotic disease on mortality, a composite ATB score ranging from 0 to 2 was created. Based on the previously described diagnoses of IHD and PAD, score 0 was given to patients with no history of any of these disorders, score 1 was given to those patients with a diagnosis of IHD or PAD, and score 2 was given to patients with the concomitant presence of both conditions. There were no exclusion criteria for the score assessment for patients with IS included in the study. The inclusion of patients with IHD or PAD was independent of other associated cardiological conditions such as AF or HF.
The t test was used to evaluate differences in continuous variables, and the χ² test was used for those in proportions. First, a univariate analysis was used to examine the demographic data, risk factors, and patients' clinical differences according to their ATB scores. Second, we performed a univariate analysis of the influence of the following variables on the 30-day mortality: age, sex, arterial hypertension, diabetes, hyperlipidemia, AF, HF, IHD, PAD, composite ATB score, smoking, glycemia, creatinine level, tissue plasminogen activator treatment, NIHSS score at admission, and in-hospital medical complications. The TOAST classification was not included in the statistical model owing to the limited number of deaths in some stroke subtypes. To estimate odds ratios and the resulting 95% confidence intervals of variables related to patients' ATB (ATB score of 0 vs ≥1) and predictors of 30-day mortality, variables that reached P<.15 were included in a multivariate logistic regression analysis. Age, NIHSS score, and glycemia were included as continuous variables and the composite ATB score was categorized into 3 groups (composite ATB scores of 0, 1, and 2). The 2 bivariate interactions between confounders were tested and no significant interactions were found. Separate multivariate logistic regression models according to sex and death causes (neurological and nonneurological) were performed to evaluate any possible influence of these factors on the effect of ATB on mortality.
Two-way repeated analysis of variance and the Kruskal-Wallis test were used to evaluate the relationship between ATB score and demographic data, vascular risk factors, and stroke severity. To assess the impact of the ATB score on 30-day mortality in those TOAST subgroups in which statistical analysis was possible, we analyzed the predictors of 30-day mortality following a similar method as previously described in this article. All of the tests were 2-tailed and statistical significance was determined at α = .05.
The information used in the study was collected from the prospective clinical protocols of our hospital, which fulfilled the local ethical guidelines. The identities of the individual patients were completely anonymous. Therefore, patients signed no specific informed consent.
A total of 1527 patients with a mean (SD) age of 73.0 (12.0) years (771 men and 756 women) were evaluated. Demographic data and vascular risk factors are shown in Table 1. The median length of hospitalization was 11 days (interquartile range, 8-17 days). The stroke severity score at admission as assessed by the NIHSS was 6 (interquartile range, 3-15). The TOAST subtypes were large-artery arteriosclerosis (359 patients [23.5%]), small-vessel occlusions (230 patients [15.1%]), cardioembolisms (448 patients [29.3%]), strokes of other determined cause (38 patients [2.5%]) and strokes of undetermined cause (452 patients [29.6%]). Two hundred patients (13.1%) had IHD (83 patients with angina pectoris and 117 with myocardial infarction), 133 patients (8.7%) had PAD, and 44 patients (2.9%) had both. Coronary angiography performed in 77 patients showed 3-vessel disease (33 patients), 2-vessel disease (14 patients), 1-vessel disease (23 patients), and no vessel disease (7 patients) without any relation to mortality (P = .55). In patients with IHD, 24 had bypass surgery, 8 had coronary artery stenting, 7 had thrombolysis, 5 had primary angioplasty, and 6 had combinations of different procedures. Heart failure was present in 221 patients (14.5%).
There were 559 patients (36.6%) with medical complications; these complications were neurological in 87 patients (5.7%), nonneurological in 441 patients (28.9%), and iatrogenic in 36 patients (2.4%). Fifty-six patients presented with more than 1 complication. The 30-day mortality rate was 13.8% (210 cases). Causes of mortality were neurological in 47 patients (36 cases of intracranial hypertension or brain edema and 11 cases of brainstem stroke, recurrences, or hemorrhagic transformation) and nonneurological in 163 patients (respiratory causes in 120 cases [pneumonia or bronchoaspiration in 111 cases; pulmonary thromboembolism in 6 cases; respiratory distress in 3 cases]; heart causes in 14 cases [congestive HF in 10 cases; sudden death in 4 cases]; systemic malignancies in 5 cases; sepsis in 3 cases; digestive hemorrhages in 7 cases; other causes in 3 cases; and undetermined causes in 11 cases). Death occurred at a mean (SD) of 7.3 (6.7) days with neurological causes and 11.5 (14.4) days with nonneurological causes (P = .005).
Table 2 shows the characteristics of patients according to the ATB score. Factors independently related to the ATB dichotomized into an ATB score of 0 vs 1 or 2 were male sex, hyperlipidemia, diabetes mellitus, arterial hypertension, HF, and current smoking (Table 3). The ATB score was associated with the NIHSS score at admission (P = .002), which increased in relation to ATB (P = .001; odds ratio, 1.03; 95% confidence interval, 1.01-1.05), but not with the creatinine level (P = .41; odds ratio, 1.12; 95% confidence interval, 0.86-1.45).
Table 1 shows the results of the univariate analysis of factors related to 30-day mortality. The 30-day mortality rate was 9.2% (33 cases) in large-artery arteriosclerosis, 0.4% (1 case) in small-vessel occlusion, 25.2% (113 cases) in cardioembolism, 15.8% (6 cases) in strokes of other determined cause, and 12.6% (57 cases) in strokes of undetermined cause. There was no relationship between the type of IHD or tissue plasminogen activator treatment (33 cases) and 30-day mortality.
Multivariate regression analysis after adjustment for confounders showed a statistical significance of the ATB score for 30-day mortality (P<.001). Other predictors of 30-day mortality were age, NIHSS score, AF, hyperlipidemia as a protector, and glycemia at admission (Table 4). Male sex (P = .60), smoking (P = .56), HF (P = .50), and the creatinine level (P = .48) did not reach statistical significance.
Analyzed by sex, ATB remained an independent predictor of 30-day mortality for men (P = .001) and women (P = .046) (Table 5). The influence of ATB on mortality also remained independent of the cause of death (neurological, P = .03; nonneurological, P<.001) (Table 5). Table 6 shows the relationship between the ATB score and 30-day mortality according to the TOAST classification controlled by confounders. The ATB score increased mortality significantly in the large-artery atherosclerotic (P = .006) and cardioembolism (P = .03) subgroups and nonsignificantly in the undetermined cause group (P = .19).
The deleterious impact that ATB, evaluated by techniques such as the measurement of the carotid intimal-medial thickness,7,8 the arterial compliance,9 the microalbuminuria determination,10 the amount of calcium in the coronary artery,11 and the endothelial function,12 has on a patient's clinical outcome after cardiovascular events has been previously demonstrated.7,9,10 The aim of our study was to analyze the influence that a simple clinical score based on the history of symptomatic atherosclerotic disease may have on first-ever IS evolution in 2 complementary aspects: the clinical characteristics of patients with IS and early mortality after IS.
Previous symptomatic atherosclerotic disease was related to male sex and all classical vascular risk factors as well as HF (Table 2 and Table 3) in accordance with the well-established relationship between vascular risk factors and atherothrombosis development. Atherosclerotic burden was associated with male patients in contrast to a recent study13 reporting an independent risk factor of female sex for thromboembolic stroke in patients with AF.
Although the deleterious impact of ATB in stroke has only been demonstrated in those of undetermined cause,14 some studies have assessed the ATB influence on the clinical outcome after acute heart disease. The OPUS-TIMI 16 (Orbofiban in Patients With Unstable Coronary Syndromes–Thrombolysis in Myocardial Infarction 16) study1 showed that patients with acute coronary syndrome and any prior stroke or PAD had more extensive coronary artery disease and worse outcome. In addition, the presence of PAD in patients with acute myocardial infarction treated with primary angioplasty was an independent predictor of in-hospital mortality and death at 1 year.2
In our study, the ATB score in patients with first-ever IS remained an independent predictor of 30-day mortality (P<.001). Other evaluated variables agree with previous reports showing that age,15,16 stroke severity,17 glycemia at admission,18 and AF19 were predictors of poor outcome after stroke, whereas higher serum cholesterol levels were related with better outcome20 (Table 4). Three facts reinforce the importance of the ATB score as a determinant of 30-day mortality after stroke. First, the ATB score after sex analysis independently predicted a poor outcome for both sexes despite the known reduced role of atherosclerosis on stroke development in women. Second, after an analysis according to neurological and nonneurological death causes, the ATB score remained a powerful predictor of death for both. However, the impact was higher for the latter, suggesting an ATB influence on very early mortality related to neurological causes as well as delayed mortality related to nonneurological complications. Third, the negative impact of the ATB score was also present in IS with nonatherothrombotic causes21,22 (Table 6), suggesting an increased susceptibility to mortality independent of the stroke cause. The low number of deaths in small-vessel occlusion stroke and those of other determined cause limited the statistical analysis in these groups.
Mechanisms explaining the deleterious impact of ATB on mortality may be multifactorial. The Second Manifestations of Arterial Disease Study23 showed that patients with IHD with stroke or abdominal aortic aneurysm had increased carotid stiffness compared with isolated IHD. Early mortality after IS has an inverse relationship with aortic compliance determined by Doppler ultrasonography,24 a sign of artery stiffness. Additionally, greater artery stiffness was related to fatal stroke in patients with essential hypertension.25 Two recent studies26,27 found that aortic pulse-wave velocity, another measure of aortic wall stiffness, was an independent predictor of IHD and stroke in apparently healthy subjects26 that can predict a composite of cardiovascular outcomes above and beyond conventional cardiovascular risk factors.27 Aortic pulse-wave velocity also predicted cardiovascular mortality in elderly hospitalized patients.28 Therefore, a direct relationship between ATB and widespread atherosclerotic disease leading to greater arterial stiffness can be hypothesized. Vasoregulatory cerebral mechanisms may be more susceptible to failing after IS in these patients, increasing the cerebral damage. Our study supports this hypothesis because the NIHSS score increased in relation to the ATB score (P = .002) and stroke severity remained independently associated with the ATB score.
Our study has some limitations. As we focused on patients with IS with previous clinical symptoms of arteriosclerosis, IHD and PAD could have been underrecognized because no systematic arterial or cardiological studies were performed in asymptomatic patients during the hospitalization period. In addition, the presence of IHD or PAD based on the clinical history as a marker of ATB may not have enough sensitivity for the systemic atherosclerotic status. Moreover, the pathological significance of PAD and IHD may differ in their influence on clinical evolution and mortality. However, these limitations may be superimposed to the simplicity of this procedure to select patients with a higher risk of mortality. Finally, we do not have data on the proportion of men and women who survive to reach the hospital.
In summary, a higher clinical ATB score may be a marker of higher stroke severity. There is an association between history of atherothrombotic disease and the 30-day mortality in patients with first-ever IS not related to the stroke cause. Further studies assessing the relationship between ATB and ischemic lesion growth during the acute phase might establish the validity of this potential marker of worse prognosis in the initial evaluation of IS.
Correspondence: Jaume Roquer, MD, PhD, Unitat d’Ictus, Servei de Neurología, IMIM–Hospital del Mar, Passeig Marítim 25-29, 08003 Barcelona, Spain (firstname.lastname@example.org).
Accepted for Publication: January 8, 2007.
Author Contributions:Study concept and design: Roquer, Ois, and Martínez-Rodríguez. Acquisition of data: Roquer, Ois, Rodríguez-Campello, Gomis, Munteis, Jiménez-Conde, and Martínez-Rodríguez. Analysis and interpretation of data: Roquer, Ois, Cuadrado-Godia, and Martínez-Rodríguez. Drafting of the manuscript: Roquer, Ois, Rodríguez-Campello, Munteis, Cuadrado-Godia, and Martínez-Rodríguez. Critical revision of the manuscript for important intellectual content: Roquer, Ois, Gomis, Munteis, Jiménez-Conde, Cuadrado-Godia, and Martínez-Rodríguez. Statistical analysis: Roquer, Ois, Jiménez-Conde, Cuadrado-Godia, and Martínez-Rodríguez. Obtained funding: Roquer. Administrative, technical, and material support: Roquer, Rodríguez-Campello, Gomis, and Munteis. Study supervision: Roquer, Ois, Rodríguez-Campello, Cuadrado-Godia, and Martínez-Rodríguez.
Financial Disclosure: None reported.
Funding/Support: This study was supported in part by grant Red HERACLES RD06/0009 from the Ministerio de Sanidad y Consumo, Instituto de Salud Carlos III.