• cerebrovascular disease;
  • computed tomography;
  • outcome;
  • severe;
  • stroke;
  • thrombolysis


  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Disclosure of Conflict of Interests
  9. References

Summary.  Background:  Severe stroke carries high rates of mortality and morbidity. The aims of this study were to determine the characteristics of patients who initially presented with severe ischemic stroke, and to identify acute and subacute predictors of favorable clinical outcome in these patients. Methods:  An observational cohort study, Acute Stroke Registry and Analysis of Lausanne (ASTRAL), was analyzed, and all patients presenting with severe stroke – defined as a National Institute of Health Stroke Scale score of ≥ 20 on admission – were compared with all other patients. In a multivariate analysis, associations with demographic, clinical, pathophysiologic, metabolic and neuroimaging factors were determined. Furthermore, we analyzed predictors of favorable outcome (modified Rankin scale score of ≤ 3 at 3 months) in the subgroup of severe stroke patients. Results:  Of 1915 consecutive patients, 243 (12.7%) presented with severe stroke. This was significantly associated with cardio-embolic stroke mechanism (odds ratio [OR] 1.74, 95% confidence interval [CI] 1.19–2.54), unknown stroke onset (OR 2.35, 95% CI 1.14–4.83), more neuroimaging signs of early ischemia (mostly computed tomography; OR 2.65, 95% CI 1.79–3.92), arterial occlusions on acute imaging (OR 27.01, 95% CI 11.5–62.9), fewer chronic radiologic infarcts (OR 0.43, 95% CI 0.26–0.72), lower hemoglobin concentration (OR 0.97, 95% CI 0.96–0.99), and higher white cell count (OR 1.05, 95% CI 1.00–1.11). In the 68 (28%) patients with favorable outcomes despite presenting with severe stroke, this was predicted by lower age (OR 0.94, 95% CI 0.92–0.97), preceding cerebrovascular events (OR 3.00, 95% CI 1.01–8.97), hypolipemic pretreatment (OR 3.82, 95% CI 1.34–10.90), lower acute temperature (OR 0.43, 95% CI 0.23–0.78), lower subacute glucose concentration (OR 0.74, 95% CI 0.56–0.97), and spontaneous or treatment-induced recanalization (OR 4.51, 95% CI 1.96–10.41). Conclusions:  Severe stroke presentation is predicted by multiple clinical, radiologic and metabolic variables, several of which are modifiable. Predictors in the 28% of patients with favorable outcome despite presenting with severe stroke include hypolipemic pretreatment, lower acute temperature, lower glucose levels at 24 h, and arterial recanalization.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Disclosure of Conflict of Interests
  9. References

Acute ischemic strokes with an initially severe neurologic deficit represent 2–10% [1] of all ischemic strokes, and are associated with poor short-term and long-term prognoses. They are usually termed ‘malignant’ or ‘massive’ if the total territory of the middle cerebral artery is infarcted [2]. Such patients have a high risk of focal edema with herniation, systemic complications such as pneumonia [3], acute cardiac failure [4], and death. These strokes seem to be more frequently caused by cardiac emboli and less frequently caused by large artery occlusive mechanisms than are less severe strokes [5]. In addition, increasing age, hyperglycemia and fever are often associated with worse outcomes. Although less evidence exists, several other factors may also predict a worse prognosis after stroke: premorbid disability, cognitive impairment, reduced consciousness at onset, certain acute clinical deficits, non-stroke unit care, and female gender [6]. The most consistently found predictors of long-term disability after ischemic stroke are increasing age and severity at onset [7]. Neuroimaging may represent a promising means to identify patients with poor outcome. The extent of early ischemic change on non-contrast computed tomography (CT) is associated with irreversible ischemia and prognosis [8–15]. In this study, we studied a population with clinically severe strokes.

Few studies have systematically evaluated multimodal factors in unselected consecutive severe stroke patients. A better knowledge of the risk factors in the acute and the subacute phases and the relationship between patient characteristics and long-term outcome would be useful for patient management. The objectives of this study were as follows: first, to study the sociodemographic, clinical, biological and radiologic factors that are associated with severe stroke in the acute phase; and second, to define predictors of good recovery in patients who presented with severe stroke.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Disclosure of Conflict of Interests
  9. References


We used data from the Acute Stroke Registry and Analysis of Lausanne (ASTRAL). The ethics commission for research on humans of the Canton of Vaud, subcommission III, approved the scientific use of ASTRAL, which collected multiple datasets on all acute ischemic stroke patients from the period 2004 to 2010, defined arbitrarily as those presenting within 24 h after last-well time to the stroke unit and/or intensive care unit of Centre Hospitalier Universitaire Vaudois (CHUV) [16], which is a community hospital for a population of ∼ 350 000, as well as the only stroke unit for 720 000 persons. In summary, we collected data on demographics, vascular risk factors, previous cerebrovascular events, prestroke disability, and medication at the time of stroke. Stroke pathophysiology was classified according to the TOAST classification [17]. The National Institute of Health Stroke Scale (NIHSS) score was prospectively recorded for each patient on admission, and at 4–6 h, 24 h, and 7 days, by NIHSS-certified medical personnel. Given the widespread use of the NIHSS to evaluate stroke severity, we used this scale with an arbitrary cut-off of ≥ 20 on admission to define severe strokes. As is the case for mild stroke, there is no consensus on how to define severe strokes. In this study, we propose this term for clinically severe strokes, and we used a cut-off of an NIHSS score of ≥ 20 points at presentation, while being aware of the limitations of the NIHSS grading of strokes associated with different cerebrovascular locations [16]. Metabolic and hematologic parameters and vital signs were recorded in the registry on arrival, and again at 24–48 h after stroke onset. Finally, we recorded silent lesions, leukoaraiosis and early ischemic lesions on acute imaging (mostly CT), as well as > 50% arterial stenosis or occlusion in cervical and cerebral arteries on acute imaging (mostly CT angiography). Most patients also underwent repeat parenchymal (magnetic resonance imaging or CT) and arterial imaging (by magnetic resonance angiography [MRA], computerized tomography angiography [CTA], or Doppler) in the subacute or chronic phase after stroke. The modified Rankin score (mRs) was assessed by mRs-certified personnel at 3 months in the outpatient clinic. At 12 months, and for patients unable to attend the 3-month outpatient clinic, mRs was assessed by a structured telephone interview by mRs-certified personnel. Acute stroke management and secondary prevention of these patient followed current European Stroke Organization guidelines [18].

Statistical analyses

In the first part of this study, all ASTRAL patients were separated into a group with a severe stroke (S) with an admission NIHSS score of ≥ 20, or other comparison group (O). We analyzed all demographic, clinical, biological and radiologic factors (total of 67 variables) that were readily available in the acute phase of stroke and were collected in ASTRAL.

For the second part of this study, all patients in the S group were separated into a group with a relatively favorable outcome and a group with an unfavorable outcome. Favorable outcome was defined as an mRS score of ≤ 3, which is a common cut-off value in studies of severe stroke, such as hemorrhagic or posterior fossa stroke studies, and, given the initially severe clinical picture, made it a priori less probable that an mRS score of 0, 1 or 2 would be achieved. We first studied the factors from the acute phase that may be associated in severe stroke with a favorable outcome, and then added variables collected in the subacute phase (24–48 h after stroke), including subacute NIHSS values, metabolic values, recanalization on repeat arterial imaging, and parenchymal findings in subacute neuroimaging.

For univariate comparison, we used Student’s t-test for normally distributed continuous data, and the Wilcoxon test for skewed distributions. For the categorical data, we used the chi-squared test. All variables with P-values of < 0.1 in the univariate analysis were entered into a multivariate logistic regression model to identify independent predictors. Using the significant variables in the multivariate analysis, we calculated the areas under the receiver operating characteristic curves (ROC-AUCs) for the two analyses. The level of significance was set at 95%. All data analyses were conducted by a statistician of the Institute of Social & Preventive Medicine at the CHUV, with stata version 9.2, 2001 (Stata, College Station, TX, USA).


  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Disclosure of Conflict of Interests
  9. References

Two hundred and forty-three of 1915 (12.7%) consecutive stroke patients had presented with severe stroke. Regarding the 58 factors examined from the acute phase, there were important baseline differences between the S and O groups (selected variables are shown in Table 1). Patients in the S group were more likely to be female (53.50%), to have atrial fibrillation, and to have a low ejection fraction. Higher percentages of dissection and cardiac etiology were found in these patients, none of whom had lacunar stroke. Also, the proportion of patients with unknown day of onset of symptoms was higher in the the S group than in the O group, as was the proportion of patients with prestroke disability and pre-existing neurologic deficit. Hypertension and chronic ischemic lesions were less frequent in the S group, but occlusion or stenosis of the cervical and/or cerebral vessels in the ischemic territory was more frequent in the S group. Patients in the S group had higher heart rates and leukocytosis at admission, but lower body temperature, hemoglobin levels, and hematocrit.

Table 1.   Selected patient characteristics in the acute phase in the ‘severe’ and ‘others’ groups with univariate comparisons
Patient characteristicsAll (n = 1915)Severe (n = 243)Others (n = 1672)OR P-value
  1. mRS, modified Rankin scale; NIHSS, National Institute of Health Stroke Scale; OR, odds ratio; SD, standard deviation; TIA, transient ischemic attack. Values are expressed as median and interquartile range for continuous variables, or absolute count and percentage for categorical variables, unless stated otherwise.

 Age (years), mean (± SD)68.8 (15.4)69.6 (16.8)68.7 (15.2)1.000.37
 Female sex840 (43.9)130 (53.5)710 (42.5)0.64< 0.01
 Prestroke disability (measured with mRS)0 (1)0 (1)0 (1)1.170.01
NIHSS and its course
 Prestroke neurologic deficit (measured with NIHSS)0 (0)0 (0)0 (0)1.130.03
 Admission NIHSS score6 (11)23 (5)5 (7)
 NIHSS score at 4–6 h after arrival5 (9)23 (6)4 (7)1.55< 0.01
 NIHSS score at 24 h4 (9)22 (8)3 (6)1.31< 0.01
 NIHSS score at 7 days3 (8)22 (21)2 (4)1.15< 0.01
Risk factors (pre-existing or newly discovered)
 Hypertension1280 (66.9)140 (57.8)1140 (68.2)0.63< 0.01
 Diabetes mellitus303 (15.9)33 (13.7)270 (16.2)0.820.33
 Active smoking444 (23.5)46 (19.8)398 (24.0)0.780.16
 Atrial fibrillation482 (25.2)94 (38.8)388 (23.2)2.09< 0.01
 Coronary heart disease284 (14.8)37 (15.2)247 (14.7)1.030.86
 Low ejection fraction (< 35%)85 (4.5)20 (8.6)65 (3.9)2.270.02
 Previous cerebrovascular event (stroke, TIA, or retinal)28 (1.4)3 (1.2)25 (1.5)0.770.11
Medication at admission
 Antiplatelet681 (35.9)81 (33.8)600 (36.2)0.900.48
 Anticoagulant192 (10.1)29 (12.1)163 (9.8)1.260.28
 Antihypertensive1045 (55.2)123 (51.9)922 (55.7)0.850.27
 Hypolipemic438 (23.1)41 (17.2)397 (24.0)0.650.02
 Antidiabetic155 (8.1)14 (5.8)141 (8.5)0.660.16
Stroke onset during night sleep445 (23.2)49 (20.1)396 (23.7)0.810.22
Unknown stroke onset, not sleep97 (5.0)32 (13.1)65 (3.8)3.74< 0.01
Stroke mechanism
 Atherosclerosis254 (13.6)30 (12.5)224 (13.7)0.900.61
 Unknown458 (24.5)50 (20.9)408 (25.0)0.910.72
 Cardiac (including patent foramen ovale as the likely cause)617 (33.0)118 (49.3)499 (30.6)1.760.01
 Lacunar277 (14.8)0 (0)277 (17.0)
 Dissection92 (4.9)18 (7.5)74 (4.7)1.810.07
 Other and rare causes77 (4.1)11 (4.6)66 (4.0)1.240.56
 Multiple and coexisting causes92 (4.9)12 (5.0)80 (4.9)1.020.76
Radiologic findings
 Acute parenchymal findings632 (35.5)149 (63.4)483 (31.2)3.80< 0.01
 Chronic radiologic strokes514 (30.8)55 (24.5)459 (31.7)0.690.03
 Significant arterial findings in ischemic territory796 (49.9)616 (88.2)180 (44.2)9.43< 0.01
Metabolic values at admission
 Admission systolic blood pressure (mmHg), mean (± SD)158.87 (28.5)156.44 (31.4)159.22 (28.0)0.990.16
 Admission heart rate (min–1), mean (± SD)79.97 (18.0)83.78 (22.3)79.41 (17.2)1.01< 0.01
 Admission temperature (°C), mean (± SD)36.39 (0.6)36.26 (0.8)36.41 (0.6)0.70< 0.01
 Admission glucose (mm), mean (± SD)7.17 (2.9)7.55 (2.3)7.11 (3.0)1.030.05
 Admission hemoglobin (g L−1), mean (± SD)138.77 (32.8)132.86 (19.6)139.65 (34.3)0.98< 0.01
 Admission white cell count (103 L−1), mean (± SD)8.65 (4.4)9.83 (4.7)8.48 (4.3)1.06< 0.01
Interventions and outcome
 Intravenous thrombolysis302 (15.7)72 (29.6)230 (13.7)2.72< 0.01
 Endovascular treatment16 (0.8)7 (2.8)9 (0.5)6.77< 0.01
 Craniectomy15 (0.7)9 (3.7)6 (0.3)10.66< 0.01
 mRS score at 3 months2 (3)5 (3)1 (2)1.93< 0.01
 Favorable outcome (mRS score at 3 months of 0–3)1405 (73)68 (2)1337 (7)0.10< 0.01

Multivariate analysis yielded seven stroke-related, metabolic and radiologic variables that were independently associated with severe stroke (Table 2). The ROC-AUC for predicting severe strokes was 86% with the significant variables from the acute phase.

Table 2.   Multivariate analysis of factors significantly associated with S
FactorsOR (95% CI) P-value
  1. CI, confidence interval; OR, odds ratio.

Cardioembolic stroke1.74 (1.19–2.54)< 0.01
Unknown onset of stroke (but not wake-up strokes)2.35 (1.14–4.83)0.02
Early ischemic changes on initial imaging2.65 (1.79–3.92)< 0.01
Chronic strokes on imaging0.43 (0.25–0.71)< 0.01
Significant arterial findings in ischemic territory27.01 (11.59–62.92)< 0.01
Hemoglobin (g L−1)0.97 (0.96–0.98)< 0.01
White cell count (× 103 L−1)1.05 (1.00–1.10)0.04

Sixty-eight of 243 (28.0%) of consecutive severe stroke patients had a favorable outcome at 3 months. Selections of the 63 acute and 31 subacute variables examined are shown in Tables 3 and 4, respectively. In a multivariate analysis of the acute phase (Table 5), favorable outcome was predicted by lower age and lower initial temperature. Private insurance status, lower pre-existing functional handicap, lower glycemia and recanalization treatment also increased the probability of favorable outcome. When subacute variables were added into the multivariate analysis (Table 6), treatment with hypolipemic agents (mostly statins) increased the probability of favorable outcome. The pre-existing functional handicap was replaced by past cerebrovascular events, acute by subacute (24–48 h) glycemia, and recanalization treatment by (therapeutic or spontaneous) recanalization. The ROC-AUCs for predicting favorable outcome after severe stroke with the acute variables were 82% and 81% when the subacute variables were added.

Table 3.   Selected patient characteristics in the acute phase in the favorable outcome and unfavorable outcome groups, with univariate comparisons
Patient characteristicsAll patients (n = 243)Favorable outcome (mRS score of 0–3 at 3 months: n = 68)Unfavorable outcome (mRS score of 4–6 at 3 months: n = 175)OR P-value
  1. EVT, endovascular treatment; IQR, interquartile range; IVT, intravenous thrombolysis; mRS, modified Rankin scale; OR, odds ratio. Values are expressed as mean and standard deviation for continuous variables, or absolute count and percentage for categorical variables, unless stated otherwise.

Age (years), median (± IQR)69.6 (16.88)72.8 (16.1)61.4 (16.2)0.96< 0.01
Private insurance42 (17.4)17 (25.0)25 (14.4)1.970.06
mRS score prestroke, median (± IQR)0 (1)0 (0.5)1 (2)1.170.01
Hypolipemic treatment on admission41 (17.2)15 (22.7)26 (15.1)1.650.17
Unknown stroke onset (not sleep)32 (13.1)7 (10.2)25 (14.2)0.680.41
Acute parenchymal findings149 (63.4)37 (56.9)112 (65.8)0.680.20
Chronic radiologic strokes55 (24.5)13 (21.6)42 (825.6)0.800.54
Significant arterial findings in ischemic territory180 (88.2)56 (90.3)124 (87.3)1.350.54
Systolic blood pressure (mmHg)156.44 (31.4)149.41 (28.7)159.19 (32.0)0.980.03
Temperature (°C)36.26 (0.8)36.07 (0.6)36.33 (0.8)0.630.02
Glucose (mm)7.55 (2.3)6.85 (1.8)7.82 (2.5)0.810.01
White blood cell count (× 103 L−1)9.83 (4.7)9.93 (4.7)9.79 (4.7)1.000.84
Total cholesterol (mm)5.59 (4.4)5.08 (1.5)5.77 (5.0)0.940.37
IVT72 (29.6)32 (47.0)40 (22.8)3.30< 0.01
EVT7 (2.8)3 (5.8)3 (1.7)5.500.03
IVT or EVT79 (32.5)36 (52.9)43 (24.5)3.45< 0.01
Timing of intervention (min)152.1 (91.6)13.41 (4.9)13.00 (5.4)1.000.39
Craniectomy9 (3.7)3 (4.4)6 (3.4)1.300.72
Table 4.   Selected factors of the acute phase in the favorable outcome and unfavorable outcome groups, with univariate comparisons
Patient characteristicsAll patientsFavorable outcome (mRS score of 0–3 at 3 months)Unfavorable outcome (mRS score of 4–6 at 3 months)OR P-value
  1. IQR, interquartile range; mRS, modified Rankin scale; OR, odds ratio; SD, standard deviation. Values are expressed as mean and SD for continuous variables, or absolute count and percentage for categorical variables, unless stated otherwise. *Not included in the multivariate analysis.

N (%)243 (100)69 (28)174 (72)
Subacute imaging findings at 24–48 h
 Any stroke-related findings (ischemia and/or hemorrhage)*150 (96.1)51 (92.7)99 (98.0)0.250.13
 Any hemorrhagic transformation (symptomatic or not)29 (18.9)10 (18.5)19 (19.1)0.950.919
 Symptomatic hemorrhagic transformation according to ECASS-2 definition6 (3.9)0 (0.0)6 (6.0)
Subacute vascular imaging at 24–48 h
 Significant arterial findings in ischemic territory78 (67.2)27 (60.0)51 (71.8)0.580.19
 Partial or complete recanalization of cervical and/or cerebral arteries (where assessed)66 (28.3)35 (51.4)31 (18.7)4.58< 0.01
 Excracranial recanalization (complete, partial)20 (8.5)6 (8.8)14 (8.4)1.050.92
 Intracranial recanalization (complete, partial)63 (27.1)37 (54.4)26 (15.8)6.33< 0.01
Subacute metabolic values at 24–48 h
 Systolic blood pressure (mmHg)143.47 (21.9)139.78 (21.9)145.07 (21.8)0.980.1
 Diastolic blood pressure (mmHg)68.70 (15.6)67.55 (15.7)69.19 (15.5)0.990.472
 Heart rate (min–1)82.55 (18.3)76.25 (16.6)85.26 (18.3)0.97< 0.01
 Temperature (°C)37.01 (0.7)36.89 (0.7)37.06 (0.6)0.69< 0.01
 Glucose (mm)6.41 (1.9)5.93 (1.2)6.65 (2.1)0.770.02
 Serum creatinine (mm)89.53 (71.7)79.19 (25.4)93.48 (82.6)0.990.16
 Hemoglobin (g L−1)126.59 (19.5)124.48 (20.7)127.53 (19.0)0.990.4
 Hematocrit mean (± SD)38.04 (5.3)37.19 (5.3)38.41 (5.3)0.950.22
 Total cholesterol (mm)8.12 (9.3)4.84 (1.0)9.76 (11.2)0.900.35
 White blood cell count (× 103 L−1)11.21 (3.9)10.69 (3.3)11.45 (4.1)0.940.29
 Platelet count (× 103 L–1) (mean)215.95 (74.3)218.71 (67.4)214.71 (77.5)1.000.77
 NIHSS score at 7 days, median (± IQR)22 (21)12 (15)24 (24)0.85< 0.01
 Death84 (36.5)0 (0)84 (50.9)
 Home11 (4.7)9 (13.8)2 (1.2)13< 0.01
 Long-term institution14 (6.0)0 (0)14 (8.4)
 Transfer to other acute-care service or hospital41 (17.8)19 (29.2)22 (13.3)2.68< 0.01
 Rehabilitation hospital80 (34.7)37 (56.9)43 (26.0)3.74< 0.01
Table 5.   Multivariate analysis of acute-phase factors significantly associated with favorable ourtcome
FactorsOR (95% CI) P-value
  1. CI, confidence interval; EVT, endovascular treatment; IVT, intravenous thrombolysis; mRS, modified Rankin scale; OR, odds ratio.

Age0.95 (0.92–0.97)< 0.01
Prestroke handicap (mRS)0.62 (0.40–0.95)0.03
Private insurance status2.51 (1.05–5.98)0.037
Acute temperature0.42 (0.23–0.74)< 0.01
Acute glucose value0.74 (0.61–0.90)< 0.01
Acute EVT within recommended time limits (IVT or EVT)2.47 (1.20–5.08)0.01
Table 6.   Multivariate analysis of factors of the acute and subacute phases combined that are significantly associated with favorable outcomeF
FactorOR (95% CI) P-value
  1. CI, confidence interval; OR, odds ratio; TIA, transient ischemic attack.

Age0.94 (0.91–0.97)< 0.01
Previous cerebrovascular event (stroke, TIA, retinal ischemia)3.00 (1.00–8.96)0.04
Hypolipemic treatement at stroke onset3.82 (1.34–10.89)0.01
Acute temperature0.42 (0.23–0.78)< 0.01
Subacute glucose values at 24–48 h0.73 (0.56–0.96)0.02
Recanalization of cervical and/or intracranial arteries (partial or complete)4.51 (1.95–10.40)< 0.01


  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Disclosure of Conflict of Interests
  9. References

Using a large number of sociodemographic, clinical, radiologic and metabolic variables, we have identified seven factors that are associated with severe stroke at the time of emergency room admission. Furthermore, we found that favorable outcome despite the patient initially presenting with severe stroke was associated, with a high degree of confidence, with demographic and metabolic factors and early recanalization.

When examining the profile of patients who suffer severe stroke, we found cardio-embolic origin of stroke to be significantly associated with these patients, similarly to previous studies [19–21]. This may be explained by the usual size of cardiac emboli as well as by the lack of collateral vascularization, which may develop and compensate for acute arterial occlusion in patients with gradual occlusion of arteries, such as in atherosclerosis of cervical or cerebral arteries. In the univariate, but not the multivariate, analysis, we also found dissection to be associated with severe stroke. It is of note that there were no patients with lacunar stroke among those presenting with severe stroke.

We found that severe strokes were more frequently of unknown time of onset, but only if they were not associated with sleep onset. This is possibly related to the fact that such patients were not alert and were unable to express details of their stroke onset. The reason why this is not the case for sleep-onset stroke may be that such strokes are more frequently lacunar [22]. Severe strokes were more frequent in patients with lower hemoglobin levels, suggesting that decreased oxygen transport capacity may increase infarct size, consistent with a previous report [23]. White cell count has previously been associated with a more severe clinical presentation and with a worse early stroke outcome [24]. It is not clear whether this is a direct effect of white cell activity and inflammation in the lesioned brain, or whether it is rather a marker of associated conditions (e.g. fever, early infection, or stress response) that may increase stroke severity and worsen outcome.

Radiologic data revealed that early signs of parenchymal ischemia and notable clinical findings within arteries of the ischemic region (e.g. occlusion or > 50% stenosis) are observed significantly more frequently in patients with severe stroke, again confirming the findings of previous studies [25,26].

We also found that chronic strokes on neuroimaging are less frequent among severe stroke patients. This seems counterintuitive, given that functional compensation for cerebral damage depends partially on the remaining ‘healthy’ brain. However, patients who have less severe strokes might have a higher likelihood of having survived other minor strokes in the past. In addition, such survivors of previous strokes may receive better secondary prevention treatments, which have been shown to decrease not only recurrence, but also the severity of recurrent strokes [27–30].

Although female gender and increased heart rate were more frequent in those patients presenting with severe stroke by univariate analysis, these factors were not independent. Similarly, pre-existing hypertension and pretreatment with hypolipemic agents were negatively associated with severe stroke in the univariate, but not in the multivariate, analysis. Interestingly, age, pre-existing risk factors, such as hypertension, diabetes, or smoking, and insurance status were not associated with severe stroke.

As an additional observation, the in-hospital course of NIHSS score differed between the two groups: in the less severe strokes, NIHSS score measured at 4 h, 24 h and 7 days showed a decreasing pattern from the beginning, which was not the case for severe strokes, where the NIHSS score remained elevated for the first 7 days. This result is probably attributable to an increased rate of complications in severe stroke patients, such as brain edema, hemorrhagic transformation (HT), and medical complications [1,2,31,32]. Regarding HT according to the ECASS-2 definition, unfavorable outcome was associated with greater HT.

Given that stroke severity is one of the most significant independent predictors of poor short-term and long-term outcome [33–35], we were interested in analyzing predictors of favorable outcome after 3 months in patients with severe stroke.

As expected, higher age and prestroke handicap (as measured by previous mRS score) were associated with a reduced likelihood of achieving a favorable outcome at 3 months in severe stroke patients. A reduction in the number of synapses in old age [36] and a diminished volume of cortical gray matter [37] may hinder recovery, because of limited neuronal plasticity. Also, acute initial temperature and elevated glucose levels had a negative influence on outcome, as previously shown [38,39]. The relationship between temperature, inflammation, stroke severity and stroke outcome is currently being elucidated [40]. However, stroke severity may be more frequently associated with dysphagia during the acute phase, which may lead to aspiration pneumonia and further worsening of stroke outcome [41].

Private insurance status – probably indicating a higher socioeconomic status – has previously been show to predict favorable outcome in stroke patients [42,43]. The reasons for this observation are not certain, although one of our previous studies showed lower initial stroke severity in privately insured patients [42]; this was not confirmed in the current dichotomized study.

It is encouraging that recanalization also favorably influenced the fate of severe stroke patients. It has been hypothesized that the decreased effect of intravenous thrombolysis in severe stroke may be related to insufficient recanalization [44,45]. Most patients with severe stroke have large occluding thrombi in intracranial arteries, which may need endovascular treatment for effective recanalization [46]; although only few of our patients underwent such treatments, an effective recanalization strategy should be offered to severe stroke patients, particularly if the relationship between viable brain tissue and core is favorable [47].

When we added covariates from the subacute phase, we found that previous clinical cerebrovascular events (stroke, transient ischemic attack, and retinal ischemia) were independently associated with favorable outcome at 3 months. This was not explained by the stroke mechanism, but could indicate that patients who recovered from a previous stroke (rather than died) also had a better chance of recovery from a subsequent severe stroke; this could be partly explained by the genetics of neurovascular factors, the anatomy of the cervico-cerebral vasculature, including collaterals, or ischemic preconditioning [48].

As recently shown for ischemic stroke of any severity, pretreatment with hypolipemic agents (mostly statins in our population) was also associated with favorable outcome [49]. This may be explained by the pleiotropic effects of statins, especially on the cerebral vasculature. Finally, hyperglycemia in the subacute phase seemed to be associated with unfavorable outcome at least as strongly as in the acute phase. This confirms previous observations of our group [50].

Interestingly, pre-existing risk factors and stroke etiology were not associated with 3-month outcome.

One limitation of our work is its observational, single-center design. Furthermore, we did not include a detailed analysis of perfusion imaging studies or of collaterals in our analysis, even though such factors have been shown to be linked to outcome prediction in stroke patients [51,52]. The fact that only a few patients had endovascular treatment limited the ability to explore its effects; recanalization (spontaneous or thrombolysis-related) was frequently measured, however, and showed significant associations with favorable outcome. We also accept that the criterion of an NIHSS score of ≥ 20 leads to a slight underrepresention of right hemispheric strokes (67 = 33% in the S group vs. 716 = 46% in the O group), owing to the inherent definition of this scale.

In summary, our results show that prestroke characteristics of the patient, stroke etiology, hematologic values and acute radiologic and arterial findings are associated with severe stroke. Although some of the results were expected, others are new, and most can be influenced before or after onset of the severe stroke. As expected, severe stroke patients had a rather poor prognosis. Nevertheless, several factors associated with favorable outcome, such as lower acute temperature, lower glucose levels at 24 h, previous cerebrovascular events, and arterial recanalization, should encourage clinicians to treat such patients aggressively and to search for new treatment modalities.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Disclosure of Conflict of Interests
  9. References

We thank A. Eskandari and S. d’Ambrogio-Remillard for their help with data collection. We thank G. Ntaios for his valuable comments on the manuscript.

Disclosure of Conflict of Interests

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Disclosure of Conflict of Interests
  9. References

The authors state that they have no conflict of interest.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Disclosure of Conflict of Interests
  9. References
  • 1
    Qureshi AIM, Suarez JIM, Yahia AMM, Mohammad YM, Uzun GM, Suri MF, Zaidat OOM, Ayata CM, Ali ZM, Wityk RJM. Timing of neurologic deterioration in massive middle cerebral artery infarction: a multicenter review. Crit Care Med 2003; 31: 2727.
  • 2
    Hacke W, Schwab S, Horn M, Spranger M, de Georgia M, von Kummer R. ‘Malignant’ middle cerebral artery territory infarction: clinical course and prognostic signs. Arch Neurol 1996; 53: 30915.
  • 3
    Berrouschot J, Rossler A, Koster J, Schneider D. Mechanical ventilation in patients with hemispheric ischemic stroke. Crit Care Med 2000; 28: 295661.
  • 4
    Kasner SE, Demchuk AM, Berrouschot J, Schmutzhard E, Harms L, Verro P, Chalela JA, Abbur R, McGrade H, Christou I, Krieger DW. Predictors of fatal brain edema in massive hemispheric ischemic stroke. Stroke 2001; 32: 211723.
  • 5
    de Jong G, van Raak L, Kessels F, Lodder J. Stroke subtype and mortality. A follow-up study in 998 patients with a first cerebral infarct. J Clin Epidemiol 2003; 56: 2628.
  • 6
    Meijer R, Ihnenfeldt DS, van Limbeek J, Vermeulen M, de Haan RJ. Prognostic factors in the subacute phase after stroke for the future residence after six months to one year. A systematic review of the literature. Clin Rehabil 2003; 17: 51220.
  • 7
    Hankey GJ. Long-term outcome after ischaemic stroke/transient ischaemic attack. Cerebrovasc Dis 2003; 16: 1419.
  • 8
    Barber PA, Demchuk AM, Zhang J, Buchan AM. Validity and reliability of a quantitative computed tomography score in predicting outcome of hyperacute stroke before thrombolytic therapy. Aspects Study Group. Alberta Stroke Programme early CT score. Lancet 2000; 355: 16704.
  • 9
    Wardlaw JM, Mielke O. Early signs of brain infarction at CT: observer reliability and outcome after thrombolytic treatment--systematic review. Radiology 2005; 235: 44453.
  • 10
    Barber PA, Demchuk AM, Zhang J, Kasner SE, Hill MD, Berrouschot J, Schmutzhard E, Harms L, Verro P, Krieger D. Computed tomographic parameters predicting fatal outcome in large middle cerebral artery infarction. Cerebrovasc Dis 2003; 16: 2305.
  • 11
    Ikeda K, Kano O, Ito H, Kawase Y, Iwamoto K, Sato R, Sekine T, Nagata R, Nakamura Y, Hirayama T, Iwasaki Y. Diagnostic pitfalls in sporadic transthyretin familial amyloid polyneuropathy (TTR-FAP). Neurology 2008; 70: 1576; author reply 1576–7.
  • 12
    von Kummer R, Bourquain H, Bastianello S, Bozzao L, Manelfe C, Meier D, Hacke W. Early prediction of irreversible brain damage after ischemic stroke at CT. Radiology 2001; 219: 95100.
  • 13
    Lyden P, Brott T, Tilley B, Welch KM, Mascha EJ, Levine S, Haley EC, Grotta J, Marler J. Improved reliability of the NIH stroke scale using video training. NINDS TPA Stroke Study Group. Stroke 1994; 25: 22206.
  • 14
    Chalela JA, Kidwell CS, Nentwich LM, Luby M, Butman JA, Demchuk AM, Hill MD, Patronas N, Latour L, Warach S. Magnetic resonance imaging and computed tomography in emergency assessment of patients with suspected acute stroke: a prospective comparison. Lancet 2007; 369: 2938.
  • 15
    Wardlaw JM, Keir SL, Seymour J, Lewis S, Sandercock PA, Dennis MS, Cairns J. What is the best imaging strategy for acute stroke? Health Technol Assess 2004; 8: ix-x, 1-180.
  • 16
    Michel P, Odier C, Rutgers M, Reichhart M, Maeder P, Meuli R, Wintermark M, Maghraoui A, Faouzi M, Croquelois A, Ntaios G. The Acute STroke Registry and Analysis of Lausanne (ASTRAL): design and baseline analysis of an ischemic stroke registry including acute multimodal imaging. Stroke 2010; 41: 24918.
  • 17
    Adams HP Jr, Bendixen BH, Kappelle LJ, Biller J, Love BB, Gordon DL, Marsh EE III. Classification of subtype of acute ischemic stroke. Definitions for use in a multicenter clinical trial. Toast. Trial of org 10172 in acute stroke treatment. Stroke 1993; 24: 3541.
  • 18
    Committee TESOEECatEW. Guidelines for management of ischaemic stroke and transient ischaemic attack 2008 (update from 16.03.2009 on http://www.Eso-stroke.Org). Cerebrovasc Dis 2008; 25: 457507.
  • 19
    Bounds JV, Wiebers DO, Whisnant JP, Okazaki H. Mechanisms and timing of deaths from cerebral infarction. Stroke 1981; 12: 4747.
  • 20
    Yoshimoto T, Ogawa A, Seki H, Kogure T, Suzuki J. Clinical course of acute middle cerebral artery occlusion. J Neurosurg 1986; 65: 32630.
  • 21
    Appelros P, Nydevik I, Seiger A, Terent A. Predictors of severe stroke: influence of preexisting dementia and cardiac disorders. Stroke 2002; 33: 235762.
  • 22
    Spengos K, Tsivgoulis G, Manios E, Synetou M, Vassilopoulou S, Zakopoulos N, Vassilopoulos D, Vemmos KN. Stroke etiology is associated with symptom onset during sleep. Sleep 2005; 28: 2338.
  • 23
    Kramer AH, Zygun DA. Anemia and red blood cell transfusion in neurocritical care. Crit Care 2009; 13: R89.
  • 24
    Ning YP, Kanai K, Tomiyama H, Li Y, Funayama M, Yoshino H, Sato S, Asahina M, Kuwabara S, Takeda A, Hattori T, Mizuno Y, Hattori N. Park9-linked parkinsonism in eastern Asia: mutation detection in atp13a2 and clinical phenotype. Neurology 2008; 70: 14913.
  • 25
    Hill MD, Rowley HA, Adler F, Eliasziw M, Furlan A, Higashida RT, Wechsler LR, Roberts HC, Dillon WP, Fischbein NJ, Firszt CM, Schulz GA, Buchan AM. Selection of acute ischemic stroke patients for intra-arterial thrombolysis with prourokinase by using ASPECTS. Stroke 2003; 34: 192531.
  • 26
    Puetz V, Dzialowski I, Hill MD, Subramaniam S, Sylaja PN, Krol A, O’Reilly C, Hudon ME, Hu WY, Coutts SB, Barber PA, Watson T, Roy J, Demchuk AM. Intracranial thrombus extent predicts clinical outcome, final infarct size and hemorrhagic transformation in ischemic stroke: the clot burden score. Int J Stroke 2008; 3: 2306.
  • 27
    Marti-Fabregas J, Gomis M, Arboix A, Aleu A, Pagonabarraga J, Belvis R, Cocho D, Roquer J, Rodriguez A, Garcia MD, Molina-Porcel L, Diaz-Manera J, Marti-Vilalta JL. Favorable outcome of ischemic stroke in patients pretreated with statins. Stroke 2004; 35: 111721.
  • 28
    Fuentes B, Martinez-Sanchez P, Diez-Tejedor E. Lipid-lowering drugs in ischemic stroke prevention and their influence on acute stroke outcome. Cerebrovasc Dis 2009; 27(Suppl. 1): 12633.
  • 29
    Dessein PH, Joffe BI, Stanwix AE. Effects of disease modifying agents and dietary intervention on insulin resistance and dyslipidemia in inflammatory arthritis: a pilot study. Arthritis Res 2002; 4: R12.
  • 30
    Lakhan SE, Sapko MT. Blood pressure lowering treatment for preventing stroke recurrence: a systematic review and meta-analysis. Int Arch Med 2009; 2: 30.
  • 31
    Vahedi K, Hofmeijer J, Juettler E, Vicaut E, George B, Algra A, Amelink GJ, Schmiedeck P, Schwab S, Rothwell PM, Bousser MG, van der Worp HB, Hacke W. Early decompressive surgery in malignant infarction of the middle cerebral artery: a pooled analysis of three randomised controlled trials. Lancet Neurol 2007; 6: 21522.
  • 32
    Michel P, Arnold M, Hungerbuhler HJ, Muller F, Staedler C, Baumgartner RW, Georgiadis D, Lyrer P, Mattle HP, Sztajzel R, Weder B, Tettenborn B, Nedeltchev K, Engelter S, Weber SA, Basciani R, Fandino J, Fluri F, Stocker R, Keller E, et al. Decompressive craniectomy for space occupying hemispheric and cerebellar ischemic strokes: Swiss recommendations. Int J Stroke 2009; 4: 21823.
  • 33
    Sato I, Wu S, Ibarra MC, Hayashi YK, Fujita H, Tojo M, Oh SJ, Nonaka I, Noguchi S, Nishino I. Congenital neuromuscular disease with uniform type 1 fiber and ryr1 mutation. Neurology 2008; 70: 11422.
  • 34
    Wahlgren N, Ahmed N, Eriksson N, Aichner F, Bluhmki E, Davalos A, Erila T, Ford GA, Grond M, Hacke W, Hennerici MG, Kaste M, Kohrmann M, Larrue V, Lees KR, Machnig T, Roine RO, Toni D, Vanhooren G. Multivariable analysis of outcome predictors and adjustment of main outcome results to baseline data profile in randomized controlled trials: Safe Implementation of Thrombolysis in Stroke-Monitoring Study (SITS-MOST). Stroke 2008; 39: 331622.
  • 35
    Schonewille WJ, Wijman CA, Michel P, Rueckert CM, Weimar C, Mattle HP, Engelter ST, Tanne D, Muir KW, Molina CA, Thijs V, Audebert H, Pfefferkorn T, Szabo K, Lindsberg PJ, de FG, Kappelle LJ, Algra A. Treatment and outcomes of acute basilar artery occlusion in the Basilar Artery International Cooperation Study (BASICS): a prospective registry study. Lancet Neurol 2009; 8: 72430.
  • 36
    Adams I. Comparison of synaptic changes in the precentral and postcentral cerebral cortex of aging humans: a quantitative ultrastructural study. Neurobiol Aging 1987; 8: 20312.
  • 37
    Good CD, Johnsrude IS, Ashburner J, Henson RN, Friston KJ, Frackowiak RS. A voxel-based morphometric study of ageing in 465 normal adult human brains. Neuroimage 2001; 14: 2136.
  • 38
    Boysen G, Christensen H. Stroke severity determines body temperature in acute stroke. Stroke 2001; 32: 41317.
  • 39
    Ntaios G, Egli M, Faouzi M, Michel P. J-shaped association between serum glucose and functional outcome in acute ischemic stroke. Stroke 2010; 41: 236670.
  • 40
    Macrez R, Ali C, Toutirais O, Le Mauff B, Defer G, Dirnagl U, Vivien D. Stroke and the immune system: from pathophysiology to new therapeutic strategies. Lancet Neurol 2011; 10: 47180.
  • 41
    Katzan IL, Cebul RD, Husak SH, Dawson NV, Baker DW. The effect of pneumonia on mortality among patients hospitalized for acute stroke. Neurology 2003; 60: 6205.
  • 42
    Rey V, Faouzi M, Huchmand-Zadeh M, Michel P. Stroke initial severity and outcome relative to insurance status in a universal health care system in Switzerland. Eur J Neurol 2011; 18: 10947.
  • 43
    Ottman R, Rosenberger L, Bagic A, Kamberakis K, Ritzl EK, Wohlschlager AM, Shamim S, Sato S, Liew C, Gaillard WD, Wiggs E, Berl MM, Reeves-Tyer P, Baker EH, Butman JA, Theodore WH. Altered language processing in autosomal dominant partial epilepsy with auditory features. Neurology 2008; 71: 197380.
  • 44
    Ntaios G, Faouzi M, Michel P. The effect of thrombolysis on short-term improvement depends on initial stroke severity. J Neurol 2012; 259: 5249.
  • 45
    Ingall TJ, O’Fallon WM, Asplund K, Goldfrank LR, Hertzberg VS, Louis TA, Christianson TJ. Findings from the reanalysis of the NINDS tissue plasminogen activator for acute ischemic stroke treatment trial. Stroke 2004; 35: 241824.
  • 46
    Mattle HP, Arnold M, Georgiadis D, Baumann C, Nedeltchev K, Benninger D, Remonda L, von Büdingen C, Diana A, Pangalu A, Schroth G, Baumgartner RW. Comparison of intraarterial and intravenous thrombolysis for ischemic stroke with hyperdense middle cerebral artery sign. Stroke 2008; 39: 37983.
  • 47
    Parsons MW, Christensen S, McElduff P, Levi CR, Butcher KS, De Silva DA, Ebinger M, Barber PA, Bladin C, Donnan GA, Davis SM. Pretreatment diffusion- and perfusion-MR lesion volumes have a crucial influence on clinical response to stroke thrombolysis. J Cereb Blood Flow Metab 2010; 30: 121425.
  • 48
    Dirnagl U, Simon RP, Hallenbeck JM. Ischemic tolerance and endogenous neuroprotection. Trends Neurosci 2003; 26: 24854.
  • 49
    Lakhan SE, Bagchi S, Hofer M. Statins and clinical outcome of acute ischemic stroke: a systematic review. Int Arch Med 2010; 3: 22.
  • 50
    Ntaios G, Abatzi C, Alexandrou M, Chatzopoulos S, Egli M, Ruiz J, Bornstein N, Michel P. Persistent hyperglycemia at 24–48 hours in acutely hyperglycemic stroke patients is not associated with worse functional outcome. Cerebrovasc Dis 2011; 32: 5616.
  • 51
    Kim JT, Park MS, Choi KH, Nam TS, Choi SM, Lee SH, Kim BC, Kim MK, Cho KH. The CBV-ASPECT score as a predictor of fatal stroke in a hyperacute state. Eur Neurol 2010; 63: 35763.
  • 52
    Odier C, Michel P. Recanalization and Collaterals Predict Outcome in Proximal MCA Occlusion, but neither Penumbra nor Core of Stroke [poster]. Int Stroke Conf 2011; Los Angeles.