Original Contribution

Stroke in elderly patients: management and prognosis in the ED

Viviane Montout MD^a, Beatrice Madonna-Py MD^a, Marie-Odile Josse MD^a, Igor Ondze MD^a, Amandine Arhan MD^a, Sophie Crozier MD^b,

Pierre Hausfater MD^a, Bruno Riou MD, PhD^a, Jacques Boddaert MD, PhD^a,c,?

^aDepartment of Emergency Medicine and Surgery, Centre Hospitalo-Universitaire (CHU) Pitie-Salpetriere,

Assistance-Publique Hopitaux de Paris (AP-HP), Universite Pierre et Marie Curie-Paris 6, Paris, France

^bStroke Center, Centre Hospitalo-Universitaire (CHU) Pitie-Salpetriere, Assistance-Publique Hopitaux de Paris (AP-HP),

Universite Pierre et Marie Curie-Paris 6, Paris, France

^cGeriatric Center, Centre Hospitalo-Universitaire (CHU) Pitie-Salpetriere, Assistance-Publique Hopitaux de Paris (AP-HP), Universite Pierre et Marie Curie-Paris 6, Paris, France

Received 19 June 2007; revised 13 October 2007; accepted 14 October 2007

Abstract

Purpose: The aim of the study was to analyze the effect of age on management and prognosis of stroke. Methods: A retrospective study was performed in consecutive patients admitted to an emergency department (ED) with a diagnosis of stroke. Comparison according to age (b75 vs >=75 years old) was done, with a 1-year follow-up including autonomy and outcome.

Results: In older patients, Brain magnetic resonance imaging (6% vs 27%, P b .001) and immediate referral to the stroke unit were less frequent (6% vs 28%, P b .001); median length of stay was longer (11 vs 8 days, P = .007); and in-hospital mortality tended to be higher (12% vs 6%, not significant). After discharge, 1-year mortality was higher (27% vs 14%, P = .004). In a multivariate analysis, severity of stroke, hemorrhagic stroke, and dementia were independent positive predictors of 1-year mortality, whereas age was not.

Conclusion: Despite the fact that age was not an independent predictor when Stroke severity was considered, our data suggest differences in the management of elderly compared with younger stroke patients admitted to the ED.

(C) 2008

Introduction

Stroke is caused by disruption of blood circulation in the brain and represents the leading cause of neurologic

* Corresponding author. Centre de geriatrie, et Service d’Accueil des Urgences et service de geriatrie, Groupe Hospitalier Pitie-Salpetriere, 75013 Paris, France. Tel.: +33 1 42 17 72 49; fax: +33 1 42 17 72 64.

E-mail address: [email protected] (J. Boddaert).

disability in the elderly [1-3]. As the proportion of the elderly in the population is expected to dramatically increase, studies regarding the effect of age on different medical diseases and conditions are necessary to determine appro- priate management. Stroke incidence has declined in persons aged younger than 75 years [4], but prevalence is still high in elderly patients; and stroke represents one of the leading causes of death in this population [5]. However, data regarding diagnostic and management decisions in older

0735-6757/$ - see front matter (C) 2008 doi:10.1016/j.ajem.2007.10.023

patients who present with signs and symptoms of stroke are scarce.

Previous reports showed differences in management concerning computed tomographic (CT) scan use and care access [6] and treatments including antithrombotic and lipid- lowering agents [7]. In practice, the best criteria of stroke management include new parameters (time before medical care, magnetic resonance imaging [MRI], and thrombolysis) that were not considered at that time. To our knowledge, no reported study has previously investigated the effect of age on stroke management in an urban teaching hospital.

In France, most of strokes in elderly patients are referred to and treated in an emergency department (ED). The aim of the present study was to analyze the effect of age on management and prognosis of stroke, and to identify factors affecting stroke survival in elderly patients referred initially to an ED. Time to be seen by a physician, CT scan and MRI use, antithrombotic and thrombolysis agents use, care and in particular access to a stroke center were the variables used to assess stroke management. Length of stay, autonomy, 3-month hospital-free days, and in-hospital and 1-year mortality were the outcome measures of interest.

Material and methods

The study was designed as a retrospective study performed during a period from January 2003 to July 2003 to analyze age influence on stroke management and prognosis in elderly patients admitted to an ED with a diagnosis of stroke. In 2005, we completed this study to determine patient outcome, including autonomy and death. In compliance with French law, this observational study did not require approval from the Comite de Protection des Personnes (equivalent to the ethical review Board). We included all consecutive patients with acute stroke admitted to the ED of a 1826-bed teaching urban hospital. All patients presenting with primary diagnosis of stroke were included in our cohort. Primary diagnosis of stroke was defined as rapidly developing signs of focal or global disturbance of cerebral function either transient or lasting for more than 24 hours, or leading to death, with no apparent cause other than that of vascular origin [8,9].

For every patient, medical care by the physician in charge included medical history especially Cardiovascular risk factors, previous neurologic, cognitive and cardiac diseases, living conditions (home living, institution), autonomy (daily help, and daily activity using the validated activities of daily living scale [ADL] [10]), physical examination, and brain imaging. Autonomy, as ascertained by questionnaire, was defined as an ADL score = 6. Hypertension was considered present if a patient had known hypertension or treatment of hypertension at admission. Diabetes was considered present if a patient had known diabetes or treatment of diabetes on admission. History of atrial fibrillation was considered

present if present on the electrocardiogram at admission. Diagnosis of stroke was performed according to the World Health Organization definition, using clinical and imaging data from ED and discharge units including neurologic, geriatric, and other units. Data sources were obtained from complete ED medical charts of patients admitted for suspicion of stroke, hospitalization reports, and medical charts from all health care services. Stroke severity was assessed at admission using the validated Modified Rankin scale because of its feasibility in ED [11]. All final diagnoses were performed according to in-hospital management and final conclusions before out-hospital discharge of patients. Patients were compared according to their age, those younger than 75 years classified into the younger group and those 75 years or older into the older group.

Management analysis

Criteria were chosen for management analysis, including time to be seen by a physician in the ED; emergency stroke center admission; emergency brain imaging including CT scan and MRI; treatments including antithrombotic, antic- oagulant treatment and fibrinolysis; and admission after ED to a stroke center, or neurologic, geriatric, or intensive care unit. All decisions were made by the emergency physician according to standard clinical practice. All events occurring in the study population were identified by monitoring of all inpatient health services. After hospitalization report analy- sis, missing data were obtained from the complete patient chart and, when necessary, by a phone call to the health care services physician. This monitoring was performed by a physician who was not involved in immediate management during the study (IO).

Outcome analysis

We determined the length of stay in the hospital and in- hospital mortality. To enable us to compare the length of hospitalization between the 2 groups while taking into account the deaths, we calculated the number of hospital-free days during the 3-month period after admission to the ED, all dead patients being scored 0 hospital-free days, as previously described [12]. In 2005, to analyze outcome for each living patient after stroke discharge, a phone call and a written questionnaire to the patient himself or herself or to the family or caregiver was used. This questionnaire, including parameters about autonomy (home living, needs for help, institution, daily living activity) and evolution (any new admission) during the year after stroke, was administered. At this stage, we obtained information on ADL status for demented patients from the correspondents.

Data are expressed as number of patients (percentage) and mean +- SD for normally distributed variables, and median and 95% confidence interval (CI) for nonnormally distrib- uted variables (Kolmogorov-Smirnov test). We used single-

entry data methods. Comparison of proportions was performed using the ?2 test or the Fisher exact method, comparison of 2 means was performed using the Student t test, and comparison of 2 medians was performed using the Mann-Whitney test. Survival was estimated by the Kaplan- Meier method, and differences in survival between groups were assessed by the log-rank test. For subgroup analysis, the Bonferroni correction was applied to the log-rank test.

Univariate and multivariate Cox proportional-hazards models were used to determine variables associated with stroke death within 1 year after admission. To avoid overfitting, we used a conservative approach and included only the significant variables in the univariate analysis (P value of entry b.10). The type of stroke was entered as hemorrhagic vs ischemic, the reference being a transient stroke. Interactions were not tested. The Spearman coefficient matrix correlation was used to identify significant colinearity (N0.70) between variables. A back- ward-elimination model was applied. The hazards ratio and the 95% CI of variables selected by the Cox model were calculated. Then, we compared the adjusted estimates of survival between young and elderly patients. We compared the estimates of survival with expected survival of the population studied in France, obtained from demographic data corresponding to the period study (2003-2005) [13].

All P values were 2-tailed, and a P value of less than

.05 was considered significant. Statistical analysis was performed using NCSS 2004 (Statistical Solutions Ltd, Corve, Ireland).

Results

During the study period, we identified 253 patients in our ED with a suspected stroke. After the exclusion of 47 patients (19%) with a diagnosis other than stroke (n = 29) and without final diagnosis (n = 18), 206 patients met the inclusion criterion and were included in our study with a final diagnosis of stroke.

Baseline characteristics are represented in Table 1. Hypertension, atrial fibrillation, coronary artery disease, dementia, and antithrombotic or anticoagulant treatment were significantly more frequent in elderly patients, but tobacco smoking was significantly less observed. Symptom duration was significantly different between the younger and older groups, but there was no significant difference for patients arriving before 3 hours after the onset of stroke (P =

.63). Type of stroke (ischemic or hemorrhagic) was similar between groups (Table 1).

Management of stroke

All older patients were managed within the first hour after admission, and more than 95% received a CT scan (Table 2). There was no significant difference for CT scan use and

Younger group (n = 101)		Older group (n = 105)	P
Age (y)	59 +- 13	83 +- 5	-
Men	36 (36)	49 (47)	.12
Women	65 (65)	56 (53)
Previous history
Hypertension	52 (51)	74 (70)	.005
Dyslipidemia	37 (37)	43 (41)	.57
Diabetes mellitus	22 (22)	19 (18)	.6
Smoking	53 (52)	37 (35)	.02
Dementia	10 (10)	24 (23)	.01
Previous stroke	20 (20)	27 (26)	.32
Coronary artery disease	6 (6)	21 (20)	.003
Atrial fibrillation	16 (16)	30 (29)	.03
Symptoms duration			.04
b3 h	27 (27)	24 (23)	.63
3-24 h	33 (33)	35 (33)
N24 h	29 (29)	23 (22)
Unknown	12 (12)	23 (22)
Treatment
None	70 (69)	48 (46)	b.001
Antithrombotic	23 (23)	42 (41)	.01
Anticoagulant	6 (6)	15 (15)	.06
Both	28 (29)	55 (53)	b.001
Unknown	3 (3)	2 (2)
Type of stroke
Ischemic	76 (75)	72 (69)
Hemorrhagic	7 (7)	8 (8)	.54
Transient	18 (18)	25 (24)
Living conditions	(n = 100)	(n = 105)
Home living	99 (99)	89 (84)
Help for daily living	0 (0)	8 (8)	b.001
Institution	1 (1)	8 (8)
Autonomy	(n = 65)	(n = 56)
ADL score	6.0 +- 0.2	5.8 +- 0.7	.13
Autonomy-free	64 (99)	51 (91)	.09
Data are expressed as number of patients (percentage), mean +- SD, or median (95% CI).

treatment, but older patients tended to be managed later (P =

.06) and had brain MRI less often (6% vs 27%, P b .001). In patients without brain MRI, only 1 patient in the young group and 3 patients in the older group had pacemakers. Stroke was more severe in elderly patients, as reflected by higher Rankin scores (Table 2). Thrombolytics were used in 3 young patients seen within 3 hours. When considering MRI results, type of stroke and stroke severity and the decision to use thrombolytics in patients seen within 3 hours remained unexplained in 4 patients in the younger group and 13 in the older group. The weight of previous treatment with anticoagulant was insignificant (1 patient in the younger group and 3 patients in the older group). Immediate referral and post-ED admissions were different between groups except for the intensive care unit. Immediate referral to the stroke unit was less frequent in

Table 1 Baseline characteristics

Younger group (n = 101)		Older group (n = 105)	P
Management
Time before medical	35 (30-50)	45 (30-60)	.06
care in the ED (min)
Brain imaging
CT scan	84 (83)	100 (95)	.06
MRI	27 (27)	6 (6)	b.001
Fibrinolysis	3 (3)	0 (0)	.12
Immediate referral
Stroke center	28 (28)	6 (6)	b.001
Intensive care unit	1 (1)	2 (2)	1.
ED admission	71 (70)	98 (93)	b.001
Admissions after ED
Stroke center	35 (35)	13 (12)	b.001
Any neurologic units	63 (62)	21 (20)	b.001
Geriatric unit	4 (4)	33 (31)	b.001
Length of stay (d)	8 (6-9)	11 (9-12)	.008
Hospital-free	82 (81-84)	78 (75-81)	.002
days (3 mo)
In-hospital death	6 (6)	13 (12)	.15
	(n = 85)	(n = 94)
Rankin score	2 (2-3)	3 (3-4)	.009
Rankin score N3	21 (25)	40 (43)	.017
	(n = 95)	(n = 92)
Treatment out of the	70 (74)	55 (60)	.06
hospital in survivors
Antithrombotic	13 (14)	20 (22)	.18
Anticoagulant	1 (1)	2 (2)	.62
Both	13 (14)	19 (21)	.25
None
Long-term outcome	(n = 101)	(n = 105)
Lost on follow up	1 (1)	2 (2)	1.00
	(n = 100)	(n = 103)
One-year mortality	14 (14)	28 (27)	.004
One-year results in	(n = 86)	(n = 72)
surviving patients
Home living	79 (92)	36 (51)	b.001
Help for daily living	4 (5)	18 (25)
Institution	3 (3)	18 (25)
	(n = 55)	(n = 52)
ADL	5.7 +- 0.8	5.2 +- 1.6	.02
Autonomy free	48 (87)	38 (73)	.09
(ADL = 6)
Data are expressed in number of patients (percentage), mean +- SD, or and median (range).

older patients (6% vs 28%, P b .001). After the ED, hospitalizations for older patients were mainly represented by the geriatric unit (P b .001) compared to the neurologic unit for younger patients (b.001) (Table 2).

Table 2 Management of stroke and outcome of patients

survival analysis“>Outcome and survival analysis

Regarding living conditions, information was available for 100 (99%) younger patients and all older patients at

baseline (Table 1), and for 86 (100%) surviving young patients and 72 (96%) surviving old patients after 1 year (Table 2). Regarding ADL and autonomy-free results, information was available for 65 (64%) younger patients and 56 (53%) older patients at baseline (Table 1) and for 55

(64%) surviving younger patients and 52 (69%) surviving older patients (Table 2). At baseline, older patients were less frequently living at home with a significant higher value for disabilities on the ADL score and there was a lower number of autonomy-free patients. After 1 year, only 51% of older patients were still living at home, compared to 92% in the younger group (P b.001), and ADL score and autonomy-free patients in this group worsened (Table 2).

After admission, length of stay was longer in older patients (P = .008); in-hospital mortality tended to be higher in older patients (12% vs 6%, nonsignificant); and the 3- month postdischarge, hospital-free days period was shorter in older patients (P = .002).

At the time of analysis, 14 patients in the younger group and 28 patients in the older group had died. The 1-year mortality rate after diagnosis of stroke was 16% in younger patients (95% CI, 8-25%) compared with 33% in older

patients (95% CI, 23%-43%) (P = .014, Fig 1). Compared with the Expected mortality in the general population in France, the mortality in our study cohort was significantly higher in both older (27% vs 11%, P b .01) and younger (14% vs 1%, P b .001) patients or when considering the whole cohort (20% vs 12%, P b .001).

Predictors of mortality

Comparison between deceased and living patients is shown in Table 3. In the multivariate analysis, hemorrhagic stroke, dementia, and a high Rankin score were independent positive predictors of a 1-year mortality (Table 4). Fig 2 compares the survival among patients with the presence of these variables to those without.

Fig. 1 Kaplan-Meier estimates of death in patients aged b or >=75 years (P = .014). time zero is the onset of the stroke. Curves show survival considering all death from any cause.

Living		Deceased	P	Hazard ratio (95%CI)?
	(n = 164)	(n = 39)
Men	97 (59)	21 (54)	.59	1
Women	67 (41)	18 (46)		1.23 (0.64-2.38)
Age	69 +- 16	80 +- 11	b.001
Age >=75	75 (46)	28 (72)	.004	2.47 (1.19-5.12)
Hypertension	102 (62)	23 (59)	.72	0.69 (0.36-1.32)
Dyslipidemia	70 (43)	10 (26)	.07	0.49 (0.23-1.04)
Diabetes mellitus	34 (21)	7 (18)	.83	0.91 (0.40-2.07)
Coronary artery disease	20 (12)	7 (18)	.43	1.24 (0.52-2.99)
Atrial fibrillation	29 (18)	16 (41)	.003	2.28 (1.17-4.45)
Previous stroke	40 (24)	6 (15)	.29	0.70 (0.29-1.69)
Dementia	19 (12)	13 (33)	.002	3.36 (1.70-6.64)
Tobacco	80 (49)	9 (23)	.004	0.32 (0.15-0.71)
Ischemic stroke	118 (72)	29 (74)	.84	1.10 (0.52-2.35)
Hemorrhagic stroke	7 (4)	7 (18)	.007	4.38 (1.91-10.01)
Poor autonomy	11 (7)	4 (10)	.49	1.37 (0.49-3.89)
	(n = 141)	(n = 36)
Rankin score	2 (2-3)	4 (4-5)	b.001
Rankin score N3	35 (25)	26 (72)	b.001	6.61 (2.98-14.66)
Previous antithrombotic or	(n = 159)	(n = 39)
anticoagulation treatment	60 (38)	21 (54)	.07	0.69 (0.36-1.33)
Data are expressed as number of patients (percentage), mean +- SD, or median (95% CI). P values are for the comparison of living patients with deceased patients and were calculated using the unpaired Student t test, the Mann-Whitney test, the ?2 test or Fisher exact method as appropriate. Hazards ratio with corresponding 95% CI were calculated using univariate Cox regression analysis and are for the comparison of living patients with deceased patients.

When old age was entered as a forced variable in the Cox model, the hazards ratio was 1.57 (95% CI, 0.72-3.43) and was not significant (P = .25). The comparison of adjusted estimates of survival between young and elderly patients showed no significant difference (Fig 3).

Table 3 Comparison between deceased and survivors 1 year after admission

Discussion

Very few studies have included elderly patients with a mean age of 75 years or older [6,7,14-21], particularly in the ED where most elderly patients with acute stroke are referred. Our study was performed to analyze collectively the management of stroke in ED-referred elderly patients considering mortality, autonomy outcomes, and factors associated with death in at the time of thrombolytics and MRI. Our findings indicate that management of stroke in elderly patients in the ED is different compared with younger patients, but brings very positive results. Elderly stroke patients seen in the ED were managed with the same delay and received the same treatment as younger patients. Moreover, elderly patients all benefited from a CT scan at the time of the diagnosis, in a same proportion as the younger patients. This result represents important progress compared with previous data that reported a significant decrease CT scan and Antithrombotic treatment use with age [6,7]. However, the explanations for this difference are numerous and could be related to individual practice in various countries, as well as evolution in stroke management.

Despite this positive result, some points remain significantly different for elderly patients. At admission, assessment with MRI and admission to the stroke unit were less frequently observed in this group despite a greater severity of stroke in older patients as shown by Rankin-modified scores, ICU referral, and initial mortality. The difference observed for MRI assessment was not explained by MRI contraindication. At discharge from the ED, referral to the neurologic department, particularly to the stroke unit, was significantly lower in elderly patients.

Our results support the fact that despite progress in stroke management, age is still associated with a difference in management using modern methods of brain imaging in patients presenting with stroke in the ED.

However, this age-related difference can be influenced by other factors. First, we found a difference between groups for

Table 4 Multivariate Cox proportional-hazards analysis of

variables associated with death within 1 year after admission for stroke

Prevalence Hazards ratio P

Because data for Rankin were missing for some patients, the final sample used in the multivariate analysis consisted of 179 patients, 36 (20%) of whom died.

	(n = 179)	(95%CI)
Rankin score N3	61 (34)	7.17 (3.18-16.15)	b.001
Hemorrhagic stroke	31 (17)	5.67 (2.25-13.69)	b.001
Dementia	12 (7)	2.66 (1.30-5.45)	.007

Fig. 2 Kaplan-Meier estimates of death for the three variables that were significant in the multivariate analysis: A: hemorrhagic stroke; B: dementia; C: high (N3) vs low (<=3a Rankin score. Time zero is the onset of the stroke. Curves show survival considering all death from any cause.

duration of symptoms, but the nonsignificant difference observed for patients arriving to ED before 3 hours of symptoms duration does not explain a 4-fold less use of MRI and immediate stroke unit referral in the older group. Several studies have shown a beneficial effect of stroke unit or

neurologic department management compared with general units for all subgroups of elderly patients [19,22], but these differences have been inconsistently observed [20]. There- fore, this can reflect an individual physician’s management without consideration of underinvestigation. Last, the hemorrhagic risk of thrombolytic agents in the stroke unit in older patients could contribute to the difference observed in management through its influence on stroke unit admission [23]. However, thrombolysis was very rare in our study, and the association between age, thrombolysis, and stroke Hemorrhagic transformation is not so clear because some studies did not observe any significant difference between groups according to age [24-26].

Overall, the age-related differences in management revealed by our study can be explained by several factors. Ageism has been previously reported in the management of cerebrovascular disease for prevention [27,28], investiga- tion, and treatment [29]. However, we cannot affirm that observed differences are related with ageism despite the fact that these differences do not appear to be justified by our results or previous studies in elderly stroke patients.

We found that the modified Rankin Score, type of stroke, and dementia were predictors of a 1-year mortality, whereas age was not [21,30-33]. The results overall highlight the weight of comorbidities compared to age in elderly patients and the need to evaluate and to consider stroke severity more than age, and does not support the notion that age is essentially a negative determining factor in the management of stroke. Previous studies have reported an association of mortality with type of stroke [32,34-36], atrial fibrillation [17,32,33,38], diabetes mellitus, increased blood glucose or discontinuous diabetes treatment [31,33,37], cognitive impairment [31,37], living conditions [38], and autonomy [32]. In our study, hemorrhagic stroke was associated with mortality, which could have been favored by antithrombotic

Fig. 3 Adjusted estimates of survival in patients aged b or >=75 years (P = .014). The survival estimates have been adjusted for presence or absence of hemorrhagic stroke, dementia, and a high (N3) Rankin score. Time zero is the onset of the stroke. Curves show survival considering all death from any cause.

or anticoagulant treatment (53% of older patients) or falls, a frequent event in this population [39-41]. Previous dementia was independently associated with mortality in stroke patients in our study, as with risk of stroke [42,43] and severe stroke [38]. Both dementia and stroke can share aggravating common factors such as hypertension [44] and apolipoprotein E e4 [45,46]. However, this association could reflect the patient’s frail condition and common shared factors that all contribute to risk of death in stroke patients. Finally, we observed a higher mortality rate in elderly patients compared to younger patients, but this difference was no longer significant when other significant risk factors were taken into account. This particular finding is markedly different from previous studies, which reported a higher mortality for in-hospital death [15,19,21] and 1-year mortality [19,20,30] in older patients. Age has been found to be independently associated with greater mortality [21,30] using multiple logistic regression and survival analyses including initial stroke severity, but one of these studies excluded transient ischemic attacks and cerebral hemorrhage

[21] and both used different assessment scales.

Our study has some limitations. This was a retrospective study conducted in a single institution, and this might reflect local medical management. In particular, we cannot exclude that different practice patterns reflecting different health care services may have influenced prognosis or outcome of patients. However, with a 1826-bed teaching urban hospital including an ED, a MRI, and a stroke center, our practice can be considered as representative of stroke management in relation to stoke center admission, MRI, and use of thrombolytics.

Because our study was performed in an ED, and patients directly referred to the stroke unit were not considered, most of the stroke patients included within a 3-hour admission and fibrinolysis time window were probably underrepresented in our study because of pre-ED selection for stroke manage- ment in France.

The choice of the modified Rankin score compared with other scales for stroke severity may be criticized but was known to be easily feasible for all physicians in our ED. Its value in our study could reflect comorbidity consequences together with stroke severity. However, a large majority of our patients were living independently at home, suggesting that the Rankin score mainly reflected stroke severity. Overall, our methods were essentially designed to highlight an association between factors (age and underuse), but not a causative effect. To analyze the effect of age on physician’s decision in management and treatment of stroke, a prospective study is needed with consideration for all factors implicated in medical decision making.

In conclusion, our results suggest that comorbidities and disease severity have to be considered more than age for the management of acute stroke in elderly patients. The age- related differences observed in our study highlight the need of new studies dedicated to stroke elderly patients and its optimal management.

Acknowledgment

We thank Dr David Baker, DM, FRCA (Staff Anesthesiol- ogist, Department of Anesthesiology, CHU Necker-Enfants Malades, Paris, France) for reviewing the manuscript.

References

1. Bonita R. Epidemiology of stroke. Lancet 1992;339:342-4.
2. Barker WH, Mullooly JP. Stroke in a defined elderly population, 1967- 1985. A less lethal and disabling but no less common disease. Stroke 1997;28:284-90.
3. Pohjasvaara T, Erkinjuntti T, Vataja R, et al. Comparison of stroke features and disability in daily life in patients with ischemic stroke aged 55 to 70 and 71 to 85 years. Stroke 1997;28:729-35.
4. Lehtonen A, Salomaa V, Immonen-Raiha P, et al. Declining incidence and mortality of stroke in persons aged N or = 75 years in Finland; the FINSTROKE study. Eur J Cardiovasc Prev Rehabil 2004;11:466-70.
5. American Heart Association. 2001 Heart and Stroke Statistical Update. Dallas (Tex): American Heart Association; 2000.
6. Bhalla A, Grieve R, Tilling K, et al. Older stroke patients in Europe: stroke care and determinants of outcome. Age Ageing 2004;33: 618-24.
7. Rudd AG, Lowe D, Hoffman A, et al. Secondary prevention for stroke in the United Kingdom: results from the National Sentinel Audit of Stroke. Age Ageing 2004;33:280-6.
8. Aho K, Harmsen P, Hatano S, et al. Cerebrovascular disease in the community: results of a WHO collaborative study. Bull World Health Organ 1980;58:113-30.
9. Warlow C, Sudlow C, Dennis M, et al. Stroke. Lancet 2003;362: 1211-24.
10. Katz S, Ford AB, Moskowitz RW, et al. Studies of illness in the aged. The index of ADL: a standardized measure of biological and psychosocial function. JAMA 1963;185:914-9.
11. Bonita R, Beaglehole R. Modification of Rankin scale: recovery of motor function after stroke. Stroke 1988;19:1497-500.
12. Le Manach Y, Perel A, Coriat P, et al. Early and delayed myocardial infarction following aortic surgery. Anesthesiology 2005;102: 885-91.
13. INSEE. La France en faits et en chiffres. www.insee.fr (access on July 20, 2007).
14. Glader EL, Stegmayr B, Johansson L, et al. Differences in long-term outcome between patients treated in stroke units and in general wards: a 2-years follow-up of stroke patients in Sweden. Stroke 2001;32: 2124-30.
15. Henon H, Pasquier F, Durieu I, et al. Preexisting dementia in stroke patients. Baseline frequency, associated factors, and outcome. Stroke 1997;28:2429-36.
16. Kong KH, Chua KS, Tow AP. Clinical characteristics and functional outcome of stroke patients 75 years old and older. Arch Phys Med Rehabil 1998;79:1535-9.
17. Jorgensen HS, Nakayama H, Reith J, et al. Acute stroke with atrial fibrillation. Copenhagen Stroke Study. Stroke 1996;27:1765-9.
18. Jorgensen HS, Reith J, Nakayama H, et al. What determines good recovery in patients with the most severe strokes? The Copenhagen Stroke Study. Stroke 1999;30:2008-12.
19. Jorgensen HS, Kammersgaard LP, Houth J, et al. Who benefits from treatment and rehabilitation in a stroke Unit? A community-based study. Stroke 2000;31:434-9.
20. Fagerberg B, Claesson L, Gosman-Hedstrom G, et al. Effect of acute stroke unit care integrated with care continuum versus Conventional treatment: A randomized 1-year study of elderly patients: the Goteborg 70+ Stroke Study. Stroke 2000;31:2578-84.
21. Kammersgaard LP, Jorgensen HS, Reith J, et al. Short-and long-term prognosis for very old stroke patients. The Copenhagen Stroke Study. Age Ageing 2004;33:149-54.
22. Kaste M, Palomaki H, Sarna S. Where and how should elderly stroke patients be treated? A randomized trial. Stroke 1995;26:249-53.
23. Bateman BT, Schumacher HC, Boden-Albala B, et al. Factors associated with in-hospital mortality after administration of thrombo- lysis in acute ischemic stroke patients: an analysis of the nationwide inpatient sample 1999 to 2002. Stroke 2006;37:440-6.
24. Berrouschot J, Rother J, Glahn J, et al. Outcome and severe hemorrhagic complications of intravenous thrombolysis with tissue plasminogen activator in very old (Nor = 80 years) stroke patients. Stroke 2005;36:2421-5.
25. Sylaja PN, Cote R, Buchan AM, et al. Thrombolysis in patients older than 80 years with acute ischaemic stroke: Canadian Alteplase for Stroke Effectiveness Study. J Neurol Neurosurg Psychiatry 2006;77:826-9.
26. Engelter ST, Reichhart M, Sekoranja L, et al. Thrombolysis in stroke patients aged 80 years and older: Swiss survey of IV thrombolysis. Neurology 2005;65:1795-8.
27. Brass LM, Krumholz HM, Scinto JD, et al. Warfarin use following ischemic stroke among Medicare patients with atrial fibrillation. Arch Intern Med 1998;158:2093-100.
28. Rudd AG, Lowe D, Hoffman A, et al. Secondary prevention for stroke in the United Kingdom: results from the National Sentinel Audit of Stroke. Age Ageing 2004;33:280-6.
29. Fairhead JF, Rothwell PM. Underinvestigation and undertreatment of carotid disease in elderly patients with transient ischaemic attack and stroke: comparative population based study. Br Med J 2006;333:525-7.
30. Shen Q, Cordato D, Chan DK, et al. Identifying the determinants of 1- year post-stroke outcomes in elderly patients. Acta Neurol Scand 2006; 113:114-20.
31. Wang SL, Pan WH, Lee MC, et al. Predictors of survival among elders suffering strokes in Taiwan: observation from a nationally representa- tive sample. Stroke 2000;31:2354-60.
32. Longstreth Jr WT, Bernick C, Fitzpatrick A, et al. Frequency and predictors of stroke death in 5,888 participants in the Cardiovascular Health Study. Neurology 2001;56:368-75.
33. Marini C, De Santis F, Sacco S, et al. Contribution of atrial fibrillation to incidence and outcome of ischemic stroke: results from a population-based study. Stroke 2005;36:1115-9.
34. Elneihoum AM, Goransson M, Falke P, et al. Three-year survival and recurrence after stroke in Malmo, Sweden: an analysis of Stroke registry data. Stroke 1998;29:2114-7.
35. Abu-Zeid HA, Choi NW, Hsu PH, et al. Prognostic factors in the survival of 1,484 stroke cases observed for 30 to 48 months. I. Diagnostic types and descriptive variables. Arch Neurol 1978;35: 121-5.
36. Brown RD, Whisnant JP, Sicks JD, et al. Stroke incidence, prevalence, and survival: secular trends in Rochester, Minnesota, through 1989. Stroke 1996;27:373-80.
37. Fried LP, Kronmal RA, Newman AB, et al. Risk factors for 5-year mortality in older adults: the Cardiovascular Health Study. JAMA 1998;279:585-92.
38. Appelros P, Nydevik I, Seiger A, et al. Predictors of severe stroke: influence of preexisting dementia and cardiac disorders. Stroke 2002; 33:2357-62.
39. Broderick JP, Phillips SJ, O’Fallon WM, et al. Relationship of cardiac disease to Stroke occurrence, recurrence, and mortality. Stroke 1992; 23:1250-6.
40. Lin HJ, Wolf PA, Kelly-Hayes M, et al. Stroke severity in atrial fibrillation. The Framingham Study. Stroke 1996;27:1760-4.
41. Whittle J, Wickenheiser L, Venditti LN. Is warfarin underused in the treatment of elderly persons with atrial fibrillation? Arch Intern Med 1997;157:441-5.
42. Zhu L, Fratiglioni L, Guo Z, et al. Incidence of stroke in relation to Cognitive function and dementia in the Kungsholmen Project. Neurology 2000;54:2103-7.
43. Ferrucci L, Guralnik JM, Salive ME, et al. Cognitive impairment and risk of stroke in the older population. J Am Geriatr Soc 1996;44: 237-41.
44. Skoog I, Lithell H, Hansson L, et al. Effect of baseline cognitive function and Antihypertensive treatment on cognitive and cardiovas- cular outcomes: Study on COgnition and Prognosis in the Elderly (SCOPE). Am J Hypertens 2005;18:1052-9.
45. Kalmijn S, Feskens EJ, Launer LJ, et al. Cerebrovascular disease, the apolipoprotein e4 allele, and cognitive decline in a community-based study of elderly men. Stroke 1996;27:2230-5.
46. Slooter AJ, Tang MX, van Duijn CM, et al. Apolipoprotein E epsilon4 and the risk of dementia with stroke. A population-based investigation. JAMA 1997;277:818-21.