Stroke in the Very Elderly: Characteristics and Outcome in Patients Aged ≥85 Years with a First-Ever Ischemic StrokeGur A.Y.a · Tanne D.b, d · Bornstein N.M.c, d · Milo R.a · Auriel E.c, d · Shopin L.c, d · Koton S.d · on behalf of the NASIS Investigators
aDepartment of Neurology, Barzilai Medical Center, Ashkelon, Faculty of Health Sciences, Ben-Gurion University of the Negev, Beer-Sheva, and Departments of Neurology, bSheba Medical Center, cTel-Aviv Sourasky Medical Center, and dThe Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel
Background: Epidemiological and clinical features of very elderly patients with stroke are still uncertain. Our aim was to study the patient characteristics and outcomes in the very elderly (aged ≥85 years) with a first-ever ischemic stroke in the National Acute Stroke Israeli Survey (NASIS) registry. Methods: The NASIS registry is a nationwide prospective hospital-based study performed triennially (2004, 2007, 2010). Patients with ischemic stroke aged ≥85 years were compared with those 65–84 years old regarding their baseline characteristics, stroke severity, etiology of stroke and stroke outcomes. Logistic regression analyses were used to adjust for potential confounders. Stroke severity was determined according to the National Institute of Health Stroke Scale (NIHSS) score. Results: The proportion of very elderly (≥85 years) patients among the NASIS population increased from 18.3% in 2004 to 19.9% in 2007 and 24.5% in 2010 (p for trend = 0.005). The percentage of women was higher in patients aged ≥85 years (p < 0.0001). Atrial fibrillation, congestive heart disease and prior disability were significantly more common, while diabetes, current smoking and dyslipidemia were less frequent in the very elderly. The very elderly presented with more severe strokes: 36.3% of the ≥85-year-old patients had an NIHSS score ≥11 compared with 22.0% in the younger age group (p < 0.05). Conclusions: There is an increasing proportion of very elderly subjects, mostly women, among first-ever ischemic stroke patients. Current information on age-specific aspects of stroke in the very elderly is crucial to set up successful prevention pathways and implementing well-organized stroke care for this population.
Copyright © 2012 S. Karger AG, Basel
Age is the most significant non-modifiable risk factor for stroke. With the increase in life expectancy, the majority of stroke patients are expected to be elderly or even very elderly subjects. First attempts to draw attention to old-age stroke patients have already begun in the early 90s of the last century [1,2]. The absence of consensus regarding the epidemiological, clinical and prognostic features of elderly stroke patients makes it difficult to cope with the challenges posed by the increasing age of patients. Some studies have shown different risks and clinical profiles of stroke in elderly [3,4] while according to others, there are no significant differences between older and younger stroke populations [5,6]. Obviously, this inconsistency is partially explained by the lack of uniform definitions of ‘elderly’ and ‘very elderly’. Whereas in several studies the age cutoff for old stroke patients is 70 years [7,8], in others, it varies from 75 to 88 years [9,10]. Moreover, elderly stroke patients in most studies have been compared with the entire population of younger patients, although it is well known that young (aged ≤55 years) stroke patients share specific etiological and prognostic features different from those of older patients [11,12,13]. Our aim was to study patient characteristics, clinical features and outcomes in the very elderly (aged ≥ 85 years) in comparison with patients aged 65–84 years with a first-ever ischemic stroke in the prospective National Acute Stroke Israeli Surveys (NASIS) registry.
NASIS is a prospective hospital-based nationwide registry performed triennially . In total, 6,279 unselected patients with acute ischemic stroke, intracerebral hemorrhage and transient ischemic attack aged ≥18 years hospitalized during 2-month periods were included in the first three surveys (NASIS 2004, 2007 and 2010). A first-ever ischemic stroke was diagnosed in 3,125 patients; of those, 2,056 (65.8%) patients were ≥65 years old and thus included in the present study.
The NASIS registry was approved by the ethical committees of all participating hospitals.
Data were collected prospectively by coordinating physicians in all the wards of all the hospitals admitting patients with acute cerebrovascular diseases. A data collection form especially designed for the NASIS registry was completed for all patients. Stroke and transient ischemic attack were reported according to the medical report on discharge. Ischemic stroke and intracerebral hemorrhage were differentiated by brain CT or MRI, and cases were regarded as undetermined stroke if CT or MRI were not performed. If the coordinating physician raised doubt regarding the diagnosis, a central adjudication committee made the final decision.
Stroke severity was determined using the National Institute of Health Stroke Scale (NIHSS) score  and functional disability using the modified Rankin scale (mRS) . Etiology of ischemic stroke was determined based on the Trial of Org 10172 in Acute Stroke Treatment classification . Vascular risk factors were defined as follows: (1) hypertension: use of antihypertensive agents or systolic blood pressure ≥140 mm Hg or diastolic blood pressure ≥90 mm Hg; (2) diabetes mellitus: use of oral hypoglycemic agents, insulin or glycosylated hemoglobin (HbA1c) >7.0%; (3) dyslipidemia: use of antihyperlipidemic agents or a serum cholesterol level >220 mg/dl; (4) current smoking of any degree; (5) ischemic heart disease: history of acute myocardial infarction or angina pectoris, coronary artery bypass graft or percutaneous coronary intervention; (6) congestive heart failure (CHF): previously diagnosed by a clinician; (7) peripheral artery disease: previously diagnosed by a clinician; (8) atrial fibrillation: diagnosed during hospitalization or history of chronic or paroxysmal atrial fibrillation.
Three outcomes were studied: complications during hospitalization (including infectious, neurological and cardiac), severe disability or death (mRS >3) and in-hospital mortality.
Patients aged ≥85 years were compared with those 65–84 years old regarding their baseline characteristics, stroke severity, etiology of stroke and stroke outcomes using the χ2 test. Risk estimates by odds ratios (ORs), 95% confidence intervals (CIs) for complications during hospitalization, severe disability or death (mRS >3), and in-hospital mortality in the very elderly were evaluated with logistic regression analysis. The registry period (2004, 2007, 2010) was entered as a covariable in the multivariable analysis. Two models were studied: the first adjusting for sex and the registry period and the second additionally adjusting for potential confounders identified in the univariate analysis including atrial fibrillation, diabetes, current smoking, dyslipidemia, ischemic heart disease, CHF, prior disability (mRS ≥2), and stroke severity (by 5 NIHSS categories as predefined by the NASIS protocol). The final calibration and discrimination of the models were assessed using the Hosmer-Lemeshow (H-L) goodness-of-fit test and the c-statistic, representing the area under the receiver operating characteristic curve, respectively. Trends in patient characteristics, stroke severity, stroke etiology and outcomes in the very old (≥85 years) by registry period were studied. Analyses were performed with the SAS 9.2 software (SAS Institute Inc., Cary, N.C., USA).
The proportion of very elderly (≥85 years) patients among the NASIS population aged ≥65 years increased during the 6-year period from 18.3% in 2004 to 19.9% in 2007 and 24.5% in 2010 (p for trend = 0.005).
In analyses by age group, the proportion of women was higher in patients aged ≥85 years compared with those 65–84 years old (p < 0.0001). Atrial fibrillation, CHF and prior disability were significantly more common while diabetes, current smoking and dyslipidemia were less frequent in the very elderly (table 1). Anticoagulants on admission were more frequently reported by younger patients (p < 0.002; table 1). In the analysis of trends by registry period, rates of dyslipidemia increased from 25.4% in 2004 to 40.8% in 2007 and 63.7% in 2010 (p for trend <0.0001) and hypertension increased from 74.8% in 2004 to 82.3% in 2007 and 90.5% in 2010 (p for trend = 0.0004).
|Table 1. Baseline characteristics of patients with first ischemic stroke by age group (n = 2,056)|
Stroke characteristics differed by age group (table 2). The very elderly presented with more severe strokes: 36.3% of the ≥85-year-old patients had a severe stroke (NIHSS score ≥11) compared with 22.0% in the younger group (p < 0.05). We also found a different distribution of the various etiological subtypes by age group. More cardioembolic strokes and fewer strokes related to large-vessel atherosclerosis were diagnosed in the older patients. Analysis by registry period of stroke etiology among ≥85-year-old patients showed that the proportion of small-vessel infarctions decreased significantly from 20.1% in 2004 to 17.7% in 2007 and 8.9% in 2010 (p for trend = 0.0007).
|Table 2. Characteristics of stroke by age group (n = 2,056)|
Rates of in-hospital complications, functional disability and death during hospitalization were significantly increased for very elderly patients (p < 0.0001 for all; table 3). Findings were consistent following adjustment for sex and survey (model 1). Following further adjustment for additional potential confounders (model 2), the older group had a statistically significant risk for in-hospital complications (OR 1.6, 95% CI 1.2–2.1, p = 0.0004; p for H-L test = 0.4, c = 0.778) and a borderline significant risk for severe disability or death (OR 1.4, 95% CI 1.0–1.9, p = 0.05; p for H-L test = 0.5, c = 0.890), whereas the risk for in-hospital death was not significantly increased (OR 1.1, 95% CI 0.7–1.7, p = 0.7; p for H-L test = 0.8, c = 0.876) (table 4).
|Table 3. Stroke outcome in patients with first ischemic stroke by age group (n = 2,056)|
|Table 4. Risk for complications, severe disability and in-hospital death for patients ≥85 years old compared with 65- to 84-year-old patients with first ischemic stroke|
Findings of the analysis of trends in outcomes for ≥85-year-old patients by registry period are presented in figure 1. A significant decrease in the rate of in-hospital mortality among the very old patients is evident: rates decreased from 18.7% in 2004 to 5.7% in 2010 (p for trend = 0.0005).
|Fig. 1. Trends in stroke outcome for patients ≥85 years old by registry period. p for trend values are presented.|
In this nationwide study, we observed several epidemiological, clinical and outcome features of very elderly patients with a first-ever ischemic stroke. To the best of our knowledge, this is the first triennially repeated observation of the same population showing a significant trend of increasing proportion of very elderly subjects among stroke patients. In China, the proportion of the very elderly in hospitalized stroke patients has been reported lower than that in western countries . Moreover, in a report from Taiwan, the percentage of old Asian stroke patients decreased from 2002 to 2005 despite the increase in the total number of stroke patients . However, this finding was related to the fact that the study was conducted in a single hospital located in a new growing area with a large population of young immigrants. Obviously, our results more accurately represent the true weight of very elderly subjects in the stroke population.
The majority of very old stroke patients in the present study were women. Previous studies have documented sex differences in stroke morbidity and mortality . Although stroke is more prevalent in men than in women, age-specific stroke incidence rates for women aged 35–44 years and those >85 years old are higher than for men .
Our study demonstrates that very old stroke patients show a different vascular risk factor profile than younger patients. The very elderly group presented more frequently with atrial fibrillation and heart disease, while diabetes, dyslipidemia and current smoking was less frequent. These results are in good agreement with previous studies [19,20] and can be explained by selective survival and lifestyle differences between very elderly and younger stroke patients. Only in the Martinique Island study, atrial fibrillation and heart disease showed a trend toward being higher in the old Caribbean black stroke patients, but this finding was not statistically significant . Differences between ethnic groups regarding the prevalence of atrial fibrillation as well as other cerebrovascular risk factors have already been shown . Our findings are consistent with the Framingham study findings in which the age-specific incidence rate of non-valvular atrial fibrillation steadily increased from 0.2 per 1,000 in the forth decade to 3.9 per 1,000 for ages 80–89 years . One of the reasons for our results may be the significantly lower use of anticoagulants for atrial fibrillation prior to the ischemic stroke in the very elderly compared to younger patients.
In the analysis by stroke etiology, we found significantly more cardioembolic-type and less large-vessel atherosclerosis and small-vessel occlusive types in the very elderly compared with the younger patients. Our findings differ from previous reports showing no significant difference in stroke subtypes between elderly and younger stroke patients [6,7]. However, those studies used different classifications of stroke subtypes. In the present study, rates of in-hospital complications, functional disability and death after discharge were significantly increased for very elderly patients compared with their younger counterparts. Poorer outcomes in the older-age group have been documented in previous studies as well [23,24,25].
Our study demonstrates several important trends among very elderly stroke patients over the period of the NASIS registry from 2004 to 2010. Prevalence of dyslipidemia and hypertension increased significantly in the very elderly stroke patients. Obviously, our results reflect serum total cholesterol and low-density lipoprotein cholesterol increase with age and higher concentrations of low-density lipoprotein cholesterol in elderly females compared with males due to menopause-related hormone changes . Hypertension is also frequent in the elderly, with prevalence rates as high as 60–80% . There was also a significant decrease in the rate of in-hospital mortality during the registry period in the very elderly stroke patients. Possible explanations for this trend may be more active treatment in the acute-stroke period and faster transfer to rehabilitation or chronic care facilities.
We acknowledge that our study has some limitations. Data were collected during two winter months in the studied years. Seasonal differences in stroke incidence have been reported in other populations and potentially might have influenced the NASIS results. In addition, although our study is based on national data, the NASIS is a hospital-based project, and thus, we had no information on patients not admitted to the hospital.
In conclusion, there is an increasing proportion of very elderly patients among first-ever ischemic stroke patients. The very elderly differ from younger (but still elderly) patients in their baseline characteristics. Risk of complications and poor functional outcome are increased in the very elderly patients. In-hospital mortality rates for the very elderly have considerably decreased in the last years. Current information on age-specific aspects of stroke in the very elderly is crucial to set up successful prevention pathways and implementing well-organized stroke care for this population. Improving stroke outcome by establishing appropriate preventive and therapeutic programs in the very elderly will constitute a challenge for physicians in the future.
The authors declare that there are no conflicts of interest and no competing financial interests.
A.Y. Gur, MD, PhD
Department of Neurology, Barzilai Medical Center
2 Hahistadrut St.
Ashkelon 78278 (Israel)
Tel. +972 8 865 1289, E-Mail email@example.com
Received: March 12, 2012
Accepted: May 7, 2012
Published online: July 8, 2012
Number of Print Pages : 6
Number of Figures : 1, Number of Tables : 4, Number of References : 27
Vol. 39, No. 1, Year 2012 (Cover Date: July 2012)
Journal Editor: Feigin V.L. (Auckland)
ISSN: 0251-5350 (Print), eISSN: 1423-0208 (Online)
For additional information: http://www.karger.com/NED