If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
The rate of recurrent ischemic stroke seems unchanged during the last 20 years despite increasing use of secondary prevention.
•
The highest recurrence rate is seen in large artery atherosclerosis and cardio embolic stroke subtypes with recurrent stroke most frequently of the same subtype.
•
Small vessel occlusion stroke subtype has the lowest recurrence rate and a more diverse recurrence pattern. A thorough diagnostic work-up is therefore highly relevant in patients with small vessel occlusion stroke.
•
Smoking, hypertension, diabetes mellitus, atrial fibrillation, prior cerebrovascular event and increasing stroke severity are independent risk factors of stroke recurrence.
Abstract
Objectives
Recurrent stroke remains a challenge though secondary prevention is initiated immediately post-stroke. Stroke subtype may determine the risk of recurrent stroke and require specific preventive measures. We aimed to identify subtype-specific stroke recurrence and associated risk factors over time.
Methods and materials
A systematic review was performed using PubMed and Embase for studies including adults >18 years, first-ever ischemic stroke in population-based observational studies or registries, documented TOAST-criteria and minimum 1-year follow-up. Meta-analysis on stroke recurrence rate was performed. Final search: November 2019.
Results
The search retrieved 26 studies (between 1997 and 2019). Stroke recurrence rate ranged from 5.7% to 51.3%. Recurrent stroke was most frequent in large artery atherosclerosis (LAA) and cardioembolic (CE) stroke with recurrent stroke similar to index stroke subtype. We identified a lower recurrence rate for small vessel occlusion (SVO) stroke with recurrence frequently of another stroke subtype. Based on a meta-analysis the summary proportion recurrence rate of recurrent stroke in studies using TOAST-criteria = 0.12 and = 0.14 in studies using TOAST-like criteria. Hypertension, diabetes mellitus, atrial fibrillation previous transient ischemic attack, and high stroke severity were independent risk factors for recurrence.
Conclusion
Stroke recurrence rates seem unchanged over time despite the use of secondary prevention. The highest recurrence rate is in LAA and CE stroke eliciting same subtype recurrent stroke. A lower recurrence rate is seen with SVO stroke with a more diverse recurrence pattern. Extensive workup is important in all stroke subtypes - including SVO stroke. Future research needs to identify better preventive treatment and improve compliance to risk factor prevention to reduce stroke recurrence.
Recurrent ischemic stroke (IS) remains a challenge even with improved neuroimaging, acute stroke management, and disseminated use of secondary prevention.
The consequences of recurrent IS are more extensive cerebrovascular impairments resulting in additional physical and cognitive disability, and significant social impact.
The symptoms at onset, the subsequent recovery process, and the resulting functional outcome of ischemic stroke is very heterogenic related to the diverse etiologies and vascular mechanism underlying the subtype of stroke. Since 1993 a stroke subtype classification system according to The Trial of Org 10172 in Acute Stroke Treatment (TOAST)-criteria
Classification of subtype of acute ischemic stroke. Definitions for use in a multicenter clinical trial. TOAST. Trial of Org 10172 in Acute Stroke Treatment.
has been widely used for in clinical application and in research.
The risk of recurrent stroke as well as the optimal preventive strategies may relate to the particular stroke subtype. The SVO stroke subtype typically display few or mild symptoms as only small cerebral lesions are induced. However, SVO is suspected to cause recurrent and often clinically “silent” strokes.
In contrast, large artery atherosclerosis (LAA) and cardioembolic (CE) stroke often cause extensive vascular deficits and ischemia due to the inherent occlusion of large cerebral arteries locally or by an embolus. The risk of recurrent stroke is closely linked to the underlying pathophysiology. The significant risk factors of stroke in general are well-known.
The risk of recurrence of stroke may relate to previously undetected risk factors that are overlooked due to assumptions made about the etiology. Relevant risk factors of stroke recurrence and the recurrence patterns of the different stroke subtypes could be mapped systematically. This mapping could improve the diagnostic stroke work-up and preventive strategies, by indicating which risk factors and treatments to address to reduce stroke recurrence.
The aim of this systematic review was to identify the stroke recurrence rate for each stroke subtype over time as well as risk factors for recurrent IS identified through the currently available literature.
Methods
Literature search
An electronic search was performed according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines for systematic reviews in PubMed using the search strategy specified in Table 1. ClinicalKey, MEDLINE, Embase, Cochrane Library and ScienceDirect were searched additionally. Reference lists of included studies were searched for relevant studies. The final search was performed the 18. November 2020.
Classification of subtype of acute ischemic stroke. Definitions for use in a multicenter clinical trial. TOAST. Trial of Org 10172 in Acute Stroke Treatment.
Classification of subtype of acute ischemic stroke. Definitions for use in a multicenter clinical trial. TOAST. Trial of Org 10172 in Acute Stroke Treatment.
study design: observational cohort studies, prospective- or retrospective registry studies with at least one year follow-up. Recurrent stroke was defined as a new neurological deficit presenting after a period of clinical stability, lasting for more than 24 h and with attributable new ischemic or hemorrhagic lesions verified either by CT or MRI of the brain.
Studies not explicitly using TOAST-criteria but describing similar subgroups (atherothrombotic (ATI), cardioembolic, lacunar infarct (LI) and uncertain/other etiology) were included but analyzed separately as “TOAST-like criteria”. ATI corresponds to LAA, LI corresponds to SVO and uncertain/other etiology corresponds to undetermined or other etiology. Randomized controlled trials, studies in cohorts of transitory ischemic attack (TIA) patients, selected subgroups of IS (e.g. only index SVO or LAA) or patients (e.g., diabetes mellitus or endovascular procedures) were excluded. The corresponding author for the selected full-text articles was approached if data were incomplete. If no response was received within two weeks, then the study was excluded (one study). The search strategy and electronic search were repeated by a second reviewer (LC) to ensure consistent findings. No disagreements occurred between the two reviewers.
Critical appraisal and assessment of bias
All authors individually considered the risk of bias in the individual studies and across studies during full-text reading using the Critical Appraisal Skills Program (CASP) checklist for observational studies.
See supplemental material S1 for the checklist and table S2 for the critical appraisal.
Synthesis of results
A descriptive synthesis was performed from studies in which the recurrence rate for each TOAST- or TOAST-like subtype in addition to important risk factors were stated. A quantitative synthesis was also performed in which a meta-analysis of proportions was conducted (using R version 3.6.1). Analyses were made separately for the studies using TOAST vs. TOAST-like criteria. Data were log-transformed and the effect size was calculated based on the method of moments (DerSimonian and Laird method) and a random effects model was applied. Heterogeneity was calculated by I-square statistic, I2, (the ratio of between-study variance to the observe variance, i.e., the sum of the between- and within-study variance).
Data were not suitable for any regression- or trend analysis. Thus, changes in incidence of recurrent stroke for the various subtypes over time, could not be assessed. Instead, a meta-analysis of proportions for the specific subtypes was conducted for the studies stating the quantitative recurrence rate for each stroke subtype.
A detailed description of the calculation and analysis in R can be provided by the first author.
Findings
In total 7688 titles were identified. After screening, removal of duplicates and full-text reading, 26 articles were included for final analysis (Fig. 1).
Study designs were either prospective observational cohort studies (20 studies) or retrospective registry studies (10 studies). Follow-up time ranged from one year to 23 years.
The number of participants ranged from 168 to 11560 stroke survivors. Mean age at index stroke ranged from 62 to 75 years. The majority of studies (16 studies) used the TOAST-criteria for stroke subtypes. The remaining 10 studies classified stroke subtypes using diagnostic criteria similar to TOAST-criteria and were analyzed separately as studies using “TOAST-like” criteria. See Table 2 for a summary of results.
Epidemiology of ischemic stroke subtypes according to TOAST criteria: incidence, recurrence, and long-term survival in ischemic stroke subtypes: a population-based study.
Prospective cohort study (1994-98) in City of Erlangen. To determine the incidence, recurrence, and long-term survival rates of ischemic stroke subtypes by a mechanism-based classification scheme.
First-ever IS; n =531 Mean age: 73 (SD =12.9) Male: 42%
2
11.5%
The highest recurrence rate in CE followed by SAO and UND. Lowest in LAA. TOAST-subtype was not a significant predictor of long-term recurrence.
Prospective observational cohort study (1995– 1999). To determine the influence of pre-stroke and poststroke dementia on the 3-year risk of stroke recurrence.
First-ever IS; n =202 Median age: 75 (NA) Male: 48%
3
14.4%
Leucoaraiosis was the only independent predictor of stroke recurrence, not dementia. Limitation: Score determined by CT-scan not MRI.
Prospective observational cohort study (1997-2001). To estimate the risk of stroke recurrence and mortality in the community of Bajo Aragón in, Spain and to compare it with previous studies conducted in other countries.
First ever IS; n =425 Mean age: 75.4 (SD =9) Male: 59.8%
1, 5
1-year: 9.5% 5-year: 26%
Highest risk of recurrence in LAA followed by CE and SVO. UND had the lowest recurrence rate. LAA-LAA, CE-CE, SVO-SVO recurrence patterns were most frequent. Age was main predictor of recurrence and death. Atrial fibrillation + either valvular disease or congestive heart failure increased risk of recurrent CE stroke.
Prospective registry study (2000-2003). To compare the mechanisms of the index and recurrent stroke.
Patients with IS; n =901 Mean age: 62 ±11 Male: 59%
2
9.5%
In the index intracranial LAA- and SAO group recurrent stroke was most often followed by same subtype. Index extracranial LAA had heterogenic recurrence pattern. For index CE-group recurrent stroke was CE again in 2 out of 3.
Retrospective cohort registry study (based on an earlier epidemiological study, 1961-1993). To examine 10-year cumulative recurrence rates for stroke in a Japanese cohort according to pathological type and clinical subtype of brain infarction.
Patients with IS; n =410 Mean age: 73.9 ±10.1 Male: 49%
1, 5, 10
1-year: 12.8% 5-year: 35.3% 10-year: 51.3%
Most frequent LAA-LAA and CE-CE recurrence pattern. CE stroke had a significantly higher risk of 10-year recurrence than LI.
Registry study based on China National Stroke Registry (unknown observation period). To explore whether hypertension contributes to the recurrence of certain subtypes of ischemic stroke (IS).
Patients with IS; n =11560 Mean age: 66 (SD =12.3) Male: 57.8%
1
17.7%
The stroke recurrence rate in the SAO subgroup was significantly greater in patients with hypertension than in those without hypertension. No statistically significant differences between patients with and without hypertension in the LAA, CE, or “other” subgroups.
Prospective registry study from the Joinville Stroke Registry (2009-2013). To determine the recurrence rates, survival rates and the cause of death in 3 years after their first-ever incident in a urban population-based setting.
Patients with first-ever IS; n =407 Mean age: 63 ±16 Male: NA
3
8.6%
No difference in stroke recurrence rate between the TOAST-subtypes.
Prospective registry study (the Ege Stroke Registry, 1998-2009). To identify whether leukoaraiosis contributes to the recurrence of certain subtypes of stroke at long term.
Patients with IS; n =9522 Mean age: 65 ±12 Male: 57%
5
22.9%
Leukoaraiosis was significantly associated with stroke recurrence in the LAD, SAD and ICH subtypes, CE and ‘other determined cause’ subgroups. Recurrence rate was significantly lower in patients treated with antithrombotics during follow-up compared to no antithrombotics.
Recurrent stroke, myocardial infarction, and major vascular events during the first year after acute ischemic stroke: the multicenter prospective observational study about recurrence and its determinants after acute ischemic stroke I.
Prospective observational cohort study of 12 university hospitals or regional stroke centers (2010-2013). To estimate rates of recurrent stroke, myocardial infarction and major vascular events during the first year after acute ischemic stroke in Korea.
First ever IS; n =9687 Mean age: 70 ±13 Male: 58.7%
1
5.7%
3-fold lower risk of recurrence with index SVO compared to the other stroke subtypes. Risk factors for recurrence were diabetes mellitus, atrial fibrillation, previous stroke and ischemic stroke subtype other than SVO.
Retrospective registry study (sub-study of the Mashhad Stroke Incidence Study (MSIS) (2006-2007)). To characterize ischemic stroke subtypes in a Middle Eastern population.
First ever IS; n =512 Mean age: 65 (SD =14) Male: 51.6%
5
14.3%
Highest risk of recurrence in the LAA and the CE subgroup.
Two independent prospective cohort studies; OXVASC (2004-2014) and a Chinese cohort from University of Hong Kong (HKU, 2008-2014). To validate the total small vessel disease (SVD) score by determining its prognostic value for recurrent stroke.
IS/IS; n =2002 (OXVAC n =1028, HKU n = 974) Mean age: 68 (OXVASC) ±14 69 (HKU) ±12 Male: 52.3% (OXVASC) 59.9% (HKU)
3.5
8.8% (OXVASC) 11.2% (HKU)
Small vessel disease (SVD)-score higher in Asian cohort. Increasing SVD-score associated with an increased risk of recurrent IS and intracerebral hemorrhage for both index SVD- and non-SVD stroke. Risk of recurrent stroke was almost doubled in patients with >20 basal ganglia perivascular space and increased in patients with ≥5 cerebral micro bleeds and periventricular white matter hyperintensities.
Retrospective analysis of pt. admitted to the Hospital at the Federal University at Paraná (2012-2015). To analyze the long-term recurrence rate in patients with a first-ever ischemic stroke secondary to intracranial large artery atherosclerosis (LAA) in a Brazilian population.
First-ever IS; n =359 Mean age: 64.6 ±13.6 Male: 50.7%
1.8
13.4%
The only variables correlated with stroke recurrence were intracranial and extracranial LAA stroke mechanism.
Retrospective registry study of the NORSTROKE cohort (2007-2012). To study the five-year incidence and risk of all-cause readmission, cause-specific readmission and mortality after IS or TIA by stroke subtype.
Patients with IS; n =1303 Mean age: 72 (SD =12.8) Male: 54%
5
14%
No significant differences for risk or recurrence rate between stroke-subtypes. LAA subtype did have an increased risk of early stroke recurrence, but the long-term risk was found to be equal for all stroke subtypes
Prospective cohort study of pt admitted to Guangdong Second Provincial General Hospital. Assessment of lifestyle associated with recurrent stroke in Chinese patients with first-ever IS.
Pt with first-ever IS; n =421. Mean age: 65 (interquartile range 56-73 years). Male: 69.4%
1
13.5%
Smoking, ischemic heart disease increased risk of recurrent stroke whereas HDL-level, housework, daily consumption of fresh fruit and good sleep quality had reduced recurrence rate. LAA har largest recurrence rate followed by SVO. CE had lowest recurrence rate.
Prospective cohort study of the South London Stroke Register (SLSR) including patients from 1995 to 2018 stratified in 3-month, 1-year, 5-year and for some years 10-year recurrence rate. To examine the risk and trends longitudinally of recurrence and death after stroke and identify patient groups at risk of stroke recurrence.
Pt with first-ever IS; n =6052. Mean age: NA Male: 51%
23 (5-year intervals)
12.6% (cumulated 5-year risk)
The cumulative risk of recurrence reduced from 1995 to 2005 with unchanged rate since. Risk of recurrence or death was highest for CE followed by LAA and ICH. Recurrence rates seems reduced for LAA and UND since year 2000 but remains stable for CE and SVO stroke. For CE and ICH the recurrent stroke was most frequently the same as index stroke. 26% of LAA had recurrent LAA stroke and 35% of SVO had recurrent SVO stroke. Older age, AF, HT, ischemic heart disease, and smoking were independent risk factors of recurrent stroke.
Prospective cohort study of the Erlangen Stroke Project (pt included from 1996-2015) stratified into 3-months, 1-year and 5-year recurrence rates. To investigate the long-term trends in case-fatality rates and stroke recurrence rates of IS over a 20-year period.
Pt with first-ever IS; n =3346. Mean age: 74±13 Male: 46.8%
20 (5-year intervals)
20.1% (cumulated 5-year risk)
In general, a declining mortality rate during the follow-up period. Survival probability was highest in pt. with SVO stroke. UND had the highest recurrence rate followed by SVO stroke. However, there was a decline in stroke recurrence in all pt.
Prospective registry study (FEL-IS, 1991-1992). To analyze the risk of stroke recurrence in Polish patients with ischemic stroke and to identify the predictors of recurrent stroke at one-year follow-up.
First ever IS; n =209 Mean age: 66 ±12.4 Male: 50.2% Stroke classification: Atherosclerosis of intra- or extracranial artery, CE, LI, other causes
2
11.4%
Recurrence most frequent with extra- and intracranial atherosclerosis and CE, and less frequent for LI. Untreated hypertension was independent predictor of stroke recurrence.
Retrospective registry study (1994-98). To determine the subtype of recurrent stroke in relation to subtype of index stroke, the prognosis of all stroke recurrences observed in our hospitals and the possible responsible causes.
Primary IS; n =1108 Mean age: 74 (SD =8) Male: 60% Stroke classification: ATI, LI, CE, undetermined, BH
2-6
12.2 %
Most frequent pattern: ATI-ATI and CE-CE IS. In patients with LI the risk of recurrent LI was similar to that observed for other types, especially ATI. Among patients with recurrent ischemic stroke, 25% quit antiplatelet medication after index stroke for different reasons.
Prospective registry study (Linköping Stroke Unit, 1986). To determine the long- term outcome in subgroups of patients with stroke and predictors of recurrent stroke and death in patients who were admitted to the Stroke Unit, Department of Neurology, Linköping, in 1986.
Patients with IS; n =309. Patients with ICH; n =30 Mean age: 71.2 (M) 73.6 (F). Male: 45.4% Stroke classification: ATI, CE, LI (and ICH)
14
43.7%
The annual occurrence of stroke was highest for CE, followed by LI and ATI. Age, severity of stroke, previous stroke and systolic blood pressure are each of importance in predicting recurrent stroke.
Substudy of the TULIPS-study (prospective, multicentre cohort, 1999-2002). To identify determinants of recurrence after ischemic stroke in the Japanese population.
Patients with IS; n =831 Mean age: 72.6 (NA) Male: 57.3% Stroke classification LVO, CE, LI or uncertain
1
8.7%
Significant difference in the incidence of recurrent stroke among the four stroke subgroups. Diabetes mellitus predicted recurrence in the LI group and prior stroke predicted recurrence in the LI and LAA group. The study did not how association between hypertension and recurrent stroke.
Retrospective registry study (1978-1997). To clarify the long-term prognosis by stroke subtypes and to identify determinants of recurrence and death after first-ever stroke.
First-ever-acute IS; n =1565 Mean age: 64.4 (NA) Male: 64.4% Stroke classification: ATI, CE, LI, other, BH
3
13%
Highest recurrence rate for CE and ‘other’ etiology followed by ATI and LI. Lowest for intracerebral hemorrhage. After the first year no significant difference in recurrence rate between stroke subtypes.
Sub study of the Atherosclerosis Risk in Communities (ARIC) prospective study (1987-89). To characterize mortality, recurrence, and hospital readmissions after stroke according to etiologic subtype.
First-ever IS; n =987 Mean age: 66.8 (NA) Male: 44% Stroke classification: Thrombotic-, CE-, lacunar IS, SAH and ICH
5.3
18.5%
70% of recurrent events were of same subtype, except with index LI where only 28% recurrent stroke were LI-subtype. LI had higher recurrence than thrombotic ischemic stroke (1.3, 0.0-1.9), but lower all-cause readmission and lowest mortality of the subtypes.
Prospective cohort study (2004-2009). To investigate the recurrent stroke types associated with white matter lesions.
Patients with IS; n =807 Mean age: 69.8 ±12 Male: 56.5% Stroke classification: LI, ATI, CE, unknown, ICH.
2.6
13.7%
Incidence of all types of stroke recurrence in patients with gr 2 and 3 white matter hyperintensities was significantly greater than in patients with grade 0-1. Recurrences presenting as deep intracerebral hemorrhage and lacunar infarctions were significantly elevated in stroke patients with white matter hyperintensities gr 2 and 3.
Prospective cohort study (2004-2011). To prospectively observe stroke recurrences in patients with histories of LI and WMH.
Patients with LI; n =305 Mean age: 70.2 ±11.7 Male: 59.5% Stroke classification: LI, ATI, ICH, CE, other. WMH Grade 0, 1, 2 or 3
4.2
20.3%
Risk of recurrent LI was significantly and independently associated with gr 2-3 white matter hyperintensities. No significant difference among grade 0, 1, and 2-3 in incidences of recurrent stroke presenting as lobar intracerebral hemorrhage, ATI, or CE infarctions.
Prospective observational cohort study (Aichi and Hiroshima Stroke Center, 206-2009). To examine potential factors affecting stroke recurrence, retrospectively.
First ever IS or TIA; n =1087 Mean age:70.4 ±11.6 Male: 66.1% Stroke classification: LI, ATI, CE and other
2
21.3%
Recurrence rate highest for CE stroke followed by SVO and LAA. Gender, mRS at discharge, age and family history were common risk factors of recurrence. National Institute of Health Stroke Scale (NIHSS), hypertension, diabetes mellitus, and family history were risk factors for recurrent CE stroke while mRS at discharge and gender were risk factors for LAA stroke.
Prospective observational study (11 stroke centers, the Korean Nationwide Image-Based Stroke Database Project). To investigate the impact of WMH on 1-year recurrence of stroke (hemorrhage vs ische- mic stroke) and mortality.
First ever IS; n =7101 Mean age: 67.9 (SD =12.7) Male: 58.7% Stroke classification: LAA, SVO, CE, undetermined, other
1
5.6%
HT, DM and AF were more prevalent in pt with recurrent stroke. Cerebral MWH were an independent risk factor of recurrent stroke. The highest MWH quartile had 2.2 fold increased risk of recurrent IS and haemorrhagic stroke regardless of subtype. LAA were most frequent as recurrent stroke followed by CE and undetermined.
The recurrence rate was given as cases of recurrence divided by total sample size. See Fig. 2 for a graphical illustration of the recurrence rate as a function of follow-up time for all studies.
Fig. 2Graph illustrating follow-up time in years and recurrence rate of all studies. TOAST-criteria: Black dots. TOAST-like criteria: Black triangles.
Recurrent stroke, myocardial infarction, and major vascular events during the first year after acute ischemic stroke: the multicenter prospective observational study about recurrence and its determinants after acute ischemic stroke I.
Epidemiology of ischemic stroke subtypes according to TOAST criteria: incidence, recurrence, and long-term survival in ischemic stroke subtypes: a population-based study.
Another study had 23 years of follow-up stratified into 3-month, 1-year, 5 year-recurrence and 10-year recurrence rates with a cumulated 5-year recurrence rate of 12.6%.
This study also found that recurrence rate dropped from 17.8% in 1995-1999 to a stable rate of approximately 10% 5-year recurrence rate from 2000-2018.
One study reported results from an Asian versus a non-Asian cohort.
TOAST-criteria were applied only to the Western cohort and showed a recurrence rate of 8.8% after a mean follow-up of 3.5 years. In the Asian cohort, only TOAST-like criteria were applied.
Studies using TOAST-like subtypes
Two studies reported a one-year recurrence rate of 8.7% and 5.6%.
with a follow-up time of between two to six years found a cumulative stroke recurrence rate of 15%. Two studies reported a four to five-year recurrence rate ranging from 18.5% to 20.3%.
Epidemiology of ischemic stroke subtypes according to TOAST criteria: incidence, recurrence, and long-term survival in ischemic stroke subtypes: a population-based study.
Recurrent stroke, myocardial infarction, and major vascular events during the first year after acute ischemic stroke: the multicenter prospective observational study about recurrence and its determinants after acute ischemic stroke I.
Recurrent stroke, myocardial infarction, and major vascular events during the first year after acute ischemic stroke: the multicenter prospective observational study about recurrence and its determinants after acute ischemic stroke I.
Recurrent stroke, myocardial infarction, and major vascular events during the first year after acute ischemic stroke: the multicenter prospective observational study about recurrence and its determinants after acute ischemic stroke I.
In one study the recurrence patterns differed between intracranial- and extracranial LAA, respectively; in the intracranial stenosis group the recurrent stroke occurred more often in the same vascular territory by the same mechanism for the majority of patients than in the extracranial group (89% vs 35%).
Recurrent stroke, myocardial infarction, and major vascular events during the first year after acute ischemic stroke: the multicenter prospective observational study about recurrence and its determinants after acute ischemic stroke I.
Recurrent stroke, myocardial infarction, and major vascular events during the first year after acute ischemic stroke: the multicenter prospective observational study about recurrence and its determinants after acute ischemic stroke I.
Recurrent stroke, myocardial infarction, and major vascular events during the first year after acute ischemic stroke: the multicenter prospective observational study about recurrence and its determinants after acute ischemic stroke I.
Recurrent stroke, myocardial infarction, and major vascular events during the first year after acute ischemic stroke: the multicenter prospective observational study about recurrence and its determinants after acute ischemic stroke I.
One study reported lifestyle factors such as sleep quality, daily intake of fresh fruit and engaging in housework as independent predictors of recurrent stroke.
Recurrent stroke, myocardial infarction, and major vascular events during the first year after acute ischemic stroke: the multicenter prospective observational study about recurrence and its determinants after acute ischemic stroke I.
Epidemiology of ischemic stroke subtypes according to TOAST criteria: incidence, recurrence, and long-term survival in ischemic stroke subtypes: a population-based study.
Epidemiology of ischemic stroke subtypes according to TOAST criteria: incidence, recurrence, and long-term survival in ischemic stroke subtypes: a population-based study.
Epidemiology of ischemic stroke subtypes according to TOAST criteria: incidence, recurrence, and long-term survival in ischemic stroke subtypes: a population-based study.
Epidemiology of ischemic stroke subtypes according to TOAST criteria: incidence, recurrence, and long-term survival in ischemic stroke subtypes: a population-based study.
Even though it fulfilled the inclusion criteria the study had a long follow-up time (10 years) in a decade (1990’ies) with fewer opportunities for acute treatment (intravenous thrombolysis was introduced in the late 1990’ies and endovascular therapy almost two decades later) and secondary prevention (use of antiplatelets or oral anticoagulants). These three studies were therefore not included in the meta-analysis.
Results are presented in the forest plot in Fig. 3a and 3b (for ‘TOAST-criteria’ and ‘TOAST-like criteria’ respectively). The heterogeneity of the included studies was in high (for studies using TOAST-criteria and TOAST-like criteria I2 = 93%, and 99% respectively). For the studies using TOAST-criteria and TOAST-like criteria the summary proportion recurrence rate was 12% and 14% respectively. As the two groups had a high heterogeneity a pooled summary proportion was not calculated. The proportion of stroke recurrence did not decrease as the publication year increased.
Fig. 3a-b: Forest plots of the meta-analysis of proportions in studies using TOAST-criteria (top) and TOAST-like criteria (bottom). Heterogeneity (I2) is estimated for the individual group. As this forest plot is based on proportions the vertical line is placed at the point value of the summary proportion, depicted as the diamond at the bottom with the tips representing the 95-% confidence interval (CI) of the summary proportion.
A sub-analysis of recurrent stroke by subtype was conducted based on the studies providing absolute numbers of recurrences per subtype. It was only possible to get the absolute numbers of LAA, CE and SVO strokes (see Fig. 4a and 4b). LAA stroke recurrence was most frequent with a proportion of 0,04 of the total population, whereas recurrence frequency of CE and SVO were similar. However, this sub-analysis should be interpreted with caution as only a part of all the included studies could be applied to this analysis.
Fig. 4a-b: Forest plots of the meta-analysis of proportions of studies including the number of cases of recurrence for each stroke subtype. As data were only appropriate for large artery atherosclerosis (LAA), cardio embolic (CE) and small vessel occlusion stroke (SVO) these were the only groups compared. Heterogeneity (I2) is estimated for the individual group. As this forest plot is based on proportions the vertical line is placed at the point value of the summary proportion, depicted as the diamond at the bottom with the tips representing the 95-% confidence interval (CI) of the summary proportion.
This review had three important findings: (1) the recurrence rate seems unchanged over more than 20 years, even though new and more effective acute treatments have been developed; (2) LAA and CE stoke had the highest recurrence rate most often following a LAA-LAA or CE-CE recurrence pattern; (3) SVO stroke had a lower recurrence rate of clinical stroke compared to other TOAST-subtypes, and SVO-stroke was often followed by another stroke subtypes than SVO.
Recurrence rate was similar between the studies using TOAST-criteria compared to those using TOAST-like criteria. This was also supported in a meta-analysis where the summary proportions were 12% and 14% for the studies using TOAST-criteria and TOAST-like criteria respectively. Furthermore, the forest plots in Fig. 3a-b show a lower recurrence rate with increasing sample size, perhaps reflecting less selection bias in large cohorts compared to smaller cohorts.
The recurrence pattern and recurrence rate differed from that of SVO-, LAA- and CE stroke. With a few exceptions
patients with SVO showed a lower risk of clinically detectable stroke recurrence, and a more diverse pattern of recurrent stroke subtype. In sub-studies of SVO stroke where MRI detectable (silent) lesions are accounted for, the occurrence may however be similar to that of other TOAST-subtypes.
Recurrent stroke after index SVO stroke was however often either due to LAA, CE or another stroke subtype. A recent study has revealed that the collateral recruitment is rather poor in stroke patients with cerebral SVD compared to stroke patients without SVD.
This might also explain the fact that index SVO stroke is often followed by LAA stroke and that an increased level of WMH is associated to increased risk of recurrent stroke. The SVO stroke subgroup may therefore represent a more composite group of stroke patients and therefore requires a more thorough work-up of both the intracranial and cardiovascular status including brain angiography as well as echocardiography.
The apparent stagnant rates of stroke recurrence during the last 20 years are striking given the ongoing advances in acute diagnostics as well as medical- and surgical interventions for stroke patients guided by improved neuroimaging techniques.
This review found that smoking, HT, DM, AF, prior cerebrovascular event, and increased severity of stroke were independent risk factors of recurrent stroke. These findings support the conclusions of the large INTERSTROKE-study which found that 10 modifiable primary risk factors accounted for more than 90% of the global stroke burden; HT, smoking, DM, abdominal obesity, poor diet, physical inactivity, unfavorable psychosocial factors, alcohol abuse, cardiac diseases, and a high apolipoprotein ApoB/ApoA1 ratio.
In contrast to the acute interventions, the last 20 years has provided little has change in the secondary preventive medical treatment, with one exception being the development of the novel anticoagulants (NOACs) for prevention of cardioembolic stroke in AF patients. In the recent decade there has been a larger focus on the treatment of cardiovascular risk factors, such as blood pressure, cholesterols and tobacco use. Furthermore, the ageing population and increased stroke survival rate result in an over-all increased cardiovascular burden at the population level as well as a larger number of people surviving to get a second stroke.
The small progress within non-AF secondary prevention and the increasing age and load of cardiovascular risk factors in the general population may therefore balance out the effect of the advances in stroke diagnostics and acute interventions regarding the stroke recurrence rate.
Future research needs to explore new treatment targets to reduce stroke recurrence and optimize knowledge how to improve non-medical preventive strategies, such as physical activity, lifestyle changes and increase patient compliance, on stroke recurrence.
Limitations
Although the recurrence rate seems unchanged during the last 20 years, the data may reflect outdated recurrence rates, which do not consider the recent developments in acute diagnostics and treatment which could also affect stroke recurrence.
One selection bias could be the inclusion of only papers in English language. However, the studies included in this review covers a wide range of publication years and populations (North- and South America, Europe, Asia, Middle East) and we expect selection bias to be minimal. Nonetheless, different patient ethnic groups are missing or underrepresented in this study; the African continent, Australia, Eastern Europe, and Russia. This highlights the need for future cohort studies also including a larger diversity of populations.
Furthermore, mean age at onset of index stroke was approximately 70 years across all included studies. As the age of the participants increases so does the risk of recurrent stroke and mortality.
In this review we chose to include studies using TOAST-criteria and studies using TOAST-like criteria. The latter was chosen since the studies applying TOAST-like criteria basically fulfilled the TOAST-criteria but did not completely define the criteria accordingly. Useful insight would be lost were these studies excluded though they were evaluated separately.
Underlying SVD increases risk of both ischemic and hemorrhagic strokes.
Though some studies did include information on hemorrhagic stroke, there was no information on the extent of SVD lesions. It is therefore unclear, if the cohorts differed regarding SVD-load and the risk of recurrent stroke due to SVD comorbidity.
Classification of subtype of acute ischemic stroke. Definitions for use in a multicenter clinical trial. TOAST. Trial of Org 10172 in Acute Stroke Treatment.
A meta-analysis of proportions of recurrent stroke in each study showed a substantial between-study heterogeneity and a summary proportion of recurrent stroke of 0.12 in studies using TOAST-criteria and 0.14 in studies using TOAST-like criteria.
Smoking, HT, DM, AF, prior cerebrovascular event, and increased stroke severity were independent risk factors for stroke recurrence. This demonstrates the importance of mapping cardiovascular risk factors and maintaining secondary preventive treatment as well as the need to perform a thorough diagnostic work up of index SVO stroke in order to prevent stroke of other subtypes.
Future studies need to address challenges of stroke subtyping, diagnostics after first-ever-ischemic stroke and improvement of drug compliance and secondary prevention.
Funding
All authors were funded by the Capital Region of Copenhagen. CK was partly funded by the Novo Nordic Foundation grant number: NNF 18O 003 1840 and MK was partly funded by the Novo Nordic Translational Research Foundation grant number: NNF19OC0058812.
Declaration of Competing Interest
None.
Acknowledgements
The authors thank Morten Ib Ingstrup for preparation of Figs. 2, 3a-b and 4a-b as well as M.D., PhD, Philip Hywel Thompson for English proofreading.
Classification of subtype of acute ischemic stroke. Definitions for use in a multicenter clinical trial. TOAST. Trial of Org 10172 in Acute Stroke Treatment.
Recurrent stroke, myocardial infarction, and major vascular events during the first year after acute ischemic stroke: the multicenter prospective observational study about recurrence and its determinants after acute ischemic stroke I.
Epidemiology of ischemic stroke subtypes according to TOAST criteria: incidence, recurrence, and long-term survival in ischemic stroke subtypes: a population-based study.
Recently, in a systematic review and meta-analysis, Kolmos et al found that the rates of recurrent ischemic stroke seem unchanged over time.1 This is disappointing, especially as the benefits of secondary prevention have become more and more recognized over the recent decades. Possibly, this knowledge has not been materialized in practice everywhere. However there are promising results from Sweden.
We thank you for the very relevant and important comment on the results of our systematic review and meta-analysis on recurrent ischemic stroke related to stroke subtypes.
00338-4&id=gr1.jpg){kind=link}
00338-4&id=gr2.jpg){kind=link}
00338-4&id=gr3.jpg){kind=link}
00338-4&id=gr4.jpg){kind=link}