Clinical Outcomes of Posterior Versus Anterior Circulation Infarction With Low National Institutes of Health Stroke Scale Scores
Background and Purpose—We compared baseline characteristics and outcomes at 3 months between patients with minor anterior circulation infarction (ACI) versus minor posterior circulation infarction (PCI), including the influence of large vessel disease on outcomes.
Methods—This study is an analysis of a prospective multicenter registry database in South Korea. Eligibility criteria were patients with ischemic stroke admitted within 7 days of stroke onset, lesions in either anterior or posterior circulation, and National Institutes of Health Stroke Scale score of ≤4 at baseline. Patients were divided into 4 groups for further analysis: minor ACI with and without internal carotid artery/middle cerebral artery large vessel disease and minor PCI with and without vertebrobasilar large vessel disease.
Results—A total of 7178 patients (65.2±12.6 years) were analyzed in this study, and 2233 patients (31.1%) had disability (modified Rankin Scale score 2–6) at 3 months. Disability was 32.3% in minor PCI and 30.3% in minor ACI (P=0.07), and death was 1.3% and 1.5%, respectively (P=0.82). In a multivariable logistic regression analysis, minor PCI was significantly associated with disability at 3 months when compared with minor ACI (odds ratio, 1.23; 95% confidence interval, 1.09–1.37; P<0.001). In pairwise comparisons, minor PCI with vertebrobasilar large vessel disease was independently associated with disability at 3 months, compared with the other 3 groups.
Conclusions—Our study showed that minor PCI exhibited more frequent disability at 3 months than minor ACI. Especially, the presence of vertebrobasilar large vessel disease in minor PCI had a substantially higher risk of disability. Our results suggest that minor PCI with vertebrobasilar large vessel disease could require more meticulous care and are important targets for further study.
Posterior circulation infarcts (PCI) have been generally considered to have a poor outcome with high mortality and morbidity,1,2 but several studies have conflicting results.3–6 The comparative functional outcome of patients with posterior versus anterior circulation stroke has been little studied, especially that of the subset of patients with minor initial deficits. Most previous studies have instead investigated potential differences in the presenting features and the frequency of recurrent stroke between the 2 circulations.7–9 Among all patients who initially present with minor stroke or transient ischemic attack (TIA), a meta-analysis found a suggestion that, compared with anterior circulation infarct (ACI) patients, PCI patients were at increased risk for recurrent stroke in the acute period and at reduced risk for recurrent stroke in the chronic period.8 However, functional outcomes were not investigated. Among posterior circulation minor stroke and TIA patients, some studies have found that the presence of vertebral or basilar artery large vessel disease (VBD) is associated with a higher early risk of recurrent stroke.9,10 However, the sample size of the analyzed cohorts was modest, and functional outcomes were not delineated.
Whether functional differences in outcome occur between minor PCI and minor ACI is an important topic for bedside prognostication and clinical trial design. Therefore, we analyzed a prospective registry of 14 academic hospitals in South Korea and compared outcomes at 3 months between patients with minor ACI and PCI. In addition, we investigated whether concurrent large vessel steno-occlusive disease influenced functional outcomes in these patients.
This report presents a study based on data from the Clinical Research Center for Stroke-5th division registry, which is a prospective, web-based registry of consecutive patients with acute stroke or TIA who were admitted to 14 academic hospitals in South Korea (http://www.stroke-crc.or.kr/ecrf). Detailed information from the Clinical Research Center for Stroke-5th division registry was reported previously.11,12 Fourteen stroke centers with nationwide coverage registered 28 348 acute stroke patients consecutively between April 2008 and November 2013. For the current study, 17 976 patients registered between January 2011 and November 2013 were initially screened. The following eligibility criteria were used for the current study: (1) patients with ischemic stroke or TIA (by <24 hours symptom duration criterion) who exhibited positive lesions on diffusion-weighted imaging (DWI) and who were admitted within 7 days of stroke onset and (2) patients with National Institutes of Health Stroke Scale (NIHSS) scores of ≤4 at baseline. The following patients were excluded: (1) patients with premorbid disability (defined as prestroke modified Rankin Scale [mRS] score >1), (2) lesions in the both anterior circulation and posterior circulation on DWI, and (3) missing data on mRS at both discharge and 3 months.
Collection of clinical information for the Clinical Research Center for Stroke-5th division registry for improving the quality of stroke care was approved by the local institutional review boards of all participating centers with a waiver of informed consent because of study subject anonymity and minimal risk to participants. Also, use of the registry database and additional review of medical records for this study was approved further by the local institutional review boards.
Demographic, clinical, and imaging data were collected prospectively. The following stroke risk factors were identified: age, sex, hypertension, diabetes mellitus, dyslipidemia, current smoking, previous coronary artery diseases, previous peripheral artery diseases, previous stroke, and previous TIA. Baseline data, including NIHSS score, were collected from all patients on admission and at discharge, and stroke subtypes were stratified according to the Trial of Org 10172 in Acute Stroke Treatment (TOAST) criteria with some modifications after complete diagnostic profiling.13,14 Vascular territories and anatomic lesion locations were recorded by the participating investigators. All patients in the current study received DWI and either magnetic resonance angiography or computed tomography angiography to determine lesion locations and vascular status. Lesion locations in anterior or posterior circulation distributions were determined from the DWI results (Method section in the online-only Data Supplement). ACI was diagnosed if the lesions were located in the anterior cerebral artery, middle cerebral artery, or internal carotid artery, including watershed areas. PCI was diagnosed if the lesions were located in the posterior cerebral artery, cerebellar arteries, basilar artery, or vertebral artery. Relevant arterial disease (RAD) was defined as arterial steno-occlusion, in a vessel supplying the ischemic field, with moderate-to-severe stenosis (>50% of luminal narrowing) or occlusion on magnetic resonance angiography, computed tomography angiography, or both. Major large vessel disease was defined as symptomatic disease involving the middle cerebral artery, internal carotid artery, vertebral artery, and basilar artery, locations with well-known high risks of poor outcomes.15,16 Patients were further grouped based on the presence or absence of any anterior circulation major large vessel disease (AC-LVD; internal carotid artery and middle cerebral artery disease) or any posterior circulation major large vessel disease (PC-LVD; vertebral artery and basilar artery disease). This analysis had 4 subgroups: minor ACI with AC-LVD, minor ACI without AC-LVD, minor PCI with PC-LVD, and minor PCI without PC-LVD.
Dedicated research nurses or physicians certified in mRS assessment gather mRS outcomes through direct or telephone interviews. Leading outcomes were disability (mRS score 2–6) at 3 months and the full range of disability (all 7 mRS score levels) at 3 months. Dependency (mRS score 3–6) and mortality by 3 months were also outcomes of interest. We assessed changes between baseline and discharge NIHSS scores. Several parameters, such as NIHSS ≥4 at discharge (severe neurological deficit at discharge) and an increase in the NIHSS score by ≥1 between baseline and discharge (neurological worsening during hospitalization) were used for exploratory analysis.
Data are presented as means±SD or the frequency of categorical variables. Categorical variables were analyzed using the χ2 test and Fisher exact test, where appropriate. Continuous variables were analyzed using the independent sample t test or the Mann–Whitney U test, where appropriate. Both multivariable ordinal and binary logistic regression analyses were performed to evaluate the association between infarct location (PCI versus ACI) and outcomes (mRS score distribution and disability) at 3 months (adjusted for age, sex, initial NIHSS score, prestroke mRS score, hypertension, diabetes mellitus, dyslipidemia, atrial fibrillation, previous TIA, previous stroke, previous coronary artery diseases, previous peripheral arterial diseases, previous statin use, TOAST criteria, RAD, thrombolysis, discharge antihypertensive use, discharge antidiabetic medication, discharge statin medication, clopidogrel at discharge, warfarin at discharge, and revascularization). Factors associated with binary outcome of death within 3 months were adjusted for age, male sex, initial NIHSS score, prestroke mRS score, hypertension, diabetes mellitus, atrial fibrillation, previous TIA, previous peripheral arterial diseases, TOAST criteria, RADs, thrombolysis, discharge antihypertensive medication, discharge antidiabetic medication, and discharge statin medication. Pairwise comparison of the 4 groups according to the presence or absence of AC-LVD and PC-LVD was performed. In addition, associations between infarct location (PCI versus ACI) and the presence of major LVD (no AC-LVD versus AC-LVD, no PC-LVD, and PC-LVD) and disability/death at 3 months were examined in subgroups according to the discharge status and hospital course using multivariable logistic regression analysis. Also, analyses of outcomes for each of the 5 possible entry NIHSS scores were performed. A P value <0.05 was considered statistically significant. Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated. All statistical analyses were performed using SPSS for Windows, version 17 (SPSS Inc, Chicago, IL).
A total of 17 976 stroke patients were registered in the outcome database in Clinical Research Center for Stroke-5th division registry during the study period. Of these, 8380 were candidates for the current study by having minor neurological deficits within 7 days of onset, with positive DWI findings. Among these patients, 659 patients were excluded because of premorbid disability (prestroke mRS score >1), 337 patients because of lesions in both ACI and PCI, and 206 patients because of no mRS data at 3 months. Ultimately, 7178 patients (mean age, 65.1±12.6 years; men, 4472, 62.3%) were included in this study (Figure I in the online-only Data Supplement).
Table 1 shows the general characteristics of minor ACI (4305, 60.0%) and minor PCI (2873, 40.0%) patients. RAD was observed in 2549 (35.5%) of 7178 patients: 1647 patients with ACI and 902 patients with PCI. In patients with RAD, magnetic resonance angiography (both extra- and intracranial) was performed in 88.5% (n=2256) and computed tomography angiography in 38.4% (n=979). Of 4305 patients with minor ACI, 1555 (36.1%) had AC-LVD, and of 2873 patients with minor PCI, 645 (22.5%) had PC-LVD.
Patients with minor ACI were older, more often woman, and had a higher baseline NIHSS score than patients with minor PCI. Previous TIA and previous peripheral artery diseases were more frequently observed in patients with minor ACI than in patients with minor PCI. Atrial fibrillation, RAD, thrombolysis, revascularization, and clopidogrel and warfarin medication at discharge were also significantly more frequent in patients with minor ACI than in patients with minor PCI. In contrast, diabetes mellitus, large artery atherosclerosis using the TOAST classification, and antihypertensive and antidiabetic medication at discharge were more frequently observed in patients with minor PCI than in patients with minor ACI.
Outcomes at 3 Months
A total of 2233 of the 7178 patients (31.1%) had disability (mRS 2–6) at 3 months. Disability was numerically more frequent in patients with minor PCI than patients with minor ACI, but this difference was not statistically significant (32.3% versus 30.3%; P=0.07). Dependency and death at 3 months were not different between patients with minor ACI and minor PCI. The unadjusted and adjusted distributions across the entire mRS at discharge and at 3 months in patients with minor ACI and minor PCI are shown in Figure II in the online-only Data Supplement and Figure. Outcomes across the entire ordinal range of mRS were independently linked to minor PCI. For minor PCI, compared with minor ACI, worse mRS status at discharge (OR, 1.39; 95% CI, 1.27–1.52; P<0.001) and at 3 months (OR, 1.24; 95% CI, 1.13–1.35; P<0.001) was more likely to occur. For disabled outcome at 3 months (mRS 2–6), minor PCI, compared with minor ACI, was independently associated with disability at 3 months (OR, 1.23; 95% CI, 1.09–1.37; P<0.001) by multivariable binary logistic regression analysis (Table 2).
Table I in the online-only Data Supplement presents other factors that were associated with disability and death at 3 months. Briefly, older age, male sex, higher initial NIHSS score, diabetes mellitus, previous stroke, and RAD were independently associated with disability at 3 months. In contrast, previous statin use and small vessel occlusion using the TOAST classification compared with undetermined cause were negatively associated with disability at 3 months. Death within 3 months was independently associated with older age, diabetes mellitus, thrombolysis, and RAD. Small vessel occlusion and statin use at discharge were negatively associated with death within 3 month. Combined aspirin and clopidogrel antiplatelet therapy was used in 30% of the cohort and was not associated with outcomes (Table II in the online-only Data Supplement).
In addition, the frequency of disability at 3 months increased both in minor ACI and PCI as the initial NIHSS increased (P for trend <0.001), but death within 3 months did not exhibit this trend (Figure III in the online-only Data Supplement).
We performed further analyses according to the presence of major LVD to investigate high-risk groups in minor ACI and minor PCI (Table 3). Patients with minor ACI with AC-LVD and minor PCI with PC-LVD had higher frequencies of disability at 3 months compared with patients with minor ACI without AC-LVD and minor PCI without PC-LVD by univariate analysis (32.9%, 39.2%, 28.8%, and 30.3%, respectively). In the pairwise comparison of other groups, minor PCI with PC-LVD was more likely to have disability and dependency at 3 months. Minor PCI with PC-LVD was independently associated with disability (OR, 1.58; 95% CI, 1.28–1.95; P<0.001) and dependency (OR, 1.36; 95% CI, 1.05–1.77; P<0.001) at 3 months, compared with minor ACI with AC-LVD. Minor PCI with PC-LVD were also independently associated with death within 3 months compared with minor ACI without AC-LVD and minor PCI without PC-LVD.
Regarding individual LVDs, individual vessel occlusions were more likely to have disability, dependency, and death at 3 months (Table III in the online-only Data Supplement). Particularly, basilar artery occlusion and extracranial internal carotid artery occlusion had significantly higher frequency of disability, dependency, and death at 3 months.
Outcome According to Subgroups
Patients with major LVD, both AC-LVD and PC-LVD, more often had severe neurological deficits and neurological deterioration at discharge than those without major LVD (Table IV in the online-only Data Supplement). The increased risk for disability at 3 months associated with posterior circulation location was pronounced in patients with minor neurological deficits at discharge (NIHSS score 0–4; Table 4). Among patients with more severe deficits at discharge, ACI and PCI patients experienced 3-month disability at similar rates, but death was more frequent in posterior circulation events. Increased disability in PCI patients was present not only among patients who had neurological deterioration during the acute hospitalization but also among those with a stable or improving course.
Our study confirms that acute ischemic stroke patients presenting with minor deficits have high rates of disability at 3 months. Fully 3 of every 10 patients had disability or death (mRS score 2–6) at 3 months. We further found that minor PCI patients were more likely than minor ACI patients to exhibit disability at 3 months, with posterior circulation location increasing the odds of disabled outcome by 23%. Furthermore, the presence of large artery disease supplying the territory of the ischemic stroke substantially increased the risk of disability and death, especially for vertebrobasilar disease in the posterior circulation. Minor PCI with PC-LVD (vertebrobasilar disease) exhibited a substantial risk of disability and death at 3 months. The large size of our study provides stronger evidence for these effects that have been previously investigated but not settled.
Our study provides information helpful in considering the potential mechanisms by which PCI confer more disability risk than ACI. Several factors standardly associated with worse functional outcome were actually more common among ACI than among PCI patients, including older age, higher presenting NIHSS score, previous TIA, previous peripheral artery diseases, and presence of RAD. Nonetheless, disability tended to occur more frequently among PCI than among ACI patients in unadjusted analysis and clearly occurred more frequently in analyses adjusting for these factors.
Another potential factor not supported by our data is differential frequency of early stroke progression or early recurrent stroke. Early progression or recurrence during hospitalization, as estimated by NIHSS changes between discharge and admission in our study, did not occur more frequently among PCI than among ACI patients; and among patients without early progression/recurrence during hospitalization, PCI location remained an independent predictor of final disability. These findings suggest that differences in the types of initial deficits among PCI and ACI patients play an important role in determining differences in disability outcome. For the same level of NIHSS scores, PCI-related deficits may be more disabling than ACI-related deficits. In part, this effect may be mediated by PCI-related deficits that are not well captured by the NIHSS, such as gait difficulty or disabling dizziness. The NIHSS is well known to be less sensitive in detecting posterior than anterior circulation deficits.17,18
The results of our study provide reliable information from a large population using a multicenter registry on the magnitude of differences in functional outcomes between minor PCI and ACI. Patients with minor PCI exhibited, with a relatively 26% greater odds of disability, than patients with minor ACI. Among 1000 patients matched for other prognostic factors, minor PCI was associated with 46 more patients having disability at 3 months and 23 more patients having dependence at 3 months than minor ACI.
Our study also found that the presence of major, relevant large artery disease, especially in the posterior circulation, was an important potentiator of risk of disability or death. Odds of disability at 3 months in patients with minor PCI with PC-LVD were increased 58% compared with minor ACI with AC-LVD. Minor PCI with PC-LVD also exhibited a >2 to 3-fold higher risk of death within 3 months than minor ACI and minor PCI without major LVDs. Absolute rate differences were modest but still clinically relevant. The frequency of disability at 3 months was an absolute 10% higher in patients with PC-LVD than those with no PC-LVD. These results suggest that major LVD is an important contributor to worse outcome and that intensive antiatherosclerotic therapy such as high-dose statin and dual antiplatelet therapy, especially of vertebrobasilar disease, has the potential to improve patient course. A possible mechanism by which large artery disease contributes to disability at 3 months is a higher incidence of early recurrent stroke. The frequency of early recurrent strokes was not evaluated in the present study. However, a population-based study showed that the risk of early recurrent stroke is higher in symptomatic large artery vertebrobasilar disease than symptomatic large artery cervical carotid disease.9
For acute minor stroke or high-risk TIA patients, combined antiplatelet therapy with aspirin and clopidogrel was superior to aspirin monotherapy in the CHANCE trial (Clopidogrel in High-Risk Patients With Acute Non-Disabling Cerebrovascular Events).19 However, the population of the current study was not routinely treated with dual antiplatelet therapy. Although combined aspirin and clopidogrel therapy was more frequently used in patients with ACI than in patients with PCI (31.4% versus 27.9%; P=0.001) and most frequently used in patients with AC-LVD followed by PC-LVD, combined aspirin and clopidogrel therapy was not associated with improved functional outcome at 3 months (Table III in the online-only Data Supplement). However, as the CHANCE study differed from the current study in not evaluating functional outcome and not including late postqualifying stroke ≤7 days, combined aspirin and clopidogrel therapy might still have the potential to improve patient course in acute minor ischemic stroke. Because our study was not designed to evaluate the effects of any potential therapy, further study will be warranted to find which therapy could be beneficial for these patients.
Our study has several strengths, including large sample size; prospective, multicenter data collection; and direct comparison between ACI and PCI patients drawn from the same consecutively encountered population. Data were collected by trained research nurses or physicians, with central data audits for outliers and missing values. In addition, DWI was used to ensure accurate localization of cerebral infarction within the defined vascular territories. Most previous studies classified many patients among posterior circulation infarcts and anterior circulation infarcts based on presenting neurological deficits and computed tomography rather than only diffusion magnetic resonance imaging. Because diffusion magnetic resonance imaging studies have shown that the posterior circulation ischemia patients frequently present with nonclassical posterior circulation deficits,7 it is likely that some patients in these series were not accurately classified.
However, there are several limitations to this study. First, although this is a multicenter study performed at 14 sites, all were in South Korea. Further studies are needed in other, especially non-Asian, populations. Second, our statistical adjustment for patient differences may be incomplete because of residual or unmeasured confounding. Third, the results of our study cannot be generalized to all patients with minor stroke or TIA because we excluded patients with diffusion magnetic resonance imaging–negative TIA and patients with diffusion lesions occurring concurrently in both the anterior and posterior circulation. Therefore, the results of our study should be interpreted with caution.
In conclusion, our study showed that long-term disability after minor stroke is common in both ACI and PCI occurring in one third of patients. The risk of disability at 3 months was slightly higher in minor PCI patients than in minor ACI patients, and further research is needed to fully understand the reasons for this. In addition, the presence of large artery vertebrobasilar disease in minor PCI was especially strongly associated with disability and death at 3 months. Our results suggest that minor PCI patients, especially those with PC-LVD, may be candidates for meticulous care and aggressive medical management and merit focused attention in randomized treatment trials.
Sources of Funding
This study was supported by a grant of the Korea Healthcare Technology R&D Project, Ministry of Health and Welfare, Republic of Korea (HI10C2020).
Guest Editor for this article was Stephen M. Davis, MD, FRACP.
The online-only Data Supplement is available with this article at http://stroke.ahajournals.org/lookup/suppl/doi:10.1161/STROKEAHA.116.013432/-/DC1.
- Received March 11, 2016.
- Revision received September 16, 2016.
- Accepted October 21, 2016.
- © 2016 American Heart Association, Inc.
- Jones HR Jr,
- Millikan CH,
- Sandok BA
- McDowell FH,
- Potes J,
- Groch S
- Turney TM,
- Garraway WM,
- Whisnant JP
- Marshall J
- Tao WD,
- Liu M,
- Fisher M,
- Wang DR,
- Li J,
- Furie KL,
- et al
- Flossmann E,
- Rothwell PM
- Marquardt L,
- Kuker W,
- Chandratheva A,
- Geraghty O,
- Rothwell PM
- Gulli G,
- Marquardt L,
- Rothwell PM,
- Markus HS
- Kim BJ,
- Kim HJ,
- Lee DH,
- Kwon SU,
- Kim SJ,
- Kim JS,
- et al
- Adams HP Jr,
- Bendixen BH,
- Kappelle LJ,
- Biller J,
- Love BB,
- Gordon DL,
- et al
- Coutts SB,
- O’Reilly C,
- Hill MD,
- Steffenhagen N,
- Poppe AY,
- Boyko MJ,
- et al