Abstract
Aims: We aimed to validate the predictive value of the ABCD3-I score for short-term and long-term stroke risk after transient ischemic attack (TIA) and to evaluate the influence of symptomatic intracranial artery stenosis (sICAS) on the performance of ABCD3-I.
Methods: We recruited TIA patients from the Third China National Stroke Registry study. Outcome parameters were stroke events during the 14-day, 3-month, 6-month, and 12-month points. The area under the curve (AUC) was calculated as a measure of predictive ability. A multivariable-adjusted Cox proportional hazards model was used to assess the hazard ratio of risk factors for stroke.
Results: Among 986 patients, 3.9%, 5.1%, 6.5 %, and 8.2% of participants experienced a stroke event during the 14-day, 3-month, 6-month, and 12-month points post TIA, respectively. The AUCs of ABCD3-I score for the prediction of stroke were 0.786, 0.732, 0.715, and 0.699 at the 14-day, 3-month, 6-month, and 12-month points, respectively. The AUCs were 0.774, 0.690, 0.617, and 0.611 in patients with sICAS, 0.789, 0.748, and 0.758 and 0.734 in those without sICAS. P values of the interaction between ABCD3-I categories and sICAS were 0.0618 for 14-day, 0.0098 for 3-month, 0.0318 for 6-month, and 0.0294 for 12-month.
Conclusions: ABCD3-I score performed well in predicting short-term risk of a stroke after an index TIA in patients with or without sICAS. However, the predictive power decayed with the prolonged period, and the decayed extent was more pronounced among those with sICAS. The assessment of sICAS is a non-ignorable item when using the ABCD3-I score for long-term stroke risk prediction in patients with TIA.
Introduction
Symptoms of a transient ischemic attack (TIA) are typically temporary-lasting but with a high risk of a subsequent stroke event in the subsequent short-term and long-term period1-4). The ABCD system is a prognostic tool developed to predict stroke risk in the short-term after onset of TIA, and it is used in primary and emergency care settings for identifying high-risk individuals to facilitate triage to specialist care and target secondary prevention5-7). The ABCD2 score was the most widely used in patients with a TIA, and it is recommended by guidelines for aiding clinical management of patients in community and emergency department settings8-10). However, the ABCD2 score is mainly based on clinical features identifiable at the time of initial and does not fulfill the purpose of guiding risk-based treatment decisions after a stroke specialist assessment11, 12). In order to improve the prediction value of the ABCD2 score, a imaging-based score- ABCD3-I score was established with a better predictive performance7, 13-16), and it was validated by several studies17-19). The 13-points ABCD3-I score is calculated by evaluating medical records as follows: age, blood pressure, clinical features, duration of symptoms, diabetes, and dual TIA, as well as imaging criteria including ipsilateral ≥ 50% ICA stenosis and an acute brain lesion.
Although the ABCD3-I score exhibits an excellent predictive value for short-term stroke recurrence in patients with TIA, two other important issues remain to be clarified. First, monitoring high long-term risk of cardiovascular in patients with TIA would be necessary. However, few studies analyzed performance of ABCD3-I at long-term stroke risk prediction after TIA15, 20). Whether ABCD3-I score exhibits the same excellent predictive value for long-term stroke risk as short-term in patients with TIA is unclear. Second, intracranial atherosclerosis is associated with a high risk of stroke event in patients with TIA21-24). Unlike carotid stenosis that is the most common cause for large vessel occlusion in Caucasians, intracranial stenosis (ICAS) is more prevalent in Asians25-27), Hispanics, and African-Americans28). The presence of stenosis of extracranial carotid was included in the ABCD3-I score as an imaging parameter, but assessments of intracranial findings are not incorporated. Whether symptomatic intracranial stenosis (sICAS) affects the predictive value of ABCD3-I score is unknown, especially in the Asian population.
Based on a hospital based study, The Third China National Stroke Registry (CNSR-III)29), we aimed to validate the predictive value of the ABCD3-I score to predict short-term and long-term stroke risk after TIA and to evaluate the influence of sICAS on the performance of ABCD3-I score.
Methods
Study Participants
The CNSR-III is a nationwide prospective registry for patients presented to hospitals with acute ischemic cerebrovascular events29). From August 2015 to March 2018, 15,166 stroke patients with a primary diagnosis of ischemic stroke or TIA within 7 days after symptom onset were enrolled from 201 participating hospitals. In CNSR-III, TIA was primary diagnosed according to time-based definitions on admission (acute onset of neurologic deficit ≤ 24h as TIA, after assessment by stroke specialists on admission)30). In the present study, 986 patients with a definite TIA were analyzed. The study was approved by the central institutional review board at Beijing Tiantan Hospital.
Data Collection
The baseline data were collected after admission through face-to-face interviews by neurologists at participating hospitals, following a standard data collection protocol developed by the steering committee. Pre-hospital care information, pre-stroke modified Rankin scale (mRS) score, in-hospital variables including demographics, medical history, ABCD2 score, family history, previous medication, physical examination, primary diagnosis, laboratory tests, and risk factor assessment were collected at baseline. Risk factors included history of stroke or TIA, hypertension, diabetes, dyslipidemia, atrial fibrillation, heart disease, and previous or current smoking. Next, height and weight were measured with participants wearing a scrub suit and no shoes, and then, body-mass index was calculated by dividing weight in kilograms by the square of height in meters. Hypertension was defined as systolic blood pressure ≥ 140 mmHg, diastolic blood pressure ≥ 90 mmHg, any use of antihypertensive drug continuously for more than 2 weeks before index cerebrovascular diseases, or self-reported history of hypertension. Dyslipidemia was defined as serum triglyceride ≥ 150 mg/dL, low-density lipoprotein cholesterol ≥ 130 mg/dL, high-density lipoprotein cholesterol ≤ 40 mg/dL, any use of lipid-lowering drugs, or any self-reported history of dyslipidemia. Diabetes mellitus was defined as fasting glucose level ≥ 7.0 mmol/L (126 mg/dL), the nonfasting glucose concentration ≥ 11.1 mmol/L (200 mg/dL) with classic symptoms of hyperglycemia or hyperglycemic crisis, any use of glucose-lowering drugs, or any self-reported history of diabetes mellitus. Atrial fibrillation was diagnosed based on electrocardiographic findings on admission or during hospitalization or a previous history of atrial fibrillation. ABCD3-I score range from 0 to 13, with higher scores indicating a greater risk of stroke; an age of 60 years or older, a blood-pressure level of 140/90 mm Hg or higher, a clinical feature with unilateral weakness or speech impairment, a duration of symptoms of 10 to 59 minutes, and diabetes are each assigned 1 point; and a duration of symptoms of 60 minutes or more, dual TIA in the preceding 7 days, ipsilateral ≥ 50% stenosis of internal carotid artery, acute diffusion-weighted imaging hyperintensity are assigned 2 points. Then, dual TIA was defined as the occurrence of ≥ 2 TIAs (i.e., the index TIA and one other TIA) within 7 days. The ABCD3-I scores were analyzed as ordinal variables and classified into low-, moderate-, and high-risk categories (0–3, 4–7, and 8–13).
Neuroimaging Findings
All patients received standard imaging exams during hospitalization, including brain MR imaging (MRI) with DWI (3.0T or 1.5T), intracranial artery imaging (TOF-MRA, CT angiography or digital subtraction angiography), and extracranial artery imaging (carotid ultrasound, CT/MR angiography or digital subtraction angiography). All image data were collected and analyzed centrally at Tiantan Neuroimaging Center of Excellence (T-NICE), Beijing Tiantan Hospital. DWI findings on MRI on admission or during hospitalization were considered abnormal if there was ≥ 1 hyperintense area consistent with the presence of an infarcted lesion. Presence of sICAS and symptomatic extracranial arterial stenosis (sECAS) was defined as 50%–99% stenosis or occlusion of any intra- and extracranial artery, which was responsible for the acute infarction lesion and neurological symptoms, and respectively by the Warfarin-Aspirin Symptomatic Intracranial Disease trial (WASID)31) and the North American Symptomatic Carotid Endarterectomy Trial criteria32). The intracranial artery included bilateral intracranial internal carotid artery, anterior cerebral artery A1/A2, middle cerebral artery M1/M2, bilateral intracranial vertebral artery, posterior cerebral artery P1/P2, and basilar artery, while the extracranial artery included the extracranial internal carotid artery and bilateral extracranial vertebral artery.
Outcome Assessment
The main outcome of this study was a stroke after TIA at the 14-day, 3-month, 6-month, and 12-month points. A stroke event included an ischemic or hemorrhagic stroke. Ischemic stroke was defined as new onset of focal neurological deficit that cannot be attributed to the presenting lesion and is confirmed with radiographic (CT and/or MRI) evidence. Hemorrhagic stroke was defined as new onset of focal neurological deficit and radiographic (CT and/or MRI) findings showing hemorrhagic signs.
Statistical Analysis
Participant demographics and characteristics were analyzed using descriptive statistics. Continuous variables were presented in medians (interquartile ranges) and compared between groups using the nonparametric Wilcoxon test. Categorical variables were presented as percentages and tested by the chi-square test. The incidence of stroke after TIA was compared by using a Cox proportional hazards model. The hazard ratios (HRs) and their 95% confidence intervals (CIs) for the development of stroke at 14-day, 3-month, 6-month, and 12-month after TIA were estimated by using a multivariable-adjusted Cox proportional hazards model. Next, interactions between ABCD3-I categories and sICAS were checked. A two-tailed P value ＜0.05 was considered to be statistically significant. All analysis was conducted with SAS version 9.4 software (SAS Institute).
Data Availability
Anonymized data will be shared by request from any qualified investigator.
Results
Study Participants and Characteristics
Of 15,166 patients participated in CNSR-III, 1,184 presented with a TIA. A total of 986 patients were included into the current study after excluding 198 with missing information on ABCD3-I items or intracranial artery imaging assessment (Fig.1). Supplemental Table 1 showed the baseline characteristics of TIA patients included and excluded in this study. Excluded patients were slightly older, more likely to exhibit a shorter time from event to enrolment, more likely to take antiplatelet agent and clopidogrel with aspirin at discharge, and less likely to present with a medical history of diabetes. The demographic and clinical characteristics of included patients with and without sICAS was present in Table 1. The mean (IQR) age was 62.0 (53.0–70.0) years, and 646 patients (65.5%) were men. Patients with sICAS were older, more likely to present with a history of hypertension, diabetes, stroke, and TIA, and less likely to be smoker or drinker. Other demographic and clinical characteristics of patients can be found in Table 1.
Supplemental Table 1.
Demographic and clinicalcharacteristics
|
Inclusion n = 986
|
Exclusion n = 198
|
P value
|
| Median age (IQR) — yr |
62.0 (53.0-70.0) |
65.0 (56.0-72.0) |
0.01 |
| Male sex - no. (%) |
646 (65.5) |
172 (66.4) |
0.52 |
| Median days from event to enrolment (IQR) |
1.0 (1.0-3.0) |
2.0 (1.0-4.0) |
0.03 |
| Medical history - no. (%) |
|
|
|
| Hypertension |
570 (57.8) |
118 (59.6) |
0.64 |
| Diabetes |
173 (17.5) |
53 (26.8) |
0.003 |
| Dyslipidemia |
117 (11.9) |
19 (9.6) |
0.36 |
| Current smoker |
267 (27.1) |
56 (28.3) |
0.73 |
| Current drinker |
141 (14.3) |
26 (13.1) |
0.67 |
| Stroke |
197 (20.0) |
35 (17.7) |
0.46 |
| TIA |
116 (11.8) |
15 (7.6) |
0.09 |
| Coronary artery disease |
118 (12.0) |
24 (12.1) |
0.95 |
| Heart failure |
2 (0.2) |
0 (0.0) |
0.53 |
| Atrial fibrillation |
46 (4.7) |
7 (3.5) |
0.48 |
| Peripheral artery disease |
13 (1.3) |
2 (1.0) |
0.72 |
| Modified Rankin score before admission - no. (%) |
|
|
0.54 |
| 0 |
805 (81.6) |
168 (84.8) |
|
| 1 |
153 (15.5) |
25 (12.6) |
|
| Body-mass index (IQR) |
24.6 (22.9-26.7) |
24.8 (23.4-26.6) |
0.49 |
| Blood pressure - mm Hg |
|
|
|
| Systolic |
142.0 (130.0-157.5) |
140.0 (130.0-157.5) |
0.37 |
| Diastolic |
83.5 (77.5-91.5) |
81.0 (77.0-90.0) |
0.35 |
| Medication use at discharge - no. (%) |
|
|
|
| Antiplatelet agent |
875 (88.7) |
165 (83.3) |
0.03 |
| Lipid-lowering agent |
869 (88.1) |
167 (84.3) |
0.14 |
| Anticoagulant agent |
26 (2.6) |
2 (1.0) |
0.17 |
| Antihypertensive agent |
432 (43.8) |
81 (40.9) |
0.45 |
sICAS indicates symptomatic intracranial stenosis and TIA, transient ischemic attack.
Table 1.
Demographic and clinical characteristics
|
Total n = 986
|
With ICAS n = 259
|
Without ICAS n = 727
|
P value
|
| Median age (IQR) — yr |
62.0 (53.0–70.0) |
65.0 (56.0–72.0) |
61.0 (52.0–68.0) |
|
| Male sex- no. (%) |
646 (65.5) |
172 (66.4) |
474 (65.2) |
0.73 |
| Median days event to enrolment (IQR) |
1.0 (1.0–3.0) |
1.0 (1.0–3.0) |
1.0 (1.0–3.0) |
0.4 |
| ABCD3-I |
|
|
|
＜0.001 |
| 0 to 3 |
313 (31.7) |
53 (20.5) |
260 (35.8) |
|
| 4 to 7 |
616 (62.5) |
189 (73.0) |
427 (58.7) |
|
| 8 to 13 |
57 (5.8) |
17 (6.6) |
40 (5.5) |
|
| Medical history - no. (%) |
|
|
|
|
| Hypertension |
570 (57.8) |
176 (68.0) |
394 (54.2) |
＜0.001 |
| Diabetes |
173 (17.5) |
54 (20.8) |
119 (16.4) |
0.1 |
| Dyslipidemia |
117 (11.9) |
27 (10.4) |
90 (12.4) |
0.4 |
| Current smoker |
267 (27.1) |
63 (24.3) |
204 (28.1) |
0.25 |
| Current drinker |
141 (14.3) |
30 (11.6) |
111 (15.3) |
0.15 |
| Stroke |
197 (20.0) |
70 (27.0) |
127 (17.5) |
0.001 |
| TIA |
116 (11.8) |
38 (14.7) |
78 (10.7) |
0.09 |
| Coronary artery disease |
118 (12.0) |
39 (15.1) |
79 (10.9) |
0.07 |
| Heart failure |
2 (0.2) |
1 (0.4) |
1 (0.1) |
0.46 |
| Atrial fibrillation |
46 (4.7) |
15 (5.8) |
31 (4.3) |
0.32 |
| Peripheral artery disease |
13 (1.3) |
4 (1.5) |
9 (1.2) |
0.75 |
| Modified Rankin score before admission - no. (%) |
|
|
|
0.16 |
| 0 |
805 (81.6) |
203 (78.4) |
602 (82.8) |
|
| 1 |
153 (15.5) |
47 (18.1) |
106 (14.6) |
|
| Body-mass index (IQR) |
24.6 (22.9–26.7) |
24.8 (22.5–26.9) |
24.6 (22.9–26.7) |
0.82 |
| Blood pressure - mm Hg |
|
|
|
|
| Systolic |
142.0 (130.0–157.5) |
145.0 (130.0–160.0) |
141.0 (130.0–156.0) |
0.05 |
| Diastolic |
83.5 (77.5–91.5) |
82.5 (76.5–91.0) |
84.0 (78.0–92.0) |
0.12 |
| Medication use at discharge - no. (%) |
|
|
|
|
| Antiplatelet agent |
875 (88.7) |
236 (91.1) |
639 (87.9) |
0.16 |
| Lipid-lowering agent |
869 (88.1) |
240 (92.7) |
629 (86.5) |
0.009 |
| Anticoagulant agent |
26 (2.6) |
8 (3.1) |
18 (2.5) |
0.6 |
| Antihypertensive agent |
432 (43.8) |
130 (50.2) |
302 (41.5) |
0.02 |
sICAS indicates symptomatic intracranial stenosis and TIA, transient ischemic attack.
As shown in Fig.2, the incidence of stroke was 3.9%, 5.1%, 6.5% and 8.2% at 14-day, 3-month, 6-month, and 12-month after onset of TIA. The rate of stroke event was 4.3%, 7.0%, 9.7% and 10.2% in patients with sICAS, while 3.7%, 4.4%, 5.4%, and 7.3% in patients without sICAS at four follow-up time point. A higher rate of stroke was observed in patients with sICAS. As compared to patients without sICAS, a steeper increase in stroke incidence as follow-up time longed. When stratified by the ABCD3-I score, the stroke incidence was higher in the high-risk group than in the low-risk group (2.9%, 9.6% and 22.8% for low-, moderate-, and high-risk categories at 12-month follow-up). When stratified by the ABCD3-I score and measurement of sICAS, an extremely high stroke incidence was observed in high-risk group patients with sICAS (29.4% at 14-day and 41.2% at 12-month).
Validity of and ABCD3-I Scores
Predictive ability of ABCD3-I reduced as time extends using the receiver-operating. As shown in Table 2, the area under the curves (AUCs) were 0.786 (95% CI, 0.718–0.855), 0.732 (95% CI, 0.658, 0.807), 0.715 (95% CI, 0.649, 0.781), and 0.699 (95% CI, 0.640, 0.758) for the prediction of 14-day, 3-month, 6-month, and 12-month occurrence, respectively. In patients with sICAS, the AUCs were 0.774 (95% CI, 0.614, 0.933), 0.690 (95% CI, 0.537, 0.844), 0.617 (95% CI, 0.488 0.747), and 0.611 (95% CI, 0.489, 0.732) for the prediction of 14-day, 3-month, 6-month, and 12-month occurrence, respectively. In patients without sICAS, the AUCs were 0.789 (95% CI, 0.710, 0.868), 0.748 (95% CI, 0.663, 0.833), 0.758 (95% CI, 0.684, 0.833), and 0.734 (95% CI, 0.669, 0.799) for the prediction of 14-day, 3-month, 6-month, and 12-month occurrence, respectively. In patients with sICAS, a clear reduction was observed in AUC value by time. However, in patients without sICAS, no noticeable decline in performance of ABCD3-I was found over time.
Table 2.
Predictive accuracy of ABCD3-I for stroke risk at different time points of follow-up and in different population
| Stroke Incidence |
AUC value |
| Total |
sICAS |
sICAS free |
| 14-Day |
0.786 (0.718, 0.855) |
0.774 (0.614, 0.933) |
0.789 (0.710, 0.868) |
| 3-Month |
0.732 (0.658, 0.807) |
0.690 (0.537, 0.844) |
0.748 (0.663, 0.833) |
| 6-Month |
0.715 (0.649, 0.781) |
0.617 (0.488, 0.747) |
0.758 (0.684, 0.833) |
| 12-Month |
0.699 (0.640, 0.758) |
0.611 (0.489, 0.732) |
0.734 (0.669, 0.799) |
AUC indicates area under the curve and sICAS, symptomatic intracranial stenosis
Table 3 presents odds ratios with 95% CI of high-, moderate-, versus low-risk category of ABCD3-I for stroke risk in patients stratified by the presence of sICAS. In patients with sICAS, compared to low-risk category, no significant predictive effect was found for stroke risk at all time points for moderate-risk category. P values of the interaction between ABCD3-I categories and sICAS were 0.0618 for 14-day, 0.0098 for 3 months, 0.0318 for 6 months, and 0.0294 for 12 months stroke.
Table 3.
Multivariable-adjusted odds ratios for stroke after TIA at different time points of follow-up
|
Odds Ratios (95% CI) |
Interaction p
value
|
| sICAS |
sICAS free |
| Crude model
|
|
|
|
| 14-day |
|
|
0.0621 |
| ABCD3-I(4–7) ＊
|
NA |
14.74 (1.98, 109.83) |
|
| ABCD3-I(8–13)＊
|
NA |
21 (2.13, 207.19) |
|
| 3-month |
|
|
0.0079 |
| ABCD3-I(4–7)＊
|
1.28 (0.27, 6.09) |
8.71 (2.05, 36.93) |
|
| ABCD3-I(8–13)＊
|
17.85 (3.22, 98.83) |
10.46 (1.69, 64.69) |
|
| 6-month |
|
|
0.0285 |
| ABCD3-I(4–7)＊
|
1.44 (0.4, 5.16) |
10.45 (2.48, 43.98) |
|
| ABCD3-I(8–13)＊
|
11.67 (2.57, 52.99) |
18.43 (3.44, 98.62) |
|
| 12-month |
|
|
0.0294 |
| ABCD3-I(4–7)＊
|
1.44 (0.4, 5.16) |
5.56 (2.17, 14.25) |
|
| ABCD3-I(8–13)＊
|
8.58 (2.11, 34.92) |
9 (2.61, 31.09) |
|
| Adjusted model
|
|
|
|
| 14-day |
|
|
0.0618 |
| ABCD3-I (4–7)＊
|
NA |
16.66 (2.23, 124.5) |
|
| ABCD3-I (8–13)＊
|
NA |
28.21 (2.8, 284.37) |
|
| 3-month |
|
|
0.0098 |
| ABCD3-I (4–7)＊
|
1.12 (0.23, 5.45) |
9.67 (2.27, 41.17) |
|
| ABCD3-I (8–13)＊
|
15.37 (2.71, 87.06) |
13.21 (2.09, 83.39) |
|
| 6-month |
|
|
0.0318 |
| ABCD3-I (4–7)＊
|
1.22 (0.33, 4.48) |
11.52 (2.73, 48.65) |
|
| ABCD3-I (8–13)＊
|
10.67 (2.29, 49.83) |
22.78 (4.17, 124.32) |
|
| 12-month |
|
|
0.0311 |
| ABCD3-I (4–7)＊
|
1.01 (0.32, 3.2) |
5.81 (2.26, 14.94) |
|
| ABCD3-I (8–13)＊
|
8.08 (1.92, 34) |
9.68 (2.76, 33.89) |
|
＊Compared with low-risk category ABCD3-I (0–3) Age and gender were adjusted in adjusted model
CI indicates confidence interval and sICAS, symptomatic intracranial stenosis
Discussion
Our data confirmed that the ABCD3-I score performed well in predicting short-term risk of stroke after an index TIA with or without sICAS. However, the predictive power of the ABCD3-I score decayed with the prolonged follow-up period and the decayed extent was more pronounced among those with sICAS.
The results of the present study indicate that early and late stroke risk after TIA in this population was similar to the previous study3, 33, 34), In our study, we found a reliable predictive utility of the ABCD3-I score for short-term stroke risk (within 14-day) in TIA patients. The performance of ABCD3-I in predicting early stroke risk after TIA was good (0.786 for 14-day and 0.732 for 3 months), consistent with some other studies in which the AUCs fluctuated between 0.70-0.76 16-18). In contrast, the predictive ability of the score for long-term stroke risk was limited. Also, another study reported the ability of the ABCD3-I in prediction of l3-year stroke risk after TIA was relatively low, with an AUC value of 0.61 15).
In addition, the predictive performance of ABCD3-I for late stoke risk could be influenced by the presence of sICAS. It is well known that patients with sICAS exhibit an extremely high risk of subsequent ischemic cerebrovascular events in both Eastern and Western population35, 36). A major reason for the inconsistencies predictive value of ABCD3-I score between different periods is that patients with sICAS may exhibit more unstable vessel plaques, which leads to a high risk of stroke event within short-term follow-up. Previous studies showed that as the number of risk factors increases, stroke risks increase in patients with sICAS35). In our study, patients with sICAS and more risk factors (ABCD3-I score 8–13) exhibited a higher risk of stroke risk when compared with low-risk factors (ABCD3-I score 0–7), and the stroke events were mostly occurred within 3 months. These may lead to the sharp AUC decay over time in patients with sICAS after TIA.
In the present study, we found a potential interaction between ABCD3-I score and the presence of sICAS for stroke recurrence after 3 months, suggesting that the association between ABCD3-I score and stroke risk were modified by sICAS. Most studies indicated Caucasians exhibited a relatively low incidence of ICAS. However, the prevalence of intracranial stenosis in Caucasians may have been underestimated. A recent UK cohort study reported that about 18% of individuals presented with 50–99% symptomatic or asymptomatic intracranial stenosis ICAS in patients with TIA or minor stroke36). In addition, it has been proposed that patients with ipsilateral atherosclerotic stenosis ＞50%（including intracranial artery）could play as an important supplementary item for screening out high risk TIA37, 38). Like extracranial internal carotid artery stenosis screening, it should be emphasized for the utility of routine intracranial stenosis screening after TIA for secondary prevention39). Furthermore, considering the inconsistent performance of ABCD3-I score in the two populations, sICAS should be taken into account for long-term stroke risk prediction in patients with TIA.
Although this is a nationwide registration study, some limitations in this study also cannot be avoided. First, 2,154 (14%) patients without a baseline MRI examination in CNSR-III were excluded. However, the baseline characteristics were similar between patients with and without MRI in CNSR-III (Supplemental Table 2). Furthermore, the CNSR-III was conducted exclusively among Chinese patients in whom a high proportion of ICAS exists. Thus, whether the results of this study can generalize to other non-Asia populations and require further validation in other ethnicities remains unknown. Finally, we only reported the results of 1-year follow-up, for the study has not completed the follow-up beyond the 1-year point, and 5-year follow-up results will be released in the future.
Supplemental Table 2.
Baseline characteristics of patients with and without MRI exam in CNSR-III
| Characteristics |
With MRI (N = 13,012)
|
Without MRI (N = 2,154)
|
P Value
|
| Age - year |
62.3±11.2 |
61.6±11.8 |
0.01 |
| Male sex - no. (%) |
8890 (68.3) |
1474 (68.4) |
0.92 |
| Interval from onset to enrolment - day |
2.3±2.0 |
2.4±2.5 |
0.56 |
| ABCD2 (for TIA patients) |
4.2±1.3 |
3.1±1.8 |
＜0.001 |
| NIHSS (for MIS patients) |
4.2±4.1 |
4.3±4.6 |
0.50 |
| Medical history - no. (%) |
|
|
|
| Hypertension |
8167 (62.8) |
1327 (61.6) |
0.30 |
| Diabetes |
2964 (22.8) |
546 (25.3) |
0.009 |
| Dyslipidemia |
1024 (7.9) |
167 (7.8) |
0.85 |
| Current smoker |
4097 (31.5) |
655 (30.4) |
0.32 |
| Current drinker |
1827 (14.0) |
299 (13.9) |
0.84 |
| Stroke |
2850 (21.9) |
505 (23.4) |
0.11 |
| TIA |
355 (2.7) |
61 (2.8) |
0.78 |
| Coronary artery disease |
1315 (10.1) |
293 (13.6) |
＜0.001 |
| Heart failure |
80 (4.6) |
14 (3.6) |
0.40 |
| Atrial fibrillation |
856 (6.6) |
163 (7.6) |
0.09 |
| Peripheral artery disease |
98 (0.8) |
20 (0.9) |
0.39 |
| mRS before admission - no. (%) |
|
|
0.81 |
| 0 |
9647 (74.1) |
1600 (74.3) |
|
| 1 |
2224 (17.1) |
351 (16.3) |
|
| Body-mass index |
24.7±3.4 |
24.7±3.2 |
0.82 |
| Blood pressure - mm Hg |
|
|
|
| Systolic |
150.2±22.1 |
149.5±22.0 |
0.09 |
| Diastolic |
87.4±13.1 |
87.0±13.1 |
0.15 |
| TOAST classification - no. (%) |
|
|
＜0.001 |
| Large-artery atherosclerosis |
3431 (26.4) |
425 (19.7) |
|
| Small-artery occlusion |
2968 (22.8) |
197 (9.1) |
|
| Cardioembolism |
768 (5.9) |
149 (6.9) |
|
| Other determined etiology |
174 (1.3) |
8 (0.4) |
|
| Undetermined etiology |
5671 (43.6) |
1375 (63.8) |
|
| Medication use at discharge - no. (%) |
|
|
|
| Antiplatelet agent |
11818 (91.0) |
1893 (88.4) |
＜0.001 |
| Lipid-lowering agent |
11908 (91.7) |
1923 (89.8) |
0.005 |
| Anticoagulant agent |
362 (2.8) |
87 (4.1) |
0.001 |
| Antihypertensive agent |
6383 (49.1) |
1035 (48.3) |
0.49 |
MRI, Magnetic Resonance Imaging; DWI, Diffusion-Weighted Imaging; TIA, transient ischemic attack; MIS, minor ischemic stroke; BMI, body mass index (calculated as weight in kilograms divided by height in meters squared); NIHSS, National Institute of Health Stroke Scale; TOAST classification: The Trial of Org 10172 in Acute Stroke Treatment classification indicates the probable cause of the index TIA/MIS, including five categories, large-artery atherosclerosis, cardioembolism, small-vessel occlusion, other determined cause, and undetermined cause; ABCD2, a score (0-7 points) to predict stroke risk following a TIA, including age ≥ 60 years (1 point), blood pressure ≥ 140/90 mm
Hg (1 point), symptoms of unilateral weakness or speech impairment (1-2 points), duration of symptoms of ＜10 to 59 minutes or ≥ 60
minutes (1-2 points), and diabetes (1 point).
＊Values are expressed as means±SD or number (%).
Conclusion
ABCD3-I score performs well in predicting short-term risk of stroke after an index TIA with or without sICAS. However, the predictive power of the ABCD3-I decayed over time, and the decayed is more pronounced among TIA with sICAS. The item sICAS should not be ignored when using the ABCD3-I score for long-term stroke risk prediction in patients with TIA.
Contributors
Yongjun Wang designed the study. Xuewei Xie and Jing Jing interpreted analysis of the data and prepared the report. Zixiao Li, Yilong Wang, Yuesong Pan, Pan Chen, Liping Liu, Yong Jiang, and Hao Li contributed to comments on the draft manuscript and revised the report. Xia Meng coordinated the study and revised the report. Aoming Jin conducted the statistical analysis.
Acknowledgements
We thank all the participating centers in CNSR-III for their hard work in data collection.
Conflict of Interest
The authors declare no financial or other competing interests relevant to the manuscript.
Funding
This work was supported by grants from National Key R&D Program of China (2018YFC1312903), grants from National Science and Technology Major Project (2017ZX09304018), grants from Beijing Municipal Science & Technology Commission (D171100003017002, Z181100001818001).
References
- 1) Coull AJ, Lovett JK, Rothwell PM and Oxford Vascular S: Population based study of early risk of stroke after transient ischaemic attack or minor stroke: implications for public education and organisation of services. BMJ, 2004; 328: 326
- 2) Johnston SC, Gress DR, Browner WS and Sidney S: Short-term prognosis after emergency department diagnosis of TIA. JAMA, 2000; 284: 2901-2906
- 3) Giles MF and Rothwell PM: Risk of stroke early after transient ischaemic attack: a systematic review and meta-analysis. Lancet Neurol, 2007; 6: 1063-1072
- 4) Amarenco P, Lavallee PC, Monteiro Tavares L, Labreuche J, Albers GW, Abboud H, Anticoli S, Audebert H, Bornstein NM, Caplan LR, Correia M, Donnan GA, Ferro JM, Gongora-Rivera F, Heide W, Hennerici MG, Kelly PJ, Kral M, Lin HF, Molina C, Park JM, Purroy F, Rothwell PM, Segura T, Skoloudik D, Steg PG, Touboul PJ, Uchiyama S, Vicaut E, Wang Y, Wong LKS and Investigators TIo: Five-Year Risk of Stroke after TIA or Minor Ischemic Stroke. N Engl J Med, 2018; 378: 2182-2190
- 5) Rothwell PM, Giles MF, Flossmann E, Lovelock CE, Redgrave JN, Warlow CP and Mehta Z: A simple score (ABCD) to identify individuals at high early risk of stroke after transient ischaemic attack. Lancet, 2005; 366: 29-36
- 6) Johnston SC, Rothwell PM, Nguyen-Huynh MN, Giles MF, Elkins JS, Bernstein AL and Sidney S: Validation and refinement of scores to predict very early stroke risk after transient ischaemic attack. Lancet, 2007; 369: 283-292
- 7) Merwick A, Albers GW, Amarenco P, Arsava EM, Ay H, Calvet D, Coutts SB, Cucchiara BL, Demchuk AM, Furie KL, Giles MF, Labreuche J, Lavallee PC, Mas JL, Olivot JM, Purroy F, Rothwell PM, Saver JL, Sheehan OC, Stack JP, Walsh C and Kelly PJ: Addition of brain and carotid imaging to the ABCD(2) score to identify patients at early risk of stroke after transient ischaemic attack: a multicentre observational study. Lancet Neurol, 2010; 9: 1060-1069
- 8) Easton JD, Saver JL, Albers GW, Alberts MJ, Chaturvedi S, Feldmann E, Hatsukami TS, Higashida RT, Johnston SC, Kidwell CS, Lutsep HL, Miller E and Sacco RL: Definition and evaluation of transient ischemic attack: a scientific statement for healthcare professionals from the American Heart Association/American Stroke Association Stroke Council; Council on Cardiovascular Surgery and Anesthesia; Council on Cardiovascular Radiology and Intervention; Council on Cardiovascular Nursing; and the Interdisciplinary Council on Peripheral Vascular Disease. The American Academy of Neurology affirms the value of this statement as an educational tool for neurologists. Stroke, 2009; 40: 2276-2293
- 9) Clinical Guidelines for Stroke and TIA Management. 2008;
- 10) Stroke - National clinical guideline for diagnosis and initial management of acute stroke and transient ischaemic attack (TIA) 2008;
- 11) Wardlaw JM, Brazzelli M, Chappell FM, Miranda H, Shuler K, Sandercock PA and Dennis MS: ABCD2 score and secondary stroke prevention: meta-analysis and effect per 1,000 patients triaged. Neurology, 2015; 85: 373-380
- 12) Diener HC and Frank B: Stroke: Stroke prevention--time to say goodbye to the ABCD2 score? Nat Rev Neurol, 2015; 11: 552-553
- 13) Purroy F, Jimenez-Caballero PE, Mauri-Capdevila G, Torres MJ, Gorospe A, Ramirez Moreno JM, de la Ossa NP, Canovas D, Arenillas J, Alvarez-Sabin J, Martinez Sanchez P, Fuentes B, Delgado-Mederos R, Marti-Fabregas J, Rodriguez Campello A, Masjuan J and Promapa study: Stroke Project CDSGSNS: Predictive value of brain and vascular imaging including intracranial vessels in transient ischaemic attack patients: external validation of the ABCD3-I score. Eur J Neurol, 2013; 20: 1088-1093
- 14) Song B, Fang H, Zhao L, Gao Y, Tan S, Lu J, Sun S, Chandra A, Wang R and Xu Y: Validation of the ABCD3-I score to predict stroke risk after transient ischemic attack. Stroke, 2013; 44: 1244-1248
- 15) Kiyohara T, Kamouchi M, Kumai Y, Ninomiya T, Hata J, Yoshimura S, Ago T, Okada Y, Kitazono T and Fukuoka Stroke Registry I: ABCD3 and ABCD3-I scores are superior to ABCD2 score in the prediction of short- and long-term risks of stroke after transient ischemic attack. Stroke, 2014; 45: 418-425
- 16) Kelly PJ, Albers GW, Chatzikonstantinou A, De Marchis GM, Ferrari J, George P, Katan M, Knoflach M, Kim JS, Li L, Lee EJ, Olivot JM, Purroy F, Raposo N, Rothwell PM, Sharma VK, Song B, Tsivgoulis G, Walsh C, Xu Y and Merwick A: Validation and comparison of imaging-based scores for prediction of early stroke risk after transient ischaemic attack: a pooled analysis of individual-patient data from cohort studies. Lancet Neurol, 2016; 15: 1238-1247
- 17) Merwick Á, Albers GW, Amarenco P, Arsava EM, Ay H, Calvet D, Coutts SB, Cucchiara BL, Demchuk AM, Furie KL, Giles MF, Labreuche J, Lavallée PC, Mas J-L, Olivot JM, Purroy F, Rothwell PM, Saver JL, Sheehan ÓC, Stack JP, Walsh C and Kelly PJ: Addition of brain and carotid imaging to the ABCD2 score to identify patients at early risk of stroke after transient ischaemic attack: a multicentre observational study. The Lancet Neurology, 2010; 9: 1060-1069
- 18) Dai Q, Sun W, Xiong Y, Hankey GJ, Xiao L, Zhu W, Ma M, Liu W, Liu D, Cai Q, Han Y, Duan L, Chen X, Xu G and Liu X: From clinical to tissue-based dual TIA: Validation and refinement of ABCD3-I score. Neurology, 2015; 84: 1426-1432
- 19) Knoflach M, Lang W, Seyfang L, Fertl E, Oberndorfer S, Daniel G, Seifert-Held T, Brainin M, Krebs S, Matosevic B, Toll T, Kiechl S, Willeit J, Ferrari J and Austrian Stroke Unit C: Predictive value of ABCD2 and ABCD3-I scores in TIA and minor stroke in the stroke unit setting. Neurology, 2016; 87: 861-869
- 20) Ildstad F, Ellekjær H, Wethal T, Lydersen S, Fjærtoft H and Indredavik B: ABCD2 and ABCD3-I scores in a TIA population with low stroke risk. 2020; 2021: 8845898
- 21) Gorelick PB, Wong KS, Bae HJ and Pandey DK: Large artery intracranial occlusive disease: a large worldwide burden but a relatively neglected frontier. Stroke, 2008; 39: 2396-2399
- 22) Kang DW, Kwon SU, Yoo SH, Kwon KY, Choi CG, Kim SJ, Koh JY and Kim JS: Early recurrent ischemic lesions on diffusion-weighted imaging in symptomatic intracranial atherosclerosis. Arch Neurol, 2007; 64: 50-54
- 23) Bang OY: Intracranial atherosclerosis: current understanding and perspectives. J Stroke, 2014; 16: 27-35
- 24) Ssi-Yan-Kai G, Nasr N, Faury A, Catalaa I, Cognard C, Larrue V and Bonneville F: Intracranial artery stenosis or occlusion predicts ischemic recurrence after transient ischemic attack. Am J Neuroradiol, 2013; 34: 185-190
- 25) Hoshino T, Uchiyama S, Wong LKS, Sissani L, Albers GW, Bornstein NM, Caplan LR, Donnan GA, Ferro JM, Hennerici MG, Labreuche J, Lavallee PC, Molina C, Rothwell PM, Steg PG, Touboul PJ, Vicaut E, Amarenco P and Investigators TIo: Differences in Characteristics and Outcomes Between Asian and Non-Asian Patients in the TIAregistry.org. Stroke, 2017; 48: 1779-1787
- 26) Uchiyama S, Hoshino T, Charles H, Kamiyama K, Nakase T, Kitagawa K, Minematsu K, Todo K, Okada Y, Nakagawara J, Nagata K, Yamagami H, Yamaguchi T and Amarenco P: Japanese and Non-Japanese Patients with Transient Ischemic Attack or Minor Stroke: A Five-Year Risk Analysis of Stroke and Vascular Events. J Atheroscler Thromb, 2021; 28: 656-664
- 27) Uchiyama S, Hoshino T, Sissani L, Linsay MT, Kamiyama K, Nakase T, Kitagawa K, Minematsu K, Todo K, Okada Y, Nakagawara J, Nagata K, Yamagami H, Yamaguchi T, Amarenco P and Investigators TIo: Japanese Versus Non-Japanese Patients with Transient Ischemic Attack or Minor Stroke: Subanalysis of TIA registry.org. J Stroke Cerebrovasc Dis, 2019; 28: 2232-2241
- 28) Caplan LR, Gorelick PB and Hier DB: Race, sex and occlusive cerebrovascular disease: a review. Stroke, 1986; 17: 648-655
- 29) Wang Y, Jing J, Meng X, Pan Y, Wang Y, Zhao X, Lin J, Li W, Jiang Y, Li Z, Zhang X, Yang X, Ji R, Wang C, Wang Z, Han X, Wu S, Jia Z, Chen Y and Li H: The Third China National Stroke Registry (CNSR-III) for patients with acute ischaemic stroke or transient ischaemic attack: design, rationale and baseline patient characteristics. Stroke Vasc Neurol, 2019; 4: 158-164
- 30) World Health O: Cerebrovascular diseases: prevention, treatment, and rehabilitation, report of a WHO meeting [held in Monaco from 25 to 29 May 1970]. 1971;
- 31) Samuels OB, Joseph GJ, Lynn MJ, Smith HA and Chimowitz MI: A standardized method for measuring intracranial arterial stenosis. AJNR Am J Neuroradiol, 2000; 21: 643-646
- 32) Fox AJ: How to measure carotid stenosis. Radiology, 1993; 186: 316-318
- 33) Amarenco P, Lavallee PC, Labreuche J, Albers GW, Bornstein NM, Canhao P, Caplan LR, Donnan GA, Ferro JM, Hennerici MG, Molina C, Rothwell PM, Sissani L, Skoloudik D, Steg PG, Touboul PJ, Uchiyama S, Vicaut E, Wong LK and Investigators TIo: One-Year Risk of Stroke after Transient Ischemic Attack or Minor Stroke. N Engl J Med, 2016; 374: 1533-1542
- 34) Coull AJ, Lovett JK and Rothwell PM: Population based study of early risk of stroke after transient ischaemic attack or minor stroke: implications for public education and organisation of services. BMJ, 2004; 328: 326
- 35) Wang Y, Zhao X, Liu L, Soo YO, Pu Y, Pan Y, Wang Y, Zou X, Leung TW, Cai Y, Bai Q, Wu Y, Wang C, Pan X, Luo B, Wong KS and Group CS: Prevalence and outcomes of symptomatic intracranial large artery stenoses and occlusions in China: the Chinese Intracranial Atherosclerosis (CICAS) Study. Stroke, 2014; 45: 663-669
- 36) Hurford R, Wolters FJ, Li L, Lau KK, Kuker W, Rothwell PM and Oxford Vascular Study Phenotyped C: Prevalence, predictors, and prognosis of symptomatic intracranial stenosis in patients with transient ischaemic attack or minor stroke: a population-based cohort study. Lancet Neurol, 2020; 19: 413-421
- 37) Johnston SC, Amarenco P, Denison H, Evans SR, Himmelmann A, James S, Knutsson M, Ladenvall P, Molina CA, Wang Y and Investigators T: The Acute Stroke or Transient Ischemic Attack Treated with Ticagrelor and Aspirin for Prevention of Stroke and Death (THALES) trial: Rationale and design. Int J Stroke, 2019; 1747493019830307
- 38) Johnston SC, Amarenco P, Albers GW, Denison H, Easton JD, Held P, Jonasson J, Minematsu K, Molina CA and Wong LK: Acute Stroke or Transient Ischemic Attack Treated with Aspirin or Ticagrelor and Patient Outcomes (SOCRATES) trial: rationale and design. Int J Stroke, 2015; 10: 1304-1308
- 39) Kernan WN, Ovbiagele B, Black HR, Bravata DM, Chimowitz MI, Ezekowitz MD, Fang MC, Fisher M, Furie KL, Heck DV, Johnston SC, Kasner SE, Kittner SJ, Mitchell PH, Rich MW, Richardson D, Schwamm LH, Wilson JA, American Heart Association Stroke Council CoC, Stroke Nursing CoCC and Council on Peripheral Vascular D: Guidelines for the prevention of stroke in patients with stroke and transient ischemic attack: a guideline for healthcare professionals from the american heart association/american stroke association. Stroke, 2014; 45: 2160-2236
