Intravenous Thrombolysis for Acute Mild Ischemic Stroke Patients: Higher ABCD2 Score Associated with Better Outcome

Background: Up to 30% of patients with mild ischemic stroke suffer neurologic deterioration. However, optimal medical approaches of such patients remain controversial given the efficacy and safety of intravenous thrombolysis (IVT). The purpose of this study was to evaluate whether patients with acute mild stroke stratified with ABCD2 score (the risk of stroke on basis of age, blood pressure, clinical features, duration of symptoms, and presence of diabetes mellitus) could benefit from IVT. Methods: Among 3321 patients with a final diagnosis of acute ischemic stroke, we retrospectively included 224 patients identified with acute mild neurologic deficits (National Institution of Health Stroke Scale, NIHSS ≤5) treated with or without IVT. Odds ratios (OR) with their confidence intervals (CI) for outcomes between groups were assessed by using multivariable binary logistic regression analyses. And the heterogeneity of treatment effect magnitude for excellent outcome at 90d (modified Rankin Scale mRS 0-1) was estimated in different subgroups. Results: A total of 224 cases were enrolled, 106 receiving IVT and 118 treated with secondary stroke prevention strategies alone. At 7d, 30 (28.3%) patients with IVT treatment versus 16 (13.6%) patients not receiving IVT achieved significant improvement (≥4-point NIHSS score decrease or complete resolution; OR, 2.448; 95%CI, 1.204-4.977; P=0.013). At 90d, excellent outcome was achieved in 83 (78.3%) patients treated with IVT versus 77 (65.35%) patients without IVT treatment (OR, 3.156; 95%CI, 1.526-6.528; P=0.002), especially in those with ABCD2 score ≥5 (OR, 2.768; 95%CI, 1.196-6.406; P=0.017) and with stroke subtype of large artery atherosclerosis (OR, 5.616; 95%CI, 1.080-29.210; P=0.040). Besides, 7(6.6%) IVT-treated patients versus 2 (1.7%) non-IVT-treated ABCD 2 score: The risk of stroke on basis of age, blood pressure, clinical features, duration of symptoms, and presence of diabetes mellitus; ABCD 3 -I score: Addition of dual symptom onsets within 7d, ipsilateral carotid artery stenosis≥50% and positive brain imaging on the basis of ABCD 2 score; AIS: Acute ischemic stroke; ARAMIS: Antiplatelet vs rt-PA for acute mild ischemic stroke; CI: Confidence interval; END: Early neurological deterioration; FPG, fasting plasma glucose; HDL-C high-density lipoprotein cholesterol; ICH: Intracranial hemorrhage; IQR: Interquartile range; IVT: Intravenous thrombolysis; LDL-C, low-density lipoprotein cholesterol; MaRISS: Mild and rapidly improving stroke study; mRS: Modified Rankin scale; NIHSS: National Institutes of Health Stroke Scale; OR: Odds ratio; PRISMS: Potential of rtPA for ischemic strokes with mild symptoms; rt-PA: Recombinant tissue plasminogen activator; SD: Standard deviation; sICH: Symptomatic intracranial hemorrhage; TC, total cholesterol; TG, total glyceride; TIA: Transient ischemic stroke; TOAST: Trial of org 10172 in acute stroke treatment.

]. Nevertheless, minor neurological deficiency or rapidly improving symptoms are the most common reasons for withdraw from IVT in otherwise guideline-based eligible patients [ 3 ]. In addition, mild stroke is considered to be the beginning of a potential dynamic process. Previous data indicated that 30% of such patients had persistent disability at 90d [ 4 ], and the risk of recurrent stroke at 90d, 1 year and 5 years were 3.7%, 5.1% and 9.5%, respectively [ 5 , 6 ].
Although the proportion of thrombolytic therapy for mild stroke patients has increased over the past decade, the efficacy and safety of recombinant tissue plasminogen activator (rt-PA) remains controversial. A series of studies suggested patients with mild deficiency could benefit from IVT [7][8][9][10][11]. Currently, the American Stroke Association gives a level I (strong) and IIb (weak) evidence-based score respectively for IVT in patients with mild disabling and nondisabling ischemic stroke symptoms within 3h [ 12 ]. But there were also studies indicating no significance in the efficacy between IVT and antiplatelet therapy in patients with mild stroke [ 14 ] , especially in those with non-disabling deficiency [ 14 ]. And the patient treated with IVT was not recommended given antithrombotic therapy within 24 hours, even though an aggravation of the condition. In consideration of the controversial therapeutic decision-making, it is necessary to screen high-risk mild stroke patients for IVT.
It is well known that the ABCD scoring system is designed for evaluating the likelihood of recurrent stroke in transient ischemic attack and minor stroke patients [ 6 , 15 , 16 ]. However, it is not clear the effectiveness of ABCD 2 score to evaluate the prognosis in patients with mild stroke.
The purpose of this observational study was to investigate whether patients with acute mild stroke symptoms (NIHSS score≤5) could achieve 7-and 90-day favorable outcome from intravenous rt-PA therapy by comparing with patients treated with timely secondary stroke prevention strategies not receiving rt-PA. We also hypothesized that ABCD 2 score might be related to the prognosis of mild stroke patients.

Study Design
This study was designed as a retrospective study evaluating the efficacy and safety of IVT administered within 4.5 hours of symptom onset. Consecutive acute ischemic stroke (AIS) patients with minor-to-mild stroke symptoms (NIHSS score≤5) were enrolled from our hospital between August 2016 and May 2018. Ethical approval for this study was obtained from the First Affiliated Hospital of Soochow University Institutional Review Board (No.

Patients Selection
Eligible cases were collected using the following inclusion criteria: (1) Clinical final diagnosis of AIS with baseline NIHSS score 0 to 5; (2) Age 18 years or older; (3) Time from symptom onset within 4.5h for rt-PA group, however, it could be extended up to 24 hours after symptom onset in non-rt-PA group; (4) Available for a telephone interview at 90d.
The exclusion criteria were as follows: (1) Clinical final diagnosis of transient ischemic stroke (TIA); (2) Pre-stroke mRS score of ≥2; (3) ICH on baseline computed tomography, and other contraindications to IVT and antithrombotic therapy.

Patient Management
All patients were given appropriate standardized treatment in accordance with the guidelines [ 12 ]. Participants were divided into two groups according to different approaches about the management of mild stroke patients: (1) rt-PA group: Intravenous rt-PA within 4.5 hours of AIS onset or last known well time (0.9 mg/kg, maximum dose 90 mg with initial 10% of total dose given as bolus during 1 minute), followed by appropriate secondary stroke prevention strategies; (2) Non-rt-PA group: Optimal secondary stroke prevention strategies within 24h from AIS onset: including initiation of dual antiplatelets among patients with NIHSS≤3 (clopidogrel 75 mg per day for 90d + aspirin 100 mg per day for the first 3 weeks); clopidogrel 75mg/ aspirin 100mg per day; or anticoagulation agents in the event of cardioembolism, and other measures such as statins and antihypertensive medication, etc.

Patient Data Collection
Baseline demographic and clinical information were collected by experienced stroke neurologists: including age, gender, previous history (hypertension, diabetes mellitus, hyperlipoidemia, atrial fibrillation, coronary heart disease, previous stroke or TIA and current smoking), medication history, blood pressure at baseline, laboratory data, time from symptom onset, stroke severity (assessed by NIHSS score at admission and 7d), ABCD 2 score (chosen due to easiest-to-manipulate in emergency room), disabled stroke [ 14 ], and stroke subtype (assessed using TOAST classification [Trial of Org 10172 in Acute Stroke Treatment]) [ 17 ]. Neurological imaging information was collected at baseline, 24-36h, 7d or discharge if sooner in stroke unit, including infarct location, severity of responsible artery stenosis, and intracranial hemorrhage transformation.

Clinical Assessment and Definition of Functional Outcomes
The follow-up was conducted by two trained neurological doctors who were blinded to the baseline information via telephone or face-to-face interviews. (1) Efficacy outcomes: The achievement of an mRS of 0 to 1 at 90d was primary outcome defined as excellent outcome, and an mRS of 0 as perfect outcome. Significant improvement was referred as complete resolution of the neurologic deficit or an improvement of at least 4 points over baseline NIHSS score [ 18 ]. (2) Adverse outcomes: Early neurological deterioration (END) was defied as a NIHSS score increase of 2 or more within 7d after symptom onset excluding any CT-or MRI-documented ICH [ 19 ]. sICH was defined as CT-based ICH within 7d with a NIHSS score increase of at least 4 or death [ 20 ]. The recurrent ischemic stroke within 3 months was assessed by stroke specialists. outcomes, multivariate logistic regression analysis was used to determine the independent predictors for all prognosis, especially the relationship between intravenous rt-PA treatment and clinical outcomes. Moreover, receiver operator characteristic curves (ROCs) were performed to estimate the best cut-off values, sensitivity and specificity of baseline NIHSS score and ABCD 2 score for predicting adverse outcomes. According to the best cut-off values, we intended to estimate the heterogeneity of therapeutic effect magnitude for excellent outcome stratified by baseline NIHSS score (≤3 versus >3), stroke subtype assessed by TOAST classification, disabling neurologic deficits (yes versus no), and baseline ABCD 2 score ( 5 versus ≥5).
The α-level of significance was P<0.05 (bilateral). All analyses were performed using the SPSS software version 25.0.

Population and Baseline Characteristics
In the First Affiliated Hospital of Soochow University, a total of 3321 patients with a final diagnosis of AIS were derived from the Electronic Patient Record system between August 2016 and May 2018. And 808 patients with NIHSS score ≤5 were extracted from the cohort. Among them, 584 were excluded on the basis of inclusion and exclusion criteria, resulting in 224 participants enrolled in the study. According to different therapeutic approaches, 106 patients were included in the rt-PA group and 118 patients in the non-rt-PA group.
In the rt-PA group, patients had a significantly higher proportion of atrial fibrillation (18.9% vs 7.6%, P = 0.012), while the proportion of hyperlipidemia was significantly lower than that in patients not receiving rt-PA treatment (35.8% vs 54.2%, P = 0.006). In addition, there was no significant difference in other aspects summarized in Table 1 (P > 0.05), indicating that the two groups were generally balanced and comparable.

Univariate Analysis of Functional Outcomes at 90 Days
Among the enrolled 224 patients with acute mild stroke, excellent outcome at 90d was obtained in 160 (71.4%) patients and poor outcome in 64 (28.6%) patients. In univariate analysis, patients with excellent outcome at 90d had a significantly higher percentage of receiving rt-PA treatment than those with poor outcome (51.9% vs 35.9%, P=0.031). Also, female, higher baseline NIHSS score and ABCD 2 score, history of diabetes mellitus and antidiabetic agents, higher level of FPG and TC, stroke subtype, and severity of responsible artery stenosis were associated with poor outcome (Table 2).

Multivariate Analysis Between Covariates and Different Clinical Outcomes
After adjusting for confounders in multivariable logistic regression analysis, intravenous rt-PA treatment was independently corelated with both excellent and perfect outcome at 90d, also with significant improvement at 7d. But we didn't detect any independent association between intravenous rt-PA treatment and adverse outcomes showed in Figure   1. Table 3  negatively associated with perfect outcome. We detected significant improvement at 7d in 30 (28.3%) patients with rt-PA treatment versus 16 (13.6%) patients without rt-PA treatment, which was independently related with male and worse stroke severity (OR, 0.740; 95%CI, 0.558-0.983; P=0.038). The rate of END within 7d was 10.2% in patients untreated with rt-PA, whereas 13.2% in patients treated with rt-PA, with no statistically significant difference between two groups (P=0.472). It is worthy noting that higher ABCD 2 score was independently associated with END at 7d (OR, 2.293; 95%CI, 1.396-3.766; P=0.001). Any ICH within 7d, occurred in 7 (6.6%) patients treated with rt-PA compared with 2 (1.7%) patients without rt-PA treatment, with no significant association with intravenous thrombolysis (P=0.359). Among them, only 1 (0.9%) was sICH in rt-PA group. Older age, high level of systolic blood pressure at admission and worse stroke severity were independently corelated with ICH. While the recurrent ischemic stroke in 90d was less frequent in patients treated with rt-PA (2.8%) than the other group (5.9%).
Only male and both circulation infarction were strongly related with incidence of recurrent ischemic stroke.

Association Between Subgroups and Excellent Outcome
According to the best cut-off values of independent predictors, we carried out multivariate analysis or excellent outcome in subgroups. It was showed that patients treated with rt-PA could still obtain a higher 90-day excellent outcome rate than that without rt-PA, even if NIHSS score ≤ 3 (OR, 3.603; 95%CI, 1.249-10.392; P=0.018). In patients of ABCD 2 ≥5, intravenous rt-PA treatment was independently associated with higher proportion of excellent outcome at 90d, with 57 (72.2%) patients receiving rt-PA versus 58 (62.4%) not receiving rt-PA (OR, 2.768; 95%CI, 1.196-6.406; P=0.017). For patients with ABCD 2 score 5, significance of treatment effect was not observed between two groups. Among patients with stroke subtype of large artery atherosclerosis, the excellent outcome rate at 90d in rt-PA group (63.9%) was significantly higher than that in the other group (52.6%) with an adjusted OR of 5.616 (95%CI, 1.080-29.210; P=0.040). Besides, in patients with disabling stroke, we didn't detect independent association between rt-PA treatment and excellent outcome at 90d. (Figure 2).

Discussion
Our study implied a high proportion of 78.3% acute mild ischemic stroke (NIHSS score≤5) patients receiving intravenous rt-PA within 4.5 hours achieved excellent outcome (mRS 0-1) at 90d, and 28.3% of such patients had significant improvement at 7d. The 90d excellent outcome rate reached 90.3% even if in patients with NIHSS score ≤3. And the encouraging results were both detected in the 0-3h (83.9%) and 3-4.5h (73.1%) time windows. The above proportion was significantly higher than that of the patients without IVT, among them only 1 case developed sICH in patients with intravenous rt-PA treatment, indicating IVT being relatively effective and safe for patients with mild neurological deficits. Notably, we found that patients with mild stroke whose ABCD 2 score ≥5 and stroke subtype of large artery atherosclerosis could benefit more from IVT. As for the safety of intravenous rt-PA in mild stroke, we detected 7 (6.6%) receiving intravenous rt-PA patients occurred ICH within 7d compared with the lower rate of 1.7% in non-rt-PA group, yet among these only 1 (0.9%) developed sICH from rt-PA group, lower than the proportion of 1.8%-4.1% previous studies had reported [ 14 , 21 , 22 ]. The significant association between post-IVT ICH with older age, worse stroke severity and higher admission glucose level was reported in a systematic review from 55 studies [ 23 ], similar with our study data. The incidence of sICH in a large single-center cohort [ 24 ] stratified by baseline NIHSS score (≤6 versus 6) was 2.0% and 8.1% (P 0.001), respectively, while the difference was not detected with increasing NIHSS score within the range of 0 to 5 (P=0.51) [ 21 ], indicating that intravenous rt-PA is relatively safe but not risk-free in mild stroke patients.
Previous studies suggested that TOAST classification [ 13 ], disability [ 14 ], abnormal ischemic perfusion [ 25 ] and penumbra [ 26 ] can be used to screen high-risk patients with mild stroke. Currently, the preliminary results of Potential of rt-PA for Ischemic Strokes with Mild Symptoms (PRISMS) [ 14 ], a double-blind, multicenter, randomized controlled trial, revealed that likelihood of excellent outcome at 90d didn't increase among mild nondisabling AIS patients receiving IVT treatment, yet a higher risk of sICH. However, the early termination with only 313 participants enrolled might preclude the definitive conclusions. While our study detected significant difference for excellent outcome in patients with nondisabling deficits. The disabling deficit is commonly defined as follows: complete hemianopia (NIHSS-3≥2), severe aphasia (NIHSS-9≥2), neglect NIHSS-11≥1), limb weakness that cannot resist gravity (NIHSS-5/6≥2), functional impairment with NIHSS>5, or any potential disabling deficit judged by experienced physician [14]. Nevertheless, the NIHSS scale may not be adequate to evaluate severity of mild stroke, especially in posterior circulation AIS Our study demonstrated that acute mild stroke with ABCD 2 score ≥5, a fresh perspective to screen appropriate patients, might potentially benefit more from intravenous rt-PA treatment. ABCD 2 score was intended to aid clinical management and estimate the risk of stroke recurrence in patients with TIA and minor stroke [ 15 ]. In addition, ABCD 3 -I score (range 0-13, addition of dual TIA within 7d, ipsilateral carotid artery stenosis≥50% and positive brain imaging) showed better validation for prediction of early and 90-day stroke recurrent risk, with clinical presentation (C), symptom duration (D) and cerebral/carotid imaging (I) being the most essential components [ 16 , 28 , 29 ]. Recent trial revealed the ABCD 2 score of 4 and more in TIA or minor stroke was significantly associated with longer-term risk of another stroke [ 6 ]. Also, these scores were used to select patients for intensive therapy[ The non-randomized design due to retrospective study is the most notable limitation of our research. Although, the baseline characteristics between two groups were matched generally, with significant difference only in history of atrial fibrillation and hyperlipemia, the well-known limitations of observational study still not be avoided. And the small sample size of 224 enrolled subjects in this single-center study limited the power to assess the efficacy and safety of IVT among mild stroke patients, resulting in the inadequacy to generalize the results of our research to nation-wide population with mild stroke. So, well-designed multicenter randomized clinical trials are needed urgently.

Conclusions
A substantial proportion of patients deemed mild stroke (NIHSS≤5) with intravenous rt-PA therapy has excellent outcome both at 7-and 90-days, especially among patients with 5 or more of baseline ABCD 2 score and stroke subtype of large artery atherosclerosis. While, accurate identification of high-risk stroke is so essential to screen patients with mild deficiency for IVT, that further researches are warranted. In conclusion, given the significant percentage of excellent outcome and low risk of sICH, our results reassure the efficacy and safety of IVT therapy for acute mild ischemic stroke patients.

Consent for publication
Not applicable.

Availability of data and material
The datasets used and analyzed during the current study are available from the corresponding author on reasonable request.

Competing interests
The authors declare that they have no competing interests.  Tables   Table 1 Baseline      Subgroup Analysis with Acute Mild Stroke for Excellent Outcome. §Multivariable binary logistic regression analysis was used. §Multivariable binary logistic regression analysis was used after adjusting for intravenous rt-PA treatment and variables with P <0.1 in univariate analysis for different outcomes.
Abbreviations: ABCD2 score (age, blood pressure, clinical features, duration of symptoms, and presence of diabetes mellitus; range, 0 to 7, with higher scores indicating a higher risk of stroke); CI, confidence interval; NIHSS, National Institutes of Health Stroke Scale; OR, odds ratio; and rt-PA, recombinant tissue plasminogen activator.

Figure 3
Subgroup Analysis with Mild Stroke for Excellent Outcome. Abbreviations: ABCD2 score (age, blood pressure, clinical features, duration of symptoms, and presence of diabetes mellitus; range, 0 to 7, with higher scores indicating a higher risk of stroke); CI, confidence interval; and OR, odds ratio.