Outcomes in minor stroke patients treated with intravenous thrombolysis

Abstract Aims Our study aimed to describe the short‐, medium‐, and long‐term outcomes of intravenous thrombolysis in minor stroke, and to explore the relationship between thrombolysis and clinical outcomes. Methods Our study included ischemic minor stroke patients (National Institutes of Health Stroke Scale score ≤ 5) within 4.5 h from symptom onset from the Third China National Stroke Registry (CNSR‐III) between August 2015 and March 2018. The primary outcome was a favorable functional outcome, defined as a modified Rankin Scale (mRS) score of 0–1 at 3 months. The secondary outcomes included mRS score of 0–1 at discharge, 6 months, and 1 year. The safety outcomes were symptomatic intracerebral hemorrhage (sICH) at 24–36 h and all‐cause mortality. The association between intravenous thrombolysis and clinical outcomes was studied using multivariable models. Results A total of 1905 minor ischemic stroke patients were included. Overall 527 patients (28%) received intravenous t‐PA (IV t‐PA) and 1378 patients (72%) in the non‐IV t‐PA group. Of them, 18.85% (359/1905) participants had a disabled outcome (defined as mRS score ≥ 2) at discharge, 12.8% (242/1885) at 3 months, 13.9% (262/1886) at 6 months, and 13.9% (260/1871) at 1 year. In multivariable analysis, IV t‐PA was associated with favorable functional outcomes at discharge (adjusted odds ratio [aOR] 1.49; 95% confidence interval [CI] 1.13–1.96; p = 0.004), 3 months (aOR 1.51; 95% CI 1.09–2.10; p = 0.01), 6 months (aOR 1.64; 95% CI 1.19–2.27; p = 0.003), and 1 year (aOR 1.52; 95% CI 1.10–2.10; p = 0.01). Symptomatic ICH occurred in 3 (0.6%) patients in IV t‐PA versus 2 (0.1%) in the non‐IV t‐PA group. No significant differences were found in all‐cause mortality between the two groups. Conclusions Intravenous t‐PA may be safe and effective in minor stroke (NIHSS ≤ 5) within a 4.5‐h window and further randomized controlled trials are warranted.


| INTRODUC TI ON
More than 50% of strokes present with minor stroke (NIHSS ≤ 5) on admission. 1,2 And about 30% of these patients were unable to ambulate independently at discharge and had a disabled outcome at 90 days. 3,4 Intravenous alteplase was recommended for minor disabling stroke within 4.5 h window according to the latest guidelines. 5 Nevertheless, the subjective nature of minor disabling stroke and potential hemorrhagic risk contributed to the discrepancy in clinical practice. 6,7 Less than half of the patients with minor stroke received intravenous thrombolysis treatment from American cohorts. 3,4 And the rate of thrombolysis in these populations was even lower in developing and undeveloped countries, including China. 8 To date, there has been limited evidence of thrombolysis for acute minor ischemic stroke. A meta-analysis of 6756 patients from contemporary completed randomized phase 3 trials of thrombolysis showed the benefit of intravenous alteplase for minor stroke patients of an NIHSS score of 0-4. 9 In addition, previous registries reported intravenous thrombolysis was associated with short-term outcomes including discharge to home, ambulation independently or modified Rankin score (mRS) at discharge. 3,8,10 Other observational studies found intravenous thrombolysis was associated with 90-day favorable outcomes in the analysis of relative higher baseline NIHSS. 2,4,11 However, the Potential of rt-PA for Ischemic Strokes With Mild Symptoms (PRISMS) trial aimed to evaluate the efficacy and safety of alteplase with non-disabling minor ischemic stroke (NIHSS ≤ 5) and revealed that alteplase was not superior to aspirin for a 90-day favorable functional outcome. Instead, alteplase increased the three-fold risk of symptomatic intracerebral hemorrhage (sICH). 7 Nevertheless, this trial was terminated early and did not provide strong evidence. So, whether thrombolysis is necessary for all minor stroke is inconsistent. Moreover, data on the association between thrombolysis and medium-and long-term functional outcomes in Asian patients with minor stroke are lacking.
Our study aimed to analyze the short-, medium-, and long-term outcomes of minor stroke with intravenous thrombolysis utilizing a large-scale prospective registry in China, and identify the association between intravenous alteplase and clinical outcomes.

| ME THODS
The CNSR-III was a nationwide prospective registry that included The study was approved by institutional ethics committees and written informed consent was obtained. The detailed design and main description of the CNSR-III have been published previously. 12

| Study population
We included the eligible patients fulfilling the following inclusion criteria: (1) minor ischemic stroke, which was defined by NIHSS ≤ 5; (2) time window: defined by presenting with stroke symptoms within 4.5 h of symptom onset or within 4.5 h of awakening after the point when last seen well. The exclusion criteria were (1) admission diagnosis of TIA or with complete symptom relief on initial evaluation; (2)

| Data collection and definitions
Patients were divided into IV t-PA and non-IV t-PA groups based on whether intravenous t-PA (0.9 mg/kg to a maximum dose of 90 mg, with 10% as an initial bolus and the remaining over one-hour intravenous infusion) administered within the 4.  13 We obtained imaging variables of symptomatic extra-and intracranial arterial stenosis (sEICAS), which was defined by 50%-99% stenosis or occlusion of any extra-and intracranial artery accounting for the acute infarction lesion and neurological symptoms. The detailed definitions of sEICAS in CNSR-III were described previously. 14

| Outcomes
The primary outcome of this study was mRS score of 0-1 at 3 months. The secondary outcomes included mRS score of 0-1 at discharge, 6 months, and 1 year. Outcomes were obtained through face-to-face interviews at discharge and 3 months and telephone by well-trained research coordinators at 6 months and 1 year. Regarding the safety outcome, we recorded all-cause mortality, intracerebral hemorrhage (ICH) which was detected on brain imaging at 24-36 h, and symptomatic intracerebral hemorrhage (sICH) defined as clinical deterioration with an increase in the NIHSS of at least four points or death attributed to hemorrhage on brain imaging, according to the European Cooperative Acute Stroke Study III (ECASS III) criteria. 15

| Statistical analysis
The data were tested for normal distribution using the Kolmogorov-Smirnov test. Continuous variables with normal distribution were described as mean ± standard deviation (SD) or as median with interquartile range (IQR) otherwise, and differences were assessed using the t-test if normally distributed or Mann-Whitney U test. Categorical variables were expressed as frequencies with percentages, and the χ 2 analysis or Fisher exact test was performed to compare the difference between groups. Three different multivariable-adjusted logistic regression models were used to balance the influence of covariates on the associations between IV t-PA and dichotomous outcomes. Model 1 adjusted age, sex, ethnicity, and baseline NIHSS.

| Baseline characteristics
The baseline characteristics of the 1905 patients were described in Table 1

| Discharge outcomes
The medium length of stay was 12 days. The change in NIHSS from baseline to discharge was shown in Figure S2. Improvement, worsening, and no change in NIHSS occurred in 75.71% (399/527), 5.50% (29/527), and 18.79% (99/527) patients, respectively, in the IV t-PA group and 51.7% (712/1377), 12.9% (178/1377), 35.37% (487/1377) patients, respectively, in the non-IV t-PA group before PSM. These factors were balanced after PSM. There was a significant difference in improvement in NIHSS at discharge by treatment status. Table 2  In the aforementioned multivariable model plus imaging marker of sEICAS, a similar association of IV t-PA with improvement in NIHSS was observed. All covariates were well-balanced after the PSM (Table S1). In a propensity score analysis, the association of IV t-PA with improvement in NIHSS was also identified (OR, 1.87; 95% CI, 1.40-2.51; p < 0.001).
At discharge, 81.15% (1546/1905) of participants had a favorable functional outcome (defined as mRS score ≤ 1). After adjustment for age, sex, ethnicity, and baseline NIHSS, IV t-PA was associated with favorable functional outcomes at discharge (aOR, 1.49; 95% CI, 1.13-1.96; p = 0.004). In following different multivariable models and propensity score analysis, a similar association of IV t-PA with the favorable functional outcome at discharge was observed (Table 2).
The associations of IV t-PA with 3-, 6-month, and 1-year functional outcomes were described in Table 3. There was no association between IV t-PA and functional outcomes compared with non-IV t-PA in univariate analysis. After adjusting for age, sex, ethnicity, and baseline NIHSS, IV t-PA was associated with 3-, 6-   Fasting glucose at admission, mmol/L 5.5 (4.9-6.8) 5.5 (4.9-6.8  antihypertensive, lipid-lowering agents, glucose-lowering agents, stroke unit, SBP at admission, and TOAST subtype, we found IV t-PA was with a robust independent predictor for 3-, 6-month, and

| Hemorrhagic events
Eleven (2.1%) patients developed ICH in IV t-PA group and 12 (0.9%) in non-IV t-PA group, of which 3 (0.6%) patients developed sICH in IV t-PA group versus 2 (0.1%) in non-IV t-PA group. No fatal event was observed due to sICH. There was no significant difference in sICH rates by treatment status (Table 4).

| All-cause mortality
The rates of all-cause mortality were no significant difference between IV t-PA group and non-IV t-PA group at short-, medium, and  Table 4).

| Subgroup analysis
Subgroup analysis was shown in Figure S3. In subgroup analysis restricted to NIHSS score of 0-2, after adjusting for age, sex,

| DISCUSS ION
Our study found that discharge, 3-, 6-month, and 1-year favorable functional outcomes occurred in more than 80% of minor stroke patients with NIHSS score of 0-5 within 4.5 h from the onset, accompanied by a low rate of sICH. Also, we identified IV t-PA was associated with favorable functional outcomes in the overall cohort.
Previous studies have indicated that minor strokes might be related to greater probabilities of impairment. The Get With The Guidelines Stroke (GWTG-S) registry which was an American quality improvement program including 5910 patients with a baseline NIHSS score of 0-5 treated with IV t-PA reported over 30% could not ambulate independently at discharge. 16 Another retrospective analysis from the GWTG-S registry described 25% of patients with an NIHSS score of 0-5 not treated with thrombolysis could not be released to ambulate independently or straight to home. 17 The Mild and Rapidly Improving Stroke Study (MaRISS) which was a prospective observational study including 1765 minor stroke patients with a baseline NIHSS score of 0-5 reported 21% of patients were unable to ambulate independently at discharge and 37% had a disabled outcome at 3 months. 4 Nevertheless, we found the rate of disabled outcome was less than 20% in the short-, medium-or longterm follow-up, which was lower than that in the abovementioned observational studies. It was in accordance with our previous study from the Chinese Stroke Center Alliance (CSCA), which was similar to the GWTG-S registry and included 6752 minor stroke patients (NIHSS ≤ 5) treated with IV t-PA reported only 10.1% could not ambulate independently at discharge. 8 The potential causes of the low disabled rate might encompass the longer length of stay, early intervention of rehabilitation, race or ethnicity, etc. 8,18 The proportion of thrombolysis in our study was 28%, which was similar to 25% alteplase administration in minor stroke reported in large population-based studies. 19 It meant more than 70% of patients with minor stroke did not receive thrombolysis, even if they presented within the 4.5-h window and had no contraindications for treatment. Potential reasons for this low thrombolysis rate might be TA B L E 3 Association of alteplase with 3-, 6-month, and 1-year mRS 0-1 in all subjects. related to low NIHSS score, being perceived as rapidly improving, and being afraid of intracerebral hemorrhage risk.
We also observed that IV t-PA was associated with short-, medium, and long-term favorable functional outcomes in the overall cohort compared with non-IV t-PA. Our findings of the association of IV t-PA and short-and medium-functional outcomes were in conformity with some observational studies. 3,11,20,21  with different study purposes showed no significant difference was detected between IV t-PA and favorable functional outcome at 3 months. In this study, minor stroke was defined by an NIHSS score of 0-3, which was different from our study. 24 In addition to demographic profiles, the underlying reasons for this discrepancy in results might include thrombus characteristics, early recanalization, hemispheric cerebral blood flow, etc. 25,26 Besides, so far, the association between IV t-PA and long-term functional outcome in minor stroke patients was seldom reported. Our study provided the rate of 1-year favorable functional outcome (86.1%) in minor stroke patients. Furthermore, the association of IV t-PA with functional outcome was still observed at 1 year and IV t-PA was associated with 1-year favorable functional outcome, compared with non-IV t-PA.
In the subgroup analyses, we found IV t-PA was statistically associated with 90-day mRS 0-1 in the baseline NIHSS of 3-5 group, while the association was borderline significant in the baseline NIHSS of 0-2 subgroup. Similarly, the MaRISS study found alteplase was associated with a favorable 3-month outcome of Stroke Impact Scale-16 in the prespecified subgroup of baseline NIHSS score of 3-5. 4 We noticed the percentage with an NIHSS score of 0-2 in our study was numerically higher than that in the MaRISS alteplase- A recent retrospective study showed mild neurologic deficits might mean less penumbra and lower hemispheric cerebral blood flow, and even given reperfusion, the benefit was limited, 25 which needed further validation. Also, in the NIHSS score of 0-2 subgroup, we observed the effect size of IV t-PA on functional outcome was enlarged from 3-, 6-month to 1-year follow-up, which was not obvious in the NIHSS score of 3-5 subgroup. The recent subsequent study from the MaRISS population exhibited 17% of patients improved on the mRS between 1-and 3-month follow-up term poststroke which meant a minor stroke might induce a greater possibility of longer-term functional improvement. 27 However, this has to be validated in future studies. Our study also found a favorable trend but no significant difference of IV t-PA on functional outcome in the restricted sample of sEICAS ≥ 50%, partly because of the limited sample size (n = 124).
In our study, although the overall rate of ICH within 36 hours was higher in patients with IV t-PA than non-IV t-PA group, the rate of sICH was comparable and no significant difference was observed between the two groups. The risk of sICH in our study was low and in line with the previous studies. 4,16,20 Furthermore, IV t-PA did not increase the rate of all-cause mortality.

| Limitations
Our study also has limitations. First, it has potential bias out of a cohort study, which was based on hospitals' voluntary and convenience in nature. However, the CNSR-III study covered 201 hospitals in 22 provinces and four municipalities and was representative of Chinese clinical practice. Second, we did not differentiate disabling from non-disabling minor strokes attributed to lacking these variables in the CNSR-III database. Third, the limited sample in the subgroups underpowered the association between IV t-PA and functional outcomes at different follow-up terms. Fourth, due to heterogeneity and inadequate drug compliance in clinical practice, our study is only an exploratory analysis based on the registry cohort and the results need to be verified in randomized controlled trials. The ongoing ARAMIS (Antiplatelet vs R-tPA for Acute Mild Ischemic Stroke) (NCT03661411), 28  trial will hopefully provide more robust evidence of thrombolysis in minor stroke.

| CON CLUS IONS
Our findings suggest minor stroke with NIHSS score of 0-5 treated with IV t-PA, as compared to non-IV t-PA, was associated with short-, medium-, and long-term favorable functional outcomes with infrequent sICH. Further large randomized controlled trials are warranted.

ACK N OWLED G M ENTS
We appreciate all the participating centers in the CNSR-III for their hard work on data collection.

CO N FLI C T O F I NTER E S T S TATEM ENT
The authors declare no conflict of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data supporting the study's conclusions are accessible from the respective authors upon reasonable request.