Intravenous thrombolysis using alteplase (0.9 mg/kg) for acute ischemic stroke (AIS) within 4.5 hours of onset has been proven effective (1, 2). Despite the solid evidence on the efficacy of alteplase, a higher risk of symptomatic intracranial hemorrhage (sICH) after thrombolysis still cannot be ignored. Hence, some Japanese trials compared two alteplase dosages of low dose (0.6 mg/kg) and standard dose (0.9 mg/kg) in AIS patients and indicated that a low dose of alteplase might be more suitable for Asian people (3, 4). Furthermore, the low-dose alteplase treatment has been strongly recommended for Japanese AIS patients within 4.5 hours by Japan Stroke Society Guidelines (5).

However, some studies in China suggested that the standard dosage might have better clinical outcomes than the low-dose alteplase without increasing the incidence of sICH (6, 7). The ENhanced Control of Hypertension And Thrombolysis strokE study (ENCHANTED) trial, as the only randomized controlled trial (RCT), has reported a lower incidence of sICH in the low-dose alteplase group but not achieved non-inferiority of low-dose to standard-dose alteplase for death or disability at 90 days (8). In our previous pool-data analysis, the low dosage was still considerable based on several regional databases (9).

Due to the high cost of a new RCT and the narrow gap expected to be found between dosages, we tried to investigate whether low-dose alteplase can bring better safety and a comparable functional outcome than standard dosage in Chinese patients with AIS using our real-world database.

From Jan 2019 to Dec 2020, a total of 4995 patients received thrombolysis with alteplase. After excluding the cases with missing data (ie, the un-noted dosage of alteplase, treatment extended to the 4.5-hours window, absence of the mRS score or the NIHSS score at discharge), 3,179 patients were included in the study. The flowchart was shown in Figure 1. Baseline characteristics were shown in Supplementary Table 1. The patients in the low-dose group were older than those in the standard-dose group. Also, the low-dose group had a higher proportion of patients with a history of stroke, AF, or disability. After adjustment of PSM, 1,334 AIS patients were included in the final analysis. Among them, 368 (27.8%) patients received low-dose alteplase. The median age of the patients was 71 years, and 38.8% were female. The baseline characteristics in both groups after PSM were presented in Table 1. After PSM, the baseline characteristics were balanced except for age and TOAST classification; the absolute standardized differences were basically within an acceptable margin of 0.1 (Supplementary Table 2).

The primary and secondary outcomes were summarized in Table 2. The low-dose group showed its association with a higher rate of mortality or disability (adjusted odds ratio (aOR) = 1.49, 95%CI [1.12, 1.98], p = 0.006) than the standard-dose group. Treatment with low-dose alteplase also resulted in a lower rate of mRS score 0-2 than that with standard-dose alteplase (aOR = 0.71, 95% CI [0.52, 0.97], p = 0.03). The results were similar to the univariate regression model. The distribution of mRS scores at discharge was shown in Figure 2. The common OR across the mRS scores between the low- and standard-dose alteplase was 1.36 (95% CI [1.10, 1.69]) in the univariate analysis, but without statistical significance after adjustment of baseline characteristics. In addition, no significant difference was found in the risk of in-hospital mortality or sICH for two groups. The sensitivity analysis for a 1:1 matched design showed generally similar results (Supplementary Table 3). Other serious adverse events during the hospitalization according to assigned treatment were presented in Table 3. There was a trend toward a higher incidence of pulmonary embolism in the low-dose alteplase group.

We performed subgroup analyses by age, gender, stroke subtypes, stroke severity, time from onset to treatment, and premorbid mRS score (Figure 3). All characteristics did not modify the treatment effect, except for stroke severity (p = 0.02). The patients with baseline NIHSS score ≤ 10 benefited more from the standard-dose alteplase therapy (OR = 1.56, 95%CI [1.16, 2.11], p = 0.003).

Our study demonstrated that patients receiving low-dose alteplase were associated with a higher risk of death or disability than those receiving standard-dose alteplase. Moreover, patients treated with low-dose alteplase were not associated with a lower sICH rate, and tend to have a higher rate of pulmonary embolism. Mild and moderate stroke patients with a baseline NIHSS score ≤ 10 may have better outcomes when treated with standard-dose alteplase.

Despite that a dose of 0.9 mg/kg alteplase is widely recommended by most guidelines for AIS patients (12–14), the high risks of sICH after thrombolysis with alteplase cannot be ignored, particularly among Asians (15). The J-ACT trial showed that the rate of mRS 0-1 at 90 days and the incidence of sICH in Japanese AIS patients receiving 0.6 mg/kg alteplase were comparable to 0.9 mg/kg alteplase treatment of the NINDS trial (1, 4). The results of several non-randomized trials in Japan subsequently showed the similar efficacy and safety outcome of low dosage thrombolysis with alteplase when compared with the standard dosage (3, 16, 17). In 2016, the ENCHANTED trial, as the only RCT involving 3310 AIS patients (63% Asian), aimed to compare low-dose with standard-dose alteplase. Although the trial failed to achieve the non-inferiority of low-dose as compared to standard-dose alteplase concerning the outcome of death or disability at 90 days (8), it also did not prove the superiority of standard-dose alteplase. There were significantly fewer sICH and trends toward the lower mortality rate with low-dose alteplase. Therefore, low-dose alteplase performed well in the safety outcome, which provided an alternative approach for both doctors and patients to make the personalized decision for the individual patient with AIS.

Compared with the ENCHANTED trial, our study achieved a similar efficacy outcome. What is new in our study, clinical physicians were more likely to administrate low dosage due to elder age and combined with AF. Our study found no association between alteplase dosage and the rate of sICH or mortality, which was different from the ENCHANTED trial. However, our findings were consistent with the results of TIMS-China (6, 7). An Asian stroke registry study also found no association between alteplase dosage and sICH risk (9). Our results might be the reflection of the real-world or false-negative error, attributed to the confounding bias of the observational study.

In subgroup analyses, we found that patients with mild-moderate stroke (NIHSS ≤ 10) in the standard-dose group were more likely to show a favorable outcome. However, the significant heterogeneity of stroke severity might be caused by a small sample of patients with a baseline NIHSS > 10. Our previous study showed that in mild stroke patients (NIHSS ≤ 4), there was no difference in the sICH risk and clinical improvement between both dosages (18). Therefore, more studies on the effectiveness and safety of low-dose thrombolysis for patients with different stroke severity are still needed.

The ENCHANTED trial raised the concern of the high risk of sICH in standard-dose alteplase treatment. Considering the bleeding-prone constitution of the Asian population, a secondary analysis of the ENCHANTED trial showed similar efficacy and safety outcomes of both doses of alteplase in both Asians and non-Asians (19). Also, the clinical use of low-dose alteplase in AIS patients with older age and other high-bleeding risks remained uncertain. Our subgroup study suggested that low-dose alteplase might be related to unfavorable functional outcomes even in patients older than 80.

Our study had several limitations. Firstly, our research population was limited to the Shanghai metropolitan. The management of risk factors and other demographic characteristics may not extend to that of other regions in China. Secondly, it was an observational retrospective study based on a registry database, the confounding bias cannot be ignored even with PSM. Thirdly, the relatively small sample size of the low-dose group might have impacted the statistical results. Fourthly, our study lacked mortality and functional outcomes in the long-term follow-up. Considering these limitations, further well-designed studies for patients with specialized characteristics are needed to provide more reference to medical practice.

Our observational study indicated that Chinese AIS patients receiving low-dose alteplase might be associated with an unfavorable functional outcome without lowering the risk of sICH. Our findings from the real-world dataset might provide more evidence to support standard-dose alteplase regimen for AIS patients in clinical practice.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

This study was approved by the Institutional Review Board in Huashan Hospital, with the waiver of consent as no identifying information was collected.

JX, XC, YX, YW, and SC performed the analysis. JX, XC, and YX drafted the manuscript. QD, YD, and KF revised the manuscript. YD and KF concepted this study, supervised the analysis, and finalized the manuscript. All authors contributed to the article and approved the submitted version.