Efficacy and safety of Panax notoginseng saponins (Xuesaitong) for patients with acute ischemic stroke: a systematic review and meta-analysis of randomized controlled trials

Abstract

Background: Stroke is the major cause of mortality and permanent disability and is associated with an astonishing economic burden worldwide. In the past few decades, accumulated evidence has indicated that Xuesaitong (XST) has therapeutic benefits in cases of acute ischemic stroke (AIS). Our study aimed to provide the best current body of evidence of the efficacy and safety of XST for patients with AIS.

Methods: This is a systematic review and meta-analysis of randomized controlled trials (RCTs). We searched eight electronic databases from inception to 17 July 2023 for relevant RCTs. The investigators independently screened trials, extracted data, and assessed the risk of bias. A meta-analysis was conducted using RevMan 5.3 and STATA 16.0 software.

Results: In total, 46 RCTs involving 7,957 patients were included. The results showed that XST improved the long-term functional outcomes with lower modified Rankin Scale (mRS) scores (MD = −0.67; 95% CI [−0.92 to −0.42]; p < 0.00001) and a higher proportion of functional independence (mRS ≤2) (RR = 1.08; 95% CI [1.05 to 1.12]; p < 0.00001). Low-quality evidence indicated that XST improved the activities of daily living (MD = 10.17; 95% CI [7.28 to 13.06]; p < 0.00001), improved the neurological impairment (MD = −3.39; 95% CI [−3.94 to −2.84]; p < 0.00001), and enhanced the total efficiency rate (RR = 1.19; 95% CI [1.15 to 1.23]; p < 0.00001). No significant difference was found in the all-cause mortality or incidence of adverse events between the XST and control groups. The certainty of evidence was estimated as moderate to very low.

Conclusion: Presently, the administration of XST within 14 days of AIS is associated with favorable long-term functional outcomes. In addition, XST can improve activities of daily living, alleviate neurological deficits, and has shown good tolerability. However, the current evidence is too weak, and the confidence of evidence synthesis was restricted by the high risk of bias. Given the insufficient evidence, appropriately sized and powered RCTs investigating the efficacy and safety of XST for patients with AIS are warranted.

Systematic Review Registration:https://www.crd.york.ac.uk/PROSPERO/display_record.php?RecordID=446208, CRD42023446208.

1 Introduction

Stroke is the major cause of mortality and permanent disability worldwide and is associated with a high lifetime risk (Feigin et al., 2017; GBD 2019 Diseases and Injuries Collaborators, 2020). The high incidence and disability of stroke lead to an astonishing economic burden annually (Wu et al., 2019a; Rajsic et al., 2019). Ischemic stroke accounts for 69.6% of incident strokes and 77.8% of prevalent strokes and is regarded as the most common stroke subtype (Wang et al., 2017b).

In select patients with non-minor acute ischemic stroke (AIS), intravenous thrombolysis within 4.5 h and mechanical thrombectomy initiated within 24 h of symptom onset could salvage the ischemic penumbra and improve functional outcomes (Mendelson and Prabhakaran, 2021). Despite the clear benefit within a specified time window, stroke thrombolysis is highly time-critical and has been limited by the unknown time from symptom onset (Meretoja et al., 2014), high cost, and limited medical level. Otherwise, short-term dual antiplatelet therapy administered within 24 h of symptom onset could reduce the risk of stroke in patients with minor AIS (Wang et al., 2013). However, the side effects associated with antiplatelet agents, including damage to the liver and kidney, gastrointestinal injuries (Nema and Kato, 2010), and an increased risk of moderate to severe bleeding (Bhatia et al., 2021), should be taken into account; limitations, including aspirin resistance (AR) (Hankey and Eikelboom, 2006) and CYP2C19 genetic variants (Wang et al., 2016), restrict the clinical applications. Given the clinical dilemma, it is imperative to optimize stroke medication by developing and confirming safer and more effective therapies benefiting more patients with AIS.

Research on neuroprotective agents for AIS has been ongoing but has frequently failed to achieve the anticipated benefits in several clinical trials (Paul and Candelario-Jalil, 2021). Panax notoginseng (Burk.) F.H. Chen, also called Sanqi or Tianqi in Chinese, is an extremely valued herbal medicine in Asia. Panax notoginseng saponins (PNSs), the bioactive ingredients of P. notoginseng, consist of multiple active components and include five main bioactive ingredients accounting for 90% of the total PNSs: notoginsenoside R1, ginsenoside Rg1, ginsenoside Rb1, ginsenoside Rd, and ginsenoside Re (Qu et al., 2020). PNSs have been used for the clinical treatment of AIS since antiquity and have exerted obvious anti-inflammatory effects on atherosclerosis-related cardiac–cerebral vascular disease (Wan et al., 2009; Wang et al., 2011b). The pathophysiology of cerebral ischemic injury is correlated with a rapid cascade of energy failure, excitotoxicity, oxidative stress, nitrative stress, and inflammatory responses (Dirnagl et al., 1999; Chamorro et al., 2016). Neuroinflammation is considered a potential treatment target for such complex and dynamic processes (Jayaraj et al., 2019). PNS and notoginsenoside R1 exhibit versatile biological activities, including anti-inflammatory activity (Shi et al., 2017), antioxidant capacity (Zhang et al., 2019), alleviation of blood–brain barrier (BBB) disruption (Wu et al., 2019b), antiapoptosis (Chen et al., 2011), and endothelial cell protection (Hu et al., 2018). Xuesaitong (XST), one of the most commonly used medicinal products of PNS-related preparations for treating AIS, was licensed for ischemic stroke by the National Medical Products Administration in China in 1999. The experimental studies indicated that the neuroprotective mechanisms of XST included antioxidation (Zhou et al., 2014) and antiapoptosis (Li et al., 2009), and significant reduction was found in the infarct volume and neurologic impairment in mice models with middle cerebral artery occlusion when XST was administered during the acute phase of ischemic stroke (Li et al., 2019). The reported quality control (Yang et al., 2017) and the previous post-marketing surveillance study (He et al., 2020) provided some evidence of the effectiveness and safety of XST for clinical applications. Consequently, XST, composed of multiple active components, might produce clinically effective neuroprotection for the treatment of AIS.

In the past few decades, accumulated evidence has indicated that XST has therapeutic benefits in cases of AIS. Recent meta-analyses of randomized controlled trials (RCTs) (Zhang et al., 2015; Geng et al., 2022) have evaluated the efficacy of XST for patients with AIS; however, the findings were discordant and inconclusive. Previous meta-analyses analyzed the efficacy of XST oral preparation or XST injection. However, the safety of XST and whether XST improves long-term functional outcomes and reduces all-cause mortality remain unknown, which has probably led to inadequate comprehension of the clinical benefits of XST for patients with AIS. Moreover, it is worth mentioning that the latest multicenter, double-blind, placebo-controlled randomized clinical trial conducted by our team provided strong new evidence of XST efficacy and safety in patients with AIS (Wu et al., 2023). To summarize and provide the best current evidence regarding the benefits and harm of XST treatment for patients with AIS, we conducted this systematic review to examine the efficacy and safety of XST on AIS without differentiating dosage forms.

2 Materials and methods

We performed and reported this systematic review and meta-analysis based on the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) 2020 statement (Page et al., 2021). The protocol was already registered in the International Prospective Register of Systematic Reviews (PROSPERO) (CRD42023446208).

2.1 Search strategy

A comprehensive search was conducted to identify published studies of RCTs indexed in PubMed, Embase, the Cochrane Library, Web of Science, the Chinese National Knowledge Infrastructure (CNKI), the Chinese Science and Technology Journals Database (VIP), the Wanfang Database, and SinoMed without language limitations from their respective inception dates to 17 July 2023. The Medical Subject Heading (MeSH) terms and free-text keywords were utilized. We also searched the registered clinical trials, dissertations, and gray literature. In addition, a secondary manual search was conducted according to the references of the included articles. The detailed search strategies are presented in the Supplementary Material.

2.2 Eligibility criteria

2.2.1 Inclusion criteria

2.2.2 Exclusion criteria

RCTs with crossover and N-of-1 designs were excluded.

2.3 Study selection

After removing duplicate studies in all records retrieved, two reviewers (XS and ZC) screened the titles and abstracts independently, and three reviewers (YL, ZC, and CZ) independently selected the articles meeting eligibility criteria through full-text search. Disagreements were discussed, and a third author (YG) was contacted to arbitrate.

2.4 Data extraction

Reviewers, in pairs (YL and ZC, SC, and CZ), independently performed the data extraction using a preformulated data collection form as follows: 1) information from the included studies concerning the authors, publication year, and title; 2) patient characteristics, including the number of participating sites, sample sizes, age, sex, and onset time; 3) intervention details, including dosage form, dosage, frequency, duration, and combination treatment; and 4) outcomes.

2.5 Assessment of risk of bias

Two reviewers (TL and MQ) independently evaluated and cross-checked the risk of bias for eligible RCTs according to the Cochrane risk of bias tool 2.0. We evaluated five items as follows: “randomization process,” “deviations from intended interventions,” “missing outcome data,” “outcome measurements,” and “selective reporting.” Finally, each item was classified into “low risk of bias,” “some concerns,” and “high risk of bias.” Any disagreement was resolved by discussion and in consultation with a third author (CZ).

2.6 Data synthesis and analysis

Statistical analyses were conducted using RevMan 5.3 software and STATA 16.0. The results were expressed herein as the relative risk (RR) for dichotomous variables, whereas the mean difference (MD) was used for continuous data. The effect estimates were measured with a 95% confidence interval (CI), and p < 0.05 was considered to be statistically significant.

Statistical heterogeneity among studies was evaluated using the I-square (I²) statistic test. Data with I² ≤ 25% were defined as insignificant heterogeneity, and we selected a fixed-effects model. When the baseline characteristics were acceptable and statistical heterogeneity was comparable (I² > 25%), a random-effects model was adopted.

When I² > 25%, we conducted the sensitivity analyses, iteratively omitting each study one at a time. Furthermore, subgroup analyses were performed regarding the duration of treatment and dosage form. If the statistical heterogeneity could be successfully explained by the sensitivity analysis or the subgroup analysis (I² ≤ 25%), we applied a fixed-effects model. If not, a random-effects model was adopted. Considering that I² could be biased in small meta-analyses, we adopted a random-effects model for such analysis (von Hippel, 2015).

2.7 Publication bias

Potential publication bias was detected by visually inspecting the funnel plot symmetry, and we conducted Begg’s statistical tests for ≥20 included studies and Egger’s statistical tests for <20 included studies.

2.8 Quality of evidence

According to the Grading of Recommendations, Assessment, Development, and Evaluation (GRADE) (Balshem et al., 2011), two independent reviewers (XS and LF) evaluated the quality of the evidence derived from the meta-analysis result. We rated the evidence as “high,” “moderate,” “low,” or “very low.” Disagreements regarding upgrades or downgrades were resolved by a third reviewer (YG).

3 Results

3.1 Literature search

The electronic search identified 5,592 potentially relevant publications. Of these, 3,131 duplicates were removed, and 2,081 were excluded after screening the titles and abstracts. Of the remaining 380 articles that were subjected to a full-text review, 334 ineligible studies were excluded for the following reasons: unclear onset time (154 studies), non-RCTs (127 studies), non-target population (30 studies), inappropriate interventions (17 studies), unavailable full-text report (5 studies), and duplicate data (1 study). Ultimately, a total of 46 studies were eventually included in the quantitative analysis; the PRISMA flow diagram is shown in Figure 1.

FIGURE 1

3.2 Characteristics of the included studies

Overall, the 46 eligible RCTs were published between 2011 and 2023 and involved 7,957 participants, with 3,983 classified in the experimental groups and 3,974 in the control groups. The sample size ranged from 50 to 3,072, and most of the participants were middle-aged or elderly, with a mean age ranging from 54.5 to 72.5 years. In all, 45 RCTs were single-center trials, and 1 RCT was a multicenter trial. Regarding the dosage form, 43 studies used XST injections, whereas only 3 studies used XST oral preparations. We summarize the characteristics of the included studies in Table 1.

3.3 Risk of bias assessment

We identified the overall bias as “low risk of bias” in one study (Wu et al., 2023) and judged the remaining 45 studies to have a “high risk of bias”, indicating the poor quality of the selected RCTs. The results of the assessment of bias risk are presented in Figure 2 and Supplementary Figure S1.

FIGURE 2

We identified the “randomization process” as having a “high risk of bias” for the inappropriate methods of allocation concealment in two studies. On the contrary, we identified one study as having a “low risk of bias” because the allocation sequence was stored by researchers who were not involved in the observation. Regarding the risk of bias due to the “deviations from the intended interventions”, we rated 36 studies as having a “high risk of bias” because they did not report blinding and the per-protocol principle was used in analyses. Conversely, we judged one study as having a “low risk of bias” for its double-blind study design and appropriate analyses such as intention to treat. In addition, we judged all studies as having a “low risk of bias” in the case of the “missing outcome data”. Because no reported loss to follow-up was detected, or we found negligible losses to follow-up, the missing data were balanced between the experimental group and control group. We identified the “outcome measurements” as having a “high risk of bias” in 40 studies, considering that the total efficiency rate is the composite endpoint. In contrast, we rated one study as having a “low risk of bias” for its objective outcome and blinded outcome assessors. With regard to the risk of bias due to the “selective reporting”, one study was identified as having a “low risk of bias” for its transparent report of the observations planned, while 45 studies were rated as having “some concerns” for the lack of relevant reporting.

3.4 Efficacy outcomes

3.4.1 Long-term functional outcomes

For the long-term functional outcomes, two studies (Zhang et al., 2022; Wu et al., 2023) comprising a total of 3,158 participants reported the grading or the proportion as per the mRS, and we were unable to synthesize the data. Patients in the XST group were more likely to have better long-term functional outcomes with lower mRS scores (MD = −0.67; 95% CI [−0.92 to −0.42]; p < 0.00001) or a higher proportion of functional independence (mRS ≤2) (RR = 1.08; 95% CI [1.05 to 1.12]; p < 0.00001) (Figure 3).

FIGURE 3

3.5 Secondary outcomes

3.5.1 Reduction of all-cause mortality

Two studies (Zhang et al., 2022; Wu et al., 2023) containing 3,162 cases reported all-cause mortality, whereas there was no significant difference between the XST group and the control group (RR = 0.43; 95% CI [0.06 to 2.93]; p = 0.39; I² = 0%) (Figure 4). As such, no evidence was found to indicate that XST could reduce the all-cause mortality of AIS.

FIGURE 4

3.5.2 Improvement in activities of daily living

In all, 13 studies (Wang, 2015; Zhou et al., 2015; Li et al., 2016; Guo, 2017; Zhong and Deng, 2017; Wang and Ning, 2018; Zhang, 2018; Liu et al., 2019; Wang et al., 2021; Xu, 2021; Wang, 2022; Zhang et al., 2022; Wu et al., 2023) comprising 1,372 participants used the BI score; however, one of the studies (Wu et al., 2023) reported data on the BI score change from baseline to 90 days, and we were unable to synthesize this study. The pooled data of the other 12 studies clarified that XST improved the BI score (MD = 10.17; 95% CI [7.28 to 13.06]; p < 0.00001) (Figure 5). In view of the significant heterogeneity in the meta-analysis of the BI score (I² = 94%, p < 0.00001), a random-effects model was used. Further sensitivity analysis showed that statistical heterogeneity was not significantly reduced when we excluded a single study in sequence. We performed subgroup analyses by the duration of treatment (14 days, MD = 12.40; 95% CI [7.85 to 16.95]; p < 0.00001; 28 days, MD = 6.66; 95% CI [2.42 to 10.90]; p < 0.00001) (Supplementary Figure S2) and by the combination treatment (conventional treatment, MD = 12.17; 95% CI [8.51 to 15.84]; p < 0.00001; neuroprotective agents plus conventional treatment, MD = 5.36; 95% CI [3.23 to 7.50]; p < 0.00001) (Supplementary Figure S3).

FIGURE 5

3.5.3 Improvement in neurological impairment

Regarding the improvement in neurological impairment, 30 studies (Luo et al., 2011; Han et al., 2014; Li et al., 2015; Wang, 2015; Zhou et al., 2015; Cao et al., 2016; Li et al., 2016; Zhang, 2016; Zhu et al., 2016; Guo, 2017; Hu et al., 2017; Liu, 2017; Zhong and Deng, 2017; Wang and Ning, 2018; Xue, 2018; Ye et al., 2018; Zhang, 2018; Fan et al., 2019; Liu et al., 2019; Li and Liu, 2020; Wang et al., 2020; Zhang et al., 2020; Wang et al., 2021; Xu, 2021; Huang et al., 2022; Ping et al., 2022; Wang, 2022; Zhang et al., 2022; Zhao, 2022; Wu et al., 2023) containing 3,385 cases reported the grading according to the NIHSS score, and we excluded one study (Wu et al., 2023) that reported data on the NIHSS score change from baseline to 90 days. The outcome indicated that XST reduced the NIHSS score (MD = −3.39; 95% CI [−3.94 to −2.84]; p < 0.00001), and a random-effects model was applied due to the high heterogeneity (I² = 94%, p < 0.00001) (Figure 6). Sensitivity analysis indicated that the statistical heterogeneity was not significantly reduced through the sequential removal of any study (Supplementary Figure S4). Subgroup analyses were then performed, respectively, by the duration of treatment (14 days, MD = −3.42; 95% CI [−4.12 to −2.73]; p < 0.00001; 28 days, MD = −3.30; 95% CI [−4.19 to −2.42]; p < 0.00001) (Supplementary Figure S5), by the dosage form (XST injection, MD = −3.36; 95% CI [−3.91 to −2.80]; p < 0.00001; XST oral preparation, MD = −4.78; 95% CI [−7.25 to −2.31]; p = 0.0002) (Supplementary Figure S6), and by the combination treatment (conventional treatment, MD = −3.58; 95% CI [−4.30 to −2.86]; p < 0.00001; neuroprotective agents plus conventional treatment, MD = −3.10; 95% CI [−4.03 to −2.18]; p < 0.00001) (Supplementary Figure S7).

FIGURE 6

We also analyzed one study (Deng, 2018) containing 50 participants, the data of which showed that XST reduced the ESS score (MD = 11.85; 95% CI [2.07 to 21.63]; p = 0.02) (Supplementary Figure S8).

3.5.4 Total efficiency rate

A total of 40 studies (Fu et al., 2011; Luo et al., 2011; Han et al., 2014; Li et al., 2015; Wang, 2015; Zhou et al., 2015; Bu, 2016; Cao et al., 2016; Huang, 2016; Li and Chang, 2016; Li et al., 2016; Zhang, 2016; Zhu et al., 2016; Gao, 2017; Guo, 2017; Hu et al., 2017; Li, 2017; Liu, 2017; Wang and Ning, 2018; Xue, 2018; Zhang, 2018; Fan et al., 2019; Liu et al., 2019; Xiao et al., 2019; Zhang, 2019; Li and Liu, 2020; Wang et al., 2020; Wu, 2020; Xiao et al., 2020; Zhang et al., 2020; Wang et al., 2021; Xu, 2021; Yang, 2021; Huang et al., 2022; Ouyang and Zhang, 2022; Song et al., 2022; Wang, 2022; Zhang et al., 2022; Zhao, 2022; Yan et al., 2023) comprising 4,473 participants reported the total efficiency rate, and the pooled data showed that XST improved the total efficiency rate (RR = 1.19; 95% CI [1.15 to 1.23]; p < 0.00001) (Supplementary Figure S9). Considering that high heterogeneity (I² = 52%, p < 0.0001) could not be changed significantly through the sensitivity analysis (Supplementary Figure S10), we performed subgroup analyses, respectively, by the duration of treatment (7 days, RR = 1.17; 95% CI [1.02 to 1.34]; p = 0.03; 14 days, RR = 1.16; 95% CI [1.12 to 1.21]; p < 0.00001; 28 days, RR = 1.27; 95% CI [1.20 to 1.34]; p < 0.00001) (Supplementary Figure S11), by the dosage form (XST injection, RR = 1.18; 95% CI [1.14 to 1.22]; p < 0.00001; XST oral preparation, RR = 1.55; 95% CI [1.24 to 1.94]; p < 0.00001) (Supplementary Figure S12), by the combination treatment (conventional treatment, RR = 1.18; 95% CI [1.13 to 1.24]; p < 0.00001; neuroprotective agents plus conventional treatment, RR = 1.19; 95% CI [1.14 to 1.25]; p < 0.00001) (Supplementary Figure S13), and by the time of administration (treatment initiated within 72 h, RR = 1.19; 95% CI [1.15 to 1.23]; p < 0.00001; treatment initiated within 14 days [except for studies initiated within 72 h only], RR = 1.14; 95% CI [1.05 to 1.24]; p < 0.00001) (Supplementary Figure S14).

3.5.5 Blood rheology indicators

The meta-analysis results of XST on HBV, LBV, FIB, PV, and Hct are shown in Supplementary Table S1 and Supplementary Figure S15. The detailed contents are presented in Supplementary Material.

3.5.6 Adverse events

Of all studies, 15 studies (Luo et al., 2011; Zhu et al., 2016; Shen and Zhang, 2017; Xue, 2018; Liu et al., 2019; Zhang, 2019; Wang et al., 2020; Wu, 2020; Wang et al., 2021; Huang et al., 2022; Ping et al., 2022; Song et al., 2022; Zhao, 2022; Wu et al., 2023; Yan et al., 2023) comprising 4,288 cases reported adverse events. No heterogeneity was found (I² = 0%, p = 0.97); thus, a fixed-effects model was adopted. There was no significant difference between the XST group and the control group (RR = 0.97; 95% CI [0.70 to 1.35]; p = 0.85) (Figure 7). No increased rate of adverse events was observed in patients who received XST treatment. Among the 16 studies, a total of 135 participants reported detailed information on adverse events before the end of the follow-up. Nausea, dizziness, and skin irritation were the most frequently reported adverse events.

FIGURE 7

3.6 Additional data from the latest large-scale RCT

Our review demonstrated that XST might have clinical efficacy in the improvement of activities of daily living and neurological impairment. Additional data from the latest large-scale RCT (Wu et al., 2023) reported the NIHSS score change from baseline to 90 days [XST: −4 (IQR −5 to −3); placebo: −4 (IQR −5 to −3); p = 0.02] and the BI score change from baseline to 90 days [XST: 15 (IQR, 5–35); placebo: 15 (IQR, 5–30); p = 0.006]. The evidence was substantial, indicating that XST was more effective in enhancing neurologic deficits. Notably, this trial provided new evidence of the symptomatic intracranial hemorrhage (XST: 1/1,488 (0.1%); placebo: 0/1,482 (0); p = 0.32), indicating that XST may not increase the risk of bleeding.

3.7 Publication bias

The funnel plot (Supplementary Figure S16) and statistical test indicated that no obvious publication bias was found in included trials regarding the BI score (Egger’s test, p = 0.441), the NIHSS score (Begg’s test, p = 0.358), HBV (Egger’s test, p = 0.193), LBV (Egger’s test, p = 0.478), FIB (Egger’s test, p = 0.774), PV (Egger’s test, p = 0.460), Hct (Egger’s test, p = 0.179) levels, and adverse events (Egger’s test, p = 0.099). However, a publication bias risk was present for the total efficiency rate (Begg’s test, p = 0.000). The publication bias of the long-term functional outcomes, all-cause mortality, and ESS score could not be estimated for only one or two included trials.

3.8 Quality of evidence

The certainty of the evidence of XST on adverse events was rated as “moderate”; that on functional independence, all-cause mortality, the BI score, and the NIHSS score was “low”; and that on the mRS score, the ESS score, the total efficiency rate, and blood rheology indicators was “very low” (Table 2). We judged the quality of evidence as moderate to very low, mainly due to the high risk of bias and the serious inconsistency.

4 Discussion

4.1 Summary of the main results

This meta-analysis consisted of 46 RCTs on the efficacy and safety of XST for patients with AIS, including a total of 7,957 participants and two dosage forms. Regarding the long-term functional outcomes, most initial RCTs did not prespecify or report on long-term functional outcomes, and only two trials reported relevant outcomes that could not be synthesized. One of the RCTs reported the proportion of functional independence at 90 days [XST: 1,328/1,487 (89.3%); placebo: 1,218/1,479 (82.4%); OR, 1.95; p < 0.001], while the other low-evidence quality study reported the mean of the mRS score. Additionally, pharmacological studies have proven that XST can promote the polarization of microglia to an M2 phenotype, inhibit neuronal cell death via the downregulation of the STAT3 signaling pathway, reduce Nogo-A expression, and inhibit the ROCKII pathway, exerting long-term neuroprotective effects (Li et al., 2019; Zhou et al., 2021). Even though we could not synthesize the effect sizes of the two studies, we believe that XST is highly likely to have a superior therapeutic benefit in the long-term functional outcomes. For the mRS, the FDA accepted the dichotomous approach as the primary outcome measure for subsequent AIS trials since it was convenient for physicians and researchers and had the advantage of being translatable into a number needed to treat (Broderick et al., 2017). We suggest that researchers should conduct relevant RCTs with more rigorous and internationally recognized methodological designs for better evidence synthesis and clinical practice in the future.

As for secondary outcomes, this study did not indicate that XST could reduce all-cause mortality by pooling a few corresponding data, while low-certainty evidence of most studies revealed that XST enhanced the total efficiency rate. Compared with other outcome indices, authors of previous studies seemed to prefer to use the total efficiency rate instead of an objective outcome index such as all-cause mortality, and we hold a dialectical perspective. The total efficiency might provide an intuitional impression of the outcomes. However, standardized approaches are not generally accepted and validated for interpretation, and it is an inadequate strategy to evaluate a composite endpoint as if it were a single primary endpoint (McCoy, 2018). The pooling data might lead to error accumulation of the total efficiency rate, and we recommend that future studies should avoid such a subjective outcome index, as to date, little guidance exists on how to interpret the aggregated endpoints (Armstrong and Westerhout, 2017). Additionally, low-certainty evidence suggested that XST improved the BI score and reduced the NIHSS score. These estimates might be very imprecise, as high heterogeneity existed and did not decrease after the application of sensitivity analyses and subgroup analyses. Furthermore, the subgroup analyses showed that regardless of the type of XST dosage form used in the acute phase of ischemic stroke, XST might be an effective alternative therapy in the improvement of the activities of daily living and neurological impairment at different durations of treatment. Notably, the improvement in neurological impairment and activities of daily living seemed to be more obvious in the XST with conventional treatment group than XST with neuroprotective agents plus conventional treatment. In addition, we detected that treatment initiated within 72 h showed more effective results according to the subgroup analysis of the total efficiency rate.

The safety outcomes of XST in patients treated for AIS remained unknown according to the previous meta-analyses. Our meta-analytical evidence from RCTs revealed that there was no significant difference in safety outcomes. The XST group and the control group both showed good tolerability, and the reported adverse reactions might be relevant to the disease or other therapeutic procedures. A large-scale, population-based post-marketing study showed that the XST injection is well tolerated and has favorable safety, with a mean cumulative medication time of 7.53 ± 5.39 days (He et al., 2020). However, most of the included RCTs used XST injection with a duration of 14 days or even 28 days, while no increasing adverse events were found. Furthermore, only one study (Wu et al., 2023) reported bleeding events, which limited us to drawing the relevant conclusion. Indeed, we look forward to a more rigorous design and more transparent reporting so that we can clarify the application of different dosage forms and specify the dosage and duration. Additionally, the latest large-scale RCT showed XST did not increase the risk of safety events [XST: 15/1,488 (1.0%); placebo: 16/1,482 (1.1%); p = 0.85], and we expect more reliable trials of the safety of XST in the future to inform this field.

4.2 Comparison with previous studies

Compared with the two previous reviews regarding the effectiveness of XST, the present systematic review and meta-analysis included all dosage forms of XST and more recent RCTs, especially the latest large-scale RCT from our team (Wu et al., 2023). Previous low-quality trials might have overestimated the efficacy of XST. In addition, the previous systematic reviews were merely concerned with the total effective rate, the NIHSS score, the CSS score, and blood rheology indicators. However, we attempted to explore whether XST could improve long-term functional outcomes and reduce all-cause mortality, which are more objective and vital for patients with AIS. The comparisons of the studies mentioned previously are shown in Table 3. We made efforts to contact the authors and tried to obtain the generation of random sequences through e-mail and telephone. Ultimately, we excluded the articles in which “selection of participants” or “retrospective analysis” was mentioned in addition to “randomization” if the authors were unavailable to provide the generation of random sequences. We aimed to provide this field with a more comprehensive and specific evaluation of XST for patients with AIS.

The latest systematic review and meta-analysis (Geng et al., 2022) published in 2022 synthetically assessed the efficacy and safety of XST oral preparation, including eight published RCTs up to August 2021. However, these eight studies were excluded during our screening for the following reasons: probably not RCTs (Li et al., 2013; Wang et al., 2017a) (n = 2), non-target population (Li and Liang, 2002; Mi and Wang, 2009) (n = 2), unclear onset time (Liu et al., 2005; Lu, 2010; Chang and He, 2017) (n = 3), and unavailable full-text report (Lin, 2007) (n = 1). Among the 23 RCTs included in the meta-analysis published in 2015 (Zhang et al., 2015), only one RCT (Fu et al., 2011) overlapped with the 46 studies included in our study. We excluded the other 22 RCTs for the following reasons: probably not RCTs (Yuan, 2003; Li and Qin, 2006; Yuan and Jiang, 2006; Wang and Li, 2007; Zhang and Zhang, 2008; Duan and Ai, 2009; Wang et al., 2011a; Yang and Cheng, 2012; Song, 2013) (n = 9), wrong randomization (Zhao, 2006; Rong and Zhi, 2008; He et al., 2011) (n = 3), inappropriate intervention (He, 2006; Wang, 2006; Li and Jiang, 2007; Wang, 2007; Zi et al., 2008; Ma and Wang, 2009) (n = 6), non-RCT (Zhang and Zhang, 2003) (n = 1), and unavailable full-text report (Li et al., 1999; Li, 2003; Cai, 2011) (n = 3). Although we attempted to contact the authors during our procedure, the information was still unavailable.

4.3 Limitations

Our study has some potential limitations. We pooled the data of the NIHSS score, the BI score, the total efficiency rate, and blood rheology indicators on conditions of significant statistical heterogeneity being observed, which lowered the evidence grade. This is likely because acceptable clinical heterogeneity existed in several aspects of the included studies, such as age, sex, onset time, cointerventions, treatment duration, and follow-up period. Although sensitivity and subgroup analyses were performed, confounding statistical results caused by heterogeneity could not be completely excluded. In addition, only two studies reported long-term functional outcomes that could not be synthesized, and we expect new relevant trials to update the meta-analysis. In addition, we expected to evaluate the XST administration during the acute phase of ischemic stroke (within 14 days of onset), but most of the included studies involved participants within 72 h of onset. We found the early time of XST administration might be associated with a higher total efficiency rate, and we failed to draw more conclusions due to the lack of relevant data. Furthermore, almost all of the included studies were at “high risk of bias,” which limited the interpretation of the previous results and further clinical application. We will be monitoring large-scale RCTs of XST to update this systematic review and meta-analysis if any high-quality trial emerges. Although we conducted this review rigorously and systematically, the results should be interpreted with caution before being recommended for clinical practice.

4.4 Implications for future research

Well-designed and properly conducted RCTs provide the gold standard for producing primary evidence, and fully reporting trial outcomes is vital for result-replication and knowledge-synthesis efforts (Butcher et al., 2022). Poorly reported findings have affected the conclusions drawn from systematic reviews and meta-analyses (Mayo-Wilson et al., 2017). We suggest that future RCTs register protocols prospectively and report the prespecified outcomes rigorously according to the CONSORT-CHM Formulas 2017 (Cheng et al., 2017). Similar to this review, future studies should strictly apply and transparently report the allocation concealment mechanism and double-blind methods. In addition, researchers should take into consideration the most appropriate and scientific method of aggregation of the outcomes, devoting attention to subsequent evidence synthesis and informing evidence-based clinical decision-making. If researchers have to use the composite outcome, it is advisable to determine an acknowledged definition of the composite outcome and all individual components of the composite outcome. Furthermore, high-quality evidence of the effectiveness of XST in patients with AIS is still insufficient, and the efficacy and safety of XST for AIS with proper intervention and long-term follow-up should be investigated to provide more robust and objective evidence.

5 Conclusion

In conclusion, the present systematic review and meta-analysis of 46 RCTs reveals that the administration of XST within 14 days for AIS is associated with favorable long-term functional outcomes. Additionally, XST can improve activities of daily living, alleviate neurological deficits, and has good tolerability. Nevertheless, the current evidence is too weak and needs to be proven by further high-quality evidence. The positive effects have been restricted by the poor methodological quality and the high risk of bias, weakening the confidence in evidence synthesis. Considering that the current evidence is too weak and that XST is a promising agent against AIS, researchers should conduct RCTs with more rigorous methodological designs and more transparent reporting to provide more evidence with moderate to high certainty.

References

• 1

  ArmstrongP. W.WesterhoutC. M. (2017). Composite end points in clinical research: A time for reappraisal. Circulation135, 2299–2307. 10.1161/circulationaha.117.026229

  • CrossRef
  • Google Scholar

• 2

  BalshemH.HelfandM.SchünemannH. J.OxmanA. D.KunzR.BrozekJ.et al (2011). GRADE guidelines: 3. Rating the quality of evidence. J. Clin. Epidemiol.64, 401–406. 10.1016/j.jclinepi.2010.07.015

  • CrossRef
  • Google Scholar

• 3

  BhatiaK.JainV.AggarwalD.VaduganathanM.AroraS.HussainZ.et al (2021). Dual antiplatelet therapy versus aspirin in patients with stroke or transient ischemic attack: meta-Analysis of randomized controlled trials. Stroke52, e217–e223. 10.1161/strokeaha.120.033033

  • CrossRef
  • Google Scholar

• 4

  BroderickJ. P.AdeoyeO.ElmJ. (2017). Evolution of the modified Rankin scale and its use in future stroke trials. Stroke48, 2007–2012. 10.1161/strokeaha.117.017866

  • CrossRef
  • Google Scholar

• 5

  BuC. (2016). Clinical effect analysis of Xuesaitong combined with Edaravone in the treatment of cerebral infarction patients. J. Front. Med.6, 173–174.

  • Google Scholar

• 6

  ButcherN. J.MonsourA.MewE. J.ChanA. W.MoherD.Mayo-WilsonE.et al (2022). Guidelines for reporting outcomes in trial reports: The CONSORT-outcomes 2022 extension. Jama328, 2252–2264. 10.1001/jama.2022.21022

  • CrossRef
  • Google Scholar

• 7

  CaiS. (2011). Observation on effect of xuesaitong injection in the treatment of acute cerebral injection. Chin. Foreign Med. Res.9, 50–51. 10.14033/j.cnki.cfmr.2011.24.137

  • CrossRef
  • Google Scholar

• 8

  CaoY.ZhaoX.WangH.HuoH.LiJ. (2016). Effect of Xuesaitong injection on nerve function and hemorheology in elderly patients with ischemic stroke. Mod. J. Integr. Traditional Chin. West. Med.25, 1523–1525. 10.3969/j.issn.1008-8849.2016.14.014

  • CrossRef
  • Google Scholar

• 9

  ChamorroÁ.DirnaglU.UrraX.PlanasA. M. (2016). Neuroprotection in acute stroke: Targeting excitotoxicity, oxidative and nitrosative stress, and inflammation. Lancet Neurol.15, 869–881. 10.1016/s1474-4422(16)00114-9

  • CrossRef
  • Google Scholar

• 10

  ChangD.HeQ. (2017). Effect of xuesaitong soft capsule on neurological function and quality of life in patients with cerebral infarction. Strait Pharm. J.29, 134–136. 10.3969/j.issn.1006-3765.2017.08.068

  • CrossRef
  • Google Scholar

• 11

  ChenS.LiuJ.LiuX.FuY.ZhangM.LinQ.et al (2011). Panax notoginseng saponins inhibit ischemia-induced apoptosis by activating PI3K/Akt pathway in cardiomyocytes. J. Ethnopharmacol.137, 263–270. 10.1016/j.jep.2011.05.011

  • CrossRef
  • Google Scholar

• 12

  ChengC. W.WuT. X.ShangH. C.LiY. P.AltmanD. G.MoherD.et al (2017). CONSORT extension for Chinese herbal medicine Formulas 2017: recommendations, explanation, and elaboration. Ann. Intern Med.167, 112–121. 10.7326/m16-2977

  • CrossRef
  • Google Scholar

• 13

  DengJ. (2018). Clinical safety and effect analysis of Xuesaitong in the treatment of cerebral infarction. Healthmust-Readmagazine18, 36–37.

  • Google Scholar

• 14

  DirnaglU.IadecolaC.MoskowitzM. A. (1999). Pathobiology of ischaemic stroke: an integrated view. Trends Neurosci.22, 391–397. 10.1016/s0166-2236(99)01401-0

  • CrossRef
  • Google Scholar

• 15

  DuanH.AiM. (2009). Observation on the effect of 69 cases of xuesaitong injection in treatment of acute cerebral infarction. Pract. J. Card. Cereb. Pneumal Vasc. Dis.17, 192. 10.3969/j.issn.1008-5971.2009.03.020

  • CrossRef
  • Google Scholar

• 16

  FanR.ZhaoJ.MaM.HaoY. (2019). Effect of Xuesaitong injection combined with aspirin on senile ischemic stroke and serum Thl/Th2 cytokine level. Shaanxi J. Traditional Chin. Med.40, 1032–1035. 10.3969/j.issn.1000-7369.2019.08.013

  • CrossRef
  • Google Scholar

• 17

  FeiginV. L.NorrvingB.MensahG. A. (2017). Global burden of stroke. Circ. Res.120, 439–448. 10.1161/circresaha.116.308413

  • CrossRef
  • Google Scholar

• 18

  FuF.YangM.LiJ.ChengB.LiW. (2011). 122 cases of xuesaitong combined with sodium ozagrel on treating acute cerebral infarction. Guangming J. Chin. Mede26, 2087–2088. 10.3969/j.issn.1003-8914.2011.10.084

  • CrossRef
  • Google Scholar

• 19

  GaoF. (2017). Effect of Xuesaitong combined with Edaravone on hemorheology and recovery of nerve function in patients with cerebral infarction. Mod. Diagnosis Treat.28, 4541–4542. 10.3969/j.issn.1001-8174.2017.24.017

  • CrossRef
  • Google Scholar

• 20

  GBD2019 Diseases and Injuries Collaborators (2020). Global burden of 369 diseases and injuries in 204 countries and territories, 1990-2019: a systematic analysis for the global burden of disease study 2019. Lancet396, 1204–1222. 10.1016/s0140-6736(20)30925-9

  • CrossRef
  • Google Scholar

• 21

  GengH.ZhangL.XinC.ZhangC.XieY. (2022). Xuesaitong oral preparation as adjuvant therapy for treating acute cerebral infarction: A systematic review and meta-analysis of randomized controlled trials. J. Ethnopharmacol.285, 114849. 10.1016/j.jep.2021.114849

  • CrossRef
  • Google Scholar

• 22

  GuoG. (2017). Clinical effect of naloxone combined with Xuesaitong on elderly patients with cerebral infarction. China J. Pharm. Econ.12, 54–56. 10.12010/j.issn.1673-5846.2017.02.019

  • CrossRef
  • Google Scholar

• 23

  HanY.ZhongX.LiuY. (2014). Effect and clinical value of Xuesaitong on serum C-reactive protein level in acute stage of cerebral infarction. Chin. J. Basic Med. Traditional Chin. Med.20, 1529–1530. 10.19945/j.cnki.issn.1006-3250.2014.11.029

  • CrossRef
  • Google Scholar

• 24

  HankeyG. J.EikelboomJ. W. (2006). Aspirin resistance. Lancet367, 606–617. 10.1016/s0140-6736(06)68040-9

  • CrossRef
  • Google Scholar

• 25

  HeQ.TanC.ZhaoQ. (2011). Clinical observation of xuesaitong injection combined with conventional treatment on patients with cerebral infarction. Drugs Clin.26, 234–236.

  • Google Scholar

• 26

  HeR. (2006). Effect of xuesaitong injection on treating 40 patients with acute cerebral infarction. Hebei Med.12, 756–758. 10.3969/j.issn.1006-6233.2006.08.028

  • CrossRef
  • Google Scholar

• 27

  HeY.GaoX. M.LiL.LiuX. G.LiuW.HongX. J.et al (2020). Safety of the xuesaitong injection in China: results from a large-scale multicentre post-marketing surveillance study in a real-world setting. Curr. Med. Res. Opin.36, 1947–1953. 10.1080/03007995.2020.1832056

  • CrossRef
  • Google Scholar

• 28

  HuS.WuY.ZhaoB.HuH.ZhuB.SunZ.et al (2018). Panax notoginseng saponins protect cerebral microvascular endothelial cells against oxygen-glucose deprivation/reperfusion-induced barrier dysfunction via activation of PI3K/Akt/Nrf2 antioxidant signaling pathway. Molecules23, 2781. 10.3390/molecules23112781

  • CrossRef
  • Google Scholar

• 29

  HuY.WangX.ShiH. (2017). Clinical curative effect of Xuesaitong Injection combined with edaravone in treatment of elderly patients with acute cerebral infarction and influence on Hemorheology of patients with plasma C reactive protein. J. Qiqihar Univ. Med.38, 170–172. 10.3969/j.issn.1002-1256.2017.02.020

  • CrossRef
  • Google Scholar

• 30

  HuangY. (2016). Effect of Xuesaitong injection on acute cerebral infarction and thrombosis. Chin. J. Pract. Nerv. Dis.19, 102–103. 10.3969/j.issn.1673-5110.2016.05.066

  • CrossRef
  • Google Scholar

• 31

  HuangY.ZhangF.YuanX.LuanX.DongZ.GaiY.et al (2022). Evaluation on the efficacy and safety of Xuesaitong (freeze-dried) for injection in elderly patients with acute ischemic stroke. China's Naturop.30, 93–102. 10.1080/10410236.2020.1824662

  • CrossRef
  • Google Scholar

• 32

  JayarajR. L.AzimullahS.BeiramR.JalalF. Y.RosenbergG. A. (2019). Neuroinflammation: friend and foe for ischemic stroke. J. Neuroinflammation16, 142. 10.1186/s12974-019-1516-2

  • CrossRef
  • Google Scholar

• 33

  LiC.ChangX. (2016). Analysis on the clinical effect of Xuesaitong combined with Edaravone in the treatment of cerebral infarction. China Pract. Med.11, 158–159. 10.14163/j.cnki.11-5547/r.2016.23.113

  • CrossRef
  • Google Scholar

• 34

  LiC.KongL.YuanF.FanZ.HuL. (2016). Effect of Xuesaitong injection on senile ischemic stroke and its influence on quality of life. Chin. J. Gerontology36, 4183–4184. 10.3969/j.issn.1005-9202.2016.17.021

  • CrossRef
  • Google Scholar

• 35

  LiC.TangC.FuR. (2013). Clinical observation of xuesaitong dropping pills in adjuvant treatment of acute cerebral infarction. J. Front. Med.00, 383–384. 10.3969/j.issn.2095-1752.2013.24.485

  • CrossRef
  • Google Scholar

• 36

  LiD. (2017). Clinical effect of Xuesaitong combined with Edaravone in the treatment of cerebral infarction. China J. Pharm. Econ.12, 61–63. 10.12010/j.issn.1673-5846.2017.02.022

  • CrossRef
  • Google Scholar

• 37

  LiF.GuD.LiY.ShiJ. (1999). Observation on curative effects of treatment with panax notoginsenoside injection combined with batroxobin for acutely cerebral infarction in 31 patients. Chin. J. Integr. Tradit. West. Med. Intensive Crit. Care6, 470–472.

  • Google Scholar

• 38

  LiF.ZhaoH.HanZ.WangR.TaoZ.FanZ.et al (2019). Xuesaitong may protect against ischemic stroke by modulating microglial phenotypes and inhibiting neuronal cell apoptosis via the STAT3 signaling pathway. CNS Neurol. Disord. Drug Targets18, 115–123. 10.2174/1871527317666181114140340

  • CrossRef
  • Google Scholar

• 39

  LiH.DengC. Q.ChenB. Y.ZhangS. P.LiangY.LuoX. G. (2009). Total saponins of Panax notoginseng modulate the expression of caspases and attenuate apoptosis in rats following focal cerebral ischemia-reperfusion. J. Ethnopharmacol.121, 412–418. 10.1016/j.jep.2008.10.042

  • CrossRef
  • Google Scholar

• 40

  LiJ.LiuF. (2020). Curative effect of Xuesaitong combined with Salvia ligustrazine in the treatment of acute cerebral infarction. Contemp. Med. Forum18, 196–197. 10.3969/j.issn.2095-7629.2020.09.144

  • CrossRef
  • Google Scholar

• 41

  LiK.QinX. (2006). Observation on effect of xuesaitong injection on treating acute cerebral infarction. J. Clin. Med. Pract.15, 293.

  • Google Scholar

• 42

  LiM.JiangY. (2007). Clinical observation on xuesaitong injection in treatment of patients with acute cerebral infarction. Mod. Med. J. China9, 93–94. 10.3969/j.issn.1672-9463.2007.10.037

  • CrossRef
  • Google Scholar

• 43

  LiM.LiaoZ.YangX.FangL. (2015). Influence factors and gene expression patterns during MeJa-induced gummosis in peach. China J. Mod. Med.25, 49–61. 10.1016/j.jplph.2015.03.019

  • CrossRef
  • Google Scholar

• 44

  LiP.LiangL. (2002). Clinical observation of xuesaitong soft capsule in treating cerebral infarction. Zhejiang J. Integr. Traditional Chin. West. Med.12, 146–147. 10.3969/j.issn.1005-4561.2002.03.007

  • CrossRef
  • Google Scholar

• 45

  LiY.SongD.ZhangY.LeeS. S. (2003). Effect of neonatal capsaicin treatment on haemodynamics and renal function in cirrhotic rats. China Hydropower Med.15, 293–299. 10.1136/gut.52.2.293

  • CrossRef
  • Google Scholar

• 46

  LinX. (2007). “80 cases of cerebral infarction treated by integrated traditional Chinese and western medicine,” in Proceedings of the Sixth National Conference on Integrative Medicine and Neurology (Berlin, Germany: Springer).

  • Google Scholar

• 47

  LiuB.ZhangR.XieN.ChangW. (2019). Curative effect of Xuesaitong combined with Edaravone on patients with cerebral infarction. Hainan Med. J.30, 2625–2628.

  • Google Scholar

• 48

  LiuL. (2017). Observation on the effect of Xuesaitong combined with Edaravone injection in 120 patients with cerebral infarction. Guide China Med.15, 199–200. 10.15912/j.cnki.gocm.2017.23.158

  • CrossRef
  • Google Scholar

• 49

  LiuY.GeX.LiangJ.YanW.FeiL.JiaoY.et al (2005). Cross-inhibition to heterologous foot-and-mouth disease virus infection induced by RNA interference targeting the conserved regions of viral genome. Yunnan J. Traditional Chin. Med.26, 51–59. 10.1016/j.virol.2005.01.051

  • CrossRef
  • Google Scholar

• 50

  LuS. (2010). Clinical effect of xuesaitong soft capsule on lacunar cerebral infarction. Chin. Med. Guide8, 216–217. 10.3969/j.issn.1671-8194.2010.33.153

  • CrossRef
  • Google Scholar

• 51

  LuoX.WangP.CengX. (2011). Xue Sai Tong injection plus routine therapy for acute cerebral infarction and the influence on plasma C-reactive protein. Pract. J. Clin. Med.8, 96–98. 10.3969/j.issn.1672-6170.2011.05.038

  • CrossRef
  • Google Scholar

• 52

  MaC.WangR. (2009). Observation on the effect of 100 cases of xuesaitong injection in the treatment of acute cerebral infarction. Chin. J. Coal Ind. Med.12, 595–596.

  • Google Scholar

• 53

  Mayo-WilsonE.FuscoN.LiT.HongH.CannerJ. K.DickersinK.et al (2017). Multiple outcomes and analyses in clinical trials create challenges for interpretation and research synthesis. J. Clin. Epidemiol.86, 39–50. 10.1016/j.jclinepi.2017.05.007

  • CrossRef
  • Google Scholar

• 54

  MccoyC. E. (2018). Understanding the use of composite endpoints in clinical trials. West J. Emerg. Med.19, 631–634. 10.5811/westjem.2018.4.38383

  • CrossRef
  • Google Scholar

• 55

  MendelsonS. J.PrabhakaranS. (2021). Diagnosis and management of transient ischemic attack and acute ischemic stroke: A review. Jama325, 1088–1098. 10.1001/jama.2020.26867

  • CrossRef
  • Google Scholar

• 56

  MeretojaA.KeshtkaranM.SaverJ. L.TatlisumakT.ParsonsM. W.KasteM.et al (2014). Stroke thrombolysis: save a minute, save a day. Stroke45, 1053–1058. 10.1161/strokeaha.113.002910

  • CrossRef
  • Google Scholar

• 57

  MiG.WangJ. (2009). Clinical observation on the treatment of 45 cases of cerebral infarction with xuesaitong soft capsule. Yunnan J. Traditional Chin. Med.30, 22–23. 10.3969/j.issn.1007-2349.2009.03.015

  • CrossRef
  • Google Scholar

• 58

  NemaH.KatoM. (2010). Investigation of gastroduodenal mucosal injuries caused by low-dose aspirin therapy in patients with cerebral infarction. J. Gastroenterol. Hepatol.25 (1), S119–S121. 10.1111/j.1440-1746.2010.06229.x

  • CrossRef
  • Google Scholar

• 59

  OuyangZ.ZhangH. (2022). Clinical efficacy of Xuesaitong injection combined with Edaravone in patients with acute cerebral infarction. Heilongjiang J. Traditional Chin. Med.51, 52–54.

  • Google Scholar

• 60

  PageM. J.MckenzieJ. E.BossuytP. M.BoutronI.HoffmannT. C.MulrowC. D.et al (2021). The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. Bmj372, n71. 10.1136/bmj.n71

  • CrossRef
  • Google Scholar

• 61

  PaulS.Candelario-JalilE. (2021). Emerging neuroprotective strategies for the treatment of ischemic stroke: An overview of clinical and preclinical studies. Exp. Neurol.335, 113518. 10.1016/j.expneurol.2020.113518

  • CrossRef
  • Google Scholar

• 62

  PingL.LiW.WuZ. (2022). Effect of Xuesaitong injection combined with butylphthalide in the treatment of acute cerebral infarction and its influence on neurological function of patients. Clin. Res. Pract.7, 28–31. 10.19347/j.cnki.2096-1413.202227007

  • CrossRef
  • Google Scholar

• 63

  QuJ.XuN.ZhangJ.GengX.ZhangR. (2020). Panax notoginseng saponins and their applications in nervous system disorders: a narrative review. Ann. Transl. Med.8, 1525. 10.21037/atm-20-6909

  • CrossRef
  • Google Scholar

• 64

  RajsicS.GotheH.BorbaH. H.SroczynskiG.VujicicJ.ToellT.et al (2019). Economic burden of stroke: a systematic review on post-stroke care. Eur. J. Health Econ.20, 107–134. 10.1007/s10198-018-0984-0

  • CrossRef
  • Google Scholar

• 65

  RongZ.ZhiH. (2008). Case control study of treatment of acute cerebral infarction with xuesaitong injection. J. Neurol. Neurorehabil5, 85–86+96. 10.3969/j.issn.1672-7061.2008.02.007

  • CrossRef
  • Google Scholar

• 66

  ShenW.ZhangY. (2017). Clinical effect of Naoxintong capsule combined with Xuesaitong injection on cerebral infarction. Clin. J. Med. Officers45, 1279–1281. 10.16680/j.1671-3826.2017.12.22

  • CrossRef
  • Google Scholar

• 67

  ShiX.YuW.LiuL.LiuW.ZhangX.YangT.et al (2017). Panax notoginseng saponins administration modulates pro-/anti-inflammatory factor expression and improves neurologic outcome following permanent MCAO in rats. Metab. Brain Dis.32, 221–233. 10.1007/s11011-016-9901-3

  • CrossRef
  • Google Scholar

• 68

  SongY. (2013). Observation on the effect of 80 cases of xuesaitong in the treatment of acute cerebral infarction. Chin. J. Urban Rural. Ind. Hyg.28, 40–41. 10.16286/j.1003-5052.2013.05.064

  • CrossRef
  • Google Scholar

• 69

  SongY.YuH.CaoJ. (2022). Study on the effect of sodium chloride butylphthalein injection combined with Xuesaitong in treating acute cerebral infarction. OUR Health22, 61–63.

  • Google Scholar

• 70

  Von HippelP. T. (2015). The heterogeneity statistic I(2) can be biased in small meta-analyses. BMC Med. Res. Methodol.15, 35. 10.1186/s12874-015-0024-z

  • CrossRef
  • Google Scholar

• 71

  WanJ. B.LeeS. M.WangJ. D.WangN.HeC. W.WangY. T.et al (2009). Panax notoginseng reduces atherosclerotic lesions in ApoE-deficient mice and inhibits TNF-alpha-induced endothelial adhesion molecule expression and monocyte adhesion. J. Agric. Food Chem.57, 6692–6697. 10.1021/jf900529w

  • CrossRef
  • Google Scholar

• 72

  WangC. (2015). Clinical efficacy of Xuesaitong combined with Biaspirin in acute cerebral infarction. Fam. Ther.11, 11–12.

  • Google Scholar

• 73

  WangG.MaoQ.ZhouZ.ChiuC. Y.ShenJ. F.LinY. J.et al (2011a). The HADS and the DT for screening psychosocial distress of cancer patients in Taiwan. J. Emerg. Tradit. Chin. Med.20, 639–646. 10.1002/pon.1952

  • CrossRef
  • Google Scholar

• 74

  WangG.WanJ. B.ChanS. W.DengY. H.YuN.ZhangQ. W.et al (2011b). Comparative study on saponin fractions from Panax notoginseng inhibiting inflammation-induced endothelial adhesion molecule expression and monocyte adhesion. Chin. Med.6, 37. 10.1186/1749-8546-6-37

  • CrossRef
  • Google Scholar

• 75

  WangJ.ChengY.LiuD.HuX.LiF. (2020). Therapeutic effect of xuesaitong injection combined with butylphthalide injection on cerebral infarction and its effect on NIHSS score and hemodynamics. Chin. Archives Traditional Chin. Med.38, 122–124. 10.13193/j.issn.1673-7717.2020.08.030

  • CrossRef
  • Google Scholar

• 76

  WangM.JiangB.SunH.RuX.SunD.WangL.et al (2017b). Prevalence, incidence, and mortality of stroke in China: results from a nationwide population-based survey of 480 687 adults. Circulation135, 759–771. 10.1161/circulationaha.116.025250

  • CrossRef
  • Google Scholar

• 77

  WangM.NingT.GaoT.ZhaoX.LvZ. (2018). Effects of Xiao yao san on interferon-α-induced depression in mice. Psychol. Dr.24, 197–202. 10.1016/j.brainresbull.2017.12.001

  • CrossRef
  • Google Scholar

• 78

  WangM.WengQ.ChaQ. (2017a). Evaluation of clinical curative effect of Xuesaitong soft capsule on patients with acute lacunar infarction combined with cerebral microhemorrhage. Chin. Clin. Pharmacol. Ther.22, 574–579.

  • Google Scholar

• 79

  WangQ. (2007). Xuesaitong injection on treating acute cerebral infarction. J. Med. Forum28, 59. 10.3969/j.issn.1672-3422.2007.20.038

  • CrossRef
  • Google Scholar

• 80

  WangR. (2022). Study on the effect of Xuesaitong in the treatment of acute cerebral infarction in the elderly. Self-Care22, 169–171.

  • Google Scholar

• 81

  WangT.LiuH.LuH.LiY. L.XuK.LouH. X. (2021). Two new quinazoline derivatives from the moss endophytic fungus Aspergillus sp. and their anti-inflammatory activity. Chin. J. Pract. Med.48, 105–110. 10.1007/s13659-020-00287-5

  • CrossRef
  • Google Scholar

• 82

  WangW.LiH.TsengT.HsuW. Y.WangC. F.HsuC. C.et al (2007). Effects of apomorphine on the expression of learned helplessness behavior. J. Med. Forum28, 63–68.

  • Google Scholar

• 83

  WangW. (2006). Xuesaitong injection in the treatment of acute cerebral infarction. J. Med. Forum27, 56. 10.3969/j.issn.1672-3422.2006.15.035

  • CrossRef
  • Google Scholar

• 84

  WangY.WangY.ZhaoX.LiuL.WangD.WangC.et al (2013). Clopidogrel with aspirin in acute minor stroke or transient ischemic attack. N. Engl. J. Med.369, 11–19. 10.1056/NEJMoa1215340

  • CrossRef
  • Google Scholar

• 85

  WangY.ZhaoX.LinJ.LiH.JohnstonS. C.LinY.et al (2016). Association between CYP2C19 loss-of-function allele status and efficacy of clopidogrel for risk reduction among patients with minor stroke or transient ischemic attack. Jama316, 70–78. 10.1001/jama.2016.8662

  • CrossRef
  • Google Scholar

• 86

  WuH. (2020). Effect of Xuesaitong combined with Edaravone in the treatment of cerebral infarction and its influence on hemorheology. Henan Med. Res.29, 1458–1460. 10.3969/j.issn.1004-437X.2020.08.059

  • CrossRef
  • Google Scholar

• 87

  WuL.SongH.ZhangC.WangA.ZhangB.XiongC.et al (2023). Efficacy and safety of panax notoginseng saponins in the treatment of adults with ischemic stroke in China: A randomized clinical trial. JAMA Netw. Open6, e2317574. 10.1001/jamanetworkopen.2023.17574

  • CrossRef
  • Google Scholar

• 88

  WuS.JiaZ.DongS.HanB.ZhangR.LiangY.et al (2019b). Panax notoginseng saponins ameliorate leukocyte adherence and cerebrovascular endothelial barrier breakdown upon ischemia-reperfusion in mice. J. Vasc. Res.56, 1–10. 10.1159/000494935

  • CrossRef
  • Google Scholar

• 89

  WuS.WuB.LiuM.ChenZ.WangW.AndersonC. S.et al (2019a). Stroke in China: advances and challenges in epidemiology, prevention, and management. Lancet Neurol.18, 394–405. 10.1016/s1474-4422(18)30500-3

  • CrossRef
  • Google Scholar

• 90

  XiaoJ.FengJ.RongT.LiF. (2019). Effect of Xuesaitong injection on ECG changes and hemorheology in patients with acute cerebral infarction. Mod. J. Integr. Traditional Chin. West. Med.28, 202–205. 10.3969/j.issn.1008-8849.2019.02.024

  • CrossRef
  • Google Scholar

• 91

  XiaoS.LiuJ.WangB. (2020). Effect of citicoline combined with Xuesaitong Injection on the level of serum TLR4/NF-κB signaling pathway in patients with ACI. J. Mol. Diagnostics Ther.12, 737–745.

  • Google Scholar

• 92

  XuN. (2021). To observe the effect of xuesaitong combined with alprostadil in the treatment of patients with cerebral infarction. Guide China Med.19, 121–122. 10.15912/j.cnki.gocm.2021.22.055

  • CrossRef
  • Google Scholar

• 93

  XueC. (2018). Clinical observation of Xuesaitong combined with aspirin in the treatment of cerebral infarction complicated with middle cerebral artery stenosis. Chin. J. Ethnomedicine Ethnopharmacy27, 93–95. 10.3969/j.issn.1007-8517.2018.7

  • CrossRef
  • Google Scholar

• 94

  YanH.LiuY.LingS. (2023). Effect of Xuesaitong injection combined with atorvastatin on hemorheology and serum related indexes in patients with acute cerebral infarction. Mod. Diagnosis Treat.34, 538–540.

  • Google Scholar

• 95

  YangH. (2021). Effect and safety analysis of Xuesaitong injection in the treatment of senile ischemic stroke. Chin. Health Care39, 146–148.

  • Google Scholar

• 96

  YangM.ChengY. (2012). 30 cases of xuesaitong injection combined with sodium ozagrel on treating acute cerebral infarction. China Mod. Med.19, 124. 10.3969/j.issn.1674-4721.2012.10.071

  • CrossRef
  • Google Scholar

• 97

  YangZ.ShaoQ.GeZ.AiN.ZhaoX.FanX. (2017). A bioactive chemical markers based strategy for quality assessment of botanical drugs: xuesaitong injection as a case study. Sci. Rep.7, 2410. 10.1038/s41598-017-02305-y

  • CrossRef
  • Google Scholar

• 98

  YeD.ShiB.LinL. (2018). Effect of Xuesaitong assisted hyperbaric oxygen therapy on hemodynamics and nerve function in patients with cerebral infarction. World J. Integr. Traditional West. Med.13, 74–76+80. 10.13935/j.cnki.sjzx.180121

  • CrossRef
  • Google Scholar

• 99

  YuanL. (2003). Observation on 34 cases of xuesaitong injection in the treatment of acute cerebral infarction. J. Emerg. Tradit. Chin. Med.12, 133–145.

  • Google Scholar

• 100

  YuanQ.JiangJ. (2006). Observation on 49 cases of xuesaitong injection combined with defibrase in treating with acute cerebral infarction. J. Snake18, 20–21. 10.3969/j.issn.1001-5639.2006.01.006

  • CrossRef
  • Google Scholar

• 101

  ZhangA.ZhangX. (2008). Clinical study on xuesaitong injection in the treatment of acute cerebral infarction. J. Pract. Med. Tech.15, 2508–2509. 10.3969/j.issn.1671-5098.2008.19.042

  • CrossRef
  • Google Scholar

• 102

  ZhangF.LiuJ.LiJ. (2022). Clinical study of Xuesaitong injection combined with agattriban and clopidogrel in the treatment of acute cerebral infarction. Chin. J. Integr. Med. Cardio/Cerebrovascular Dis.20, 3789–3793. 10.12102/j.issn.1672-1349.2022.20.031

  • CrossRef
  • Google Scholar

• 103

  ZhangL.LiW.JinM.YangC. (2020). Study on the therapeutic effect of Xuesaitong on cerebral infarction and the mechanism of hemorheology. Int. Med. Health Guid. News26, 2230–2233. 10.3760/cma.j.issn.1007-1245.2020.15.018

  • CrossRef
  • Google Scholar

• 104

  ZhangL.ZhangS. (2003). Observation on 34 cases of xuesaitong injection in the treatment of acute cerebral infarction. J. Emerg. Tradit. Chin. Med.12, 133–145. 10.3969/j.issn.1008-7044.2002.01.065

  • CrossRef
  • Google Scholar

• 105

  ZhangM.GuanY.XuJ.QinJ.LiC.MaX.et al (2019). Evaluating the protective mechanism of panax notoginseng saponins against oxidative stress damage by quantifying the biomechanical properties of single cell. Anal. Chim. Acta1048, 186–193. 10.1016/j.aca.2018.10.030

  • CrossRef
  • Google Scholar

• 106

  ZhangS. (2018). Clinical effect of naloxone combined with Xuesaitong in the treatment of senile cerebral infarction. Guide China Med.16, 108–109. 10.15912/j.cnki.gocm.2018.01.086

  • CrossRef
  • Google Scholar

• 107

  ZhangT. (2016). The clinical effect of xuesaitong combined with aspirin in acute cerebral infarction. J. Aerosp. Med.27, 428–429. 10.3969/j.issn.2095-1434.2016.04.011

  • CrossRef
  • Google Scholar

• 108

  ZhangX.WuJ.ZhangB. (2015). Xuesaitong injection as one adjuvant treatment of acute cerebral infarction: a systematic review and meta-analysis. BMC Complement. Altern. Med.15, 36. 10.1186/s12906-015-0560-4

  • CrossRef
  • Google Scholar

• 109

  ZhangY.BishopP. A. (2019). Risks of heat illness in athletes with spinal cord injury: Current evidence and needs. Electron. J. Clin. Med. Literature6, 68–69. 10.3389/fspor.2019.00068

  • CrossRef
  • Google Scholar

• 110

  ZhaoB. (2006). Observation on 40 cases of xuesaitong injection combined with defibrase and qinger injection in the treatment of acute cerebral infarction. Chin. Med. Guid.3, 63–64.

  • Google Scholar

• 111

  ZhaoW.HuH.LiuP.TanM. (2022). Optimization and evaluation of protein C-terminal peptide enrichment strategy based on arginine cleavage. Se Pu40, 17–27. 10.3724/SP.J.1123.2021.03030

  • CrossRef
  • Google Scholar

• 112

  ZhongD.DengX. (2017). Observation of clinical effect of urinary Kallidinogenase combined with Xuesaitong, local mild hypothermia in treatment of cerebral infarction. Jilin Med. J.38, 660–662. 10.3969/j.issn.1004-0412.2017.04.023

  • CrossRef
  • Google Scholar

• 113

  ZhouD.CenK.LiuW.LiuF.LiuR.SunY.et al (2021). Xuesaitong exerts long-term neuroprotection for stroke recovery by inhibiting the ROCKII pathway, in vitro and in vivo. J. Ethnopharmacol.272, 113943. 10.1016/j.jep.2021.113943

  • CrossRef
  • Google Scholar

• 114

  ZhouN.TangY.KeepR. F.MaX.XiangJ. (2014). Antioxidative effects of Panax notoginseng saponins in brain cells. Phytomedicine21, 1189–1195. 10.1016/j.phymed.2014.05.004

  • CrossRef
  • Google Scholar

• 115

  ZhouQ.LiuJ.LiY.YuanL. (2015). Naloxone combined Xuesaitong in the treatment of senile cerebral infarction. Shaanxi J. Traditional Chin. Med.2015, 964–966. 10.3969/j.issn.1000-7369.2015.08.013

  • CrossRef
  • Google Scholar

• 116

  ZhuY.LuG.ZhouJ. (2016). Effect of xuesaitong injection combined with butylphthalide on neurological function in patients with acute cerebral infarction. Chin. J. Thrombosis Hemostasis22, 618–620. 10.3969/j.issn.1009-6213.2016.06.006

  • CrossRef
  • Google Scholar

• 117

  ZiJ.YueY.MaF. (2008). Clinical observation on xuesaitong injection in treating acute cerebral infarction. Chin. J. Misdiagnostics8, 7343–7354. 10.3969/j.issn.1009-6647.2008.30.036

  • CrossRef
  • Google Scholar