A commentary on “Efficacy and safety of intra-arterial thrombolysis after endovascular reperfusion for acute ischemic stroke: a systematic review and meta-analysis of randomized trials”

Dear Editor,

We read with great interest the recent article titled “Efficacy and safety of intra-arterial thrombolysis after endovascular reperfusion for acute ischemic stroke: a systematic review and meta-analysis of randomized trials”^[1] published in the International Journal of Surgery. In this study, patients with acute ischemic stroke due to large vessel occlusion (AIS-LVO) had their intra-arterial thrombolysis (IAT) safety and effectiveness evaluated after near-complete to complete reperfusion by endovascular thrombectomy (EVT). It offers a comprehensive assessment of the transcriptome and new data suggesting that IAT, when used in conjunction with effective EVT, appears to improve the outstanding functional result in patients with AIS-LVO without significantly raising the risk of symptomatic intracranial hemorrhage (sICH) or death. Future studies must, however, address a number of important concerns.

First, Limitations of the study design. The study, despite having seven randomized controlled trials (RCTs), demonstrates significant heterogeneity. For example, the POST-TNK study had a greater sample size (n = 540) than the CHOICE trial, which was prematurely stopped and had an inadequate sample size (n = 114)^[2]. This could have an impact on the results’ balance. Furthermore, subgroup analysis has not been able to effectively explain the significant variations in the types of thrombolytic drugs employed (antithrombin, tranexamic acid, and urokinase) and dosages between trials (for instance, tranexamic acid dosages range from 0.03125 to 0.125 mg/kg). The paucity of data from European and American populations in all trials, which were carried either in China or Spain, restricts how broadly the findings may be applied. Future studies should include more regional data and standardize the thrombolytic agent regimens.

Secondly, possible issues with statistical techniques. No non-linear relationship tests for continuous variables (such time from onset to randomization) were reported, even though random effects models and I² tests (I² = 0) were used. Although the Z-curve for mRS 0-1 findings passed the monitoring boundary, according to TSA analysis, it was unclear if the pre-set expected effect size (RR = 1.2, for example) was appropriate^[3]. The statistical power for indirect comparisons in the network meta-analysis is inadequate because there are only three thrombolytic medications and only seven studies, which could result in misleading negative results. To improve robustness in small sample comparisons, Bayesian framework studies should be supplemented.

Thirdly, the clinical variables are not adequately defined; the definition of symptomatic intracranial hemorrhage (sICH) is not clear whether it follows SITS-MOST or HELPSS standards, which may affect the consistency of safety assessments; the moderating effect of baseline NIHSS or ASPECTS scores on efficacy has not been analyzed, despite the fact that these are important predictors of prognosis; the success criteria for reperfusion are not uniform (TICI 2b50-3 vs. 2c-3); and the subgroup analysis indicates a significant difference in efficacy between the two (P = 0.03), but the anatomical mechanism was not investigated.

In summary, although the present study offers significant information regarding the safety and effectiveness of mechanical thrombectomy for stroke patients who have major vascular blockage, a number of areas need more explanation and research. To guarantee uniformity across investigations, a single standard for safety evaluations that takes into account both SITS-MOST and HELPSS criteria is required. Furthermore, a thorough examination of how baseline NIHSS and ASPECTS scores affect treatment results is essential for improving patient selection and prognostication. These gaps should be filled by future studies in order to enhance clinical results and manage stroke patients as best as possible. All this was observed regarding the transparency in the reporting of Artificial Intelligence – the TITAN guideline^[4].

Ethical approval

Not applicable.

Consent

None.

Sources of funding

None.

Author contributions

R.M., J.L., and B.Y. collaborated on the creation and structure of this letter.

Conflicts of interest disclosure

There are no conflicts of interest.

Guarantor

Bobo Yuan.

Research registration unique identifying number (UIN)

Not applicable.

Provenance and peer review

This manuscript is a comment without being invited.

Data availability statement

No data was used in this Letter to the Editor.