Endovascular thrombectomy vs tPA: better function, same mortality

View on the News

Meta-analyses have inherent limitations

It is important to note some limitations with this well-conducted meta-analysis. First, functional outcomes showed significant heterogeneity, which the authors attributed to variations in patient-, treatment-, and study-related factors.

Second, the confidence intervals for mortality and intracranial hemorrhage were wide, indicating that more data are necessary to fully inform these outcomes.

Third, five of the eight trials were halted early because of the evident superiority of endovascular thrombectomy, which means they fell substantially short (by up to 74%) of their planned sample sizes. This tends to cause overestimation of treatment effects. Fourth, nearly all these strokes involved carotid territory, nearly all the patients were on the young end of the age spectrum, and very few participants had comorbidities such as AF or diabetes. Such favorable characteristics do not reflect real-world experience with ischemic stroke.

Dr. Joanna M. Wardlaw and Dr. Martin S. Dennis are at the Centre for Clinical Brain Sciences at the University of Edinburgh (Scotland). They reported having no relevant financial disclosures. Dr. Wardlaw and Dr. Dennis made these remarks in an editorial accompanying Dr. Badhiwala’s meta-analysis (JAMA 2015;314:1803-4).




Endovascular mechanical thrombectomy yielded better function and revascularization rates but similar mortality and intracranial hemorrhage rates as standard medical therapy using tissue plasminogen activator (tPA) in a meta-analysis of eight high-quality randomized clinical trials comparing the two approaches for acute ischemic stroke.

The results were published online Nov. 3 in JAMA.

Copyright American Stroke Association

This meta-analysis included only large multicenter trials published from 2013 to the present. Previous trials and meta-analyses “had several well-recognized limitations” including inconsistent use of vascular imaging to confirm vessel occlusion before randomization, variable use of tPA in patients who eventually were assigned to endovascular therapy, and reliance on less effective and now outdated mechanical devices, said Dr. Jetan H. Badhiwala of the division of neurosurgery, University of Toronto, and his associates.

The eight trials included 2,423 patients (mean age, 67.4 years); 46.7% were women. A total of 1,313 patients underwent endovascular therapy, defined as the intra-arterial use of a microcatheter or other device for mechanical thrombectomy, with or without the local use of a chemical thrombolytic agent. The remaining 1,110 received standard medical therapy (tPA). The interval between stroke onset and endovascular treatment varied from 5 to 12 hours across these studies, with a mean of 3.8 hours.

Patients who had endovascular thrombectomy showed significantly higher rates of functional independence at 90 days (44.6%) than did those who had tPA (31.8%), for an OR of 1.71 and a number needed to treat of 8. The rate of angiographic revascularization at 24 hours also was markedly higher for endovascular thrombectomy (75.8% vs 34.1%), for an OR of 6.49, the investigators said (JAMA 2015;314:1832-43).

However, there were no significant differences between the two study groups in rates of symptomatic intracranial hemorrhage at 90 days (5.7% vs 5.1%) or all-cause mortality at 90 days (15.8% vs 17.8%), and overall morbidity including in-hospital rates of deep venous thrombosis, MI, and pneumonia also were similar.

No sponsor or source of financial support was reported for this study. Dr. Badhiwala and his associates reported having no relevant financial disclosures.

Next Article: