Mechanical Thrombectomy Outcomes With or Without Intravenous Thrombolysis

Insight From the ASTER Randomized Trial

Florent Gariel, MD, MSc; Bertrand Lapergue, MD, PhD; Romain Bourcier, MD, PhD; Jérôme Berge, MD; Xavier Barreau, MD; Mikael Mazighi, MD, PhD; Maéva Kyheng, BST; Julien Labreuche, BST; Robert Fahed, MD; Raphael Blanc, MD, MSc; Benjamin Gory, MD, PhD; Alain Duhamel, PhD; Suzana Saleme, MD; Vincent Costalat, MD, PhD; Serge Bracard, MD, PhD; Hubert Desal, MD, PhD; Lili Detraz, MD; Arturo Consoli, MD; Michel Piotin, MD, PhD; Gaultier Marnat, MD; for the ASTER Trial Investigators


Stroke. 2018;49(10):2383-2390. 

In This Article


In this study of patients among the ASTER trial population, we compared clinical and angiographic outcomes and safety between those having received or not IVT before MT.

We observed no significant difference in favorable outcome and angiographic recanalization between the 2 groups. However, we found a higher mortality rate in the MT alone group. The effect of IVT before MT remains a matter of debate in the literature, which includes meta-analysis, retrospective analysis, and post hoc analysis of randomized studies.[3,9–12]

The reason for not using IVT constitutes a major bias in the interpretation of the present data. Contraindication to IVT inevitably selects a particular population with higher comorbidity rates, such as intercurrent pathology and medication or a recent history of surgery. This may explain a higher morbi-mortality in the MT-alone group. Several studies have already reported a decrease in mortality in patients who received IVT before MT, but the explanation for lower mortality remains unclear.[3]

In our study, as expected, baseline characteristics of our 2 groups are not comparable. Patients in the MT-alone group were older, with a higher prestroke modified Rankin Scale and more often had previous antithrombotic medication. Prestroke anticoagulant medication was one of the most relevant differences between groups. Therefore, we performed a subgroup analysis, comparing patients with and without prestroke anticoagulant medication. These 2 subgroup populations were comparable (Table in the online-only Data Supplement). Among patients without anticoagulants, we found a higher mTICI2b/3 reperfusion rate after first-line strategy in the IVT+MT group. This effect was not found on final recanalization rate, but interestingly, the number of MT passes was significantly lower in the IVT+MT group. As some authors previously suggested,[3,13] thrombolysis can, therefore, be thought to facilitate recanalization by reducing the number of passes required to obtain favorable recanalization. Among patients not on anticoagulants, our results demonstrate that functional outcome was significantly better in the IVT+MT group compared with MT alone, both in the short and long terms. We hypothesize that IVT may be beneficial in cases of incomplete recanalization with MT thanks to its action on distal clots inaccessible to endovascular treatment. This may contribute to achieve better final reperfusion. However, preclinical data suggest that the role of IVT is more complex than a simple proximal clot lysis. In an animal transient middle cerebral artery occlusion model, IVT acts on the downstream microvascular thrombosis that starts immediately after the occlusion, limiting the infarct extension and allowing better functional results.[14] Such elements may potentially explain better functional outcome and lower 90-day mortality for patients in the IVT+MT group not on anticoagulants before stroke onset.

Our stratified analysis according to first-line therapy in patients without oral anticoagulation suggests that thrombolysis facilitates thrombectomy by SR but not by CA. Indeed, reperfusion after the first-line strategy was significantly better in the IVT+MT group with a lower number of passes only in SR first-line subgroup patients. This may be explained by a potential lower sensitivity of SR compared with CA to thrombolysis-induced thrombus fragmentation. This hypothesis is also sustained by the greater number of passes after thrombolysis in the CA first-line subgroup. These elements could thus justify the use of SR rather than CA in patients receiving IVT before MT.

The relationship between IVT and emboli in a new territory remains unclear. On the one hand, IVT could induce thrombus fragmentation and thus facilitate emboli during thrombectomy. But, conversely, in a post hoc analysis of the ESCAPE trial (The Endovascular Treatment for Small Core and Anterior Circulation Proximal Occlusion With Emphasis on Minimizing CT to Recanalization Times), Ganesh et al[15] concluded that IVT before thrombectomy reduced by approximately two-thirds the likelihood of an infarct in a new, previously unaffected territory, complicating MT. Their findings have not been reported in the other randomized clinical trials to date. In our study, we found no effect of thrombolysis on the occurrence of distal emboli.

Despite thrombolysis-induced coagulopathy, we found no difference in hemorrhagic complication rate in IVT+MT versus MT alone groups. Likewise, there was no difference in terms of procedural complications. These findings are consistent with the majority of previous publications[3,9–11] although one showed the possibility of an increased risk of asymptomatic intracranial hemorrhage.[16]

In our study, almost two-thirds of patients followed a drip-and-ship paradigm. Drip-and-ship patients in the MT+IVT group received IVT before transfer and therefore potentially benefited from the action of IVT on distal microthrombosis during transfer to the comprehensive stroke center. This would potentially be a further argument for thrombolysis in the drip-and-ship strategy. On the other hand, Gerschenfeld et al[17] recently reported that patients treated with the drip-and-ship approach had significantly longer process times (onset to IVT, onset to puncture, IVT to puncture, and onset to recanalization) compared with those benefiting from a mothership approach, without any effect on clinical outcome. In the meta-analysis of Mistry et al,[3] there was no clear argument in favor of extending the treatment times (onset-to-groin) associated with the use of IVT. In the present study, IVT was not responsible for any significant management delays as the different treatment times (onset-to-groin puncture, onset to imaging, and imaging to recanalization) were similar between the 2 groups.

The present study experiences several limitations. First, measured or unmeasured variables may represent potential confounding factors that cannot be ruled out despite our prespecified adjustment. Furthermore, we cannot exclude false positive results because of the multiple testing issues. In addition, the results of subgroups analysis should be taken with caution as we lack statistical power. Finally, despite the intention-to-treat study design, some patients were not included in the ASTER trial because of spontaneous intracranial recanalization after IVT alone, thereby underestimating the effect of thrombolysis. Tsivgoulis et al[18] recently showed that spontaneous recanalization after IVT and before thrombectomy occurred in 11% of cases and in 17% of cases if tandem occlusions are excluded.

Another potential limitation is the relatively small number of patients in the subgroup without prestroke anticoagulation among the MT alone population. Otherwise, in the MT-alone group, ≈25% of patients were ineligible to IVT because of an arrival outside the IVT window. Even if here door-to-groin times are similar, this is a potential bias. Indeed, patients with onset-imaging >4.5 hours could be overrepresented among MT alone group.

Despite our results, a randomized comparative study in patients eligible for IVT remains necessary to determine the exact impact of IVT in patients undergoing MT. This study should include patients in centers with comparable door-to-needle and door-to-groin times.

Finally, this is a post hoc analysis of patients randomized for MT first-line strategy and not for prethrombectomy IVT.