Yang P, Zhang Y, Zhang L, et al. Endovascular Thrombectomy with or without Intravenous Alteplase in Acute Stroke. N Engl J Med. 2020 May 21;382(21):1981-93.
Reviewed by Wazim Mohamed, MD
Read the article.*
*You will need to log in to US National Library of Medicine National Institutes of Health to read this article.
As thrombectomy has become the mainstay of treatment for large vessel occlusion strokes, the additional benefit offered by thrombolytics has come into question. Thrombolysis with alteplase may aid in early reperfusion prior to thrombectomy and may have a role in dissolving distal thrombi post-thrombectomy but is ineffective for large proximal clots. Moreover, partial lysis may lead to migration of clots distally, rendering them inaccessible to thrombectomy.
The current study (DIRECT-MT) sought to determine if thrombectomy alone was non-inferior to combination therapy (thrombectomy preceded by intravenous alteplase) in patients with ischemic stroke due to large vessel occlusion in the anterior circulation. This was a multicenter, randomized, open label trial with blinded outcomes assessment conducted at 41 tertiary care centers in China. Patients were randomized (1:1) to thrombectomy alone or combination aletplase and thrombectomy. Adult patients with large vessel occlusion of the first two segments of the MCA presenting within 4.5 hours of stroke symptoms and an NIHSS of ≥2 were eligible. Patients with a pre-stroke modified Rankin scale (mRS) >2 were excluded. Stent retrievers were primarily used for thrombectomy, with aspiration devices and intra-arterial thrombolytics also available.
The primary outcome was a shift analysis on the mRS at 90 days analyzed for noninferiority. Physicians assessing outcomes were blinded to the treatment group assignments. Secondary outcomes included: (1) death within 90 days, (2) successful reperfusion before thrombectomy, (3) eTICI >2b, (4) percentage of patients with recanalization at 72 hours, (5) final lesion volume on CT, (6) comparison of mRS scores, (7) EuroQoL questionnaire for quality of life and (8) Barthel index scores at 90 days. Safety outcomes included all hemorrhages, symptomatic hemorrhages, embolization/infarction in new vascular territories, pseudoaneurysms, groin hematoma and mortality at 90 days. The sample size of 636 patients was estimated to obtain a noninferiority margin of 0.8. All analyses were performed on an intention-to-treat basis. The primary outcome was determined with an odds ration adjusted for age, NIHSS, pre-stroke mRS, cerebral collateral blood flow and time to randomization.
Over a period of 16 months, 656 patients were randomized (327 in the thrombectomy group and 329 in the combination group). The baseline characteristics were similar between both the groups with a median age of 69 years, 56.4% men, median NIHSS of 17 and median ASPECTS of 9. The combination group had 88% of patients undergo thrombectomy defined as successful contact with the thrombus by any device and 91.4% in the thrombectomy alone group. A stent retriever was used in the majority of patients (95.8%). There was minimal intergroup crossovers – four patients in each arm. In the combination group, most patients (86.5%) completed alteplase infusion during the procedure. General anesthesia was administered in 32.4% of patients and intraarterial thrombolysis was used in seven patients in each group. The time from hospitalization to groin puncture was similar in both groups (less than 90 minutes). Only two patients were not assessed for primary outcome due to missing data. The unadjusted OR for the primary outcome in the thrombectomy group was 1.07 (95% CI 0.81-1.40). Thrombectomy alone was non-inferior to combination therapy because the lower end of CI was greater than the prespecified value of 0.8. Among secondary outcomes, the 90-day mortality was similar between both groups (17.7% in thrombectomy group vs. 18.8% in combination group). As expected, successful reperfusion prior to intervention was lower in the thrombectomy alone group (2.4%% vs. 7%; OR 0.33 [95% CI 0.14 – 0.74]). However, an eTICI score > 2b and recanalization at 72 hours were similar between both groups. All other secondary outcomes appeared to be similar in both groups, however, the authors chose not to comment on its significance due to lack of a prespecified plan for adjustments. Serious adverse events and procedural complications were also similar between both groups. Both symptomatic and asymptomatic hemorrhages were slightly lower in the thrombectomy alone group, but neither reached statistical significance.
The authors conclude that thrombectomy was non-inferior to combination therapy in patients with acute ischemic strokes who were eligible for both modalities. However, since the lower boundary of the OR was above the prespecified cut off of 0.8, the benefit of alteplase cannot be ruled out. Even though noninferiority was established for the thrombectomy group, the margin to establish this was quite generous and the wide confidence interval did not exclude benefit to almost 20% in the combination group. The authors also distinguish this study from other observational studies and meta-analyses by eliminating confounders and providing adjustments. They hypothesize this as a reason why the previous studies may have shown benefit to combination therapy. There was significantly more reperfusion prior to procedure in the combination group. At the end of the procedure as well, there was more reperfusion in the combination group, albeit nonsignificant. It is unclear why these findings did not translate to clinical outcomes as combination therapy appears to achieve earlier reperfusion. In this study, alteplase was always given at tertiary centers, which differs from United States standard of care where thrombolytics may be started prior to transfer from a primary stroke center to a comprehensive stroke center. It is encouraging that administration of alteplase prior to thrombectomy did not significantly increase the time to groin puncture. With the advent of newer aspiration devices and the recent traction gained by tenecteplase, thrombolytics combined with thrombectomy may still have a role to play, specifically in patients with failed thrombectomies. Larger trials in different patient populations, hospital triage systems and healthcare delivery systems are required to effectively compare thrombectomy to combination therapy. For now, adhering to the current guidelines that include thrombolysis may be the more pragmatic approach.