Hold the tPA on Some Thrombectomy Candidates, Trials Suggest

Two small trials added to the evidence that some stroke patients could fare well going straight to mechanical thrombectomy without IV thrombolysis, though the studies were confined to Asian populations and non-inferiority was not shown in the strictest sense.

For patients with acute ischemic stroke due to large vessel occlusion, results were inconclusive (but somewhat favorable) for a strategy of skipping the IV tissue plasminogen activator (tPA) usually recommended prior to mechanical thrombectomy in the Japanese SKIP trial.

However, a similar strategy did prove to be non-inferior to combined thrombectomy and IV tPA in the Chinese DEVT trial.

IV thrombolysis is thought to benefit thrombectomy candidates by increasing the chances of reperfusion. The trade-offs, however, include an increased risk of bleeding complications, potential thrombus fragmentation, and worsened distal perfusion, and extra cost and work for the stroke team.

Both SKIP and DEVT were published online in JAMA.

SKIP: Thrombectomy Alone Neither Non-inferior Nor Inferior?

Thrombectomy alone did not meet non-inferiority criteria against the standard of thrombectomy plus IV alteplase, but SKIP investigators couldn’t say it was inferior, either.

A good functional outcome, a modified Rankin Scale score 0-2 at 90 days, was similarly likely between stroke patients getting thrombectomy alone and peers who had combined therapy (59.4% vs 57.3%), yet the confidence interval around the effect estimate was wide and crossed the non-inferiority margin of 0.74 (P=0.18 for non-inferiority).

“Although this study hypothesis could not be proved, the point estimates of treatment effect for mechanical thrombectomy alone was nominally slightly better, not worse, compared with combined therapy. Accordingly, a larger trial or meta-analysis of trials is needed to conclusively assess noninferiority,” according to Kazumi Kimura, MD, PhD, of Nippon Medical School in Tokyo, and colleagues.

Primary results were similar across subgroups and study sites.

There were no between-group differences in 90-day mortality (7.9% vs 8.7%). Intracerebral hemorrhage rates at 36 hours favored the group receiving mechanical thrombectomy alone (33.7% vs 50.5%, P=0.02), but symptomatic intracerebral hemorrhage (sICH) did not favor either group (5.9% vs 7.7%).

SKIP was a noninferiority trial conducted at 23 hospital networks in Japan. Participants were randomized to mechanical thrombectomy with or without IV alteplase 0.6 mg/kg.

The SKIP cohort included 204 stroke patients (median age 74; 62.7% men). All had acute stroke with internal carotid artery (ICA) or M1 occlusion.

Roughly 8 minutes separated lytic drug start and endovascular therapy start in those randomized to both treatments.

The trial’s open-label design was a major limitation, Kimura’s group acknowledged. Additionally, favorable outcomes were more frequent than expected, potentially reducing the study’s power.

SKIP’s main findings were previously reported at the 2020 International Stroke Conference.

DEVT: Thrombectomy Alone Reaches Noninferiority

In the DEVT trial, stroke patients randomized to mechanical thrombectomy alone had no worse outcomes than controls, and investigators were able to show non-inferiority.

Patients deemed functionally independent at 90 days (mRS 0-2) reached 54.3% of the thrombectomy alone group versus 46.6% of peers undergoing combined IV tPA and thrombectomy, according to Qingwu Yang, MD, PhD, of Xinqiao Hospital and The Second Affiliated Hospital, Army Medical University in Chongqing, China, and colleagues.

Given that the lower boundary of the CI was -5.1% and fell within the fairly generous noninferiority margin of -10%, the thrombectomy-only strategy met non-inferiority (P=0.003 for non-inferiority).

Both groups experienced similar rates of sICH within 48 hours (6.1% vs 6.8%) and 90-day mortality (17.2% vs 17.8%).

A major caveat of DEVT was that it had been terminated early because of efficacy, when 234 out of a planned 970 patients were randomized, Yang’s group noted.

DEVT participants were patients at 33 experienced stroke centers in China (mean age 68; 43.6% women). The cohort was randomized to endovascular thrombectomy alone or with IV alteplase.

Unlike SKIP, the DEVT trial had participants randomized to alteplase receive the full dose of 0.9 mg/kg. Additionally, centers took longer, approximately 40 minutes, between lytic drug start and thrombectomy start.

Drawing Conclusions from Limited Evidence

Ultimately, the two trials are in line with prior retrospective studies and a recent randomized trial, DIRECT-MT, which had suggested that outcomes after stroke were comparable whether mechanical thrombectomy was or was not offered in tandem with IV thrombolysis within 4.5 hours of symptom onset.

“With its higher drug dose and longer interlude for drug action, the DEVT trial more strongly probed the ability of IV lytics to improve outcome by quickly dissolving the target occlusion before EVT [endovascular thrombectomy] can be performed,” commented JAMA editor Jeffrey Saver, MD, of University of California Los Angeles, and Opeolu Adeoye, MD, of the University of Cincinnati.

“The SKIP trial, in comparison, more fully explored the possibility that lower-dose, shorter interlude IV lytics can improve outcome by dissolving residual small thrombi in the distal vasculature after incomplete endovascular reperfusion while minimizing the risk of hemorrhagic transformation,” they wrote in an accompanying editorial.

Yang and colleagues acknowledged that DEVT, SKIP, and DIRECT-MIT all included Asian patients only, limiting the generalizability of their findings given the higher prevalence of intracranial atherosclerotic disease in Asian than in Western populations.

“Importantly, none of these three trials sought to demonstrate noninferiority in the strongest sense of formally excluding the minimal clinically important difference (MCID),” Saver and Adeoye added. They noted that an MCID trial would need an “infeasibly large” sample size to show “true indistinguishability” in the present setting.

“Nonetheless, taken together, the accumulated results from these studies suggest that the simpler EVT alone strategy is broadly noninferior to combined EVT and IVT [IV thrombolysis] and accordingly may be reasonable to consider for patients who present directly to thrombectomy-capable centers,” the editorialists said.

The duo offered two points of caution: that patients with small-to-medium vessel occlusions are not eligible for endovascular therapy, and that people with large vessel occlusions should only skip IV tPA if they are sure to receive thrombectomy quickly and reliably.