Thrombectomy for Acute Stroke Works Across Subgroups, but Best When Done Early
LOS ANGELES—The benefits of endovascular treatment for acute ischemic stroke extend across several patient subsets, including older patients, late presenters, and those ineligible for IV thrombolysis, but the impact on functional outcomes is greatest when interventions are performed as soon as possible after symptom onset, according to 2 meta-analyses presented at the International Stroke Conference held here last week.
Both studies came out of the HERMES collaboration, which pools patient-level data from 5 recently published endovascular trials: MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT. All of the trials showed that adding endovascular therapy—mostly with stent retrievers—to medical therapy that usually included IV thrombolysis improved functional outcomes at 90 days compared with medical therapy alone in patients with large vessel occlusions in the anterior circulation.
“I think people are pretty much of the belief that these treatments are generalizable,” Wade Smith, MD, PhD, of the University of California, San Francisco, who was not involved in either analysis, commented to TCTMD. “Now that you can put all the data together, you can get much more accurate predictions of how much patients will benefit using those numbers.”
That increased precision can then provide the opportunity to do cost-effectiveness analyses and to better advise patients’ families about the benefits of the procedure, he added.
Consistent Treatment Effects
To illustrate the collaborative nature of the HERMES group, Michael Hill, MD, of the University of Calgary (Calgary, Canada), and Tudor Jovin, MD, of the University of Pittsburgh Medical Center Stroke Institute (Pittsburgh, PA), jointly presented the results of the subgroup meta-analysis, which was published simultaneously online in The Lancet.
The 5 trials included a total of 1,287 patients, with baseline characteristics well balanced between the endovascular and control arms. Most occlusions (71%) were in the M1 segment of the middle cerebral artery, the median ASPECTS score was 8.3, and the median NIH Stroke Scale (NIHSS) score was 17. The rate of IV tPA use was slightly higher in the control arm (87.1% vs 83.0%).
The pooled effect of adding endovascular therapy on the shift in modified Rankin Scale (mRS) score at 90 days was expressed as a common odds ratio, which was 2.49 (95% CI 1.76-3.53). That equated to a number needed to treat (NNT) of 2.6. Similar relationships were observed for the proportion of patients achieving the following outcomes: mRS score of 0-1 or 0-2 at 90 days, NIHSS score of 0-2 at 24 hours, and early neurological recovery at 24 hours.
The subgroup analyses showed that the treatment effect was not modified by age, sex, occlusion location, ASPECTS or NIHSS score, treatment with tPA, or time from randomization. There was, however, a nonsignificant trend (P = .07 for interaction) toward a less robust benefit in young patients and a larger effect in older patients.
“We can say with a high degree of certainty that endovascular treatment is a highly effective therapy across all subgroups,” Jovin said during his presentation, urging caution in interpreting the findings in subgroups with small patient numbers. “These data may provide additional support for endovascular treatment in subgroups not addressed in individual trials.”
Time Is Brain
In the second meta-analysis, Mayank Goyal, MD, also of the University of Calgary, and Jeffrey Saver, MD, of the University of California, Los Angeles, took a closer look at the impact of time from symptom onset to treatment.
Across the 5 trials, patients were enrolled relatively shortly after stroke onset, with average times ranging from 2 hours 48 minutes to 3 hours 45 minutes.
Baseline characteristics of the patients varied somewhat based on the interval between onset and randomization. Those enrolled within 2 hours, for example, tended to be older, to be directly admitted rather than transferred from a different center, to have an occlusion in the internal carotid artery, and to have a higher ASPECTS score.
Looking at workflow speeds in the endovascular arm, median intervals were 104 minutes from door to puncture at the start of the procedure, 76 minutes from imaging to puncture, 238 minutes from onset to puncture, and 285 minutes from onset to reperfusion. Overall time to reperfusion was longer in transferred patients because of the extended time from onset to the emergency department.
The analysis showed that the likelihood of achieving a particular mRS score in the endovascular arm declined as the time from onset to randomization increased, although such a relationship was not seen in the control group. The benefit of endovascular therapy relative to control was clear through about the first 6 hours from onset to randomization, with an advantage likely seen in patients randomized between 6 and 8 hours, Goyal said.
There was also a relationship between treatment effect and the time from onset to substantial reperfusion in the endovascular arm (defined as modified TICI 2b/3 flow). For example, about 43% of patients achieved an mRS score of 0-1 at 90 days if they achieved reperfusion at 2 hours, with the rate dropping to about 23% in patients achieving reperfusion at 8 hours.
Time to reperfusion did not interact with either mortality or symptomatic hemorrhage, however.
“These data can hopefully be utilized to support further intensive quality improvement and set time lines of expectation of workflow,” Goyal said. “And essentially, once again it provides compelling evidence of ‘time is brain.’”
Some Uncertainty Remains
In an editorial accompanying the Lancet paper, Keith Muir, MD, of the University of Glasgow (Glasgow, Scotland), and Philip White, MD, of the University of Newcastle (Newcastle-upon-Tyne, England), said the NNT of 2.6 observed for the overall treatment effect “is extraordinary in itself and a huge advance in care for patients with stroke caused by occlusion of the major intracranial carotid territory vessels, who often respond poorly to optimal medical care with intravenous thrombolytic drugs and have high risk of mortality or disability.”
They note, however, that the results of these trials may not be broadly generalizable in routine practice. “Replicating the highly efficient networks of care in these trials, exemplified by the very short median time from symptom onset to intravenous thrombolysis of 100 min, will be a major challenge for many healthcare systems,” they write.
Muir and White also voice some caution about the interpretation of results from subgroups.
“In subgroup analyses, lack of statistical heterogeneity is not the same as evidence of efficacy when many major subgroups are so small; implementation of endovascular thrombectomy in circumstances such as late presentation, extensive early ischemic change on CT, or ineligibility for [IV tPA] should be done with caution, and ideally supported in due course by further trial evidence,” they argue.
Additional areas of uncertainty, they say, include the benefit of thrombectomy in more distal anterior circulation vessels and in the posterior circulation, whether general anesthesia or sedation is preferred, and the roles for aspiration techniques, balloon guiding catheters, adjunctive stenting of coexisting carotid stenosis, and additional intraarterial thrombolytic drugs.
“Nonetheless, the efficacy of endovascular thrombectomy in patients with anterior circulation stroke due to large artery occlusion is established, and implementation should be the priority,” they conclude. “Further trials are needed to clarify the boundaries of treatment.”
Goyal M, Menon BK, van Zwam WH, et al. Endovascular thrombectomy after large-vessel ischaemic stroke: a meta-analysis of individual patient data from five randomized trials. Lancet. 2016;Epub ahead of print.
Saver JL, Goyal M, van der Lugt A, et al. Time is brain in endovascular thrombectomy: results from the individual patient pooled data analysis of MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, and REVASCAT. Presented at: International Stroke Conference; February 17, 2016; Los Angeles, CA.
Muir KW, White P. HERMES: messenger for stroke interventional treatment. Lancet. 2016;Epub ahead of print.
- Medtronic funded the HERMES collaboration through an unrestricted grant to the University of Calgary.
- Goyal reports receiving grants and personal fees from Covidien during the conduct of the study and having a patent for systems and methods for diagnosing strokes licensed to GE Healthcare.
- Saver reports having served as an unpaid site investigator in trials run by Medtronic and Stryker for which the University of California Regents received payments on the basis of clinical trial contracts for the number of subjects enrolled and having relationships with multiple other pharmaceutical and device companies. The University of California has released the Rankin Focused Assessment for free use under a Creative Commons license, and has copyright for Rankin Scale training vignettes. The University of California has patent rights in retrieval devices for stroke.
- Hill reports receiving unrestricted grant funding for the ESCAPE trial to the University of Calgary from Covidien (Medtronic), and active/in-kind support from a consortium of public and charitable sources (Alberta Health Services, Alberta Innovates Health Solutions, Heart & Stroke Foundation) and the University of Calgary (Calgary Stroke Program, and Hotchkiss Brain Institute, Departments of Clinical Neurosciences and Radiology); personal fees from Merck; and nonfinancial support from Hoffmann-La Roche Canada. He has patents pending to the US Patent Office for systems and methods for assisting in decision-making and triaging for acute stroke patients (number 62/086,077) and owns stock in Calgary Scientific, a company that focuses on medical imaging software.
- Jovin reports having consulted for Codman Neurovascular and Neuravi; holding stock in Blockade and Silk Road; having acted as an unpaid consultant to Stryker as principal investigator of the DAWN trial; and having served as an unpaid member of a Medtronic advisory board.
- Muir and White make no statements regarding conflicts of interest.
- Smith reports receiving compensation for serving on an independent data and safety monitoring board for an ongoing trial by Stryker.