Angioplasty and stenting for mechanical thrombectomy in acute ischemic stroke

Time window.

It has been well-established that time to recanalization is one of the most important predictors for improved mortality and morbidity in stroke patients.¹⁰ Since the National Institute of Neurological Disorders and Stroke (NINDS) trial, IV tPA has been the cornerstone of therapy in AIS.¹⁰ However, even after 15 years of implementation, few AIS patients receive IV tPA.¹¹ The recent extension of the time window for administering IV tPA to 4.5 hours will possibly help in increasing the access to this therapy,¹² but the clinical impact of this extended time window still remains to be seen. Following the NINDS trial, several IA trials (Prolyse in Acute Cerebral Thromboembolism [PROACT] I and II)^2,3 and trials of combined IA and IV thrombolytics (Interventional Management of Stroke [IMS I and II])⁴ were conducted, extending the time window of ischemic stroke treatment to 6 hours. Median time to therapy for PROACT II was 5.3 hours, whereas mean time to therapy for the IMS I trial was 217 minutes, or 3 hours and 37 minutes.¹³ Thereafter, mechanical means of recanalization for acute cerebrovascular occlusions such as the Merci retriever^5,6 and the Penumbra^7,13 systems further extended the time window for recanalization to 8 hours for the anterior circulation. Median time to recanalization for the pooled data of Mechanical Embolus Removal in Cerebral Ischemia (MERCI) and Multi-MERCI trials was reported to be 4.3 hours for a total of 305 patients.^5,–,7,13 In our pooled data for 127 patients, the mean time to recanalization was 320 minutes, or 5 hours and 20 minutes, putting stenting well within the range of the 8-hour window. Current studies are trying to extend this time window for a certain subset of patients, based on the advanced imaging protocol.¹⁴ As the science of perfusion imaging advances, these windows are bound to advance, which will significantly improve access to thrombectomy therapy for a significant number of stroke patients.

Recanalization rates.

Following an acute stroke, recanalization along with the presenting NIHSS score is the best predictor for a good outcome.¹⁵ Even though IV tPA is the primary mode of therapy for strokes presenting in the first 4.5 hours, recanalization rates with IV thrombolytics have been reported to be only about 10% in proximal internal carotid artery occlusions and 30% in middle cerebral artery occlusions.¹³ The use of IA therapy in PROACT I and II advanced these rates to 57.7% for middle cerebral artery M1 and M2 lesions.^7,13,16 However, the use of IA thrombolytics in PROACT I and II was limited to only middle cerebral artery lesions.^2,3 Combination therapy in IMS I and II achieved rates of thrombolysis in cerebral infarction grade 2–3 in 61.3% of cases.^7,13,16 Use of the Merci retriever with adjunctive therapy further advanced the recanalization rates to 64.6%, and even in lesions of the proximal internal carotid artery the rates were 62.6%. A major gain in the recanalization rates was shown in the use of the Penumbra system, where rates of 81.6% were observed with the adjunctive use of IA thrombolytic therapy. In our analysis, 89% of the lesions treated had a recanalization TIMI grade of 2–3, and in studies published after 2008, this rate was as high as 92.5%, suggesting the possibility that stenting might be superior to other methods of endovascular recanalization.

Postprocedural morbidity and mortality.

When compared to the placebo group with a symptomatic intracranial hemorrhage rate of 0.6%, patients receiving IV tPA had a hemorrhage rate of 6.4%.¹⁰ The rate following use of IA prourokinase in the PROACT II trial was 10.9% in the treatment group,^3,13 whereas in the combined groups of the IMS I and II trials the rates were 6.3% to 9.9%.⁴ Data from the MERCI and Multi-MERCI trials had symptomatic intracranial hemorrhage rates of 7.8% to 9.8%, and this rate increased to 11.2% with use of the Penumbra Aspiration System.¹³ In the current pooled analysis we found that this rate was 9.4%.

Another important outcome measure reported in these studies was mortality. The mortality at 3 months for the NINDS placebo group was 24%.¹⁶ The mortality rates for various chemical thrombolysis groups, namely, IV tPA, PROACT I, PROACT II, and the IMS I and II trial groups, were 21%, 26.9%, 25%, and 16%, respectively.^7,13,16 For the mechanical means the data were reported to be 43.5%, 34%, and 32.8% for the MERCI, Multi-MERCI, and Penumbra systems, respectively.¹³ Mortality in our study group was 30%, comparable to that with other mechanical methods of recanalization.

Favorable clinical outcome.

The studies we included did not all report the same outcome variable for clinical outcome, which made comparison to data from other studies difficult. Also, a good functional outcome in most of these studies was defined by an mRS score of ≤3, whereas in all studies including NINDS it was defined by an mRS score ≤2. Three studies reported data on mRS scores ≤3 only at discharge or 30 days, and in these studies this outcome was achieved in 49% of patients. Three other studies reported data for the 3-month follow-up, and their pooled results showed an mRS score ≤3 in 41% of patients. One study reported only a change in NIHSS score by ≥4 points at time of discharge as the good functional outcome measure, and this finding was seen in 75% of cases in that study. Whereas in the NINDS trial only 33% of patients had a good functional outcome (mRS scores ≤2),¹⁰ in the IA trials (PROACT and IMS), good functional outcome (mRS scores ≤2) was achieved in 40% to 46% of patients.^7,16 Rates from the Multi-MERCI and Penumbra trials were lower; good outcomes (mRS scores ≤2) were seen in only 36% and 25% of patients, respectively.⁷

The only available prospective trial in our series, the SARIS study by Levy at al.,¹⁷ reveals that a good outcome was noted in 60% of patients. In a recent article, the study investigators reported 6-month outcome data. A favorable outcome of mRS score ≤3 was seen in 60% of patients, and an mRS score ≤2 in 55%; mortality at the end of 6 months was 35%.¹⁸

Stenting beyond 8 hours (subacute).

Recent reports have shown that endovascular therapy including stenting can be done in patients presenting with AIS beyond 8 hours chosen carefully on the basis of diffusion-perfusion–based protocols.^14,19 These are based on reports that there is salvageable penumbra sustained by collateral support. The recanalization rates reported vary between 90% and 100%, higher than with most mechanical or chemical methods.^20,21 Favorable outcomes (mRS score <3) were observed in 50% to 66% at 3 months, with postprocedure intracranial hemorrhage/subarachnoid hemorrhage rates of 10% to 25%.^20,21 It is possible that the lesions treated had an underlying atherosclerotic nidus and that stenting helped in stabilizing the endothelium and providing quick and sustained recanalization. Thus, it appears that stenting can be done beyond 8 hours with reliable safety and efficacy in a carefully selected population of patients (figures 1 and 2).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1 A 71-year-old man presented with dysarthria, imbalance, and weakness

NIH Stroke Scale score was 12 on presentation. Angiography showed trickle flow in distal basilar artery, with occluded distal vertebral artery (A). Revascularization was done with use of 3.5 × 12–mm multilink vision and poststenting thrombolysis in cerebral infarction flow was restored to grade 3 (B). Modified Rankin Scale score was 0 at 3 months.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2 A 42-year-old man presented with right upper- and lower-extremity weakness

NIH Stroke Scale score was 9. Angiography revealed complete occlusion of left middle cerebral artery M1 (A). Poststenting imaging revealed thrombolysis in cerebral infarction 3 flow grade restoration (B). Modified Rankin Scale score was 0 at discharge.

Limitations.

There are several important limitations in this review. Most of the studies included are retrospective case series with few patients, except one. We excluded the articles and series where stenting was used as part of multimodal therapy but the role of stenting/angioplasty was not specified separately. Even in the articles included in our analysis, stents were used after multiple chemical and mechanical therapies, which makes it difficult to analyze the solitary impact of stenting. Another limitation was that outcome measures defined by these studies were often different, making it difficult to compare these data with other, historical study data. Most of the lesion treatments with stents were done as salvage therapy. This introduces a selection bias in choosing lesions that would be less amenable to other conventional therapies, thus influencing the radiologic and possibly the clinical outcome. In addition, multiple types of stents with different mechanical properties were included, which in turn might have influenced the rates of recanalization and hemorrhage independently.