Prognostic value of recanalization attempts in endovascular therapy for M2 segment middle cerebral artery occlusions

Abstract

Background: There is growing evidence suggesting efficacy of endovascular therapy for M2 occlusions of the middle cerebral artery. More than one recanalization attempt is often required to achieve successful reperfusion in M2 occlusions, associated with general concerns about the safety of multiple maneuvers in these medium vessel occlusions.

Aim: The aim of this study was to investigate the association between the number of recanalization attempts and functional outcomes in M2 occlusions in comparison with large vessel occlusions (LVO).

Methods: Retrospective multicenter cohort study of patients who underwent endovascular therapy for primary M2 occlusions. Patients were enrolled in the German Stroke Registry at 1 of 25 comprehensive stroke centers between 2015 and 2021. The study cohort was subdivided into patients with unsuccessful reperfusion (mTICI 0-2a) and successful reperfusion (mTICI 2b-3) at first, second, third, fourth, or ⩾fifth recanalization attempt. Primary outcome was 90-day functional independence defined as modified Rankin Scale score of 0-2. Safety outcome was the occurrence of symptomatic intracranial hemorrhage. Internal carotid artery or M1 occlusions were defined as LVO and served as comparison group.

Results: A total of 1078 patients with M2 occlusion were included. Successful reperfusion was observed in 87.1% and 90-day functional independence in 51.9%. The rate of functional independence decreased gradually with increasing number of recanalization attempts (p < 0.001). In both M2 occlusions and LVO, successful reperfusion within three attempts was associated with greater odds of functional independence, while success at ⩾fourth attempt was not. Patients with ⩾4 attempts exhibited higher rates of symptomatic intracranial hemorrhage in M2 occlusions (6.5% vs 2.7%, p = 0.02) and LVO (7.2% vs 3.5%, p < 0.001).

Conclusion: This study suggests a clinical benefit of successful reperfusion within three recanalization attempts in endovascular therapy for M2 occlusions, which was similar in LVO. Our findings reduce concerns about the risk-benefit ratio of multiple attempts in M2 medium vessel occlusions.

Data access statement: The data that support the findings of this study are available on reasonable request after approval of the German Stroke Registry (GSR) steering committee.

Clinical trial registration information: ClinicalTrials.gov Identifier: NCT03356392.

Keywords: Cerebral infarction; infarction; ischemic stroke; stroke; thrombectomy.

Figure 1.

Association between number of recanalization attempts, successful mechanical reperfusion, and functional outcomes in M2 segment occlusions. (a) Cumulative rates of successful reperfusion (mTICI 2b–3; black line) and functional independence at 90 days (mRS score 0–2; blue line) stratified by the number of recanalization attempts. The cumulative rate of functional independence at 90 days did not increase substantially with four or more attempts. (b) Distribution of mRS scores at 90 days stratified by the number of recanalization attempts to achieve successful reperfusion. The rate of functional independence decreased gradually when successful reperfusion was achieved with an increasing number of recanalization attempts (dotted line) and was the lowest in patients with unsuccessful reperfusion (mTICI 0–2a).

Figure 2.

Predicted probabilities of functional independence at 90 days from multivariable logistic regression analyses. Model-based estimates of averaged marginal probabilities from the multivariable logistic regression models to predict functional independence at 90 days for (a) M2 occlusions and (b) for intracranial ICA/proximal M1 occlusions. A higher number of recanalization attempts required for successful mTICI 2b–3 reperfusion decreased the predicted probability of functional independence at 90 days in both ICA/M1 and M2 occlusions. In both regression models, only successful reperfusion within three attempts was associated with greater odds of achieving functional independence at 90 days compared with unsuccessful mTICI 0–2a reperfusion (indicated by asterisks; see Supplemental Tables S2 and S5). Please note that compared with unsuccessful reperfusion, successful reperfusion at fourth or ⩾fifth attempt showed smaller point estimates in M2 occlusions, but greater point estimates in ICA/M1 occlusions (red dotted line). Results were adjusted for age, sex, admission NIHSS, baseline ASPECTS, treatment with intravenous thrombolysis, general anesthesia, procedure time, and occurrence of sICH. Error bars indicate the 95% CI.

Figure 3.

Occurrence of sICH stratified by the number of recanalization attempts. Frequency of sICH stratified by occlusion location and the number of recanalization attempts irrespective of the final mTICI grade. The rates of sICH increased gradually with an increasing number of recanalization attempts and were compared between M2 and ICA/M1 occlusions. Consistently, the rate of sICH was significantly higher in patients with more than three attempts during endovascular therapy in M2 occlusions (2.7% vs 6.5%; p = 0.02) and ICA/M1 occlusions (3.5% vs 7.2%; p < 0.001).