MR Imaging versus Noncontrast CT for Selecting Patients with Acute Ischemic Stroke of Large Vessel Occlusion for Endovascular Thrombectomy: A Systematic Review and Meta-analysis.

Background: Neuroimaging in the acute phases after the onset of stroke symptoms is necessary to determine large vessel occlusion presence as well as the extent of the ischemic insult before deeming eligibility for endovascular thrombectomy (EVT).

Purpose: To evaluate the clinical outcomes in patients with acute ischemic stroke selected for EVT based on initial imaging; NCCT compared with those selected by using MRI.

Data Sources: PubMed, Embase, Scopus, and Web of Science were searched from inception until August 2024.

Study Selection: We included observational studies comparing functional independence (mRS 0-2), successful reperfusion (TICI 2b-3), symptomatic intracerebral hemorrhage (sICH), or mortality in patients selected for EVT by using NCCT±CT angiography versus MRI±MR angiography. We excluded studies that used perfusion imaging in their patient selection for EVT.

Data Analysis: Data were pooled by using a random-effects model, and heterogeneity was assessed using I statistics. A subgroup analysis was performed to determine the effect of the treatment window (<6 hours versus >6 hours from last known well). The quality of eligible studies was assessed using Newcastle Ottawa Scale.

Data Synthesis: Seven studies ( = 3940 patients) met the inclusion criteria. Two studies had a low risk of bias, and others had some concerns. Patients with MRI selection showed better chances of functional independence (OR: 1.85; 95% CI: 1.28-2.67; < .01; I = 45%), lower rates of sICH (OR: 0.59; 95% CI: 0.39-0.89; = .01; I = 0%), reduced 90 day mortality (OR: 0.63; 95% CI: 0.51-0.78; < .01; I = 0%), and no difference in successful reperfusion (OR: 0.99; 95% CI: 0.62-1.58; = .95; I = 0%) compared with NCCT in patients treated within 6 hours of stroke onset. There were no significant differences in any end points between MRI and NCCT for patients treated beyond 6 hours.

Limitations: Our meta-analysis comprised only observational studies, with different populations and imaging protocols limiting the strength of the conclusions.

Conclusions: Within the crucial <6-hour window, the MRI's superior patient selection justifies its use despite longer acquisition times. Beyond 6 hours, the focus should shift to rapid EVT access rather than imaging technique choice as the benefits of MRI diminish.