MR RESCUE
A Trial of Imaging Selection and Endovascular Treatment for Ischemic Stroke
Year of Publication: 2013
Authors: Chelsea S. Kidwell, Reza Jahan, Jeffrey Gornbein, et al.
Journal: New England Journal of Medicine
Citation: Kidwell CS, Jahan R, Gornbein J, et al. A Trial of Imaging Selection and Endovascular Treatment for Ischemic Stroke. N Engl J Med. 2013;368:914–923.
PDF: Clinical Question
Does neuroimaging-based penumbral selection identify patients with acute ischemic stroke who differentially benefit from endovascular therapy?
Study Overview
Objective
To evaluate whether neuroimaging selection based on penumbral patterns identifies patients who benefit from endovascular therapy for acute ischemic stroke.
Study Summary
This phase 2b trial found no significant benefit of mechanical embolectomy over standard care, regardless of whether patients had favorable or non-favorable penumbral imaging patterns..
Intervention
Mechanical embolectomy using Merci Retriever or Penumbra system vs. standard medical care. Neuroimaging used to classify patients by penumbral pattern.
Patients per Arm
Embolectomy: 64, Standard care: 54
Bottom Line
Endovascular therapy with first-generation devices did not show benefit over standard care in patients with large-vessel anterior-circulation ischemic stroke within 8 hours, and a favorable penumbral imaging pattern did not identify patients who differentially benefited from embolectomy.
Major Points
- MR RESCUE assessed whether imaging-defined penumbra could identify patients who differentially benefit from endovascular therapy.
- Randomized patients within 8 hours of symptom onset to mechanical embolectomy (Merci Retriever or Penumbra System) vs standard medical care.
- Used multimodal MRI or CT to stratify patients as having favorable penumbral vs nonpenumbral pattern before randomization.
- No significant interaction between imaging pattern and treatment on 90-day mRS (P=0.14).
- Trial failed to demonstrate benefit of endovascular therapy overall (mean 90-day mRS 3.9 vs 3.9, P=0.99), possibly due to first-generation devices and a low rate of substantial revascularization.
- Laid groundwork for better patient selection methods in later trials (e.g., DEFUSE 3, DAWN).
Design
Study Type: Phase 2b, multicenter, randomized, controlled, open-label (blinded outcome) trial with imaging stratification
Randomization: 1
Blinding: Open-label with blinded outcome assessment
Enrollment Period: 2004 – 2011
Follow-up Duration: 90 days
Centers: 22
Countries: North America
Sample Size: 118
Analysis: Nonparametric two-way analysis of variance using permutational methods, with prespecified interaction test between penumbral pattern and treatment assignment on 90-day mRS
Inclusion Criteria
- Acute ischemic stroke due to anterior circulation large vessel occlusion (ICA or MCA)
- Age 18–85
- NIHSS 6–29
- Randomization within 8 hours of symptom onset
- Pretreatment multimodal CT or MRI of the brain
Exclusion Criteria
- Per Figure 1, 9 of 127 randomized patients were excluded from the primary analyses: 5 did not have a target lesion on vessel imaging, 2 did not have post–t-PA vessel imaging, and 2 had failed perfusion imaging. Additional prespecified exclusion criteria are not enumerated in the primary paper (see Supplementary Appendix).
Baseline Characteristics
| Characteristic | Control | Active |
|---|---|---|
| Age (mean) | 67.1 | 64.2 |
| Female (%) | 50% | 53% |
| NIHSS (median) | 16 (penumbral) / 20.5 (nonpenumbral); overall all-patient median 17 (IQR 13–21) — arm-aggregate median not reported in paper | 16 (penumbral) / 19 (nonpenumbral); overall all-patient median 17 (IQR 13–21) — arm-aggregate median not reported in paper |
| Favorable penumbral pattern (%) | 63% | 53% |
| IV tPA received (%) | 30% | 44% |
Arms
| Field | Endovascular Therapy | Control |
|---|---|---|
| Intervention | Mechanical embolectomy using Merci Retriever or Penumbra System (intraarterial t-PA up to 14 mg allowed as rescue within 6 hours) | Supportive care and IV tPA if eligible; no endovascular therapy |
| Duration | Single procedure within 8 hours of onset; 90-day follow-up | 90-day follow-up |
Outcomes
| Outcome | Type | Control | Intervention | HR / OR / RR | P-value |
|---|---|---|---|---|---|
| 90-day modified Rankin Scale (mRS) distribution; primary test was interaction between penumbral pattern and treatment assignment | Primary | Mean mRS 3.9 | Mean mRS 3.9 | 0.99 (overall treatment comparison); interaction P=0.14 | |
| Revascularization (TICI 2a–3) in embolectomy group | Secondary | NA | 67% | ||
| Mortality at 90 days (paper reports 4 subgroups: Embolectomy-Penumbral 6/34 [18%], Standard-Penumbral 7/34 [21%], Embolectomy-Nonpenumbral 6/30 [20%], Standard-Nonpenumbral 6/20 [30%]) | Secondary | 24% (13/54) | 19% (12/64) | 0.75 (4-group overall comparison per Table 2 footnote; no arm-aggregate P reported) | |
| Symptomatic intracerebral hemorrhage (paper reports 4 subgroups: 3 [9%], 2 [6%], 0, 0) | Secondary | 4% (2/54) | 5% (3/64) | 0.24 (4-group overall comparison per Table 2 footnote; no arm-aggregate P reported) | |
| Good outcome (mRS 0–2) at 90 days (paper reports 4 subgroups: 7 [21%], 9 [26%], 5 [17%], 2 [10%]) | Secondary | 20% (11/54) | 19% (12/64) | 0.48 (4-group overall comparison per Table 2 footnote; no arm-aggregate P reported) | |
| Symptomatic ICH | Adverse | 4% (2/54) | 5% (3/64) | 0.24 (4-group overall comparison per Table 2 footnote; no arm-aggregate P reported) | |
| Mortality | Adverse | 24% (13/54) | 19% (12/64) | 0.75 (4-group overall comparison per Table 2 footnote; no arm-aggregate P reported) | |
| Asymptomatic ICH | Adverse | 48% (26/54) | 66% (42/64) | 0.04 (4-group overall comparison per Table 2 footnote; no arm-aggregate P reported) |
Subgroup Analysis
No significant interaction between imaging pattern and treatment (P=0.14). Favorable penumbral: mean mRS 3.9 (embolectomy) vs 3.4 (standard care), P=0.23. Nonpenumbral: 4.0 vs 4.4, P=0.32.
Criticisms
- Small sample size and underpowered
- First-generation endovascular devices (Merci, Penumbra) with low rate of substantial revascularization (paper distinguishes reperfusion 37–57% from revascularization/TICI 2a–3 67–93% across subgroups)
- Long enrollment period (2004–2011) limited applicability to modern practice
- Long time from imaging to embolectomy (groin puncture >6 hours from symptom onset)
- Heterogeneity of imaging modalities (both MRI and CT)
- Real-time penumbral processing succeeded in only 58% of cases
Funding
National Institute of Neurological Disorders and Stroke (NINDS) grant P50 NS044378; Concentric Medical provided devices until August 2007
Based on: MR RESCUE (New England Journal of Medicine, 2013)
Authors: Chelsea S. Kidwell, Reza Jahan, Jeffrey Gornbein, et al.
Citation: Kidwell CS, Jahan R, Gornbein J, et al. A Trial of Imaging Selection and Endovascular Treatment for Ischemic Stroke. N Engl J Med. 2013;368:914–923.
Content summarized and formatted by NeuroTrials.ai.