Background: Most ischemic stroke patients have at least partial collateral filling of the ischemic territory or grade 2 collaterals, yet the temporal delay or limited spatial extent may vary considerably. We investigated the clinical implications of grade 2 (partial) versus grade 3 (complete) collaterals in the IMS III trial.
Methods: The angiography core lab in IMS III prospectively evaluated collateral circulation utilizing the ASITN/SIR definitions of grades 2 and 3. The clinical features, serial imaging findings, revascularization results and outcomes were contrasted.
Results: Of 203 M1 or intracranial ICA occlusions, ASITN/SIR collateral grades could be scored in 161 (79%) cases, including grade 0 or none (n=7), grade 1 or marginal (n=23), grade 2 or partial (n=62), grade 3 or complete but delayed (n=62) and grade 4 or early and complete (n=7) collaterals. History of hypertension (p=0.02) was more common in those with partial collaterals, whereas age, sex and other co-morbidities were similar. Initial stroke severity defined by baseline NIHSS was worse in those with partial versus complete collaterals (median 19 IQR 16-21 vs. median 17 IQR 13-20; p=0.02). Baseline ASPECTS was lower in partial versus complete collaterals (median 7 IQR 5-9 vs. median 8 IQR 7-10; p=0.04). Successful reperfusion (mTICI 2B-3) was markedly less frequent (16% vs. 44%, p<0.001), despite equivalent recanalization of primary vessel occlusions (AOL 2-3) (49% vs. 54%, p=0.34). ASPECTS at 24 hours was much lower in partial versus complete collaterals (median 3 IQR 1-5 vs. median 6 IQR 3-7; p<0.001), with greater ASPECTS decline (median 4 IQR 2-5 vs. median 2 IQR 1-4; p=0.002) even after successful reperfusion (median 3.5 IQR 1-4.5 vs. median 2 IQR 1-4; p=0.22). 90-day mRS was notably worse in those with partial collaterals (median 4 IQR 3-6 vs. median 2 IQR 1-4; p<0.001) with fewer good clinical outcomes (12% mRS 0-2 vs. 32%, p<0.001) and greater mortality (19% vs. 5%, p=0.003).
Conclusions: Grade 2 or partial collaterals have radically distinct imaging and clinical implications in the setting of endovascular stroke therapy.