ASPECTS as a surrogate marker of core-perfusion mismatch in late-window large vessel occlusion stroke

In acute stroke due to a large vessel occlusion (AIS-LVO), identifying a core-perfusion mismatch guides reperfusion therapy in extended time window, but perfusion imaging is not universally available. We investigated whether the Alberta Stroke Program Early CT Score (ASPECTS) is associated with core-perfusion mismatch and could serve as a surrogate imaging marker.

We retrospectively analysed a large multicentre international unselected cohort of consecutive AIS-LVO patients imaged between 4.5 and 24 h from last time seen well, with baseline CT- or MR-perfusion imaging. Core-perfusion mismatch was defined as a mismatch volume ≥ 15 mL and a mismatch ratio ≥ 1.8 using automated software. CT and MRI cohorts were analysed separately.

Among 531 included patients, 182 underwent CT and 349 MRI. Core-perfusion mismatch was present in 86% of CT patients and 62% of MRI patients. Alberta Stroke Program Early CT Score was strongly associated with mismatch in both cohorts (P < .001). In the MRI cohort, ASPECTS predicted mismatch with an area under the receiver operating characteristic curve (AUC) of 0.843 (95% CI, 0.797-0.890), with an optimal cutoff ≥ 6, yielding sensitivity 0.87, specificity 0.72, positive predictive value (PPV) 0.84 and negative predictive value (NPV) 0.77. In the CT cohort, the AUC was 0.801 (95% CI, 0.698-0.904), with an optimal cutoff ≥ 7, yielding sensitivity 0.74, specificity 0.81, PPV 0.96 and NPV 0.34.

The high PPV of ASPECTS ≥ 6 on MRI and ASPECTS ≥ 7 on CT supports their potential use as pragmatic enrichment criteria for patient selection in future late-window AIS-LVO trials, when perfusion imaging is unavailable, as surrogate markers of salvageable tissue.