DiscussionAlthough each DWI-ASPECTS point corresponded to a wide range of Vol DWI , all patients with extensive changes on DWI-ASPECTS (0-3) had large Vol DWI , whereas all patients with limited DWI-ASPECTS changes (≥7) had Vol DWI <70 mL.DWI-ASPECTS is increasingly used for description or prognostic purposes in stroke populations. Although not designed to substitute for Vol DWI , DWI-ASPECTS does provide some semiquantitative estimate of it. However, DWI-ASPECTS overlooks lesions within the striatocapsular region and only partially covers the middle cerebral artery territory. This explains the wide range of true lesion volumes for a given DWI-ASPECTS point found here, in line with other studies. 2,7 Our finding that DWI-ASPECTS <4 invariably predicted Vol DWI ≥93 mL is entirely consistent with 1 previous report 7 and highly relevant to the Diffusion and perfusion imaging Evaluation For Understanding Stroke Evolution (DEFUSE)-2 malignant profile 100 mL cut point. 4 However, patients with DWI-ASPECTS ≥7 all had Vol DWI <70 mL, which corresponds to the cut point incorporated in the target mismatch definition.4 Although debated, 11 these volume cut-offs are proposed to identify poor or good responders to reperfusion therapy, 4 and particularly the 100-mL cut point serves as an exclusion criterion in several ongoing trials. However, fully automated softwares to calculate Vol DWI are not yet commonly used and can fail in real time. This may lead to imbalanced groups on baseline characteristics in trials where randomization is based on automated MR-image segmentation.14 Failure of automated volumetry may also restrain patient's inclusion in trials. To overcome these difficulties, DWI-ASPECTS <4 could replace the poorly reproducible greater than one third of the middle cerebral artery territory CT rule as an alternative exclusion criterion in MR-based trials.The tight relationships between extreme DWI-ASPECTS values (ie, <4 or ≥7) and the >100-or <70-mL, respectively, cut points found here suggest that DWI-ASPECTS could serve as a surrogate for these volumes. This concerned almost 3 quarters (241/330) of the studied population and may have clinical relevance. However, in those patients with intermediate DWI-ASPECTS (4-6), Vol DWI straddled widely across the above cut point volumes, indicating that intermediate DWI-ASPECTS cannot substitute for Vol DWI to identify patients with target mismatch or malignant profile. Of note, no DWI-ASPECTS cut point identified lesion volume >145 mL, 15 above witch decompressive hemicraniectomy is indicated.Limitations of our study include its retrospective and single-center nature, and the focus on patients who underwent thrombolysis, which limits generalizability to nonthrombolized patients and may, in part, explain the low proportion of patients with large Vol DWI and consequently the relative large 95% confidence interval for patients with low ASPECTS.In conclusion, in the first 6 hours, each DWI-ASPECTS point corresponds to a wide range of Vol DWI . However, extreme DWI-ASPECTS scor...