A new era of the treatment for acute ischemic stroke (AIS) patients has opened with the development in neurointerventional technique such as endovascular thrombectomy (EVT) using a stent-retriever technique. In a number of prospective, randomized trials, the clinical outcome of selected AIS patients with large vessel occlusion in anterior circulation improved with EVT, as long as 24 h after stroke onset [1,2,3].

Despite the safety and effectiveness of EVT verified across the several randomized clinical trials, the rate of functional independence after EVT has been reported to be approximately 50% [4]. Given the varying clinical outcome of the patients following EVT, imaging-based prognostic biomarkers to accurately predict the clinical outcome have also been of growing interest over the past few years. Infarction volume on follow-up imaging would be the most intuitive parameter; and thus, several previous studies have investigated the parameter and proven its close association with the clinical outcome [5, 6]. However, the labor-intensive nature of manual delineation and the suboptimal performance of automated software in complicated cases with hemorrhagic transformation have been acknowledged as the major drawbacks in using the parameter.

Alberta Stroke Program Early Computed Tomography Score (ASPECTS) is a simple and reliable semi-quantitative grading system, which was first developed based on pre-treatment non-contrast CT images to assess the extent of early ischemic changes in the middle cerebral artery territory [7]. Subsequently, the scoring system has been extended to diffusion-weighted imaging (DWI) and Yoshimoto et al have demonstrated that pre-treatment DWI ASPECTS correlated with pre-treatment DWI infarction volume and helped predict an unfavorable outcome [8]. Nonetheless, the discrepancy between pre-treatment infarction volume and final infarction volume can occur depending on the time to reperfusion and the recanalization state following EVT [9]. Recently, ASPECTS, which was evaluated on post-treatment non-contrast CT images mostly scanned at 24–72 h after EVT, has shown value in predicting the functional outcome in AIS patients after EVT [10]. However, the infarction area depicted on the non-contrast CT images scanned at a relatively short interval after EVT might also underestimate the infarction area and not accurately depict the final infarction core [10].

In this issue of European Radiology, Xu et al evaluated ASPECTS and infarction volume on post-treatment DWI in 98 patients who underwent EVT and assessed the prognostic value of post-treatment DWI ASPECTS, as compared with post-treatment DWI infarction volume [11]. The authors chose DWI, which is considered more sensitive for the detection and delineation of infarct lesion, as the imaging modality to calculate the post-treatment ASPECTS score. Moreover, the scores were calculated at a relatively later time window than the previous study, so that they could be more close estimates of the final infarction volumes. Xu et al found a strong negative correlation between post-treatment DWI ASPECTS and infarction volume [11]. Furthermore, both post-treatment DWI ASPECTS and infarction volume were found to be independent predictors of good clinical outcome, which was defined as modified Rankin Scale score of 0–2 at 90 days. In the receiver operating characteristic curve analysis, the diagnostic performance of post-treatment DWI ASPECTS (cutoff score, ≥ 6; AUC = 0.836; sensitivity, 87.1%; specificity, 66.7%) was comparable with that of infarction volume (cutoff volume, ≤ 94.87 mL; AUC = 0.821; sensitivity, 90.3%; specificity, 55.6%). These results may suggest that post-treatment DWI ASPECTS might be an eligible surrogate of infarction volume for predicting the clinical outcome of stroke patients following EVT [11].

However, as the authors acknowledged, the intrinsic limitation of ASPECTS due to its semi-quantitative nature is worth mentioning. While several ASPECTS regions (lentiform, insular, caudate, and internal capsule) are of small volumes, all 10 regions are equally assigned 1 point each according to the scoring system. Therefore, some patients with infarctions involving at the lentiform, insular, caudate, and internal capsule areas could have an ASPECT score of 6 despite the small infarction volume.

A few other limitations of the study deserve mention. First, the location of infarction, which is also known to influence the clinical outcome, was not taken into consideration. Second, follow-up MRI was scanned at a varying time interval with a median of 4 days after EVT. The infarction volume measured near the fourth day might not accurately capture the final infarction volume due to cerebral edema. Third, exclusion of patients with poor consciousness and symptomatic hemorrhage transformation could have introduced selection bias.

Nevertheless, the authors have contributed to advancing knowledge and facilitating the clinical workflow in routine practice, by suggesting a convenient and relatively reliable surrogate marker of infarction volume for predicting the clinical outcomes of AIS patients following EVT, particularly in patients with higher ASPECTS and smaller infarction volume.