ASPECTS and Other Neuroimaging Scores in the Triage and Prediction of Outcome in Acute Stroke Patients

The Alberta Stroke Program Early CT Score (ASPECTS) and hyperdense vessel sign (HDVS) on baseline non-contrast CT (NCCT) may benefit prognosis of acute ischemic stroke (AIS). We aimed to investigate the agreement of ASPECTS between automated and manual interpretations, and further understand the roles of NCCT and CT Perfusion (CTP) in prognosis.

From January 2019 to May 2020, thrombolysis-treated AIS patients undergoing NCCT and Perfusion imaging before treatment were retrospectively reviewed. A radiologist, a senior neuroradiologist and a neurologist blindly interpreted ASPECTS from NCCT images and a prototypical software produced automated results. Another independent radiologist determined presence of HDVS and CTP-ASPECTS. Three-month modified Rankin scale (mRS) ≤ 2 indicated good functional outcome. NCCT ASPECTS were compared against CTP-ASPECTS using squared weighted kappa. Univariable, multivariable and receiver operating characteristics (ROC) analysis were conducted to evaluate the prognostic value of clinical risk factors, NCCT and CTP findings.

Seventy-five patients were included in this study, of whom 35 (46.7%) presented favorable outcome. Fair to substantial agreement with CTP-ASPECTS was witnessed for automated and manual interpretations (0.685, automated; 0.778, radiologist; 0.830, neuroradiologist; 0.313, neurologist). ASPECTS, HDVS, infarct core volume and mismatch ratio were univariably related to functional outcome, and infarct core volume remained as an independent prognostic factor in the multivariable analysis. The multivariable model achieved an area under ROC (AUC) of 0.768 (95% CI, 0.666–0.870).

Automated ASPECTS achieves substantial agreement with reference CTP-ASPECTS, and comprehensive CT assessment may benefit AIS prognosis after intravenous thrombolysis.

The Alberta Stroke Program Early Computed Tomography Score (ASPECTs) on noncontrast CT (NCCT) is widely used for the decision of endovascular therapy (ET) in randomized controlled trials. CT angiography source images (CTA-SI) is another method to evaluate ischemic brain damage. We aimed to evaluate the association between pretreatment ASPECTs on CTA-SI and post-treatment clinical and radiological outcomes in acute stroke patients treated with ET. The association between both scores along with final infarct and outcome were analyzed. A total of 90 patients with successful recanalization were included in the study. The mean age was 59 ± 11.8. According to the results, CTA-SI ASPECTs was better correlated with final infarct than NCCT ASPECTs (p < 0.001). In univariate analyzes, factors associated with good outcome were age, baseline NIHSS score, and presence of diabetes mellitus (p = 0.001, p = 0.002, p = 0.034, respectively). On the other hand, when an analysis differentiating patients by age was performed, 40 patients below 60 years of age had significantly better outcomes despite having higher baseline NIHSS scores (p = 0.002). Finally, in multivariate analyzes, only age and baseline NIHSS score were found to be independent predictors of good outcome (p = 0.003 for both). In conclusion, CTA-SI ASPECTs in patient selection for ET seems to be more useful than NCCT ASPECTs. However, both scoring modalities were not found to be independent predictors of good outcome. Outcomes are changeable for the younger population who could continue their lives with mild or no deficits despite having a relatively low initial ASPECTs.

While computer-aided diagnosis (CAD) tools are widely used in stroke imaging routines already, their influence on actual decision-making is still underexplored. We analyzed the effect of a simulated CAD tool on ASPECT-Scoring on acute-stroke CT-scans with respect to experience level.

Baseline CT scans of 100 stroke patients from the MR CLEAN trial with consensus-ASPECTS as ground truth were independently ASPECTS graded by three readers with different levels of experience. Weeks later the same CTs were re-analyzed with additional displaying of simulated ASPECTS (s-ASPECTS, by adding or subtracting 2 points from the ground truth). Readers were told that the score was generated by an automatic ASPECT-Scoring algorithm. The influence of the displayed s-ASPECTS on the readers’ second ASPECT-Scoring was analyzed by using a linear mixed model and the reliability was assessed. Performance was measured as the absolute difference between readers ASPECTS and consensus-ASPECTS.

The influence of the s-ASPECTS on the second ASPECT-Scoring was the lowest for the reader with the most experience in neuroradiology, while the other readers were significantly more influenced. All readers veered further away from the ground truth in their second ASPECT-Scoring with the s-ASPECTS, though not significantly. Overall interrater reliability was excellent (ICC = 0.94 [0.92–0.96]).

ASPECT-Scoring may be significantly influenced by simulated ASPECTS displayed by a suboptimal CAD tool, especially in readers with less experience, and performance tends to decrease.