Reliability and Utility of the Alberta Stroke Program Early Computed Tomography Score in Hyperacute Stroke

doi:10.1016/j.jstrokecerebrovasdis.2017.05.042

Journal of Stroke and Cerebrovascular Diseases

Volume 26, Issue 11, November 2017, Pages 2547-2552

https://doi.org/10.1016/j.jstrokecerebrovasdis.2017.05.042 Get rights and content

Goal

The Alberta Stroke Program Early Computed Tomography Score (ASPECTS) on non-contrast computed tomography (NCCT) is dependent on the visibility of early ischemic change. The goal of our study was to evaluate whether time from ischemic stroke onset to initial NCCT influences the inter-rater variability and prognostic accuracy of ASPECTS for a 3-month functional outcome.

Materials and Methods

Ischemic stroke patients treated with intravenous tissue plasminogen activator (IV-tPA) from 2007 to 2014 at the Royal Melbourne Hospital were included. ASPECTS were blindly assessed by 2 independent raters with inter-rater agreement determined by weighted kappa. Onset time to computed tomography time was dichotomized at the median (≤100 and >100 minutes). Outcome was assessed using the modified Rankin Scale. Logistic regression and receiver operating characteristic analysis were used to assess the prognostic utility of ASPECTS in the early and later time periods.

Results

There were 379 patients included. Inter-rater agreement was significantly lower in the early time period: kappa = .75 (95% confidence interval (CI), .59-.84) ≤ 100 minutes versus .92 (95% CI, .91-.93) > 100 minutes, P < .001. The distributions of absolute inter-rater differences in ASPECTS differed significantly between time epochs (P = .03). The prognostic accuracies of ASPECTS across time epochs were area under the receiver operating characteristic curve ≤ 100 minutes = .57 (95% CI, .50-.64) and >100 minutes = .66 (95% CI, .59-.73), P = .055.

Conclusions

This study demonstrated a significantly lower inter-rater agreement and a trend toward reduced prognostic accuracy of ASPECTS in earlier time periods. The use of ASPECTS to select patients for revascularization in early time windows may be unreliable.

Introduction

Stroke interventions are time critical, and patients who present to the hospital within the first 60 minutes of onset have the greatest chance of benefitting from revascularization therapies.¹ Thus, it has been clinically crucial to develop systems, tools, and technologies to optimize the number of patients receiving treatment earlier.² For example, education in recognizing stroke symptoms, efficient code stroke methods, and the introduction of telemedicine have led to an increase in the proportion of stroke patients arriving at the hospital earlier.³ The preliminary implementation of mobile stroke units has allowed the use of pre-hospital treatment to reduce the median time from stroke onset to therapy decision to as little as 35 minutes.⁴ Currently, the Alberta Stroke Program Early Computed Tomography Score (ASPECTS) on non-contrast computed tomography (NCCT) represents a commonly used imaging tool for detecting early ischemic change (EIC). However, the use of ASPECTS is highly time dependent, with optimal detection on NCCT hours to days from stroke onset.5, 6, 7 As such, and with a shift to earlier stroke treatment times, it is currently unknown whether the utility of ASPECTS will persist as a simple and reliable method for assessing ischemic change.

Five randomized controlled trials demonstrated the superiority of intra-arterial and intravenous tissue plasminogen activator (IV-tPA) treatment over standard treatment for acute ischemic stroke patients with large artery occlusion.8, 9, 10, 11, 12 For most of these trials, the ASPECTS was used to grade the degree of EIC, allowing the prediction of irreversible ischemic injury.¹³ Similarly, for patients only eligible for IV-tPA, the ASPECTS on NCCT has proven a simple and reliable method to identify stroke patients unlikely to make an independent recovery despite this treatment.14, 15 Important advantages of ASPECTS on NCCT are its geographical and temporal ubiquity and its high sensitivity for the detection of intracerebral hemorrhage.¹⁴ EIC on NCCT is measured by a reduction in Hounsfield units (HU), which are time dependent. Parenchymal hypoattenuation is due to an increase in brain tissue water content as a result of ischemic injury, leading to ionic or vasogenic edema.16, 17 Thus, the visibility of ischemic change on NCCT varies from hours to days later depending on the magnitude of water uptake in ischemic tissues.5, 6, 7 As it is established that time affects the visibility of ischemic change, we hypothesized that the ASPECTS in hyperacute times is not a reliable measure for selecting patients for reperfusion therapies. Consequently, the objective of our study was to evaluate whether hyperacute time from ischemic stroke onset to initial NCCT influences the inter-rater variability and prognostic accuracy of ASPECTS for a 3-month functional outcome.

Section snippets

Study Population

This was a retrospective, single-center, cohort study of acute anterior circulation ischemic stroke patients admitted to the Royal Melbourne Hospital between December 26, 2007, and April 20, 2014, who received thrombolysis. Patients were excluded if baseline demographic or follow-up clinical data were unavailable during the patients' time in the ward or if a 3-month follow-up modified Rankin Scale (mRS) was unavailable. Patients were also excluded if the NCCT was severely movement degraded. The

Subject Characteristics

There were a total of 379 participants who met the inclusion criteria and were included in this single-center, retrospective, observational study. The median (interquartile range (IQR)) time of acute stroke onset to CT scan (OCT) was 100 (73-142) minutes. Thus, 100 minutes was used as the time point to define the earlier (≤100 minutes) and later time periods (>100 minutes) of OCT. Baseline characteristics were similar between time periods except for the higher NIHSS score in the early group

Discussion

This study has demonstrated that the NCCT ASPECTS in the early period after stroke is both more difficult to assess, as reflected in reduced inter-rater agreement, and less reliable for prognostication. This is a clinically important finding given the recommendation to use ASPECTS 6-10 as an eligibility criterion for thrombectomy in the recent American Heart Association guidelines update.¹⁹

When we assessed the prognostic accuracy of ASPECTS for a 3-month functional outcome, as measured by the

Conclusions

This study confirms that ASPECTS in earlier times (within 100 minutes) have greater inter-rater variability, but it has additionally demonstrated that the prognostic accuracy of ASPECTS is weaker in the early window. Extra care should be exercised in adopting the ASPECTS on NCCT to ultra-early times as a tool for selecting patients for revascularization. We propose further investigation and validation of the ASPECTS as well as the role of alternative imaging modalities in hyperacute stroke.

References (29)

K. Fassbender et al.
Streamlining of prehospital stroke management: the golden hour
Lancet Neurol
(2013)
S. Walter et al.
Diagnosis and treatment of patients with stroke in a mobile stroke unit versus in hospital: a randomised controlled trial
Lancet Neurol
(2012)
P.A. Barber et al.
Validity and reliability of a quantitative computed tomography score in predicting outcome of hyperacute stroke before thrombolytic therapy. ASPECTS Study Group. Alberta Stroke Programme Early CT Score
Lancet
(2000)
P. Barber et al.
Validity and reliability of a quantitative computed tomography score in predicting outcome of hyperacute stroke before thrombolytic therapy
Lancet
(2000)
J.M. Simard et al.
Brain oedema in focal ischaemia: molecular pathophysiology and theoretical implications
Lancet Neurol
(2007)
J.L. Saver et al.
The “golden hour” and acute brain ischemia: presenting features and lytic therapy in >30,000 patients arriving within 60 minutes of stroke onset
Stroke
(2010)
J.L. Saver
Time is brain—quantified
Stroke
(2006)
J.M. Wardlaw et al.
Can stroke physicians and neuroradiologists identify signs of early cerebral infarction on CT?
J Neurol Neurosurg Psychiatry
(1999)
R. von Kummer et al.
Interobserver agreement in assessing early CT signs of middle cerebral artery infarction
AJNR Am J Neuroradiol
(1996)
P.D. Schellinger et al.
Imaging-based decision making in thrombolytic therapy for ischemic stroke: present status
Stroke
(2003)

J.L. Saver et al.

Stent-retriever thrombectomy after intravenous t-PA vs. t-PA alone in stroke

N Engl J Med

(2015)

T.G. Jovin et al.

Thrombectomy within 8 hours after symptom onset in ischemic stroke

N Engl J Med

(2015)

O.A. Berkhemer et al.

A randomized trial of intraarterial treatment for acute ischemic stroke

N Engl J Med

(2015)

M. Goyal et al.

Randomized assessment of rapid endovascular treatment of ischemic stroke

N Engl J Med

(2015)

Cited by (22)

Comparison of consistency in acute stroke: Automated ASPECT score software tools and manual consensus scores
2024, European Journal of Radiology
The Alberta Stroke Program Early Computed Tomography Score (ASPECTS) is a standardized semi-quantitative scoring system. It is widely used to assess the extent of acute ischemic stroke (AIS). We evaluated the consistency of three automatic software packages with the overall and region-specific manual ASPECTS scores of AIS patients.
Retrospectively, we gathered patients who presented with stroke symptoms between February 2019 and June 2022, and 174 cases were eventually included in the trial. Two radiologists reviewed the NCCT images independently; After four weeks, the same two radiologists began randomly reviewing the DWI images, discussed different scores and give consistent results as ground truth.
Median ASPECTS of the expert consensus reading was 7 (5–9). Good to excellent correlation of ASPECTS total scores among the three software tools (0.70, 0.74 and 0.83). Correlation among ground truth and Rapid-ASPECTS, RealNow-ASPECTS, ShuKun-ASPECTS (ICC = 0.51, Cronbach's α = 0.53), (ICC = 0.60, Cronbach's α = 0.70) and (ICC = 0.52, Cronbach's α = 0.64) respectively. The AUCs for Rapid, RealNow and ShuKun were 0.61, 0.67, and 0.62 respectively. The region-specific results showed a poor to good correlation. The correlations between the non-dominant and dominant cerebral hemispheres and the ground truth were statistically different (P < 0.05).
Overall, the scoring consistency between the three automated scoring software and the ground truth is comparable, with RealNow-ASPECTS being no less consistent and effective than Rapid-ASPECTS and ShuKun-ASPECTS, and even better than both. But the consistency grade that still is developable.
Update on imaging in Code Stroke
2023, Radiologia
El «código ictus» es un procedimiento multidisciplinar diseñado para detectar los ictus agudos de origen isquémico y trasladar a los pacientes para una terapia de reperfusión temprana.
Para la selección de estos pacientes es necesario un estudio radiológico multimodal indistintamente con TC o RM: 1) exploración convencional sin contraste, obligatoria para detectar sangrado; también identifica y cuantifica las áreas de infarto precoz utilizando la escala ASPECTS; 2) estudio angiográfico, necesario en pacientes candidatos a tratamiento mecánico intraarterial; identifica estenosis y obstrucciones y valora la circulación colateral; 3) estudio de perfusión para los pacientes con evolución de entre 6 y 24 h o tiempo indeterminado desde el inicio del evento, que discrimina el tejido infartado del tejido isquémico recuperable.
Los softwares semiautomáticos facilitan el diagnóstico, pero precisan la interpretación del radiólogo.
“Code Stroke” is a multidisciplinary procedure designed to detect acute ischemic strokes and transfer patients for early reperfusion. Selecting these patients requires multimodal imaging with either CT or MRI. 1) Conventional studies without contrast material are obligatory to detect bleeding. Applying the ASPECTS scale, these studies can also identify and quantify areas of early infarction. 2) In candidates for mechanical thrombectomy, angiographic studies are necessary to identify stenoses and obstructions and to evaluate the collateral circulation. 3) Patients with known onset between 6 and 24 hours or with unknown onset require perfusion studies to distinguish between infracted tissue and recoverable ischemic tissue. Semi-automatic software facilitates diagnosis, but radiologists must interpret its output.
Associations between early ischemic signs on non-contrast CT and time since acute ischemic stroke onset: A scoping review
2022, European Journal of Radiology
Citation Excerpt :
An often-used semi-quantitative grading method for the extent of hypodensity is the Alberta Stroke Program Early CT Score (ASPECTS) [40,41]. Visual assessment of hypodensity suffers from a low inter-observer agreement and low sensitivity, especially early since onset [40,42]. This is due to the subtlety of hypodensity in the first few hours since onset, and the significant uptake in water that has to occur before a change in HU can be visible by the human eye [43].
Onset to imaging (OTI) time is a crucial factor in determining treatment eligibility for acute ischemic stroke patients, since the treatments are time-dependent. Patients with an unknown OTI time are often excluded from treatment, or advanced imaging is needed, which is not widely and readily available. As non-contrast CT (NCCT) is part of the standard stroke protocol, estimating OTI time using only NCCT would be valuable for patients with an unknown OTI time. Early ischemic signs (EISs) visible on NCCT might be fit for this purpose if an association between these signs and OTI time exists. This scoping review aims to provide an overview of the literature that associated OTI time with qualitative or quantitative EISs, including the hyperdense artery sign (HAS), decrease in grey matter-white matter differentiation, hypodensity, and mass effect.
The prevalence of the EISs at specific OTI times is assessed, and previously presented associations between the EISs and OTI time are reported.
The EIS prevalence varied between the studies. The HAS prevalence decreased after 6 h since onset. The hypodensity prevalence increased with increasing OTI time. Studies quantifying the extent of hypodensity could distinguish patients within and beyond treatment time windows, indicating its potential to estimate OTI time. Finally, mass effect prevalence was seen more often at later OTI times.
It is concluded that, despite the high prevalence variability between studies, some associations between EISs and OTI time can be observed. These are potentially valuable in estimating OTI time and supporting treatment decisions.
Assistance from Automated ASPECTS Software Improves Reader Performance
2021, Journal of Stroke and Cerebrovascular Diseases
Citation Excerpt :
Defining the best reference standard for an ASPECTS study is challenging. Numerous studies have shown that expert readers often have suboptimal agreement with each other regarding the total score or the individual regions affected.8-13 To help improve the accuracy of the readers, we had 3 expert neuroradiologists review the CT scan along with a follow-up MRI scan that showed the final infarct.
To compare physicians’ ability to read Alberta Stroke Program Early CT Score (ASPECTS) in patients with a large vessel occlusion within 6 hours of symptom onset when assisted by a machine learning-based automatic software tool, compared with their unassisted score.
50 baseline CT scans selected from two prior studies (CRISP and GAMES-RP) were read by 3 experienced neuroradiologists who were provided access to a follow-up MRI. The average ASPECT score of these reads was used as the reference standard. Two additional neuroradiologists and 6 non-neuroradiologist readers then read the scans both with and without assistance from the software reader-augmentation program and reader improvement was determined. The primary hypothesis was that the agreement between typical readers and the consensus of 3 expert neuroradiologists would be improved with software augmented vs. unassisted reads. Agreement was based on the percentage of the individual ASPECT regions (50 cases, 10 regions each; N=500) where agreement was achieved.
Typical non-neuroradiologist readers agreed with the expert consensus read in 72% of the 500 ASPECTS regions, evaluated without software assistance. The automated software alone agreed in 77%. When the typical readers read the scan in conjunction with the software, agreement improved to 78% (P<0.0001, test of proportions). The software program alone achieved correlations for total ASPECT scores that were similar to the expert readers who had access to the follow-up MRI scan to help enhance the quality of their reads.
Typical readers had statistically significant improvement in their scoring of scans when the scan was read in conjunction with the automated software, achieving agreement rates that were comparable to neuroradiologists.
Alberta Stroke Program Early CT Score Calculation Using the Deep Learning-Based Brain Hemisphere Comparison Algorithm
2021, Journal of Stroke and Cerebrovascular Diseases
Citation Excerpt :
It is important to assess the extent of a region of stroke on brain imaging during the acute ischemic period to determine eligibility for reperfusion therapy. However, ASPECTS in the early period after stroke is difficult to assess because of subtle changes in an ischemic lesion, leading to reduced inter-rater agreement in human readers.29,30 RAPID-ASPECTS and human readers showed minimal agreement in the early period after stroke, although this was improved with time from stroke onset.20,21
The Alberta Stroke Program Early Computed Tomography Score (ASPECTS) is a promising tool for the evaluation of stroke expansion to determine suitability for reperfusion therapy. The aim of this study was to validate deep learning-based ASPECTS calculation software that utilizes a three-dimensional fully convolutional network-based brain hemisphere comparison algorithm (3D-BHCA).
We retrospectively collected head non-contrast computed tomography (CT) data from 71 patients with acute ischemic stroke and 80 non-stroke patients. The results for ASPECTS on CT assessed by 5 stroke neurologists and by the 3D-BHCA model were compared with the ground truth by means of region-based and score-based analyses.
In total, 151 patients and 3020 (151 × 20) ASPECTS regions were investigated. Median time from onset to CT was 195 min in the stroke patients. In region-based analysis, the sensitivity (0.80), specificity (0.97), and accuracy (0.96) of the 3D-BHCA model were superior to those of stroke neurologists. The sensitivity (0.98), specificity (0.92), and accuracy (0.97) of dichotomized ASPECTS > 5 analysis and the intraclass correlation coefficient (0.90) in total score-based analysis of the 3D-BHCA model were superior to those of stroke neurologists overall. When patients with stroke were stratified by onset-to-CT time, the 3D-BHCA model exhibited the highest performance to calculate ASPECTS, even in the earliest time period.
The automated ASPECTS calculation software we developed using a deep learning-based algorithm was superior or equal to stroke neurologists in performing ASPECTS calculation in patients with acute stroke and non-stroke patients.
Detection of ischemic changes on baseline multimodal computed tomography: expert reading vs. Brainomix and RAPID software
2020, Journal of Stroke and Cerebrovascular Diseases
Citation Excerpt :
CT perfusion (CTP) has a potential to discriminate between irreversibly damaged tissue, infarct core, and tissue at risk of infarction, penumbra (11,12). It has been demonstrated that visual applying of ASPECTS into CTP parametric maps has a strong correlation with good clinical outcome (defined as modified Rankin scale/mRS 0-2), with a prognostic value greater than NCCT ASPECTS (13–16). All previous studies have shown the highest correlation of good clinical outcome with cerebral blood volume (CBV) ASPECTS (13–16).
The aim of the study was to compare the assessment of ischemic changes by expert reading and available automated software for non-contrast CT (NCCT) and CT perfusion on baseline multimodal imaging and demonstrate the accuracy for the final infarct prediction.
Early ischemic changes were measured by ASPECTS on the baseline neuroimaging of consecutive patients with anterior circulation ischemic stroke. The presence of early ischemic changes was assessed a) on NCCT by two experienced raters, b) on NCCT by e-ASPECTS, and c) visually on derived CT perfusion maps (CBF<30%, Tmax>10s). Accuracy was calculated by comparing presence of final ischemic changes on 24-hour follow-up for each ASPECTS region and expressed as sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV). The subanalysis for patients with successful recanalization was conducted.
Of 263 patients, 81 fulfilled inclusion criteria. Median baseline ASPECTS was 9 for all tested modalities. Accuracy was 0.76 for e-ASPECTS, 0.79 for consensus, 0.82 for CBF<30%, 0.80 for Tmax>10s. e-ASPECTS, consensus, CBF<30%, and Tmax>10s had sensitivity 0.41, 0.46, 0.49, 0.57, respectively; specificity 0.91, 0.93, 0.95, 0.91, respectively; PPV 0.66, 0.75, 0.82, 0.73, respectively; NPV 0.78, 0.80, 0.82, 0.83, respectively. Results did not differ in patients with and without successful recanalization.
This study demonstrated high accuracy for the assessment of ischemic changes by different CT modalities with the best accuracy for CBF<30% and Tmax>10s. The use of automated software has a potential to improve the detection of ischemic changes.

View all citing articles on Scopus

¹: These two authors contributed equally.

View full text

Reliability and Utility of the Alberta Stroke Program Early Computed Tomography Score in Hyperacute Stroke

Goal

Materials and Methods

Results

Conclusions

Introduction

Section snippets

Study Population

Subject Characteristics

Discussion

Conclusions

Lancet Neurol

Lancet Neurol

Lancet

Lancet

Lancet Neurol

The “golden hour” and acute brain ischemia: presenting features and lytic therapy in >30,000 patients arriving within 60 minutes of stroke onset

Stroke

Time is brain—quantified

Stroke

Can stroke physicians and neuroradiologists identify signs of early cerebral infarction on CT?

J Neurol Neurosurg Psychiatry

Interobserver agreement in assessing early CT signs of middle cerebral artery infarction

AJNR Am J Neuroradiol

Imaging-based decision making in thrombolytic therapy for ischemic stroke: present status

Stroke

Stent-retriever thrombectomy after intravenous t-PA vs. t-PA alone in stroke

N Engl J Med

Thrombectomy within 8 hours after symptom onset in ischemic stroke

N Engl J Med

A randomized trial of intraarterial treatment for acute ischemic stroke

N Engl J Med

Randomized assessment of rapid endovascular treatment of ischemic stroke

N Engl J Med