Prediction of recurrence in patients with non-acute atherosclerotic middle cerebral artery occlusion: A longitudinal study using computed tomography perfusion and magnetic resonance imaging

Abstract

Objective

This study aimed to assess the performance of time to maximum concentration (Tmax) using computed tomography perfusion for predicting stroke recurrence in patients with symptomatic non-acute atherosclerotic middle cerebral artery occlusion.

Methods

This retrospective monocentric study enrolled 138 patients diagnosed with non-acute atherosclerotic middle cerebral artery occlusion. In addition to conventional computed tomography angiography or digital subtraction angiography, all patients underwent head computed tomography perfusion and magnetic resonance imaging 7–14 days after stroke and repeat head magnetic resonance imaging at 3–6 weeks. Volume of time to maximum concentration >4 s was considered the hypoperfusion area. The association between imaging characteristics, new ischemic lesion or lesion size increase on magnetic resonance imaging, and patient prognosis was assessed.

Results

Increased number/area of infarct lesion was observed in 52/138 (37.7%) patients on diffusion weighted imaging 3–6 weeks after the first stroke. The volume of time to maximum concentration >4 s at baseline was strongly associated with infarct lesion development (odds ratio = 1.22 per 10-mL increase, 95% confidence interval: 1.10–1.34, p < 0.001). Volume of time to maximum concentration >4 s exhibited high discriminative ability for poor prognosis (area under the curve = 0.84, 95% confidence interval: 0.78–0.91, p < 0.001) with volume of time to maximum concentration >4 s larger than 99.8 mL being an optimal cutoff (odds ratio = 14.31, 95% confidence interval: 5.84–40.89, p < 0.001, sensitivity = 0.88 and specificity = 0.65).

Conclusions

The risk of stroke recurrence in patients with symptomatic non-acute atherosclerotic middle cerebral artery occlusion gradually increases with the volume of time to maximum concentration >4 s. Patients with volume of time to maximum concentration >4 s larger than 99.8 mL may have a significantly higher risk of stroke. Volume of time to maximum concentration >4 s on computed tomography perfusion can be used to predict the prognosis in patients with symptomatic non-acute atherosclerotic middle cerebral artery occlusion.

Keywords

Stroke computed tomography perfusion diffusion weighted imaging prognosis middle cerebral artery occlusion

Introduction

For acute intracranial large artery occlusion, positive results from studies on endovascular therapy (EVT) published since 2015 have confirmed that mechanical thrombectomy can significantly improve the prognosis of patients with intracranial large artery occlusion.^1–5 The treatment time frame was expanded to 24 h in 2018, according to the findings published in the DEFUSE 3 and DAWN studies.^6,7 However, <5% of patients in China were able to undergo EVT within 24 h of stroke.⁸ Intracranial large artery occlusion lasting >24 h is unified as non-acute intracranial large artery occlusion (NILAO).^9,10 Owing to good collateral circulation, some of these NILAO patients are asymptomatic. However, a considerable proportion of these patients exhibit symptoms, which mainly manifest as recurrent ischemic stroke, progressive neurological deterioration as well as cognitive and emotional disorders.¹¹

Previous studies have reported that NILAO is more common in stroke patients in China.¹² Epidemiological studies have reported an NILAO incidence of 34.5% among Asian stroke/transient ischemic attack (TIA) patients.¹² The Chinese Intracranial Atherosclerosis Study (CICAS) has reported the presence of intracranial large artery disease (stenosis ≥50%) in 46.6% stroke patients.¹³ Furthermore, the annual stroke risk of patients with symptomatic NILAO was 23.4%. In addition to being associated with a high recurrence risk, NILAO can also lead to cognitive decline and emotional disorder, which seriously affect the patient’s quality of life.^12,13

Several studies have identified the middle cerebral artery (MCA) as the most common site of intracranial arterial occlusion.^14,15 The mechanism of non-acute atherosclerotic middle cerebral artery occlusion (NAMCAO) leading to stroke recurrence remains unclear; however, most scholars believe that hemodynamic changes following middle cerebral artery occlusion (MCAO) are the most important cause of recurrent stroke.¹⁶ Regional cerebrovascular reactivity (rCVR) refers to the ability of blood vessels in a specific area of the brain to dilate or constrict in response to metabolic demands or external stimuli. This ability is used to assess cerebrovascular reserve capacity.¹⁷ For patients with conditions such as carotid artery stenosis or MCAO, reduction or loss of rCVR serves as a warning signal and is a significant predictor of ischemic stroke recurrence.¹⁸ Previous studies have reported that the annual risk of stroke recurrence in patients with NAMCAO accompanied with a decrease in rCVR was as high as 35.6%, significantly higher than that in patients with normal rCVR.^17,18 Although the recurrence rate is high, it is still challenging to define hypoperfusion and predict stroke recurrence in stroke patients with NAMCAO. Therefore, there is an urgent need to find a simple and reliable predictor of stroke recurrence in these patients.

In one clinical study, the hypoperfusion area was defined as time to maximum concentration (Tmax) >6 s on computed tomography perfusion (CTP).⁷ Due to the influence of brain tissue tolerance, collateral circulation, and other factors, the hypoperfusion area of NAMCAO may be different from that of acute cerebral artery occlusion. Due to long-term ischemic preconditioning and establishment of collateral circulation, patients with NAMCAO usually present a small area of Tmax >6 s and cerebral blood flow (CBF) <30%, which is inadequate for assessing the hypoferfusion area.¹⁹ In this study, we defined hypoperfusion area as volume of Tmax >4 s on CTP. EVT can improve the hemodynamic alterations caused by NAMCAO, thereby reducing the risk of stroke recurrence. We attempted to predict stroke recurrence in stroke patients with NAMCAO using CTP parameters, which may provide new evidence for EVT in these patients.

Patients and methods

Patient selection

We conducted our study in accordance with the Helsinki Declaration of 1975, as revised in 2024. The institutional review board (IRB) and Ethics Committee of Nanjing Drum Tower Hospital approved this study, number: 2021-399-02. We have deidentified all patient details. The reporting of this study conforms to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines.²⁰

This was a retrospective monocentric study. The following selection criteria were applied for patient inclusion: (a) MCAO confirmed on digital subtraction angiography (DSA), defined as no forward blood flow; (b) presenting with occluded artery related stroke; and (c) MCA atherosclerotic occlusion. The exclusion criteria were as follows: (a) clinical, laboratory, or imaging findings not suspicious for atherosclerotic lesions, such as vasculitis, moyamoya syndrome, and arterial dissection; (b) a coexisting cardioembolic source (e.g. atrial fibrillation, mitral stenosis, prosthetic valve, myocardial infarction within 6 weeks, intracardiac clot, ventricular aneurysm, and bacterial endocarditis); (c) a concomitant intracranial aneurysm or any bleeding disorder; (d) large infarct core, defined as an Alberta Stroke Program Early computed tomography (CT) Score (ASPECTS) of <6 points; (e) progressive stroke; (f) stenosis of the contralateral artery >70%; and (g) no follow-up results available. All patients provided informed consent for study participation.

Clinical assessment

All patients underwent head CTP and magnetic resonance (MR) imaging 7–14 days after the first stroke and head MR imaging 3–6 weeks later as part of standard follow-up. Lesion development was defined as increased number of lesions or enlarged lesions on diffusion weighted imaging (DWI) 3–6 weeks after first stroke.

We obtained data regarding age, sex, functional status (evaluated using the modified Rankin scale (mRS)), and baseline stroke severity (evaluated using the National Institutes of Health Stroke Scale (NIHSS)). Stroke recurrence was defined as increased number of lesions or enlarged lesions on DWI. Imaging results were evaluated by two independent, qualified clinicians. The intraclass correlation coefficient (2, 1) was 0.91.

Imaging

Estimates of the hypoperfusion area obtained from CTP scans were calculated using RAPID software (iSchemaView), an automated image postprocessing system. The size of the hypoperfusion area was estimated based on the volume of tissue for which there was delayed arrival of an injected tracer agent, Tmax >4 s. Furthermore, volumes of Tmax >4 s and Tmax >6 s were measured. More than 50% patients showed no volume of Tmax >6 s on CTP; although this parameter has been considered the gold standard according to previous studies involving acute MCAO patients, we did not take it into consideration.

Postprocedural antiplatelet therapy

NAMCAO patients received aspirin (100 mg/day) and clopidogrel (75 mg/day). For patients who could not tolerate clopidogrel, cilostazol was considered.

Statistical analyses

Continuous variables were summarized as medians with interquartile ranges, while categorical variables were presented as counts (n) and percentages (%). Baseline and procedural characteristics and treatment outcomes were compared between stroke recurrence and control groups. Categorical and binary variables were analyzed using χ² test, and continuous variables were analyzed using Mann–Whitney U test. Receiver operating characteristic (ROC) curves were used to evaluate the prediction performance of Tmax >4 s volume. Additionally, multinomial logistic regression analysis was performed to assess the correlation between hypoperfusion and stroke recurrence. The covariates were age, volume of Tmax >3 s, volume of Tmax >4 s, volume of Tmax >5 s, and smoking; the goodness-of-fit of the logistic regression model was assessed using Hosmer–Lemeshow and Tjur’s R squared analyses. Variance inflation factors (VIFs) were used to examine the multicollinearity among covariates, VIF <5 was considered acceptable. Multivariable logistic regression analysis was performed to assess the association between Tmax >4 s volume and stroke recurrence. Youden index was used to determine the optimal cutoff point; the largest Youden index was considered the optimal cutoff. All statistical analyses were performed using Statistical Package for Social Sciences (SPSS) software (version 25.0; IBM SPSS, Chicago, IL), and p-value <0.05 was considered statistically significant.

Results

Baseline characteristics

From February 2019 to June 2022, in total, 215 patients at our stroke centers with non-acute MCAO were reviewed. A total of twenty-five patients were excluded because of poor quality of follow-up data, 19 were ruled out due to severe stenosis of the contralateral intracranial artery, 15 were excluded because they presented with moyamoya syndrome, and 18 were ruled out because of poor image quality. Finally, 138 patients were enrolled (Figure 1).

Figure 1.

Flowchart of patient selection.

In total, 52 (37.7%) patients showed stroke recurrence on DWI after 3–6 weeks of the first stroke. As shown in Table 1, the baseline NIHSS and mRS scores were not different between the two groups. The stroke recurrence group (181.5, 117.9–245.1 mL) showed a larger volume of Tmax >4 s on CTP than the control group (83.75, 37.03–124.8 mL) after the first stroke (p < 0.001). The volume of Tmax >5 s was also larger in the recurrence group (p = 0.031).

Table 1.

Characteristics of the study population.

	All patients(n = 138)	Control group(n = 86)	Stroke recurrence(n = 52)	p-value
Age (years)(median) (IQR)	62.0 (55.0–70.0)	63 (55.0–72.25)	60.5 (54.0–68.0)	0.28
Sex, male (N) (%)	92 (66.7)	61 (70.9)	37 (71.2)	0.98
Risk factors
Hypertension (N) (%)	101 (73.2)	61 (70.9)	40 (76.9)	0.44
Diabetes (N) (%)	27 (19.6)	16 (18.6)	11 (21.2)	0.71
Smoking (N) (%)	59 (42.8)	41 (47.7)	18 (34.6)	0.13
Dyslipidemia (N) (%)	22 (15.9)	16 (18.6)	6 (11.5)	0.34
Baseline NIHSS score(median) (IQR)	2 (0–3)	1 (0–3)	2 (0–4)	0.42
Baseline mRS score(median) (IQR)	1 (0–1)	1 (0–1)	1 (0–1)	0.98
Postprocedural oral antiplatelet medication (N) (%)	137 (99.3)	86 (100.0)	51 (98.1)	0.38
Volume of Tmax >4 s(median) (IQR)	165.5 (64.78–165.50)	83.75 (37.03–124.80)	181.50 (117.90–245.10)	0.001
Volume of Tmax >3 s(median) (IQR)	554.70 (358.70–694.60)	493.80 (302.40–690.80)	583.9 (379.5–721.3)	0.293
Volume of Tmax >5 s(median) (IQR)	18.71 (4.20–40.85)	11.00 (0.00–40.00)	23.00 (10.70–42.53)	0.031

NIHSS: National Institutes of Health Stroke Scale; IQR: interquartile range; Tmax: time to maximum concentration; mRS: Modified Rankin scale.

Multivariable logistic regression analysis (Table 2) show that volume of Tmax >4 s had a strong association with stroke recurrence (1.22 per 10-mL increase, 95% confidence interval (CI): 1.10–1.34, p < 0.001).

Table 2.

Multivariable logistic regression analysis of the association between Tmax >4 s volume and stroke recurrence.

	Variable	OR	95% CI	p-value	Hosmer–Lemeshow analysis
Stroke recurrence	Volume of Tmax >4 s	1.22	1.10–1.34	0.001	p-value 0.323
	Volume of Tmax >5 s	0.95	0.79–1.26	0.36
	Smoking	0.739	0.687–1.012	0.139
	Age	0.984	0.943–1.026	0.446

OR: odds ratio; CI: confidence interval; Tmax: time to maximum concentration.

Tmax >4 s volume predicts stroke recurrence

As shown in Figure 2, ROC curve analysis of Tmax >4 s volume for predicting stroke recurrence showed an area under the curve of 0.84 (95% CI = 0.78–0.91, p < 0.001) (Table 3). We chose volume of Tmax >4 s as the predictor of stroke recurrence. The best cutoff point for volume of Tmax >4 s was 99.8 mL (Table 4), with a sensitivity of 0.88, specificity of 0.65, and Youden index of 0.53. Multivariable logistic regression analysis (Table 5) showed that volume of Tmax >4 s larger than 99.8 mL was strongly related with stroke recurrence (odds ratio (OR) = 14.31, 95% CI = 5.84–40.89, p < 0.001).

Figure 2.

ROC curve analysis of Tmax >4 s volume for predicting stroke recurrence. Details of ROC curve analysis of volume of Tmax >4 s are shown in Table 3.

Table 3.

ROC curve analysis of Tmax >4 s volume.

Area under the ROC curve
Area	0.84
Standard error	0.03
95% confidence interval	0.78–0.91
p-value	<0.001

ROC: receiver operating characteristic; Tmax: time to maximum concentration.

Table 4.

Cutoff point of Tmax >4 s volume.

Cutoff point (mL)	Sensitivity (%)	Specificity (%)
>99.80	88.46	65.12
>98.05	88.46	63.95
>93.40	90.38	61.63
>139.6	69.23	82.56
>101.2	86.54	65.12
>96.35	88.46	62.79
>127.0	73.08	77.91
>91.45	90.38	60.47
>163.7	57.69	93.02
>137.2	69.23	81.4

Tmax: time to maximum concentration.

Table 5.

Multivariable logistic regression analysis of Tmax >4 s volume and stroke recurrence.

	Variable	OR	95% CI	p-value	Hosmer–Lemeshow analysis
Stroke recurrence	Tmax >4 s larger than 99.8 mL	14.31	5.842–40.890	0.001	p-value = 0.260
	Smoking	0.994	0.980–1.221	0.199
	Age	0.991	0.950–1.033	0.654
	Baseline NIHSS	1.113	0.910–1.413	0.348
	Baseline mRS	0.783	0.311–1.855	0.588

Tmax: time to maximum concentration; OR: odds ratio; CI: confidence interval; NIHSS: National Institutes of Health Stroke Scale; mRS: modified Rankin scale.

Patients were divided into mild or severe hypoperfusion group according to whether the volume of Tmax >4 s was larger than 99.8 mL. Six patients (9.68%) in the mild hypoperfusion group and 46 (60.52%) in the severe hypoperfusion group exhibited stroke recurrence. The rate of stroke recurrence was much higher in the severe hypoperfusion group (p < 0.001) (Table 6).

Table 6.

Stroke recurrence in the mild and severe hypoperfusion groups.

	All patients(n = 138)	Mild hypoperfusion group(n = 62)	Severe hypoperfusion group(n = 76)	p-value
Stroke recurrence	52 (37.68)	6 (9.68)	46 (60.52)	<0.001
Age	62.00 (55.00–70.22)	65.50 (55.00–74.00)	61.00 (54.25–67.75)	0.07
NIHSS	2 (0–3)	1 (0–3)	2 (0–4)	0.34
mRS	1 (0–1)	1 (0–1)	1 (0–1)	0.90

mRS: modified Rankin scale; NIHSS: National Institutes of Health Stroke Scale.

Discussion

The main finding of the present study is that the CTP parameter, Tmax, is independently associated with stroke recurrence in patients with NAMCAO. We defined hypoperfusion as the area with Tmax >4 s and found that volume of Tmax >4 s was significantly larger in the recurrence group than in the control group. Logistic regression analysis showed that a large hypoperfusion area is an important risk factor for stroke recurrence in patients with NAMCAO and can predict stroke recurrence in these patients. As the baseline mRS score of all patients who underwent EVT ranged from 0 to 3, no significant differences were found in the follow-up mRS and NIHSS scores between the two groups. The severity of neurological impairment may be influenced not only by the volume of hypoperfusion but also by its location in relation to eloquent brain areas.

Non-acute atherosclerotic intracranial large artery occlusion, especially NAMCAO, is common among stroke patients in China^21,22 and is accompanied with a higher risk of stroke recurrence.²³ However, appropriate indicators to evaluate the perfusion status of the affected side are still lacking.

Inspired by the DEFUSE 3 study, CTP has gained popularity in the evaluation of stroke patients with acute anterior large artery occlusion at 6–16 h after symptom onset.²⁴ Additionally, the area with Tmax >6 s was considered the hypoperfusion area, which might develop into core infarction without timely treatment.²⁴ Venous drainage identified on computed tomographic angiography is reportedly correlated with collateral status and infarct growth.^25–27 However, due to long-term ischemic preconditioning and establishment of collateral circulation, patients with NAMCAO usually present a small area of Tmax >6 s and CBF <30%, which is inadequate for assessing the hypoperfusion area.²⁸ Herein, we chose the volume of Tmax >4 s to assess the hypoperfusion area and found that it could be used to predict stroke recurrence in patients with NAMCAO.

Our study was the first attempt to predict stroke recurrence in NAMCAO patients using the volume of Tmax >4 s on CTP; however, certain study limitations should be noted. First, the present findings should be interpreted with caution because of the limited sample size and the fact that data were obtained from a single center. Second, Tmax is sensitive to several hemodynamic effects, and its physiological interpretation is complex.²⁷ Third, patients did not undergo CTP after stroke recurrence; therefore, we were unable to evaluate the perfusion status in case of stroke recurrence. Fourth, ASPECTS <6 as an exclusion criterion may have led to the exclusion of more severe cases, potentially leading to a selection bias toward milder stroke cases. Fifth, we did not count the size or quantity of lesions for quantitative analysis on follow-up DWI. MR DWI was only used to verify whether there was recurrence of cerebral infarction and for qualitative analysis. We cannot say with certainty whether a larger volume of Tmax >4 s corresponds to larger or more disabling strokes or a higher number of new lesions. Sixth, in this study, we only paid attention to stroke recurrence and did not correlate Tmax volumes or recurrence with clinical endpoints. In the future, we plan to perform an randomized controlled trial (RCT) to further confirm that the volume of Tmax >4 s is useful for assessing hypoperfusion and predict stroke recurrence in NAMCAO patients.

Conclusion

The risk of stroke recurrence in patients with symptomatic NAMCAO gradually increases with the volume of Tmax >4 s. Patients with a Tmax >4 s volume exceeding 99.8 mL may have a significantly higher risk of stroke. Volume of Tmax >4 s on CTP can be used to predict the prognosis in patients with symptomatic NAMCAO because patients with larger volume of Tmax >4 s are likely to experience recurrence.

Footnotes

Acknowledgments

We extend our sincere gratitude to all the participants of this study.

Author contributions

Xi Zhang conceived and designed the experiments, performed the experiments, analyzed and interpreted the data and wrote this study.

Yuxiang Zhang performed the experiments as well as analyzed and interpreted the data.

Rongcheng Zou performed the experiments as well as analyzed and interpreted the data.

Zhibin Chen analyzed and interpreted the data as well as contributed reagents, materials, analysis tools, or data.

Guangxin Duan analyzed and interpreted the data as well as contributed reagents, materials, analysis tools, or data.

Zhibin Chen performed the experiments as well as analyzed and interpreted the data.

Jingwei Li conceived and designed the experiments, performed the experiments, and wrote the manuscript.

Yun Luo conceived and designed the experiments, performed the experiments, and wrote the manuscript.

Data availability statement

Data will be made available on request.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have influenced the work reported in this study.

Funding

The authors received no financial support for the research, authorship, and/or publication of this study.

ORCID iDs

Yun Luo

Jingwei Li

References

Berkhemer

Fransen

Beumer

; MR CLEAN Investigatorset al. A randomized trial of intraarterial treatment for acute ischemic stroke. N Engl J Med 2015; 372: 11–20.

Campbell

Mitchell

Kleinig

; EXTEND-IA Investigatorset al. Endovascular therapy for ischemic stroke with perfusion-imaging selection. N Engl J Med 2015; 372: 1009–1018.

Goyal

Demchuk

Menon

; ESCAPE Trial Investigatorset al. Randomized assessment of rapid endovascular treatment of ischemic stroke. N Engl J Med 2015; 372: 1019–1030.

Jovin

Chamorro

Cobo

; REVASCAT Trial Investigatorset al. Thrombectomy within 8 hours after symptom onset in ischemic stroke. N Engl J Med 2015; 372: 2296–2306.

Saver

Goyal

Bonafe

; SWIFT PRIME Investigatorset al. Stent-retriever thrombectomy after intravenous t-PA vs. t-PA alone in stroke. N Engl J Med 2015; 372: 2285–2295.

Nogueira

Jadhav

Haussen

; DAWN Trial Investigatorset al. Thrombectomy 6 to 24 hours after stroke with a mismatch between deficit and infarct. N Engl J Med 2018; 378: 11–21.

Albers

Marks

Kemp

; DEFUSE 3 Investigatorset al. Thrombectomy for stroke at 6 to 16 hours with selection by perfusion imaging. N Engl J Med 2018; 378: 708–718.

Gorelick

Wong

Bae

, et al. Large artery intracranial occlusive disease: a large worldwide burden but a relatively neglected frontier. Stroke 2008; 39: 2396–2399.

Mattioni

Cenciarelli

Eusebi

, et al. Transcranial Doppler sonography for detecting stenosis or occlusion of intracranial arteries in people with acute ischaemic stroke. Cochrane Database Syst Rev 2020; 2: CD010722.

10.

Mattioni

Cenciarelli

Biessels

, et al. Prevalence of intracranial large artery stenosis and occlusion in patients with acute ischaemic stroke or TIA. Neurol Sci 2014; 35: 349–355.

11.

Wang

Zhao

Liu

; CICAS Study Groupet al. Prevalence and outcomes of symptomatic intracranial large artery stenoses and occlusions in China: The Chinese Intracranial Atherosclerosis (CICAS) Study. Stroke 2014; 45: 663–669.

12.

Marshall

Festa

Cheung

, et al. Cerebral hemodynamics and cognitive impairment: baseline data from the RECON trial. Neurology 2012; 78: 250–255.

13.

Khatri

Abdallah

Ellis

Management of coronary chronic total occlusion. Cleve Clin J Med 2017; 84: 27–38.

14.

Hernández-Pérez

Pérez de la Ossa

Aleu

, et al. Natural history of acute stroke due to occlusion of the middle cerebral artery and intracranial internal carotid artery. J Neuroimaging 2014; 24: 354–358.

15.

Rotzinger

Mosimann

Meuli

, et al. Site and rate of occlusive disease in cervicocerebral arteries: a CT angiography study of 2209 patients with acute ischemic stroke. AJNR Am J Neuroradiol 2017; 38: 868–874.

16.

Han

Lam

, et al. Total cerebral blood flow estimated by color velocity imaging quantification ultrasound: a predictor for recurrent stroke? J Cereb Blood Flow Metab 2007; 27: 850–856.

17.

Wong

Gao

Chan

, et al. Mechanisms of acute cerebral infarctions in patients with middle cerebral artery stenosis: a diffusion-weighted imaging and microemboli monitoring study. Ann Neurol 2002; 52: 74–81.

18.

Kuroda

Houkin

Kamiyama

, et al. Long-term prognosis of medically treated patients with internal carotid or middle cerebral artery occlusion: can acetazolamide test predict it? Stroke 2001; 32: 2110–2116.

19.

Hou

, et al. Tmax >4 s volume predicts stroke recurrence in symptomatic intracranial atherosclerotic stenosis with optimal medical treatment. J Neurointerv Surg 2025; 18: 99–107.

20.

Von Elm

Altman

Egger

; STROBE Initiativeet al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Ann Intern Med 2007; 147: 573–577.

21.

Menon

Qazi

Nambiar

; Interventional Management of Stroke III Investigatorset al. Differential effect of baseline computed tomographic angiography collaterals on clinical outcome in patients enrolled in the Interventional Management of Stroke III Trial. Stroke 2015; 46: 1239–1244.

22.

Zhu

Michel

Aghaebrahim

, et al. Computed tomography workup of patients suspected of acute ischemic stroke: perfusion computed tomography adds value compared with clinical evaluation, noncontrast computed tomography, and computed tomography angiogram in terms of predicting outcome. Stroke 2013; 44: 1049–1055.

23.

Zhang

Han

Sun

, et al. Endovascular recanalization and standard medical management for symptomatic non-acute intracranial artery occlusion: study protocol for a non-randomized, 24-month, multicenter study. Front Neurol 2021; 12: 729534.

24.

Tan

Parsons

Bivard

, et al. Comparison of computed tomography perfusion and multiphase computed tomography angiogram in predicting clinical outcomes in endovascular thrombectomy. Stroke 2022; 53: 2926–2934.

25.

Robben

Boers

AMM

Marquering

, et al. Prediction of final infarct volume from native CT perfusion and treatment parameters using deep learning. Med Image Anal 2020; 59: 101589.

26.

Katyal

Calic

Killingsworth

, et al. Diagnostic and prognostic utility of computed tomography perfusion imaging in posterior circulation acute ischemic stroke: a systematic review and meta-analysis. Eur J Neurol 2021; 28: 2657–2668.

27.

Fabritius

Reidler

Froelich

, et al. Incremental value of computed tomography perfusion for final infarct prediction in acute ischemic cerebellar stroke. J Am Heart Assoc 2019; 8: e013069.

28.

Divel

Christensen

Segars

, et al. A dynamic simulation framework for CT perfusion in stroke assessment built from first principles. Med Phys 2021; 48: 3500–3510.