Quantitative Intracerebral Iodine Extravasation in Risk Stratification for Intracranial Hemorrhage in Patients with Acute Ischemic Stroke

Prediction of cerebral hemorrhagic transformation after thrombectomy using a deep learning of dual-energy CT

OBJECTIVES

To develop and validate a deep learning model for predicting hemorrhagic transformation after endovascular thrombectomy using dual-energy computed tomography (CT).

MATERIALS AND METHODS

This was a retrospective study from a prospective registry of acute ischemic stroke. Patients admitted between May 2019 and February 2023 who underwent endovascular thrombectomy for acute anterior circulation occlusions were enrolled. Hemorrhagic transformation was defined using follow-up magnetic resonance imaging or CT. The deep learning model was developed using post-thrombectomy dual-energy CT to predict hemorrhagic transformation within 72 h. Temporal validation was performed with patients who were admitted after July 2022. The deep learning model's performance was compared with a logistic regression model developed from clinical variables using the area under the receiver operating characteristic curve (AUC).

RESULTS

Total of 202 patients (mean age 71.4 years ± 14.5 [standard deviation], 92 men) were included, with 109 (54.0%) patients having hemorrhagic transformation. The deep learning model performed consistently well, showing an average AUC of 0.867 (95% confidence interval [CI], 0.815-0.902) upon five-fold cross validation and AUC of 0.911 (95% CI, 0.774-1.000) with the test dataset. The clinical variable model showed an AUC of 0.775 (95% CI, 0.709-0.842) on the training dataset (p < 0.01) and AUC of 0.634 (95% CI, 0.385-0.883) on the test dataset (p = 0.06).

CONCLUSION

A deep learning model was developed and validated for prediction of hemorrhagic transformation after endovascular thrombectomy in patients with acute stroke using dual-energy computed tomography.

CLINICAL RELEVANCE STATEMENT

This study demonstrates that a convolutional neural network (CNN) can be utilized on dual-energy computed tomography (DECT) for the accurate prediction of hemorrhagic transformation after thrombectomy. The CNN achieves high performance without the need for region of interest drawing.

KEY POINTS

• Iodine leakage on dual-energy CT after thrombectomy may be from blood-brain barrier disruption. • A convolutional neural network on post-thrombectomy dual-energy CT enables individualized prediction of hemorrhagic transformation. • Iodine leakage is an important predictor of hemorrhagic transformation following thrombectomy for ischemic stroke.

Post-endovascular treatment, blood-brain barrier disruption, predicts patient outcomes better than pre-treatment status

OBJECTIVES

Blood-brain barrier (BBB) disruption is an important pathological change after cerebral infarction that exacerbates brain injury. We aimed to investigate and compare the predictive utility of pre-treatment BBB permeability (BBBP) and BBBP within 1 h after endovascular treatment (EVT) for hemorrhagic transformation (HT) and 90-day prognosis.

METHODS

Patients underwent preoperative computed tomography perfusion (CTP) and non-contrast CT (NCCT) within 1 h after EVT. Preoperative BBBP was determined by the relative permeability surface area product (rPS) in the hypoperfusion area. Postoperative BBBP was determined by the post-EVT Alberta Stroke Program Early CT Score (Post-ASPECTS), which is based on brain parenchymal hyperdensity on the postoperative NCCT.

OUTCOMES

We included 100 patients. Univariate logistic regression analysis revealed correlations of preoperative rPS with HT, poor outcomes, and death. However, these correlations were not observed in multivariate logistic regression. A Post-ASPECTS ≤7 and could independently predict poor outcomes, while Post-ASPECTS ≤6 could independently predict death and HT. The baseline National Institutes of Health Stroke Scale (NIHSS) score could independently predict poor outcomes and death but not HT. A combined model using the baseline NIHSS and Post-ASPECTS scores had better predictive performance for poor outcomes and death than baseline NIHSS score alone; however, it was not superior to the predictive performance of the Post-ASPECTS score.

CONCLUSION

The preoperative rPS cannot independently predict clinical outcomes in EVT-treated patients; contrastingly, the Post-ASPECTS score could independently predict poor outcomes, death, and HT. This parameter could inform prompt postoperative treatment decisions.

Determinants and Clinical Relevance of Iodine Contrast Extravasation after Endovascular Thrombectomy: A Dual-Energy CT Study

BACKGROUND AND PURPOSE

Iodine contrast extravasation (ICE) is common in patients with acute ischemic stroke (AIS) after endovascular-thrombectomy (EVT). The aim of our study was to evaluate the incidence of ICE assessed by dual-energy CT (DECT), its determinants, and associations with clinical outcome.

MATERIALS AND METHODS

We retrospectively examined imaging parameters and clinical factors from consecutive patients with AIS treated with EVT who had a DECT 24 hours thereafter, identified at a single academic center. Associations between ICE, clinical, imaging, and procedural parameters, as well as clinical outcome were explored by using univariable and multivariable models.

RESULTS

A total of 197 consecutive patients were included (period 2019-2020), of which 53 (27%) demonstrated ICE that was pure ICE in 30/53 (57%) and mixed with intracranial hemorrhage (ICH) in 23/53 (43%). Low initial-ASPECTS, high per-procedural-contrast volume injected, and high admission-glycemia were independently associated with ICE (respectively, OR = 0.43, 95% CI, 0.16-1.13, P = .047; OR = 1.02, 95% CI, 1.00-1.04, P = .003; OR = 8.92, 95% CI, 0.63-125.77, P = .043). ICE was independently associated with ICH (P = .047), but not with poorer clinical outcome (6-month mRS >2, P = .223). Univariate analysis demonstrated that low ADC, higher ischemic volume, ICA occlusion, mass effect, longer procedure duration, combined thrombectomy technique, higher number of device passes, and lower recanalization rate were associated with ICE (respectively, P = .002; <.001; .002; <.001; .002; 0.011; <0.001; 0.015).

CONCLUSIONS

ICE evaluated with DECT is a relatively frequent finding after EVT, present in almost one-third of patients. Lower admission ASPECTS, higher glycemia, and high contrast volume injected per procedure were associated with ICE. We also found an association between ICE and ICH, confirming blood-brain barrier alteration as a major determinant of ICH.

Prognostic value of contrast staining on dual-energy CT after endovascular therapy in acute ischemic stroke: a meta-analysis

BACKGROUND

Contrast staining (CS) on dual-energy CT (DECT) is common after endovascular therapy (EVT) in acute ischemic stroke (AIS). We performed a meta-analysis to investigate the prognostic significance of CS detected by DECT after EVT in AIS.

METHOD

MEDLINE, EMBASE, Cochrane Central Register of Controlled Trials, Web of Science and Scopus databases were searched from inception to July 2023 for publications on the prognostic significance of CS on DECT after EVT in patients with AIS. Prognostic outcomes were hemorrhage transformation (HT) and poor functional outcome (modified Rankin Scale [mRS] Score of 3-6 at the 90-day follow-up). Data are presented as odds ratios (OR) and 95% confidence intervals (CI).

RESULTS

Eleven studies including 1123 cases of AIS were included. Pooled results indicated a higher risk of HT in patients with CS than in those without CS (OR = 2.22; 95% CI 1.41-3.51, P = 0.001; I2 = 45.4%). No association between CS and symptomatic HT was observed (OR = 2.10; 95% CI 0.64-6.95, P = 0.223; I2 = 67.3%). Moreover, there was also higher odds of poor functional outcome in patients with CS than in those without CS (OR = 2.76; 95% CI 1.53-4.97, P = 0.001; I2 = 44.9%).

CONCLUSIONS

The presence of contrast staining on DECT after EVT is associated with a higher risk of hemorrhage transformation and poor functional outcome. However, further high-quality studies with standardized processes are required to confirm these results.