Collateral circulation as a predictor of reperfusion injury: insights from mechanical thrombectomy in stroke patients
Abstract
Background/aim: The purpose of this study was to evaluate the association between collateral circulation (CC) status on preprocedural brain CT angiography (BCTA) and the development of postthrombectomy hemorrhagic transformation (HeT) and contrast extravasation (CE) in patients with acute ischemic stroke. Materials and methods: This retrospective, single-center study included 185 patients with middle cerebral artery occlusion who underwent mechanical thrombectomy (MT) between February 2019 and October 2022 and achieved complete angiographic reperfusion with a final Thrombolysis in Cerebral Infarction grade of 3. Preprocedural CC was assessed on biphasic BCTA using a modified threecategory grading approach and classified as good, moderate, or poor. Newly developed hyperdense lesions on noncontrast CT obtained within 24 h of thrombectomy were classified as CE or HeT according to their evolution on follow-up CT performed 24–48 h later. Associations were evaluated using exact tests and multivariable binary logistic regression. Interreader agreement was assessed using Cohen’s kappa coefficient. Results: Postthrombectomy hyperdense lesions were detected in 93 patients (50.3%). Of the entire cohort, 62 patients (33.5%) developed HeT and 31 (16.8%) developed CE. HeT occurred in 12 of the 53 patients (22.6%) with good CC, 43 of the 123 (35.0%) with moderate CC, and seven of the nine (77.8%) with poor CC. CC category was significantly associated with HeT (Fisher’s exact test, p = 0.005), whereas no significant association was observed between CC and CE. In the multivariable analysis, poor CC was the only factor independently associated with HeT and was associated with sevenfold higher odds of HeT (adjusted odds ratio 7.00; 95% confidence interval (CI) 1.41–34.85; p = 0.017). CC status was also significantly associated with in-hospital clinical outcome (p = 0.004). Interreader agreement for CC grading was excellent (Cohen’s κ = 0.910, p < 0.001). Conclusion: Poor preprocedural CC was the most prominent factor independently associated with HeT after MT in patients who achieved complete reperfusion. Assessment of CC on preprocedural BCTA may contribute to postthrombectomy risk stratification; however, the small number of patients with poor CC and the absence of several established prognostic variables warrant cautious interpretation and prospective validation.