Predictive Value of Pooled Cerebral Blood Volume Mapping for Final Infarct Volume in Patients with Major Artery Occlusions. A Retrospective Analysis

Purpose: Detection of ischemic core and collateral status is helpful to predict clinical success of thrombectomy in acute cerebral artery occlusion. Angiographic flat panel detector computed tomography (CT) with pooled blood volume (PBV) data acquisition was described to be helpful to estimate ischemic core in stroke patients prior to thrombectomy and to depict cerebral vessels. We therefore retrospectively evaluated preinterventional PBV data of a large collective of ischemic stroke patients prior to thrombectomy to test its predictive value on final infarct considering PBV maps and collateral status.

Methods: We used PBV data from 101 patients with acute cerebral artery occlusion prior to successful thrombectomy to reconstruct PBV maps and collateral status maps. Suspected ischemic core and collateral status were correlated to final infarct on follow-up multislice CT. Furthermore, the influence of time window and patient age was taken into consideration.

Results: In 75.2% (95% confidence interval CI 66-82%), suspected ischemic core (PBV) matched with final infarct and in 24.8% (95% CI 17-34%) final infarct was overestimated. In all patients, collateral status could be evaluated, and the better the collateral status, the smaller the final infarct (p = 0.016). Although not statistically significant, poor collaterals seem to be a risk factor for overestimation of final infarct on PBV maps. In patients >80 years old predictive value of PBV and collateral status is better than in patients ≤80 years old (p = 0.04). Increasing time window did not have significant impact on predictive value of PBV and collateral status.

Conclusion: The PBV data are useful to expeditiously exclude infarct growth and estimate collateral status prior to thrombectomy after a longer interval between initial multislice CT magnetic resonance imaging (MRI) and intervention. However, because of overestimating final infarct in 25% of patients, PBV data presuming large infarct should not be used as the only basis for excluding patients from effective stroke treatment at this point in time.