Work

Brain edema progression rate on serial CT predicts neurological outcome after cardiac arrest: a region-specific biomarker of hypoxic-ischemic brain injury

Year: 2026

Type: article

Abstract: Objectives To evaluate the prognostic performance of the attenuation change rate on serial brain computed tomography (CT), a novel quantitative marker of temporal cerebral attenuation change, for predicting poor neurological outcome after cardiac arrest (CA). Methods This was a retrospective analysis of a prospective cohort. CA patients who underwent cranial CT both within 12 hours and between 12 hours and one week after return of spontaneous circulation (ROSC) were included. The attenuation change rate, calculated as the HU difference between paired scans divided by the time interval, was measured in 12 predefined brain regions. Neurological outcome was assessed using the Cerebral Performance Category (CPC) at discharge. Results Of 53 patients enrolled, 17 (32.1%) had favorable neurological outcomes at discharge. Patients with poor outcomes exhibited significantly greater attenuation decline rates in all gray matter regions except for gray matter at the centrum semiovale level, including the caudate nucleus (−0.01 vs. 0.02 HU/h), putamen (−0.02 vs. 0 HU/h), thalamus (−0.02 vs. 0.01 HU/h), gray matter of the high convexity (−0.02 vs. 0 HU/h), and cerebellar gray matter (−0.02 vs. 0.03 HU/h) (all P < 0.01), whereas no statistically significant differences were observed in any of the white matter regions. ROC analysis demonstrated AUCs ranging from 0.725 to 0.778, with cerebellar gray matter achieving the highest AUC (0.778; sensitivity, 75.0%; specificity, 76.5%). At the 100% specificity threshold, the caudate nucleus showed the highest sensitivity for predicting poor outcome (36.1% at a decline rate of -0.040 HU/h). Conclusion Serial CT-derived attenuation change rate may serve as an exploratory quantitative marker of cerebral edema progression after cardiac arrest. However, this single-center, small-sample study does not support its use for withdrawal of life-sustaining treatment decisions. Multicenter validation is warranted.

Source: Resuscitation Plus

DOI: 10.1016/j.resplu.2026.101509