Performance and clinical relevance of continuous glucose monitoring during early brain injury in subarachnoid hemorrhage

Glycemic dysregulation after spontaneous subarachnoid hemorrhage (SAH) has been associated with poor outcomes, potentially through glucose-driven neurotoxicity. This study assessed the performance of Continuous Glucose Monitoring (CGM) during the Early Brain Injury (EBI) phase and explored associations between glucose metrics, functional outcome, and serum Neurofilament Light Chain (NFL) levels as a surrogate biomarker of neuroaxonal injury. In this prospective single-center study, 65 non-diabetic SAH patients admitted within 24 h of onset underwent 72-hour CGM and finger-prick glucose testing every six hours. CGM agreement with finger-prick was evaluated through Median Absolute Relative Difference (MARD) and Clarke Error Grid. Multivariate models assessed associations between glucose metrics, 3-month modified Rankin Scale (mRS) outcomes, and NFL levels at 72 h. CGM showed acceptable agreement (MARD 14.4%) with 95.4% of paired values in Clarke Error Grid zones A/B. Poor outcome (mRS > 2) occurred in 35% of patients and was associated with higher glucose burden and variability. Several CGM-derived metrics, but not finger-prick measures, correlated with elevated NFL levels, particularly reduced time in range and increased glycemic variability. CGM is a feasible and well-performing tool during EBI after SAH, providing a sensitive assessment of early glycemic disturbances potentially linked to neuroaxonal injury and prognosis.