Abstract Background and aims Hyperglycemia during the early brain injury (EBI) period after subarachnoid hemorrhage (SAH) is common and correlates with poor outcome. However, the prognostic relevance of the amount of time spent above specific glucose thresholds remains uncertain. We aimed to identify the optimal definition of time above glucose range (TAR) derived from continuous glucose monitoring (CGM) for predicting functional outcome. Methods A prospective cohort of 81 nondiabetic SAH patients were included. CGM devices were placed on admission and kept for 72 hours to track the entire EBI period. TAR was defined as the percentage of glucose readings above predefined thresholds ranging from 100 to 200mg/dL in 5mg/dL increments. Functional outcome was assessed using the modified Rankin Scale (mRS) at 90 days. For each TAR threshold, discriminative performance was evaluated using receiver operating characteristic curve analysis and the maximum Youden index. Ordinal regression models were constructed to assess the association between TAR and functional outcome (mRS shift). Results Among all evaluated thresholds, TAR130 mg/dL had the highest discriminative performance for poor outcome (AUC=0.79), with an optimal cut-off of 20.5% (sensitivity 0.69, specificity 0.85 for mRS 2). Increasing TAR130 mg/dL was associated with higher odds of worse functional outcome (adjustedOR per 10%-increase=1.25, 95%CI=1.01-1.55; p=0.036) in models adjusted by confounders including age and WFNS grade. Conclusions Prolonged exposure to glucose levels above 130 mg/dL during the EBI period, as measured by CGM, is independently associated with worse functional outcome. These data may inform future research on glucose management strategies after SAH. Conflict of interest Daniel Santana: nothing to disclose. Alejandra Mosteiro: nothing to disclose. Inés Bartolomé: nothing to disclose. Emilio Ortega: nothing to disclose. Luigi Zattera: nothing to disclose. Gabriel Pujol: nothing to disclose. Mariano Werner: nothing to disclose. Ramon Torné: nothing to disclose. Laura Llull: nothing to disclose. Sergio Amaro: nothing to disclose.
Santana et al. (Fri,) studied this question.