Abstract Background and aims Endovascular thrombectomy restores cerebral perfusion but may cause hemorrhagic reperfusion injury. We investigated whether chronic hyperglycemia, reflected by elevated HbA1c, increases this risk. Methods We performed a post hoc analysis of a prospective registry of consecutive adults treated with endovascular thrombectomy for anterior circulation large vessel occlusion between 1/2017 and 1/2023 at a tertiary stroke center serving as the hub of a regional telemedicine network with 13 referring hospitals in Germany. Patients were categorized by HbA1c (normal, prediabetes, diabetes). Intracranial hemorrhage was classified using the Heidelberg Bleeding Classification. Causal effects were estimated using double-robust inverse probability weighted regression with adjustment for demographic, clinical, imaging, and procedural confounders. Dose-response relationships and robustness were assessed using continuous HbA1c models and sensitivity analyses. Results Among 983 patients, intracranial hemorrhage occurred in 35.5% (hemorrhagic infarction 24.9%, parenchymal hematoma 7.4%). After weighting with excellent covariate balance, diabetic HbA1c levels (≥6.5%) were associated with a significantly higher risk of hemorrhagic reperfusion injury compared with normal HbA1c (5.7%) (Average Treatment Effect 14.7%, 95% CI 3.9–25.6%, p=0.008), corresponding to an absolute risk increase from 32.9% to 47.6% (Number-Needed-to-Harm=6). Dose-response analyses demonstrated a progressive rise in hemorrhage risk at HbA1c values above the diabetes threshold (Figure 1). These findings were consistent across sensitivity analyses using alternative HbA1c cut points, propensity score matching, and adjusted regression models. Conclusions Elevated HbA1c is associated with a dose-dependent increase in intracranial hemorrhage after thrombectomy, identifying chronic hyperglycemia as a clinically relevant and potentially modifiable risk factor of cerebral reperfusion injury. Conflict of interest "Norma J. Diel: nothing to disclose", "Martin Arndt: nothing to disclose", " Isabella Püschel: nothing to disclose", "Erik Simon: nothing to disclose", "Kristian Barlinn: nothing to disclose", "Daniel P. O. Kaiser: nothing to disclose", "Hagen B. Huttner: nothing to disclose", "Annahita Sedghi: nothing to disclose", "Timo Siepmann: nothing to disclose" Figure 1 - belongs to Results
Diel et al. (Fri,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: