Background Current clinical decision tools for assessing the risk of symptomatic intracranial hemorrhage (sICH) in patients with vertebrobasilar artery occlusion (VBAO) who received endovascular treatment (EVT) have limited performance. This study develops and validates a clinical risk score to precisely estimate the risk of sICH in VBAO patients. Methods The derivation cohort recruited patients with VBAO who received EVT from the Posterior Circulation IschemIc Stroke Registry in China. Based on the posterior circulation-Alberta Stroke Program Early CT Score (pc-ASPECTS) evaluation method, the cohort was further divided into non-contrast CT (NCCT) and diffusion weighted imaging (DWI) cohorts to construct predictive models. sICH was diagnosed according to the Heidelberg Bleeding Classification within 48 hours of EVT. Clinical signature was constructed in the derivation cohort using machine learning and was validated in two additional cohorts from Asia and Europe. Results We enrolled 1843 patients who underwent EVT and had complete data. pc-ASPECTS of 1710 patients was evaluated on NCCT and 699 patients on DWI. In the NCCT cohort, 1364 individuals made up the training set, of whom 101 (7.4%) developed sICH. In the DWI cohort, the training set consisted of 560 individuals, with 44 (7.9%) experiencing sICH. Predictors of sICH were: glucose, pc-ASPECTS, time from estimated occlusion to groin puncture (EOT), poor collateral circulation, and modified Thrombolysis in Cerebral Infarction (mTICI) score. From these predictors, we derived the weighted poor collateral circulation-EOT-pc-ASPECTS-mTICI-glucose (PEACE) score. The PEACE score showed good discrimination in the training set (area under the curve (AUC)NCCT=0.85; AUCDWI=0.86), internal validation set (AUCNCCT=0.81; AUCDWI=0.82), and two additional external validation set (Asia: AUCNCCT=0.78, AUCDWI=0.80; Europe: AUCNCCT=0.74, AUCDWI=0.78). Conclusion The PEACE score reliably predicted the risk of sICH in VBAO patients who underwent EVT.
Xu, Y., Alexandre, A. M., Pedicelli, A., Huang, X., Wei, M., Zhang, P., Hu, M., Chen, X., Guo, Z., Zhu, J., Chen, H., Ni, C., Fan, L., Wang, R., Wang, Q., Wen, J., Yang, Y., Chu, W., Dai, Z., Tan, S., Broccolini, A., Camilli, A., Abruzzese, S., Cirelli, C., Bergui, M., Romi, D. A., Scarcia, L., Kalsoum, E., Frauenfelder, G., Meder, G., Scalise, S., Ganimede, M. P., Bellini, L., Sette, B. D., Arba, F., Sammali, S., Salcuni, A., Vinci, S. L., Cester, G., Roveri, L., Wang, L., Duan, Z., Zhang, S., Xu, G., Li, S., Liang, Y., Wu, Z., Qin, S., Luo, G., Huang, Z., Xiao, L., Sun, W., Predicting symptomatic intracranial hemorrhage after endovascular treatment of vertebrobasilar artery occlusion: PEACE score, <<JOURNAL OF NEUROINTERVENTIONAL SURGERY>>, 2025; (N/A): N/A-N/A. [doi:10.1136/jnis-2024-022802] [https://hdl.handle.net/10807/307899]
Predicting symptomatic intracranial hemorrhage after endovascular treatment of vertebrobasilar artery occlusion: PEACE score
Pedicelli, AlessandroData Curation
;Broccolini, AldobrandoData Curation
;Camilli, AriannaData Curation
;Abruzzese, SerenaData Curation
;
2025
Abstract
Background Current clinical decision tools for assessing the risk of symptomatic intracranial hemorrhage (sICH) in patients with vertebrobasilar artery occlusion (VBAO) who received endovascular treatment (EVT) have limited performance. This study develops and validates a clinical risk score to precisely estimate the risk of sICH in VBAO patients. Methods The derivation cohort recruited patients with VBAO who received EVT from the Posterior Circulation IschemIc Stroke Registry in China. Based on the posterior circulation-Alberta Stroke Program Early CT Score (pc-ASPECTS) evaluation method, the cohort was further divided into non-contrast CT (NCCT) and diffusion weighted imaging (DWI) cohorts to construct predictive models. sICH was diagnosed according to the Heidelberg Bleeding Classification within 48 hours of EVT. Clinical signature was constructed in the derivation cohort using machine learning and was validated in two additional cohorts from Asia and Europe. Results We enrolled 1843 patients who underwent EVT and had complete data. pc-ASPECTS of 1710 patients was evaluated on NCCT and 699 patients on DWI. In the NCCT cohort, 1364 individuals made up the training set, of whom 101 (7.4%) developed sICH. In the DWI cohort, the training set consisted of 560 individuals, with 44 (7.9%) experiencing sICH. Predictors of sICH were: glucose, pc-ASPECTS, time from estimated occlusion to groin puncture (EOT), poor collateral circulation, and modified Thrombolysis in Cerebral Infarction (mTICI) score. From these predictors, we derived the weighted poor collateral circulation-EOT-pc-ASPECTS-mTICI-glucose (PEACE) score. The PEACE score showed good discrimination in the training set (area under the curve (AUC)NCCT=0.85; AUCDWI=0.86), internal validation set (AUCNCCT=0.81; AUCDWI=0.82), and two additional external validation set (Asia: AUCNCCT=0.78, AUCDWI=0.80; Europe: AUCNCCT=0.74, AUCDWI=0.78). Conclusion The PEACE score reliably predicted the risk of sICH in VBAO patients who underwent EVT.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
