Meta-analysis of diffusion-weighted imaging in detecting postoperative ischemic brain injury in patients with carotid artery stenosis

Objective This work systematically evaluated the incidence and influencing factors of new ischemic brain lesions detected by diffusion-weighted imaging (DWI) following carotid endarterectomy (CEA) or carotid artery stenting (CAS) in patients with CA stenosis through a systematic review and meta-analysis. Methods Relevant studies on new ischemic brain lesions detected by DWI after CEA or CAS were retrieved from multiple databases. After literature screening and quality assessment, eleven eligible studies were included. Meta-analysis of key indicators, including the incidence of DWI-detected ischemic lesions and the postoperative DWI-positive rate, was performed using Stata MP18.0 and RevMan5.3 . Results The pooled estimate of the detection rate of new DWI lesions per post-operative scan was 58.64 % (95 % CI: 44.30 %–72.98 %) under the fixed-effect model, with moderate heterogeneity among studies ( I 2 = 49.0 %, P = 0.047). Subgroup analysis showed that the incidence of DWI lesions after CAS was 37.84 % (95 % CI: 24.68 %–51.01 %), while after CEA it was 27.05 % (95 % CI: 25.10 %–28.99 %). Among CAS studies reporting the use of embolic protection devices (EPD), the pooled risk difference was 93.82 (95 % CI: 45.34–142.31), with no heterogeneity ( I 2 = 0.0 %). The pooled effect size for the analysis of DWI lesion characteristics was 58.50 (95 % CI: 28.59–88.40), also with no heterogeneity among studies ( I 2 = 0.0 %). Conclusion This meta-analysis confirmed that new DWI ischemic lesions are common after carotid revascularization, with incidence varying by surgical method (higher in CAS than in CEA). Despite considerable heterogeneity, DWI appears effective for detecting acute postoperative cerebral ischemic injury. Its routine inclusion in assessment protocols should be considered alongside standardization efforts to reduce variability in imaging and procedural protocols.