HDLox and neuronal injury biomarkers as early signals of silent strokes after high thromboembolic risk transcatheter procedures
A P Evangeliou, N Pagonas, E Sotiroglou, N Charitakis, M Xenitopoulou, D Kotzadamis, V Schismenos, S Chiotis, P Mamolis, E Pagourelias, G Giannopoulos, Ch Papadopoulos, Ch Karkos, V Vassilikos, S TzikasAbstract
Background
High thromboembolic risk transcatheter procedures are frequently associated with silent ischemic brain lesions detected on diffusion-weighted MRI. Although clinically asymptomatic, these lesions are linked to cognitive decline and adverse long-term outcomes. Early identification of patients at risk remains a clinical challenge. Blood biomarkers may provide valuable prognostic information.
Purpose
To investigate whether peri-procedural changes (Δ = 24 hours post-procedure minus 24 hours pre-procedure) in HDL oxidation (HDLox), neuron-specific enolase (NSE), and neurofilament light chains (NFL) predict silent ischemic brain lesions and early cognitive decline following high thromboembolic risk transcatheter procedures.
Methods
This prospective observational study included 100 patients (55 undergoing carotid artery stenting and 45 undergoing catheter ablation). Plasma HDLox, NSE, and NFL levels were measured 24 hours before and 24 hours after the intervention. NSE and NFLs assays were available only for ablation group, and HDLox for all patients. Biomarker change (Δ) was defined as the difference between post-procedural and pre-procedural levels (24h post – 24h pre). Brain MRI was performed pre- and post-procedurally. Cognitive function was assessed using the Montreal Cognitive Assessment (MoCA) and Mini-Mental State Examination (MMSE), 24h pre - 24h post the procedure. Associations between biomarker Δ and presence of new ischemic lesions were evaluated using non-parametric testing and ROC analysis. Correlations with post procedural cognitive decline were assessed using Spearman coefficients. Analyses were stratified by procedure type.
Results
A total of 100 patients were included. Among those, 64 patients had MRI follow-up and new ischemic lesions were detected in 40.6% (26/64). Peri-procedural increase in HDLox (Δ HDLox) was significantly higher in patients with lesions (p=0.0026), demonstrating good discriminative performance (AUC=0.77). The predictive value was stronger in the carotid stenting subgroup (AUC=0.83; p=0.00196), while weaker in the ablation group (AUC=0.68). Peri-procedural increase in NSE (Δ NSE) was also significantly associated with new lesions in the carotid cohort (p=0.011), whereas ΔNFL showed no significant predictive value. A combined biomarker model yielded an AUC of 0.83 in the carotid subgroup. Regarding cognitive impairment, HDLox Δ was strongly correlated with MoCA decline after the procedure in carotid patients (Spearman ρ=−0.584, p<0.001), as was NSE Δ (ρ=−0.565, p<0.001). No significant association was observed with MMSE in the ablation group.
Conclusion
Peri-procedural increases (24h post – 24h pre) in HDLox and NSE were significantly associated with new MRI-detected lesions and MoCA decline in the overall cohort. The association remained statistically significant in patients undergoing carotid stenting, whereas in the ablation subgroup biomarker changes showed no significant association.TableFigure