Latent class analysis identifies three biologically distinct phenotypes among patients undergoing endovascular thrombectomy: a derivation and temporal validation study
- 1Department of Neurology, Taizhou Hospital of Zhejiang Province, Affiliated to Wenzhou Medical University
Abstract
Background: Despite successful vascular recanalization, approximately half of patients undergoing endovascular thrombectomy (EVT) fail to achieve functional independence at 3 months. The biological determinants of this heterogeneity remain poorly characterized.
Methods: Latent class analysis (LCA) was applied to 825 consecutive EVT-treated patients (development cohort, January 2021–December 2024) using 12 routine post-procedural clinical and laboratory indicators. Phenotype associations with favorable 3-month outcome (modified Rankin Scale 0–2) were assessed using multivariable logistic regression. Reproducibility was evaluated in an independent temporal validation cohort ( n = 211).
Results: LCA identified three phenotypes. The Low-burden phenotype (Class 1, 44.4%) had the lowest inflammatory and coagulatory burden and achieved favorable outcome in 70.8% of patients. The Metabolic phenotype (Class 2, 42.7%), characterized by hyperglycemia and elevated neutrophil-to-lymphocyte ratio, had a favorable outcome rate of 44.9% and symptomatic intracranial hemorrhage (sICH) rate of 7.1%. The hypercoagulable phenotype (Class 3, 13.0%), defined by markedly elevated D-dimer and atrial fibrillation, had a favorable outcome rate of 40.2% and sICH rate of 8.4%. After multivariable adjustment, both Class 2 (OR 0.49, 95% CI 0.35–0.69) and Class 3 (OR 0.41, 95% CI 0.25–0.67) were independently associated with unfavorable outcome (both p < 0.001). Class distribution and outcome associations were reproduced in the validation cohort.
Conclusion: LCA identified three biologically distinct phenotypes among EVT-treated patients with divergent functional outcomes. Multicenter prospective validation is needed before these findings can inform clinical practice.
Related
- Endovascular thrombectomy activity as a candidate marker of organised acute stroke-service capacity: a global burden of disease-based European panel study
- Endovascular thrombectomy for mild vertebrobasilar artery occlusion: a hierarchical win-ratio analysis
- Technical considerations in endovascular thrombectomy for distal vessel occlusions
- Endovascular mechanical thrombectomy for the treatment of extensive postpartum cerebral venous sinus thrombosis: a case report
- Intra-Arterial Tenecteplase After Endovascular Thrombectomy for Acute Stroke
BibTeX
@article{yang2026latent,
title = {{Latent class analysis identifies three biologically distinct phenotypes among patients undergoing endovascular thrombectomy: a derivation and temporal validation study}},
author = {Li-Li Yang and Chao Chen and Li-Li Sun and Meng-Xia Li and Li Han},
journal = {Frontiers in Neurology},
year = {2026},
month = oct,
doi = {10.3389/fneur.2026.1855033},
url = {https://www.frontiersin.org/articles/10.3389/fneur.2026.1855033/full},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Latent class analysis identifies three biologically distinct phenotypes among patients undergoing endovascular thrombectomy: a derivation and temporal validation study
AU - Li-Li Yang
AU - Chao Chen
AU - Li-Li Sun
AU - Meng-Xia Li
AU - Li Han
PY - 2026
DA - 2026/10/15
JO - Frontiers in Neurology
DO - 10.3389/fneur.2026.1855033
UR - https://www.frontiersin.org/articles/10.3389/fneur.2026.1855033/full
AB - Background: Despite successful vascular recanalization, approximately half of patients undergoing endovascular thrombectomy (EVT) fail to achieve functional independence at 3 months. The biological determinants of this heterogeneity remain poorly characterized. Methods: Latent class analysis (LCA) was applied to 825 consecutive EVT-treated patients (development cohort, January 2021–December 2024) using 12 routine post-procedural clinical and laboratory indicators. Phenotype associations with favorable 3-month outcome (modified Rankin Scale 0–2) were assessed using multivariable logistic regression. Reproducibility was evaluated in an independent temporal validation cohort ( n = 211). Results: LCA identified three phenotypes. The Low-burden phenotype (Class 1, 44.4%) had the lowest inflammatory and coagulatory burden and achieved favorable outcome in 70.8% of patients. The Metabolic phenotype (Class 2, 42.7%), characterized by hyperglycemia and elevated neutrophil-to-lymphocyte ratio, had a favorable outcome rate of 44.9% and symptomatic intracranial hemorrhage (sICH) rate of 7.1%. The hypercoagulable phenotype (Class 3, 13.0%), defined by markedly elevated D-dimer and atrial fibrillation, had a favorable outcome rate of 40.2% and sICH rate of 8.4%. After multivariable adjustment, both Class 2 (OR 0.49, 95% CI 0.35–0.69) and Class 3 (OR 0.41, 95% CI 0.25–0.67) were independently associated with unfavorable outcome (both p < 0.001). Class distribution and outcome associations were reproduced in the validation cohort. Conclusion: LCA identified three biologically distinct phenotypes among EVT-treated patients with divergent functional outcomes. Multicenter prospective validation is needed before these findings can inform clinical practice.
SN - 1664-2295
ER -