Efficacy of endovascular treatment for patients with acute large vessel occlusion stroke from the Western Sichuan Plateau and machine learning prediction models: a prospective study protocol

Prediction models DON'T GET SURVIVORS RECOVERED! Useless. Or are you blind to facts?

 Efficacy of endovascular treatment for patients with acute large vessel occlusion stroke from the Western Sichuan Plateau and machine learning prediction models: a prospective study protocol

Chuanxi Duan^1†Yu Hu^2,3†Yuding Luo³Jiali Zhang³Pingchuan Liu^2,3Junhao Li^2,3Hai Xiong^2,3Mengling Duan¹Yupeng Niu⁴Ke Huang³Zhao Chen³Jian Wang^2,3*

• ¹North Sichuan Medical College, Nanchong, China
• ²Department of Neurology, The Affiliated Hospital, Southwest Medical University, Luzhou, China
• ³Department of Neurology, Ya'an People's Hospital, Ya'an, China
• ⁴Sichuan Agricultural University, Ya'an, China

Objectives: Stroke is the second leading cause of death and the third leading cause of disability among non-communicable diseases globally. The prevalence, incidence, and mortality rates of stroke are higher in high-altitude regions compared to lowland areas. However, compared to plain areas, the efficacy of endovascular therapy for large vessel occlusive acute ischemic stroke (LVO-AIS) in high-altitude regions remains unclear.

Methods and Design: This study is a multicenter, prospective, endpoint-blinded cohort study. From January 2025 to December 2027, a total of 1,052 patients with acute large vessel occlusion ischemic stroke (LVO-AIS) from the Western Sichuan Plateau will be prospectively enrolled, including those who receive endovascular treatment and those who do not. Baseline characteristics and endovascular treatment details will be documented. Treatment decisions are guided by clinical practice guidelines, taking into account high-altitude real-world constraints such as patient or proxy refusal and delays in interhospital transfer. Medical records will be established for each patient, and a 180-day follow-up will be conducted. The primary outcome was the proportion of patients achieving functional independence [modified Rankin scale (mRS) range from 0 to 2] at 90 days. The secondary outcomes included the mRS score at 90 days, early neurological improvement rate [defined as a National Institutes of Health Stroke Scale (NIHSS) score of 0–2 or a reduction of ≥8 points from baseline within 24 h of enrollment], changes in NIHSS scores between day 7 ± 1 or discharge and baseline, quality of life as assessed by the five-level EuroQol five-dimensional questionnaire at 90 days, and cognitive function at 180 days will be assessed using mini-mental state examination and montreal cognitive assessment scores. Imaging outcomes will include the rate of successful reperfusion (defined as a modified Thrombolysis in Cerebral Infarction score ≥2b) and infarct volume measured within 5–7 days. Statistical analysis and fused optimized multimodal learning were blinded to the group assignments.

Conclusion: This study aims to evaluate the efficacy of endovascular treatment compared with standard medical therapy in patients with LVO-AIS in the Western Sichuan Plateau and to develop an artificial intelligence–based prognostic model to refine treatment strategies for this and other high-altitude regions.

Clinical trial registration: https://www.chictr.org.cn/showproj.html?proj=241870, ChiCTR2400092762.