Delirium is a common acute encephalopathy syndrome characterized by acute or subacute onset of impaired attention and consciousness, with fluctuating cognitive function that may affect the sleep-wake cycle (Stollings et al., 2021, Wilson et al., 2020). There are three typical clinical types of delirium: hyperactive, hypoactive, and mixed delirium (Goldberg et al., 2020). Several factors, including non-interventional predisposing factors, triggers, and pharmacologic influences, can induce delirium, with stroke being an important contributor (Fialho et al., 2021, Stollings et al., 2021). Stroke is one of the primary diseases endangering human health and life, primarily classified into ischemic and hemorrhagic stroke (Herpich and Rincon, 2020). Reports indicate that ischemic strokes account for approximately 80 % of all stroke cases, with about 15 %-20 % of patients admitted to stroke units requiring treatment in intensive care units (ICU)(De Georgia and Patel, 2011, Mansutti et al., 2019). Additionally, ischemic strokes can lead to a range of complications, such as aphasia, stroke-related pneumonia, post-stroke depression, and post-stroke delirium (Barthels and Das, 2020, Campbell et al., 2019, Herpich and Rincon, 2020). These complications pose significant threats to the patient's life and health and also exacerbate the global economic burden of stroke (Bonkhoff et al., 2022). The prevalence of delirium is notably higher in stroke patients, likely due to brain tissue damage and alterations in neurotransmitter levels (Shi et al., 2012). Approximately 10 % to 48 % of patients experience post-stroke delirium (Fleischmann et al., 2023, Klimiec et al., 2016, Shi et al., 2012). This condition is associated with adverse outcomes, including prolonged length of stay, escalated healthcare costs, increased mortality, and permanent cognitive and functional impairment (Ojagbemi and Ffytche, 2016, Oldenbeuving et al., 2011, Qu et al., 2018, Shi et al., 2012). Currently, the treatment of delirium primarily involves pharmaceutical and non-pharmaceutical methods. To prevent or treat delirium, the Society of Critical Care Medicine guidelines recommend a comprehensive non-pharmaceutical strategy (Devlin et al., 2018). Non-pharmaceutical interventions typically address adjustable risk factors, such as regulating the physical environment (e.g., noise or lighting), encouraging early mobility, and including family involvement (Chen et al., 2022). However, because of the complexity and incomplete understanding of delirium's pathophysiology, clinical observation remains the primary method for diagnosis, leading to frequently missed or misdiagnosed cases (Guldolf et al., 2021, Inouye et al., 2014, Ko et al., 2023). International guidelines strongly stress the importance of routinely monitoring delirium through validated tools (Barr et al., 2013, Devlin et al., 2018). Early detection and prevention of delirium are essential for improving outcomes for stroke patients. Novel clinical assessment tools can simplify the evaluation process, enabling earlier identification and management of patients at high risk for delirium.

The Braden Score was developed in 1987 as a bedside assessment tool to predict the risk of skin pressure injuries (pressure ulcers)(Bergstrom et al., 1987). The score covers six domains related to skin integrity and the patient's overall condition. Because it is easy to obtain and does not require laboratory data, it provides a wealth of information for healthcare providers. Recently, researchers found its association with patient frailty, defined as a state of reduced physiological reserve and increased vulnerability to stressors (Afilalo et al., 2014, Bandle et al., 2017). Studies have shown its utility in predicting several medical outcomes, including mortality in the cardiac intensive care unit (Jentzer et al., 2019), hospitalization for heart failure (Bandle et al., 2017), rehabilitation placement after pancreatectomy (Watkins et al., 2019), and the risk of acute kidney injury in patients with acute coronary syndrome (Li et al., 2022). Specifically, Ding et al. conducted a review of 414 ischemic stroke patients and found that the Braden score, with a cutoff of 18, could be an effective tool for predicting post-stroke pneumonia (Ding et al., 2019). Growing evidence suggests that the Braden score has potential beyond its original scope, particularly for assessing complex conditions in critical care settings. This prompted us to explore the association between the Braden score and delirium. If confirmed, the association between Braden score and delirium risk in patients with ischemic stroke could offer a straightforward and efficient method for identifying and intervening in delirium cases.

The primary aim of this study is to examine the relationship between the Braden score and delirium in ischemic stroke patients in the ICU. Our hypothesis, based on clinical experience and literature review (Jentzer et al., 2019), is that there is a significant, potentially inverse relationship between the Braden score and the risk of delirium. As secondary objectives, this study aims to examine the associations between Braden scores and sociodemographic and clinical variables, and to determine the incidence rates of delirium in different risk groups defined by Braden score cutoffs.