Advanced neuromonitoring approaches may improve the detection of neurologic complications in patients in the neurologic or medical intensive care unit compared with standard clinical examination alone.
In the review, investigators synthesized evidence from prior observational studies, clinical trials, and meta-analyses involving patients admitted to the intensive care unit (ICU) with primary neurologic or nonneurologic conditions at risk of experiencing secondary or new-onset neurologic complications. The investigators evaluated monitoring strategies—including continuous electroencephalography (EEG), brain tissue oxygenation, and automated pupillometry—and their associations with neurologic detection, mortality, and clinical outcomes.
Across the studies reviewed, traditional neurologic assessment was frequently insufficient in detecting subtle neurologic changes, particularly in patients who had a lack of consciousness or were sedated or intubated. The investigators noted that up to 40% of patients with impaired consciousness had nonconvulsive status epilepticus that couldn't be detected through routine examination alone, highlighting the role of continuous electroencephalography (EEG) in identifying seizures among comatose patients.
Neurologic complications were also reported among patients hospitalized with a nonneurologic condition. In one cohort referenced in the review, just over 12% of these patients developed neurologic signs such as stroke symptoms or seizures. Delirium rates reached up to 80% among patients receiving mechanical ventilation. Complications like convulsive seizures following cardiac surgery were associated with increased mortality rates and longer hospital stays.
Specific monitoring modalities contributed differently to detection and management. Continuous EEG improved identification of seizures over time, while jugular venous oxygen saturation monitoring plus intracranial pressure monitoring was associated with lower mortality compared with intracranial pressure monitoring alone in patients with traumatic brain injury (23% vs 35%); however, the combination strategy was linked to a higher risk of unfacorable discharge (56% vs 47%). Automated pupillometry provided more precise and reproducible assessment of pupillary responses compared with manual examination with a pen torch and demonstrated strong prognostic accuracy in patients with hypoxic ischemic brain injury.
The investigators emphasized that no single modality was able to capture the full range of neurologic changes. Instead, they described a shift toward multimodal neuromonitoring, integrating electrical, hemodynamic, and metabolic data. The investigators also highlighted emerging applications of artificial intelligence (AI) to process continuous monitoring data, citing models that predicted episodes of elevated intracranial pressure with 77% classification accuracy and provided advance warning prior to clinical deterioration.
The investigators noted several limitations across the evidence base. Most of the studies were observational, limiting causal inference. Much of the literature focused on patients with primary neurologic injury, with less data available for broader ICU populations. In addition, many neuromonitoring tools require specialized expertise and resources, which may limit widespread implementation.
The findings suggested that targeted use of multimodal neuromonitoring may improve detection of neurologic complications in patients admitted to the neurologic or medical ICU, particularly when clinical examination is limited.
"Continuous neuromonitoring can aid in the early detection and management of these conditions, thereby preventing further neurological phenomenon and associated complications,” wrote lead study author Reza Eshraghi, of the Social Determinants of Health Research Center at Isfahan University of Medical Sciences in Iran, and colleagues.
The study authors reported no conflicts of interest and no external funding.
Source: Brain Research Bulletin
