Neurocritical care
-
Big data (BD) and artificial intelligence (AI) have increasingly been used in neurocritical care. "BD" can be operationally defined as extremely large datasets that are so large and complex that they cannot be analyzed by using traditional statistical modeling. "AI" means the ability of machines to perform tasks similar to those performed by human intelligence. We present a brief overview of the most commonly applied AI techniques to perform BD analytics and discuss some of the recent promising examples in the field of neurocritical care. The latter include the following: cognitive motor dissociation in disorders of consciousness, hypoxic-ischemic injury following cardiac arrest, delayed cerebral ischemia and vasospasm after subarachnoid hemorrhage, and monitoring of intracranial pressure. ⋯ These collaborations will allow us to share data, combine predictive algorithms, and analyze multiple and cumulative sources of data retrospectively and prospectively. Once AI algorithms are validated at multiple centers, they should be tested in randomized controlled trials investigating their impact on clinical outcome. The neurocritical care community must work to ensure that AI incorporates standards to ensure fairness and health equity rather than reflect our biases present in our collective conscience.
-
Large and complex data sets are increasingly available for research in critical care. To analyze these data, researchers use techniques commonly referred to as statistical learning or machine learning (ML). The latter is known for large successes in the field of diagnostics, for example, by identification of radiological anomalies. ⋯ To conclude, there are important opportunities but also pitfalls to consider when performing clustering or predictive studies with ML techniques. We advocate careful valuation of new data-driven findings. More interaction is needed between the engineer mindset of experts in ML methods, the insight in bias of epidemiologists, and the probabilistic thinking of statisticians to extract as much information and knowledge from data as possible, while avoiding harm.
-
Beta-lactam neurotoxicity is a relatively uncommon yet clinically significant adverse effect in critically ill patients. This study sought to define the incidence of neurotoxicity, derive a prediction model for beta-lactam neurotoxicity, and then validate the model in an independent cohort of critically ill adults. ⋯ In this single center cohort of critically ill patients, beta-lactam neurotoxicity was demonstrated less frequently than previously reported. We identified obesity as a novel risk factor for the development of neurotoxicity. The prediction model needs to be further refined before it can be used in clinical practice as a tool to avoid drug-related harm.
-
Meta Analysis
Safety and Efficacy of Vinpocetine as a Neuroprotective Agent in Acute Ischemic Stroke: A Systematic Review and Meta-Analysis.
Vinpocetine as a neuroprotective agent is effective in acute ischemic stroke in some randomized controlled trials (RCTs). Since the last systematic review has been published in 2008, which didn't find conclusive evidence favoring its use, two more RCTs have also been completed. ⋯ Vinpocetine has some promising efficacy in patients with ischemic stroke when used in the acute stage in reducing the disability, but presently there is not enough evidence to suggest that it also reduces case fatality. More double-blind, placebo-controlled RCTs of adequate sample size are needed before making recommendations for the routine administration of vinpocetine for all patients with acute ischemic stroke.