By Peter B. Forgacs, MD
Assistant Professor of Neuroscience and Neurology, Feil Family Brain and Mind Research Institute and Department
of Neurology, Weill Cornell Medical College
Dr. Forgacs reports no financial relationships relevant to this field of study.
In this retrospective cross-sectional study involving more than 7 million critically ill adult patients from the National Inpatient Sample database, researchers showed that the use of continuous electroencephalography is associated with lower in-hospital mortality.
Hill CE, Blank LJ, Thibault D, et al. Continuous EEG is associated with favorable hospitalization outcomes for critically ill patients. Neurology 2019;92:e9-e18. doi:10.1212/WNL.0000000000006689.
Continuous electroencephalography (cEEG) monitoring has been used increasingly to monitor brain function of patients admitted to intensive care units (ICUs). cEEG offers real-time, noninvasive, and direct assessment of ongoing brain activity. Therefore, it plays a unique and well-established role in diagnosing ongoing or fluctuating functional cerebral disturbances, such as seizures/status epilepticus (including seizures without overt clinical symptoms) and encephalopathies, in the detection of delayed cerebral ischemia after subarachnoid hemorrhage and in assessment of prognosis after severe brain injuries, including post-cardiac arrest hypoxic brain injury. However, the relationship between the use of cEEG and hospital outcomes, including mortality, is not well-characterized.
Hill et al designed this study to characterize the association between cEEG use and outcomes for specific diagnostic categories using administrative claims data recorded between 2004 and 2013 in the National Inpatient Sample. More than 7 million critically ill patients were identified in the database. Over 10 years, cEEG use increased more than 10-fold, but cEEG remains underused; overall, only 0.3% of the critically ill patients included in the study underwent cEEG monitoring at any point during their hospitalization.
Importantly, while patients in the cEEG cohort appeared to be more ill with higher comorbidity scores, more frequent palliative care consultation, and longer length of stay, cEEG use was associated with decreased in-hospital mortality (22.8% of cEEG cohort vs. 27.8% for no cEEG). In subgroups divided by diagnosis, a similar relationship between cEEG use and mortality was observed in patients with subarachnoid or intracerebral hemorrhage (26.3% vs. 54.0%) and altered levels of consciousness (21.5% vs. 28.3%), but no significant correlation was found in patients with seizures/status epilepticus (18.2% vs. 19.8%). The authors postulated that patients who had a lethal primary diagnosis leading to status epilepticus did not benefit from further monitoring. The use of cEEG was associated with an increase in total cost of hospitalization and longer length of stay. Patients admitted to large or urban teaching hospitals located in the Northeast or Midwest were more likely to undergo cEEG monitoring than patients who were admitted to non-teaching or rural hospitals or in other regions of the country.
COMMENTARY
Hill et al documented that the use of cEEG in critically ill patients is strongly associated with decreased in-hospital mortality, but it remains underused based on data from an administrative claims database. Despite the demonstrated benefit in reduced mortality in ICU patients, many barriers hinder the wider implementation of cEEG monitoring in critically ill patients. The overwhelming majority of U.S. hospitals with ICUs currently do not have the infrastructure to perform cEEG studies. Despite the 10-fold increase in cEEG use over the nine-year period analyzed in this study, within the 3,054 unique hospitals involved, 93.9% never used cEEG. Major barriers to more widespread use of cEEG include the cost of acquiring and maintaining EEG equipment and providing 24-hour availability of technologists and specialized neurologists.
The scarcity of cEEG use among all hospitals represents a limitation in the interpretation of the study results. It is possible that the observed mortality benefit is related to unmeasured characteristics of the hospitals with cEEG availability (i.e., the differences in management strategies in larger volume, urban hospitals in academic centers). These hospitals also are more likely to have dedicated neurocritical care units with specialized teams using advanced therapeutic approaches and technologies. Other limitations include similarities to all studies using claims databases and the dependence of the study on billing codes used for administrative purposes; it cannot be determined if coding practices may have influenced the results. Finally, long-term functional outcomes, such as level of independence and quality of life, could not be assessed in this study; such measures are just as important for patients and caregivers as in-hospital mortality.
Nevertheless, the Hill et al study strongly highlights the potential benefits of continuous neurological monitoring of electrical brain activity in critically ill patients. Further prospective studies should aim to assess the value of conventional and newer electrophysiological brain monitoring techniques in all ICU patients, similar to cardiac monitoring in intensive care settings worldwide. Many new technologies offer cost savings in cEEG monitoring. More affordable, fast, secure, and reliable remote access of studies may allow physicians to cover multiple lower-volume hospitals at the same time. In addition, many ongoing studies use a limited number of EEG channels and easily apply electrode caps to assess the accuracy of simpler techniques in acute and intensive care settings. Furthermore, automatic seizure detection algorithms and newer visualization methods using quantitative EEG features may allow quick bedside assessment of large segments of EEG data by nurses or other ICU staff.
In the future, deep learning algorithms using artificial intelligence may offer automated assessment, further decreasing the cost of monitoring and analyzing large amounts of data. Hospitals with ICUs currently not equipped to perform cEEG studies should be encouraged to implement neurological monitoring capabilities to expand these benefits to all critically ill patients.
