New clinical tool identifies patients likely to have seizures
UW Health News Dec 04, 2017
A new scoring system developed by a UW School of Medicine and Public Health neurologist may help physicians identify which critically ill patients are likely to have seizures.
Dr. Aaron Struck, assistant professor of neurology, identified patterns of electrical behavior that predict which patients are more likely to have seizures by reviewing data from more than 5,400 patients hospitalized for conditions other than epilepsy at three large academic medical centers. The study was published online in the journal JAMA Neurology.
Patients who are critically ill frequently have seizures. Struck explains that more than 75% of the time, doctors cannot detect the seizures by observing the patient, so special brain monitoring with electroencephalography (EEG) is necessary.
“Undetected seizures are associated with worse patient outcomes,” he said. “This tool helps doctors figure out who is at greatest risk of having seizures so they can decide on appropriate work-up and treatment. The goal is to improve cognitive outcomes in patients with critical illness.”
Struck and his co-authors created something they call the 2HELPS2B score that predicts which patients are likely to have seizures and may need monitoring and medication. The name comes from 2Hz (frequency greater than two cycles per second); epileptiform discharges; lateralized periodic discharges; plus features (such as rhythmic, sharp or fast brain activity); seizure history; and 2B, which adds two points for brief rhythmic discharges.
The researchers used a new machine-learning method called RiskSLIM that is designed to produce accurate, risk-calibrated scoring systems with a limited number of variables and small integer weights. They validated the accuracy and risk calibration of the model and compared its performance with models built with state-of-the-art logistic regression methods. The patient data came from the Critical Care EEG Research Consortium and used data collected over three years at Emory University Hospital, Brigham and Women's Hospital, and Yale University Hospital.
Go to Original
Dr. Aaron Struck, assistant professor of neurology, identified patterns of electrical behavior that predict which patients are more likely to have seizures by reviewing data from more than 5,400 patients hospitalized for conditions other than epilepsy at three large academic medical centers. The study was published online in the journal JAMA Neurology.
Patients who are critically ill frequently have seizures. Struck explains that more than 75% of the time, doctors cannot detect the seizures by observing the patient, so special brain monitoring with electroencephalography (EEG) is necessary.
“Undetected seizures are associated with worse patient outcomes,” he said. “This tool helps doctors figure out who is at greatest risk of having seizures so they can decide on appropriate work-up and treatment. The goal is to improve cognitive outcomes in patients with critical illness.”
Struck and his co-authors created something they call the 2HELPS2B score that predicts which patients are likely to have seizures and may need monitoring and medication. The name comes from 2Hz (frequency greater than two cycles per second); epileptiform discharges; lateralized periodic discharges; plus features (such as rhythmic, sharp or fast brain activity); seizure history; and 2B, which adds two points for brief rhythmic discharges.
The researchers used a new machine-learning method called RiskSLIM that is designed to produce accurate, risk-calibrated scoring systems with a limited number of variables and small integer weights. They validated the accuracy and risk calibration of the model and compared its performance with models built with state-of-the-art logistic regression methods. The patient data came from the Critical Care EEG Research Consortium and used data collected over three years at Emory University Hospital, Brigham and Women's Hospital, and Yale University Hospital.
Only Doctors with an M3 India account can read this article. Sign up for free or login with your existing account.
4 reasons why Doctors love M3 India
-
Exclusive Write-ups & Webinars by KOLs
-
Daily Quiz by specialty -
Paid Market Research Surveys -
Case discussions, News & Journals' summaries
