Seizures that occur in patients admitted to the hospital for nonseizure reasons are likely to recur during the hospital stay, usually carry a high risk of mortality and morbidity, and often are not treated optimally, according to a report published online Jan. 14 in JAMA Neurology.
In a retrospective study of 218 patients who developed seizures when they were inpatients at medical, surgical, or emergency departments at two New York hospitals, the overall rate of death or discharge to hospice care was 14%, said Dr. Madeline C. Fields of the department of neurology at Mount Sinai School of Medicine, New York, and her associates.
Dr. Madeline Fields
The investigators undertook this study to describe the epidemiology of hospital-onset seizures because until now, no studies in the United States have done so. No management guidelines are available for clinicians, and even the usefulness of antiepileptic drugs (AEDs) is uncertain. "This is surprising considering that hospital-onset seizures are dramatic events that often lead to increased intensity of medical care, consultations, and prolonged hospital stays," they noted.
The Agency for Healthcare Research and Quality estimates that seizures develop during approximately 1.4 million hospitalizations each year, or about 4% of all annual hospitalizations. When they develop in people with no history of seizure, it is usually because of stroke, infection, or metabolic disturbances. When they develop in people with a history of seizure, it is thought to be because stress, medication, sleep deprivation, fever, or other factors related to the hospitalization exacerbate the underlying seizure disorder.
Dr. Fields and her colleagues reviewed 1 year of medical records from a large nonprofit and a large municipal hospital affiliated with New York University, identifying 218 cases in which adults admitted for nonseizure indications developed hospital-onset seizures. Most of these cases (64%) occurred in patients without a history of seizures.
Among the 79 (36%) patients with a history of seizures, 16 (20%) were not taking AEDs at hospitalization, 32 (41%) were taking a single AED, and 31 (39%) were taking two or more AEDs.
Most patients (61%) in the entire study population had multiple seizures during the hospital stay. A total of 39% had seizures on multiple days, while another 22% had multiple seizures during a single day. Close to half (43%) of the patients who had new-onset seizures had recurrences during the hospital stay, as did 32% of patients who had a history of seizure.
These high rates of seizure recurrence "have never been reported and may be important" for clinicians trying to manage such cases, the researchers said (JAMA Neurol. 2013 Jan. 14 [doi:10.1001/2013.jamaneurol.337]).
Status epilepticus occurred in 8% of patients overall and was the index seizure in 6% overall.
The most common types of seizure were generalized tonic-clonic convulsions (33%) and complex partial seizures (21%). This was true in both patients with no history of seizures and in those with known seizure disorders.
The most frequent identifiable reasons for the seizures, other than exacerbation of an existing seizure disorder, were stroke, metabolic derangement, and brain tumor. Stroke was the most common etiology in patients who had multiple seizures during a single day, whereas metabolic abnormalities were the most common etiology in patients who had isolated seizures on multiple days.
Death or discharge to hospice care was considered "common" in this study, occurring in 14% of patients overall. Among patients with no history of seizures, this rate was even higher at 19%. This outcome occurred more often in patients who had recurrences during their stay (21%) than in those who had only a single seizure while hospitalized (10%).
In other settings, newer AEDs are recommended for patients with comorbidities because they are less likely than are older agents to interact with other medications, less likely to provoke adverse reactions, and do not carry the risk of protein-binding abnormalities that older agents do. However, in this study, older AEDs were much more likely to be prescribed.
Phenytoin was the first-line choice in half of the patients who had no history of seizures, as well as in 28% of those who had a history of seizures.
Moreover, "phenytoin was not always used in a manner commensurate with current standards," Dr. Fields and her associates wrote.
The loading dose of IV phenytoin often was not individualized to the patient, resulting in a subtherapeutic dose in 21% of cases and a supratherapeutic dose in 9%. The loading dose was not checked in another 29% of cases.
In addition, 26% of the patients discharged from the hospital were given prescriptions for phenytoin at that time. It may be that selection of this hepatic-enzyme-inducing drug was "predicated on acute choice rather than consideration of the consequences of long-term therapy in what is most likely an ill population receiving concomitant medications whose metabolism could be affected by phenytoin," the researchers wrote.