By Kimberly Pargeon, MD
Assistant Professor of Clinical Neurology, Weill Cornell Medical College
Dr. Pargeon reports no financial relationships relevant to this study.
SYNOPSIS: In this retrospective review of mortality and hospitalization related to status epilepticus from 1999 to 2010, the authors found that the overall mortality remained relatively stable, whereas the rate of related hospitalizations significantly increased, particularly in patients who were intubated and in whom status epilepticus was not the primary diagnosis.
SOURCE: Betjemann JP, et al. Trends in status epilepticus-related hospitalizations and mortality: Redefined in U.S. practice over time. JAMA Neurol 2015;72:650-655.
Status epilepticus (SE) is a common neurological emergency that can have significant effects on morbidity, mortality, and healthcare costs. Although definitions have changed, SE is presently defined as a prolonged seizure or multiple seizures with incomplete return to baseline function lasting longer than 5 minutes. The incidence in the United States is about 10-41 per 100,000 people with higher rates among some ethnic minorities. 1 The overall reported mortality in adults is about 20%, with worse outcomes seen in older adults.
SE can be further categorized as either convulsive or nonconvulsive, although nonconvulsive SE is often not well-defined in the literature. Recent efforts have focused on early seizure termination, typically with pre-hospital benzodiazepines, and early seizure detection, particularly for nonconvulsive events, through the increased use of continuous EEG monitoring.
Betjemann et al sought to examine trends in SE-related hospitalizations and mortality in the United States using retrospective data from large national administrative databases, from January 1999 to December 2010, for 408,304 hospitalized patients. The overall and age-standardized mortality rates were determined using data from the Centers for Disease Control and Prevention, and population-adjusted hospitalization rates were determined using the Nationwide Inpatient Sample.
They further categorized hospitalizations as to whether SE was the principal or secondary diagnosis, whether the patient was intubated, and by insurance type. The use of video-EEG monitoring was identified using a specific ICD-9 procedure code.
The findings show
One of the primary findings was that overall mortality from SE (two deaths per 1,000,000 persons) remained relatively stable from 1999 to 2010, particularly when it was the primary or underlying cause of death, increasing by only 5.6%. However, the rate of related hospitalizations increased by 56.2%, with the most significant increase seen for patients with a secondary diagnosis of SE as compared to a primary diagnosis (102.0% as compared to 33.4%).
The use of video-EEG in SE-diagnosed patients also increased from 1.1% in 1999 to 4.3% in 2010, with the largest relative increase among intubated patients with a secondary diagnosis. Hospital discharges were also categorized by diagnosis type (principal vs secondary) and whether the patient was intubated. Although each of the four categories demonstrated an increase, the largest increase (181.8%) was seen again in intubated patients with a secondary diagnosis.
The investigators also found that SE hospitalizations increased for all insurance types from 1999 to 2010, but initially remained relatively stable for Medicare until an abrupt 81.1% increase occurred around 2005.
COMMENTARY
Much of the previous epidemiology of SE has been based on observational studies with relatively small populations with limited generalizability. The current study, however, used national administrative databases to evaluate trends in diagnosis and mortality over a 12-year span. The authors found overall mortality related to SE was relatively stable, while related hospitalizations significantly increased, particularly after 2005, and most dramatically in intubated patients whose primary diagnosis was not SE.
The authors postulate that the findings were likely attributable to the changes in coding practices, particularly after 2007, when coding rules and incentives changed for Medicare and Medicaid, and increased detection through the broader use of video-EEG monitoring. Thus, we are identifying more patients with nonconvulsive seizures and nonconvulsive SE, particularly in critically ill patients. Although overall mortality may not be significantly increased, nonconvulsive SE may be a marker of underlying brain injury leading to increased morbidity.
The authors also note that the related mortality was significantly lower in their study, at about 0.5-2%, as compared to previously reported mortality rates, as high as 20%. This difference highlights a shortcoming in our present coding system. The authors used ICD-9 codes for one of the databases, specifically 345.3 for grand mal status epilepticus and 345.2 for petit mal status epilepticus, the latter technically intended for absence status epilepticus. Many patients with secondary diagnoses in non-neurological ICUs could be alternately coded as seizure NOS (780.39) or altered mental status (780.02), so this study may underestimate the actual incidence of SE.
Although retrospective and possibly underestimating the true incidence of SE-related mortality, this study highlights a shift in our practice to better identify SE, particularly nonconvulsive events, through the increased use of continuous EEG monitoring. However, we know that both convulsive and nonconvulsive SE can lead to extensive physiologic and neuronal damage, so mortality rates alone may paint a limited picture, particularly in critically ill patients.
REFERENCE
- Dham BS, et al. The epidemiology of status epilepticus in the United States. Neurocrit Care 2014;20:476-483.
