4.1, FileMaker, Inc., Santa Clara, CA). Descriptive statistics are reported. RESULTS. Demographics. During the study there were. 31,508 patient visits in the 12 ...
622
SEIZURES
Huff et al. • SEIZURES IN ED PATIENTS
Emergency Department Management of Patients with Seizures: A Multicenter Study J. STEPHEN HUFF, MD, DEXTER L. MORRIS, MD, PHD, RASHMI U. KOTHARI, MD, MICHAEL A. GIBBS, MD, FOR THE EMERGENCY MEDICINE SEIZURE STUDY GROUP (EMSSG)*
Abstract. Objectives: Patients with seizure disorders are common in the emergency department (ED), yet little is known regarding the management of these patients. This study was performed to define the frequency of patients with seizure disorders in the ED patient population and to determine possible seizure etiologies, characteristics of diagnostic activities, treatments, and dispositions. Methods: Twelve EDs monitored all patients with a chief complaint related to seizure disorders presenting over 18.25 days (5% of the calender year) in late 1997. Retrospective chart review was used to gather specific data regarding these consecutive cases. Results: Of the 31,508 patients who presented to these 12 EDs during the study period, 368 (1.2%) had complaints related to seizures. Three hundred sixty-two charts were available for analysis and make up the study population. Two hundred fifty-seven (71%) utilized emergency medical services (EMS) for transport and care. Ad-
From the Department of Emergency Medicine, University of Virginia Medical Center, Charlottesville, VA (JSH); Department of Emergency Medicine, University of North Carolina Health Care System, Chapel Hill, NC (DLM); Department of Emergency Medicine, University of Cincinnati Medical Center, Cincinnati, OH (new affiliation: Borgess Research Institute, Kalamazoo, MI) (RUK); and Department of Emergency Medicine, Carolinas Medical Center, Charlotte, NC (MAG). Received October 3, 2000; revision received December 28, 2000; accepted January 8, 2001. Presented in preliminary form at the SAEM Mid-Atlantic Regional meeting, Charlottesville, VA, April 1998, and at the annual meeting, American College of Emergency Physicians, San Diego, CA, May 1998. Address for correspondence and reprints: J. Stephen Huff, MD, Department of Emergency Medicine, P.O. Box 800699, University of Virginia Medical Center, Charlottesville, VA 229080699. Fax: 804-243-2924; e-mail: jshuff @ virginia.edu *Study group principal investigators at each site (listed alphabetically by investigator name): Michael A. Gibbs, MD, Carolinas Medical Center; J. Stephen Huff, MD, University of Virginia; Andy Jagoda, MD, Mount Sinai Medical Center; Rashmi U. Kothari, MD, University of Cincinnati Medical Center; Jonathan A. Maisel, MD, Bridgeport Hospital and Yale University School of Medicine; Moss H. Mendelson, MD, Maryview Medical Center and Eastern Virginia Medical School; Dexter L. Morris, MD, PhD, University of North Carolina Health Care System; Edward Sloan, MD, Our Lady of Resurrection Medical Center and University of Illinois College of Medicine; Corey M. Slovis, MD, Vanderbilt University Medical Center; Rodney W. Smith, MD, St. Joseph Mercy Hospital and Michigan State University; and Jerald A. Solot, MD, University of Pittsburgh Medical Center, Shadyside Hospital.
vanced care, including intravenous access, laboratory work, cardiac monitoring, or oxygen administration, was utilized in 304 (84%) patients. Antiepileptic drugs were given in 199 (55%) patients. Ethanol withdrawal or low antiepileptic drug levels were implicated as contributing factors in 177 (49%) of patients. New-onset seizures were thought to be present in 94 (26%) patients. Status epilepticus occurred in only 21 (6%) patients. Ninety-eight (27%) of all patients were admitted to the hospital. Conclusions: Patients with presenting complaints related to seizures are frequent in the ED population and make considerable demands on EMS and ED resources. Six percent of patients with seizure-related presentations were in status epilepticus and more than a fourth of all patients required hospitalization. Key words: seizures; epilepsy; status epilepticus; emergency medical services. ACADEMIC EMERGENCY MEDICINE 2001; 8:622–628
P
ATIENTS with seizures and presenting complaints related to seizure disorders are frequent in the emergency department (ED) population, but little is known of their demands on emergency medical services (EMS) and ED resources. The acuity of problems ranges from status epilepticus to requests for medication refills. Surprisingly, little research has been performed on seizure disorders in the emergency patient population; one study describes this population from the perspective of a single ED.1 This study was undertaken to determine the frequency of patients with a chief complaint related to seizures at a broad range of institutions and to determine characteristics of clinical care provided, including diagnostic tests, treatment actions, and disposition. Investigators were instructed to cast a wide net to capture all patients with presenting complaints related to seizures or a seizure disorder. The widest spectrum of patient presentations was to be included for study, ranging from active seizures to requests for medication refills. The project was a descriptive study providing a snapshot in time of seizure care in diverse EDs. To the best of our knowledge, this is the only multicenter study to date that attempts to describe the needs of patients with seizures in EDs.
623
ACADEMIC EMERGENCY MEDICINE • June 2001, Volume 8, Number 6
METHODS Study Design. Investigators were solicited from the Neurologic Emergencies Interest Group of the Society for Academic Emergency Medicine and from other emergency physicians with an interest in neurologic emergencies. The principal investigator at each study site determined the most thorough methods of identifying all patients with seizure-related conditions at his or her ED during the study period. Each investigator used at least two of the following methods to identify patients: ED admission logbook review, chart review, or quality assurance review. Two centers used research assistants present several hours a day in the ED to enroll patients and collect data at the time of ED visit. The selection of methods at each site for patient identification was made by the site investigator with knowledge of the nuances of his or her institution regarding the two processes most likely to capture all patients with seizure-related chiefcomplaints. The study was approved by the institutional review board (IRB) at each participating center. The study was considered exempt from informed consent by each IRB since the study consisted of chart review and did not involve interventions. Database entry did not include any personal patient identifiers and there was no further patient contact outside of that which might have occurred in routine medical care. Study Setting and Population. Twelve EDs monitored all patients with seizures or a presenting complaint related to seizures for 18.25 days (representing 5% of the calendar year) in late 1997. Both community hospitals and EDs of academic centers participated in the study. Adults and children were included reflecting the emergency patient population of the different centers (Table 1).
Study Protocol. The study was a retrospective chart review of ED records. Each site used a standard templated form for data collection; additional free text comments were added at the investigators’ discretion. The form was developed in consensus with the different site investigators. The investigator at each site reviewed the charts and data forms for completeness. Variables entered into the data collection form included center identification, a unique confidential identifying number for each patient, demographic information, basic characteristics of any EMS care, whether seizures were thought to be of new onset, seizure type, suspected etiology of the seizures, medications administered, laboratory testing, neuroimaging procedures, and disposition. The assignment of seizure type was by history and bedside determination in the ED without benefit of electroencephalography (EEG) or video analysis using standard clinical classification. Suspected seizure etiology was assigned by the site investigator after evaluating the ED record. At eight of the 12 sites, the physician-investigator was the data collector; at the other four sites, ancillary personnel (usually advanced-degree nurses) collected information under the supervision of the physician investigator, who reviewed the material prior to submission to central data collection. Ten of the 12 sites used just one collector (again, usually the site investigator); the other two sites used real-time enrollers for many hours of the day. Because of the limited number of data collectors at each site, no interrater reliability measures were attempted. Investigators at each site contributed to the development of the final data collection sheet. Beyond these frequent prestudy interactions, no formal education on appropriate data extraction was conducted.
TABLE 1. Participating Centers in the Emergency Medicine Seizure Study Group Hospital Name Bridgeport Hospital, Bridgeport, Connecticut Carolinas Medical Center, Charlotte, North Carolina Maryview Medical Center, Portsmouth, Virginia Mount Sinai Medical Center, New York, New York Our Lady of Ressurection Medical Center, Chicago, Illinois St. Joseph Mercy Hospital (adult ED only), Ann Arbor, Michigan University of Cincinnati Medical Center, Cincinnati, Ohio University of Illinois College of Medicine, Chicago, Illinois University of North Carolina Health Care System, Chapel Hill, North Carolina University of Pittsburgh Medical Center, Shadyside, Pittsburgh, Pennsylvania University of Virginia Health System, Charlottesville, Virginia Vanderbilt University Medical Center, Nashville, Tennessee
Type of Hospital*
Study Period Visits
Number of Seizure Visits
Pediatrics % Annual
CTH CTH CTH UTH CTH
2,824 4,406 2,014 3,428 1,665
37 49 25 32 33
31 25 15 20 23
CTH UTH UTH
2,713 3,478 1,866
18 33 22
0 0 37
UTH
2,516
29
7
CH UTH UTH
1,415 2,649 2,538
8 33 49
5 28 38
*UTH = university teaching hospital; CTH = community teaching hospital; CH = community hospital.
624
SEIZURES
Huff et al. • SEIZURES IN ED PATIENTS
TABLE 2. Clinical Assignment of Seizure Types Seizure Type Generalized convulsive Partial with secondary generalizations Partial with simple symptoms Partial complex and other nonconvulsive types Infantile spasm Pseudoseizures
Number (Percentage) 312 (86%) 9 (2%) 24 (7%) 6 (2%) 1 (
