|
|
| |||||||||||||||||||||||||||||||||||||||||||||||
Background Acute repetitive seizures are readily recognizable episodes involving increased seizure frequency. Urgent treatment is often required. Rectal diazepam gel is a promising therapy.
Methods We conducted a randomized, double-blind, parallel-group, placebo-controlled study of home-based treatment for acute repetitive seizures. Patients were randomly assigned to receive either rectal diazepam gel, at doses ranging from 0.2 to 0.5 mg per kilogram of body weight on the basis of age, or placebo. Children received one dose at the onset of acute repetitive seizures and a second dose four hours later. Adults received three doses — one dose at onset, and two more doses 4 and 12 hours after onset. Treatment was administered by a care giver, such as a parent, who had received special training. The number of seizures after the first dose was counted for 12 hours in children and for 24 hours in adults.
Results Of 125 study patients (64 assigned to diazepam and 61 to placebo) with a history of acute repetitive seizures, 91 (47 children and 44 adults) were treated for an exacerbation of seizures during the study period. Diazepam treatment was superior to placebo with regard to the outcome variables related to efficacy: reduced seizure frequency (P<0.001) and improved global assessment of treatment outcome by the care giver (frequency and severity of seizures and drug toxicity) (P<0.001). Post hoc analysis showed diazepam to be superior to placebo in reducing seizure frequency in both children (P<0.001) and adults (P=0.02), but only in children was it superior with regard to improvement in global outcome (P<0.001). The time to the first recurrence of seizures after initial treatment was longer for the patients receiving diazepam (P<0.001). Thirty-five patients reported at least one adverse effect of treatment; somnolence was the most frequent. Respiratory depression was not reported.
Conclusions Rectal diazepam gel, administered at home by trained care givers, is an effective and well-tolerated treatment for acute repetitive seizures.
Family members are often able to predict the onset of acute repetitive seizures according to the time of day when the seizures begin, the occurrence of a particular type of seizure, the severity of the seizure, and accompanying behavioral changes in the patient. If they are left untreated, acute repetitive seizures can evolve into more serious problems, including status epilepticus.4,5,6
Benzodiazepines are the treatment of choice for management of acute seizures.4,7 They are active against many types of seizure, have a rapid onset of action, and are safe.7 Oral diazepam and lorazepam, sublingual lorazepam, rectal solutions of lorazepam and diazepam, and diazepam suppositories have been used to treat acute repetitive seizures.8,9,10,11,12,13 Oral or sublingual administration is frequently difficult and hazardous when the patient is actively convulsing, and absorption of diazepam and lorazepam tablets and rectal lorazepam solution is slower than that of rectal diazepam solutions.8,14,15
Rectal diazepam solutions have many characteristics sought in the ideal drug to treat acute repetitive seizures. Their high lipid solubility permits both prompt absorption and rapid penetration into the central nervous system.16 Peak serum plasma concentrations are reached within 5 to 45 minutes, with bioavailability averaging between 80 and 100 percent.8,9,13,16,17,18,19,20 In contrast, diazepam suppositories have slow, erratic absorption, which limits their use in the management of acute seizures.13,17,19 Rectal diazepam solutions can be administered easily and safely by nonmedical personnel, irrespective of the patient's ability to cooperate.
Both European and North American clinicians have used rectal diazepam to treat acute repetitive, prolonged, and febrile seizures.12,16,21,22 In the United States, rectal diazepam solutions have been prepared from commercial parenteral products.8,23,24,25
In 1988, Upsher–Smith Laboratories, in collaboration with the University of Minnesota, began developing a rectal diazepam delivery system to treat acute repetitive seizures. We conducted a clinical trial to assess the safety and efficacy of diazepam rectal gel for acute repetitive seizures.
Methods
Patients
The patients were boys and girls 2 to 14 years of age and adults, defined as 15 to 60 years of age, with a maximal weight of 100 kg, who had had at least four episodes of acute repetitive seizures during the preceding year and at least one in the preceding three months. We defined acute repetitive seizures as an episode of multiple complex partial or generalized (tonic, clonic, tonic–clonic, atypical absence, or myoclonic) seizures occurring within a 24-hour period in adults or a 12-hour period in children, with a pattern distinguishable from the patient's usual seizure pattern, and with onset readily recognizable by a care giver, such as a parent. All patients had been on a stable antiepileptic regimen for at least four weeks before enrollment. Brain computed tomography or magnetic resonance imaging and laboratory screening had shown no evidence of a treatable cause of the seizures. Women of childbearing potential were eligible if they used contraception and had a negative pregnancy test.
Patients were ineligible if they met any of the following criteria: plasma phenobarbital concentrations greater than 30 mg per liter (130 µmol per liter), current treatment with drugs other than anticonvulsants, long-term use of benzodiazepines, use of central nervous system depressants or drugs interacting with diazepam, more than one previous treatment with rectal diazepam, nonepileptic seizures within the preceding five years, habitual progression to status epilepticus, clinically significant psychiatric disorder, lack of a suitable care giver, or use of an investigational drug or device within the preceding five months. Each patient, parent, or legal guardian provided written informed consent, and children gave assent.
Study Design
The study was conducted between October 1991 and January 1995. We used a prospective, randomized, double-blind, placebo-controlled, parallel-group design. A sample size of 144 patients was planned on the basis of estimates of the sample size required for a two-sided test to detect a 50 percent reduction in seizure frequency with diazepam at a significance level of 0.05 and a power of 0.80.
Each center's institutional review board approved the protocol and consent procedures. Oversight included regular review by the National Institute of Neurological Disorders and Stroke Performance and Safety Monitoring Board. Because of the board's concern about the safety of the investigational formulation and the response of the acute repetitive seizures to treatment, it requested an unblinded interim analysis of efficacy and all safety data for the first 36 patients. The board did not disclose the results or change the study at the time of this analysis. Thereafter, the blinding was maintained until the trial had been completed and decisions about the inclusion of patients in the analyses had been made. Randomization was based on permuted blocks, with stratification according to center and age group.
Procedures
Using an instructional videotape, study nurses taught care givers how to identify episodes of acute repetitive seizures, give medication, and record respiration, skin color, seizures, adverse events, and global assessment of treatment outcome in a booklet. During episodes of acute repetitive seizures, study nurses maintained telephone contact with the care givers to review procedures, monitor patients, and intervene if patients needed additional treatment. The observation period began immediately after the first dose and continued for 12 hours in children and 24 hours in adults. The lengths of the observation periods were based on preliminary data showing that most episodes of acute repetitive seizures ended within these times. The care givers and patients returned to the clinic 72 hours after treatment to review the recorded data.
We removed patients from the study before treatment if they no longer met eligibility criteria or if they withdrew consent. All data were analyzed, including those on patients who did not complete the protocol.
Study Treatments
We employed an age-adjusted dosage regimen of diazepam rectal gel (Diastat, Athena Neurosciences, San Francisco), as follows: 0.5 mg per kilogram of body weight for children 2 to 5 years of age, 0.3 mg per kilogram for children 6 to 11 years of age, and 0.2 mg per kilogram for patients 12 or older. These schedules were based on the results of clinical trials showing that diazepam clearance in children declines until about the age of 12 years, when adult values are reached.8,16,19,26 A second dose was given four hours after the first, and, for adults, a third dose was given eight hours after the second, since previous studies indicated that these schedules should maintain target plasma diazepam concentrations (150 to 300 ng per milliliter).8,19,26 Diazepam rectal gel (5 mg per milliliter) and identical-looking placebo were supplied by the manufacturer in 2-ml syringes containing 0.5, 1.0, or 1.5 ml and 5-ml syringes containing 2.0, 2.5, 3.0, 3.5, or 4.0 ml. Diazepam doses ranged from 2.5 to 20 mg, in 2.5-mg increments. The doses were rounded up to the nearest 2.5 mg.
The medication kits contained two syringes for children and three syringes for adults, with one dose per syringe. A syringe with half the regular dose was available to be used if the regular dose was expelled within five minutes.
Outcome Variables Related to Efficacy
Seizure frequency and global assessment of treatment outcome by the care giver were the outcome variables related to efficacy. Seizure frequency was defined as the number of seizures per hour of observation. Global treatment outcome was a composite evaluation based on the care giver's assessment of the frequency and severity of seizures and drug toxicity. The global outcome was rated as better than, the same as, or worse than the outcome after previous episodes.
Two efficacy outcome variables were retrospectively defined: the time to the first recurrence of seizures after the initial treatment, and the number of patients remaining seizure-free during the first 12 hours after treatment.
Safety Variables
At the return visit, the notes in the care giver's booklet about respiratory rate, skin color, and adverse events were reviewed. All adverse events were categorized according to the preferred terms of the Food and Drug Administration's COSTART system.27
Statistical Analysis
All patients who received at least one dose of study medication were included in the analysis. For base-line comparisons between the treatment groups, we used Fisher's exact test for sex, numbers of adults and children, and race and analysis of variance for age and body weight. The seizure-frequency data failed to support the assumption of normality required for a t-test. Therefore, we used a stratified Wilcoxon rank-sum test28 to analyze both continuous data (seizure frequency) and categorical data (global assessment of treatment outcome), incorporating "child" and "adult" as the stratifying variables. The latter statistical analyses (investigation of data for normality, t-test, and rank-based methods) were specified by the protocol before the study began.
A Spearman rank-correlation test29 was used to determine the relation between seizure frequency and the global assessment of treatment outcome, and the Wilcoxon rank-sum test30 was used for both primary efficacy variables for a post hoc analysis of treatment differences for children and for adults.
The statistical tests for the two primary efficacy variables were two-sided at the 0.025 level of significance, with a Bonferroni correction31 for testing two outcomes. The significance level was 0.0246, with an adjustment by O'Brien–Fleming procedures32 to incorporate a penalty for the interim analysis performed at the 0.001 level of significance. The time to a first seizure recurrence after initial treatment was displayed as Kaplan–Meier survival estimates.33
Fisher's exact test and the Wilcoxon rank-sum test30 with exact P values were computed with StatXact software (version 3, Cytel Software, Cambridge, Mass.), and analyses of variance by JMP software (version 3.1, SAS Institute, Cary, N.C.).
Results
Characteristics of the Patients
One hundred twenty-five patients (64 in the diazepam group and 61 in the placebo group) were enrolled and randomly assigned to treatment. This number was less than the 144 patients originally planned, since the eligibility criteria were very stringent and funding considerations limited the period of recruitment. Ninety-one of these patients were treated. The median times from randomization to treatment were 20 days for the diazepam group and 19 days for the placebo group (exact P=0.95 by the Wilcoxon rank-sum test). The diazepam and placebo groups did not differ significantly in sex, age, and weight (Table 1). Significantly more black patients were randomly assigned to receive diazepam than were assigned to receive placebo, and significantly more white patients were assigned to placebo than to diazepam (P=0.04 by Fisher's exact test). Data from all 91 treated patients were analyzed. A child in the placebo group who required medical intervention for seizures during treatment was assigned the most conservative scores (zero seizures, better global assessment).
|
Study Medication
Forty-one of the 45 treated patients assigned to diazepam and 40 of the 46 treated patients assigned to placebo received all their scheduled doses. One of the 20 adults given diazepam received two doses instead of three, and 3 of the 25 children given diazepam received one dose instead of two. The reasons included reluctance to give additional doses to a lethargic patient (one patient), rash (one patient), and the care giver's noncompliance with the protocol (two patients). Four of the 24 adults and 2 of the 22 children in the placebo group received fewer doses than planned. In five instances, medical intervention was required for seizures; one care giver misunderstood the treatment directions. Seven patients (three in the diazepam group and four in the placebo group) received the replacement dose.
The median total dose of diazepam was 20 mg (range, 7.5 to 25) for the 25 children and 37.5 mg (range, 15 to 60) for the 20 adults. Most patients received a slightly higher dose than the targeted dose because of rounding. The median increase over the target dose was 13.6 percent (range, 4.3 percent decrease to 51.5 percent increase) for the diazepam group. Six patients, all children, received doses at least 25 percent greater than the target dose, with one adverse reaction (moderate somnolence).
Efficacy
Diazepam was significantly more effective than placebo both for reducing seizure frequency and for improving the care giver's global assessment of the treatment outcome (P<0.001) (Figure 1 and Figure 2). Seizure frequency and global assessment of treatment outcome were strongly correlated (Spearman r=0.62, P<0.001).
|
|
Patients in the diazepam group had greater protection from seizure recurrence throughout the observation period than those in the placebo group. Differences between the groups were apparent as early as 30 minutes after initial treatment and continued throughout the 12-hour observation period (Figure 3). The time to the first seizure recurrence was significantly longer in the diazepam group than in the placebo group (chi-square=13.75 with 1 df, P<0.001 by the modified Wilcoxon test).
|
Fifty-six patients reported no adverse effects of treatment. There were no reports of respiratory difficulty in patients receiving diazepam. One patient in the placebo group reported cyanosis. Thirty-five patients reported at least one adverse effect, but the difference between the diazepam and placebo groups was not significant: 46.7 percent of the diazepam group and 30.4 percent of the placebo group (P= 0.13 by Fisher's exact test). Table 2 lists the adverse effects reported by more than one patient.
|
Two patients receiving diazepam and two receiving placebo did not complete treatment because of adverse effects. The third dose was withheld because of marked lethargy in an adult given diazepam, and the second dose was withheld because of rash in a child given diazepam. The second and third doses were withheld because of sedation in an adult given placebo. The second dose was withheld because of seizures requiring intravenous diazepam in a child given placebo.
Discussion
We conducted a rigorous, placebo-controlled trial evaluating the use of diazepam rectal gel in patients with exacerbations of epileptic seizures. Clinical observations of patients during hospitalization and at home have suggested that rectal diazepam shortens the duration of seizures, prevents recurrence, or both.6,8,16,20,21,22,23,24,25,34,35,36,37,38,39,40,41 The importance of these studies is difficult to gauge, however. They were retrospective, unblinded, or nonrandomized, and they did not use placebo controls or rigorously define treatment end points.42 A study similar to ours was conducted by Milligan et al.,12 who gave 20-mg diazepam suppositories to adults with seizure clusters in a residential epilepsy center. The investigators used a double-blind, placebo-controlled design, with each subject as his or her own control. The nursing staff administered the medications and observed the patients for 24 hours. Seventy-eight percent of the patients given diazepam remained seizure-free, as compared with 19 percent of the patients given placebo. The analysis was not conducted on intention-to-treat principles, however, since it excluded data from patients from whom blood specimens were not obtained. Also, results from institutionalized adults treated by nurses are difficult to extrapolate to patients treated at home, many of whom are children.
We found that diazepam rectal gel was effective and safe when given outside the hospital for the treatment of acute repetitive seizures. The superiority of diazepam over placebo was found for both measures of efficacy: reduction in seizure frequency and the care giver's global assessment of treatment outcome (combining seizure frequency, seizure severity, and drug toxicity). Also, the time to a recurrence of seizures after the initial treatment and the proportion of patients remaining seizure-free during the observation period were greater in the diazepam group. When the efficacy variables were analyzed separately for children and adults, the reduction in seizure frequency was significantly greater with diazepam in both age groups, but the care giver's global assessment of treatment outcome was significantly better with diazepam only among children. The difference in outcome between children and adults may be related to the study design. Three doses were given to adults but only two to children, regardless of treatment response, and the extra dose of diazepam may have made the adults more sedated. Also, the care givers may have watched the children more closely, resulting in better recognition of differences in global outcome.
The majority of patients had no adverse effects. The patients given diazepam had more adverse effects involving the nervous system but fewer digestive and systemic adverse effects than the patients given placebo. There were no respiratory difficulties.
The benefits of a safe and effective treatment for acute repetitive seizures were highlighted by Alldredge et al.,43 who reviewed 45 episodes of convulsive status epilepticus in 38 children given prehospital treatment with rectal or intravenous diazepam. Treatment shortened the duration of status epilepticus by one half and reduced the likelihood that seizures would recur in the emergency department by one third. In our study, no patients given diazepam and 13 percent of the patients in the placebo group required emergency medical care. Kriel et al.23 computed the cost savings associated with in-home rectal diazepam treatment of cluster and prolonged seizures. They found a 67 percent decrease in emergency room visits and a cost savings of $1,000 per family in the year after the families begin using rectal diazepam, with reduced family disruption and improved quality of life.
We also showed that informed and educated care givers can identify the onset of acute repetitive seizures, safely and successfully administer diazepam rectally, and assess the patient's response. The number of doses was fixed to standardize conditions for this trial. Administration of the second dose in children and adults and of the third dose in adults was not based on response. In practice, the number and timing of doses might better be based on the individual patient's seizure profile and initial response to treatment.
We did not assess rectal diazepam gel in patients with acute repetitive seizures who typically progress to status epilepticus; such patients were ineligible. Although one patient in the placebo group was apparently in status epilepticus when given the first dose, no patient in either study group had status epilepticus after treatment was begun.
Supported by contracts with the National Institute of Neurological Disorders and Stroke and Athena Neurosciences.
Dr. Dreifuss, Dr. Cloyd, and Dr. Pellock have served as consultants to Athena Neurosciences and Upsher–Smith Laboratories, which market and developed, respectively, rectal diazepam gel.
We are indebted to E. Bebin, M.D., N. Santilli, M.N., P.N.P., and R. Homzie-Schlesinger, R.N., at the University of Virginia; P. Bruno, R.N., and J. Paolini, R.N., at the New England Medical Center; C. Jones-Saete, R.N., R. Kriel, M.D., D. Wolff, Pharm.D., P. Ahmann, M.D., and L. Kraemer, at the University of Minnesota; W. Garnett, Pharm.D., S. Driscoll-Bannister, R.N., and K. O'Hara, R.N., at the Medical College of Virginia; R. Faught, M.D., and G. Thompson, R.N., at the University of Alabama at Birmingham; A. Tsao, M.D., F. Wright, M.D., M. Drake, M.D., S. Borror, R.N., C. Schumer, R.N., and P. Snider, R.N., at the Children's Hospital of Columbus; J. Madsen, M.D. (deceased), and C. Roberts, L.P.N., at the University of Utah; D. Bates, B.S.N., at Arkansas Children's Hospital; M. Kolodgie, M.S.N., C.P.N.P., at Children's National Medical Center, Washington, D.C.; R. Ristanovic, M.D., A. Kanner, M.D., and L. Smith, R.N., at Rush–Presbyterian–St. Luke's Medical Center; L. Groves, Ph.D., and J. Franklin, R.N., at Athena Neurosciences; and J. Sahlroot, Ph.D., H. Moore, M.S., K. Reese, M.S., and M. Matula, M.A., at the National Institute of Neurological Disorders and Stroke.
Source Information
From the Department of Neurology, University of Virginia, Charlottesville (F.E.D.); the Division of Pediatric Neurology, New England Medical Center, Boston (N.P.R.); the College of Pharmacy, University of Minnesota, Minneapolis (J.C.C.); the Department of Neurology, Medical College of Virginia, Richmond (J.M.P.); the Department of Neurology, University of Alabama at Birmingham, Birmingham (R.I.K.); the Department of Pediatrics, Children's Hospital, Columbus, Ohio (W.D.L.); the Department of Neurology, University of Utah, Salt Lake City (F.M.); the Department of Neurology, Arkansas Children's Hospital, Little Rock (G.B.S.); the Department of Neurology, Children's National Medical Center, Washington, D.C. (J.A.C.); the Department of Neurological Sciences, Rush–Presbyterian–St. Luke's Medical Center, Chicago (D.C.B.); and the National Institutes of Health, Bethesda, Md. (W.E.B.).
Address reprint requests to Dr. Rosman at the Division of Pediatric Neurology, NEMC 330, Floating Hospital for Children, 750 Washington St., Boston, MA 02111.
References
| |||||||||||||||||||||||||||||||||||||||||||||||
Related Letters:
Treatment of Acute Repetitive Seizures
Orr R. D., Shafrir Y., Rosman N. P., Cloyd J. C., Bell W. E.
Extract |
Full Text
N Engl J Med 1998;
339:1856-1857, Dec 17, 1998.
Correspondence
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2009 Massachusetts Medical Society. All rights reserved. |
Previous
