ORIGINAL CONTRIBUTION alcohol withdrawal delirium; phenytoin; seizure
Emergency Department Treatment of Alcohol Withdrawal Seizures With Phenytoin Study objective: Prevention of recurrent alcohol withdrawal seizures is a common emergency department problem. A prospective, randomized, placebo-controlled, double-blind study of adequate size was designed to assess the efficacy of phenytoin in preventing recurrence of alcohol withdrawal seizures. Methods: Fifty-five patients who had seized from alcohol withdrawal were randomly assigned to treatment with IV phenytoin or placebo. Patients with known seizure disorders and those receiving any anticonvulsant were excluded. The study was terminated after seizure recurrence or passage of a six-hour, high-risk seizure interval. Results: Six of 28 phenytoin-treated patients (21%) had recurrent seizures compared with five of 27 placebo-treated patients (19%). The 95% confidence interval for the difference in response probabilities was + 16% to - 2 0 % . There was no statistically significant difference between the response rates for the two treatments (P > .05). Conclusion: Phenytoin does not show significant benefit over placebo in preventing recurrence of alcohol withdrawal seizures. ]Chance JF: Emergency department treatment of alcohol withdrawal seizures with phenytoin. Ann Emerg Med May 1991;20:520-522.]
Joseph F Chance, MD Charlottesville, Virginia From the Division of Emergency Medicine, University of Virginia Hospital, Charlottesville. Received for publication April 25, 1990. Revisions received July 30, and August 24, 1990. Accepted for publication September 21, 1990. Address for reprints: Joseph F Chance, MD, Division of Emergency Medicine, Box 523-21, Charlottesville, Virginia 22908.
INTRODUCTION Alcohol withdrawal is a common cause of seizures in patients presenting to emergency departments. 1 Such patients are at risk of recurrent seizures With associated traumatic injury or aspiration. 2 Patients with alcohol withdrawal seizures pose a therapeutic dilemma because no treatment has been shown to be effective in preventing recurrent seizures. Phenytoin is useful in treating various seizure types, can achieve therapeutic levels quickly through the IV route, and has reported prophylactic efficacy in preventing onset of alcohol withdrawal seizures.3, 4 It thus appears reasonable to believe that phenytoin might be effective in preventing recurrent alcohol withdrawal seizures in patients presenting to the ED. A prospective, randomized, placebo-controlled, double-blind study of adequate size was conducted to test this hypothesis.
MATERIALS A N D METHODS Patients presenting to an ED with a complaint of seizure whose clinical diagnosis was that the seizure was due exclusively to alcohol withdrawal were admitted to the study. The diagnosis of alcohol withdrawal seizure was made by the ED attending physician and confirmed by the study investigator. Patient history and physical examination, chart review, and witness and rescue squad accounts were all considered in the diagnostic evaluation. Patients had at least one computed tomography scan, electrolyte.profile, and formal neurologic consultative opinion diagnosing seizures as resulting from alcohol withdrawal. Patients with known seizure disorders unrelated to alcohol withdrawal whose seizure thresholds could have been lowered by alcohol use were excluded. Patients with initial detectable phenytoin levels (> 2.5 ~g/mL) or allergies or contraindications to hydantoin derivatives also were excluded. Patients who had received any anticonvulsants to control an acute seizure episode (eg, benzodiazepines) were excluded. Women were ex-
20:5 May 1991
Annals of Emergency Medicine
520/57
PHENYTOIN Chance
cluded because of the inability of even serum pregnancy tests to detect early pregnancy and because of the teratogenic risk of phenytoin. Patients were randomly assigned to receive either phenytoin 50 rag/ mL or 0.9% normal saline. The phenytoin and normal saline solutions were in identical 20-mL vials prepared by the pharmacy in an order determined by random-number tables and sent to the ED labeled only by test solution numbers. The test solution dose was determined by body weight (0.3 mL test solution/ kg), added to 250 mL of normal saline, and administered by infusion pump at a rate of 10 mL/min through a 0.22-~m filter. If the test solution was phenytoin, this procedure resulted in 1) a dose of 15 mg/kg to a maximum of 1,000 mg for a 70-kg man, 2) a concentration of no more than 3.7 mg/min, and 3) a rate of no more than 37 m g / m i n . The dose, concentration, and rate were chosen to p r e v e n t side effects of ataxia, b u r n i n g at IV line site, and arrhythmias, respectively.g, 5 All patients had cardiac and blood pressure monitoring during infusion. Neither patient nor physician knew if the test solution administered was phenytoin or normal saline. The study was terminated if 1) the patient seized after test solution infusion or 2) at least six hours had passed since the initial seizure. This time period was chosen because the majority of recurrent alcohol withdrawal seizures o c c u r w i t h i n six hours of the initial seizure. 2 Choice of treatment after study termination was left to the treating physician. The study protocol was approved by the H u m a n Investigation Committee. The code for the test solution vials was not broken until study completion. The code was broken, and data were analyzed when the sample size reached the n u m b e r necessary to have a power of 95% of detecting the efficacy of phenytoin in eliminating alcohol withdrawal seizures assuming a recurrent rate of 40% (e~ = .05). The response probabilities for phenytoin and placebo were evaluated by the 95% confidence interval for the difference in response probabilities between two groups and by Fisher's exact test. 6
58/521
RESULTS Fifty-five patients completed the study protocol over a two-year period. No side effects were observed during test solution infusions. Six of 28 patients (21%) receiving phenytoin seized compared with five of 27 receiving placebo (19%). The 95% confidence interval for the difference in response rates between phenytoin and placebo was +16% to - 2 0 % . The associated difference in response probabilities was P = .53 (one-tailed) and P = .99 ( t w o - t a i l e d ) u s i n g Fisher's exact test. DISCUSSION Alcohol withdrawal is a common cause of seizures in patients presenting to hospital EDs. l Patients who have experienced one seizure as a result of withdrawal from alcohol are at risk of recurrent seizures. The true incidence of recurrent seizures is uncertain, varying from 20% in the present study to as high as 60% in a patient population under medical supervision at time of first seizure. 2 Seizure patients are at risk of aspiration with acute airway obstruction or subsequent pneumonia. They often experience traumatic injuries such as laceration, dislocation, or fracture. A small percentage develop status epilepticus. 2 The physician seeing a patient after an alcohol withdrawal seizure is faced with the task of preventing m o r b i d i t y from recurrent seizures. This could be done by observing the patient with seizure precautions or by administering treatment effective in preventing recurrent seizures. The present study was conducted to see if phenytoin might be such a treatment. Standard referenced reports are not useful to the physician faced with caring for a patient who has just suffered an alcohol withdrawal seizure. The subject is mentioned briefly, if at all. Medication use is discouraged on the premise that seizures will have ceased by the time the medication becomes effective. Informal neurologic c o n s u l t a t i o n s u s u a l l y echo these sentiments. Published reports reveal the lack of directly pertinent data on the use of phenytoin in preventing recurrence of alcohol withdrawal seizures.7, 8 Published data do, however, pro-
Annals of Emergency Medicine
vide a hint that phenytoin could be effective in preventing recurrent alcohol withdrawal seizures and thus prevent hospitalization. Given in combination with chlordiazepoxide, phenytoin can prevent seizures in patients with a history of alcohol withdrawal seizures if administered before the first seizure compared with p a t i e n t s given c h l o r d i a z e p o x i d e alone. 4 Given alone, phenytoin can prevent alcohol withdrawal seizures in an animal model if administered before the first seizure. 9 These studies cannot be applied directly to the clinical dilemma of treating a patient who has already had his first seizure. A clinically useful study would be one in which human subjects were used instead of animal models, phenytoin was given alone and not in combination with drugs that might have anticonvulsant properties, and phenytoin was administered after the first seizure and not prophylactically. The results of this study indicate that phenytoin is not helpful in preventing recurrent alcohol withdrawal seizures. This supports the conclusions of Alldredge et al, who found no advantage of phenytoin over placebo in preventing recurrent alcohol withdrawal seizures, lo The emergency physician is thus left with a dilemma when seeing a patient who presents after an alcohol withdrawal seizure. Recurrent seizures will occur in a substantial percentage of patients. If unobserved, the patient is at risk of upper airway obstruction, aspiration, soft-tissue or orthopedic injury, and status epilepticus. P h e n y t o i n c a n n o t p r e v e n t these seizures. Anecdotal and uncontrolled evidence has concerned the use of carbamazepine, valproic acid, phenobarbital, primidone, and benzodiazepines.ll,12 No clinically useful controlled trials indicate efficacy of a n y therapeutic agent in alcohol withdrawal seizures. Discharge of patients from the ED after a period of observation is one option. The duration of such an observation period is uncertain, although there is a suggestion that the first six hours after the initial convulsion is the highest risk period. 2 Hospital admission on seizure precaution~s using IV benzodiazepines for prolonged acute seizure episodes
20:5 May 1991
PHENYTOIN Chance
is another option meriting consideration. CONCLUSION Patients presenting after an alcohol withdrawal seizure are frequently encountered in the ED. Prevention of recurrent seizures w i t h associated morbidity is desirable. Conflicting opinion has existed concerning the efficacy of phenytoin in preventing seizure recurrence. This prospective, randomized, placebo-controlled, double-blind study of sufficient size to detect desirable therapeutic efficacy showed phenytoin to be ineffective in preventing recurrence of alcohol withdrawal seizures. The author thanks the emergency physi-
20:5 May 1991
cians, housestaff, and nursing personnel of the University of Virginia Hospital for their e n t h u s i a s m and cooperation. Thanks go to Olivette Hasty for manuscript preparation, and special thanks go to William R Wimbish, pharmacist, for his persistence in producing and stocking the test solution.
5. Earnest M E Marx ]A, Drury LR: Complications of intravenous phenytoin for acute treatment of seizures. lAMA 1983;249:762-765.
REFERENCES
9. Chn NS: Prevention of alcohol withdrawal seizures with phenytoin in rats. Epilepsia 1981;22:179-184.
1. Earnest MP, Yarnell PR: Seizure admissions to a city hospital: The role of alcohol. Epilepaia 1976;17:387-393. 2. Victor M, Brausch C: The role of abstinence in the genesis of alcoholic epilepsy. EpiJepsia 1967;8:1-20.
6. Simon R: Confidence intervals for reporting the results of clinical trials. A n n Intern M e d 1986;105: 429-435. 7. Morris JC, Victor M: Alcohol withdrawal seizures. Emerg Med Clin North A m 1987;5:827-839. 8. Essig CI:, Carter WW: Failure of diphenylhydantoin in preventing barbiturate withdrawal convulsions in the dog. Neurology 1962;12:481-484.
10. Alldredge BK, Lowenstein DH, Simon RP: Placebocontrolled trial of intravenous diphenylhydantoin fol short-term treatment of alcohol withdrawal seizures A m J Med 1989;87:645-648.
3. Carducei B, Hedges JR, BeaI JC, et al: Emergency phenytoin loading by constant intravenous infusion. Ann Emerg Med 1984;13:1027-1031.
11. Young GP, Rozes C, Murphy C, et al: Intravenou., phenobarbital for alcohol withdrawal and convulsions Ann Emerg Med 1987;16:847 850.
4. Sampliner R, Iber FL: Diphenylhydantoin control of alcohol withdrawal seizures: Results of a controlled study. JAMA 1974;230:1430-1432.
12. Wilburn R, Kulik F: Anticonvulsant drugs in alco hol w i t h d r a w a l . A m J Hosp PharmacoI 1981;38 i138-1143.
Annals of Emergency Medicine
522/E
Emergency Department Treatment of Alcohol Withdrawal Seizures With Phenytoin Study objective: Prevention of recurrent alcohol withdrawal seizures is a common emergency department problem. A prospective, randomized, placebo-controlled, double-blind study of adequate size was designed to assess the efficacy of phenytoin in preventing recurrence of alcohol withdrawal seizures. Methods: Fifty-five patients who had seized from alcohol withdrawal were randomly assigned to treatment with IV phenytoin or placebo. Patients with known seizure disorders and those receiving any anticonvulsant were excluded. The study was terminated after seizure recurrence or passage of a six-hour, high-risk seizure interval. Results: Six of 28 phenytoin-treated patients (21%) had recurrent seizures compared with five of 27 placebo-treated patients (19%). The 95% confidence interval for the difference in response probabilities was + 16% to - 2 0 % . There was no statistically significant difference between the response rates for the two treatments (P > .05). Conclusion: Phenytoin does not show significant benefit over placebo in preventing recurrence of alcohol withdrawal seizures. ]Chance JF: Emergency department treatment of alcohol withdrawal seizures with phenytoin. Ann Emerg Med May 1991;20:520-522.]
Joseph F Chance, MD Charlottesville, Virginia From the Division of Emergency Medicine, University of Virginia Hospital, Charlottesville. Received for publication April 25, 1990. Revisions received July 30, and August 24, 1990. Accepted for publication September 21, 1990. Address for reprints: Joseph F Chance, MD, Division of Emergency Medicine, Box 523-21, Charlottesville, Virginia 22908.
INTRODUCTION Alcohol withdrawal is a common cause of seizures in patients presenting to emergency departments. 1 Such patients are at risk of recurrent seizures With associated traumatic injury or aspiration. 2 Patients with alcohol withdrawal seizures pose a therapeutic dilemma because no treatment has been shown to be effective in preventing recurrent seizures. Phenytoin is useful in treating various seizure types, can achieve therapeutic levels quickly through the IV route, and has reported prophylactic efficacy in preventing onset of alcohol withdrawal seizures.3, 4 It thus appears reasonable to believe that phenytoin might be effective in preventing recurrent alcohol withdrawal seizures in patients presenting to the ED. A prospective, randomized, placebo-controlled, double-blind study of adequate size was conducted to test this hypothesis.
MATERIALS A N D METHODS Patients presenting to an ED with a complaint of seizure whose clinical diagnosis was that the seizure was due exclusively to alcohol withdrawal were admitted to the study. The diagnosis of alcohol withdrawal seizure was made by the ED attending physician and confirmed by the study investigator. Patient history and physical examination, chart review, and witness and rescue squad accounts were all considered in the diagnostic evaluation. Patients had at least one computed tomography scan, electrolyte.profile, and formal neurologic consultative opinion diagnosing seizures as resulting from alcohol withdrawal. Patients with known seizure disorders unrelated to alcohol withdrawal whose seizure thresholds could have been lowered by alcohol use were excluded. Patients with initial detectable phenytoin levels (> 2.5 ~g/mL) or allergies or contraindications to hydantoin derivatives also were excluded. Patients who had received any anticonvulsants to control an acute seizure episode (eg, benzodiazepines) were excluded. Women were ex-
20:5 May 1991
Annals of Emergency Medicine
520/57
PHENYTOIN Chance
cluded because of the inability of even serum pregnancy tests to detect early pregnancy and because of the teratogenic risk of phenytoin. Patients were randomly assigned to receive either phenytoin 50 rag/ mL or 0.9% normal saline. The phenytoin and normal saline solutions were in identical 20-mL vials prepared by the pharmacy in an order determined by random-number tables and sent to the ED labeled only by test solution numbers. The test solution dose was determined by body weight (0.3 mL test solution/ kg), added to 250 mL of normal saline, and administered by infusion pump at a rate of 10 mL/min through a 0.22-~m filter. If the test solution was phenytoin, this procedure resulted in 1) a dose of 15 mg/kg to a maximum of 1,000 mg for a 70-kg man, 2) a concentration of no more than 3.7 mg/min, and 3) a rate of no more than 37 m g / m i n . The dose, concentration, and rate were chosen to p r e v e n t side effects of ataxia, b u r n i n g at IV line site, and arrhythmias, respectively.g, 5 All patients had cardiac and blood pressure monitoring during infusion. Neither patient nor physician knew if the test solution administered was phenytoin or normal saline. The study was terminated if 1) the patient seized after test solution infusion or 2) at least six hours had passed since the initial seizure. This time period was chosen because the majority of recurrent alcohol withdrawal seizures o c c u r w i t h i n six hours of the initial seizure. 2 Choice of treatment after study termination was left to the treating physician. The study protocol was approved by the H u m a n Investigation Committee. The code for the test solution vials was not broken until study completion. The code was broken, and data were analyzed when the sample size reached the n u m b e r necessary to have a power of 95% of detecting the efficacy of phenytoin in eliminating alcohol withdrawal seizures assuming a recurrent rate of 40% (e~ = .05). The response probabilities for phenytoin and placebo were evaluated by the 95% confidence interval for the difference in response probabilities between two groups and by Fisher's exact test. 6
58/521
RESULTS Fifty-five patients completed the study protocol over a two-year period. No side effects were observed during test solution infusions. Six of 28 patients (21%) receiving phenytoin seized compared with five of 27 receiving placebo (19%). The 95% confidence interval for the difference in response rates between phenytoin and placebo was +16% to - 2 0 % . The associated difference in response probabilities was P = .53 (one-tailed) and P = .99 ( t w o - t a i l e d ) u s i n g Fisher's exact test. DISCUSSION Alcohol withdrawal is a common cause of seizures in patients presenting to hospital EDs. l Patients who have experienced one seizure as a result of withdrawal from alcohol are at risk of recurrent seizures. The true incidence of recurrent seizures is uncertain, varying from 20% in the present study to as high as 60% in a patient population under medical supervision at time of first seizure. 2 Seizure patients are at risk of aspiration with acute airway obstruction or subsequent pneumonia. They often experience traumatic injuries such as laceration, dislocation, or fracture. A small percentage develop status epilepticus. 2 The physician seeing a patient after an alcohol withdrawal seizure is faced with the task of preventing m o r b i d i t y from recurrent seizures. This could be done by observing the patient with seizure precautions or by administering treatment effective in preventing recurrent seizures. The present study was conducted to see if phenytoin might be such a treatment. Standard referenced reports are not useful to the physician faced with caring for a patient who has just suffered an alcohol withdrawal seizure. The subject is mentioned briefly, if at all. Medication use is discouraged on the premise that seizures will have ceased by the time the medication becomes effective. Informal neurologic c o n s u l t a t i o n s u s u a l l y echo these sentiments. Published reports reveal the lack of directly pertinent data on the use of phenytoin in preventing recurrence of alcohol withdrawal seizures.7, 8 Published data do, however, pro-
Annals of Emergency Medicine
vide a hint that phenytoin could be effective in preventing recurrent alcohol withdrawal seizures and thus prevent hospitalization. Given in combination with chlordiazepoxide, phenytoin can prevent seizures in patients with a history of alcohol withdrawal seizures if administered before the first seizure compared with p a t i e n t s given c h l o r d i a z e p o x i d e alone. 4 Given alone, phenytoin can prevent alcohol withdrawal seizures in an animal model if administered before the first seizure. 9 These studies cannot be applied directly to the clinical dilemma of treating a patient who has already had his first seizure. A clinically useful study would be one in which human subjects were used instead of animal models, phenytoin was given alone and not in combination with drugs that might have anticonvulsant properties, and phenytoin was administered after the first seizure and not prophylactically. The results of this study indicate that phenytoin is not helpful in preventing recurrent alcohol withdrawal seizures. This supports the conclusions of Alldredge et al, who found no advantage of phenytoin over placebo in preventing recurrent alcohol withdrawal seizures, lo The emergency physician is thus left with a dilemma when seeing a patient who presents after an alcohol withdrawal seizure. Recurrent seizures will occur in a substantial percentage of patients. If unobserved, the patient is at risk of upper airway obstruction, aspiration, soft-tissue or orthopedic injury, and status epilepticus. P h e n y t o i n c a n n o t p r e v e n t these seizures. Anecdotal and uncontrolled evidence has concerned the use of carbamazepine, valproic acid, phenobarbital, primidone, and benzodiazepines.ll,12 No clinically useful controlled trials indicate efficacy of a n y therapeutic agent in alcohol withdrawal seizures. Discharge of patients from the ED after a period of observation is one option. The duration of such an observation period is uncertain, although there is a suggestion that the first six hours after the initial convulsion is the highest risk period. 2 Hospital admission on seizure precaution~s using IV benzodiazepines for prolonged acute seizure episodes
20:5 May 1991
PHENYTOIN Chance
is another option meriting consideration. CONCLUSION Patients presenting after an alcohol withdrawal seizure are frequently encountered in the ED. Prevention of recurrent seizures w i t h associated morbidity is desirable. Conflicting opinion has existed concerning the efficacy of phenytoin in preventing seizure recurrence. This prospective, randomized, placebo-controlled, double-blind study of sufficient size to detect desirable therapeutic efficacy showed phenytoin to be ineffective in preventing recurrence of alcohol withdrawal seizures. The author thanks the emergency physi-
20:5 May 1991
cians, housestaff, and nursing personnel of the University of Virginia Hospital for their e n t h u s i a s m and cooperation. Thanks go to Olivette Hasty for manuscript preparation, and special thanks go to William R Wimbish, pharmacist, for his persistence in producing and stocking the test solution.
5. Earnest M E Marx ]A, Drury LR: Complications of intravenous phenytoin for acute treatment of seizures. lAMA 1983;249:762-765.
REFERENCES
9. Chn NS: Prevention of alcohol withdrawal seizures with phenytoin in rats. Epilepsia 1981;22:179-184.
1. Earnest MP, Yarnell PR: Seizure admissions to a city hospital: The role of alcohol. Epilepaia 1976;17:387-393. 2. Victor M, Brausch C: The role of abstinence in the genesis of alcoholic epilepsy. EpiJepsia 1967;8:1-20.
6. Simon R: Confidence intervals for reporting the results of clinical trials. A n n Intern M e d 1986;105: 429-435. 7. Morris JC, Victor M: Alcohol withdrawal seizures. Emerg Med Clin North A m 1987;5:827-839. 8. Essig CI:, Carter WW: Failure of diphenylhydantoin in preventing barbiturate withdrawal convulsions in the dog. Neurology 1962;12:481-484.
10. Alldredge BK, Lowenstein DH, Simon RP: Placebocontrolled trial of intravenous diphenylhydantoin fol short-term treatment of alcohol withdrawal seizures A m J Med 1989;87:645-648.
3. Carducei B, Hedges JR, BeaI JC, et al: Emergency phenytoin loading by constant intravenous infusion. Ann Emerg Med 1984;13:1027-1031.
11. Young GP, Rozes C, Murphy C, et al: Intravenou., phenobarbital for alcohol withdrawal and convulsions Ann Emerg Med 1987;16:847 850.
4. Sampliner R, Iber FL: Diphenylhydantoin control of alcohol withdrawal seizures: Results of a controlled study. JAMA 1974;230:1430-1432.
12. Wilburn R, Kulik F: Anticonvulsant drugs in alco hol w i t h d r a w a l . A m J Hosp PharmacoI 1981;38 i138-1143.
Annals of Emergency Medicine
522/E
