Curriculum.toxicology.wikispaces.net

Management of Alcohol Withdrawal Delirium
An Evidence-Based Practice Guideline
Michael F. Mayo-Smith, MD, MPH; Lee H. Beecher, MD; Timothy L. Fischer, DO; David A. Gorelick, MD, PhD;Jeanette L. Guillaume, MA; Arnold Hill, MD; Gail Jara, BA; Chris Kasser, MD; John Melbourne, MD;for the Working Group on the Management of Alcohol Withdrawal Delirium,Practice Guidelines Committee, American Society of Addiction Medicine Background: Alcohol withdrawal delirium is the most
significant differences among various benzodiazepines serious manifestation of alcohol withdrawal. Evidence sug- and barbiturates were not found. No deaths were re- gests that appropriate care improves mortality, but sys- ported in 217 patients from trials using benzodiazepines Methods: Articles with original data on management
Conclusions: Control of agitation should be achieved
of alcohol withdrawal delirium underwent structured re- using parenteral rapid-acting sedative-hypnotic agents that are cross-tolerant with alcohol. Adequate doses shouldbe used to maintain light somnolence for the duration Results: Meta-analysis of 9 prospective controlled
of delirium. Coupled with comprehensive supportive trials demonstrated that sedative-hypnotic agents are medical care, this approach is highly effective in pre- more effective than neuroleptic agents in reducing du- ration of delirium and mortality, with a relative risk ofdeath when using neuroleptic agents of 6.6. Statistically Arch Intern Med. 2004;164:1405-1412 From the Primary Care Service,Veterans Administration ALCOHOLDEPENDENCEIS liriumtypicallydoesnotdevelopuntil2
to 3 days after cessation of drinking. Al- cohol withdrawal delirium usually lasts 48 to 72 hours, but there have been case re- ports6-8 of much longer duration. Initial tings.1-3 Alcohol withdrawal is among the many medical problems associated with al- 15%,8 but with advances in treatment, mor- tality rates have fallen, with more recent studies9 indicating mortality of 0% to 1%.
Tri County Commission onAlcohol and Drug Abuse, known as delirium tremens or “DTs,” is the most serious manifestation of alcohol utility. The purpose of this guideline, there- fore, is to assist physicians and other health change in cognition or perceptual distur- care professionals in providing appropri- ate treatment for all patients with AWD.
Marlboro Medical Center,Marlborough, Mass (Dr Hill); shortly after withdrawal from heavy alco- hol intake (Table 1).4 The classic clini-
cal presentation of AWD also includes hy- AWD as these topics are covered in a pre- perpyrexia, tachycardia, hypertension, and Management of AWD was a topic identified for trials of inpatient drug treatment for al- guideline development by the American Soci- cohol withdrawal.5 Clinical features of al- ety of Addiction Medicine Committee on Prac- financial interest in this article. pointed that included individuals with training (REPRINTED) ARCH INTERN MED/ VOL 164, JULY 12, 2004 2004 American Medical Association. All rights reserved.
Table 1. DSM-IV Diagnostic Criteria for Alcohol Withdrawal
Table 2. Methods of Grading Levels of Evidence
and Alcohol Withdrawal Delirium*
and Recommendations12
Alcohol Withdrawal
Definition
A. Cessation of (or reduction in) alcohol use that has been heavy and Levels of Evidence
Randomized trials with low false-positive and low B. Two (or more) of the following, developing within several hours to a Randomized trials with high false-positive or high (1) Autonomic hyperactivity (eg, sweating or pulse rate Nonrandomized, concurrent cohort comparisons Nonrandomized, historical cohort comparisons (4) Nausea or vomiting(5) Transient visual, tactile, or auditory hallucinations or illusions Recommendations
Supported by level I studies or by a meta-analysis in which the lower limit of the confidence interval for the effect of treatment exceeds the C. The symptoms in criterion B cause clinically significant distress or impairment in social, occupational, or other important areas of Supported by level II studies or by a meta-analysis in which the estimate of treatment effect exceeds D. The symptoms are not due to a general medical condition and are the minimal clinically significant benefit but the not better accounted for by another mental disorder.
lower limit of the confidence interval does not Specify whether with perceptual disturbances.
controlled trials, including secondary analyses Alcohol Withdrawal Delirium†
A. Disturbance of consciousness (ie, reduced clarity of awareness of the environment), with reduced ability to focus, sustain, or shiftattention.
B. A change in cognition (such as memory deficit, disorientation, or ences from the selected articles, including those from before language disturbance) or the development of a perceptual 1966, from review articles, and from textbooks were also ex- disturbance that is not better accounted for by a preexisting, amined and included when appropriate. Members of the work- ing group, using a structured data collection form, abstracted C. The disturbance develops in a short period (usually hours to days) and tends to fluctuate during the day.
all articles meeting the initial inclusion criteria. Articles iden- D. There is evidence from the history, physical examination, or tified as prospective controlled trials with patients meeting ex- laboratory findings that the symptoms in criteria A and B developed plicit inclusion criteria, including the basic elements of the during, or shortly after, a withdrawal syndrome.
DSM-IV criteria for AWD, underwent further independent re-view by a second member, with abstraction of data for meta- *Data from the American Psychiatric Association.4 analysis. Any differences of interpretation were resolved by con- †This diagnosis should be made instead of a diagnosis of substance sensus. Meta-analysis was performed when possible using the withdrawal only when the cognitive symptoms are in excess of those usually associated with the withdrawal syndrome and when the symptoms aresufficiently severe to warrant independent clinical attention.
RECOMMENDATIONS
in internal medicine, family practice, psychiatry, and pharma- Recommendations based on the evidence were drafted and cology and individuals involved in primary care medicine, ad- graded according to a published system (Table 2).12 In sev-
diction medicine, and research on alcohol withdrawal.
eral areas, it was recognized that a single recommendation could The primary outcomes considered by the working group not be formulated to guide the treatment of all patients but that included (1) mortality rate, (2) duration of delirium, (3) time the decisions should be guided by a series of clinical consid- required for control of agitation, (4) adequate control of de- erations. In such areas, the level of evidence supporting these lirium, (5) treatment complications, and (6) costs. Acquisi- considerations was identified. In formulating recommenda- tion costs were determined by averaging wholesale prices listed tions, greater weight was given to studies with higher grades of evidence, as defined in Table 2. When no evidence from con- The options considered for managing AWD included phar- trolled studies was available, expert opinion was considered.
macologic and nonpharmacologic strategies. Any pharmaco- Among outcomes, greatest value was given to patient safety, logic agent that has been studied in the management of AWD followed by patient comfort, and then cost. Given the serious- was considered. Nonpharmacologic strategies included the ness of the outcomes involved, it was believed that there would choice of the setting for treatment, evaluation, monitoring, and be little or no variation in patient preference for treatment and that patients would prefer improved medical outcomes (de-creased mortality, shorter duration of delirium, etc).
REVIEW OF THE EVIDENCE
GUIDELINE REVIEW
Searches of the English-language medical literature were con-ducted through MEDLINE using the key words “substance with- The draft guideline was sent for review to first authors of ar- drawal syndrome and alcohol,” “alcohol withdrawal de- ticles from the past 10 years that met the inclusion criteria and lirium,” and “delirium tremens” from the initial entries in to representatives of organizations of medical interest (drawn MEDLINE (January 1, 1966, through September 30, 2001). Ar- from the list published by the American Medical Association) ticles were selected if they involved human subjects and included for whom this guideline may have been of interest. The Ameri- new clinical data on the management of AWD (ranging from a can Society of Addiction Medicine Board of Directors ap- single case report to a prospective randomized trial). Refer- proved the final version in October 2002, with review and re- (REPRINTED) ARCH INTERN MED/ VOL 164, JULY 12, 2004 2004 American Medical Association. All rights reserved.
vision scheduled for November 2007, unless new informationwarrants revision before then.
Table 3. Prospective Controlled Trials Reporting Mortality
as an Outcome*
Deaths, No./
Forty-three articles were identified as having original data, Patients,
Intervention
Administration
Total No.
including 9 prospective controlled trials. In the follow-ing subsections, data are reviewed according to the spe- BENZODIAZEPINES AND OTHER
SEDATIVE-HYPNOTIC AGENTS
Mortality
No controlled trials comparing sedative-hypnotic agents with placebo in treating AWD were identified. How- ever, 5 controlled trials13-17 compared sedative-hypnotic agents and neuroleptic drugs in reducing mortality with AWD (Table 3). Meta-analysis indicated that sedative-
hypnotic use is more effective than neuroleptic use in re- ducing mortality from AWD, with a summary relative risk of mortality with neuroleptic treatment compared with sedative-hypnotic treatment of 6.6 (95% confidence in- The effectiveness of different sedative-hypnotic agents (diazepam, chlordiazepoxide, pentobarbital, par- aldehyde, and barbital) in reducing mortality with AWD was evaluated in 5 controlled trials (Table 3).15,17-20 Twodeaths were reported (both patients were treated with par- Abbreviations: IM, intramuscular; IV, intravenous; PO, oral.
aldehyde); thus, overall, these trials do not demonstrate *Neuroleptic agents are shown in italic.
statistically significant differences among them. The small number of deaths in these trials, however, limits the power ‡The summary relative risk of neuroleptics vs sedative-hypnotics of the 2 studies with mortality is 6.6 (95% confidence interval, 1.2-34.7).
to detect differences in this outcome.
Duration of Delirium
peak action than other benzodiazepines.21 The onset of Table 4 summarizes the results of prospective trials evalu-
action of all benzodiazepines injected IV is rapid, rang- ating different agents in reducing the duration of AWD.
ing from 15 seconds to a few minutes. Peak action of IV Three of 4 trials13,14,17 comparing sedative-hypnotic agents benzodiazepines is 5 to 15 minutes.21 Intramuscular in- with neuroleptic agents demonstrated that the former are jection of chlordiazepoxide and diazepam is associated superior to the latter in reducing the duration of AWD.
with erratic absorption, which can lead to difficulty in (In the fourth trial,16 there was insufficient data in the rapid control of symptoms.22,23 An exception is loraze- original article to calculate P values.) Differences among pam, which has good intramuscular and sublingual ab- sedative-hypnotic agents in reducing duration of AWD sorption.24 Continuous infusion of shorter-acting agents, such as midazolam and lorazepam, has also been used,25with the hypothesis that this may facilitate rapid titra- Time Required to Control Agitation
tion of the dose. However, continuous infusion has notbeen directly compared with intermittent dosing in any Only 2 studies were identified that considered the time required to control agitation. In a study19 comparing rec-tal paraldehyde use with intravenous (IV) diazepam use, Adequate Control of Delirium
the time to achieve adequate sedation, defined as the pa-tient being quiet but awake, was significantly shorter with In the study19 comparing rectal paraldehyde use and IV diazepam (1.1 vs 3.0 hours; P = .02). In contrast, in a diazepam administration, satisfactory control of agita- study20 comparing intramuscular diazepam use and oral tion was achieved in all 17 patients in the diazepam arm barbital therapy, there was no significant difference in but in only 12 of 17 in the paraldehyde arm. In a large, the mean number of hours to achieve adequate seda- multicenter Veterans Affairs study,17 there were no sig- tion, defined as a light sleep from which the patient could nificant differences in achieving adequate control of easily be aroused (11 hours for diazepam vs 8 hours for delirium, but the rate of failure was low. Two of 46 pa- tients taking perphenazine and 1 of 41 taking pentobar- General pharmacokinetic studies have shown that bital were “unresponsive to treatment” with their as- oral diazepam has slightly shorter times to onset and to signed medication. Studies have demonstrated that the (REPRINTED) ARCH INTERN MED/ VOL 164, JULY 12, 2004 2004 American Medical Association. All rights reserved.
Table 4. Prospective Controlled Trials Reporting Duration of Delirium*
Intervention
Route of Administration
Patients, No.
Duration, h
P Value
Abbreviations: IM, intramuscular; IV, intravenous; PO, oral.
*Neuroleptic agents are shown in italic.
†Fisher exact test, 2-tailed.
‡Insufficient data provided in the original article to calculate P value.
required dose of medication can vary substantially among ated with the occurrence of withdrawal seizures if dis- patients and within the same patient over time. In one study,19 the doses for initial calming ranged from 15 to Several case series have reported on the use of other 215 mg of diazepam. Cumulative doses of more than 2000 sedative-hypnotic agents in managing AWD, including mg of diazepam in 2 days,26 more than 2000 mg of diaz- chlormethiazole,34-37 lorazepam,38,39 flunitrazepam,40 epam in 4 days, and more than 20 000 mg of oxazepam pentobarbital,41 propofol,31,42-45 and midazolam.29 Chlor- in 9 days27 have been required for the management of methiazole and flunitrazepam are not available in the AWD. In one published case,28 the patient required 2850 United States. The shorter-acting agents—propofol, pen- mg of midazolam in a 50-day period. Another patient re- tobarbital, lorazepam, and midazolam—were thought to quired 12 424.4 mg of diazepam, 121 mg of lorazepam, be advantageous owing to ease of titration and lower 3050 mg of chlordiazepoxide, and 2025 mg of mid- risk of excess sedation. However, there are no controlled trials comparing short- and longer-acting agents in Although studies have shown no difference in over- all rates of achieving control of delirium among differ-ent sedative-hypnotic agents, case series describe pa- tients whose agitation was refractory to even massive dosesof benzodiazepines but then responded to pentobarbi- Costs can vary greatly depending on the selected drug tal30 or IV infusions of propofol.31 The authors hypoth- and the route of administration. For example, the aver- esized that the benzodiazepine receptors that mediate age wholesale cost of different agents in oral form at ap- ␥-aminobutyric acid–A activity became saturated with proximately equivalent dosages are as follows: chlordi- high doses of benzodiazepines and that further in- azepoxide, 25 mg, $0.07; diazepam, 5 mg, $0.10; and creases thus had little effect on control of delirium. Bar- lorazepam, 1 mg, $0.80.10,29 Intravenous medication, which biturates and propofol act via a different set of recep- is usually needed for adequate control of AWD, is often tors, and, thus, their addition could yield beneficial results.
more than 3 times as expensive as oral medication. For Furthermore, propofol has additional effects on N-methyl- example, the average wholesale cost of these agents in D-aspartate and glutamate receptors that also are be- equivalent dosages are as follows: diazepam, 10 mg, $2.40; lieved to play a role in alcohol withdrawal symptoms.
lorazepam, 2 mg, $2.74; pentobarbital, 350 mg, $4.90; Thus, propofol may be able to modify withdrawal symp- and midazolam, 5 mg, $5.60. (Midazolam would need toms by a different pathway than benzodiazepines.
continuous infusion, with published doses at 0.75 to 10.0µg/kg per minute, or $3.36 to $47.04 per hour for a 70-kg Treatment Complications
person, although prices are expected to decrease as thegeneric form becomes available.) Some practition- In the study19 comparing rectal paraldehyde use and IV ers28,29,46 have described the use of continuous infusion diazepam use, 2 of 17 patients in the paraldehyde group of short-acting benzodiazepines, such as lorazepam or developed respiratory arrest requiring resuscitation. In midazolam. Such infusions can require very large amounts another study,17 1 patient treated with pentobarbital de- of medication over several hours or days. Direct drug costs veloped lethargy progressing to coma. In the remainder (excluding costs of preparation, administration, and moni- of the studies, significant complications related to treat- toring) of $50 335 for a 25-hour infusion of midazolam ment were not observed. It has also been demonstrated were reported for 1 patient,29 and a hospital stay costing in patients undergoing alcohol withdrawal, but not in $26 045 was reported for another patient.46 Further- those with AWD, that shorter-acting agents have a higher more, there are no trials reporting comparative risks and incidence of rebound symptoms32 and may be associ- benefits of intermittent vs continuous IV administra- (REPRINTED) ARCH INTERN MED/ VOL 164, JULY 12, 2004 2004 American Medical Association. All rights reserved.
tions, and no evidence could be identified documenting ciency is associated with Wernicke encephalopathy and Wernicke-Korsakoff syndrome. Thiamine administra-tion has a low risk of adverse effects and can prevent the NEUROLEPTIC AGENTS
development of these conditions. In particular, thia-mine should be given before administration of IV fluids No placebo-controlled trials of neuroleptic agents in AWD containing glucose, as the IV administration of glucose were identified. The trials reviewed earlier demon- may precipitate acute thiamine deficiency.60 strated that neuroleptic drug therapy is inferior to sedative-hypnotic drug use in reducing mortality and duration.
OTHER AGENTS
Nevertheless, neuroleptic agents, especially haloperi-dol, are commonly used with sedative-hypnotic drugs to Several articles describe the use of various other agents calm patients with AWD.47-50 However, neuroleptic agents in managing AWD, including carbamazepine,61 dexa- have the potential to cause a variety of serious adverse methasone,62 physostigmine,63 5-hydroxytrytophan,64 and effects, particularly when used in very high doses, which bromperidol.65 However, these case series have been small may be required to control severe agitation. Chlorproma- and uncontrolled. In addition, although studies of other zine, promazine, and other low-potency typical antipsy- agents (antiepileptic agents, clonidine, etc) in manag- chotic agents have been reported51 to have the greatest ing alcohol withdrawal without delirium have been pub- effect on lowering seizure threshold. Chlorpromazine and lished, no evidence regarding their effectiveness in AWD thioridazine are the most common offenders for caus- ing hypotension, and thioridazine may also prolong theQTc interval, increasing risk for torsade de pointes and SUPPORTIVE CARE
sudden death.52 All neuroleptic agents are thought to havethe potential for causing neuroleptic malignant syn- No controlled studies of nonpharmacologic interven- drome,53 and cases have been reported in patients with tions were identified in the literature search. However, AWD who have received neuroleptic drugs. No studies the literature includes recommendations from clinical ex- were identified describing the use of newer “atypical” an- tipsychotic agents, such as risperidone, olanzapine, and A comprehensive history, physical examination, and quetiapine, for AWD. These agents are at least as effica- thorough diagnostic evaluation are always recom- cious as typical antipsychotic agents for other indica- mended in view of the known morbidity and mortality tions and have a preferable adverse effect profile.
of AWD and the frequent occurrence of associated medi-cal illnesses.47,48,50,66-68 Patients usually need the stan- ␤-ADRENERGIC ANTAGONISTS
dard diagnostic tests to evaluate new-onset delirium, in-cluding neuroimaging to rule out subdural hemorrhaging The effect of ␤-adrenergic antagonists in patients with AWD or other intracranial lesions. Lumbar punctures have been has not been studied. However, delirium is a known ad- recommended in febrile patients when there are no con- verse effect of ␤-adrenergic blocker therapy,54 and in at least traindications.67,68 Further diagnostic evaluation can be 1 controlled study55 of propranolol in alcohol withdrawal undertaken for any indication of commonly coexisting syndrome, there was an increased incidence of delirium.
conditions, such as gastrointestinal hemorrhage, pan-creatitis, and infectious diseases.47,50,67,68 Most experts have MAGNESIUM
recommended general supportive care that includes aquiet, well-lit room, reassurance and reorientation, fre- Low serum magnesium levels have repeatedly been re- quent monitoring of vital signs, and restraints as ported56-59 in patients with AWD. It has been suggested needed.47,49,50,67,68 Dehydration and metabolic abnormali- that magnesium administration reduces neuromuscular ties, such as magnesium and phosphorus deficiency, are activity. However, its use has not been evaluated in con- common with AWD, and it is generally recommended that fluid status and electrolyte levels be monitored care-fully and any abnormalities be corrected.48-50,67,68 ETHYL ALCOHOL
Although there have been small case series describing ad-ministration of alcohol for the prevention and treat- CHOICE OF PHARMACOLOGIC AGENT
ment of withdrawal symptoms, there are no controlledtrials evaluating its use in the prevention or treatment The initial therapeutic goal in patients with AWD is con- of AWD. Ethyl alcohol is known to have the potential trol of agitation, the symptom that should trigger use of for several adverse effects, including hepatic, gastroin- the medication regimens described in this guideline. Rapid testinal, hematologic, and neurologic toxic effects.
and adequate control of agitation reduces the incidenceof clinically important adverse events. Sedative- THIAMINE
hypnotic drugs are recommended as the primary agentsfor managing AWD (grade A recommendation). These Patients with alcohol dependence are often thiamine de- drugs reduce mortality, reduce the duration of symp- ficient, and it has been reported57,59 that patients with AWD toms, and are associated with fewer complications com- have even more substantial deficiencies. Thiamine defi- pared with neuroleptic agents in controlled trials.
(REPRINTED) ARCH INTERN MED/ VOL 164, JULY 12, 2004 2004 American Medical Association. All rights reserved.
somnolence as the recommended therapeutic end point (grade C recommendation). Light somnolence is char-acterized by a state in which the patient is awake but tends Several different benzodiazepines and dosing regimens to fall asleep unless stimulated or is sleeping but easily have been used and recommended. The following are ex- aroused. The amount of medication required for ad- amples of medications and dosing regimens.
equate sedation varies greatly from patient to patient and Benzodiazepines
over time in the same patient. Sedative-hypnotic drug Diazepam, 5 mg intravenously (2.5 mg/min). If the ini- doses needed to suppress AWD are commonly much tial dose is not effective, repeat the dose in 5 to 10 min- higher than doses used to treat severe anxiety or to se- utes. If the second dose of 5 mg is not satisfactory, use date patients presurgically. Tolerance, age, severity of signs 10 mg for the third and fourth doses every 5 to 10 min- and symptoms, and medical comorbidity affect the quan- utes. If not effective, use 20 mg for the fifth and subse- tity of medication needed for adequate control. When us- quent doses until sedation is achieved. Use 5 to 20 mgevery hour as needed to maintain light somnolence.
ing shorter-acting agents, medication should be tapered Lorazepam, 1 to 4 mg intravenously every 5 to 15 min- carefully even after AWD resolves to prevent the devel- utes, or lorazepam, 1 to 40 mg intramuscularly every 30 opment of breakthrough symptoms or the occurrence of to 60 minutes, until calm, then every hour as needed to The medication should be administered by a route Neuroleptics
that supports achievement of rapid control of agitation Haloperidol, 0.5 to 5 mg intravenously/intramuscularly and maintenance of appropriate sedation (light somno- every 30 to 60 minutes as needed for severe agitation.
lence). Intravenous administration has the quickest on- (Only to be used as adjunctive therapy with sedative- set compared with other routes. Intramuscular injec- tion of most benzodiazepines is not recommended owing Haloperidol, 0.5 to 5 mg orally every 4 hours as needed to erratic absorption (grade C). Lorazepam, however, is for agitation not controlled by sedative-hypnotic agents an option in patients with stable cardiovascular status, as it has good intramuscular absorption. Intermittent IVadministrations of long-acting medications and continu-ous IV infusion of short-acting medications seem effec-tive and thus are acceptable. However, continuous IV in- Current evidence does not clearly indicate that a spe- fusion is considerably more expensive, and there is no cific sedative-hypnotic agent is superior to others or that existing evidence of therapeutic superiority.
switching from one to another is helpful. Benzodiaz-epines are most commonly used and recommended by OTHER AGENTS
addiction specialists because of a favorable therapeutic/toxic effect index. Examples of commonly used regi- Neuroleptic agents are not recommended as the sole phar- mens are shown in the Box. However, reported clinical macologic agents in the treatment of AWD because they experience indicates that barbiturates may be consid- are associated with higher mortality, longer duration of ered an option. Owing to difficulties in administration delirium, and more complications compared with seda- and titration of dose, paraldehyde is not recommended tive-hypnotic agents in controlled trials13-17 (grade A (grade A recommendation). Choice among benzodiaz- recommendation). Neuroleptic agents may be consid- epines may be guided by the following considerations: ered for use in conjunction with benzodiazepines when (1) agents with rapid onset control agitation more quickly, agitation, perceptual disturbances, or disturbed think- for example, oral or IV diazepam has a more rapid onset ing are not adequately controlled by benzodiazepine than other agents (level II evidence); (2) agents with long duration of action (eg, diazepam) provide a smooth treat- ␤-Adrenergic antagonists may be considered for use ment course with less breakthrough symptoms; (3) agents in conjunction with benzodiazepines in selected pa- with shorter duration of activity (eg, lorazepam) may have tients for control of persistent hypertension or tachycar- lower risk when there is concern about prolonged seda- dia (grade C recommendation). They are not recom- tion, such as in patients who are elderly or who have sub- mended for routine use in all patients with AWD, however, stantial liver disease or other serious concomitant medi- as there is no evidence that they improve outcomes in cal illness (level III evidence); and (4) the cost of different AWD, and ␤-adrenergic antagonists, particularly pro- benzodiazepines can vary considerably.
pranolol, may worsen delirium (level V evidence).
If a patient demonstrates agitation that is not con- Ethyl alcohol is not recommended because there trolled with extremely large doses of benzodiazepines, are no controlled trials and there are well-known ad- use of pentobarbital or propofol can be considered (grade verse effects (grade C recommendation).
There is no evidence that magnesium therapy spe- cifically benefits the delirium in alcohol withdrawal. How- DETERMINATION OF DOSE
ever, magnesium deficiency is common in patients with AND ROUTE OF ADMINISTRATION
AWD. Magnesium should be provided for demon-strated hypomagnesemia, and it is also safe and reason- It is recommended that the dose be determined specifi- able to include it in IV fluids given for volume repletion cally for each individual patient and that medications be provided renal function is normal and levels are moni- given in doses sufficient to achieve and maintain light (REPRINTED) ARCH INTERN MED/ VOL 164, JULY 12, 2004 2004 American Medical Association. All rights reserved.
Parenteral administration of thiamine (100 mg daily Accepted for publication September 4, 2003. for at least 3 days, IV or intramuscularly) is recom- This study was supported by the American Society of mended to prevent or treat Wernicke-Korsakoff syn- Addiction Medicine and the Stepping Stones Foundation, Bed- We thank Emily Williams for assistance with manu- This guideline is not a substitute for the experience and judgment of a physician. It has been developed to enhance The following recommendations are based on the clinical the physician’s ability to practice evidence-based medicine. experience of recognized experts; they have not been the Presented authors’ opinions are not necessarily representa- subject of controlled studies (grade C recommendations).
tive of the agencies for which they work. Correspondence: Michael F. Mayo-Smith, MD, MPH, EVALUATION
Veterans Administration Medical Center, 718 Smyth Rd,Manchester, NH 03104 (Michael.Mayo-Smith@med.va.gov). On admission or transfer of a patient from one setting toanother, a thorough medical evaluation is needed to de-termine appropriate diagnostic tests, monitoring, and medication. Elderly patients and those with concurrentmedical conditions, acute and chronic, are at higher risk 1. Buschbaum DG, Buchanan RG, Poses RM, Schnoll SH, Lawton MJ. Physician of complications. Concurrent medical conditions are com- detection of drinking problems in patients attending a general medical practice.
J Gen Intern Med. 1992;7:517-521.
mon and may include dehydration, unrecognized head 2. Graham AW. Screening for alcoholism by life-style risk assessment in a com- trauma, electrolyte abnormalities, infections (including munity hospital. Arch Intern Med. 1991;151:958-964.
meningitis), gastrointestinal hemorrhage, pancreatitis, liver 3. Moore RD, Bone LR, Geller G, Mamon JA, Stokes EJ, Levine DM. Prevalence, disease, and myocardial infarction. These conditions may detection and treatment of alcoholism in hospitalized patients. JAMA. 1989;261: not be obvious or self-reported in delirious patients.
4. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition. Washington, DC: American Psychiatric Association; MONITORING
5. Mayo-Smith MF. Pharmacological management of alcohol withdrawal. JAMA. • Close monitoring by nursing personnel is criti- cal in providing protection for the patient and for main- 6. Stendig-Lindberg G, Rudy N. Stepwise regression analysis of an intensive 1-year taining accurate information to guide ongoing medical study of delirium tremens. Acta Psychiatr Scand. 1980;62:273-297.
management. In many cases, continuous, one-to-one 7. Cutshall B. The Saunders-Sutton syndrome: an analysis of delirium tremens.
Q J Stud Alcohol. 1965;26:423-448.
observation and monitoring may be required to ensure 8. Victor M, Adams RD. The effect of alcohol on the nervous system. Res Publ Assoc safe and adequate management of agitated and disori- Res Nerv Ment Dis. 1953;32:526-573.
9. Ferguson JA, Suelzer CJ, Eckert GJ, Zhou XH, Dittus RS. Risk factors for delirium • Vital signs should be monitored regularly in all tremens development. J Gen Intern Med. 1996;11:410-414.
patients. The appropriate frequency of monitoring de- 10. 2001 Red Book. Montvale, NJ: Medical Economics Data Inc; 2001.
11. Hedges LV, Olkin I. The Logit Method: Statistical Methods for Meta-analysis.
pends on the frequency of medication administration, Orlando, Fla: Harcourt Brace Jovanovich; 1985:40-41.
concurrent medical conditions, and the degree of ab- 12. Cook DJ, Guyatt GH, Laupacis A, Sackett DL. Rules of evidence and clinical recom- mendations on the use of antithrombotic agents. Chest. 1992;102:305S-311S.
13. Friedhoff AJ, Zitrin A. A comparison of the effects of paraldehyde and chlor- When high doses of benzodiazepines are needed, promazine in delirium tremens. N Y State J Med. 1959;59:1106-1063.
or when continuous infusions of medication are used, 14. Thomas DW, Freedman DX. Treatment of the alcohol withdrawal syndrome: com- or when patients have significant concurrent medical parison of promazine and paraldehyde. JAMA. 1964;188:244-246.
15. Chambers JF, Schultz JD. Double-blind study of three drugs in the treatment of conditions, cardiac monitoring and oximetry should be acute alcoholic states. Q J Stud Alcohol. 1965;26:10-18.
in place and resuscitative equipment should be readily 16. Golbert TM, Sanz CJ, Rose HD, Leitschuh TH. Comparative evaluation of treat- ments of alcohol withdrawal syndromes. JAMA. 1967;201:113-116.
17. Kaim SC, Klett CJ. Treatment of delirium tremens: a comparative evaluation of MANAGEMENT
four drugs. Q J Stud Alcohol. 1972;33:1065-1072.
18. Brown JH, Moggey DE, Shane FH. Delirium tremens: a comparison of intravenous • A quiet room with good lighting and environ- treatment with diazepam and chlordiazepoxide. Scot Med J. 1972;17:9-12.
mental cues (eg, a clock and a calendar) may help reduce 19. Thompson WL, Johnson AD, Maddrey WL. Diazepam and paraldehyde for treat- ment of severe delirium tremens: a controlled trial. Ann Intern Med. 1975;82: • Physical restraints may be needed temporarily to 20. Kramp P, Rafaelsen OJ. Delirium tremens: a double-blind comparison of diaz- protect agitated patients from injuring themselves and to epam and barbital treatment. Acta Psychiatr Scand. 1978;58:174-190.
protect staff. Guidelines have been formulated on the ap- 21. Greenblatt DJ, Shader RI, Abernethy DR. Drug therapy: current status of ben- propriate use of restraints to ensure patient safety.69,70 zodiazepines. N Engl J Med. 1983;309:410-416.
22. Kanto J. Plasma concentrations of diazepam and its metabolites after peroral, If patients cannot take oral medications or maintain ad- intramuscular, and rectal administration: correlation between plasma concen- equate oral intake, or if more rapid sedation is needed, IV tration and sedatory effect of diazepam. Int J Clin Pharmacol Biopharm. 1975; fluids and medications are recommended. Fluid and electrolyte balance should be maintained, and monitor- 23. Greenblatt DJ, Shader RI, MacLeod SM, Sellers EM. Clinical pharmacokinetics of chlordiazepoxide. Clin Pharmacokinet. 1978;3:381-394.
ing of fluid input and output and laboratory variables 24. Greenblatt DJ, Shader RI, Franke K, et al. Pharmacokinetics and bioavailability may be required. Occasionally, endotracheal intubation of intravenous, intramuscular, and oral lorazepam in humans. J Pharm Sci. 1979; and ventilatory support may be required.
(REPRINTED) ARCH INTERN MED/ VOL 164, JULY 12, 2004 2004 American Medical Association. All rights reserved.
25. Favazza AR, Martin P. Chemotherapy of delirium tremens: a survey of physi- 48. Rosenbloom A. Emerging treatment options in the alcohol withdrawal syn- cians’ preferences. Am J Psychiatry. 1974;131:1031-1033.
drome. J Clin Psychiatry. 1988;49:28-32.
26. Nolop KB, Natow A. Unprecedented sedative requirements during delirium tre- 49. Mayo-Smith MF. Management of alcohol intoxication and withdrawal. In: Gra- mens. Crit Care Med. 1985;13:246-247.
ham AW, Schultz TK, eds. Principles of Addiction Medicine. Chevy Chase, Md: 27. Woo E, Greenblatt DJ. Massive benzodiazepine requirements during acute alco- American Society of Addiction Medicine; 1998.
hol withdrawal. Am J Psychiatry. 1979;136:821-823.
50. Adinoff B, Bone GHA, Linnoila M. Acute ethanol poisoning and the ethanol with- 28. Lineaweaver WC, Anderson K, Hing DN. Massive doses of midazolam infusion drawal syndrome. Med Toxicol. 1988;3:172-196.
for delirium tremens without respiratory depression. Crit Care Med. 1988;16: 51. Hyman SE, Arana GW, Rosebaum JF. Handbook of Psychiatric Drug Therapy. 3rd ed. Philadelphia, Pa: Lippincott Williams & Wilkins; 1995.
29. Wolf KM, Shaughnessy AF, Middleton DB. Prolonged delirium tremens requir- 52. Glassman AH, Bigger JT Jr. Antipsychotic drugs: prolonged QTc interval, tor- ing massive doses of medication. J Am Board Fam Pract. 1993;6:502-504.
sade de pointes, and sudden death. Am J Psychiatry. 2001;158:1774-1782.
30. Hill A, Williams D. Hazards associated with the use of benzodiazepines in alco- 53. Francis A, Chandragiri S, Petrides G. Risk factors for neuroleptic malignant syn- hol detoxification. J Subst Abuse Treat. 1993;10:449-451.
drome. Am J Psychiatry. 1998;155:1639-1640.
31. Coomes TR, Smith SW. Successful use of propofol in refractory delirium tre- 54. Zechnich RJ. Beta blockers can obscure diagnosis of delirium tremens. Lancet. mens. Ann Emerg Med. 1997;30:825-828.
32. Ritson B, Chick J. Comparison of two benzodiazepines in the treatment of alco- 55. Zilm DH, Jacob MS, MacLeod SM, Sellers EM, Ti TY. Propranolol and chlordi- hol withdrawal: effects on symptoms and cognitive recovery. Drug Alcohol De- azepoxide effects on cardiac arrhythmias during alcohol withdrawal. Alcohol Clin 33. Mayo-Smith MF, Bernard D. Late-onset seizures in alcohol withdrawal. Alcohol 56. Kramp P, Ronsted P, Hansen T. Barbital and diazepam plasma levels during treat- Clin Exp Res. 1995;19:656-659.
ment of delirium tremens. Acta Psychiatr Scand. 1979;59:263-275.
34. Athen D. Comparative investigation of chlormethiazole and neuroleptic agents 57. Hoes M. Plasma concentrations of vitamin B-1 in alcoholism and delirium tre- in the treatment of alcoholic delirium. Acta Psychiatr Scand. 1986;73:167-170.
mens: pathogenic and prognostic implications. Acta Psychiatr Belg. 1981;81: 35. Schied HW, Kimmerle K, Braunschweiger M. A retrospective comparison of de- lirium tremens cases before and after the availability of chlormethiazole. Acta 58. Sullivan JF, Lankford HG, Swartz MJ, Farrell C. Magnesium metabolism in alco- Psychiatr Scand. 1986;73:157-161.
holism. Am J Clin Nutr. 1963;13:297-302.
36. Schied HW, Braunschweiger M, Schupmann A. Treatment of delirium tremens 59. Hoes MJ. The significance of the serum levels of vitamin B-1 and magnesium in in German psychiatric hospitals: results of a recent survey. Acta Psychiatr Scand. delirium tremens and alcoholism. J Clin Psychiatry. 1979;40:476-479.
60. Marcus R, Coulston AM. Water soluable vitamins. In: Hardman JG, Limbird LE, 37. Feuerlein W, Reiser E. Parameters affecting the course and results of delirium Gilman AG, eds. Goodman and Gilman’s The Pharmacological Basis of Thera- tremens treatment. Acta Psychiatr Scand Suppl. 1986;329:120-123.
peutics. 10th ed. New York, NY: McGraw-Hill Co; 2001:1753-1771.
38. Spencer J. Use of injectable lorazepam in alcohol withdrawal. Med J Aust. 1980; 61. Brune F, Busch H. Anticonvulsive-sedative treatment of delirium alcoholicum.
Q J Stud Alcohol. 1971;32:334-342.
39. Hosein IN, de Freitas R, Beaubrun MH. Intramuscular/oral lorazepam in acute 62. Fischer DK, Simpson RK, Smith FA, Mattox KL. Efficacy of dexamethasone in alcohol withdrawal and incipient delirium tremens. West Indian Med J. 1979;28: benzodiazepine-resistant delirium tremens [letter]. Lancet. 1988;1:1340-1341.
63. Powers JS, Decoskey D, Kahrilas PJ. Physostigmine for treatment of delirium 40. Pycha R, Miller C, Barnas C, et al. Intravenous flunitrazepam in the treatment of tremens. J Clin Pharmacol. 1981;21:57-60.
alcohol withdrawal delirium. Alcohol Clin Exp Res. 1993;17:753-757.
64. Campioni A, Russo Perez G. Treatment of delirium tremens with 5-hydroxytryp- 41. Hillbom ME, Hjelm-Jaeger M. Should alcohol withdrawal seizures be treated with tophan. Ital J Neurol Sci. 1981;2:307-308.
anti-epileptic drugs? Acta Neurol Scand. 1984;69:39-42.
65. Schmatolla E. Interim report: high dosage bromperidol therapy of delirium tre- 42. Stiebel VG, Crippen D, Ermakov S. Treatment of delirium tremens with continu- mens. Acta Psychiatr Belg. 1978;78:180-187.
ous propofol infusion [abstract]. Psychosomatics. 1994;35:193.
66. Romach MK, Sellers EM. Management of the alcohol withdrawal syndrome. Annu 43. Crippen D, Ermakov S. Titrated treatment of delirium tremens using continuous propofol infusion, clonidine, esmolol and cerebral function monitoring [ab- 67. Turner RC, Lichstein PR, Peden JG, Busher JT, Waivers LE. Alcohol withdrawal stract]. Intensive Care Med. 1994;20:1.
syndromes: a review of pathophysiology, clinical presentation and treatment. J Gen 44. Crippen DW. Strategies for managing delirium tremens in the ICU. J Crit Illn. 1997; 68. Thompson WL. Management of alcohol withdrawal syndromes. Arch Intern Med. 45. Ermakov S, Crippen DW. Continuous propofol infusion for sedation in delirium tremens [abstract]. Crit Care Med. 1994;20(suppl):S37.
69. American Psychiatric Association. Seclusion and Restraint: The Psychiatric 46. Hoey LL, Nahum A, Vance-Bryan K. A retrospective review and assessment of Uses. Washington, DC: American Psychiatric Association; 1991. Task Force re- benzodiazepines in the treatment of alcohol withdrawal in hospitalized patients.
Pharmacotherapy. 1994;14:572-578.
70. Allen MH, Currier GW, Hughes DH, Reyes-Harde M, Docherty JP, Expert Con- 47. Lewis DC, Femino J. Management of alcohol withdrawal. Ration Drug Ther. 1982; sensus Panel for Behavioral Emergencies. The Expert Consensus Guideline Se- ries: treatment of behavioral emergencies. Postgrad Med. 2001;May:1-88.
(REPRINTED) ARCH INTERN MED/ VOL 164, JULY 12, 2004 2004 American Medical Association. All rights reserved.
phosphodiesterase type 5 (PDE5 ) inhibitors in patients sions can most likely be drawn for therapy with silde- with cardiac disease. Until such trials become available, PDE5 inhibitors should be used with caution in patients Cardiologists and other physicians treating pa- tients with CHF therefore have to anticipate 1 more chal-lenge. The prevalence of ED in these patients is high. Be- cause of its prognostic superiority to ␤-blockers, the useof the ␣␤-blocker carvedilol in patients with reduced left Correspondence: Dr Nasser Mikhail, Endocrinology Divi- ventricular ejection fraction has increased worldwide. The sion, UCLA School of Medicine, Olive View—UCLA Medi- quantitative proportion of ␣-blocker in 1 carvedilol tab- cal Center, Sylmar, CA 91342-1495 (nasser.mikhail let is approximately one tenth. To the best of our knowl- edge, no study has so far assessed the safety of combi-nation therapy with carvedilol and PDE5 inhibitors.
1. Webster LJ, Michelakis ED, Davis T, Archer SL. Use of sildenafil for safe im- However, Webster and colleagues1 may be in the posi- provement of erectile function and quality of life in men with New York HeartAssociation classes II and III congestive heart failure: a prospective, placebo- tion of having data that can enlighten us on this impor- controlled, double-blind crossover trial. Arch Intern Med. 2004;164:514-520.
tant issue. In their study,1 33 (94%) of the 35 patients 2. Mulhall JP. Deciphering erectile dysfunction drug trials. J Urol. 2003;170: took ␤-blockers, and we would like to ask the authors if 3. Boulton AJM, Selam JL, Sweeny M, Ziegler D. Sildenafil citrate for the treat- any of these patients actually were treated with the “mod- ment of erectile dysfunction in men with type II diabetes mellitus. Diabeto- ern ␤-blocker” carvedilol? If so, were there any differ- ences in the blood pressure and heart rate response be- 4. Goldstein I, Lue TF, Padma-Nathan H, et al; the Sildenafil Study Group. Oral sildenafil in the treatment of erectile dysfunction. N Engl J Med. 1998;338: tween patients taking carvedilol and patients being treated In their study, Webster et al1 measured blood pres- sure and heart rate at 15-minute intervals after the in-gestion of sildenafil. However, there appear to be no data Use of Sildenafil Is Safe in Men
telling the readers how often the 35 patients actually took With Congestive Heart Failure
sildenafil during the 12-week study period. We wouldappreciate if the authors could also inform us on the total W ebsteretal1foundthatthePDE5inhibitor numberof50-mgdosesofsildenafilcitratethatweretaken
sildenafil citrate can be safely used in the by the 35 patients. These data are of importance in the overall evaluation of the scientific strength of the study II and III heart failure. The study by Webster and col- and its potential clinical consequences.
leagues1 may include additional data of further clinicalimportance. It is generally recognized that the concomi- tant use of nitrates and PDE5 inhibitors is strictly con- traindicated. Caution in using PDE5 inhibitors in pa-tients receiving treatment with ␣-blockers has also been Correspondence: Dr Mickley, Department of Cardiology, warranted, and combination therapy cannot be recom- Odense University Hospital, 5000 Odense C, Denmark mended. In the product summary of the most recent com- mercial available PDE5 inhibitor vardenafil hydrochlo-ride (launched in 2003), it is emphasized that clinical data 1. Webster LJ, Michelakis ED, Davis T, Archer SL. Use of sildenafil for safe im- addressing the safety of combination therapy with vard- provement of erectile function and quality of life in men with New York HeartAssociation classes II and III congestive heart failure: a prospective, placebo- enafil and ␣-blockers are insufficient. Similar conclu- controlled, double-blind crossover trial. Arch Intern Med. 2004;164:514-520.
Error in Box. In the Original Investigation by Mayo-Smith et al published in
the July 12 issue of the ARCHIVES (2004;164:1405-1412), titled “Management
of Alcohol Withdrawal Delirium: An Evidence-Based Practice Guideline,” there
was an error in the box on page 1410. The example medication regimen for
lorazepam should have read as follows: Lorazepam, 1 to 4 mg intravenously
every 5 to 15 minutes, or lorazepam, 1 to 4 mg intramuscularly every 30 to 60
minutes, until calm, then every hour as needed to maintain light somnolence.
(REPRINTED) ARCH INTERN MED/ VOL 164, OCT 11, 2004 2004 American Medical Association. All rights reserved.

Source: http://curriculum.toxicology.wikispaces.net/file/view/delirium+etoh.pdf