Dr. Baron is professor and chair in the Department of Psychiatry, and Drs. Garbely and Boyd are assistant professors of psychiatry, all at Temple University School of Medicine in Philadelphia, Pennsylvania.
Disclosures: Dr. Baron is consultant to the California Academy of Family Practice and Eli Lilly; is on the speaker’s bureau of the California Academy of Family Practice; and is a member of the Food and Drug Administration data monitoring team for Pfizer (contracted through AC-I). Dr. Garbely is on the speaker’s bureaus of AstraZeneca and Pfizer. Dr. Boyd reports no affiliation with or financial interest in any organization that may pose a conflict of interest.
Off-label disclosure: This article includes discussion of unapproved/investigational treatments for substance intoxication and withdrawal disorders.
Please direct all correspondence to: David Baron, MSEd, DO, Professor and Chair, Department of Psychiatry, Temple University School of Medicine, Philadelphia, PA 19125; Tel: 215-707-8483; Fax: 215-707-0726; E-mail: firstname.lastname@example.org.
• Evaluating substance abuse is critical in emergency psychiatry.
• Intoxication is substance specific.
• Managing withdrawal is important.
• Triage of substance abusers requires a comprehensive work-up.
This article focuses on the clinical evaluation and initial treatment of patients with substance abuse problems who present to an emergency department. The importance of making an accurate differential diagnosis and consideration of all relevant biopsychosocial factors is highlighted. The authors offer a treatment algorithm for emergency department clinicians and psychiatrists working in an emergency department setting to consider when assessing patients with intoxication or withdrawal from drugs of abuse. As emergency departments serve an important triage function, level of care determinates are highlighted. Three important clinical federal guidelines developed by the Center for Substance Abuse Treatment are presented and discussed, which are relevant to all emergency department clinicians who evaluate patients who are intoxicated or in withdrawal. These include the following: First, never give medications to an intoxicated patient and immediately discharge them. Second, avoid discharging any intoxicated patient to the street. Last, differentiate between acute intoxication and withdrawal and assess the potential for self harm, intentional and non-intentional.
Alcohol-related emergencies are very prevalent, accounting for a rate of 28.7 per 1,000 United States population emergency department visits (1992–2000) according to a study published in 2004.1 In fact, a vast majority of traumas are alcohol related, not to mention the myriad medical complications associated with alcohol use. Fifty percent acute trauma deaths due to injury are related to excessive drinking and up to 40% of all admissions to medical and surgical wards are alcohol related.2 Illicit drug use, on the other hand, poses distinct but no less challenging dilemmas regarding associated emergencies. Often, the individual will present with medical and/or psychiatric manifestations of intoxication or withdrawal, which may be compounded by the use of multiple substances.3 Nonetheless, as this article has a focus on psychiatric emergencies, concentration is placed on substance intoxication and withdrawal presentations.
Evaluation of the patient with alcohol and drug intoxication and withdrawal requires discipline. First, obtain a careful and meticulous history from the patient and any available credible collateral sources. The latter is often of critical importance for numerous reasons. Patients tend to minimize their alcohol and drug use in an effort to avoid unwanted retribution or inquiry into their personal lives.4 They are generally afraid to be labeled as having an alcohol- and/or drug-related problem due to the stigma associated with those diagnoses. Patients who are intoxicated or in severe withdrawal are often unable to give an intelligible history.5
Second, perform a careful and complete physical and mental status examination looking for signs of trauma, autonomic instability, potentially treatable neurologic and/or general medical conditions, and co-occurring mental disorders. Data from step two can often give valuable clues as to the substances that the patient has ingested (Table 1). For example phencyclidine (PCP) intoxication produces vertical nystagmus which is not seen in any other substance intoxication.6 The data from step two can give an estimate of the severity of the ongoing process. As will be described in more detail subsequently, clinical instruments based on symptoms and signs, such as the Clinical Institute Withdrawal Assessment-Revised (CIWA-Ar) and the Clinical Opiate Withdrawal Scale (COWS), can be very helpful in management decisions.7
Third, obtain laboratory testing and imaging studies based on the data obtained from the first two steps. Since historic data is often inaccurate or incomplete in patients heavily intoxicated or in severe withdrawal states, a urine drug screen and a blood-alcohol level (BAL) or breathalyzer testings are indicated in virtually all instances.3 Although urine drug assays provide only qualitative information (ie, tell the clinician if the patient has used a substance, but not how much he or she has used), blood alcohol levels provide quantitative data that can be very helpful in clinical decision making (Table 2). Since alcohol is metabolized at a constant rate (zero order kinetics) of ~.015 g/dL (1 drink) per hour, the fall in BAL over a period of time can be predicted.
Substance abuse emergencies are often associated with trauma and acute medical problems, requiring careful screening.3 For example, patients with an abnormal mental status, focal neurologic symptoms, or head trauma should have imaging of their central nervous system performed before they are transferred to a detoxification, rehabilitation, or mental health facility. Patients with stimulant intoxication and chest pain need to have myocardial damage excluded. Similarly, patients intoxicated with PCP, methylenedioxymethamphetamine (MDMA; ecstasy), or stimulants should be screened for the possibility of rhabdomyolysis, particularly if they have incurred soft tissue damage or are febrile at presentation. A complete blood count, serum electrolytes, blood urea nitrogen, creatinine, and hepatic function tests are helpful since major hematologic abnormalities, electrolyte imbalances, or severe renal or hepatic dysfunction will significantly impact clinical decision making.
Together, the history, physical, and mental status examinations, coupled with appropriate laboratory analysis and imaging studies, should allow for the identification of patients who have co-occurring illnesses that would necessitate initial treatment in a non-psychiatric setting such as a general medical or surgical inpatient unit.8 Substance-abusing patients who are hemodynamically unstable, whether resulting from intoxication or withdrawal, must be managed in an intensive care unit. Patients who are actively suicidal, psychotic, or manic should be treated in an acute psychiatric setting. A useful way to discuss management of these emergencies is to group syndromes in terms of their potential lethality and treatability of symptoms.
Group I consists of syndromes that can produce severe morbidity and mortality if not recognized and appropriately managed. Syndromes include alcohol withdrawal; sedative, hypnotic, and anxiolytic withdrawal; and opiate overdose.
Group II consists of syndromes that usually do not result in death or severe morbidity if untreated but for which specific treatment is effective and beneficial. Syndromes include opiate withdrawal as well as intoxication with stimulants, dissociative agents, or hallucinogens
Group III consists of syndromes that do not result in severe morbidity or mortality if untreated and for which nothing beyond supportive care is helpful. Syndromes include mild to moderate alcohol or sedative, hypnotic, and anxiolytic intoxication; withdrawal from stimulants, dissociative agents, or hallucinogens; and cannabinoid intoxication and withdrawal.
Alcohol withdrawal occurs in patients with heavy and prolonged alcohol use who abruptly stop or drastically reduce their alcohol intake. Symptoms generally first appear within 6–8 hours after the last drink, peak at 1–2 days, and generally resolve within 7 days. However, the onset of alcohol withdrawal symptoms and signs can be delayed in certain circumstances. For example, it is not unusual for alcohol withdrawal to be first diagnosed in chronic drinkers 24–36 hours after admission to the hospital for unrelated medical or surgical conditions. Thus, the absence of symptoms 8 hours after the last drink does not exclude the possibility that alcohol withdrawal will occur in the future. The manifestations of alcohol withdrawal can include autonomic hyperactivity; tremor; insomnia; nausea and vomiting; transient visual, tactile, or auditory hallucinations or illusions; agitation; anxiety; and grand mal seizures. An excellent clinical instrument to help assess the severity of alcohol withdrawal is the CIWA-Ar (Table 3). Fortunately, the vast majority of patients with alcohol dependence will only develop mild to moderate withdrawal symptoms and, in uncomplicated cases, can be safely managed in the outpatient setting. However, ~5% of patients will develop severe alcohol withdrawal symptoms including alcoholic hallucinosis (hallucinations in the presence of a clear sensorium) or delirium tremens. Although alcoholic hallucinosis is not life threatening, delirium tremens are associated with a mortality rate of 5% to 15%. Alcohol withdrawal seizures occur in 5% to 15% of patients with a greatly increased incidence in those with a prior history of withdrawal seizures.
The treatment of alcohol withdrawal may be provided in a variety of settings and should be individualized. Table 4 lists the relative contraindications for outpatient alcohol detoxification. Additionally, all patients should be assumed to be chronically malnourished and prophylactic supplementation with thiamine 100 mg/day should be started immediately to prevent the development of Wernicke’s encephalopathy (Table 5). While Wernicke’s is a potentially reversible syndrome if aggressively treated, if missed it can progress to become Korsakoff’s syndrome (Table 5) which is irreversible. Benzodiazepine substitution therapy is considered the standard of care for alcohol detoxification. However, there are several dosing strategies that can be employed and no one benzodiazepine has been shown to be superior in treating alcohol withdrawal. One such regime that has been reported to be used with success is symptom-triggered treatment using the CIWA-Ar to guide the need for lorazepam administration.
Patients in alcohol withdrawal may become agitated, violent, and psychotic. The first pharmacologic approach to behavioral control in such patients should be the use of benzodiazepines.9 Benzodiazepines suppress the withdrawal process and have intrinsic antiepileptic properties. If a patient remains psychotic despite benzodiazepine therapy, the judicious addition of a high-potency antipsychotic is often helpful. However, the administration of antipsychotics as the initial approach to treating disruptive patients in alcohol withdrawal is not recommended. Neuroleptics do not ameliorate the basic withdrawal process and often lower the seizure threshold. Another important caveat is to be sure that a patient’s agitation and/or psychosis is not due to alcohol intoxication as opposed to withdrawal. The administration of benzodiazepines to inebriated patients will increase the level of intoxication, with potentially fatal consequences.
Sedative, hypnotic, or anxiolytic withdrawal (henceforth referred to as sedative withdrawal) presents with clinical symptoms and signs that are virtually identical to those from alcohol withdrawal. However, because different sedatives have different half lives, the timing of the onset, peak, and duration of symptoms can be quite variable. Also, even patients who abruptly stop sedatives with a very short half life, such as alprazolam (t1⁄2 6–12 hours), tend to have withdrawal symptoms that persist for much longer (3–4 weeks) than those of patients withdrawing from alcohol. Seizures are far more common in sedative withdrawal than with alcohol.
Sedative use disorders have a particularly high co-occurrence rate with both primary psychiatric disorders and other substance use disorders, which can greatly complicate management of withdrawal. An example of this problem is that benzodiazepines are often prescribed to treat anxiety, and yet anxiety is a prominent symptom of benzodiazepine withdrawal. It can be very difficult to discern whether a patient who presents with anxiety on alprazolam is in withdrawal or under-medicated for his psychiatric condition. Similarly, Strain and colleagues10 have found that ~50% of patients in methadone maintenance programs were abusing benzodiazepines. In dually diagnosed patients, the general approach is to aggressively treat the most lethal problems first while stabilizing the less dangerous entities. For example, in the heroin addict dependent on alprazolam in withdrawal from both substances, the approach should be to stabilize the patient’s opiate withdrawal with methadone or buprenorphine while concentrating on benzodiazepine detoxification.
There are several possible strategies for sedative detoxification. One is to restabilize the patient on the abused substance and then slowly taper him off that drug. This strategy is often successful when sedatives with long half-lives are involved. A major problem with this approach is that the taper needs to be conducted very slowly, which usually requires the patient to control the use of his drug of choice over many months. A second approach, applicable to sedatives with long and short half-lives, is to switch patients to phenobarbital, which is rarely abused because of its slow onset of action and long half-life. Phenobarbital detoxification should be initiated in the hospital if any of the conditions in Table 4 are present or in an addiction specialist’s office if they are not. Phenobarbital detoxification is almost never initiated in the emergency department because of the low therapeutic index of barbiturates. When the degree of a patient’s sedative tolerance is unclear, a pentobarbital challenge test can be very helpful in determining the initial phenobarbital dose. No matter which of the above approaches are chosen, long-term follow up of sedative-dependent patients is mandatory.
Opiate intoxication produces a characteristic constellation of clinical symptoms and signs (Table 6). Respiratory depression can be lethal and, therefore, is a medical emergency. Most overdoses with opiates with a rapid onset of action become symptomatic almost immediately and are triaged away from psychiatrists until they are out of danger. The patients most likely to be triaged directly to psychiatrists are those who have ingested oral opioids with a slow onset and long half-life. Methadone has been particularly problematic because it has a half-life of 24 hours and it takes 5 days to establish steady state drug levels. Patients are at risk for overdose if they are initially started on daily doses of methadone >30 mg/day or if their daily doses are titrated upward at intervals shorter than 5 days. Particularly troublesome are patients with chronic pain being managed with methadone. Methadone is administered at 6-hour intervals in many analgesic regimes which can lead to the accumulation of very high drug levels. Similar problems may occur with buprenorphine, which also has slow onset of action and a terminal half-life of 37 hours.
It is important to recognize the symptoms and signs of opiate intoxication (Table 6) and to consider it in the differential in polysubstance abusers. Urine assays for methadone and buprenorphine can be diagnostically helpful in such instances. A trial of parenteral naloxone can be both diagnostically and therapeutically useful with the caveat that the abrupt onset of an opiate withdrawal syndrome might be precipitated. If a slow onset, long half-life opioid has been ingested, it is important to realize that symptoms could worsen later when peak drug levels are reached.
The abrupt cessation or rapid reduction in the dose of opioids in chronic users produces a characteristic clinical picture that is the mirror image of that produced by opiate intoxication (Table 6). In uncomplicated cases, opiate withdrawal is not life threatening but does cause significant suffering in the affected individual. The time of onset, duration, and severity of opioid withdrawal is variable and depends on the drug used, its half-life, the amount used per day, and the duration of use. For example, heroin withdrawal usually starts with in 8–12 hours after the last dose and lasts 3–5 days, whereas methadone withdrawal does not begin until 36–48 hours after the last dose and takes ≥3 weeks to resolve.
It is recommended that opiate withdrawal be managed with medications because of the unnecessary suffering that can occur.10 The choice of which medication regimen to use depends on the patient’s treatment goals, the availability of appropriately licensed facilities and personnel, and the expected duration and severity of the withdrawal syndrome. The use of COWS (Table 7) can be very helpful in quantitating the initial severity of opiate withdrawal and in assessing its progression over time.
Patients whose ultimate goal is long-term abstinence and who wish to avoid further exposure to opioids are candidates for detoxification with clonidine. Clonidine is an a2 agonist which, when given at high enough doses, is effective at suppressing the symptoms of adrenergic hyperactivity seen in opiate withdrawal. Unfortunately, clonidine does not effectively treat non-adrenergic opioid withdrawal symptoms which necessitate the use of additional medications for symptom relief.
Methadone substitution is another common and effective treatment for opioid withdrawal symptoms. Patients taking unknown quantities of opiates who manifest withdrawal symptoms are judiciously given small doses of methadone and their symptomatic response is closely observed. When withdrawal symptoms are suppressed, the daily dose of methadone needed to keep the patient comfortable without sedation can be calculated and administered. If the patient desires to become abstinent, he can be gradually tapered off methadone by decreasing the dose by 5 mg/day, watching closely for the reemergence of objective signs of withdrawal. If the patient wishes to go directly into a methadone maintenance program, tapering is not necessary. The main drawbacks to methadone treatment of opioid withdrawal are pharmacokinetic and regulatory. Because of methadone’s long half-life, a new steady state blood concentration will not be achieved until 5 days after a dose change. Therefore, patients stabilized on a given dose of methadone must be closely watched thereafter for the development of symptoms and signs of opiate intoxication. By Federal law, methadone can only be administered for opioid withdrawal by specially licensed programs except when the patient is hospitalized for the treatment of another acute medical condition. Therefore, treating opioid withdrawal through a general psychiatrist’s office or an emergency room using methadone is illegal.
The third option for treating opioid withdrawal is buprenorphine substitution therapy. Buprenorphine is a partial opioid mu receptor agonist and a kappa receptor antagonist that is Food and Drug Administration approved for both detoxification and maintenance treatment of opioid dependence. Buprenorphine comes as a sublingual pill either with or without naltrexone. Buprenorphine, like methadone, has a long half-life and therefore can be taken once daily. However, unlike methadone, buprenorphine administration can precipitate withdrawal in dependent patients by displacing a full mu agonist from the opioid receptor. Therefore, it is important that patients are in significant opioid withdrawal (a COWS score of ≥13) before they receive buprenorphine. Patients in opioid withdrawal are given small doses of buprenorphine at hourly intervals until they become asymptomatic. Stabilizing doses of buprenorphine usually are between 8–24 mg/day. Depending on the patient’s treatment plan, buprenorphine can either be tapered off to achieve abstinence or continued as maintenance therapy. The advantages of buprenorphine over methadone are that it causes less respiratory depression and can be administered in the outpatient setting by specially trained and certified physicians. The disadvantages of buprenorphine compared to methadone are that it may not fully suppress withdrawal symptoms in patients chronically on very high opioid doses and that it is significantly more costly. Both treatments are very effective.
Patients intoxicated with stimulants, dissociative agents, and/or hallucinogens also fall into Group II. Very severe intoxication with alcohol, sedatives, stimulants, hallucinogens, and “club drugs” usually present as medical emergencies and should be triaged to medicine for stabilization before being referred to the psychiatrist. However, patients with mild to moderate levels of intoxication with those substances are often directly triaged to psychiatrists. The main clinical problems associated with these intoxications are aggression, agitation, psychosis, and suicidality. The treatment of all of these conditions involves providing a safe environment for the patient, the administration of psychotropic drugs to control behavior, and careful monitoring of the patient during the period of intoxication. Important clinical pearls to remember for this group of emergencies include the following. First, the use of benzodiazepines in patients with alcohol or sedative intoxication is contraindicated. Second, patients intoxicated with stimulants and PCP have a high proclivity for violence and verbal de-escalation alone is often ineffective and can be dangerous. The mainstay of treatment is psychotropic drug administration. Third, patients with PCP and stimulant intoxication are at high risk of developing rhabdomyolysis, making the use of physical restraints hazardous. Fourth, patients intoxicated with LSD and other pure hallucinogens manifest hallucinations in the presence of a clear sensorium. Verbal de-escalation is often effective in these patients. Fifth, MDMA intoxication can be accompanied by dehydration, hyperpyrexia and, in extreme cases, rhabdomyolysis. Sixth, intoxicated patients are at increased risk for suicidality and should be carefully evaluated psychiatrically. Seventh, intoxicated patients should not be discharged until symptoms and signs of intoxication have resolved. Last, assaults on healthcare professionals by intoxicated patients are a significant problem. The warning signs of an impeding assault and how to appropriately react to them are discussed in detail by Novitsky and colleagues in this issue.11
Group III disorders are not substance abuse emergencies and, beyond general supportive care, do not require specific treatment. It is important to remember that patients withdrawing from long-term stimulant use often become very depressed when abstinent and, therefore, arranging for appropriate follow-up care is essential.
Unhealthy alcohol use and substance abuse are significant contributors to the utilization of medical and psychiatric emergency services. Efficiently managing these patients is critical to the successful operation of an emergency service. Treating substance abuse emergencies requires a focused medical evaluation and psychiatric work-up, including a mental status exam and relevant laboratory tests. Treatment strategies will vary based on the drug used, so a “one size fits all” approach is not recommended. This article reviewed the most commonly encountered drugs of abuse and the classic presenting signs and symptoms. The suggested treatment algorithm should assist the clinician in making acute treatment decisions, including triage strategies.
There is little literature on the problem of polysubstance use presenting in emergency settings. Since drug-drug interactions have not been well studied, evaluating and ultimately managing a polydrug abuser remains very empiric and challenging.
Likewise, there is little data on the potential modulating effects of age, race, gender, menstrual cycle phase, diet, sleep, and comorbid psychopathology on the presentation and clinical course of substance abuse emergencies. Until more data becomes available, the evaluation and management of substance abuse emergencies will remain anecdotal rather than evidence based. PP
1. McDonald III A, Wang N, Camargo C. US emergency department visits for alcohol-related diseases and injuries between 1992 and 2000. Arch Intern Med. 2004;164(5):531-537.
2. Sadock B, Sadock V. Comprehensive Textbook of Psychiatry. 8th ed. Philadelphia, PA: Lippincott Williams & Wilkins; 2005.
3. Graham A, Schultz T. Principles of Addiction Medicine. 3rd ed. Chevy Chase, MD: American Society of Addiction Medicine; 2003.
4. Lowinson J, Ruis P. Substance Abuse: A Comprehensive Textbook. 4th ed. Philadelphia, PA: Lippincott Williams & Wilkins; 2005.
5. Galanter M, Klebr H. Textbook of Substance Abuse Treatment. 4th ed. Washington, DC: American Psychiatric Publishing; 2008.
6. O’Brien C. Drug Addiction and drug abuse. In: Goodman and Gilman’s The Pharmacological Basis of Therapeutics. 11th ed. New York, NY: McGraw-Hill Medical Publishing Division: 2006:607-627.
7. Kosten T, O’Connor P. Management of drug and alcohol withdrawal. N Engl J Med. 2003(18);348:1786-1795.
8. Detoxification and Substance Abuse Treatment. A Treatment Improvement Protocol (TIP 45). Rockville, MD: US Department of Health and Human Services. Substance Abuse and Mental Health Services Administration Center for Substance Abuse Treatment; 2006.
9. Swift R. Drug Therapy for Alcohol Dependence. N Engl J Med. 1999;340(19):1482-1490.
10. Strain E, Stitzer M. The Treatment of Opioid Dependence. Baltimore, MD: Johns Hopkins University Press; 2006.
11. Novitsky MA, Julius RJ, Dubin WR. Non-pharmacologic management of violence in psychiatric emergencies. Primary Psychiatry. 2009;16(9):49-53.