Key Clinical Questions
How common are alcohol and sedative dependence in the general hospital setting?
How is alcohol withdrawal optimally managed?
What factors make it more likely for a patient to develop delirium tremens?
What are the differences in management of withdrawal from alcohol, benzodiazepines, and barbiturates?
What should be done with the patient dependent on alcohol or a sedative following the inpatient medical hospitalization?
Introduction
Sedatives constitute a diverse group of agents, including alcohol (ethanol, ethyl alcohol, EtOH), benzodiazepines, nonbenzodiazepine hypnotic medications (so-called Z-drugs such as zolpidem [Ambien]), barbiturates, and several other compounds, including chloral hydrate and meprobamate. The class also includes gamma hydroxybutyrate (GHB), a short-acting sedative, which is specifically addressed in Chapter 236, “Other Drugs of Abuse.” The agents covered in this chapter collectively are often referred to as sedatives, sedative-hypnotics, or sometimes hypnotics. Here the term sedatives will be used for this class of compounds; it will not apply to all classes of medications (eg, antihistamines, various antidepressants, antipsychotics, opioids, and others) that may produce sedation. Note that while the Diagnostic and Statistical Manual, Fourth Edition, Text Revision (DSM-IV-TR) uses the term sedative-hypnotic, this chapter employs the term sedative for the same class of agents. While all sedatives may produce drowsiness, sleep, and a feeling of relaxation, they do not always do so, particularly when tolerance has developed. Many sedatives also produce muscle relaxation, and all may produce addiction (see Diagnosis section for definition), which will likely complicate and exacerbate the course of the underlying disorder(s) that led to hospitalization.
The sections in this chapter specifically cover alcohol, benzodiazepines, Z-drugs (zolpidem, zaleplon, zopiclone, eszopiclone), barbiturates, and others, as appropriate. All sedatives, including alcohol, have the common effect of dose-dependently depressing neuronal function, though this may occur by one or more mechanisms. Most of these agents, in overdose, may produce fatal respiratory depression, but this is rarely the primary reason for use. Human beings have used alcohol for countless millennia to produce many desired effects, including diminished anxiety, relaxation, sleep, analgesia, and intoxication. Sedating medicinal plants have also been employed for millennia, usually to facilitate sleep and less often for any of many other purposes, some of which reflect the actual pharmacology of one or more compounds present in the plant(s). It is beyond the scope of this chapter, however, to address sedating herbs and other medicinal plants.
Estimates of the prevalence of alcohol use disorders in the general population vary, in part depending on the country being surveyed and the diagnostic criteria used, but a recent estimate put the lifetime prevalence of alcohol dependence (according to DSM-IV-TR) at 12.5% in the United States. Estimates of the prevalence of alcohol dependence among hospitalized patients also vary, partly because studies have employed different methods to ascertain and define alcohol use disorders, and because some studies have included measures of “at risk” drinking rather than alcohol use disorders, and because different populations have been studied. One study from the 1980s suggested that as many as 25% of all admissions to a tertiary hospital center had an alcohol use disorder (sometimes quiescent, or in remission) and that most physicians did not make the diagnosis of an alcohol use disorder in those patients likely affected by one. Further, even when the diagnosis was made, there was frequently no intervention documented to address the alcohol use disorder. There are few data from the United States on the prevalence of alcohol use disorders (AUDs) among hospitalized patients, but a recentstudy from Germany suggested that 5.3% of a very large sample of patients had a current AUD, that 3.6% were currently drinking above recommended levels, with an additional 3.1% having alcohol dependence in remission. The prevalence of dependence on sedatives is difficult to estimate because of limited data, but it is almost certainly less than the prevalence of alcohol dependence and probably in the range of 1 to 5% of hospitalized patients, most of whom have co-occurring anxiety disorders or other substance use disorders (SUDs).
Ingestion of sedatives or alcohol typically produces reductions in anxiety and muscle tension, and impairments in memory; intoxication may result in uncoordinated motor functioning (gait ataxia, finger to nose incoordination, positive Romberg sign), nystagmus, slurred speech, and various aberrant behaviors, including behavioral disinhibition, impairment of consciousness, reduced respirations, and drowsiness or sleep. Nevertheless, intoxication may also be difficult to detect, particularly when patients have developed significant tolerance to these agents or are medically or psychiatrically ill and not previously known to the doctors caring for them. Even individuals with high levels of tolerance, however, are not immune to the potential memory disturbances and frank amnesia associated with intoxication with alcohol and other sedatives. These periods of amnesia are frequently referred to as blackouts, and they appear to be more likely with short-acting sedatives, alcohol, or the combination of alcohol with other sedatives. Some predatory individuals attempt to take advantage of this by spiking the drink of a potential victim (of robbery, rape, or other crimes) with a sedative. This possibility should be considered in the evaluation of crime victims who appear intoxicated and/or who cannot remember the crime(s) purportedly perpetrated against them.
Overdoses with many sedatives and alcohol may produce sleep, unconsciousness, coma and even death, which results from the respiratory depression produced by all these agents. Overdoses with alcohol, often referred to as alcohol poisoning, may result in coma and death. Alcohol poisoning is the cause of about 50 deaths in the United States annually, often involving an adolescent or young adult who is not a regular drinker but who consumes a large volume of alcohol in a short period of time, producing very high blood levels of alcohol. Benzodiazepine and Z-drug overdoses by themselves are rarely, if ever, lethal, but their presence in combination with other sedative agents, including alcohol, works synergistically to produce respiratory depression. Combination overdoses of benzodiazepines or Z-drugs with nonlethal doses of barbiturates or opioids may also be lethal. Barbiturates very effectively suppress respiration and are indeed lethal in overdose, even when taken as a single agent without any other respiratory depressants. Phenobarbital doses of 6 g to 10 g may be fatal, while smaller doses of short-acting barbiturates (eg, secobarbital, pentobarbital) in the range usually of 2 g to 3 g may be fatal. The older, nonbarbiturate sedatives, chloral hydrate, meprobamate, ethchlorvynol, and glutethimide, produce effects similar to barbiturates during intoxication, and all can be fatal in overdose, even when taken without other medications or alcohol.
The history, either directly from the patient or from his or her associates, usually provides sufficient evidence to confirm intoxication with alcohol or sedatives. The smell of alcohol on the patient’s breath or clothing is an obvious clue to alcohol intoxication, which may then be confirmed by the blood alcohol content (BAC, discussed in detail in the Diagnosis section), obtained directly from a serum sample or indirectly by measurement of the breath alcohol content with a breathalyzer. Urine testing for sedatives (see later discussion) will help confirm the diagnosis of sedative intoxication.
An old joke defines an alcoholic as someone who drinks more than his or her doctor. Fortunately, better screening instruments exist to identify those who are at risk for alcohol dependence, and to identify individuals with an AUD. The best method available to identify an individual with an AUD is eliciting a good history. For a variety of reasons, many physicians do not develop the skill and habit to take a careful and thorough alcohol use history. Many physicians simply have not learned what questions to ask beyond quantity, frequency, and duration of use questions. While these questions are important, they infrequently elicit the kind of information that helps practitioners sort out whether an individual actually has an alcohol use disorder. And these quantity, frequency, duration questions have led many to create unfortunate rules of thumb, such as, “If the patient tells you he drinks x drinks daily, double it.” (Some variations on this include unreferenced and unsupportable rules such as “Multiply the amount the patient admits to drinking by 2.4”). Among the problems with these rules of thumb is that they encourage practitioners to expect patients to lie to them and thus make the practitioners less likely to ask the questions necessary to make a diagnosis of an alcohol or sedative use disorder. When this happens, of course, it reduces further the likelihood of a meaningful dialogue between physicians and patients about alcohol use. Another of the many phenomena that may make physicians less likely to screen for alcohol and other SUDs is the mistaken belief, often supported unwittingly by the press, which only publicizes celebrity relapses and never sustained sobriety, that treatment for SUDs is usually fruitless. This is very far from the truth. Substance dependence syndromes are chronic, life threatening, frequently remitting and relapsing illnesses that definitely respond to treatment (at rates similar to the chronic illnesses commonly encountered in general medicine).
Use of a few lists of questions or criteria for making substance abuse diagnoses will dramatically improve the efficiency of discussions between patients and their providers about alcohol use (and, with minor modifications, the use of other substances). Specifically, the CAGE questions (Table 233-1), so named because the acronym CAGE includes the first letter of the word containing the main idea of each of four questions related to alcohol effects, have been shown to be a very efficient screening instrument to identify individuals who may have an alcohol use disorder. The CAGE questions should arguably be incorporated into every clinical history, given their utility and simplicity in helping identify patients in need of additional screening to determine whether an alcohol use disorder is present. These questions also encourage a broader and more meaningful discussion about the role of alcohol in a person’s life, which discussion in turn supports even brief efforts to help patients identify their own reasons to change drinking behaviors. The import of such a discussion cannot be underestimated because alcohol dependence does respond to treatment, and it directly impacts dozens of other conditions addressed by hospitalists.
To identify an individual with an alcohol or sedative use disorder, it is helpful to keep in mind both the criteria for abuse and dependence. These are listed in Table 233-2. Recalling these criteria when taking a substance use history can help identify those individuals more likely to have an SUD, because the diagnostic criteria can guide a physician to explore the various consequences of and associated findings in addictive use of any substance. In addition, physicians can identify potential problem sedative use by inquiring about the following: use of sedatives for any reason other than the prescribed indication, continued use despite resolution of the target symptoms identified by a prescribing physician, continued use despite harm due to use, and acquisition of the sedatives from multiple physicians or from other illicit sources.
| Dependence | Abuse |
|---|---|
|
|
|
|
Laboratory testing is helpful in assessing individuals with alcohol use disorders, and it can provide important clues that may assist in making a diagnosis of alcohol or sedative dependence. First, the blood alcohol content (BAC), may be measured directly in serum, or it may be approximated, usually quite closely, by use of a breathalyzer to measure the amount of alcohol in exhaled air, using a simple, common (if somewhat variable) ratio of breath to BAC. The BAC is usually reported in units of g/dL (sometimes referred to as grams percent) or mg/dL (milligrams percent), reflecting the mass of alcohol present in 100 mL (1 dL) of blood. Table 233-3
Related posts:
Strategies for Cost-Effective Care
Building, Growing, and Managing a Hospitalist Practice
Designing a Hospitalist Compensation and Bonus Plan
The Face of Health Care Emerging Issues for Hospitalists
Medical Malpractice
Preventing and Managing Adverse Patient Events: Patient Safety and the Hospitalist
Full access? Get Clinical Tree
