The mental status changes associated with opioid toxicity are those of general CNS depression and “dozing off.” The term narcotic, often used as a synonym for opioid, derives from the Greek word narke, “to make numb.” The typical patient with opioid toxicity may drift off and appear to be sleeping during the interview or even fail to complete sentences while talking.

Opioids frequently produce a characteristic set of physical examination findings, commonly referred to as the “opioid toxidrome.” These findings are the constellation of bradypnea (slow breathing), hypopnea (shallow breathing), and/or apnea (not breathing); pinpoint pupils; and decreased or absent bowel sounds.

A positive urine drug screening for opiates in the setting of symptoms of opioid toxicity supports the diagnosis. A negative urine drug screen in a patient who appears to be intoxicated with opioids does not rule out this diagnosis, however, because several commonly abused opiates (including oxycodone and methadone) fail to react with many commercially available opioid drug screens. A positive drug screen in a patient whose symptoms are not consistent with opioid intoxication likely represents recent use and persistent metabolites in the urine, rather than a true intoxication.

Opiate toxicity is effectively reversed with the competitive antagonist naloxone (Narcan). In patients who are not known or presumed to be addicted to opiates, a starting dose of 0.4 to 2.0 mg is reasonable. Naloxone has an extraordinary safety profile.

Indiscriminate dosing of naloxone in the known or presumed opiate addict, however, will quickly precipitate withdrawal and should be avoided. In such patients, we recommend an initial dose of 0.01 mg intravenously, repeated every 1 to 2 minutes until the patient is effectively oxygenating (pulse oximetry of 93% or greater) on room air.

In the setting of presumed opiate toxicity, a positive response to naloxone is essentially diagnostic.

The hallmark of sedative-hypnotic toxicity is a general decrease in mental status. Thought processes are slowed considerably. Speech may be slurred. Patients may be excessively emotional or angry as a result of the disinhibiting effects of these agents. GHB is somewhat unique in that profoundly sedated patients may rouse easily in response to noxious stimulation, only to become profoundly sedated again after the stimulus is removed.

Vital sign abnormalities depend in large part on the substance ingested. Benzodiazepines rarely produce significant vital sign abnormalities. Most patients with uncomplicated alcohol toxicity have normal vital signs. Barbiturates are not nearly as safe as benzodiazepines (it is for this reason that they have largely been replaced by benzodiazepines for virtually every clinical indication). Barbiturate toxicity may result in bradycardia, hypotension, and respiratory depression or arrest. GHB toxicity may result in bradycardia, coma, respiratory depression, or apnea.

Patients with alcohol toxicity may have the sweet, characteristic odor of this substance on their breath. Alcohol may also produce a significant horizontal nystagmus. Barbiturate intoxication may be associated with cutaneous blisters. With all of these agents, coordination as assessed on the neurological examination is impaired.

Qualitative urine toxicology screening may identify both barbiturates and benzodiazepines. Care should be taken not to overinterpret such results, however. Urine testing usually identifies metabolites of the drug class in question and does not confirm that the observed behavior is due to the drug identified. For example, a patient who has not taken diazepam for several days may still have a positive urine toxicology screen, despite the fact that he or she is no longer under the influence of that drug.

Care should also be taken not to overinterpret serum ethanol levels. Unlike urine toxicology screens, serum ethanol levels reliably quantify the blood concentration of the substance in question. However, the serum ethanol level is an imperfect gauge of the degree of the patient’s intoxication.

Ethanol is a substance to which patients may develop functional and physiological tolerance, occasionally to a profound extent. A naïve drinker with a blood alcohol level of 100 mg/dL might be profoundly inebriated, whereas a chronic alcoholic may be awake and conversant with an alcohol level of 300 mg/dL. The latter patient may be in acute ethanol withdrawal when his or her blood alcohol level is 150 mg/dL.

The most appropriate measure of sobriety in a patient who has been consuming ethanol, therefore, is the patient’s clinical status. Arbitrary numerical cutoffs designed to “determine” a patient’s suitability for discharge from the emergency department or for a psychiatric interview are not based on sound physiological principles. Patients who are awake, responding appropriately, and not slurring their speech and are ambulatory with a steady gait are no longer clinically intoxicated. Serum ethanol levels play virtually no role in that determination.

Treatment of the patient with sedative hypnotic toxicity is generally supportive and in mild to moderate cases consists of intravenous fluids and supplemental oxygen. Patients who have used long-acting agents (e.g., phenobarbital) should be admitted (usually to the general medical service) when they are symptomatic, because they are unlikely to regain a clear sensorium in a reasonable time period. Patients who are intoxicated with short-acting agents (such as ethanol) can be observed until resolution of their mental status changes.

Flumazenil, a benzodiazepine antagonist, should almost never be used in the setting of known or questioned benzodiazepine intoxication. When given indiscriminately to a patient who is habituated to benzodiazepines, flumazenil may precipitate acute benzodiazepine withdrawal. This syndrome is clinically similar to—and as dangerous as—ethanol withdrawal. Furthermore, patients to whom flumazenil has been administered may prove refractory to further treatment with benzodiazepines because of the antagonist effects of the flumazenil.

Patients who are intoxicated with hallucinogens usually complain of vivid visual hallucinations, in contradistinction to the predominantly auditory hallucinations that characterize the primary psychoses or intoxication with sympathomimetic substances. When asked, patients frequently describe the appearance of geometric patterns. Inanimate objects may seem to pulsate or move, and concrete boundaries may be described as “melting.” Synesthesias, a melding of sensory perceptions, are also common; patients may “hear” colors or “see” sounds. Time perception may be distorted or completely lost.

Hallucinogens produce few peripheral physical examination findings; the presence of such findings should prompt a search for other substances or medical conditions. There are exceptions to this general rule, however. Ingestion of nutmeg, for example, typically produces severe nausea and vomiting before the onset of hallucinogenic symptoms.

Laboratory testing is generally unremarkable in patients who have used hallucinogens. These medications produce no pathognomonic metabolic or other laboratory abnormalities. The presence of significant laboratory abnormalities should prompt a search for a concurrent or alternative condition.

Patients who are intoxicated with dissociative agents may present in different ways, depending on the extent of intoxication. Patients who are mildly intoxicated may appear confused and aggressive, whereas severe intoxication may produce complete dissociation and unresponsiveness. Phencylcidine users may alternate between severe agitation and an outwardly calm exterior. Ketamine users try to self-titrate to a level of perceptual distortions without producing complete dissociation. When such users overshoot, they describe the dissociative state as a “K-hole.”

Dissociative agents, particularly phencyclidine and ketamine, frequently produce hypertension and tachycardia. One important corroborative physical examination finding is vertical and/or rotatory nystagmus. Vertical or rotatory nystagmus strongly suggests intoxication with one of a few drugs, including the dissociative agents. Ketamine may also produce excessive salivation.

Urine tests for phencyclidine are commonly available, but the results must be interpreted with caution. Some patients ingest phencyclidine congeners, which possess all of the chemical activity of phencyclidine but do not react with the phencyclidine screening test. Although the results of such a test are truly negative (because the patient did not actually ingest phencyclidine), they may mislead the clinician from the true diagnosis of dissociative agent toxicity. Alternatively, dextromethorphan (in doses used to treat cough, not doses used in abuse) may generate a false positive test because of cross-reactivity with the phencyclidine assay.

There are no other laboratory tests that help confirm dissociative agent toxicity or routinely guide management. Patients who are severely agitated and physically aggressive should be screened for rhabdomyolysis via testing for urine myoglobin (which cross-reacts with the hemoglobin test of the urine dipstick) or serum creatinine kinase.