Stimulants in the ER: differential and what to avoid

It is three in the morning. A man in his early thirties arrives at the emergency department with chest pain, cold sweats and a restlessness that keeps anyone from touching the monitor. At triage, someone writes "cocaine user" before any other information. From there, the story seems ready: drug, adrenaline, anxiety, observation and discharge. The problem is that this same phrase fits an acute coronary syndrome, an aortic dissection, a hyperthermia that will worsen, a hypoglycemia, a trauma that no one saw and an infection that started before the use. This text uses stimulant use as a clinical reasoning scenario: how to build the differential when the history of use comes first, which biases appear on shift and how to talk to the person without closing the door. The scene above and the examples throughout the text are fictional and were written for teaching.

NEXO is a daily medical diagnosis game: one clinical case a day, up to six progressive clues, and ICD-10 proximity feedback across five levels.

The text also avoids being a guide to management or doses. Protocols of each service and society guidelines guide treatment. Here the focus is the way of thinking: what question to ask before accepting the most obvious explanation, and how to turn information about substance use into a useful clue, and not into a label that closes the case. If you play the case of the day in NEXO, you will recognize the move: each clue changes the weight of the hypotheses, and no clue alone should close the diagnosis.

Key points

• The history of stimulant use is a strong clue, and for that very reason it can close the reasoning too early.
• Chest pain, agitation and heat are signs with many possible causes; the use of cocaine or amphetamines changes the probability of some of them, without eliminating the others.
• Anchoring, affective bias and diagnostic overshadowing appear frequently when the patient uses substances.
• Asking about use in a neutral way improves the quality of the history and the safety of management.
• The only treatment with consistent evidence for stimulant use disorder, according to systematic reviews, is behavioral; the conversation in the emergency department can be the first door.

Why the word "cocaine" changes reasoning on shift

When the history of use comes before the physical examination, it acts as an anchor. The anchor is the first strong information you receive and that starts pulling all the others toward it. The phenomenon has been described for decades in the diagnostic error literature, and Pat Croskerry placed it among the most common cognitive errors in emergency medicine, alongside premature closure and confirmation bias (Croskerry, 2003). If you want to review the concept of heuristic before continuing, the post on examples of heuristics in clinical reasoning shows how useful shortcuts become a trap.

In the stimulant scenario, the anchor has two faces. The first is useful: cocaine and amphetamines do increase the risk of myocardial ischemia, arrhythmias, aortic dissection and hyperthermia, and the classic New England Journal of Medicine review on cardiovascular complications of cocaine summarizes these mechanisms (Lange and Hillis, 2001). Knowing this should increase the pre-test probability of some serious hypotheses. The second face is dangerous: the same information can serve as an explanation for everything, including findings that have nothing to do with the drug.

In practice, the difference between the two faces lies in the question you ask afterward. "Does the use explain this picture?" leads to closure. "Does the use change the probability of which hypotheses, and what still needs explaining?" keeps the differential open. The second question is the same one the post on pre-test probability uses to interpret tests: the clue moves the needle, and you still need to decide what to do with the shift.

The label at triage travels with the patient

A sentence written at triage is usually read by every professional who comes afterward. If the first line of the chart says "cocaine intoxication", the resident taking over the shift reads the case through that lens. This phenomenon of label transmission is known as diagnostic momentum: the hypothesis gains strength with each handoff, even without new data. The review by Graber and colleagues on diagnostic error in internal medicine identified premature closure as the most frequent cognitive error in the cases analyzed (Graber, Franklin and Gordon, 2005).

Stigma enters the reasoning without asking permission

A systematic review on stigma among health professionals showed frequent negative attitudes toward people with substance use disorders, with shorter consultation time, less empathy and less involvement in care (van Boekel and colleagues, 2013). A study with emergency nurses, residents and preceptors about patients who use methamphetamine described similar perceptions in the emergency department itself (Richards and colleagues, 2019). The post on affective bias explains how irritation, fear or fatigue enter the diagnosis; with stimulants, these emotions appear easily, because the person may be agitated, suspicious or aggressive.

Chest pain with cocaine use: the differential that cannot shrink

Cocaine-associated chest pain is one of the most studied scenarios at the interface between substance use and the emergency department. The multicenter COCHPA study followed patients with chest pain after cocaine use and showed that infarction occurred in a relevant portion of them, even in young people with few classic risk factors (Hollander and colleagues, 1994). Years later, a study published in the New England Journal of Medicine validated a short observation period for low-risk patients with cocaine-associated chest pain, provided the initial assessment was careful (Weber and colleagues, 2003).

What these studies teach about reasoning is that the clue "cocaine" does not reduce the possible severity of the pain; it increases the number of mechanisms that may be at play. Coronary vasospasm, increased oxygen consumption, thrombosis and accelerated atherosclerosis coexist with other causes of chest pain that exist with or without the drug. The American Heart Association scientific statement on cocaine-associated chest pain and infarction organizes these hypotheses and discusses why management has particularities (McCord and colleagues, 2008).

Three hypotheses that are often forgotten

Aortic dissection. The pain may be described as tearing and migratory, but not always. The abrupt rise in blood pressure with stimulants is a plausible mechanism, and the hypothesis must be remembered before any decision that increases the risk of bleeding.

Pneumothorax and pneumomediastinum. The inhaled route of use and Valsalva maneuvers may be in the history. Physical examination and imaging guide the way.

Causes unrelated to the drug. Pulmonary embolism, pericarditis, reflux, musculoskeletal pain and anxiety remain possible. Cocaine use may be coincidental.

The post on differential diagnosis with clues shows a way to keep the list open: for each hypothesis, note what favors it, what weakens it and what still needs to be asked. With stimulants, the "what still needs to be asked" column is usually the longest, because the history was closed early.

When time since use counts as a clue

The interval between use and the onset of pain is a clue that changes probabilities. Pain that starts soon after use points to acute mechanisms; pain that starts hours or days later may have another origin or reflect late complications. Asking when, how much, by which route and with which other substances is part of the history just as much as asking about radiation and duration. The question only works if it is asked without judgment, the subject of the section on communication further on.

Agitation, heat and confusion: when the toxidrome misleads

The classic sympathomimetic picture brings together tachycardia, hypertension, dilated pupils, sweating, agitation and increased temperature. NIDA describes the acute effects of methamphetamine and cocaine on reference pages that serve as a basis for students (NIDA, methamphetamine; NIDA, cocaine). Recognizing this pattern is useful, but it has a limit: several other pictures look like it.

Think of a fictional woman in her early twenties brought in from a party, agitated, with fever and a high heart rate. Her friend says she used "something". The sympathomimetic pattern is there, and the temptation is to close the case. Before that, it is worth asking:

Are there signs of central nervous system infection, such as neck stiffness or altered consciousness that does not improve? Was the capillary blood glucose measured? Are there signs of alcohol or sedative withdrawal, which also cause agitation, tachycardia and sweating? Was there use of other medications that could cause serotonin syndrome or anticholinergic syndrome? Are there signs of trauma, such as a scalp hematoma or a report of a fall? Could the thyroid be involved, in a person with weight loss and palpitations for weeks?

Each of these questions opens a hypothesis that the toxidrome does not rule out. High fever deserves special attention: hyperthermia from stimulants can be severe, and meningitis or sepsis can also present with agitation. The pause reasoning described in the post diagnostic pause before closing the case fits well here: before accepting "it is the drug", ask what else it could be, and what you would lose if you were wrong.

What the monitor shows and what it hides

Heart rate, blood pressure and temperature help classify severity, but they do not tell the cause. An agitated person may have high blood pressure because they are agitated, because they used a stimulant or because they have another ongoing illness. The trend of the signs over the hours says more than a single value. If the agitation decreases and the temperature rises, the picture has changed in nature. If the blood pressure falls instead of rising, another hypothesis has entered the scene.

The biases that most appear with stimulants

The biases below appear in any specialty, and the stimulant scenario has a combination that favors all of them at once: agitated patient, busy shift, strong label and little reliable history. Graber, Kissam and colleagues reviewed cognitive interventions to reduce diagnostic error and found mixed results, with some promising strategies, such as checklists and the question "what else could it be?" (Graber and colleagues, 2012).

Anchoring and premature closure

The "cocaine" anchor closes the differential early. The antidote is to write at least two hypotheses that do not depend on use and decide what would rule them out. The post about dual process theory shows why fast recognition is useful and why it needs a slow check when the error can be serious.

Diagnostic overshadowing

Diagnostic overshadowing is attributing new symptoms to an already known condition, such as a mental disorder or a substance use disorder, without investigating other causes. A person with a history of use may receive fewer tests for chest pain than another without that history, even with the same picture.

Affective bias

When the person is aggressive or distrustful, the team tends to shorten the care. The emotion is legitimate and human; the problem is letting it decide the approach. The post about implicit bias in diagnosis shows how gender, race and age stereotypes also change the way of listening. With substance use, these stereotypes add up.

Confirmation bias

Once the "drug" hypothesis settles in, each new finding is read as proof of it. Tachycardia confirms, sweating confirms, anxiety confirms. The test is to ask whether the finding is more likely with the hypothesis than without it. If it appears equally in several hypotheses, it confirms none.

Table: clue, hypotheses it favors and what still needs checking

Use the table as a reasoning exercise, not as a protocol. It shows how a common clue in the stimulant setting points to several hypotheses and what still needs to be asked before closing the case. The examples are fictional and require clinical review.

Clue	Hypotheses it favors	What still needs checking
Chest pain shortly after cocaine use	Vasospasm ischemia, infarction, aortic dissection	Characteristics of the pain, pulses, serial electrocardiogram, other causes of chest pain
Agitation with fever	Sympathomimetic toxidrome, hyperthermia, central nervous system infection	Neck stiffness, temperature trend, blood glucose, history of other medications
Confusion that does not improve over time	Prolonged drug effect, head trauma, hypoglycemia, encephalitis	Neurological exam, signs of trauma, blood glucose, serial reassessment
Palpitations for weeks and weight loss	Chronic stimulant use, hyperthyroidism, primary arrhythmia	Timeline, thyroid exam, electrocardiogram
Sudden dyspnea after inhaled use	Pneumothorax, pneumomediastinum, bronchospasm	Auscultation, imaging, oxygen saturation
Person denies use, sympathomimetic picture	Unreported use, pheochromocytoma, alcohol or sedative withdrawal	Repeated neutral question, alcohol history, pattern of episodes

How to ask about use without ending the conversation

The quality of the history depends on the person's trust in the one asking. A randomized study showed that the way of naming people who use substances changes the judgment of mental health professionals: describing someone as a "substance abuser", instead of a "person with a substance use disorder", increased the tendency to blame and punish (Kelly and Westerhoff, 2010). If words change the judgment of the one asking, they also change the willingness of the one answering.

Some ways of asking help:

• Normalize the question: "I ask everyone who arrives with chest pain about alcohol, cigarettes and other substances, because it changes what I look for."
• Explain the clinical reason: "Some substances change the risk to the heart and to blood pressure. Knowing what you used helps me decide on the tests."
• Ask about duration, route and mixing, and not only about "yes or no".
• Prefer "person who uses cocaine" to labels such as "addict" or "junkie", including in the medical record.
• Ensure the confidentiality of the information, within local rules.

The post about empathy and clinical reasoning in the consultation shows that attentive listening and reasoning do not compete; listening improves the data that feed reasoning. In the emergency department, this means that two minutes of a well-asked question can avoid hours of investigation in the wrong direction.

What to do with the answer

The answer about use goes on the list of clues like any other. It changes probabilities, but it does not replace physical examination, electrocardiogram, blood glucose and trauma assessment when they apply. If the person denies use and the picture suggests use, the hypothesis remains possible; if the person confirms use and the picture has findings that use does not explain, other hypotheses remain alive.

After the emergency: what the evidence shows about treatment

Once the acute picture is stabilized, clinical reasoning moves from the question "what is happening now?" to "what can change from here on?". In stimulant use disorder, the most consistent evidence comes from psychosocial interventions. The Cochrane review by Minozzi and colleagues on psychosocial interventions for psychostimulant use found benefit especially for contingency management, which links incentives to confirmed abstinence (Minozzi and colleagues, 2016).

For pharmacotherapy, the situation is different. A systematic review with meta-analysis on pharmacological treatments for cocaine use disorder found no strong evidence of benefit for any of the drug classes studied (Chan and colleagues, 2019). This information changes the conversation: the student who expects to close the case with a prescription needs to learn to offer something else, such as referral to behavioral treatment, information about risks and an agreed follow-up.

The discussion about beta-blockers as an exercise in uncertainty

For many years, the recommendation to avoid beta-blockers in people intoxicated by cocaine was taught as a fixed rule, based on the theoretical risk of unopposed alpha stimulation. The AHA statement discusses the topic with caution (McCord et al., 2008), and a review on the pharmacological management of acute conditions related to methamphetamine reassessed the available evidence and showed a more nuanced picture than the rule suggests (Richards, 2018). For the student, the value of the example is methodological: a rule taught as certain may have a weaker basis than it seems. Management in each case follows local protocols and the assessment of the responsible physician. The post on diagnostic uncertainty helps to deal with situations in which the evidence does not settle the question.

Discharge with follow-up, not just discharge

Safe discharge includes warning signs explained in clear language, arranged follow-up and, when the person wants it, information about treatment services. If there is a risk to life or intense suffering, advise the person to seek the necessary help and ensure they know where to find care. The way discharge is carried out influences whether the person will return to the service when needed.

How to train with cases and with NEXO

NEXO offers a fictional clinical case per day, with a realistic clinical presentation and up to six progressive clues. You propose a diagnosis by searching for the ICD-10 code and receive feedback on proximity in five levels, which shows whether the guess was the exact code, the same category, the same group, the same chapter or another chapter. This format connects with the theme of this text: a strong clue, such as the report of use, can push you toward an entire chapter, and the feedback shows the distance between the guess and the answer.

To use the case of the day as training in reasoning about biases, try three habits:

Before each guess, write in one sentence why that hypothesis is better than the second one. When a clue seems decisive, ask what other diagnosis it would also explain. After the result, read the case explanation and note which clue you overvalued or ignored.

Three exercises to train with cases

Reasoning in the face of stimulants is trained like any differential: with repetition, with feedback and with cases that force you to keep hypotheses open. The post on case-based learning in clinical practice shows how the case organizes knowledge around decisions. Three exercises work well in a study group or alone:

Move the clue. Read the same case twice: once with the information about use at the beginning, once with it at the end. Compare the hypotheses you wrote in each reading.

Write the serious hypothesis that cannot be missed. For each case involving stimulant use, note the serious hypothesis that does not depend on the drug and the finding that would rule it out.

Rewrite the medical record without labels. Take a fictional progress note with stigmatizing terms and rewrite it in person-centered language. Note what changes in the reading of the case.

These exercises align with the idea of calibration described in the post on calibration of confidence in diagnosis: knowing how much you trust the hypothesis and checking later whether the confidence was right.

The limits of use of the game are described on the NEXO economy page. If you research teaching of clinical reasoning, the page for researchers explains how to get in touch. Professionals who write or review cases find information in curation and in creators, and institutions can learn about the proposal on the partnerships page.

Frequently asked questions

Does cocaine use really increase the risk of heart attack in young people?

Yes. Prospective studies in emergency departments showed heart attack in people with chest pain after cocaine use, including young people with few classic risk factors (Hollander et al., 1994). Young age alone does not justify ruling out the hypothesis.

If the person confirms use, do I still need to investigate other causes?

Yes, whenever the condition has findings that use does not explain or when the error could be serious. Use changes probabilities; it does not exclude aortic dissection, infection, hypoglycemia, trauma or other causes of chest pain.

Is there an approved medication to treat cocaine use disorder?

The available systematic reviews found no strong evidence for any drug class (Chan et al., 2019). Psychosocial interventions, especially contingency management, have the most consistent evidence (Minozzi et al., 2016).

How do you write the medical record without stigmatizing?

Describe behaviors and facts instead of labels: "reports inhaled cocaine use two hours ago" instead of "drug user". The study by Kelly and Westerhoff (2010) shows that words change the judgment of the reader.

Does this text serve as a care protocol?

This text deals with clinical reasoning and biases. Management and doses follow your service's protocols and the guidelines of medical societies, under supervision.

How does this topic appear on exams?

Exams usually test the differential of chest pain with stimulant use, the sympathomimetic toxidrome and communication with the patient. Practicing reasoning with cases helps more than memorizing lists.

Final checklist

• I noted at least two serious hypotheses that do not depend on stimulant use.
• I asked about time, route, amount and mixing of substances, in a neutral way.
• I measured blood glucose and looked for signs of trauma and infection.
• I reassessed vital signs over time, not only on arrival.
• I reviewed the medical record to remove stigmatizing labels.
• I left the case with agreed follow-up, not just a discharge.

Want to practice reasoning with progressive clues? Play the case of the day.

References

1. Chan B, Kondo K, Freeman M, et al. Pharmacotherapy for Cocaine Use Disorder: a Systematic Review and Meta-analysis. J Gen Intern Med. 2019. PMID 31183685.
2. Croskerry P. The importance of cognitive errors in diagnosis and strategies to minimize them. Acad Med. 2003. PMID 12915363.
3. Graber ML, Franklin N, Gordon R. Diagnostic error in internal medicine. Arch Intern Med. 2005. PMID 16009864.
4. Graber ML, Kissam S, Payne VL, et al. Cognitive interventions to reduce diagnostic error: a narrative review. BMJ Qual Saf. 2012. PMID 22543420.
5. Hollander JE, Hoffman RS, Gennis P, et al. Prospective multicenter evaluation of cocaine-associated chest pain. Acad Emerg Med. 1994. PMID 7614278.
6. Kelly JF, Westerhoff CM. Does it matter how we refer to individuals with substance-related conditions? A randomized study of two commonly used terms. Int J Drug Policy. 2010. PMID 20005692.
7. Lange RA, Hillis LD. Cardiovascular complications of cocaine use. N Engl J Med. 2001. PMID 11484693.
8. McCord J, Jneid H, Hollander JE, et al. Management of cocaine-associated chest pain and myocardial infarction: a scientific statement from the American Heart Association. Circulation. 2008. PMID 18347214.
9. Minozzi S, Saulle R, De Crescenzo F, et al. Psychosocial interventions for psychostimulant misuse. Cochrane Database Syst Rev. 2016. PMID 27684277.
10. NIDA. Cocaine. National Institute on Drug Abuse. nida.nih.gov.
11. NIDA. Methamphetamine. National Institute on Drug Abuse. nida.nih.gov.
12. Richards JR. Beta-Blockers and Evidence-Based Guidelines for the Pharmacological Management of Acute Methamphetamine-Related Disorders and Toxicity. Pharmacopsychiatry. 2018. PMID 28859206.
13. Richards JR, Hawkins JA, Acevedo EW, et al. The care of patients using methamphetamine in the emergency department: Perception of nurses, residents, and faculty. Subst Abus. 2019. PMID 29595368.
14. van Boekel LC, Brouwers EP, van Weeghel J, Garretsen HF. Stigma among health professionals towards patients with substance use disorders and its consequences for healthcare delivery: systematic review. Drug Alcohol Depend. 2013. PMID 23490450.
15. Weber JE, Shofer FS, Larkin GL, et al. Validation of a brief observation period for patients with cocaine-associated chest pain. N Engl J Med. 2003. PMID 12571258.