You are called into resuscitation, where EMS has just finished transferring a 55 year old man onto the ED stretcher. They were called after his family found him unconscious at home. His current GCS is 3…
The differential diagnosis of altered mental status is huge and can be overwhelming in the face of an acutely ill, undifferentiated emergency department patient. I try to sort through diagnoses based on how quickly they could kill the patient and how quickly I can treat them.
What could kill my patient immediately?
- Cardiac arrest
- Airway obstruction
- Breathing (oxygenation)
This is textbook emergency medicine. The immediate first step is to check for a pulse. At the same time, my nurses are getting the patient on the monitor and getting a full set of vital signs. Next, I assess airway patency and breathing pattern. If necessary, I start with basic, temporizing airway maneuvers, such as positioning, oral/nasal airways, or an LMA. (I don’t want to intubate a patient who only requires D50W or narcan.) The need for c-spine precautions should also be considered.
NOTE: Don’t forget to get the history from EMS before they leave. The patient can’t communicate and the paramedics almost always have important information.
What could kill my patient in the next few minutes?
- Intracranial hypertension and herniation
My first priority is getting the glucose checked, primarily so it does not get overlooked. Next, I ask my nurses to start working on vascular access while I perform a rapid, focused primary survey:
- Neuro: Pupils, eye movements, corneal reflex, moving all 4 extremities, reflexes, muscle tone, any asymmetry?
- Signs of impending herniation: Hypertension, bradycardia, and irregular respirations (Cushing’s triad); posturing; unilateral blown pupil?
- Breathing pattern: Regular, Cheyne-Stokes, irregular, apnea?
- Toxidrome: Vital signs, pupils, skin
- Signs of shock: Cap refill, skin warm or cold?
- Abdo: Any obvious pain or masses?
- Trauma: Any clear signs of trauma?
This all takes about 1 minute to complete. At this point, I am ready to consider if any immediate therapeutic interventions are required:
- Hypoglycemia: D50W 1-2 amps IV
- Opioid toxidrome (or suspicion): Naloxone 0.2-0.4mg IV q2-3min. (If the patient is stable, I will usually start with a much lower dose (0.04mg IV) to avoid precipitating rapid opioid withdrawal.)
- Signs of impending herniation: Intubate; provide analgesia and sedation; elevated the head of the bed; respirate to a target pCO2 of 35mmHg; Mannitol 0.5-1gram IV or 3% hypertonic saline 2-3ml/kg IV bolus.
What could kill my patient in the next 10 minutes?
- Still the ABCs
- Aortic disasters
It’s easy to get lost in the differential. After the rapid primary survey and initial interventions, I remind myself to reassess the ABCs. If a rapidly reversible cause hasn’t been identified, I will start planning for a definitive airway. The next two diagnostic moves are an ECG and the ultrasound machine. The ECG will provide essential diagnostic information about ischemia, arrhythmias, overdoses, and hyperkalemia. Ultrasound examination may be a RUSH exam for hypotension, an aorta exam, or a more focal exam based depending on the findings of the primary survey.
Interventions at this point: For hypotension, I will start a fluid bolus or blood products depending on the context. If there is any suspicion of anaphylaxis, I will give epinephrine 0.5mg IM. If there is reason to suspect hyperkalemia, or any bizarre appearing ECG, I will empirically give calcium (2-3 amps of calcium gluconate IV).
What could kill my patient over the next few hours?
- Intracranial hemorrhage
- Alcohol Withdrawal
- Status epilepticus (presumably non-convulsive if I didn’t recognize it immediately)
- Necrotizing fasciitis (look everywhere)
- Abdominal catastrophes
- Metabolic problems (DKA, HHNK, hyponatremia, thyroid disorders, adrenal disorders)
After the rapid assessment and management of immediate life threats, the next step is to ensure the patient is adequately resuscitated before the inevitable trip to the CT scanner. A definitive airway should be in place before traveling to radiology. Any signs of shock are addressed with fluids, blood, and/or vasopressors. Blood work, probably already drawn reflexively by the nurses, should be sent off. Unless there is a clear alternative diagnosis, I start empiric antibiotics on everyone. (Acyclovir can also be considered for herpes encephalitis.) Non-convulsive status epilepticus is a difficult diagnosis to make, but warrants specific consideration.
It is important to use all possible sources of information, including old charts, family, friends, and EMS. I also take a few extra minutes to perform a more thorough physical exam, ensuring that I have seen every inch of the patient’s skin. If the initial temperature check was with a peripheral thermometer, I will ask for a core temperature. I also specifically search for things like medic-alert bracelets, medication lists, or contact information that might be hidden among the patient’s possessions. (The LP, although possibly a necessary test, is not an emergent test. In sick patients, it is generally better to get therapy started empirically, and worry about the LP later.) Finally, once the patient is stabilized, I will get them to the CT scanner for images of their brain (and any other organs indicated by the primary survey).
What could I be missing?
After ruling out initial life threats, starting empiric therapy, and getting the patient to the CT scanner, I focus on running through the larger differential diagnosis. As much as I hate mnemonics, if I still haven’t identified the cause at this point, I will often pull out my phone and run through the horrible AEIOU TIPS:
This mnemonic is useless as a memory aid. Almost all the letters stand for more than one possible etiology, and the letter “I” is there twice (but actually represents fewer etiologies than some of the other letters). There is no way that you will be able to consistently reproduce this list from memory in emergent situations. However, it can be useful as a checklist after completing the initial resuscitation.
In addition to empiric antibiotics, there are other empiric therapies that should occasionally be considered depending on the speed and availability of testing: thyroxine for possible myxedema, dexamethasone for adrenal crisis, benzodiazepines for possible non-convulsive status epilepticus, and specific antidotes for any suspected toxidromes.
Other FOAMed Resources
Altered Mental Status on EMBasic
Approach to: Altered Mental Status on EMin5
Killer coma cases part 1 (the found down patient) and part 2 (the intoxicated patient) on Emergency Medicine Cases
A time-based approach to elderly patients with altered mental status on ALiEM
Dazed and Confused: The Approach to Altered Mental Status in the ED on Taming the SRU
The Pupil Exam in Altered Mental Status on PEMBlog
Bassin BJ, Cooke JL, and Barsan WG. Chapter 94. Altered Mental Status and Coma. In: Adams JG ed. Emergency Medicine Clinical Essentials, 2e. Philadelphia: Elsevier Saunders; 2013.
Odiari EA, Sekhon N, Han JY, David EH. Stabilizing and Managing Patients with Altered Mental Status and Delirium. Emergency medicine clinics of North America. 33(4):753-64. 2015. PMID: 26493521
Huff JS. Altered Mental Status and Coma. In: Tintinalli JE et al eds. Tintinalli’s Emergency Medicine: A Comprehensive Study Guide, 7e. New York, NY: McGraw-Hill; 2011.
Huff JS, Stevens RD, Weingart SD, Smith WS. Emergency neurological life support: approach to the patient with coma. Neurocritical care. 17 Suppl 1:S54-9. 2012. PMID: 22932989
Stevens RD, Huff JS, Duckworth J, Papangelou A, Weingart SD, Smith WS. Emergency neurological life support: intracranial hypertension and herniation. Neurocritical care. 17 Suppl 1:S60-5. 2012. PMID: 22936079
Clinical policy for the initial approach to patients presenting with altered mental status. Annals of emergency medicine. 33(2):251-81. 1999. PMID: 14765552
Kelly MA. Chapter 8. Neurology Emergencies. In: Cameron P et al, eds. Textbook of Adult Emergency Medicine, 4e. Philadelphia: Elsevier Saunders; 2015.
Morgenstern, J. The emergency medicine approach to an unconscious patient, First10EM, February 22, 2016. Available at:
2 thoughts on “The emergency medicine approach to an unconscious patient”
I encourage all of my students and providers to fully understand the Sick/Not Sick approach to pt. assessment. Dynamic critical thinking – with a plan! Thank you for this read…it is well done!