Sunday, December 31, 2023, 1300 hours
The nurse calls about your next patient, a 20-year-old woman who is behaving erratically. From their behaviour–the patient came in with a friend–she feels the patient is withdrawing from a drug and will need to go to hospital.
You arrive and the patient’s friend immediately confronts you with talk. She catches you off guard, actually. You are taken aback but not perturbed, just uncomfortable at first. You find the patient writhing in bed and her friend sitting at the bedside, immediately trying to engage you with history and conversation. The nurse has moved on to another patient in another room, so it is just you, the patient, and her friend. The friend conveys her concern in rapid-fire sequence: that the patient may need psychiatric help and is under a lot of stress. You speak briefly with the friend as the patient rests, initially non-committal. She seems in pain, writhing in bed, unable to get comfortable. You learn she was on psychotropic medication but isn’t now, the details of which and when are unknown. The friend’s chatter is insistent and consistent: her friend may need psychiatric help.
You address the patient, and she responds, says she has abdominal pain, but her responses are erratic and staccato, and not infrequently: “I don’t know.” You ask about associated symptoms and she volunteers diarrhoea, associated nausea, no vomiting. The answers are so erratic you move quickly to a systems review: headache, fever, chest pain. A quick assessment: pupils 3mm, reactive; speaking briskly; heart sounds dual, nil added; chest clear; abdomen soft. Oriented to person and place, not time. Amongst it all, her main response increasingly becomes: “I don’t know,” retorted reflexively, indifferently. You sense she’s being defensive. But you don’t know, because she says it like she means nothing by it.
You hold off on examining further, and while she’s stable, albeit restless and erratic, you change tack, almost without thinking.
“Do you want a cup of tea?” you say.
“Yes,” she nods, sitting up, clutching her belly. You’re surprised by her reply.
“How do you have it? White?”
“Yes.”
“One sugar?”
“Yes.”
Being polite, you turn to her friend and ask her too.
“Tea?”
“Yes, thank you,” she blurts out. “Milk and one sugar.”
The nurse returns, looking busy as always. You ask her to take some obs and, as you both walk away from the bedside, you tell her you might try the “talking cure” on this one. You go to the staff room and make two cups of tea, white, one sugar each. You return with them, and they gladly accept. You order some labs, including lipase and beta-hCG, even a venous blood gas.
They take a sip or two, and you find a theatre stool just beside you, and sit.
Your demeanour is a mismatch to the patient’s presentation. You learn she’s used drugs in the past, over a year since she last used ICE, and then only for a month. She denies using today. She injected it but denies other IVDU. She smokes cigarettes, doesn’t drink, doesn’t smoke cannabis, is on no regular medications, no allergies. She remains restless, erratic, complaining of abdominal pain and feeling hot, moaning and groaning, the occasional “Oh my God” as she writhes, sits up, sits down, lies down, sits up again. Then she lies down and seems to dose off, or zone out. You question her friend, who is fixated on the idea her friend needs a psychiatrist. You ask whether she may have used anything today. She denies it.
Anxious now, you check her over once more, starting with her pupils: this time a 4, sluggish, though equal. She’s clammy. Your initial thought, on arrival, that this might be a panic attack, comes abruptly to an end. You worry. Is she about to fit? Cerebral irritation? Septic, and if so from where? Too many thoughts flood your head. You pace, and regather.
You read the nurse’s obs off the paper towelling they’ve been written on, sitting on the parcel shelf below the cardiac monitor:
- BP (sitting): 133/84
- Pulse: 125
- Temperature: 36.2
- Respiratory rate: 20
- O2 saturation: 98%
(Always confirm the oxygen saturation in any patient who is restless or erratic.)
And just then, another of her friends arrives, and you learn that earlier in the day, at about quarter past ten, she smoked some ICE. Nil other drug use, apparently.
You put in an IV and start 1L normal saline at a brisk rate, drawing bloods before starting the fluids. Remembering to always check a BSL in someone acting erratically, you do a bedside BSL: 6. You recheck her pupils, this time with a different light source: 3-4mm, reactive. You think you might send this woman to the ED after all. “Have you called an ambulance?” you ask the nurse.
You chart 2.5mg risperidone PO, and go to your room. You see some general practice patients while the fluid runs in, deciding whether to refer or keep her. Perhaps five minutes later, you hear a kerfuffle from the acute side. She has pulled her cannula out and walked out.
That’s where the story stops, because that’s where it actually stopped. No ambulance caught her at the door, no callback, no follow-up letter from a hospital she never reached. She was an adult, alert enough to pull her own line and walk, and there’s a real limit to what you can do once someone chooses to leave. It’s an uncomfortable ending to write, because case logs like to resolve, and this one doesn’t. But that’s honest, and it’s worth sitting with rather than tidying away: sometimes the last thing you know about a patient is that they left, and you never find out what happened next.
The nurse’s instinct that something serious was unfolding was right from the first phone call, even though the specific shape of it, psychiatric crisis versus drug withdrawal versus acute intoxication, only became clear once the second friend arrived with the missing history. (Lesson 4: trust that a nurse who says “this one needs to go to hospital” is telling you something real, even if the reason turns out to be different from the one first offered.)
Sympathomimetic toxidrome
This presents similarly to an anticholinergic syndrome, except for differences in skin and bowel findings.
Blind as a bat, mad as a hatter, red as a beet, hot as a hare, dry as a bone; the bowel and the bladder lose their tone, and the heart runs alone.
- Diaphoresis
- Mydriasis
- Tachycardia
- Hypertension
- Hyperthermia
- Seizures
- Increased peristalsis
Depending on the timing of presentation, you may variably see a patient in the midst of a toxidrome or in withdrawal from the drug, which, generally, presents with more or less the opposite picture.
What the tea trick actually was
It wasn’t a trick, not really. It was buying time and lowering the temperature of the room, hers and mine, while I worked out what I was looking at. A wired, defensive, possibly frightened young woman, fielding a friend who’d already decided what was wrong with her, doesn’t need more questions fired at her. She needs something ordinary. A cup of tea is ordinary in a way that a differential diagnosis isn’t, and sometimes ordinary is what gets you back in the room with someone.
On the risperidone
Looking back at this with fresh eyes, I’d reconsider that choice.
The evidence on managing acute agitation from stimulant intoxication points fairly consistently toward benzodiazepines as first-line, particularly in exactly the situation I was in: an undifferentiated presentation, where stimulant intoxication is suspected but not yet confirmed [1]. One consensus guideline is explicit that benzodiazepines are the preferred choice when there is no clear diagnosis and when stimulant intoxication is a possibility [2]. Antipsychotics, meanwhile, lower the seizure threshold, which matters given I was actively worried about her fitting [1].
There’s genuine nuance here, some ED-based evidence shows antipsychotics can be as effective as benzodiazepines for sedation in monitored settings [3], and risperidone specifically has support as a first-line option where agitation is clearly drug-related [2]. But I wasn’t in a monitored ED. I was in a general practice, with an uncertain diagnosis, a patient whose pupils and vitals were actively evolving in front of me, and no capacity to titrate or closely monitor a parenteral agent. In that context, a benzodiazepine, faster-acting, easier to titrate, and without the seizure-threshold concern, would likely have been the safer first choice.
She’d already left by the time it would have mattered. But it’s worth naming, since the next undifferentiated, agitated patient in a room like that one deserves the better-supported first move.
References
- Wodarz N, Krampe-Scheidler A, Christ M, et al. Evidence-Based Guidelines for the Pharmacological Management of Acute Methamphetamine-Related Disorders and Toxicity. Pharmacopsychiatry. 2017;50(3):87-95.
- Clinically Significant Acute Agitation: Consensus Update. US Pharmacist. Available at uspharmacist.com.
- Singh AP, Mohan NTM. Second-Generation Parenteral Antipsychotic (Olanzapine) as a First-Line Treatment for Acute Undifferentiated Agitation in the Emergency Department in Comparison With Haloperidol. Cureus. 2023;15(6):e40226.
Get new case reviews by email. No spam, just the occasional post when there’s a case worth sharing.