So Sakamoto is an emergency physician in Japan focused on high-quality emergency department care, transitions of care, bedside decision-making, and education at the emergency–primary care interface in ageing populations.
A young woman looks ‘…classic for hyperventilation.‘ Her respiratory rate is high, her hands tingle, her fingers cramp, and she tells you she ‘…can’t get enough air.‘ In that moment, how often do we reach for the same script?
“Slow your breathing.”
“Breathe in for one second, and out for two.”
We now broadly accept that paper-bag rebreathing is unsafe and out of date. But once we have decided what not to do, what should we do next? The uncomfortable truth is that the more we try to teach breathing, the more the patient often inhales. What they need is not breathing coaching. What they need is conversation.
…you do need a brief, reliable safety screen for high-stakes conditions such as pulmonary embolism, acute coronary syndrome, and pneumothorax.
Primary hyperventilation is common in younger people, often recurring, and frequently associated with anxiety or panic symptoms. Most cases seen in emergency and urgent care settings are discharged home after supportive management. One study of emergency department presentations also fits a familiar clinical pattern: first presentations in older adults are uncommon.1
None of this gives us permission to label and relax too quickly. Hyperventilation is both a diagnosis and a pattern that can be secondary to something else. So the first move is not “calm them down”, but “don’t miss a dangerous cause of breathlessness”. You do not need an encyclopaedia of differentials, but you do need a brief, reliable safety screen for high-stakes conditions such as pulmonary embolism, acute coronary syndrome, and pneumothorax. Pulmonary embolism, in particular, is a master of disguise: the “textbook” triad is often absent, and diagnostic delay remains a recognised problem.2,3
A small set of questions and observations goes a long way: Was the onset sudden? Is there chest pain? Syncope or near-syncope? Hypoxia? Unilateral leg swelling? Combine this with vital signs, focused examination, and—when indicated—an ECG and chest radiograph to make sure you are not dealing with respiratory failure masquerading as “panic”. One more safety note: brief post-hyperventilation apnoea can occur. If apnoea is prolonged, oxygen saturation falls, or consciousness deteriorates, do not force the story into “simple hyperventilation”; reassess and prioritise airway and breathing support.
Once you have done that short safety work, the clinical objective changes. Now you are not ‘diagnosing;’ you are undoing a physiology loop. Here is the counterintuitive part: generic breathing instructions often fail.
From the clinician’s side, “Slow down” sounds reassuring. From the patient’s side, it can land as blame, or as an impossible task: “If I could do that, I would.” Worse, it keeps attention pinned to breathing. That is exactly where the panic circuit wants attention to stay. As soon as the patient starts monitoring their own breathing, they tend to inhale more deeply and more frequently. Their symptoms then intensify, and the cycle tightens.
It helps to reframe the problem. Hyperventilation is not mainly a rate problem; it is an over-inhalation problem. The patient keeps taking big breaths. Carbon dioxide is washed out. Cerebral blood flow drops. Paraesthesia and carpopedal spasm appear. The body interprets those sensations as further threat, and the next breath becomes even larger.
So the core intervention is not ‘teach breathing.’ It is ‘prevent unnecessary inhalation.’
Conversation does that.
This is more than a psychological trick. Conversation is a physiological intervention. Speaking forces structured exhalation. Answering questions breaks the pattern of repeated deep inhalations. The patient does not have to control their breathing; they simply have to keep responding. If they talk, they must breathe out.
Practical implementation can be disarmingly simple. Sit the patient down (and reduce the risk of a fall). Create a little privacy if you can (move them away from the queue; minimise the ‘audience effect’). Then keep the conversation going with short, answerable questions.
“Tell me your full name.”
“What symptom came first?”
“When did it start?”
“Any chest pain?”
“What did you last eat?”
If they can only manage one or two words, start with yes/no questions and build up. The content is secondary; the speaking is the treatment.
Mirroring is allowed. You can echo a few of their words back to them and invite the next sentence. The aim is not to deliver a lecture on carbon dioxide. The aim is to provide a calm cadence that the patient can synchronise with—without either of you talking about breathing.
Your own delivery matters. Lower your voice. Slow your pace. Keep your wording simple. Hyperventilation is contagious in another sense too: the clinician’s urgency can become the patient’s urgency. Your calm is an intervention.
Paper-bag rebreathing deserves only a brief mention, because it is not the main story. It is tempting—physiologically plausible and theatrically decisive—but it is unsafe when the diagnosis is wrong. In a patient with unrecognised hypoxia, asthma, pneumothorax, pulmonary embolism, or myocardial ischaemia, rebreathing can make you catastrophically wrong. The safer alternative is already in your hands: conversation that restores carbon dioxide without depriving oxygen.
There is one more tool that fits naturally with conversation: narrate the time course. Many patients are terrified not only by symptoms, but by uncertainty. So give a short, concrete explanation and a horizon, “Your oxygen level is good. Right now your breathing is washing out carbon dioxide, which causes tingling and dizziness. This feels awful but it is reversible. It usually settles over minutes. I’ll stay with you while it does.”
Predictability is not just reassurance; it changes physiology.
Do the brief safety screen. Then stop trying to teach respiration to someone in panic. Talk to them. Keep them answering.
This also explains a common observation in emergency care: many patients who were ‘too breathless to speak‘ at the call-out are visibly better on arrival. In the ambulance, treatment has often already started—not because of a drug, but because paramedics do ABC, seat the patient safely, monitor, and talk. In good crews, questions are not only assessment; they are intervention.
In Japan, the wider context matters. Ambulance call-outs have risen, and response times from call to arrival have lengthened in many areas. Hyperventilation is a condition that is often manageable on scene when the patient is young, has a clear trigger (a needle procedure, acute anxiety, an argument with a partner, a stressful event at work), has had similar episodes before, and has stable observations without red flags. In those typical situations, early conversational control can resolve symptoms rapidly and may avoid an unnecessary ambulance call. That is not about discouraging access to emergency care; it is about protecting it—keeping scarce response capacity for truly time-critical illness.
The reverse is equally important. Hyperventilation-like symptoms in older adults, first presentations without a clear trigger, hypoxia, persistent chest pain, altered consciousness, focal neurological signs, fever, or significant systemic illness should not be “talked away” into non-conveyance. You can still use conversation to reduce distress and over-inhalation, but you should not rush past observation, testing, or referral when the story is atypical. Hyperventilation can settle while the underlying problem remains.
So the message is not “don’t investigate”. The message is “don’t coach breathing”.
Do the brief safety screen. Then stop trying to teach respiration to someone in panic. Talk to them. Keep them answering. Let speech do what commands cannot: create exhalation, limit inhalation, and restore normal ventilation without the patient having to notice they are doing it.
References
- Pfortmueller CA, Pauchard-Neuwerth SE, Leichtle AB, et al. Primary hyperventilation in the emergency department: a first overview. PLoS One. 2015;10(6):e0129562.
- Meyer G, Roy PM, Gilberg S, et al. BMJ 2010; 340 doi: https://doi.org/10.1136/bmj.c1421 (Published 13 April 2010)
- van Maanen R, Trinks-Roerdink EM, Rutten FH, et al. Diagnostic delay in pulmonary embolism: a systematic review and meta-analysis. Eur J Gen Pract. 2022;28(1):165-172.
Featured Photo by Kelly Sikkema on Unsplash
