The short version

  • Breathing exercises are powerful enough to produce symptoms, which means they are powerful enough to be contraindicated in some situations.
  • Hyperventilation-based practices can trigger panic, fainting and, in water, drowning.
  • Pregnancy, epilepsy, cardiac arrhythmia and uncontrolled hypertension are reasons to get advice before starting.
  • If a practice consistently makes you feel worse, that is data, not a lack of discipline.

Breathwork is usually presented as the safest possible intervention: free, portable, drug-free, available to anyone. Most of that is true. It is also true that breathing is a direct input to blood pH, cerebral blood flow and autonomic tone, which is precisely why changing it can produce strong effects — and why those effects are not always desirable.

The purpose of this essay is not to scare anyone away from breathing practice. It is to name the situations in which the standard advice is wrong, because the standard advice is written for a healthy adult sitting comfortably in a chair.

Hyperventilation and panic

Rapid or deep breathing lowers arterial CO2, which constricts cerebral blood vessels and raises blood pH. The result is light-headedness, tingling in the extremities, chest tightness, and a sense of unreality. These are also the core symptoms of a panic attack. A practice that deliberately produces them, in a person prone to panic, is not therapeutic; it is a rehearsal.

Many traditional breathwork styles include hyperventilation components — rapid breathing rounds, breath of fire, holotropic-style sessions. For some people these produce catharsis. For a person with panic disorder, they can trigger a full attack, and the memory of that attack makes the next practice harder.

The paradox of relaxation

Telling someone with panic disorder to "take a deep breath" often makes things worse, because deep breathing lowers CO2 further in someone who is already over-ventilating. The better instruction is usually to slow the exhale, not to deepen the inhale.

If you have panic disorder, the useful starting point is a practice that reduces ventilation rather than increasing it: quiet nasal breathing, a gentle exhale, no breath holds until you are stable. And it belongs alongside treatment, not instead of it.

Blood pressure and fainting

Breath holds and forceful exhalations against a closed airway — the Valsalva manoeuvre — raise intrathoracic pressure and can cause sudden drops in blood pressure on release. In most healthy people this produces a brief head rush. In people with a history of syncope, orthostatic intolerance, or cardiac arrhythmia, it can produce a faint.

This is the mechanism behind shallow-water blackout. A swimmer hyperventilates to lower CO2, dives, and does not feel the urge to breathe until oxygen has already fallen to the point of loss of consciousness. The CO2 signal that would have forced them to surface has been removed. It is a preventable death, and it happens to competent swimmers.

Never practise breath holds in water. Never practise them while driving. Never practise them standing if you have fainted before.

Pregnancy and breath holds

Pregnancy increases resting ventilation and reduces the CO2 buffer, which makes breath holds both less comfortable and less appropriate. Sustained maternal hypoxia is a plausible risk to the fetus, and there is no evidence supporting breath-hold training in pregnancy.

Slow breathing without holds is generally considered safe and is sometimes used for labour preparation, but the appropriate guidance comes from an obstetric clinician who knows the pregnancy. This is not a decision to make from a video.

Conditions that need a clinician in the room

  • Epilepsy. Changes in CO2 and cerebral blood flow can lower seizure threshold. Voluntary hyperventilation is used in clinical settings to provoke seizure activity for diagnostic purposes; that should tell you something.
  • Cardiac arrhythmia or recent cardiac events. Autonomic manoeuvres can provoke arrhythmia in susceptible people.
  • Uncontrolled hypertension. Breath holds raise blood pressure transiently; get advice before adding them.
  • Severe asthma or COPD. Breathing exercises can help, but only within a management plan. A practice that induces breathlessness can precipitate an exacerbation.
  • Recent surgery, pneumothorax, or retinal conditions. Forceful breath holds increase intrathoracic and intraocular pressure.
  • Children. Breath-hold practices should not be taught to children without clinical guidance; the risk of fainting and the difficulty of self-monitoring are both higher.

A safer way to start

For most people, the risk-free version of breathwork is also the least dramatic. Sit comfortably. Breathe through your nose at a rate that requires no effort. Let the exhale be slightly longer than the inhale. Do not hold your breath. Stop if you feel light-headed.

Five minutes is enough. If nothing happens, that is a normal outcome: the practice is a small input, not an event. If you feel worse — dizzy, panicky, breathless, or emotionally raw — stop and take it as information about the practice rather than a personal failing.

And if you find yourself needing progressively more extreme breathing to feel calm, that is worth discussing with someone. Breathing practice should make the ordinary breath feel easier over time. If it is making the ordinary breath feel insufficient, the practice has become part of the problem.

References

  1. Meuret AE, Ritz T. Hyperventilation in panic disorder and asthma: empirical evidence and clinical strategies. Int J Psychophysiol. 2010;78(1):68–79.
  2. Craig AB Jr. Underwater swimming and loss of consciousness. JAMA. 1961;176:255–258.
  3. Lindholm P, Lundgren CE. The physiology and pathophysiology of human breath-hold diving. J Appl Physiol. 2009;106(1):284–292.
  4. Nava S, et al. Respiratory physiology in pregnancy. Clin Chest Med. 2011;32(1):1–13.
  5. Esquivel D, et al. Voluntary hyperventilation in epilepsy: a review. Epilepsy Behav. 2019.