Fast Breathing: What the Evidence Shows and the Real Risks

What is fast breathing?

"Fast breathing" here means breathing faster and deeper than the body needs, on purpose. Researchers group these practices as high ventilation breathwork; they are linked with marked changes in subjective experience and in nervous-system function [S5]. The practices differ, and results from one do not transfer to another:

• Wim Hof breathing: rounds of deep, fast breaths followed by a breath-hold after breathing out [S6]. The full Wim Hof Method adds cold exposure and a mindset or meditation component, and most studies test the combination [S1] [S4].
• Cyclic hyperventilation with retention: a similar pattern used in laboratory and remote trials, sometimes without the cold [S3].
• Holotropic breathwork: a prolonged session of voluntary hyperventilation, usually with music, in a group or therapy setting [S13].
• Tummo: a Tibetan practice that combines a forceful breathing technique with meditative visualisation [S14].
• Bhastrika and kapalabhati: fast pranayama techniques from yoga, studied separately; they are covered in the ONDA articles on fast versus slow pranayama and bhastrika.

What happens in the body?

Carbon dioxide falls. Breathing more than the body needs washes carbon dioxide out of the blood (hypocapnia). The blood becomes more alkaline: in runners doing Wim Hof breathing, exhaled carbon dioxide dropped to levels that indicate respiratory alkalosis [S6].

Brain blood vessels narrow. Low carbon dioxide increases the resistance of brain blood vessels and reduces brain blood flow [S10]. This is the likely source of light-headedness and the "high" some people describe.

Tingling and cramps. Tingling in the fingers and lips, and sometimes cramping of the hands, are common. They are usually attributed to a fall in ionized calcium as the blood turns alkaline, but one small study found tingling in every volunteer without a significant change in ionized calcium [S11]. The exact mechanism is debated.

Adrenaline rises. In trained volunteers, practising the techniques produced intermittent alkalosis and low oxygen, with a large rise in plasma adrenaline [S1]. This is an activating, not a calming, response.

Oxygen does not go up in any useful way; during the hold it falls. At the end of Wim Hof breath-holds, finger oxygen saturation fell to about 60% on average [S6].

Why the breath-hold feels easier. The urge to breathe is driven mainly by rising carbon dioxide. Hyperventilation lowers carbon dioxide first, so the breaking point comes later and the hold feels longer and easier; meanwhile oxygen keeps falling. That is why hyperventilation is considered a major risk factor for blackout during breath-holding [S7]. In serial breath-holds after hyperventilation, oxygen fell further with each round [S7].

What does the evidence show?

Inflammatory response in the laboratory. Emerging. In the best-known trial, twelve trained and twelve untrained healthy volunteers were given an injected bacterial toxin (endotoxin). The trained group, taught meditation, breathing techniques and cold exposure together, had a faster and higher anti-inflammatory response, lower pro-inflammatory proteins and fewer flu-like symptoms [S1]. Because the training combined all three, the effect cannot be attributed to breathing alone. A later pilot study from the same group separated the parts in 48 healthy young men: cold training alone did not meaningfully change the inflammatory response, the breathing exercise did, and adding cold strengthened the breathing effect [S2]. A systematic review of nine papers covering eight trials concluded that the method may reduce inflammation, with few and mostly small studies [S4]. All of this comes from short-lived laboratory inflammation in healthy volunteers, not from people with inflammatory or autoimmune disease. The cold side of these studies is covered on the cold exposure page.

Mood, anxiety and stress. Emerging and mixed. In a remote randomised trial, daily breathwork improved mood more than mindfulness meditation; the exhale-focused cyclic sighing arm did best, and cyclic hyperventilation with retention was one of the arms compared [S3]. In women with high depressive symptoms, a Wim Hof-based programme with cold showers and a slow-breathing programme with warm showers led to equivalent improvements in depression, anxiety and stress; all participants who dropped out were in the Wim Hof group [S15]. In another small trial, daily Wim Hof practice for about two weeks did not improve heart rate variability, blood pressure or psychological measures compared with controls [S16]. Each of these is a single study.

Sport performance. Emerging; no benefit shown. A single session of Wim Hof breathing before repeated sprints did not improve peak power, average power or fatigue in amateur runners, despite large changes in oxygen and carbon dioxide [S6]. The systematic review found mixed effects on exercise performance [S4].

Body temperature in tummo. Emerging. Expert tummo meditators raised their body temperature, but only during the forceful-breath type of practice and with meditative visualisation; non-meditators using the breathing alone produced only small rises within the normal range [S14]. This is one small study.

Altered states and holotropic breathwork. Unknown. Holotropic breathwork has been proposed as an addition to psychotherapy, but the main paper on it is a hypothesis paper that calls for better-designed research to confirm or refute any benefit [S13]. Controlled trials of its effects and safety are lacking.

Fast versus slow breathing. Fast breathing and slow breathing push the body in opposite directions: fast breathing lowers carbon dioxide and raises adrenaline [S1] [S6], while slow breathing near the resonance range is the pattern linked with higher heart rate variability. See slow breathing and how breathing changes HRV. Direct comparisons are rare. The two trials that compared them head-to-head found no advantage for the fast technique on mood or depression [S3] [S15].

Does hyperventilation fill the body with oxygen?

A common claim is that fast, deep breathing oxygenates the body. The studies show something else: what extra breathing mainly does is remove carbon dioxide, producing alkalosis [S6]. During the breath-hold that follows, oxygen saturation falls well below normal [S6], and after hyperventilation oxygen falls further before the urge to breathe arrives [S7].

Does making the blood alkaline fight disease?

Some teaching says that making the blood alkaline through breathing cures disease. The alkalosis from hyperventilation is real but temporary [S1] [S6], and it narrows brain blood vessels [S10]. No study has shown that it cures any disease.

Is the method proven to treat autoimmune disease?

The endotoxin studies are sometimes presented as proof that Wim Hof breathing treats autoimmune disease. The original authors suggested possible implications for such conditions [S1], but the studies used a short laboratory inflammation model in healthy young volunteers [S1] [S2]. The systematic review calls for studies in people with inflammatory disorders; it does not report proven benefit in them [S4].

Safety

This is the most important part of the page.

• Never in water. Never practise hyperventilation or breath-holds in a pool, the sea, a bath or a cold plunge. Healthy young swimmers have died after hyperventilating and becoming unconscious underwater [S8]. Hyperventilation delays the urge to breathe, so blackout can come without warning [S7]. Deaths in breath-hold diving mostly involve loss of consciousness from low oxygen [S9].
• Never while driving, cycling, standing or doing anything where fainting could hurt you. Practise sitting or lying down, on a safe surface.
• Epilepsy: talk to a neurologist first. Hyperventilation is used during EEG tests precisely because it can provoke seizures; in childhood absence epilepsy it provokes absence seizures with brief loss of consciousness in most patients [S12].
• Talk to a doctor first if you have heart disease, an arrhythmia, high or poorly controlled blood pressure, a history of fainting or stroke, if you are pregnant, or if you have panic attacks or a psychiatric condition. Holotropic breathwork can bring up intense emotional experiences and has not been tested for safety in controlled trials [S13].
• What to do if symptoms come. Tingling and light-headedness are signs of low carbon dioxide. If they appear, stop the fast breathing and breathe slowly and normally until they pass. If hands cramp, stop. If you faint, feel chest pain, palpitations with dizziness, severe breathlessness, or symptoms that do not settle within minutes, seek medical help. Someone who loses consciousness needs emergency care.

What it does not tell you

• Small samples. Most trials have a few dozen participants or fewer [S1] [S2] [S6] [S16].
• Motivated, self-selected participants. Volunteers who sign up to learn a famous method expect it to work; in one trial, participants were even trained by the method's creator [S2].
• No blinding. People know when they hyperventilate, so self-reported mood and symptoms are open to expectation effects.
• Mixed ingredients. Many studies combine breathing with cold exposure and meditation [S1] [S4], so the effect of breathing alone is known only from a pilot study [S2].
• Laboratory models. Endotoxin responses in healthy young men do not show what happens in disease.
• Long-term effects are unknown, both benefits and harms.

In ONDA

ONDA does not measure blood carbon dioxide, oxygen or adrenaline, and does not detect fainting or seizures. ONDA shows your pulse live during a practice, from the iPhone camera or an Apple Watch. A higher pulse during fast breathing fits the adrenaline rise seen in studies [S1]; it is not a sign of benefit. If you want to see how slow breathing affects your heart rhythm, read your readings against your own baseline rather than as single values; nightly breathing rate is covered on the respiratory rate page.