Ancient practice. Modern measurement.
Measuring the power of the breath with the tools of today.
What changes when you breathe slowly
Some changes happen the moment we start breathing slowly. Other changes build over weeks of regular practice. Both are real, and both have been measured in controlled studies.
Blood pressure drops
When we breathe slowly, our blood pressure goes down. This has been measured across many studies. In 2026, researchers added up the results from 13 separate trials covering 1,097 people with high blood pressure.1
Here is what they found. People who practiced slow breathing had their top blood pressure number (the systolic) drop by 7.68 mmHg on average. The bottom number (the diastolic) dropped by 4.02 mmHg. Their heart rate also slowed slightly.1
This sounds small. It isn't. Other research has shown that even a 3.6 mmHg drop in blood pressure across a population is linked to 28% fewer strokes and 25% fewer deaths from heart disease.1 The breathing effect of 7.68 mmHg is more than double that threshold.
Why it works
Our body has automatic systems that adjust blood pressure all the time. One of them is called the baroreflex. Pressure sensors in our arteries notice when blood pressure changes, and they send signals to our heart and blood vessels to compensate. In people with high blood pressure, these sensors stop working as well.1
Slow breathing at around 6 breaths per minute brings the sensors back online. The deeper, slower breaths stretch our lungs and chest more, which sends new signals through the nervous system that tell our blood vessels to relax. The result is lower blood pressure.1
Heart rate variability roughly doubles
The heart does not beat in a perfectly steady rhythm. Even at rest, the gaps between heartbeats vary slightly. One beat might be 0.85 seconds after the last. The next might be 0.91. The next might be 0.82.
This variation is called heart rate variability (HRV). It sounds like it would be a bad sign; surely a steady heart rate is healthier? Actually the opposite is true. A heart that varies its rhythm is responding moment-by-moment to what the body needs. A heart that beats like a metronome is a heart whose nervous system has stopped doing its job.2
The standard way to measure HRV is a number called the root mean square of successive differences (RMSSD). Higher RMSSD means more variability, which means more activity in the calming side of the nervous system (the parasympathetic system).2
What slow breathing does to RMSSD
In a 2022 study, 112 healthy young adults sat in a lab and did 5 minutes of slow breathing at 6 breaths per minute. Researchers measured RMSSD before, during, and after.2
Before the breathing started, average RMSSD was about 47 milliseconds. During the 5 minutes of slow breathing, it jumped to about 91 milliseconds. That is roughly double. The calming side of the nervous system became much more active within seconds of starting.2
When the breathing stopped, RMSSD dropped back to where it started, also within seconds. The researchers described it like a switch: the practice turns on the calm state, and stopping the practice turns it off.2
Why this matters
Most of the time, our nervous system runs in the background without us noticing. We cannot directly tell our blood pressure to lower itself or our heart to slow down. Breathing is the one autonomic function we can take over and hand back.2 When we take it over and slow it down, the rest of the system follows.
Mood and anxiety improve over weeks
The blood pressure and heart rate effects above happen during practice and stop when we stop. But what about effects that build over time, the kind that change how we feel even when we are not practicing?
A 2020 study at a university in Brazil looked at this directly. 30 healthy young adults were randomly split into two groups. One group practiced Bhastrika pranayama 5 days a week, 30 minutes a session, for 4 weeks. The other group spent the same amount of time doing puzzles, crosswords, and card games with an instructor present. At the start and at the end of the 4 weeks, both groups filled out anxiety and mood questionnaires.3
What changed
The pranayama group's state anxiety dropped by about 16%. Their negative feelings (fear, nervousness, agitation) dropped by about 20%. Their positive feelings stayed steady, while the control group's positive feelings actually dropped over the same 4 weeks.3
The effects were what researchers call "large," a technical term meaning the changes were big enough to be clearly meaningful, not just a statistical blip.3
Brain scans show why
Researchers also put participants in a magnetic resonance imaging (MRI) scanner before and after the 4 weeks. They looked at parts of the brain involved in emotion: the amygdala, the insula, the anterior cingulate, the prefrontal cortex. After 4 weeks of pranayama, the way these regions communicated with each other had changed. Specifically, the connection between the insula (which monitors body sensations) and parts of the prefrontal cortex (which regulates emotion) became less tightly coupled. The participants whose brains showed the biggest loosening also reported the biggest drops in anxiety.3
A note on this study
The trial was small (30 people). The brain analyses were exploratory, meaning they should be treated as a strong hint about where the truth lies rather than a definitive proof. The Bhastrika practice used in this study combines several techniques (fast breathing, breath retention, slow alternate-nostril breathing) rather than just one. So this trial tells us that 4 weeks of regular pranayama practice produces real and measurable changes in anxiety and brain activity, but it cannot tell us exactly which component of the practice did the work.3
Why slow breathing has that effect
The body has built-in machinery that responds to how we breathe. Slow breathing reaches the nervous system through specific pathways, and the precise breathing rate matters more than people realize.
How slow breathing reaches the nervous system
When we breathe slowly, our lungs inflate more than usual. This stretches the lung walls, the airways, and the major blood vessels near the heart. These tissues are lined with sensors that fire when stretched.2
The signal they send travels up the vagus nerve, which is the body's main connection between the brainstem and the internal organs: heart, lungs, gut. The vagus is also the main line into the calming side of the nervous system. When the stretch sensors fire, they're talking directly to it.2
Where the signal goes
From the brainstem, the signal branches out to several places at once:4
The second channel
There is also a second channel. Chemical sensors in the neck, near the heart, and in the brainstem itself monitor oxygen and carbon dioxide levels in the blood. When breathing changes, these sensors respond too.4
Why this is unusual
Almost every other automatic process in the body runs without our input. We cannot decide to lower our blood pressure, slow our heart, or quiet our emotional response to a stressful thought. Breath is the exception. And because breath plugs into the same wiring those systems use, when we slow it down, we are pulling on the levers that reach all of them at once.4
Why 6 breaths per minute
When we breathe at a normal rate (12 to 20 breaths per minute), our heart rate goes up slightly during each inhale and down slightly during each exhale. This is called respiratory sinus arrhythmia, and it is a small, barely-noticeable effect.
When we slow our breathing down to around 6 breaths per minute, something different happens. The natural oscillation in heart rate lines up with the natural oscillation in blood pressure. The two waves stack and amplify each other. This synchronization is called resonance, and 6 cycles per minute (cpm) is roughly where it happens for most people.24
What resonance actually does
At resonance, the heart rate oscillations become much bigger. Researchers have measured them as roughly doubled compared to normal breathing.4 This is what heart rate variability (HRV) measurements are picking up, and bigger is better here. A heart that varies its rhythm moment by moment is a healthy, responsive nervous system. A heart that beats like a metronome is one whose regulation has gone flat.2
The bigger oscillations also stimulate the baroreflex (the body's blood pressure regulator) more strongly than normal breathing does. This is part of why slow breathing produces the blood pressure effects covered in the previous section of effects.4
How precise does the rate need to be
A 2017 trial tested whether the exact rate matters. The researchers measured each person's individual resonance frequency, then compared breathing at exact resonance to breathing one cycle per minute above resonance, and to a quiet sitting control. The exact-resonance group had measurably larger HRV changes and bigger blood pressure drops than the off-by-one group. Both breathing groups did better than control.5
The practical implication: breathing at exactly 6 cpm is close to most people's resonance frequency, and produces most of the benefit. Practicing at someone's exact individual resonance produces somewhat larger effects, but the gap between exactly right and close enough is moderate, not huge. For a daily practice without specialized measurement, 6 cpm is the right target.
The baroreflex
The baroreflex is the body's automatic blood pressure regulator. Inside the walls of certain arteries (the aorta near the heart, the carotids in the neck), there are pressure sensors called baroreceptors. They constantly monitor blood pressure and trigger compensating adjustments.4
When blood pressure rises, the baroreflex slows the heart and dilates blood vessels to bring pressure back down. When blood pressure drops, the baroreflex speeds the heart and constricts vessels to bring it back up. Healthy baroreflex function keeps blood pressure stable through changes in posture, activity, and stress.4
What goes wrong in high blood pressure
In people with high blood pressure, the baroreflex stops working as well. The sensors become less sensitive, so the corrections happen too slowly or with too small a magnitude. Blood pressure stays elevated because the regulator is not catching the problem.41
How slow breathing fixes it
Slow breathing at around 6 cycles per minute (cpm) stimulates the baroreceptors more strongly than normal breathing. The bigger blood pressure oscillations during resonance breathing repeatedly stretch the sensors, and over weeks of regular practice, this rebuilds their sensitivity.14
This is the mechanism behind the blood pressure findings. The 7.68 mmHg systolic reduction across the 13 trials in the Cheng meta-analysis is consistent with baroreflex sensitivity recovering over the course of an intervention.14
What we know and don't yet know
Breath retention (Kumbhak) sits at the heart of classical pranayama. The mechanisms are biologically plausible, and the limited modern trial evidence is consistent with the traditional framework. The research is still catching up to what practitioners have known for centuries.
What happens when we hold the breath
When we hold the breath after inhaling, two things happen quickly. Oxygen levels in the blood start to drop. Carbon dioxide (CO2) levels start to rise. This is the basic physics of retention; the body keeps consuming oxygen and producing CO2, but no new gas is coming in.4
This brief state, slightly low oxygen and slightly high CO2, activates a different set of sensors than slow breathing alone does. These are called chemoreceptors. They live in the neck, near the heart, and inside the brainstem. When CO2 rises or oxygen drops, the chemoreceptors fire and trigger a chain of responses.4
What the body does in response
Why traditional pranayama emphasizes retention
Classical pranayama treats Kumbhak as the integrating practice. The inhale brings energy in. The exhale releases what is not needed. The hold is where the work of those two phases consolidates. Modern physiology gives a reason this might be true: the chemoreceptor pathway activated during retention is different from the pathway activated during slow breathing alone, and the two together may produce effects that neither produces alone.4
The proposed mechanisms are plausible and consistent with known physiology. The direct trial evidence for classical Kumbhak at full expression is still being built.
What controlled trials show
In 2024, researchers in India ran a small pilot trial of one specific Kumbhak technique. 24 people with high blood pressure were randomly split into two groups. One group practiced Sheetali pranayama with Kumbhak: inhaling through a curled tongue, holding the breath for 10 seconds, exhaling slowly through the nose. They did this for 10 minutes. The other group did breath awareness, sitting and watching the breath without any active practice.6
What changed
After the 10-minute session, the Sheetali Kumbhak group had measurable drops in both systolic and diastolic blood pressure compared to the breath awareness group. The drops were statistically significant despite the small sample.6
Cerebrovascular blood flow (how fast blood was moving through the major artery supplying the brain) did not change significantly. The researchers noted that prior work has shown cerebral blood flow changes only after retentions exceeding 30 seconds. A 10-second hold is too short to produce the cerebral effects that longer retentions can.6
What this trial does and does not show
The trial shows that even a short Kumbhak practice (10 seconds, for 10 minutes total) produces a measurable blood pressure response in hypertensive patients within a single session.
It does not show that retention is the active ingredient (the slow exhale could also be doing work). It does not show whether the effect persists past the session. It does not test sustained Kumbhak at higher levels (30 seconds, a minute, or longer, as practiced in classical pranayama curricula).6
The trial is small, has imbalanced baselines between groups, and is the first of its kind. It justifies a larger follow-up study. It is suggestive evidence, not definitive evidence.
Beyond the studies
The research base on classical Kumbhak is genuinely small. The one direct trial used a 10-second hold practiced for 10 minutes. Pranayama at higher levels uses holds of 30 seconds, a minute, or longer, practiced consistently over months and years.
Modern science is catching up. The mechanisms point in the right direction, and the limited trial evidence is consistent with the traditional framework. The absence of large controlled trials on classical Kumbhak doesn't mean the practice is ineffective. It means the science is still measuring what practitioners have known for centuries.
The 12-level Kumbhak curriculum in Lunura is built on this traditional framework. Your own practice over months and years is its own form of evidence.
Terminology
Heart rate dynamics
The science page measures the same underlying thing, your heart's rhythm responding to breath, but reports it different ways. Here's how they connect.
The phenomenon: heart rate oscillations
Your heart rate isn't steady. It ticks up slightly when you inhale and down slightly when you exhale, with each breath cycle. Plot heart rate over a minute and you see a wave: rising on the in-breath, falling on the out-breath, repeating.
The amplitude of that wave is how far up and down the heart rate swings within each cycle. At normal breathing rate (12 to 20 per minute), this swing is small, maybe a few beats per minute. At resonance breathing (around 6 per minute), the swing roughly doubles.
This is the actual physical thing happening in your body when you breathe slowly.
The measurement: RMSSD
RMSSD (root mean square of successive differences) is the way to measure the oscillations.
The formula: take every pair of consecutive heartbeats, find the difference in their intervals, square it, average across all pairs, take the square root. The formula was specifically designed to capture short-term, breath-driven variation in heart rate.
So when one study reports heart rate oscillations doubled and another reports RMSSD went from 47 to 91 ms, they're describing the same event.
The umbrella term: HRV
HRV (heart rate variability) is the broader concept that any variation in your heart's rhythm matters. RMSSD is one specific way to measure HRV. Other formulas exist (SDNN, pNN50, frequency-domain measures), each sensitive to different aspects of variability.
When the page says HRV is high or HRV doubles, it's talking about the general concept. When it gives specific milliseconds, it's reporting RMSSD numbers.
The reason higher HRV is healthier: a heart stuck at a metronome pace means a nervous system that has stopped adjusting in real time. A heart that varies means a nervous system responding moment by moment to what the body needs.
Baroreflex and baroreceptors
The baroreflex is your body's automatic blood pressure regulator. Baroreceptors are the pressure sensors that feed it.
Baroreceptors
Inside the walls of certain arteries (the aorta near the heart, the carotids in the neck) live specialized sensors called baroreceptors. They constantly monitor blood pressure by detecting how much the artery wall is stretching with each pulse. When blood pressure rises, the artery stretches more, and the baroreceptors fire faster signals to the brainstem.
Baroreflex
The baroreflex is the loop those signals trigger. When the brainstem receives high-pressure signals, it sends commands back through the nervous system to slow the heart and dilate blood vessels, which brings pressure back down. When it receives low-pressure signals, it speeds the heart and constricts vessels, which brings pressure back up. The whole loop happens automatically and within a few heartbeats.
In high blood pressure, the baroreceptors become less sensitive. The corrections happen too slowly or too weakly, and blood pressure stays elevated because the regulator is no longer catching the problem.
How slow breathing connects
Resonance breathing doubles the size of the heart rate oscillations within each breath cycle. Those bigger oscillations carry through to bigger blood pressure oscillations, which means the baroreceptors get stretched harder with each cycle. Over weeks of repeated practice, this strong rhythmic stimulation rebuilds baroreceptor sensitivity. That's the mechanism behind the 7.68 mmHg blood pressure reduction in the meta-analysis: not a drug effect, but a regulatory system being retrained back into proper function.
References
- Cheng et al. (2026). Voluntary Slow Breathing Exercise on Cardiovascular Parameters in Patients With Hypertension: A Systematic Review and Meta-Analysis. Clinical Cardiology. Read paper
- Laborde et al. (2022). Psychophysiological effects of slow-paced breathing at six cycles per minute with or without heart rate variability biofeedback. Psychophysiology, 59(1), e13952. Read paper
- Novaes et al. (2020). Effects of Yoga Respiratory Practice (Bhastrika pranayama) on Anxiety, Affect, and Brain Functional Connectivity and Activity: A Randomized Controlled Trial. Frontiers in Psychiatry. Read paper
- Mondal (2024). Proposed physiological mechanisms of pranayama: A discussion. Read paper
- Steffen et al. (2017). The Impact of Resonance Frequency Breathing on Measures of Heart Rate Variability, Blood Pressure, and Mood. Frontiers in Public Health. Read paper
- Radhiga et al. (2024). Impact of Sheetali Pranayama with Kumbhaka on Blood Pressure and Cerebrovascular Hemodynamics in Patients with Hypertension: A Pilot Randomized Controlled Trial. Read paper