The famous 2014 PNAS study showed that trained practitioners could consciously suppress their immune response to injected endotoxin. I've called it the "4-minute mile" of breathing. Here's an honest look at what the research actually found — and what it still can't tell us.
In 1954, Roger Bannister ran a mile in under 4 minutes. Before that day, many people believed it was physically impossible. After that day, dozens of runners broke the barrier within a year. The barrier wasn't physiological — it was psychological. Bannister proved it was possible.
I think about the Wim Hof endotoxin study the same way.
For centuries, breathing practices have been used to modulate the autonomic nervous system and immune system. But scientifically? The conventional view was that these systems operated completely outside of conscious control. You could not "think" or "breathe" your way out of an immune response. That was simply not how the body worked.
Then came Wim Hof — and the 2014 study published in Proceedings of the National Academy of Sciences (PNAS), one of the most prestigious scientific journals in the world.
It showed, for the first time, that ordinary people — not just Wim — could use breath and mind to voluntarily influence their innate immune response.
That was the 4-minute mile. It proved the impossible was possible. And just like Bannister's run, it opened the door to everything that came after.
Below, I walk through the original study in detail, then cover the follow-up research that has built on it — including what we now know works, what doesn't, and what still needs answering.
An endotoxin is a substance found on the outer membrane of certain bacteria. When injected into a healthy person, it reliably triggers a well-characterized immune response: fever, headache, chills, nausea, muscle aches — essentially, a controlled case of feeling very sick for a few hours. Then it passes.
Scientists have injected this substance into hundreds of people over the years, which means they have a precise picture of what "normal" looks like. That baseline makes it an excellent tool for studying whether something — a drug, a training method, a breathing technique — can change the immune response.
Importantly, this is the same initial test that drugs for autoimmune disease must pass in early clinical trials. That context matters when we discuss what the Wim Hof results actually mean.
This is the study that started everything. Here is an honest breakdown of what they did and what they found.
What they did: Thirty healthy male volunteers were divided into a Wim Hof Method (WHM) group (n=18, with 12 completing the endotoxin portion) and a control group (n=12). The WHM group trained with Wim and a few instructors for four days — including breathing, cold exposure, meditation, yoga poses, and climbing a mountain in shorts. They then practiced independently for 5–9 days before the experiment.
The endotoxin procedure: Starting 30 minutes before injection, the WHM group performed the standard WHM breathing: 30 deep breaths → maximum breath hold after exhalation → inhale and hold for 10 seconds → repeat. They continued this for 1.5 hours. Then, for another 1.5 hours, they performed a modified technique: 30 deep breaths → full inhale → hold for 10 seconds while tightening all muscles. That is a total of 3 hours of WHM breathing. The control group received the endotoxin with no breathing intervention.
What they measured: Inflammatory cytokines (pro- and anti-inflammatory proteins), fever, blood oxygen saturation, adrenaline levels, and subjective flu-like symptoms throughout the experiment.
Key results:
With just 4 days of training, ordinary people could use Wim Hof's method to consciously influence their sympathetic nervous system and innate immune response — reducing inflammation and feeling significantly better during an endotoxin challenge.
The study revealed a plausible mechanism for what happened: adrenaline.
It seems counterintuitive at first. Doesn't stress cause inflammation? Yes — chronic stress does. But short-term, intense sympathetic nervous system (SNS) activation is a different story. Research from the fields of psychoneuroendocrinology and neuroimmunology has shown that acute spikes in adrenaline can actually suppress the inflammatory response.
The WHM breathing — 30 rapid, deep breaths followed by a breath hold — is an intense short-term stressor. The hyperventilation phase floods the body with oxygen and CO₂ fluctuates dramatically. The breath hold then creates intermittent hypoxia (low oxygen). Together, these trigger a massive adrenaline surge.
The adrenaline spike correlated strongly with the reductions in inflammatory cytokines. In other words: the breathing drove adrenaline up, and the adrenaline drove inflammation down.
"The participants' adrenaline rose higher than seen in people bungee jumping for the first time."
— Kox et al., PNAS 2014This also explains one interesting side effect: during the breath holds, participants' blood oxygen saturation dropped to around 50% — well below the 70% threshold where serious problems typically begin. The intermittent hypoxia combined with the adrenaline surge appears to be a critical part of the anti-inflammatory effect.
I want to be honest here, because the breathwork world has a tendency to take this study and run far beyond what it actually shows. There are real limitations worth understanding.
Only 12 WHM participants completed the endotoxin portion of the study. All were healthy young males. This is a small and non-representative sample. We cannot assume the results generalize to women, older adults, or people with chronic health conditions.
Wim himself led the training and was present during the experiment. He is, by any measure, a charismatic and compelling figure. A separate 2016 study found that after training with Wim, participants' "outcome expectancies" rose significantly — meaning they expected the method to work — and that this correlated with their symptom reductions. Correlation is not causation, but it raises a legitimate question about how much of the effect was psychological.
The participants performed 3 continuous hours of WHM breathing during the experiment. This is not a casual practice. Outside a controlled research setting, nobody is doing this for a routine cold or illness. The clinical relevance of a 3-hour breathing protocol requires some imagination.
The endotoxin challenge is an acute inflammatory event — hours, not weeks or years. The results tell us nothing directly about chronic inflammation associated with autoimmune disease, diabetes, or long-term conditions. Different biology, different timescales.
None of these limitations negate the findings. They are genuinely remarkable. But they're important context for understanding what comes next — and why follow-up research mattered so much.
Want a Complete Breathing System?
If you're interested in evidence-based breathing for anxiety, stress, and nervous system health, The Anxious Person's Breath Manual synthesizes 454 studies into one practical guide — covering slow breathing, nasal breathing, and more.
Get the Manual for $27The 2014 study raised as many questions as it answered. Which component was responsible — the breathing, the cold, or Wim's personal presence? How much training is needed? What if you can't stomach the breath holds? A 2022 study addressed all of these questions directly.
What they did: Forty male participants were randomized into four groups: intensive training (4 days) with Wim, intensive training with an independent instructor, short training (2 hours only) with Wim, or short training with an independent instructor. The key measure was adrenaline release during two different breathing exercises — standard WHM (with breath holds and oxygen drops) and a modified technique (with breath holds but no hypoxia).
Key findings:
You don't need Wim Hof personally, you don't need to drop your oxygen to 50%, and you may only need 2 hours of instruction. But timing matters — morning practice appears to be critical for the anti-inflammatory response.
What they did: Forty-eight participants were divided into four groups: Cold Only (4-day cold training, no breathing), Breathing Only (2-hour breath instruction, no cold), Cold + Breathing, and Control. All were then injected with the same endotoxin used in the 2014 study. Adrenaline, cytokines, and flu-like symptoms were recorded throughout.
Key findings:
Breathing is the primary driver of the anti-inflammatory response. Cold alone doesn't reduce acute inflammation, but it adds a meaningful boost when combined with the breathing. If you hate the cold, the breathing alone may still help — but you may be leaving something on the table.
The endotoxin studies cover acute inflammation — a controlled spike and recovery over hours. But what about the chronic, low-grade inflammation that underlies autoimmune disease, metabolic disease, and aging? A 2019 study tackled that question directly.
What they did: Twenty-four patients with axial spondyloarthritis (axSpA) — a form of arthritis involving chronic spinal inflammation — were randomized into early or late intervention groups. The early group trained in the full WHM (breathing, cold, meditation, physical exercises) for 8 weeks with Wim and four instructors, meeting twice a week and practicing daily at home. Inflammation markers, disease activity, and quality of life were measured before and after.
Key findings:
The WHM appears to reduce chronic inflammation and improve quality of life in autoimmune disease — not just acute inflammation in a controlled lab setting. This is important. It suggests the method has real-world relevance beyond the endotoxin experiment.
Four years before the 2014 WHM study, a separate research team looked at the effects of slow, paced breathing (heart rate variability biofeedback) on endotoxin exposure. The two studies were never designed to be compared, but the similarity of their methods makes it about as close an accidental comparison as breathing science is likely to produce.
| Outcome | Wim Hof Method | Slow Breathing |
|---|---|---|
| Pro-inflammatory cytokines | Significantly reduced (51–57%) | No significant effect |
| Anti-inflammatory cytokines | Significantly increased (194%) | No significant effect |
| Fever | Reduced | Not reduced |
| Flu-like symptoms overall | Reduced by ~56% | Partial (headache, light sensitivity only) |
| Heart rate variability / autonomic resilience | Not measured | Improved (less HRV drop) |
| Nausea, chills, muscle aches | Reduced | Not significantly reduced |
The conclusion: for acute inflammation and fighting off an injected endotoxin, the WHM was decidedly more effective. Slow breathing helped some — particularly with autonomic resilience and certain symptoms — but it did not produce the dramatic anti-inflammatory cytokine changes that the WHM did.
This is not a knock on slow breathing. Slow breathing was never designed to acutely suppress an immune challenge. Its value lies elsewhere: daily stress reduction, improved sleep, long-term nervous system regulation, and HRV improvement. The two approaches serve different purposes, and that's worth knowing.
"What might be even more fascinating is to look at the daily practice of slow breathing for increased resiliency, followed by the WHM for the acute challenge. Maybe we'd get the best of both worlds."
— Nick Heath, Ph.D.The Anxious Person’s Breath Manual
Want to understand both slow breathing and breathwork in one place? The Manual covers the full spectrum — from calming slow breathing for anxiety to understanding the science behind more intense techniques.
Get the Manual for $27I'm a type-1 diabetic. I've practiced the WHM intermittently. And I've thought a lot about the gap between the science and the lived experience of using this method with a chronic condition.
The research is genuinely exciting — the immune benefits are real, the quality-of-life improvements in autoimmune patients are real, and the chronic inflammation data from the spondyloarthritis study is meaningful. But there are several things people with diabetes (and others with metabolic or cardiovascular conditions) should understand before diving in:
None of this means the WHM is off-limits for diabetics or people with chronic conditions. It may even be especially beneficial in the long run. But it should be approached with eyes open, with a certified instructor, and ideally with your physician's awareness.
Here is what the body of evidence now shows — clearly and without overstating:
And here is what the research does not yet tell us:
The science is genuinely exciting. The 4-minute mile has been run. We now know that conscious influence over the immune system through breathing is possible. What remains is figuring out who it helps most, when, how, and at what dose.
That research is ongoing. And as a breathing nerd and a diabetic, I'll be reading every study as it comes out.
The Anxious Person’s Breath Manual
Want a complete, research-based breathing system? The Anxious Person's Breath Manual synthesizes 454 studies into one practical guide — slow breathing, nasal breathing, nervous system science, and practical protocols you can use today.
Get the Manual for $27As an Amazon Associate I earn from qualifying purchases.
The Breathing 411
Weekly breath science, wisdom, and practical tools.