Almost ten years ago, I blocked my nose for ten consecutive days.
It was a breathing experiment conducted at Stanford University. The question was straightforward: what happens to healthy adults when nasal breathing is taken away entirely? Not reduced. Removed. Mouth only, day and night, for ten days.
I agreed because I wanted to understand the consequences of something I had spent years advising people to avoid. I did not expect the experiment to become one of the most instructive experiences of my life.
What followed was a systematic unravelling, of sleep, of stress regulation, of appetite, of mood, that taught me more about how breathing and CO₂ shape human physiology than almost anything I had read before it. This article follows that unravelling and explains what was happening at each stage.
Night one: the sleep collapses
The first night felt almost normal. By night three, I was snoring an average of three hours per night. I had never snored in my life.
I woke repeatedly. I struggled to fall back asleep. Eventually I stopped wanting to go to bed at all, because I knew how awful the night would be.
The mechanism behind this is something we explored in detail in the first article in this sleep series. Mouth breathing increases breathing volume even at rest. During sleep, when metabolic demand is low, this means breathing exceeds the body’s actual needs. CO₂ drops. The instability loop begins: breathing destabilises, CO₂ drops below the apneic threshold, breathing briefly falters, CO₂ then spikes back up past the arousal threshold, and the brain surfaces. The same chemistry, the same cycle, every night. It says 2am. It will say 2am tomorrow.
My body was demonstrating this mechanism in real time, living inside it rather than reading about it.
The cleaning crew stops working
What was not happening during those broken nights matters as much as what was.
As we explored in the second article in this sleep series, deep sleep is when the brain runs its cleaning cycle. The glymphatic system, discovered only in 2012, uses the widened spaces between brain cells during deep sleep to flush out metabolic waste, including proteins associated with Alzheimer’s and Parkinson’s disease. CO₂ rising during healthy sleep is the signal that activates this process. Without that rise, the cleaning cycle never fully engages.
Every time I surfaced during those ten nights, the cycle stalled. CO₂ dropped. Blood vessels constricted. The dishwasher kept getting interrupted before it could finish. A 2026 clinical trial with 39 human participants has since confirmed what the animal research suggested: normal sleep actively clears these proteins from the brain, and sleep deprivation impairs that clearance[1]. Night after night of fragmented sleep means night after night of incomplete maintenance.
Day seven: the cravings arrive
This is the part of the experiment that surprised me most. And it is the part that pointed toward something I have been investigating ever since.
Before the experiment, I had little interest in sugar or fast carbohydrates. Even during the first two days with my nose blocked, cravings were absent. But as the sleep deprivation accumulated, something shifted in my metabolism. By the final four days, the cravings had reached seven or eight on a ten-point scale. Chocolate. Ice cream. Pizza. Beer. Foods I normally rarely touch became daily habits.
Like stepping back into an older, more stressed version of myself.
At the time I did not understand why. I do now.
What was actually happening: the fuel switch
The body stores roughly 2,000 calories as sugar and 50,000 to 300,000 calories as fat. One of these fuel sources could power you for weeks. The other barely lasts a day. Every cell can burn either one. But which fuel it chooses depends heavily on how much oxygen is available.
Fat cannot be burned without oxygen. To break down a single fat molecule, your cells require 23 molecules of oxygen. To break down a single sugar molecule, they need only six. When oxygen is scarce, the body switches to the fuel that costs less to access. It burns sugar.
And here is what most sleep and diet advice misses entirely: CO₂ governs how much oxygen actually reaches your cells. Through the Bohr effect, CO₂ is what causes haemoglobin to release oxygen into the tissues. When CO₂ drops, as it does with overbreathing or mouth breathing, less oxygen reaches the cells, regardless of how much oxygen you are inhaling. The body senses this shortfall and switches fuel sources. It burns more sugar. Sugar runs out fast. Blood sugar drops. The brain sends an alarm: eat something sweet, eat it now.
This is not a failure of willpower. It is a physiological response to a change in fuel access.
A Finnish study measured this directly. Researchers compared the respiratory quotient, the ratio that reveals which fuel the body is burning, of people with hyperventilation syndrome versus healthy controls. At rest, there was little difference. But when participants stood up, the groups diverged sharply. The hyperventilators’ breathing volume nearly doubled. Their metabolism shifted to roughly 70 percent sugar burning. The healthy controls stayed at 77 percent fat burning. Same bodies, same room, same position. The only difference was how they breathed[2].
During my ten days of forced mouth breathing, my CO₂ was chronically depleted. My cells were less able to access fat. My sugar reserves were depleted faster. My brain demanded more. The cravings were the signal.
Sleep makes it worse
Poor sleep does not just disrupt breathing and CO₂. It independently impairs the body’s ability to handle the sugar it is burning.
A randomised clinical trial at Columbia University restricted participants’ sleep to 6.2 hours per night for six weeks, not severe deprivation, just a modest reduction similar to what millions of people experience routinely. The result: insulin resistance increased by nearly 15 percent overall, and by more than 20 percent in postmenopausal women. The cells were becoming less responsive to insulin, less able to manage blood sugar efficiently. Critically, these effects were independent of any changes in body fat. The poor sleep itself was causing metabolic dysfunction[3].
When participants returned to their normal sleep duration, insulin sensitivity recovered.
The reversal matters. It means the system is not permanently damaged. It responds to conditions. Restore the conditions, and it restores.
The reversal
After ten days, the blockage came out. I taped my mouth and slept.
The difference was, literally, like night and day. I woke the next morning with enormous energy. I felt happy. I could not stop smiling or singing. My snoring had dropped from three hours to zero. The cravings had vanished. Harmony returned, immediately and completely, as if a switch had been flipped.
The experiment continued for another ten days of slow, low, nasal breathing. One evening I deliberately tested the system: a beer, pizza, ice cream, and chocolate. That night, my snoring data spiked immediately.
The system is that sensitive.
One connected system
What the Stanford experiment revealed was not several separate problems. It was one system responding to one upstream variable.
Mouth breathing depleted CO₂. Depleted CO₂ disrupted sleep. Disrupted sleep impaired the glymphatic cleaning cycle and damaged insulin sensitivity. Impaired insulin sensitivity and depleted CO₂ together shifted fuel burning from fat to sugar. Depleted sugar triggered cravings. Those cravings, if acted on, promoted further shallow breathing and CO₂ loss, which made sleep worse, which deepened the dysfunction.
The loop tightens with every turn.
Restore nasal breathing, and the CO₂ returns. CO₂ stabilises sleep. Sleep restores the cleaning cycle and insulin sensitivity. The body accesses fat again. The cravings disappear.
This is not a sleep story or a metabolism story or a breathing story. It is all three, because they are the same story. CO₂ is the thread that runs through all of it.
The Stanford experiment was one corner of a much larger investigation. How CO₂ shapes not just sleep but energy, metabolism, and the function of nearly every system in the body. Ten days showed me the outline of it. Seventeen years of study have filled in the detail. And the further I go, the more central CO₂ turns out to be.
References
[1] Dagum P, Elbert DL, Giovangrandi L, et al. The glymphatic system clears amyloid beta and tau from brain to plasma in humans. Nat Commun. 2026;17:715.
[2] Malmberg LP, Tamminen K, Sovijarvi AR. Orthostatic increase of respiratory gas exchange in hyperventilation syndrome. Thorax. 2000;55(4):295–301.
[3] Zuraikat FM, Scaccia S, Laferrère B, et al. Chronic insufficient sleep in women impairs insulin sensitivity independent of adiposity changes: results of a randomized trial. Diabetes Care. 2023;46(11):1941–1948.
The information in this article is for educational purposes only and does not constitute medical advice. If you have a diagnosed sleep disorder or are currently using medical equipment to manage your sleep, please consult your healthcare provider before making changes to your treatment.
