Rewire Weekly is one science backed protocol every Tuesday, drawn from the latest longevity and neuroscience research. I am a practitioner and researcher. The goal is not wellness content. It is clinical evidence translated into something you can actually use.

The brain is 2% of body mass and consumes 20% of resting ATP output. During sustained cognitive work — a difficult problem, a long meeting, a period of high stress — local ATP in the prefrontal cortex and hippocampus can deplete faster than mitochondria can regenerate it through oxidative phosphorylation alone.

The solution the brain evolved is the same one your muscles use to sustain a ten-second sprint: a phosphocreatine reserve. Phosphocreatine donates its phosphate group to ADP and regenerates ATP in milliseconds, faster than any other cellular energy system. This system does not require oxygen. It does not require glucose transport. It fires immediately.

The problem is that brain phosphocreatine stores are finite, they deplete with age, they compress under chronic stress and sleep deprivation, and the average diet — particularly one low in red meat — supplies significantly less creatine than the system requires to stay saturated. What the past decade of research has established is that supplementing creatine can increase brain creatine content and that this increase produces measurable improvements in the cognitive domains most sensitive to ATP availability.

Creatine enters the central nervous system via SLC6A8, a sodium-dependent transporter on the blood brain barrier. Once inside, it is phosphorylated by CKBB — the brain-specific isoform of creatine kinase — to form phosphocreatine. This stored phosphocreatine acts as a rapid energy buffer: when neuronal ATP demand spikes during high-frequency firing, CKBB catalyzes the reverse reaction (phosphocreatine + ADP → creatine + ATP), replenishing the local ATP pool faster than mitochondria can respond.

This is the primary mechanism. It explains why the cognitive benefits of creatine are most pronounced under conditions of energy stress: sleep deprivation, aging, and dietary deficit all reduce the baseline phosphocreatine pool and therefore increase the magnitude of the supplementation effect.

A 2023 review in the Archives of Razi Institute documented a second mechanism with implications beyond energy buffering. Creatine interacts directly with NMDA receptors, GABA receptors, and the sodium potassium ATPase — the ion pump responsible for maintaining the electrochemical gradients required for action potential transmission. This makes creatine a neuromodulator in the pharmacological sense: it is influencing synaptic transmission, not merely fueling it.

A third mechanism involves structural plasticity. A 2025 study published in Food Science and Nutrition demonstrated that CKBB activity is critical for hippocampal dendritic spine integrity. When CKBB was reduced by 34% in animal models, researchers documented cognitive deficits, oxidative stress markers, and visible damage to hippocampal spine morphology. Creatine supplementation reversed this — increasing CKBB activity by 36% and CKBB expression by 14.3% in an accelerated aging model. The hippocampus is the structure first compromised in Alzheimer's disease pathology. CKBB activity is a modifiable upstream variable.

The 2024 meta-analysis by Xu et al. in Frontiers in Nutrition is the most comprehensive synthesis to date: 16 randomized controlled trials, 492 participants, ages 20.8 to 76.4 years, all using creatine monohydrate. The aggregate findings:

Memory: standardized mean difference 0.31 (95% CI: 0.18 to 0.44) — statistically significant.
Attention time: SMD −0.31 (95% CI: −0.58 to −0.03) — significant (faster).
Processing speed: SMD −0.51 (95% CI: −1.01 to −0.01) — significant (faster).
Overall cognitive function: not significant.
Executive function: not significant.

The pattern is interpretable. Memory and processing speed are the cognitive domains most sensitive to ATP availability and hippocampal function — the domains where the phosphocreatine mechanism has the clearest foothold. Executive function and overall cognition are more diffuse, drawing on broader prefrontal networks less directly tied to local ATP buffering.

The memory-focused meta-analysis by Prokopidis et al. (2022, Nutrition Reviews, 8 RCTs) found an overall memory SMD of 0.29 (p = 0.02). The subgroup finding that matters for this audience: in adults aged 66 to 76, the memory SMD was 0.88 (p = 0.009) — three times the effect size seen in younger adults. The mechanism is consistent with the basic science: aging reduces both dietary creatine intake and endogenous synthesis, compressing the baseline phosphocreatine pool and increasing the magnitude of the supplementation effect.

The largest single trial, published in BMC Medicine in 2023 by Sandkühler et al. (n = 123, crossover, double blind, placebo controlled, 5g per day for 6 weeks), found a small but directionally consistent benefit on backward digit span — a working memory task — that approached statistical significance (p = 0.064, η² = 0.029). This is an honest result: a modest effect in a well-powered study that did not use an easy outcome measure.

Creatine monohydrate is the only form with cognitive trial evidence. Every study in the 2024 meta-analysis used creatine monohydrate. This is not a minor point — it is the complete evidentiary basis. Creatine HCl, creatine ethyl ester, and other delivery variants have no cognitive RCT data.

Brain creatine stores increase with supplementation, but the increase is more modest than in muscle. Muscle creatine loading can raise intramuscular creatine content by 20 to 40%. Brain creatine increases are approximately 10 to 15% under standard dosing — the blood brain barrier limits entry to what SLC6A8 can transport, and there is no equivalent of muscle's insulin-driven creatine surge in neural tissue.

This constraint changes the dosing logic. A loading protocol — the 20g per day approach used to rapidly saturate muscle — produces unnecessary GI burden and does not meaningfully accelerate brain creatine saturation given transporter kinetics. The evidence supports a lower, consistent daily dose. Five grams per day over 4 to 6 weeks saturates the available pool without the loading side effects.

The responder profile matters. The Avgerinos et al. systematic review (2018, Experimental Gerontology) found that vegetarians outperformed omnivores on memory tasks following creatine supplementation — consistent with the depletion model. People with lower dietary creatine baselines (vegetarians, those consuming minimal red meat, older adults with reduced endogenous synthesis) carry more room to fill and show larger effects. If you eat red meat daily, your baseline brain creatine is already partially saturated and the supplementation benefit will be proportionally smaller.

The Forbes et al. 2022 review in Nutrients specifically identified sleep deprivation as a primary use case. Sleep-deprived individuals show acute drops in brain phosphocreatine that correspond to measurable cognitive impairment. Creatine supplementation attenuates this decline. For anyone running chronic sleep debt — which describes most people over 40 in demanding professional roles — the phosphocreatine reserve is operating at a structural disadvantage before the day begins.

The most underreported finding in the creatine literature is its relationship to the muscle-brain axis. A 2025 review in Frontiers in Nutrition (Ribeiro et al.) documented that creatine supplementation combined with resistance or aerobic exercise upregulates BDNF expression — brain-derived neurotrophic factor, the primary molecular driver of neuroplasticity. The mechanism runs through myokines: contracting muscle releases irisin and other signaling molecules that cross the blood brain barrier and activate BDNF transcription. Creatine amplifies this signal by sustaining the muscle energy output required to drive that myokine release.

This means creatine's cognitive benefit compounds with exercise. A person taking 5g per day and completing three sessions of moderate-intensity exercise per week is generating a different neurochemical signal than someone taking the same dose at rest. The protocol below reflects this.

The oxidative stress data adds a third dimension for the 40+ audience. The aging brain accumulates oxidative damage through declining mitochondrial efficiency and reduced antioxidant enzyme activity. Creatine's anti-inflammatory and mitochondrial stabilization effects — documented in the Zhu et al. aging model and the 2026 narrative review by Li in Frontiers in Nutrition — provide a neuroprotective input that operates independently of the ATP buffering mechanism. Two separate mechanisms, both relevant to the same aging-related risk profile.

Creatine monohydrate — Look for "creatine monohydrate" on the label specifically. Not creatine HCl, not creatine ethyl ester — no cognitive trial evidence exists for those forms. Minimum 3g per serving; 5g is the clinically validated dose. Third-party tested. All 16 RCTs in the 2024 meta-analysis used monohydrate exclusively.

I have reviewed the available options against the clinical evidence and the most reliably standardized forms are available through my Fullscript dispensary at a discount below retail. Creatine is included alongside the full cognitive and longevity stack I use and recommend.

If this issue was useful, forward it to one colleague who tracks their health seriously. That is the most direct way to help this newsletter grow.

Have a question about the research or a specific stack you want me to look at? Reply to this email. I read every response.