Creatine: Dumitrescu et al., 2026
A small MR spectroscopy study of 20 people in early methamphetamine abstinence found elevated glial markers (myo-inositol and choline compounds) in fronto-striatal brain regions, but no relationship between those neurometabolite changes and impulsivity scores. The connection between stress and motor impulsivity was consistent across groups, but the brain chemistry differences appear to be markers of methamphetamine exposure rather than drivers of behavior in this sample.
Key takeaways
- Recently abstinent methamphetamine users showed elevated myo-inositol and choline in the caudate and medial prefrontal cortex — markers of glial inflammation — but normal N-acetylaspartate levels, suggesting neurons remain intact.
- Impulsivity scores were higher in the abstinent group on motor, self-control, and cognitive complexity subscales, but the elevated brain metabolites did not predict those impulsivity scores in regression analysis.
- Stress level predicted motor impulsivity across both groups; the brain chemistry changes appear to be consequences of methamphetamine exposure, not direct causes of impulsive behavior.
- This is a cross-sectional snapshot of 20 individuals — underpowered to detect subtle brain-behavior correlations, and unable to test whether metabolite changes normalize with longer abstinence.
- The study used creatine as the reference molecule for all other metabolites; if methamphetamine alters creatine levels itself, the ratios could be misleading.
The study
Dumitrescu and colleagues at Oregon Health & Science University published this magnetic resonance spectroscopy (MRS) study in NeuroImage: Clinical in 2026 (PubMed link). They recruited 20 adults in early abstinence from methamphetamine (mean age 34, 55% male, average 9 days abstinent) and 21 age-matched controls. MRS scans measured neurometabolite concentrations in four bilateral brain regions: dorsolateral prefrontal cortex, medial prefrontal cortex, anterior cingulate cortex, and caudate. The metabolites of interest were N-acetylaspartate (a neuronal health marker), myo-inositol and choline compounds (glial inflammation markers), and glutamate plus glutamine (excitatory neurotransmitter pool), all expressed as ratios to total creatine. Participants completed the Barratt Impulsiveness Scale (BIS-11), a 30-item self-report questionnaire broken into six first-order factors. The authors tested whether group differences in metabolite levels predicted impulsivity subscale scores, controlling for education, depression, anxiety, and stress.
How to read this study
What this paper does well: The authors measured four distinct brain regions bilaterally and multiple metabolites simultaneously, providing a comprehensive neurochemical profile rather than a single-voxel snapshot. They controlled for psychiatric covariates (depression, anxiety, stress) in their regression models — important because those factors independently predict impulsivity. The use of six BIS-11 subscales instead of a single total score is methodologically sound; impulsivity is not unitary, and motor impulsivity, self-control failure, and cognitive complexity may have different neural substrates. The MRS data quality was acceptable, with coefficients of variation below 15% for the primary metabolites.
What this paper is missing: The sample size of 20 recently abstinent individuals is underpowered to detect brain-behavior correlations. Neuroimaging-behavior regressions typically require 100+ participants to reliably detect effect sizes in the r = 0.2-0.3 range, which is common for brain-cognition associations. The study is cross-sectional — we see elevated glial markers but cannot tell whether they normalize with continued abstinence or whether they predate methamphetamine use. The authors used total creatine as the denominator for all metabolite ratios, a standard MRS practice, but methamphetamine could alter creatine levels themselves (some evidence suggests stimulants deplete phosphocreatine in striatum), which would artificially inflate or deflate the ratios. Absolute quantification would be more definitive.
How I'd weight this paper: This is a well-executed but underpowered exploratory study. The finding that glial markers are elevated but don't predict impulsivity is informative — it suggests the neurometabolite changes are consequences of methamphetamine toxicity rather than mechanisms of ongoing behavioral dyscontrol. The stress-impulsivity link replicates prior work and is the most robust finding here. I treat this as hypothesis-generating: the glial changes are real and warrant tracking in larger samples and longitudinal designs, but I would not conclude that myo-inositol or choline supplementation would reduce impulsivity based on this evidence.
What they found
Recently abstinent individuals scored significantly higher than controls on three BIS-11 subscales: Motor impulsivity (mean difference not reported numerically, but p < 0.05), Self-Control (p < 0.05), and Cognitive Complexity (p < 0.05). The other three subscales (Attention, Cognitive Instability, Perseverance) did not differ between groups.
MRS detected elevated myo-inositol/creatine ratios in the caudate (group effect p < 0.05) and elevated choline/creatine ratios in both the caudate and medial prefrontal cortex (both p < 0.05) in the abstinent group. N-acetylaspartate and glutamate+glutamine ratios did not differ between groups in any region. The effect sizes for the metabolite elevations were not reported as Cohen's d, but the authors described them as "modest."
In regression models predicting impulsivity subscales from metabolite levels, controlling for group, education, depression, anxiety, and stress: Motor impulsivity was predicted only by stress level (β not reported, p < 0.05). Self-Control was associated with caudate myo-inositol in the control group (higher metabolite = worse self-control, p < 0.05) but not in the abstinent group (interaction term significant). Cognitive Complexity was predicted by group membership but no metabolite levels. None of the other metabolite-impulsivity associations reached significance.
What it means for the average man
If you've never used methamphetamine, this study tells you nothing about your brain or behavior. If you're in recovery from stimulant use, the finding is that early abstinence shows elevated glial activity markers in frontal and striatal regions — signs of inflammation or repair processes — but those changes don't directly drive impulsive behavior. Stress level was the strongest predictor of motor impulsivity across all participants, regardless of drug history.
The practical takeaway: stress management is the actionable intervention target during early recovery. The neurometabolite elevations may resolve with time, but this study can't tell us that. Supplementing with creatine, choline, or myo-inositol is not supported by this evidence — the elevated levels are likely a compensatory or inflammatory response, not a deficiency.
The caveats
Twenty participants in the abstinent group is too small to rule out real brain-behavior correlations; the study may have missed true effects. The cross-sectional design means we don't know if metabolite levels normalize with longer abstinence or if they predict relapse. The study excluded people with other substance use disorders, limiting generalizability to the typical polysubstance-using population. Impulsivity was self-reported; laboratory-measured impulsivity (e.g., go/no-go tasks, delay discounting) might show different relationships. The authors did not report whether the recently abstinent group was receiving any treatment (counseling, medication-assisted treatment) during the study period, which could influence both metabolite levels and impulsivity. Funding source was NIH grants; no industry conflicts declared.
Frequently asked questions
Is a sample size of 20 people enough to trust these findings?
For detecting large group differences in brain metabolites — yes, 20 vs. 21 can show elevated myo-inositol if the effect is robust. For detecting brain-behavior correlations (does myo-inositol predict impulsivity?) — no, correlational neuroimaging studies are chronically underpowered below 100 participants. The null findings (no metabolite-impulsivity link) may be real, or the study may have simply been too small to detect them.
Does this mean creatine supplementation could help with methamphetamine recovery?
No. The study used total creatine as a reference denominator, not as an intervention target. The elevated myo-inositol and choline ratios could reflect either true increases in those metabolites or decreases in creatine — the ratio alone doesn't tell us. There is no evidence in this paper or elsewhere that exogenous creatine supplementation affects methamphetamine recovery outcomes.
Why didn't the elevated brain metabolites predict impulsivity if they're both elevated in the same people?
Because correlation at the group level (abstinent individuals have higher metabolites AND higher impulsivity) does not guarantee correlation at the individual level (within the abstinent group, those with the highest metabolites are not the most impulsive). The metabolite changes may be markers of methamphetamine exposure rather than mechanisms of behavioral dyscontrol. Stress level was the only consistent predictor of motor impulsivity across both groups.
What does "early abstinence" mean in this study?
The average participant was 9 days abstinent at the time of scanning. This is the acute withdrawal and early recovery window, when relapse risk is highest and brain chemistry is most unstable. The findings may not apply to people with months or years of sustained abstinence, whose neurometabolite profiles may have normalized.
Sources
- Dumitrescu A., Vest M., Loftis J., Murnane K. Fronto-striatal neurometabolite signatures of impulsivity in early abstinence from methamphetamine. NeuroImage Clin. 2026. PubMed
- Verdejo-García A., Lorenzetti V., Manning V., et al. A roadmap for integrating neuroscience into addiction treatment: a consensus of the Neuroscience Interest Group of the International Society of Addiction Medicine. Front Psychiatry. 2019;10:877.
These statements have not been evaluated by the Food and Drug Administration. This article is for educational purposes only and is not intended to diagnose, treat, cure, or prevent any disease. Consult your physician before starting any supplement regimen, especially if you have a medical condition or take prescription medications.