Neurological Health, Energy, and Who Is at Risk of Deficiency

Evidence Review

Neurological Consequences: Clear Causation, Dose-Dependent Recovery

Ali et al. (2025) conducted a systematic review of 10 randomized controlled trials examining B12 supplementation for neurological outcomes in people with clinical and subclinical deficiency [1]. In patients with overt deficiency and symptoms (peripheral neuropathy, cognitive impairment), supplementation produced consistent and meaningful improvements. Oral therapy at high doses (1,000–2,000 mcg daily) was equivalent in efficacy to intramuscular injection and offered superior tolerability. However — a crucial nuance — in older adults with subclinical deficiency (low-normal serum B12 without overt symptoms), supplementation did not significantly improve cognitive or neurological outcomes. This suggests that intervention matters most before clinical symptoms appear, and that not all low-B12 states will reverse fully with supplementation once the window has passed.

The established mechanistic pathway is demyelination from impaired methylation. Subacute combined degeneration of the spinal cord (SCD) — a specific degeneration of the dorsal and lateral columns — is the most severe neurological manifestation. It presents with impaired vibration and position sense, spastic gait, and eventually paralysis if untreated. MRI findings (posterior column signal changes) can precede symptoms. Early case series have consistently shown reversal of early-stage SCD with aggressive B12 repletion, partial recovery in moderate stages, and minimal recovery once demyelination is advanced.

Cognitive Decline: Associated but Not Fully Causal

Moore et al. (2012) reviewed 43 studies examining B12 and cognitive impairment or dementia [2]. Low B12 levels (particularly below 150 pmol/L) were associated with increased risk of cognitive decline and dementia, including Alzheimer's disease. B12 therapy produced cognitive improvements in patients with documented deficiency, particularly when treated early. In patients without established deficiency, however, B12 supplementation did not meaningfully improve cognition.

The association is partially mediated through homocysteine: elevated homocysteine (itself associated with low B12 and folate) is a recognized risk factor for vascular dementia and brain atrophy. B12 supplementation reliably reduces homocysteine — but whether this translates to clinical cognitive protection in people with adequate baseline B12 remains uncertain. The Oxford Homocysteine study found that B vitamin supplementation (including B12) significantly slowed brain atrophy in people with mild cognitive impairment and elevated baseline homocysteine, suggesting a targetable subpopulation.

Metformin-Induced Depletion: Large-Scale Evidence

Two large studies establish the metformin-B12 connection beyond dispute. Aroda et al. (2016) analyzed data from 2,155 participants in the DPPOS — the longest-running diabetes prevention trial [3]. After 5 years of metformin versus placebo, B12 deficiency (≤203 pg/mL) was more than twice as common in the metformin group (4.3% vs 2.3%; p = 0.02). Borderline-low B12 (≤298 pg/mL) affected 19.1% of metformin users versus 9.5% of placebo. Crucially, each year of metformin use increased the odds of deficiency by 13% (OR 1.13; 95% CI 1.06–1.20). The risk was dose-dependent.

Chapman et al. (2016) conducted a meta-analysis of 26 studies and performed a pooled analysis of 4 trials, finding that metformin reduced serum B12 by a mean of 57 pmol/L [4]. This reduction was statistically significant and large enough to push borderline individuals into deficiency. The mechanism — interference with calcium-dependent uptake of the intrinsic factor-B12 complex — is well-characterized. Notably, this can be reversed or mitigated with calcium supplementation (400–1000 mg calcium carbonate with metformin), though this is not a standard recommendation and routine B12 monitoring in metformin users remains the clearest clinical response.

Vegan and Vegetarian Deficiency: More Pervasive Than Recognized

Niklewicz et al. (2024) synthesized 22 studies examining functional (not just serum) B12 markers in vegan adults [5]. Results were consistent across markers: vegans had lower serum B12, lower holotranscobalamin, higher MMA, and higher homocysteine compared to omnivores — all pointing to functional deficiency in a substantial proportion of unsupplemented vegans. Among supplemented vegans, all four markers improved significantly, normalizing in most individuals.

The clinical significance is that functional B12 deficiency can be present even when serum B12 is in the "normal" range. Holotranscobalamin below 50 pmol/L and MMA above 0.26 µmol/L both indicate intracellular insufficiency — these tests catch people that standard serum B12 misses. This has implications not just for vegans but for any population with borderline-low serum B12 whose functional status is uncertain.

Forms Comparison: No Clear Metabolic Winner

Paul and Brady (2017) reviewed pharmacokinetic and clinical data on all four B12 forms [6]. All forms (methylcobalamin, adenosylcobalamin, hydroxocobalamin, and cyanocobalamin) were absorbed and converted to the active intracellular coenzymes — the ratio of methylcobalamin to adenosylcobalamin inside cells was not influenced by which supplement form was ingested. Studies comparing retention showed that methylcobalamin was retained in tissues longer than cyanocobalamin (which was excreted more rapidly in urine), suggesting better tissue uptake — though whether this translates to superior clinical outcomes in adequately-dosed trials has not been definitively established.

The practical conclusion: bioidentical forms (methyl, adenosyl, hydroxo) are biochemically preferred and avoid the small theoretical cyanide exposure from cyanocobalamin, but at clinically effective doses of 1,000–2,000 mcg, all forms have been shown to effectively replete B12 status. Dose matters more than form for most people.

Evidence Strength Summary