How do B12 injections compare to oral supplements?

How it works

The passive pathway is the mechanistic basis for high-dose oral therapy: at 1000-2000 mcg/day, the 1.2% absorbed by passive diffusion (around 12-24 mcg) substantially exceeds the 2.4 mcg/day RDA. In pernicious anaemia, atrophic gastritis, and post-gastrectomy or terminal-ileal-resection users, intrinsic-factor-mediated absorption is severely reduced; high-dose oral B12 can still raise serum B12 via passive diffusion in these populations but injection remains UK first-line for confirmed malabsorption.

Effective dose

Lower oral doses (1.5-50 mcg/day) require intrinsic factor and are insufficient for users with malabsorption. NICE NG239 (March 2024) reflects updated evidence and recommends considering oral instead of IM where cause is uncertain and malabsorption is not suspected. Oral monotherapy is not recommended for confirmed pernicious anaemia or established malabsorption in UK practice; high-dose oral can still effectively raise serum B12 in these populations via passive diffusion but IM remains first-line.

Forms compared

Thakkar and Billa 2015 (Indian J Pharmacol, PMC5370327) systematic review: clinical evidence does not consistently demonstrate superiority of methylcobalamin over cyanocobalamin in correcting deficiency or improving haematological or neurological outcomes. Marketing claims of methylcobalamin superiority are not anchored in clinical trial evidence. UK BNF: cyanocobalamin injection is less suitable for prescribing and cannot be prescribed in NHS primary care; hydroxocobalamin is retained in the body longer and is the standard parenteral form. Cyanocobalamin 1 mg tablets are the licensed UK oral form (Orobalin, generic equivalents).

Timing

UK regional NHS guidance (e.g. NENC ICB 2025): neurological improvement usually begins within one week of starting injection therapy; substantial resolution typically 6 weeks to 3 months. Established subacute combined degeneration of the spinal cord may not fully reverse if therapy is delayed. Sensory neuropathy and gait disturbance are most likely to improve; cognitive symptoms recover variably. Reassessment at 3 months: serum B12 alone is unreliable when supplementation is ongoing; functional markers (MMA, homocysteine) more sensitive.

Safety profile

Cyanocobalamin in renal impairment: cyanide moiety theoretical concern at very high doses; clinical significance limited at standard supplement doses. Methylcobalamin and hydroxocobalamin do not contain cyanide. Pernicious anaemia: lifelong replacement required; do not stop without specialist input. NICE NG239 emphasises starting therapy without waiting for test results when neurological symptoms are present (urgency outweighs route question).

Special populations

Pregnancy: B12 deficiency has implications for neural tube development alongside folate; standard antenatal screening per NICE. Breastfeeding mothers who are vegan: ensure adequate B12 supplementation (infant B12 deficiency can develop). Many UK primary care services now offer self-administered IM as a third option for stable maintenance users, combining reliability of IM with reduced clinic burden. Patient preference for self-administered home therapy can favour oral when cause permits.

Interactions

Nitrous oxide exposure (recreational or anaesthetic): inactivates B12 by oxidising the cobalt centre; can precipitate functional B12 deficiency. Folate co-supplementation during acute B12 repletion: folic acid 5 mg/day for 4 weeks is sometimes added to address functional folate deficiency from accelerated red cell production, but B12 must be confirmed and therapy started first or alongside; folate alone can mask B12 deficiency and allow neurological progression. Investigation before relying on oral therapy alone in unexplained deficiency: intrinsic factor antibodies, parietal cell antibodies, coeliac serology (anti-tTG IgA).

Guideline positions

Wang 2018 Cochrane updated Vidal-Alaball 2005 (CD004655.pub2, PMID 16034940). Both concluded that high doses of oral vitamin B12 (1000-2000 mcg/day) may be as effective as IM administration in producing short-term haematological and neurological responses in B12-deficient patients. The 2018 update included 3 RCTs (Bolaman 2003, Kuzminski 1998, Saraswathy 2012; total n around 150); evidence base is limited and conclusion is appropriately tentative. UK clinical practice has historically favoured IM; the 2024 NICE update is the first major UK guideline to formally consider oral as an option in non-malabsorption cases. British Society for Haematology guidance suggests reviewing the need for continuation of alternate-day IM after 3 weeks in neurological cases.

Practical framework

Reassessment at 3 months: serum B12 alone is unreliable when supplementation is ongoing (level rises whether or not tissues are using B12 effectively). Functional markers: MMA most sensitive and specific (B12 required for methylmalonyl-CoA mutase; deficiency raises MMA); plasma total homocysteine also raised but non-specific. NICE NG239 (2024) recommends MMA testing in selected cases. Investigation for cause before relying on oral alone: intrinsic factor antibodies, parietal cell antibodies, coeliac serology, consideration of relevant medications. This is a summary of published research, not personal health advice. Discuss any health or supplement decisions with a qualified healthcare professional, particularly during ongoing care, pregnancy, or with chronic conditions.

Common misconceptions

Claim: serum B12 at 3 months indicates whether therapy is working. Once supplementation is ongoing, serum B12 alone is unreliable; functional markers (MMA, homocysteine) are more sensitive.

Claim: starting folate when both B12 and folate are low is fine. The folate trap (haematological response masks while neurological damage progresses); replace B12 first or alongside.

Claim: low-dose oral cyanocobalamin (1.5-50 mcg/day) is sufficient for users with absorption issues. Lower doses require intrinsic factor; only high-dose oral (1000-2000 mcg/day) bypasses via passive diffusion.