How we wish to be cited:
Norberg B. Treatment of vitamin B12 deficiency – documentation of oral cyanocobalamin [editorial]. Rondel 2005; 22. URL: http://www.rondellen.net

Treatment of vitamin B12 deficiency
Documentation of oral cyanocobalamin

Figure 1 Serum vitamin B12 (pmol/L) after long-term treatment with cyanocobalamin, 0.5 or 1.0 mg daily. The original figure (see 1) is simplfied and reproduced by the permission of Journal of Internal Medicine (Fig 8 from Berlin et al 1968, 4). Most patients (n=64) had different kinds of pernicious anemia, a minority other sorts of impaired B12 absorption. At end of treatment, 17 out of 64 patients had serum B12 values below 300 pmol/L (400 pg/ml).

Figure 2 Calculated uptake of vitamin B12 following an oral dose of 0.5 mg radio-active cyanocobalamin taken on an empty stomach. Distribution of the total material, normal persons (n=10) and patients with impaired B12 absorption (n=64). The figure is simplified from original data of Berlin et al 1968, Table 1 and Figure 6 (4). The data are reproduced with the permission of Journal of Internal Medicine One healthy proband and 16 patients took up less than 5 microgram, i.e. less than 1% of an oral dose of cyanocobalamin.

Herman Nilsson-Ehle (1) is the first to challenge a 45-year-old global consensus, that 1 mg cyanocobalamin daily is the optimal dose for oral treatment of vitamin B12 deficiency (2-8). From about 1998, Nilsson-Ehle has preached that 0,5 mg cyanocobalamin daily is enough for treatment of B12 deficiency. His claim is based mainly on his own interpretation of the results of Berlin et al 1968 on patients with impaired B12 uptake (4) and his own trial on healthy probands with TrioBe (0,5 mg cobalamin, 0,8 mg folcic acid, 3 mg pyridoxine), one tablet daily (11). It is desirable to scrutinize his evidence, since Nilsson-Ehle (NE) is a national authority of benign hematology in Sweden.

The NE interpretation (1) of Berlin results (4) is based on means without due regard to biological variability (Figs 1,2). At long-term follow-up, 17/64 (29%) of the Berlin patients had serum cobalamin values now regarded as inclusion criterion for treatment trials (8,9). The absorption study of the Berlins supported this view; 17/74 (23%) persons absorbed less than 1% (5 microgram) of labelled cyanocobalamin (Fig 2). These data suggest that approximately 10% of "healthy" probands and 50-80% of patients with previous B12 deficiency provide potential therapy failures, if treated with 0.5 mg oral cyanocobalamin daily. These figures are fasting values. Simultaneous food reduces uptake by 50% (4). The conclusion that 0.5 mg oral cyanocobalamin daily is insufficient for deficiency treatment is supported by the Kuzminsky study 1998; two out of 18 patients treated with oral cyanocobalamin, 2 mg daily, had serum B12 values below 300 pmol/L (400 pg/ml) after 4 months (6).

The uptake of vitamin B12 in the human body by intrinsic factor (IF) from the stomach and cubilin in the distal part of ileum dominate in text-books. Most workers in the field were aware of another mechanism, "dirct uptake" of vitamin B12, about 1950 due to the use of the Schilling test at that time. In fact, the IF-mediated uptake appears to serve as a safeguard against bile losses of vitamin B12. The Berlins (4) estimated the IF-mediated uptake of vitamin B12 to 0.5-2.2 microgram a day (4). The supplementary direct uptake was thought to be located to the upper part of the small bowel.

The Baltimore group of McIntyre was first in using the direct uptake of vitamin B12 for successful oral treatment of vitamin B12 deficiency; in 1960, they reported that cyanocobalamin, 1 mg daily, was the optimal dose for oral treatment of vitamin B12 deficiency. Observation time was at most six months (2).

Berlin et al started their studies on oral cyanocobalamin about 1955 (3) and published their final report in 1968 (4), a study for approximately 12 years with specific problems such as new research reports (2) and patient dropout by death. "All patients treated for three years or less, and a few of the others, died during the observation period, for reasons not connected with the disease studied" (4). Nilsson-Ehle misunderstands the six dropouts by death as those last recruited with associated short observation time (1).

A more modern description of the observation time may be approximated from the material of Berlin et al 1968 (4):
Median 55 months
Interquartile range 45-64 monts
Range 12-70 months

By 1964, all patients (n=64) were enrolled into the study, their absorption of oral cyanocobalamin was calculated (Fig 2), and 60 of the patients were treated for more than one year (3). Tablet Behepan, cyanocoblamin 1 mg, was registered and all remaining patients were switched to the new preparation. "The patients were started on 500 or 1000 microgram of B12 daily: during the last three years of the study 1000 microgram daily were given throughout" (4).

A comparison between treatment results in the preliminary report 1965 in Swedish (3) and the final report 1968 (4) elucidates the time course of the Berlin studies. In 1965 (3), 52/64 (81%) of the patients had serum cobalamin values below 300 pmol/L (400 pg/ml). In 1968 (4), 17/64 (29%) of the patients had serum cobalamin values below 300 pmol/L (400 pg/ml). It is reasonable to assume that these 17 patients were those who had been treated with 0,5 mg, until they left the study (Fig 1). Thus, treatment strategy was changed in the period 1960-1964 from 0.5 mg cyanocobalamin daily to 1 mg daily.

It is true that the Berlins continued to praise the low dose (1), on which much of their basic research was based (4); "the harlot praises her virtue, the nun her passions." The Berlins praised "half the recommended daily dose", the love of their youth. Furthermore, there was a strain of political prudence in the Berlin communication; all researchers before McIntyre (2) had burned their fingers on low-dose oral cobalamin.

Social competence and sound science made Ragnar Berlin a professor of internal medicine in Linköping, Gunnar Brante a professor of clinical chemistry in Lund. As consultant internist, I had the favor of using the research microscope of Gunnar Brante in my cytological studies for a couple of years. The polite manners of Berlin and Brante should not be misjudged by younger workers (1). Berlin and Brante verified the original observation of McIntyre et al 1960 (2) that 1 mg cyanocobalamin daily is the optimal dose for oral treatment of B12 deficiency, nothing else.

The Berlin recommendations (4) have stood the test of time (5-8). Furthermore, the Berlin group suggested that oral cyanocobalamin, 1 mg daily, is sufficient for the treatment of neurological symptoms due to B12 deficiency (4). Their view was supported by later studies (5,9). In the period 1990-2000, the Swedish experience with oral cyanocobalamin, 1 mg, comprised approximately one million patient years (10).

In 1998, a combination of cyanocobalamin (0.5 mg), folic acid (0.8 mg), and pyridoxine (3 mg) was registered in Sweden for prophylaxis of incipient B12 deficiency (trade mark "TrioBe"). The documentation was a study on healthy probands, mean age 76 years, mean homocysteine value 18 micromol/L, observation period four months, no follow-up, no clear description of distribution of parameters (1,11).

According to my information, TrioBe was tailored for homocystein lowering in non-deficient people with vascular risks by Pharmacia Sweden in the period 1990-1995, when homocysteine was expected to prove a causal risk factor for vascular disease (cf 12). In addition, TrioBe once daily was marketed for overt B12 deficiency, mainly based on the authority of Nilsson-Ehle (1) and his interpretation of original research by previous workers (2-5). It is desirable that Nilsson-Ehle makes his documentation (1,11) complete and public; more than six years have passed with an insidious and misleading marketing but without evidence. The marketing is now subject to the scrutiny of regulatory authorities.

The marketing of TrioBe has aroused much discussion (12-14). In case anyone believes that 0.5 mg of oral cyanocobalamin daily is enough for a safe and reliable treatment, prophylactic or curative, there are simple and cheap clinical models for confirmation (14). Otherwise, it would be practicable to give TrioBe twice daily in order to adapt to previous documentation (2-5); in the Swedish context, such a step would need negotiations with regulatory authorities.

Since 1960 (2), leading hematologists have agreed that cyanocobalamin, 1 mg daily, is the optimal dose for a safe and reliable oral treatment of B12 deficiency (2-5). Hitherto, the challenge of the old consensus by Nilsson-Ehle is not supported by sound historical evidence or new clinical findings available to public examination (1,11). At present, the NE claim has an archetype structure – "teaching dad how to make kids".

1. Nilsson-Ehle H. High homocysteine – prophylaxis and treatment [debate]. Rondel 2004; 21. URL: http://www.rondellen.net/debate21_eng.htm
2. McIntyre PA, Hahn R, Masters JM, Krevans JR. Treatment of pernicious anemia with orally administered cyanocobalamin (vitamin B12). Arch Intern Med 1960; 106:280-92
3. Berlin H, Berlin R, Brante G. Peroral behandling av perniciös anemi med höga doser vitamin B12 utan intrinsic factor. Läkartidningen 1965; 62:773-81
4. Berlin H, Berlin R, Brante G. Oral treatment of pernicious anemia with high doses of vitamin B12 without intrinsic factor. Acta Med Scand 1968; 184:247-58
5. Magnus EM. Cobalamin and unsaturated transcobalamin values in pernicious anaemia; Relation to treatment. Scand J Haematol 1986; 36; 457-65
6. Kuzminski AM, Del Giacco EJ, Allen RH, Stabler SP, Lindenbaum J. Effective treatment of cobalamin deficiency with oral cobalamin. Blood 1998; 92:1191-8
7. Nyholm E, Turpin P, Swain D, Cunningham B, Daly S, Nightingale P, et al. Oral vitamin B12 can change our practice. Postgraduate Medical Journal 2003;79(930):218-9
8. Björkegren K. Studies on vitamin B12 and folate deficiency markers in elderly: A population-based study. Dissertation, Uppsala 2003
9. Nilsson K, Gustafson L, Hultberg B. Improvement of cognitive functions after cobalamin/folate supplementation in elderly patients with dementia and elevated plasma homocysteine. Internat J Geriatr Psychiatry 2001; 16:609-14
10. Nilsson M. Cobalamin communication in Sweden 1990-2000. Views, knowledge, and practice among Swedish physicians. Dissertation, Umeå University 2005
11. Lewerin C, Nilsson-Ehle H, Matoucek M, Linderstedt G, Steen B. Reduction of plasma homocysteine and serum methylmalonate concentrations in apparently healthy subjects after treatment with folic acid, vitamin B12, and vitamin B6: a randomised trial. Eur J Clin Nutr 2003; 57:1426-36
12. Jansson J-H:. Hypothesis on shaky pillars – homocysteine and vascular disease [evaluation]. Rondel 2004; 21. URL: http://www.rondellen.net/evaluation21_eng.htm
13. Rosenberg P, Hernborg A, Håkansson J, Skoglund I, Svartholm R, Söderström U, Wahlström L. B vitamins and dementia – lack of evidence in Cochrane analysis [debate]. Rondel 2004; 19. URL: http://www.rondellen.net/debate19_eng.htm
14. Norberg B. Safety and reliability of oral cobalamin [editorial]. Rondel 2004; 21. URL: http://www.rondellen.net/publisher21_eng.htm

Published January 26, 2005