How we wish to be cited:
Nilsson M. Advice for assessors – how to read context and content of the Kuzminski paper [evaluation]. Rondel 2008; 28. URL: http://www.rondellen.net

Advice for assessors
How to read context and content of the Kuzminski paper

Summary

The Kuzminski study (1) is an open, randomised, and controlled trial on 33 patients with mild-moderate deficiency of vitamin B12. Only two of the patients had overt anaemia. Oral treatment consisted of two mg cyanocobalamin daily. Parenteral treatment consisted of cyanocobalamin, 1 mg i.m. for remission treatment and subsequent maintenance treatment. For the period studied, four months, oral treatment appeared to be more efficient than parenteral treatment in the present patients. In order to be properly understood, the Kuzminski paper needs careful reading with due regard to context and content. Mats Nilsson describes some major pitfalls to be avoided by future experts. He is a statistician and epidemiologist. His doctoral theses concerned views and knowledge about B12-associated problems among Swedish physicians (http://urn.kb.se/resolve?urn=urn:nbn:se:umu:diva-416). Illustration: detail analysis of patient data in the Kuzminski paper, revealing discrepancy between the mean value and the median value of serum vitamin B12 after oral treatment.

Oral vitamin B12 spreads to the world

Oral treatment of vitamin B12 deficiency appears to be about to enter the world community. It is already common in Sweden. It is also recognised in e.g USA, Canada, France, Australia, and Chile.

The reasons for the expected expansion are complex. The trials of homocysteine lowering have demonstrated that tablets are tolerable, increase serum vitamin levels, and appear to have no side effects. However, vitamin B12 and folate do not reduce cardiovascular morbidity and mortality (2).

Another reason for the expected expansion of oral vitamin B12 is the rising standards of developing countries. These countries acquire diagnostic tools and become aware of deficiency of vitamin B12 and folate in children, fertile women, and elderly people (3). A cheap tablet is an attractive alternative for a strained health care. For much the same reasons, many post-industrial countries are interested in swapping injections for tablets.

In the adjustment of practice to the necessity of time, experts (assessors, referees) on parenteral and oral vitamin B12 have become a commodity in short supply. The old experts from the 1960s are mostly dead or at least tottering on the brink of the tomb. Old texts are not easy to understand for new experts recruited from e.g. statistics, epidemiology, and pharmacology.

Since any of us could find himself recruited as an expert in an authority, national or international, in any corner of the world, I have prepared myself by listing the pitfalls offered by a key reference, which I am likely to meet (1). I would appreciate, if others take the trouble to analyse other cornerstone references for mutual increase of competence.

The context of the Kuzminski paper

The Kuzminski paper (1) is published in Blood, the official organ of the American Society of Hematology. The journal was founded in 1946 (www.bloodjournal.org) and has dominated global haematology since then. In order to set it into a European perspective, the impact factor of Blood is more than twice the impact factor of British Journal of Haematology and slightly greater than that of British Medical Journal.

Blood is partly financed by page charges, a system applied by many scientific journals. Page charges have existed for more than one hundred years. This financing implies that the Kuzminski paper, like other articles in Blood, from the point of US law is an “advertisement” and must be labelled as such.

One national expert denounced the scientific values of the Kuzminski paper (1), suspecting that it is an everyday advertisement. Which country? Discretion is a point of honour. However, the nation has a permanent seat in the Security Council of the United Nations. The veto may be handy, in case the B12 safety of its citizens is actualised. No, it was not Russia or China.

The Kuzminski study was conducted from January 1993 to September 1996 in the central New York state. At that time, it was difficult to obtain oral cyanocobalamin in pharmacological doses in USA. The Kuzminski team managed to get 1-mg tablets from a health food store (Nature´s Bounty, Bohemia, NY) and served two such tablets each morning with an American breakfast. This combination is thought to reduce the amount of vitamin B12 available for absorption from the gut to approximately 1 mg (4:Table 3).

The patients of the Kuzminski study had mild or moderate deficiency of vitamin B12; the inclusion criterion was serum vitamin B12 below 160 pg/mL. Only 38 patients were obtained from the participating centres (25 females and 8 males); such patients are difficult to find in post-industrial societies.

Five of the patients turned out to have folate deficiency, which never should be forgotten in cases of suspected vitamin B12 deficiency. Among the remaining 33 patients, only two had classical pernicious anaemia. Some other would probably have been classified as “food-cobalamin malabsorption” nowadays (5).

The implementation of the Kuzminski study

The patients in the parenteral arm of the Kuzminski trial received treatment with intramuscular cyanocobalamin  1 mg on days 1,3, 7, 10,14, 21,30, 60, 90, i.e. remission (loading) treatment with seven injections during the first month, followed by maintenance treatment with one injection each month. The regimen falls within the accepted frames of time and society (cf 6).

It is evident from the Illustration of the present analysis that parenteral therapy according to the Kuzminski schedule failed to fill body stores of vitamin B12, in contrast to oral therapy. The observation is compatible with the hypothesis that the parenteral failure was due to urinary excretion of vitamin B12 by overloading of the transport capacity of serum proteins. Such overloading is one of the weaknesses of parenteral therapy with vitamin B12 (cf 6).

There were 15 patients in the parenteral arm of the Kuzminski study, all of whom had haematocrit values close to the lower limit of the reference range at baseline (1:Table 3). The changes during therapy in haematocrit and mean corpuscular volume support the evidence from homocysteine and methylmalonic acid that parenteral therapy was efficacious.

There were 18 patients in the oral arm of the Kuzminski study, two of whom with definite anaemia (1:Table 1). The others were borderline cases, as deemed from their haematocrit values, just as the patients of the parenteral arm. The efficacy of treatment in the oral arm was more dramatic as judged from mean corpuscular volume, serum vitamin B12 (Illustration), homocysteine and methylmalonic acid.

Although parenteral treatment failed to replenish body stores of vitamin B12 within the period of the Kuzminski study (Illustration), it cured deficiency signs and symptoms (1). “Correction of haematologic and neurologic abnormalities was prompt and indistinguishable between the two groups.” Thus, clinical response is dependant on necessary vitamins available, not on filled or overfilled body stores. 

Serum vitamin B12 during therapy

The Kuzminski paper (1) provides interesting insights into the filling of body stores with vitamin B12 during treatment for deficiency (Illustration). The curves follow each other during remission treatment the first 30 days. Thus, one mg daily, corresponding to the uptake of at most 10-16 mcg daily from the gut (cf 4:Table 3), was as efficient as parenteral vitamin B12 7 x 1000 mcg during the same period.

Between days 30 and 120, the parenteral curve sank down to a plateau level of approximately 300 pg/mL, a phenomenon reflecting insufficient filling of body stores (Illustration), despite adequate treatment of deficiency for the time being. It is reasonable to assume that body stores were eventually filled by the parenteral regimen.

In contrast, the serum vitamin B12 levels of the patients in the oral arm continued to increase (Illustration). After four months, seven out of the 18 patients had serum B12 levels above 900 pg/mL. This is a conventional upper limit of the reference range in many laboratories.

Modern experts recruited from e.g. diabetology or anticoagulation may expect that serum vitamin B12 should lie within a reference range, say 200-900 pg/mL, as a sign of a well controlled patient. This view is unhistorical and clashes with biological and clinical realities. The pioneer work of both parenteral and oral treatment with vitamin B12 were designed to guarantee daily need and adequate filling of body stores of vitamin B12 (4,6).

After an injection of vitamin B12, the serum peak exceeds the upper limit of the reference range. Then the peak is flattened by urinary excretion, storing in tissues, losses in bile. After weeks or months, the serum level falls below the upper limit and eventually below the lower limit of the reference range (6).

Only about 1% of the B12 content of a tablet is absorbed from the gut. The absorbed portion behaves like an injection of equal size and duration (4). The Kuzminski paper shows that tablets were more efficacious than injections in the patient group selected for the time period studied (Illustration).

Some modern experts are troubled by the dispersion of individual values of serum vitamin B12 in the oral arm at four months (Illustration). However, each patient of the group was on his own way in his own pace towards steady state, usually 900-2000 pg/mL. The patients of the parenteral arm were closer to their point of departure (selection range). Consequently, their dispersion was less.

Documentation of vitamin B12 therapy

Parenteral treatment with vitamin B12 is nowadays regarded as gold standard in most countries. It should, however, be emphasised that the injection of cyanocobalamin or hydroxocobalamin was never documented by modern methods. The documentation is based on historical comparisons by eye with liver preparations (6). The Kuzminsky study demonstrates that an oral regimen was superior to a parenteral regimen in remission treatment, at least in aged ambulatory patients in the central state of New York (Illustration).

The Kuzminski team appears to have been surprised by their own findings (1). They elaborate that neither tablets nor injections were regulated by FDA at the time of the study. For an external observer, the efficacy of both is supported by robust clinical and laboratory evidence. A better explanation is that the patients of the study were not regulated by FDA. “Variety is not the spice of life, it is the very stuff of it.”

One of the weak points of the Kuzminski study is the statistical analysis; distribution-free tests should have been used throughout due to small samples with wide and skew scatter of individual measurements (7). Nevertheless, the paper found its way into the leading journal of global haematology. Why? The Kuzminski paper provides basal information hitherto unrivalled.

Conclusions

Treatment of deficiency of vitamin B12 is cheap and basically unpatentable (cf 2). Thus, clinical research on subtle differences cannot be financed by the pharmaceutical industry. It is desirable that the empirical oral treatment is acknowledged like the empirical parenteral treatment in all countries, provided that the quality of the preparations are regulated and proven.

The variable quality of the experts is a major concern. It is suggested that clinical expert reports, as well as the comments of the referees of an authority, are published in an established scientific journal and subjected to rapid responses and comments from the readers. After public discussion, a new preparation could be assessed and regulated by the authority.

At present, one dyslectic expert may create a trade barrier for a preparation of potential benefit for many citizens.

Mats Nilsson
County Council Hospital of Ryhov
Futurum Academy for Health and Care, Building B4
SE-551 85 Jönköping, Sweden
E-mail: mats.p.nilsson@lj.se 

Published October 24, 2008