POSTGRAD. MED. J. (1962), 38, 3 5

THE ESTIMATION AND SIGNIFICANCE OFTHE LEVEL OF, VITAMIN B12 IN SERUM

G. H. SPRAY, B.Sc., M.A., D.PHIL..NVffield Department of Clinical AMedicitne, Radcliffe Infirmary, Oxford

SINCE Ross (1950) described the estimation ofvitamin B12 in human serum, many papers on thissubject have appeared. In this paper an attemptwill be made to review some of the methods andresults which have been described, from thestandpoints of diagnosis and significance in re-lation to the requirements of patients for vitaminB12. Lack of space precludes detailed descriptionsof methods.

MethodsNormal human serum contains a few hundred

micro-microgrammes (,u,ug.) of vitamin B12 perml. A micro-microgramme is one-millionth of amillionth of a gramme, or I/I,000,000,000 of amilligramme. No chemical method is capable ofmeasuring such small amounts. Specialized tech-niques are necessary and, with one exception, allmethods employ microbiological assays. Certainmicro-organisms require vitamin B12 for growth indefined media, which must contain all nutrients,other than vitamin B12, needed by the organism.Over a certain range, which varies according tothe organism used, the concentration of vitaminB12 is proportional to the amount of growth,which can be assessed either turbidimetrically orby measuring the changes in pH of the medium, orby titrating the amount of acid produced in themedium, after growth. The concentration ofvitamin B12 in serum can be computed by com-paring the growth responses to known amountsof serum, or suitable extracts of serum, with thoseproduced by known quantities of vitamin B12The principles underlying microbiological assays

are simple, but technically the methods presentconsiderable difficulty. Cleanliness, attention toapparently trifling details of technique and experi-ence are necessary to ensure satisfactory results.These aspects of microbiological assay methodswere discussed by Girdwood (ig6oa).

OrganismsMethods using Euglena gracilis var. bacillaris

(Ross, 1952) and Z strain (Hutner, Bach and Ross,1956) and Lactobacillus leichmannii (Rosenthal andSarett, 1952; Spray, 1955) are most common,

while Ochromonas malhamensis (MVollin and RossI955; Baker, Ziffer, Pasher and Sobotka, I958)and the Escherichia coli mutant I I3-3 (Grossowicz,Aronovitch and Rachmilewitz, I954) are also ern-ployed. Euglena and Ochromonas offer advantagesin greater sensitivity and specificity, Euglenaresponding to as little as 0.25 ,ug. vitamin B12per ml. medium (Hutner and others, 1956). How-ever, they require longer periods of incubation(three to eight days) at special temperatures, andEuglena needs intense and uniform illumination.

For general work with human subjects Lacto-bacillus leichmannii appears to be the best organism,giving results in one to two days and making up insimplicity and reliability what it lacks in specificityand sensitivity. Complete media, the use of whichfor determinations of vitamin B12 in serum was firstdescribed by Meynell, Cooke, Cox and Gaddie(1957), are available commercially (Difco Labora-tories through Baird and Tatlock (London) Ltd.;Danochemo A/S, Copenhagen, Denmark, throughGeorge T. Gurr Ltd., London). The preparationof complicated media is thus eliminated, althoughin my experience this process is relatively simple,and, having developed a method using mediaprepared in the laboratory (Spray, I955), I havenot changed since the commercial media weremarketed.The growth of Lactobacillus leichmannii is pro-

moted by analogues of vitamin B12 as well as bythe vitamin itself, and also by desoxyribosides,such as thymidine. While the latter may complicatestudies on the serum of animals (Booth and Spray,I960), human serum does not appear to containsuch interfering compounds (Spray, 1955).Preparation of Serum for AssayMost of the vitamin B12 in serum is bound to

proteins, in which state it is unavailable to thetest organism. The vitamin is released from theproteins by heat, either by adding diluted wholeserum to the medium and heating the mixture(Ross, 1952), by heating diluted serum beforeaddition to the medium (Grossowicz and others,1954) or by previously heating diluted, bufferedserum to precipitate the proteins and release the

copyright. on M

ay 23, 2020 by guest. Protected by

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.38.435.35 on 1 January 1962. D

ownloaded from

36 POSTGRADUATE MEDICAL JOURNAL Janyary 1962

vitamin B12 simultaneously. Some authors includetraces of cyanide at this stage, while others do not(Table i). The use of cyanide gives higher resultswith Lactobacillus leichmannii (Girdwood, ig6ob),and Matthews (I962) has shown that withoutcyanide vitamin B12 is not completely extractedfrom the proteins. I found it impossible to recoveradded vitamin B12 from serum quantitatively inthe absence of cyanide (Spray, I955).Whether or not to include cyanide is thus an

important question, the answer to which dependson the method used. The results with Euglena,where diluted whole serum is added to the medium,are not affected by cyanide (Hutner and others1956); when proteins are precipitated beforeadding the serum to the medium there seems tobe good evidence that cyanide should always beused in order to ensure complete release of vitaminB12 from proteins, whatever test organism is used.If all laboratories employing methods involvingpreliminary precipitation of proteins would use astandard method of extraction, including cyanide,it seems likely that results from different centreswould become comparable one with another. Thissuggestion is supported by the good agreementbetween the results of parallel estimations carriedout by Dr. D. M. Matthews at the Royal FreeHospital, London, and myself on 13 samples of

serum (Fig. i). At present, because of the dif-ferences in normal ranges (Table i), results can

01

E

a 800

400.

0 400 800.' 200 1600 2600M4 vitamin B 1per ml. serumcXFOREI

FIG. I.-Graph showing the relationship between thevalues for vitamin B12 activity obtained in twolaboratories on 13 samples of serum. The dottedlines indicate the limits of the normal ranges foundin the two laboratories. There is close agreementbetween the results in the low and normal rahges,but some discrepancy on the t'wo samples withhigher levels. This discrepancy is not sufficientto vitite, the value of the test.

TABLE I

CONCENTRATIONS OF VITAMIN B12 IN THE SERUM OF CONTROL SUBJECTS AND PATIENTS WITH PERNICIOUS ANJEMIA REPORTEDBY DIFFERENT AUTHOkS

,Aug. Vitamin B,2 Activity per ml. Serum

Authors and Country Organism Extraction Procedure Controls Pernicious AnwmiaNo. Mean Range No. Mean Range

Unglaub, Rosenthal and Goldsmith Lactobacillus Heat precipitation of 31 210 70- 420 6 40 I0- 70(x9s4}-U.S.A. leichmannii proteins, no cyanide

Meynell and others (,957)-G.B. .. ,, , , ,, " 43 281 105- 672 20 4I 5- 95-90* 244 85- 625J

Girdwood (Ig6ob)-G.B. ........... ,, ,, ,, ,, 267* 28i Ino- 8io 241 - < 50-130

Spray and Witts (Is958)-G.B. .. ,, ., Heat precipitation of 123 450 I50-1,000 92 64 10-170proteins, with cyanide

Matthews (i962)-G.B. 70...............,, ,,., 7 480 x20-1,150 40 - <50-125

Nieweg, Faber, de Vries and Kroese ,, ,, Serum diluted, auto- 36 528 310-I,050 17 <79 < S-175(1954)-Holland claved with medium

Halsted, Carroll and Rubert (I959)- ,, ,, Not stated 333 470 110-1,260 I6 37 x6-O4U.S.A.

Grossowicz and others (1954)-Israel Escherichia Serum diluted, heated 3o - 200-1,000 8 79 50-130coli 1I3-3 with cyanide

Lear, Harris, Castle and Fleming Euglena gracilis Serum diluted, heated 2o 532 292- 856 33 39 0- 8s(1954)-U.S.A. var. bacillaris with medium

Pitney and Beard (I955)-U.S.A. ,, ,, ,, ,, 212 86- 460 I6 13 0- 46

Mollin and Ross (sI9.57)-G.B. .. ,, ,, " " 223 356 1I00- 900 320 <38 < IO-I I0

Cooper (1959)-U.S.A. .. .. ,, ,, , 32* 499 - 22 6iEuglena gracili 27* 305 -I 5 44Z strain

Barakat and Ekins (i96i)-G.B. .. Saturation analysis 2 626 330- 840 3 49 I- 74

* Hospital patients used as bontrols.

copyright. on M

ay 23, 2020 by guest. Protected by

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.38.435.35 on 1 January 1962. D

ownloaded from

January 1962 SPRAY: The Estimation and Significance of the Level of Vitamin B12 in Serum

usually be referred only to the normal range of theparticular laboratory performing the investigation.However, the normal ranges obtained by thoseauthors who use cyanide (Grossowicz and others,1954; Spray and Witts, 1958; Matthews, I962)agree closely and results given by the method ofGrossowicz and others (I954) are almost identicalwith those obtained when the proteins are pre-cipitated by heat in the presence of cyanide (Rach-milewitz, Izak, Hochman, Aronovitch and Grosso-wicz, 1957). These observations provide furthersupport for the suggestion that cyanide shouldalways be used.

Accuracy and ReproducibilityMicrobiological assays are subject to inherent

errors. Whilst careful technique and attention todetail may minimize such errors, microbiologicalassays will never approach the levels of accuracyobtainable by chemical methods. It is essential,therefore, to, assess the accuracy of the method ineach laboratory by measuring the recovery ofknown amounts of vitamin B12 added to a numberof different samples of serum, and the reproduci-bility of the results by assaying the same sampleof serum a number of times. In my handsrecoveries of added vitamin B12 from io samples ofserum varied between 8o and I52%, with meansof io8% with 250 tg. vitamin B12 added per ml.serum, and 112% with 500 ,u,ug. per ml. added(Spray, I955). A sample of pooled human serumwas assayed 59 times between February I958 andJuly 1959. No deterioration in vitamin B,2activity was detected in deep-frozen batches ofthis sample during this period and the valuesranged from 250 to 440 Wg. vitamin B12 per ml.(mean 326, standard deviation 45). These resultsare comparable with data published by others(Rosenthal and Sarett, 1952; Girdwood, ig6ob).

Because of the errors of these methods, there is acertain range on either side of the limits of normalin which the results are equivocal when based ononly one or two determinations, as is usual. Thismust be borne in mind when interpreting 'lownormal' or 'slightly subnormal' results.

Other MethodsA method using radioactively-labelled vitamin

B,2 and 'saturation analysis' for the determina-tion ofvitamin B12 in plasma was recently described(Barakat and Ekins, I96I). When known amountsof radioactively-labelled vitamin B12 are added tosamples of a standard plasma some becomesbound to proteins and some remains free. Theratios of the free and bound fractions, which canbe separated b.y dialysis, are plotted against theamounts of radioactive vitamin B12 added. Theresulting calibration curve can then be used to

measure unknown amounts of non-radioactivevitamin B,2 extracted from samples of plasma tobe tested.No other reports of the use of this method have

yet appeared. It may offer advantages where thefacilities and skill for microbiological work are notavailable, but in terms of the number of manipula-tions and the apparatus required it appears to bemore cumbersome than the microbiologicalmethods. It is reported to give results comparablewith the latter (Table I).

ResultsThe validity of these methods for diagnosing

deficiency of vitamin B12 depends primarily on acomparison of the results from control subjectswith those from patients with classical untreatedAddisonian pernicious anaemia. A selection ofresults published by various authors in these twogroups of subjects is given in Table I. 'This listis not intended to be exhaustive, but to illustratethe considerable differences found in normalranges according to the test organism'and themethod of treating serum, and to show that' inpatients with untreated pernicious anemia theresults are almost always below the normal range.Low results may be encountered in a number of

other conditions, including subacute combineddegeneration of the' spinal cord due to deficiencyof vitamin B12, patients with cerebral manifesta,-tions of deficiency of vitamin B12, dietary deficiencyof vitamin B12, fish tapeworm anemia, patientswho have had a partial or total gastrectomy, themalabsorption syndrome, intestinal diverticulosisand the blind loop syndrome, pregnancy whethernormal or accompanied by megaloblastic anamia,and megaloblastic anamia due to treatment withanti-convulsant drugs (Heinrich, 1954; Wokes,Badenoch and Sinclair, 1955; Nyberg and Ostling,1956; Spray and Witts, 1958; Girdwood, I96ob;Smith, 1960). The significance of low values willbe discussed in the next section.

Pathologically high values have been reportedin various types of leukhemia, in polycythaemiavera and in myelosclerosis and hamochromatosis(Mollin and Ross, 1955; Spray and Witts, 1958).In leuklemia and polycythaemia, with the exceptionof chronic myeloid leukamia where the level seemsto be consistently high, the result in a particularpatient may be normal or raised and little diag-nostic information is obtained from the investiga-tion. High results are also found in liver disease,where the level may fall as the patient improves.The test may have some diagnostic application(Rachmilewitz, Stein, Aronovitch and Grossowicz,1958) in distinguishing hepatocellular from extra-hepatic obstructive jaundice.

37copyright.

on May 23, 2020 by guest. P

rotected byhttp://pm

j.bmj.com

/P

ostgrad Med J: first published as 10.1136/pgm

j.38.435.35 on 1 January 1962. Dow

nloaded from

POSTGRADUATE MEDICAL JOURNAL

Significance of the Level of Vitamin B12 inthe Serum in Relation to the Amount oftheVitamin in the Tissues

Mollin and Ross (I953) studied the changes inthe morphology of the bone marrow and in theserum vitamin B12 level in patients with perniciousanaemia after injections of vitamin B12. Theyconcluded that the level of vitamin B12 in theserum was maintained at the expense of that in thetissues. This hypothesis was supported by Pitneyand Beard (I955). Whilst the diagnostic value ofthe serum vitamin B12 level in pure vitamin B12deficiency is fully established, the relationship ofthe level in serum to the amount of the vitamin inthe tissues is clearly important in assessing thevalue of the estimation in conditions other thanpernicious anaemia and closely related diseases.

It is difficult to investigate this problem directlyin man. In rats we found that after total gas-trectomy the levels of vitamin B12 in both serumand liver decreased, but the concentration inserum became subnormal earlier than that inliver (Booth and Spray, I960). Evidence for asimilar relationship was found in normal rabbits(Simnett and Spray, I96I). One recent observa-tion supports the view that these relationshipsobtain in man, and other reports provide cir-cumstantial evidence pointing to the sameconclusion. Adams (personal communication)measured the vitamin B12 in the liver of a patientwho died some time after a total gastrectomy. Thelevel of vitamin B12 in the serum had been sub-normal for io months before death, but the livercontained 356 jig. vitamin B12 (0.254 [Lg./g.).These values are slightly below the lowest figuresfound in 17 control subjects, but are much higherthan the levels found in the livers of untreatedpatients with pernicious anaemia by others (Wolff,Drouet and Karlin-Weissman, I951; Nelson andDoctor, 1958). It seems, therefore, that as in ourrats, the level of vitamin B12 in the liver of thispatient decreased more slowly than that in theserum. Sobotka, Baker and Ziffer (I960) foundthat the concentration of vitamnin B12 in the redcells of three normal subjects was as much as fourtimes less than that in the plasma, while three outof four patients with untreated pernicious anamiahad a higher concentration of vitamin B12 in thered cells than in the plasma. As deficiency ofvitamin B12 develops, therefore, the concentrationin plasma appears to fall before that in red cells.If the level of vitamin B12 in red cells is a measureof that in other tissues, these results support thehypothesis. No further data on levels of vitaminB12 in red cells have appeared.Some subjects seem to maintain health and

normal blood formation in the face of low levelsof vitamin B12 in their serum due to impaired

absorption or reduced dietary intake. In thisDepartment a number of subjects with 'latentpernicious anemia' have been studied at intervalsfor up to eight years (Callender and Spray, I962).Several have shown abnormally low levels ofvitamin B12 in the serum for several years whilethey were in good health and showed little or noevidence of abnormalities in their blood. Vegans, agroup of strict vegetarians whose diet is completelydevoid ofanimal protein, may be in good health andhave a comparatively normal blood picture aftermany years on the diet, yet their serum may containabnormally low amounts ofvitamin B12 (Wokes andothers, 1955). Both types of subject probablypossess sufficient vitamin B12 in their tissues fornormal hamopoiesis to continue for a long timedespite the reduced levels in their serum. Thefall in the levels of vitamin B12 in the serum ofsuch subjects probably precedes depletion of thevitamin in their tissues, and may be the first signof their failure to absorb adequate amounts ofvitamin B12.Low results for the concentration of vitamin B12

in the serum have been reported in conditionswhere there may be increased demand for thevitamin as a result of stress or deficiency of someother substance required for normal blood form-ation. Such conditions include normal pregnancy(Heinrich, I954), iron-deficiency anamia (Cox,Meynell, Gaddie and Cooke, I959), idiopathicsteatorrhcea with megaloblastic anmemia due todeficiency of folic acid (Mollin and Waters, I96I),and in some patients who absorbed crystallinevitamin B12 normally after partial gastrectomy(Deller, Richards and Witts, I96I). The patientswith steatorrheea also absorbed vitamin B12normally, and their diet appeared to containadequate amounts of vitamin B12. The levels ofvitamin B12 in the serum increased in the patientswith iron-deficiency anaemia after treatment withiron, and in the patients with steatorrhiea aftertreatment with folic acid. If the tissues must be-come depleted of vitamin B12 before the levels inserum can fall below normal, all these groups ofpatients should have shown overt signs of de-ficiency of vitamin B12. Apart from the megalo-blastic anaemia attributable to deficiency of folicacid in the patients with steatorrhcea, no signs ofdeficiency of vitamin B12 were reported except intwo of the patients with partial gastrectomies, inwhom? the signs were minimal.Adams (I96I) found that after injections of

I000 ,ug. vitamin B12, patients with perniciousanaemia retained about six times more vitamin B12than after doses of 5o or ioo ,ug., yet the con-centration of vitamin B12 in their serum remainedwithin the normal range for only a short timelonger after the large doses than after the smaller

38 Yanuary i96zcopyright.

on May 23, 2020 by guest. P

rotected byhttp://pm

j.bmj.com

/P

ostgrad Med J: first published as 10.1136/pgm

j.38.435.35 on 1 January 1962. Dow

nloaded from

January I962 SPRAY: The Estimation and Significance of the Level of Vitamin B12 in Serum 39

ones. It was suggested that some of the vitamin B12retained after the large doses was broken down inthe body. This suggestion is unnecessary if thelarger amounts retained after injections of iooo ,ug.were taken up by the tissues, and the levels ofvitamin B12 in the serum were not maintained atthe expense of the tissues.

All these observations can be explained if thelevel of vitamin B12 in the serum becomes sub-normal before the supplies of the vitamin in thetissues are exhausted, and in the light of the datapresented above it seems reasonable to accept thisrelationship as the best explanation of the facts.Some caution is therefore necessary in the diag-nostic interpretation of low serum vitamin B12values. Although a low result will indicate de-ficiency of vitamin B12 in most patients in whomthe test is performed for diagnostic purposes, itdoes not seem to be a legitimate conclusion thatthe tissues are depleted of vitamin B12 in everypatient with a low level. The observation must beassessed in the light of the clinical picture and theresults of other laboratory tests, particularly themorphology of the bone marrow and the peri-pheral blood, the presence or absence of achlor-hydria and of gastric mucosal atrophy, and theabsorption of radioactively-labelled vitamin B12and of other substances such as fat and folic acid.It should then be possible to see whether there issome defect, such as deficiency of iron or of folicacid, which might cause a temporary reduction inthe level of vitamin B12 in the serum. If evidencefor a defect of this sort is found, the concentrationof vitamin B12 in the serum may return to normalwhen medication other than vitamin B12 is given.

Estimation of vitamin B12 in serum may be ofparamount importance in patients suspected tohave subacute combined degeneration of thespinal cord in whom the blood picture is com-paratively normal, and in patients with cerebralmanifestations of deficiency of vitamin B12. Thetest has undoubted value in confirming the diag-nosis in untreated pernicious anamia, although

low results may be obtained in some patients withmegaloblastic anamia due to other causes. How-ever, a normal value in a patient with megalo-blastic antemia suggests that the patient is notsuffering from pernicious anxamia. The possi-bilities of the patient having received vitamin B12before the blood was taken, and of contaminationof the sample from syringes or containers, must ofcourse be considered first. In the absence of anysuch explanation for an anomalous result, someintestinal lesion, dietary deficiency, or drug treat-ment must be considered. The results of othertests, such as those mentioned above, shouldenable a definite diagnosis to be made.A further point requiring consideration in

interpreting values for serum vitamin B12 inindividual patients is the standards of normalityto which the results are referred. The variationbetween laboratories has already been discussed.Most normal ranges are based on values from asfew as 20 Up to 2oo to 300 control subjects. Evenfrom these comparatively small groups the lowerlimit of the normal range may be as much as i otimes less than the upper limit, and it is evidentthat. data are required from much larger groupsof control subjects to establish a true basis forcomparison. Little is known about the influenceof racial and dietary factors on the serum vitaminB 12 values in healthy subjects (Matthews, I962).These factors, together with the inaccuracies ofthe methods, must all be considered in interpretingresults. Nevertheless, determination of the levelof vitamin B12 in serum has been, and will continueto be, a valuable tool in the study of the megalo-blastic anTemias and the roles of vitamin B12,folic acid and related compounds in bloodformation.

I am grateful to Dr. J. F. Adams, Western Infirmary,Glasgow, W.i, for permission to publish details of hispatient with total gastrectomy; to Dr. D. M. Matthews,Royal Free Hospital, London, W.C.i, for supplyingsome of the data included in Table I; and to ProfessorL. J. Witts for his interest in my work on this subjectwhich has been carried out in his department.

REFERENCESADAMS, J. F. (I96I): Therapeutic Effect, Utilization, and Fate of Injected Vitamin B,2 in Man, Brit. med. J3., i, 1735BAKER, H., ZIFFER, H., PASHER, I., and SOBOTKA, H. (1958): A Comprison of Maternal and Faetal Folic Acid and

Vitamin B,2 at Parturition, Ibid., i, 978.BARAKAT, R. M., and EKINS, R. P. (I96I): Assay of Vitamin B,2 in Blood, Lancet, ii, 25.BOOTH, M. A., and SPRAY, G. H. (I960): Vitamin B12 Activity in the Serum and Liver of Rats After Total Gastrec-

tomy, Brit. J. He-mat., 6, 288.CALLENDER, S. T., and SPRAY, G. H. (I962): Latent Pernicious Anmmia, Ibid. (in the press).CooPER, B. A. (I959): Studies with a More Rapid Method of Vitamin B,2 Assay Using Euglena gracilis, _'. clin. Path.,

12, 153-Cox, E. V., MEYNELL, M. J., GADDIE, R., and COOKE, W. T. (1959): Inter-relation of Vitamin B12 and Iron, Lancet,

ii, 998.DELLER, D. J., RICHARDS, W. C. D., and WITTs, L. J. (I96I): Changes in the Blood After Partial Gastrectomy, with

Special Reference to Vitamin B12: II, The Cause of the Fall in Serum Vitamin B12, Quart. J3. Mod., 3I, 89.GIRDWOOD, R. H. (I960a): Folic Acid, its Analogs and Antagonists, Advances in Clinical Chemistry, 3, 235. New

York: Academic Press.

copyright. on M

ay 23, 2020 by guest. Protected by

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.38.435.35 on 1 January 1962. D

ownloaded from

40 POSTGRADUATE MEDICAL JOURNAL January I962

GIRDWOOD, R. H., (Ig6ob): Microbiological Methods of Assay in Clinical Medicine with Particular Reference to theInvestigation of Deficiency of Vitamin B12 and Folic Acid, Scot. med. J., 5, io.

GRossowIcz, N., ARONOVITCH, J., and RACHMILEWITZ, M. (1954): Determination of Vitamin B12 in Human Serumby a Mutant of Escherichia coli, Proc. Soc. exp. Biol. (N. Y.), 87, 513.

HALSTED, J. A., CARROLL, J., and RUBERT, S. (1959): Serum and Tissue Concentration of Vitamin B12 in CertalnPathologic States, New Engl. Y. Med., 260, 575. 4

HEINRICH, H. C. (1954): Die Biochemischen Grundlagen der Diagnostik und Therapie der V'itamin-BL2 Mangel-zustande (B12-hypo- und avitaminosen) des Menschen und der Haustiere II. Untersuchungen zum Vitamin B12-Stoffwechsel des Menschen wahrend der Graviditait und Lactation, Klin. Wschr., 32, 205.

HUTNER, S. H., BACH, M. K., and Ross, G. I. M. (I956): A Sugar-containing Basal Medium for Vitamin B12-Assaywith Euglena: Application to Body Fluids, J. Protozool., 3, IOI.

LEAR, A. A., HARRIS, J. W., CASTLE, W. B., and FLEMING, E. M. (I954): The Serum Vitamin B12 Concentration inPernicious Anemia, Y. Lab. clin. MIed., 44, 715.

MATTHEWS, D. M. (I962): Observations on the Estimation of Serum Vitamin B12 Using Lactobacillus leichmannii,Clin. Sci., 22 (in the press).

MEYNELL, M. J., CooKE, W. T., Cox, E. V., and GADDIE, R. (1957): Serum-Cyanocobalamin Level in Chronic Intes-tinal Disorders, Lancet, i, 90I.

MOLLIN, D. L., and Ross, G. I. M. (1953): Serum Vitamin B12 Concentrations of Patients with Megaloblastic AnmmiaAfter Tteatment with Vitamin B12, Folic Acid, or Folinic Acid, Brit. med. J., ii, 640.

-- , (1955): Serum Vitamin B12 Concentrations in Leukemia and in Some Other Haematological Conditions,Brit. J. Hemat., I, 1.

- , - (1957): 'The Pathophysiology of Vitamin B1, Deficiency in the Megaloblastic Anemias, in Vitamin B12und Intrinsic Factor, p. 413. Stuttgart: Ferdinand Enke Verlag.and WATERS, A. H. (I96I): Clinical Aspects of Metabolic Interrelationships Between Folic Acid and Vitamin B12,

Int. Z. Vitaminforsch., 31, 428.NELSON, R. S., and DOCTOR, V. M. (1958): The Vitamin B1, Content of Human Liver as Determined by Bio-Assay

of Needle Biopsy Material, Ann. intern. Med., 49, 136I.NIEWEG, H. O., FABER, J. G., DE VRIES, J. A., and KROESE, W. F. S. (1954): The Relationship of Vitamin B12 and

Folic Acid in Megaloblastic Anemias, Y. Lab. clin. Med., 44, I i8.NYBERG, W., and OSTLING, G. (1956): Low Vitamin B1, Concentrations in Serum in Fish Tapeworm Anemia, Nature

(Lond.), 178, 934. 0

PITNEY, W. R., and BEARD, M. F. (1955): Vitamin B1, Deficiency Following Total Gastrectomy, Arch. intern. Med.,95, 591.

RACHMILEWITZ, M., IZAK, G., HOCHMAN, A., ARONOVITCH, J., and GROSSOWIcz, N. (I957): Serum Vitamin B12 inLeukemias and Malignant Lyphomas, Blood, 12, 804., STEIN, Y., ARONOVITCH, J., and GROSSOWICZ, N. (I958): The Clinical Significance of Serum Cyanocobalamin(Vitamin B12) in Liver Disease, Arch. intern. Med., IO3, III8.

ROSENTHAL, H. L., and SARETT, H. P. (1952): The Determination of Vitamin B1, Activity in Human Serum, 7. biol.Chem., 199, 433.

Ross, G. I. M. (1950): Vitamin B1, Assay in Body Fluids, Nature (Lond.), I66, 270.- (1952): Vitamin B1, Assay in Body Fluids Using Euglena gracilis, Y. clin. Path., 5, 250.SIMNETT, K. I., and SPRAY, G. H. (I96I): Relationship of the Concentration of Vitamin B12 in Serum to that in Liver

and Kidney in Rabbits, Nature (Lond.), in the press.SMITH, A. D. M. (I960): Megaloblastic Madness, Brit. med. J3., ii, I840.SOBOTKA, H., BAKER, H., and ZIFFER, H. (I960): Distribution of Vitamin B12 Between Plasma and Cells, Amer. J.

clin. Nuttr., 8, 283.SPRAY, G. H. (I955): An Improved Method for the Rapid Estimation of Vitamin B12 in Serum, Clin. Sci., 14, 66I.

, and WITTS, L. J. (1958): Results of Three Years' Experience with Microbiological Assay of Vitamin B12 inSerum, Brit. med. 7., i, 295.

UNGLAUB, W. G., ROSENTHAL, H. L., and GOLDSMITH, G. A. (1954): Studies of Vitamin B12 in Serum and UrineFollowing Oral and Parenteral Administration, J. Lab. clin. Med., 43, I43. N

WOKES, F., BADENOCH, J., and SINCLAIR, H. M. (1955): Human Dietary Deficiency of Vitamin B12, Voeding, I6, 590.WOLFF, R., DROUET, P.-L., and KARLIN-WEISSMAN, R. (195I): L'emploi de la Ponction-Biopsie pour L'etude de la

Vitamin B12 H6patique chez l'Homme, C.R. Acad. Sci. (Paris), 232, 568.

copyright. on M

ay 23, 2020 by guest. Protected by

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.38.435.35 on 1 January 1962. D

ownloaded from