the of howell-jolly bodies and · howell-jolly bodies (%) 0 1 1-2 2-3 3+...

J. clin. Path. (1961), 14, 374

The significance of Howell-Jolly bodies andgiant metamyelocytes in marrow smears

D. W. DAWSON AND H. P. R. BURY1

From the Department ofPathology, Crumpsall Hospital, Manchester

SYNOPSIS Howell-Jolly bodies in erythroid precursors can be found in the marrow in a variety ofdiseases. More than 1% in the later erythroblasts is rarely found except in vitamin B12 or folic aciddeficiency states.

Giant metamyelocytes are probably always the result of vitamin B12 or folic acid deficiency.A distinction is drawn between the deficiency of, and therapeutic need for, vitamin B12 and folic acid.

The detection of vitamin B12 or folic acid deficiencyis often difficult, sometimes impossible, by examin-ing marrow alone, particularly if the alteration in thenuclear pattern of the erythroblasts is the only patho-logy on which the diagnosis is to be made. Cox,Matthews, Meynell, Cooke, and Gaddie (1960) referto the many reports of patients with vitamin B12deficiency and normoblastic erythropoiesis. Wheremethods of vitamin B12 and folic acid estimation areavailable the interpretation of ambiguous erythro-poiesis may assume less importance, but these pro-cedures are not yet suitable for routine hospital work.The report of Hutchison and Ferguson-Smith (1959)on the correlation between Howell-Jolly bodies inred cell precursors and vitamin B12 or folic aciddeficiency is thus of importance, but we considered itadvisable to confirm their findings, especially incases where the type of erythropoiesis was difficultto interpret. At the same time we decided tofollow Lehmann's (1955) suggestion and reviewthe significance of giant metamyelocytes in themarrow.

Other features generally considered to be associ-ated with a megaloblastic change are prematurehaemoglobinization and a maturation arrest of theerythroblasts. The former is a most subjective assess-ment and may be valueless in the presence of con-comitant iron deficiency, and was therefore notconsidered. The degree of maturation arrest had tobe taken into account when determining the signi-ficance of Howell-Jolly bodies and the value of thisfeature is also discussed.

'Present address: Children's Hospital, Sheffield, 10.

Received for publication I November 1960.

MATERIALS AND METHODS

Marrow smears from the following conditions (numbersof cases in brackets) were examined.

GROUP I CONDITIONS RESPONDING TO VITAMIN B12 ORFOLIC ACID Pernicious anaemia (45), post-gastrectomyanaemia (5), malabsorption syndrome (5), nutritional B12deficiency (1), leukaemia with megaloblastic anaemia (2),and anticonvulsant macrocytic anaemia (6). Marrowexaminations were made on these patients before thestart of treatment, and, in addition, 12 patients withmegaloblastic anaemia under treatment were investigated.

GROUP II CONDITIONS WITHOUT APPARENT B12, FOLICACID, OR IRON DEFICIENCY Leukaemia (10), other myelo-proliferative disorders (12), carcinoma (8), thrombo-cytopenic purpura (8), chronic infection (7), refractorynormoblastic anaemia (3), cirrhosis (3), haemolyticanaemia (2), various miscellaneous disorders (7), andnormal (8). There were 22 cases of iron-deficiencyanaemia not following upon a gastrectomy.The features assessed in each smear were (1) the per-

centage of erythroid precursors earlier than the inter-mediate stage of development. Three hundred erythro-blasts were counted and the percentage calculated to thenearest whole number; (2) the number of Howell-Jollybodies in 500 erythroblasts, and the prerence of two ormore Howell-Jolly bodies in any of these erythroidprecursors. Howell-Jolly bodies were not seen in cellsearlier than the intermediate stage of development.Therefore, because of the varying degree of maturationarrest in the two groups, the counts were corrected,using the percentage of erythroid precursors to give thepercentage of Howell-Jolly bodies in the intermediateor late erythroblasts. Cells in mitosis were excludedfrom the counts in Nos. 1 and 2 above. (3) The type oferythropoiesis, as judged on 500 cells, was classified asnormoblastic, megaloblastic if any cell showed t-he

374

on July 25, 2020 by guest. Protected by copyright.

http://jcp.bmj.com

/J C

lin Pathol: first published as 10.1136/jcp.14.4.374 on 1 July 1961. D

ownloaded from

http://jcp.bmj.com/

The significance of Howell-Jolly bodies and giant metamyelocytes in marrow smears

characteristic nuclear change, and transitional megalo-blastic if any cell showed a transformation towards amegaloblast but none had reached this degree of abnor-mality. (4) Cells of the metamyelocyte or stab stagelarger than 18t were counted as giant metamyelocytes.They were searched for under a high-power objective insquash preparations of marrow particles giving highlycellular fields. Their frequency in the smears was recordedas follows:-

O=none in whole smear+ I to 5 in whole smear

+ + =several in whole smear, but less than one perfour fields

+ + + I or more per four fields

the severity of the anaemia whereas in group I therewas a relationship, although it was not close.Separating the cases according to size of the Howell-Jolly bodies gave no useful information.

There was no significant difference in the percent-age of Howell-Jolly bodies between those conditionsin group I that responded to B12 and those thatresponded to folic acid (six anticonvulsant macro-cytic anaemias, three malabsorption syndrome, twoleukaemias).Among the eight normal marrows, Howell-Jolly

bodies were found in three, though each had lessthan 1 %. In none were multiple bodies present.

RESULTS

The findings in group I (excluding cases under treat-ment) and group II are presented in Table I.

TABLE IINCIDENCE OF HOWELL-JOLLY BODIES, GIANT METAMYE-LOCYTES, AND ERYTHROID PRECURSORS IN CASES WITH(GROUP 1) AND WITHOUT (GROUP 11) APPARENT VITAMIN

B12 OR FOLIC ACID DEFICIENCY

Group I Group II(64)

Without Iron- Iron-deficiencydeficiency (68) Anaemia (22)

Howell-Jolly bodies ( %)011-22-33+

Smears showing erythro-blasts containing multipleHowell-Jolly bodies

Giant metamyelocytes0+

Erythroid precursors ( %)NormalIncreased

0367

48

392521I

51222

60

2152621

6430

0

17 6247 6

Figures under groups refer to number of cases.

HOWELL-JOLLY BODIES In group 1, Howell-Jollybodies were found in every smear. In 96% therewere 1% or more, and cells containing two or morebodies were found in 94%. In contrast, although50% of the cases in group II contained Howell-Jollybodies, only 10% had over 1 % and 8 % had multiplebodies in any cell. The highest counts of Howell-Jolly bodies were in group I, 32 4 %, and in group II,

5-4 %. The degree of anaemia did not account for thedifference in findings between the two groups. Ingroup II there was no correlation between thepresence or the number of Howell-Jolly bodies and

GIANT METAMYELOCYTES In group I giant meta-myelocytes were found in 97% of cases, and in largenumbers (+ + or + + +) in 73 %. They were notseen in two cases. In group II they were seen in 14%of the cases, and in large numbers in 3 %. The threecases of group II without iron deficiency in whichgiant metamyelocytes were seen were two withchronic infections and one with acute myeloidleukaemia. No giant metamyelocytes were seen inthe normal marrow smears.The more severely anaemic patients in group I

usually had the more numerous giant metamye-locytes, but, as with Howell-Jolly bodies, the corre-lation between the degree of anaemia and numbersfound was not close. There was no such correlationin the group II cases in which they were present.There was no difference in the frequency of giantmetamyelocytes between the cases which respondedto B12 and those which responded to folic acid.

PERCENTAGE OF ERYTHROID PRECURSORS From pre-vious experience a figure of up to 9% for thisfinding was considered to be normal. The eightnormal cases in group II fell below this level. Ingroup I, 73 % and in group II, 11 % showed an in-crease in the proportion of erythroid precursors: thehighest percentage in group I was 50 and in group II37. In group I there was a fairly close correlationbetween the percentage of erythroid precursors andthe severity of the anaemia.From these results the features suggestive but not

diagnostic of B12 or folic acid deficiency appeared tobe 1% or more cells containing Howell-Jolly bodiesor multiple bodies in any erythroblast; several giantmetamyelocytes in smear (++ or more); 100% ormore early erythroid precursors.Of the 19 cases in group II with one or more of

these features, 18 were not considered to be deficientin vitamin B12 or folic acid for the following reasons:Ten cases (nine iron-deficiency anaemias and onepyridoxine-responsive anaemia) responded promptlyand completely to appropriate therapy without

375


http://jcp.bmj.com

/J C


ownloaded from

http://jcp.bmj.com/

D. W. Dawson and H. P. R. Bury

administration of vitamin B12 or folic acid. Threecases (two refractory normoblastic anaemias andone pre-leukaemic leukaemia) did not respond tovitamin B12 or folic acid. Five cases (one each ofcongestive cardiac failure, cirrhosis, idiopathicacquired haemolytic anaemia, Hodgkin's disease,and polycythaemia rubra vera) gave no clinical orhaematological indications of vitamin B12 or folicacid deficiency during the subsequent course of theirdisease. One patient with acute myeloid leukaemia(a smear showed 1-4% and multiple Howell-Jollybodies) died soon after marrow puncture and there-fore without adequate follow-up. In this case adeficiency of vitamin B12 or folic acid cannot beexcluded with certainty.

APPLICATION OF RESULTS TO DIFFICULT MARROWS

Supportive evidence of vitamin B12 or folic aciddeficiency in the marrow smear is only needed incases where the megaloblastic transformation isabsent or incomplete. The value of this evidence canonly be judged if it is of assistance in such instances,and the cases without frankly megaloblastic marrowsin group I were therefore considered separately(Table II). In the 23 cases 1% or more Howell-Jollybodies were present in 21, multiple Howell-Jollybodies in 16, giant metamyelocytes (+ + or more) in14, and the percentage of erythroid precursorsincreased in eight. One or more of these changeswere, however, present in every case.

The problem of the investigation of an anaemiaafter vitamin B12 or folic acid has been given to thepatient is not infrequent. Return of megaloblastic tonormoblastic erythropoiesis is rapid and corrobora-tive evidence of previous vitamin B12 or folic aciddeficiency persisting after the erythroblastic reversionwould be valuable. To determine how long theabnormalities considered here remain after treat-ment a series ofmarrows taken at varying times afterthe start of vitamin B12 or folic acid therapy wasexamined (Table HI). The persistence of supportiveevidence might depend on the degree of initialchange and the pre-treatment findings in the eightcases in which they were available are also shown.There is an insufficient number of cases from whichto draw definite conclusions. Cases 10 and 11 mayhave received suboptimal therapy and the results maythus be misleading. In the others, after erythro-poiesis was no longer frankly megaloblastic, Howell-Jolly bodies and the percentage of erythroid pre-cursors were still abnormal in two, and giant meta-myelocytes in three (with an occasional one in threemore).

IRON-DEFIENCY ANAEMIA

Originally all the cases in group II were consideredtogether but it was noticed that many of them show-ing the features suggestive of vitamin B12 or folic aciddeficiency were cases of iron-deficiency anaemia.Smears from patients with this disease were therefore

BLE IITWENTY-THREE CASES WITHOUT FRANKLY MEGALOBLASTIC MARROWS RESPONDING TO

VITAMIN B12 OR FOLIC ACID THERAPY

Case No. Hb (g. %) Howell-Jolly Multiple Giant % Erythroid Erythro-Bodies (%) Howell-Jolly Metamye- Precursors poiesis

Bodies locytes

2345

6789101112131415161718

19202122

14-614-012-612-512012-011-311-211-211-011-010-610-09-69.59.49.48-4

8-07-7705-6

6-62-21-81-68-02-44-84-2402-0401-04-61-42-65 82-00-4

3-61-84-40-8

YesYesNoNoYesYesYesYesYesNoYesNoNoNoNoYesYesYes

YesYesYesYes

++

++++

++

++

++

+

++++

++

++

++

+++

23 5 6 5-6 Yes 0

13 T.M. Pernicious anaemia9 N. S.A.C.D.

11 T.M. Pernicious anaemia3 N. S.A.C.D.4 T.M. Pernicious anaemia10 T.M. Pernicious anaemia + S.A.C.D.9 T.M. Pernicious anaemia + S.A.C.D.4 T.M. Pernicious anaemia8 T.M. Idiopathic steatorrhoea6 T.M. Nutritional vitamin B1, deficiency8 T.M. Chronic lymphatic leukaemia15 T.M. Pernicious anaemia13 T.M. Pernicious anaemia9 T.M. Idiopathic steatorrhoea + iron-deficiency16 T.M. Post-gastrectomy + iron-deficiency5 T.M. Pernicious anaemia7 T.M. Pernicious anaemia + S.A.C.D.8 T.M. Post-gastrectomy + non-reactive tuber-

culosis18 T.M. Pernicious anaemia + chronic pylonephritis8 T.M. Post-gastrectomy + iron-deficiency9 T.M. Anticonvulsant macrocytic anaemia6 T.M. Pernicious anaemia (? spontaneous remis-

sion)10 T.M. Acute lymphatic leukaemia

Diagnosis

376


http://jcp.bmj.com

/J C


ownloaded from

http://jcp.bmj.com/

The significance of Howell-Jolly bodies and giant metamyelocytes in marrow smears 377

recorded separately in Table I, and those with giant splenectomy, the opportunity was taken to examinemetamyelocytes, excess Howell-Jolly bodies, or a certain post-splenectomy marrows. Three patientsraised percentage of erythroid precursors in more had undergone splenectomy during upper abdominaldetail in Table IV. These 12 showed no constant operations two to three weeks previously. In eachfeatures either in their marrow findings or clinical smear Howell-Jolly bodies were present, but less thandata, and in the latter respect did not differ from the 1% in number, and in none were they multiple.10 whose marrows had no evidence suggestive of Two patients who now presented with perniciousvitamin B12 or folic acid deficiency. anaemia had undergone splenectomy six and eight

years ago during repair of hiatus hernia and forSPLENECTOMY traumatic rupture. They showed 90% and 544%

Howell-Jolly bodies, and both had cells with multipleBecause of the known relationship between Howell- bodies. These percentages were of the same order asJolly bodies in the circulating erythrocytes and the figures recorded in patients with pernicious

TABLE IIIMARROW FINDINGS IN 12 PATIENTS RESPONDING TO VITAMIN B12 OR FOLIC ACID

Before Treatment After Treatment Diagnosis Treatment

42 42~~~~~~~~~~~-.5~~~~-i ~~~~~, a~4z a,

0118~~~~~~~~~u *~~~~~ ~~~ 8-u~~~~i

242526272829630313210

5-64-64-67-24-95-6

12-0707070

11-0

M

M

T.M.M

M

M

T.M.

32-4

7-2

2-4500-816-82-0

Yes

Yes

YesYesYesYesNo

+++

+++

++

++

11 11-0 T.M. 40 Yes ++

25

26

103115116

15i hrs.22 hrs.3 days5 days8 days10 days14 days14 days15 days18 days21 days

M

M

NT.M.T.M.NNNNNN

8 28 days N

32-8 Yes +++ 21 Pernicious anaemia4-2 Yes +++ 43 Pernicious anaemia0-8 Yes +++ 6 Pernicious anaemia6-2 Yes + + 9 Pernicious anaemia0-8 No + + 16 Pernicious anaemia0-2 No + 3 Pernicious anaemia- No + 4 Pernicious anaemia0-2 No 0 6 Pernicious anaemia- No + 10 Pernicious anaemia

No 0 8 Pernicious anaemia1-4 Yes 0 6 Nutritional vitamin

B1, deficiency1-0 No 0 5 Chromic lymphatic

leukaemia

Cytamen intramuscularlyCytamen intramuscularlyCytamen intramuscularlyCytamen intramuscularlyCytamen intramuscularlyCytamen intramuscularlyCytamen intramuscularlyCytamen intramuscularlyCytamen intramuscularlyFolic acidOral B1,

Folic acid

TABLE IVTWELVE CASES OF IRON-DEFICIENCY ANAEMIA RESPONDING TO IRON THERAPY WITH MARROW

FINDINGS SUGGESTIVE OF VITAMIN B12 OR FOLIC ACID DEFICIENCYCase Sex Age Hb M.C.H.C. Marrow Findings Free Recent Duration Aetiology of AnaemiaNo. (g.%) (%.) Acid in Iron of

Howell- Multiple Giant % Test Therapy SymptomsJolly Howell- Metamye- Erythroid Meal beforeBodies Jolly locytes Precursors Marrow

Bodies Examination

69 M 64 10-8 - 10 No + 4 - Yes Weeks Peptic ulcer, chronicbronchitis

70 M 66 9-2 26 2-0 Yes 0 3 - No Weeks Idiopathic72 F 80 8-6 27 2-0 Yes + 10 Present Yes Years Steatorrhoea, duodenal

diverticulum73 F 83 8-2 29 04 No + ++ 9 - Yes Months Hiatus hernia74 M 63 8-0 24 5-4 Yes + 11 - Yes Weeks Peptic ulcer, chronic

bronchitis75 F 72 7-4 27 0-6 No + 6 - Yes Years Idiopathic77 M 40 6-8 24 0-2 No + 6 Absent Yes Years Haemorrhoids78 F 33 6-4 - 0-2 No + + 9 Present No Weeks Idiopathic79 F 45 5-8 22 0-6 No + 7 Absent No Years Duodenal ulcer81 F 52 5-8 - 0-2 No + + 17 Present No Years Idiopathic84 F 46 5-5 - 1-2 No 0 2 - Yes Months Menorrhagia89 M 59 3-6 - 0-2 No + 10 Absent No Months Gastric ulcer


http://jcp.bmj.com

/J C


ownloaded from

http://jcp.bmj.com/


anaemia with similar degrees of anaemia who hadnot undergone splenectomy. A sixth patient with arefractory normoblastic anaemia showed 1-0y%Howell-Jolly bodies before and 0-6% I1 months aftersplenectomy.

DISCUSSION

HOWELL-JOLLY BODIES IN ERYTHROID PRECURSORSHutchison and Ferguson-Smith (1959) in a study ofthe significance of the presence of nuclear fragmentsin the erythroid precursors considered it probablethat disordered nucleic acid synthesis resulting fromvitamin B12 or folic acid deficiency was the funda-mental factor in their production. They did, however,record four cases in which infrequent solitary inclu-sions were found but in which there was no evidenceof vitamin deficiency, and they thought that thebodies were possibly artefacts. They also foundsimilar inclusions in one case of iron-deficiencyanaemia.

It might be argued that the inclusions found in ourcases without vitamin deficiency were not in factHowell-Jolly bodies, but this we think unlikely.Inclusions were only counted if they were of thesame degree of staining as the nucleus of the parentcell. Often in grossly megaloblastic marrows thebodies show the same reticulation as the cell nucleus.Pappenheimer bodies for which they might be mis-taken are usually of a lighter hue than the nucleusand occur often in greater numbers in one cell thanHowell-Jolly bodies (other than in grossly megalo-blastic cells). Howell-Jolly bodies are not refractile,a feature which distinguishes them from someartefacts. In megaloblastic cells in which there weremany bodies of all sizes, the small ones were identicalto the inclusions found in scanty numbers in non-vitamin deficiency cases.

Since occasional Howell-Jolly bodies occur in nor-mal marrows and in conditions in which deficiencyof B.2 or folic acid can be excluded, a quantitativeassessment of their presence would appear to beessential if the finding is to be of value when mostneeded. This count takes only a few minutes to per-form and at the same time permits an accurateassessment of erythropoiesis. Where erythropoiesisis obviously megaloblastic a Howell-Jolly count is ofcourse superfluous.Our findings show that a count of less than 1 % of

Howell-Jolly bodies has no significance as regardsthe aetiology of the anaemia, but that more than thisnumber or the presence of multiple bodies in onecell is highly suggestive but not pathognomonic ofvitamin B12 or folic acid deficiency. The value of thefinding of excess Howell-Jolly bodies in difficultmarrow smears is shown in the selected cases of

Table II, where an increased number was found in 21,and multiple bodies in 16, of the 23 smears. If assaymethods of vitamin B12 and folic acid levels are notavailable, an excess number of Howell-Jolly bodiesis sufficient evidence to justify a therapeutic trial ofvitamin B12 or folic acid. Unfortunately, in caseswhere therapy has been started a search for Howell-Jolly bodies is of little value since they rapidly dis-appear with the conversion of erythropoiesis.

GIANT METAMYLELOCYTES The literature on thespecificity of giant metamyelocytes is conflicting.Whitby and Britton (1957) quote Henning (1935) tothe effect that they are typical of pernicious anaemiabut note that Jones (1943) met them in a variety ofdiseases. Jones, however, had found them as well asin pernicious anaemia only in diseases associatedwith megaloblastic transformation of the marrow,namely, achrestic anaemia, tropical macrocyticanaemia, sprue, and pernicious anaemia of preg-nancy. Scott (1939) commented on their presence inone or two of 23 cases of iron-deficiency anaemia.Foy, Kondi, and Hargreaves (1950) considered themto be as pathognomonic as megaloblasts in thoseanaemias which respond to liver, folic acid, Marmite,or a high-protein diet. Later, Foy and Kondi (1952)reported that they were present in 20% of the mar-rows of their patients with hypochromic anaemia.Davidson (1954) in a survey of 500 marrows foundthem associated with iron-deficiency anaemia,leukaemia, and other diseases (not stated) as well aspernicious anaemia, and Lehmann (1955) quotesDavidson as finding giant metamyelocytes in 63 of164 iron-deficiency anaemias. Tasker (1959) foundsome abnormal myeloid cells in many normoblasticiron-deficiency anaemia marrows which respondedto iron therapy. Maclver and Back (1960), on theother hand, stated that they had never seen abnormalmyeloid cells in iron-deficiency anaemia.

Because of these differing reports which suggestthat giant metamyelocytes might not be confined tovitamin B12 or folic acid deficiency, we made aquantitative assessment of their presence in themarrow smears. This showed a definite relationshipbetween the presence of many giant metamyelocytesand the deficiency of one of the vitamins. The threeiron-deficiency anaemias containing many giantmetamyelocytes would seem to refute this relation-ship, but this is not necessarily so since the work ofCox et al. (1960) and Cox, Meynell, Gaddie, andCooke (1959) provides evidence of tissue deficiencyif vitamin B12 in apparently straightforward iron-deficiency anaemia. Their cases responded satis-factorily to iron therapy alone, as ours did.

It appears therefore that the findings of frequentgiant mctamyelocytes is, in most cases, an indication

378


http://jcp.bmj.com

/J C


ownloaded from

http://jcp.bmj.com/

The significance of Howell-Jolly bodies and giant metamyelocytes in marrow smears

for vitamin B12 or folic acid therapy. Foy et al. (1950)and Maclver and Back (1960) have reported theresponse to folic acid of normoblastic anaemiaswith giant metamyelocytes though the frequency ofsuch cells was not recorded. We have met severalnormoblastic anaemias in pregnancy with numerousgiant metamyelocytes which responded to folicacid.The value of finding numerous giant metamyelo-

cytes in difficult smears appears to be less than that ofHowell-Jolly bodies since only 14 of the 23 caseslisted in Table II had such numbers but this is still asufficiently common finding to make their assessmentof value. But their absence does not exclude thepossibility that the anaemia will respond to vitaminB12 or folic acid (Cases 8 and 23, Table II).

In cases already on vitamin therapy the number ofgiant metamyelocytes like Howell-Jolly bodiesrapidly declines, although marrows eight days afterthe start of treatment may still contain many and anoccasional one may be found two weeks later (TableIII). These times are somewhat longer than thesurvival of giant metamyelocytes recorded byLeitner (1949).Davidson (1952) commented that in iron-deficiency

anaemia the giant metamyelocytes had nuclei almost,if not quite, as complicated as those associated withpernicious anaemia. Our impression was that regard-less of the aetiology of the anaemia the more bizarreforms were found in those marrows with numerousgiant metamyelocytes but in a few instances even theoccasional one had an abnormal nucleus and is pro-bably evidence of tissue depletion of vitamin B12 orfolic acid. It is significant that of the 25 cases re-corded here with occasional giant metamyelocytes,15 responded to B12 or fofic acid (two of these werenormoblastic anaemias) and the other 10 were ofiron-deficiency anaemia (see below) or had activemyeloid marrows. It may be that in cases withincreased myeloid activity this cell series, rather thanthe erythroid mass, would show the first overt signsof deficiency of B12 or folic acid. Spray and Witts(1953) have shown that patients with leukaemiaappear to require abnormally large amounts of folicacid, though this is not a constant finding. Swendseid,Swanson, Meyers, and Bethell (1952) suggested thatthe folic acid depletion in leukaemia results from theproduction of large numbers of immature whitecells of high folic acid content.

PERCENTAGE OF ERYTHROID PRECURSORS An arrestin maturation of the erythroid precursors was onlyprominent in frankly megaloblastic anaemias ingroup I, and as an aid to the diagnosis of difficultcases it is of little value. This is in accord with theexperience of others (de Gruchy, 1958).

IRON-DEFICIENCY ANAEMIA The literature on giantmetamyelocytes in iron-deficiency anaemia has beenreviewed above. Our results support the findings ofthose authors who have recorded them to be a com-mon occurrence in iron-deficiency anaemia. Anincrease in Howell-Jolly bodies is also not uncom-mon and is probably of similar significance, althoughit is not clear why the two abnormalities were notalways present together. Of interest in this connexionis an analysis of the cases of vitamin B12 or folic aciddeficiency in Table II, in which it is seen that thehighest percentage was not necessarily associatedwith the most numerous giant metamyelocytes.The finding of Cox et al. (1959, 1960) of low serum

and marrow B12 levels in iron-deficiency anaemiahas already been cited. There have also been reportsof folic acid depletion in this disease (Spray andWitts, 1953; Chanarin, Mollin, and Anderson, 1958).We believe that the abnormal findings in our casesare the morphological results of one or more ofthese deficiencies. It is not our purpose to considerthe possible ways in which iron deficiency may pro-duce deficiency of B12 or folic acid but merely torecord the frequency with which evidence of thesedeficiencies may be found.The presence of such evidence in the marrow of a

patient with iron-deficiency anaemia does not, how-ever, mean that response to iron will be suboptimalor that the patient's anaemia will improve on vitaminB12 or folic acid. This has been our experience inanaemia of pregnancy. Cox et al. (1959) have shownthat the low serum B12 level returns slowly to normalwith iron therapy, further evidence of the relation-ship between the iron and vitamin deficiency. How-ever, in certain circumstances, e.g., anaemiasassociated with gastrectomy, the malabsorptionsyndrome, or pregnancy, giant metamyelocytes orexcess Howell-Jolly bodies may be indications of avitamin depletion concomitant with, but not resultingfrom, the iron deficiency, which will persist despiteiron therapy. A case we have encountered since com-pleting this series supports this contention. A patientwith hypochromic normoblastic anaemia followingpartial gastrectomy with a serum B12 level of less than50 ju,ug per ml. responded partly to parenteral ironbut after 17 days the marrow had become franklymegaloblastic. The original marrow containednumerous giant metamyelocytes but no excessHowell-Jolly bodies. This change in erythropoiesisfollowing iron therapy has been mentioned pre-viously by Giles and Shuttleworth (1958) andTasker (1959). Therefore, although giant meta-myelocytes and, usually, an excess of Howell-Jollybodies, are evidence of B12 or folic acid deficiency,only by knowing the indications for which mar-row aspiration was performed can their clinical

379


http://jcp.bmj.com

/J C


ownloaded from

http://jcp.bmj.com/


significance be assessed. For example, if the patienthas amild normochromic anaemiawith indeterminateerythropoiesis the finding of either abnormality is avaluable guide to thediagnosis of pernicious anaemiaor related disease. On the other hand, such findingsin an iron-deficiency anaemia, or in a marrow withmyeloid hyperplasia, although basically due tovitamin B12 or folic acid depletion, indicate thathaematological response to the appropriate therapyshould be observed closely in case the vitamindeficiency should be unrelated to the primary dis-order or irreversible without specific therapy.

SPLENECTOMY Our finding of Howell-Jolly bodiesin the erythroid precursors in all our splenectomizedcases is at variance with the report of Hutchison andFerguson-Smith (1959). But we believe our results tobe logical. If Howell-Jolly bodies were absent fromthe marrow in splenectomized patients one wouldhave to postulate that the inclusions seen in theerythrocytes are the remains of the nucleus and thatthese cells are indeed nucleated red cells. But theregular eccentricity of Howell-Jolly bodies in the redblood cells and their much smaller size than thenuclei of the late normoblasts in the marrow, with-out intervening sizes being evident, does not supportthis idea.

We wish to thank Dr. A. Womack for the serum vitaminB12 estimation.

REFERENCES

Chanarin, I., Mollin, D. L., and Anderson, B. B. (1958). Brit. J.Haemat., 4, 435.

Cox, E. V., Matthews, D. M., Meynell, M. J., Cooke, W. T., andGaddie, R. (1960). Blood, 15, 376.

-, Meynell, M. J., Gaddie, R., and Cooke, W. T. (1959). Lancet,2, 998.

Davidson, W. M. (1952). Ibid, 1, 566.-- (1954). Amer. J. clin. Path., 24, Suppl., p. 88.de Gruchy, G. C. (1958). Clinical Haematology in Medical Practice,

p. 113. Blackwell, Oxford.Foy, H., and Kondi, A. (1952). Lancet, 1, 416.-, -, and Hargreaves, A. (1950). Ibid, 1, 1172.Giles, C., and Shuttleworth, E. M. (1958). Ibid, 2, 1341.Henning, N. (1935). Dtsch. med. Wschr., 61, 1543.Hutchison, H. E., and Ferguson-Smith, M. A. (1959). J. clin. Path.,

12, 451.Jones, 0. P. (1943). Arch. Path. (Chicago), 35, 752.Lehmann, H. (1955). Trans. roy. Soc. trop. Med. Hyg., 49, 90.Leitner, S. J. (1949). Bone Marrow Biopsy. English Translation from

Swiss Edition. Revised and Edited by C. J. C. Britton and E.Neumark. Churchill, London.

Maclver, J. E., and Back, E. H. (1960). Arch. Dis. Childh., 35,134.

Scott, R. B. (1939). Quart. J. Med., 32 (n.s. 8), 127.Spray, G. H., and Witts, L. J. (1953). Clin. Sci., 12, 385.Swendseid, M. E., Swanson, A. L., Meyers, M. C., and Bethell, F. H.

(1952). Blood, 7, 307.Tasker, P. W. G. (1959). Trans. roy. Soc. trop. Med. Hyg., 53,

291.Whitby, L. E. H., and Britton, C. J. C. (1957). Disorders of the Blood,

8th ed., p. 245. Churchill, London.

380


http://jcp.bmj.com

/J C


ownloaded from

http://jcp.bmj.com/

the of howell-jolly bodies and · howell-jolly bodies (%) 0 1 1-2 2-3 3+...

Documents