Bone changes in polycythaemia and myelosclerosisBonechanges inpolycythaemia vera andmyelosclerosis...
Transcript of Bone changes in polycythaemia and myelosclerosisBonechanges inpolycythaemia vera andmyelosclerosis...
J. clin. Path. (1969), 22,696-700
Bone changes in polycythaemia vera andmyelosclerosis
B. E. ROBERTS, C. G. WOODS, D. W. MILES, AND C. R. PATERSON
From the Univetsity Departments of Haematology, Bone Pathology, Medicine and Clinical Investigation,The General Infirmary, Leeds
SYNOPSIS To study the problem of new bone formation in myelosclerosis bone biopsies have beenperformed in cases of polycythaemia vera, which may be regarded as a pre-myelosclerotic condition,and in primary myelosclerosis; morphological studies of bone and reticulin were made and osteoidand bone matrix was quantitatively assessed. Relevant metabolic studies were also carried out, butwere predominantly normal. No new bone formation was detected in polycythaemia vera and somebone was osteoporotic; no significant increase in osteoid was found. With the use of polarized lightit was found that new bone formation in myelosclerosis resulted from the ossification of wavyargyrophilic fibres in marrow: the presence of a dense network of these wavy fibres was a pre-requisite of new bone formation and ossification of the non-refractile branched reticulin fibres wasnot observed.
'Myelosclerosis' is the term commonly used todescribe a haematological syndrome the mainfeatures of which are a leucoerythroblastic anaemiaand hepatosplenomegaly due to extramedullaryhaemopoiesis. It refers in particular to the histo-logical features which comprise fibrosis of the bonemarrow (myelofibrosis) with specifically, accordingto Vaughan (1936) who introduced the term, excessbone in the spongiosa and medullary cavity.
Myelosclerosis may be primary in origin or it mayfollow polycythaemia vera. According to Szur andLewis(1966), myelosclerosis may be the most frequentterminal syndrome in a large series of cases ofpolycythaemia vera followed up for a long time.
In an attempt to observe the histological changesduring the transformation of polycythaemia vera tomyelosclerosis trephine biopsies have been performedin a series of patients with polycythaemia vera duringthe various stages of the disease. The histologicalcriteria of myelosclerosis, which were establishedfrom five myelosclerotic trephine specimens, werethen sought in the sections from these patients andmeasurements of osteoid made. Appropriate bio-chemical investigations were also performed.
PATIENTS, METHODS, AND MATERIALS
The patients with polycythaemia vera were divided intothe following groups: untreated polycythaemia vera;Received for publication 7 February 1969.
696
treated polycythaemia (all these patients had been treatedwith radioactive phosphorus for varying periods of timeup to 12 years); and spent polycythaemia (these patientshad all been treated with radioactive phosphorus andhad developed a refractory anaemia). The cases ofprimarymyelosclerosis had been diagnosed on peripheral bloodfindings; they had a chronic anaemia and there was nohistory of an antecedent polycythaemic phase. Thediagnostic criteria of these patients (Table I) were out-lined in a previous publication (Roberts, Miles, andWoods, 1969).Bone biopsies were obtained from the left anterior iliac
crest with a Sacker-Nordin trephine under local anaes-thesia in a series of patients with polycythaemia vera andprimary myelosclerosis. The cylinder of bone was fixedin barbitone-buffered 10% formol-saline and dividedtransversely into two portions. The wholly cancellousportion was used for the preparation of undecalcifiedsections; the other part was decalcified in neutral EDTAand sections were prepared from the endosteal end of thespecimen.One decalcified section was stained with Ehrlich's
haematoxylin and eosin for general histological assess-ment; another was stained by the silver impregnationtechnique of Gordon and Sweets and counterstained withvan Gieson's stain to assess the amount and distributionof reticulin and collagen.
Undecalcified sections were prepared and stained bythe von K6ssa technique and counterstained with neutralred. Osteoid measurements were recorded in threedifferent ways using polarized light: (1) the number oflaminae were counted; (2) the percentage length oftrabeculae covered by osteoid was measured with an
copyright. on January 28, 2021 by guest. P
rotected byhttp://jcp.bm
j.com/
J Clin P
athol: first published as 10.1136/jcp.22.6.696 on 1 Novem
ber 1969. Dow
nloaded from
Bone changes in polycythaemia vera and myelosclerosis
eyepiece graticule divided into 1 mm squares; (3) thearea of osteoid as a percentage of total bone matrix(osteoid + calcified bone) was estimated using a Chalkleypoint-array graticule.
Full details of the technique of osteoid measurementshave been recorded previously (Woods, Morgan, Pater-son, and Gossman, 1968).The percentage area of bone matrix in a section was
estimated with a Chalkley array in a decalcified sectionusing transmitted light.The serum values for calcium, phosphorus, and
alkaline phosphatase were measured according toWootton (1964); if the alkaline phosphatase level wasraised the heat stability index was estimated (Posen,Neal, and Clubb, 1965). Urinary hydroxyproline excretedover a 24-hour period was measured with a modificationof the technique of Prockop and Udenfriend (1960).
RESULTS
MARROW MORPHOLOGY AND RETICULIN On exam-ining the preparations stained for reticulin two maintypes of argyrophilic fibres were noted: one showeda branched pattern and was not refractile withpolarized light; the second was smooth and wavyhaving the morphology of collagen and was refractilewith polarized light. Neither fibre showed anystaining property with van Gieson's stain. It wasnoted that the density of these fibres increased withbone marrow cellularity and that the second type offibre although predominating in myelosclerosis waspresent in other myeloproliferative disorders andwas not in itself a diagnostic feature (Roberts et al,1969). However when the haematoxylin and eosinsections were examined, five cases (26, 27, 28, 29,and 30) had an obviously different morphologicalappearance from the rest; they showed completedisorganization of normal marrow architecture dueto marked proliferation of fibroblasts and reticulumcells and the presence of coarse bands of fibrousconnective tissue. This appearance was consideredto be the most important diagnostic criterion ofmyelosclerosis, and the cases with marrows showingthis appearance are shown in Table I, and includethe four cases of primary myelosclerosis and onecase of 'spent' polycythaemia.
BONE STRUCTURE IN MYELOSCLEROSIS New boneformation was sought in all sections and was con-spicuous in three cases (27, 28, and 29). In these thebone trabeculae were thickened, had irregularsurfaces, and showed cement lines which were, inparts, poorly defined and slightly irregular inpattern, though this irregularity was not so markedas to suggest the arrangement in osteitis deformans.Osteoblasts were not numerous, but in the outerparts of the trabeculae plump osteocytes lay in
TABLE IOSTEOID AND BONE MATRIX MEASUREMENTS
Case Initials Area ofNo. Osteoid( %)
Untreated polycythaemia vera1 R.S. 22 R.K. -I3 H.P. 1
4 W.B. 1
5 P.D. _16 A.Sw. 27 G.B. 1-58 J.S. 1
9 L.T. 1410 F.S. 1 511 B.C. 3
Treatedpolycythaemia vera12 A.Wa. 4131415161718192021
M.E.N.S.G.W.J.B.H.B.J.N.A.C.A.S.A.Wh.
2220-22-53-530054
Spent polycythaemia vera22 E.B. _123 C.B. 0 224 W.R. 125 L.O. 126 E.P. -'
Trabecular No. ofLength BirefringentCovered (Y.) Laminae
35
323023903050
1-2
1-21-2
1-3
1-3
4619134531549378
46
1-21-21-21-41 i
1-21-31-31l1-3
21036 1-3
Primary myelosclerosis27 O.G. _228 T.F. _229 M.L. _'
30 O.P. 62 70 1-2
Normal range 0-5 0-45 0-3(Woods, Morgan, Paterson, and Gossman, 1968)
TotalMatrix (0%)
49155_1
231010_1
610164
1616242623202015
1118
17-519 51321_1
2049_1
15 5
>15
'These were not measurable either due to fragmentation of the sectionor to the inclusion of subcortical bone."Cases with histological changes of myelosclerosis in the marrow.Osteoid was not measurable in cases 26 to 29 for reasons explained inthe text.
round or oval lacunae and were closer together thanin the central parts of the trabeculae.Examination of these sections in polarized light
showed that most of the trabecular thickening wasproduced by the addition of woven bone to the sur-
face, explaining the irregular surfaces of the trabe-culae and the plump osteocytes. It was, however,difficult, when using polarized light, to distinguishbetween the advancing border of new bone formationand the fibrous tissue of marrow and impossible tomeasure osteoid. This was due to the fact that therefractile fibres of bone marrow were continuouswith the fibres of bone matrix (Figs. 1 and 2), andthis observation led to the conclusion that the new
bone formation of myelosclerosis arises by ossi-fication of preformed smooth, wavy, argyrophilicfibres in marrow. Ossification of branched non-
697
copyright. on January 28, 2021 by guest. P
rotected byhttp://jcp.bm
j.com/
J Clin P
athol: first published as 10.1136/jcp.22.6.696 on 1 Novem
ber 1969. Dow
nloaded from
B. E. Roberts, C. G. Woods, D. W. Miles, and C. R. Paterson
FIG. 1. Argyrophilic fibres in myelosclerosis, transmitted light x 300.
FIG. 2. The same field as is shown in Fig. I but using polarized light which demonstrates thecontinuity of the fibres of bone marrow and matrix x 300.
698
copyright. on January 28, 2021 by guest. P
rotected byhttp://jcp.bm
j.com/
J Clin P
athol: first published as 10.1136/jcp.22.6.696 on 1 Novem
ber 1969. Dow
nloaded from
Bone changes in polycythaemia vera and myelosclerosis
refractile reticulin fibres was not observed. In onecase of myelosclerosis (case 30) the bone still had apredominantly lamellar construction and there wasno demonstrable increase in the quantity of bone.In this case it was possible to measure the osteoid,which was slightly in excess of normal, due mainlyto the increase of bone surface covered by osteoidbut the amount of bone matrix was at the lower limitof normal.
BONE CHANGES IN POLYCYTHAEMIA VERA No newbone formation was observed in polycythaemia vera,and, as the figures in Table I show, there was noapparent increase in trabecular bone in the sectionsfrom polycythaemia vera: indeed some were osteo-porotic as might be expected with the elderly agedistribution of this group.Osteoid measurements in the polycythaemic groups
were within normal limits (Table 1) except for a fewcases, and then the increase was slight and mainlyconfined to the percentage of trabecular lengthcovered. An exception was provided by case 9 inwhich measurements of the area and trabecularlength of osteoid were notably increased. But itshould be noted that, from measurements of totalmatrix, this patient was grossly osteoporotic and theamount of osteoid was measured as a percentage oftotal bone matrix. So in absolute terms the amountof osteoid must be much lower; and in this contextthe osteoid, when represented as a number of bire-fringent laminae, was within normal limits.
BIOCHEMICAL CHANGES Hydroxyproline excretionin all patients was normal, including those with grossmyelosclerosis. Values for serum calcium and phos-phorus were also entirely normal. Serum alkalinephosphatase levels were measured in 26 cases andthese were raised in five cases though the highestvalue recorded was 30 King-Armstrong units. Heatstability indices on the cases with raised levelsshowed that the enzyme had its origin in bone (lessthan 02) in one case (treated polycythaemia vera,case 15); in liver (greater than 0 3) in two others(untreated polycythaemia vera, cases 6 and 9); andnormal proportions of liver and bone enzyme(between 0-2 and 0 3) in the other two cases (onetreated, case 20, and one spent polycythaemia vera,case 23).
DISCUSSION
Myelosclerosis has been the subject of many pub-lications but detailed descriptions of histology andhistogenesis show wide variation.
There is little doubt that new bone is formedaround preexisting trabeculae but is there an initial
phase with osteoid formation, as Hartmann, Klima,Czitober, and Rieder (1959) suggest? Increasedosteoid was found by Oeschlin (1956) in an inter-mediate phase in the development of myelosclerosisand by Leonard, Israels, and Wilkinson (1957) in alate phase. However, Chapman (1933), Turnbull(cited by Vaughan and Harrison, 1939), Erf andHerbut (1944), and Rohr (1956) reported that osteoidwas sparse. In this series osteoid has been measuredin three different ways: the area of osteoid as apercentage of total bone matrix, the length of trabe-culae covered, and the number of birefringentlaminae. The present figures, with the exception ofcases 9 and 30, are mostly within the normal rangethough several, particularly when measuring lengthof trabeculae covered, are at the upper limit ofnormal or slightly increased. An increased amountof osteoid may be due to osteomalacia or Paget'sdisease (Paterson, Woods, and Morgan, 1968) butthere may also be a spurious rise in osteoid inosteoporotic bone, because of a reduction in bonematrix (Woods, Paterson, and Morgan, 1968). Inany series of bone specimens from elderly patients,some can be expected to have abnormal osteoidestimations because of osteoporosis and possiblythe early stages of osteitis deformans; case 9 hadsevere osteoporosis both histologically and radio-logically.
Oeschlin (1956) described an initial stage of myelo-sclerosis in which there was marked osteoclasticactivity and this was also reported in one of thecases of Beattie and Withey (1953). In the presentseries the only evidence is possibly that of pastosteoclastic activity from the irregular contour of thecement lines; no multinucleate osteoclasts wereidentified on bone surfaces.One explanation of the relationship between
polycythaemia and myelosclerosis is that they arisefrom a common stimulus to the stem cell (Vaughanand Harrison, 1939; Dameshek, 1951; Wasserman,1954). This is the concept of the 'myeloproliferativedisorders' and accordingly there should be a pro-liferation of osteoblasts. This could not be corro-borated by examination of the sections of the presentseries; indeed they were found to be sparse which is inagreement with the findings ofErfand Herbut (1944),Wyatt and Sommers (1950), and Oeschlin (1956).The serum alkaline phosphate level may be raisedin some cases of myelosclerosis (Rohr, 1956) and itmay be used as an index of osteoblastic activity,particularly if the heat stability index is below 0-2(Posen et al, 1965). However, only one case (case 15)showed a raised level of alkaline phosphatase ofbonyorigin but on examination of the bone biopsy therewas no undue osteoblastic activity, marrow fibrosis,or new bone formation, though an occasional
699
copyright. on January 28, 2021 by guest. P
rotected byhttp://jcp.bm
j.com/
J Clin P
athol: first published as 10.1136/jcp.22.6.696 on 1 Novem
ber 1969. Dow
nloaded from
B. E. Roberts, C. G. Woods, D. W. Miles, and C. R. Paterson
increase in the number of osteoid lamellae was seen.In two cases a raised alkaline phosphatase level wasconsidered to be of hepatic origin; both these caseshad appreciable hepatosplenomegaly, and this raisedlevel was possibly associated with hepatic extra-medullary haemopoiesis.
Both from the present series and reports in theliterature (Carpenter and Flory, 1941; Black-Schaffer and Stoddard, 1953; Oeschlin, 1956; Rohr,1956), there seems little doubt that although myelo-sclerotic fibrosis of the bone marrow can occurwithout new bone formation the reverse is not true.The wavy reticulin fibres termed 'pre-collagen' byRobb-Smith (1957) appear, from the polarized lightexamination, to undergo progressive ossification.This degree of organization of fibres seemed essentialfor the formation of new bone: ossification ofbranched, non-refractile reticulin was not observed.Under these conditions of ossification it was notpossible to calculate the amount of osteoid. All theorganized wavy fibres could, theoretically, ossify,and there was no demonstrable change in these fibresat the light microscope level of magnification todistinguish an advancing bone surface from a staticone.Our observations suggest that myelosclerosis
begins as a fibroblastic proliferation which leads tothe replacement of marrow by refractile wavy 'pre-collagen' fibres. These fibres are progressivelyossified by intramembranous ossification, whichdoes not require that specialized osteoblastic cellsshould proliferate on the bone surfaces. In this waymyelosclerosis differs from Paget's disease of boneand other diseases in which permanent sclerosis ofcancellous bone is a feature.We have recently noted a similar process in an
osteosclerotic secondary deposit of carcinoma ofbreast in bone, and this type of ossification appears
to be analogous to that seen in bony spurs formingat tendon insertions ('rider's bones') and in somefibrosarcomata.
This is a further indication that the primaryabnormality is in the fibroblast and that the newbone formation in myelosclerosis is essentially asecondary and subsidiary phenomenon.
We wish to thank Dr J. R. Fountain and Dr C. J. Hayterfor allowing us to investigate patients under their care,and Mr Ian Salmon for the histological preparations.One of us (C.G.W.) is in receipt of a grant from theMedical Research Council.
REFERENCES
Beattie, J. W., and Withey, J. L. (1953). Brit. toed. J., 2, 414.Black-Schaffer, B., and Stoddard, L. D. (1953). Amner. J. Path., 29, 413.Carpenter, G., and Flory, C. M. (1941). Arch. intern. Med., 67, 489.Chapman, E. M. (1933). Amier. J. tied. Sci., 185, 171.Dameshek, W. (1951). Blood, 6, 372.Erf, L. A., and Herbut, P. A. (1944). Ann. intern. Mdcl., 21, 863.Hartmann, G., Klima, R., Czitober, H., and Rieder, H. (1959). Wien.
Z. inn. Med., 40, 437.Leonard, B. J., Israels, M. C. G., and Wilkinson, J. F. (1957). Quart.
J. Med., 26, 131.Oeschlin, R. J. (1956). Acta hacuemat. (Basel), 16, 214.Paterson, C. R., Woods, C. G., and Morgan, D. B. (1968). J. Path.
Baot., 95, 449.Posen, S., Neale, F. C., and Clubb, J. S. (1965). Ai,t,. initernl. Med.,
62, 1234.Prockop, D. J., and Udenfriend, S. (1960). Ana/rt. Biochenm., 1, 228.Robb-Smith, A. H. T. (1957). In Sym171posiuni on Connective Tissiues,
edited by R. E. Tunbridge, p. 177. Blackwell, Oxford.Roberts, B. E., Miles, D. W., and Woods, C. G. (1969). Bi-it. J.
Hactinat., 16, 75.Rohr, K. (1956). Acta haeniiat. (Basel), 15, 209.Szur, L., and Lewis, S. M. (1966). Brit. J. Radiol., 39, 122.Vaughan, J. M. (1936). J. Path. Bact., 42, 541.
, and Harrison, C. V. (1939). Ibid, 48, 339.Wasserman, L. R. (1954). Bull. N. Y. Acad. Med., 30, 343.Woods, C. G., Morgan, D. B., Paterson, C. R., and Gossman, H. H.
( 1968). J. Path. Bact., 95, 441.Paterson, C. R., and Morgan, D. B. (1968). Orthopaedics, 1, 121.
Wootton, 1. D. P. (1964). MicroanalYsis in Medical Biochetiistrr,4th ed. Cht,rchill, London.
Wyatt, J. P., and Sommers, S. C. (1950). Blood, 5, 329.
700
copyright. on January 28, 2021 by guest. P
rotected byhttp://jcp.bm
j.com/
J Clin P
athol: first published as 10.1136/jcp.22.6.696 on 1 Novem
ber 1969. Dow
nloaded from