Ultrastructure marrow leishmaniasis · with visceral leishmaniasis. The patients were moderately anaemic but showed a suboptimal in-crease in the absolute reticulocyte count. Serumandred

J Clin Pathol 1987;40:267-275

Ultrastructure of bone marrow in patients withvisceral leishmaniasisS N WICKRAMASINGHE,* S H ABDALLA,t* E G KASILI$

From the *Department ofHaematology, St Mary's Hospital Medical School, London, the tNuffield Departmentof Clinical Medicine, University ofOxford, Oxford, and the tDepartment ofHuman Pathology, College ofHealth Sciences, University ofNairobi, Kenya

SUMMARY Ultrastructural studies were performed on bone marrow aspirates from three patientswith visceral leishmaniasis. The patients were moderately anaemic but showed a suboptimal in-crease in the absolute reticulocyte count. Serum and red cell folate concentrations and serum vita-min B12 concentrations were normal in all three cases, and serum ferritin concentrations werenormal or increased. The bone marrows were hypercellular and showed erythroid hyperplasia; ahigh proportion of the erythroblasts showed dyserythropoietic changes. Amastigote forms of Leish-mania donovani were found within bone marrow macrophages and within occasional neutrophil andeosinophil granulocytes. Electron microscopy showed the presence of many abnormal cells, whichprobably represented immature erythroblasts with giant lysosomes. These cells were often large,usually contained immature nuclei with relatively little condensed chromatin, had 1-20 electrondense cytoplasmic granules with an average diameter of 0-5 um, and regularly displayed substantialrhopheocytotic activity. A few abnormal cells and intermediate and late erythroblasts appeared tohave been phagocytosed by macrophages. The data indicate that dyserythropoiesis and ineffectiveerythropoiesis have a role in the pathogenesis of the anaemia of at least some cases of kala-azar.

Visceral leishmaniasis (kala-azar) is a chronic diseasecaused by the intracellular parasitic organism Leish-mania donovani. It is characterised by severe splen-omegaly and varying degrees of pancytopenia. Thepancytopenia seems to be mainly related to the splen-omegaly that results in increased pooling of bloodcells within the spleen, an increase in plasma volume(haemodilution), and some reduction in the life spanof blood cells (due to "hypersplenism"). Thus in alarge series of cases Cartwright et all showed that theseverity of pancytopenia correlated both with the sizeof the spleen and the duration of symptoms. In addi-tion, as in other conditions with substantial splen-omegaly, both an increase in plasma volume, and ashortening of the life span of red cells and plateletsand of the half life of circulating granulocytes hasbeen shown in kala-azar.2 5

Additional mechanisms have also been proposedfor the anaemia in kala-azar. These include dys-erythropoiesis and ineffective erythropoiesis, attrib-uted to the action of toxins generated by theorganisms, and haematinic deficiencies.6 Althoughseveral studies have shown the presence of com-

Accepted for publication 4 September 1986

plement components37 or IgG plus complement com-ponents on the red cells of patients with kala-azar,8there is no information as to the part played by theseimmunoproteins in the pathogenesis of the anaemia.

Little attention has been paid to the ultrastructureof the bone marrow in kala-azar; a survey of the pub-lished reports showed only electron microscopic dataon a single patient who was studied three days afterantimony treatment had been started.9 In view of thepossibility that marrow dysfunction may have somerole in the pathogenesis of the pancytopenia in kala-azar we made a detailed ultrastructural study of thebone marrows of one previously treated and two un-treated cases. The present paper describes ourfindings.

Material and methods

The study was carried out in 1981 during an outbreakcf kala-azar in the Kitui district of Kenya. The sub-jects were three children who were referred to theClinical Research Centre, Nairobi, with a diagnosis ofkala-azar. The splenic aspirates of all subjects werepositive for L donovani.Soon after admission the patients were examined

267

on May 13, 2020 by guest. P

rotected by copyright.http://jcp.bm

j.com/

J Clin P

athol: first published as 10.1136/jcp.40.3.267 on 1 March 1987. D

ownloaded from

http://jcp.bmj.com/

268and blood was taken into edetic acid and into a plaintube for haematological studies. Bone marrow was

aspirated from the posterior superior iliac crest.Smears were made at the bedside, and the remainderof the aspirate was mixed with 2 ml of Hanks's bal-anced salt solution containing 20 units of preservativefree heparin. Patients were treated with daily intra-muscular injections of sodium stibogluconate, andthe haematological studies were repeated two weeksafter treatment was started.

HAEMATOLOGICAL STUDIESA full blood count was obtained using a CoulterModel S electronic cell counter. Platelet counts were

performed using a Neubauer counting chamber. Se-rum and red cell folate concentrations were measuredby a microbiological assay based on chloramphenicolresistant Lactobacillus casei. Serum vitamin B12 andserum ferritin concentrations were determined usingthe Becton Dickinson B,2 radioassay kit (57Co) andthe Becton Dickinson ferritin (125I) radioimmuno-assay kit, respectively. The direct antiglobulin testwas performed using the microtitre plate method out-lined previously.'0 The reagents used were a specificrabbit antihuman-IgG, anti-C3, and anti-C4. Thesewere kindly supplied by Mr T Hunt of the BloodTransfusion Centre in Oxford. All reagents wereabsorbed to remove non-specific activity and tested

Wickramasinghe, Abdalla, Kasili

for specificity by using cells coated with IgG, C3, andC4, before use.The bone marrow smears were fixed in methanol

and stained using the May-Grunwald-Giemsa stainand Perls's stain for haemosiderin. A differential cellcount was carried out on 1000 consecutive nucleatedcells. In addition, the severity of dyserythropoiesiswas assessed by counting the numbers of cells thatshowed dyserythropoietic features among 200 eryth-roblasts. The dyserythropoietic features scored in-cluded cytoplasmic vacuolation or stippling,intererythroblastic cytoplasmic bridges, irregularlyshaped nuclei, karyorrhexis, binuclearity and multi-nuclearity, and intercellular chromatin bridges. Thenumber of individual intracellular and extracellular Ldonovani per 1000 nucleated bone marrow cells wasalso quantitated.

ELECTRON MICROSCOPIC STUDIESOnly the pretreatment marrow samples were used forultrastructural studies. Bone marrow fragments wereremoved from the mixture of marrow plus hepa-rinised Hanks's solution and fixed in 2-5% glu-taraldehyde in 0-1 M phosphate buffer (pH 7-3) atroom temperature. The fixed fragments were pro-cessed for transmission electron microscopy, as de-scribed by Wickramasinghe etal,'1 except that thesections were cut on a Reichert Om U4-Ultracut ul-tramicrotome.

Table I Essential clinical and haematologicalfeatures of three patients studied with kala-azar before (a) and after (b)leishmaniacidal treatment

Case I Case 2 Case 3

Variable (a)* (b) (a) (b) (a) (b)

Age (years) 10 7 9Sex M M MSpleen sizet 25 15 15Liver sizet 8 6 3

Blood count:Haemoglobin (g/dl)t1 7-0 5-8 79 85 70 6-8Red blood cells ( x 1012/1) 2-13 1-97 3-52 3-42 2-68 2-77Mean corpuscular volume (fl)$ 78 72 65 72 79 85Reticulocytes ( x 10'"/lD) 0-52 0-34 0-28 0-21 0-25 0-22White cell count ( x 10 /1) 44 7-3 14-7 3-4 3-6 4 5Neutrophils(x 109/1) 1.5 2-3 3-7 09 09 1-9Lymphocytes ( x 109/1) 2-7 4-7 10-0 2-1 2-3 1-8Monocytes (x 109l1) 0-2 0-4 1-0 0-4 0-3 0-8Eosinophils(x 10 /1) 0 0 0-03 0-04 0-1 0Platelets( x 109/1) 120 96 125

DAGT:Anti-IgG + + + + + +Anti-C3 - - + + + +Anti-C4 + + + + + +

Serum vitamin B (ng/l)tt 563 241 204Serum folate (gugh/tt 4-2 4-8 9-3Red cell folate (ug/l)tt 270 293 270Serum ferritin (pg/l)tt 2237 42 45

*Had been treated previously with three 30 day courses of stibogluconate (10 mg/kg/day); tcm below the mid point of the costal margin;$normal ranges for children between 4 and 10 years: Hb 9-2-15-6g/dl, MCV 72-97 fl 2 ; ttnormal ranges: serum vitamin B12 165-684ng/l,serum folate 2-1-28-0pg/l, red cell folate 160-640pg/l, serum ferritin 7-142 pg/I.



j.com/

J Clin P


ownloaded from

http://jcp.bmj.com/

Ultrastructure of bone marrow in patients with visceral leishmaniasisTable 2 Characteristics ofbone marrow smearsfrom three patients with kala-azar before (a) and after (b) leishmaniacidaltreatment

Case I Case 2 Case 3

Variable Normal range12 (a) (b) (a) (b) (a) (b)

Erythroblasts (%) 13-28 58 49 45 43 40 39Neutrophils and precursors (%) 45-77 32 34 35 38 39 34Eosinophils and precursors (%) 1-0 1-2 2-4 2-0 4-0 3-4Lymphocytes(%) 5-36 8-6 12-8 13-8 11-6 14-8 21.2Monocytes (%) 1-8 1.2 0-4 1-4 2-4 3-0Plasma cells(%) 0-2-06 0-4 1-2 1-6 3-6 2-8 2-0M:E ratio 1-2-5.2 0-55 0-70 0-78 0-88 0-98 085Organisms per 1000 cells 6 5 144 4 100 0Erythroblasts with dyserythropoietic features (%) 2-7's 30 7 20 8 17 8Sideroblasts (%) 20-90 51 50 15 32 22

Results

Table 1 summarises the main clinical and laboratorydata prior to and two weeks after the start of treat-ment. At presentation all three patients had massivesplenomegaly and moderately severe anaemia. Theabsolute reticulocyte counts were raised to betweenabout two and five times higher than normal. Plateletcounts were only slightly reduced. Mild neutropeniawas seen in one of the cases. There was little change inthe blood picture at the end of two weeks of treatment

-'with antimony.The pretreatment marrow fragments were grossly

hypercellular with a virtually complete absence of fatcells. Table 2 summarises data on the cellular com-position and other features of the marrow beforetreatment. All three patients showed reducedmyeloid:erythroid (M:E) ratios due to erythroid hy-perplasia, and increased numbers of megakaryocytes.They also showed an increased proportion of erythro-blasts with dyserythropoietic features. Most of themorphologically abnormal erythroblasts were earlyand late polychromatic cells; the most common eryth-roblast abnormalities were cytoplasmic stippling andirregularly shaped nuclei or karyorrhexis. Cyto-plasmic stippling- was seen in 6-14% and irregularlyshaped nuclei or karyorrhexis in 7-13% of theerythroblasts. There were 1-4% of binucleate ormultinucleate erythroblasts. Occasional giant meta-myelocytes and several megaloblasts were found incase 2. Stainable iron was present in the marrow frag-ments of all three cases. A noticeable reduction in theproportion of morphologically abnormal erythro-blasts occurred two weeks after the start of leish-maniacidal treatment (table 2).

Prior to treatment the number of intracellular andextracellular L donovani seen per 1000 nucleated mar-row cells was lowest in case 1 and highest in case 2.The prevalence of L donovani in the marrow smears,however, did not correlate with counts of the or-ganism in splenic aspirates or with the activity of thedisease, as judged by the spleen size.

ELECTRON MICROSCOPYMacrophages were prominent in the bone marrows ofall three patients. Amastigote forms of L donovaniwere found within a high proportion of the macro-phage profiles in cases 2 and 3 and in only a raremacrophage profile in case 1. The organisms seemedto be rounded or oval in outline and were limited bya trilaminar periplast consisting of an outer unitmembrane of host cell origin, a middle unit mem-brane of parasite origin, and an inner layer of sub-pellicular hollow microtubules. Other cellularstructures recognised included the nucleus, the ki-netoplast, a flagellar pocket, the flagellum, mito-chondria, a Golgi apparatus, multivesicular bodies,electron lucent vacuoles-sometimes containing vary-ing quantities of electron dense granular material,large rounded homogeneously electron dense inclu-sions, and lipid droplets. Between one and 11 or-ganisms were usually seen within a single macrophageprofile and were found singly or in groups of two orfour. In some sections, and particularly when theparasites were present in groups, the two unit mem-branes of the periplast, which are usually closely ap-posed to each other were separated to varying extentsby granular electron dense material.

Parasitic organisms, some of which showed variousdegrees of degeneration or degradation, were alsofound within occasional neutrophil and eosinophilgranulocytes (fig 1). The granulocytes containing theorganisms were usually located near a broken upparasitised macrophage. Furthermore, a few parasiteswere occasionally found extracellularly attached tothe cell membrane of the granulocytes, and these weresurrounded by a narrow fluffy rimlof residual macro-phage cytoplasm.A proportion of the profiles of intermediate and

late erythroblasts, particularly of cases I and 2, dis-played various ultrastructural abnormalities includ-ing irregularly shaped nuclei, abnormally long ormultiple intranuclear clefts, and binuclearity andmultinuclearity (figs 2a-c). Occasionally erythroblastprofiles contained abnormally large autophagic

269



j.com/

J Clin P


ownloaded from

http://jcp.bmj.com/


i

0

@ '~~~~~~~xt''tFig 1 Amastigoteforms ofLeishmania donovani within eosinophil granulocyte (a) and neutrophilgranulocyte (b). Both these granulocytes were near disrupted infected macrophage. Case 2 (a) x 16 500;(b) x 9700.

vacuoles (fig 2d). Erythroblast profiles in contact withheavily infected macrophages did not necessarilyshow ultrastructural abnormalities.

In cases 1, 2, and 3, respectively, 23-1, 10 8, and11 9% of the profiles of nucleated cells belonged to anabnormal cell type (fig 3). Such cells displayed moreor less rounded or oval outlines, were generally large,being up to 11 m in their long axis, and contained asingle large nucleus and moderate quantities of cyto-plasm. The outlines of the nuclei were roughly circu-lar or oval and, occasionally, quite irregular. Thenuclei usually contained only very small or smallquantities of heterochromatin and displayed promi-nent nucleoli. The cell membrane regularly showedevidence of substantial rhopheocytotic activity (fig 4).The electron density of the cytoplasmic matrix wasusually low, being similar to that in pronormoblasts,but was occasionally considerably higher (fig 3). Thecytoplasm contained many ribosomes, often arrangedas polysomes, several mitochondria, and a few toseveral strands of rough endoplasmic reticulum. Thecytoplasm of the abnormal cell profiles character-istically contained one to 20 (usually one to 10) largemembrane bound, moderately or strongly, electrondense granules, which were rounded or oval in outline

and which had average diameters of 0-2-0-8 pm(mean 05 pm). The granules were sometimes scat-tered throughout the cytoplasm but, more often, wereconcentrated in a fairly restricted area near the Golgiapparatus and centrioles. Some granules were ho-mogeneous in appearance; some showed one or morerounded areas of increased density in their centre; andothers contained a variable number of irregularlyshaped, more or less well defined electron lucent areas(fig 4). Occasional granules contained ferritin mole-cules that were either diffusely distributed or localisedto one region within the granule matrix. In additionto the large electron dense granules described above,some cell profiles displayed much smaller granules ofsimilar electron density. Several abnormal cells weresometimes associated with a single infected macro-phage and the erythroblasts surrounding it; the ab-normal cells were in intimate contact with themacrophage or its cytoplasmic processes, other ab-normal cells, or erythroblasts. Abnormal cells werealso found apparently unassociated with a macro-phage profile or associated with macrophage profilesnot including sectioned organisms. Some of the ab-normal cells interdigitated with adjacent normallooking erythroblasts in a manner similar to the inter-

270



j.com/

J Clin P


ownloaded from

http://jcp.bmj.com/

Ultrastructure of bone marrow in patients with visceral leishmaniasis

Fig 2 Examples ofultrastructural abnormalities affecting erythroblasts in kala-azar. (a) Binucleateerythroblast; (b) trinucleate erythroblast in which nuclear masses are stuck together; (c) erythroblast withirregularly shaped nucleus displaying three intranuclear clefts; (d) erythroblast with irregularly shaped nucleus,large intracytoplasmic granule, and large presumably autophagic intracytoplasmic vacuoles. Case 1. (a) x10950; (b) x 13250; (c) x 14000; (d) x 11650.

271



j.com/

J Clin P


ownloaded from

http://jcp.bmj.com/

272digitation observable between normal erythroblasts,and a few contained siderosomes. Occasional abnor-mal cells contained moderate or large quantities ofheterochromatin and moderately electron dense cyto-plasm and resembled intermediate or late erythro-blasts, except for the presence of the characteristiclarge cytoplasmic granules (fig 3b). The abnormalcells did not display evidence of phagocytic activity

Wickramasinghe, Abdalla, Kasiliand did not contain secondary lysosomes; nor didthey protrude cytoplasmic processes at their periph-ery.

In occasional sections of macrophages there werepresumably phagocytic vacuoles containing neu-trophils, extruded erythroblast nuclei, intermediate orlate erythroblasts, or cells containing granules of thetype described in the preceding paragraph (fig 5).

Fig 3 Examples ofabnormal cells containing granulesfound in bone marrows ofpatients with kala-azar. (a) Threelarge cells with nuclei containing small amounts ofheterochromatin andprominent nucleoli. Electron density ofcytoplasm oftwo upper cells is similar to that ofproerythroblast, and that of third cell (arrow) is considerablygreater. (b) Smaller cell with moderate quantities ofcondensed chromatin, moderately electron dense cytoplasm,and intranuclear cleft. Allfour cell profiles contained large electron dense granules and showed rhopheocytoticactivity. Case 1. (a) x 9200; (b) x 10000.

.,.-- .7k',

:1.4 ..w i

IAMl*'.



j.com/

J Clin P


ownloaded from

http://jcp.bmj.com/

Ultrastructure of bone marrow in patients with visceral I*- ;^ p g-_ %i A I.A St_. m . N .> ~ _ _r

ai t ^t t * g w} ^;>^ 4 s *t. ,, j , -*s', .gf ..*wO. id. W. tio... a.WX i5 >; - .f' r w ' 8 i VE S. fif \ * ;ffiZS e MgS +:4* sWSEYff P+ ........... i,,t,r . .r ...................... S At*;, s - 5 ;i\

3; > i. Xr <# '*.S,:. wf v ;e' *t, srY',Xt

E*~~~~~~~~~~4A ~ ~ ~ ~ ~ ~ ~

Fig4 Prominent rhopheocytotic activity (small arrows) in abnormal cell containing granules. Some profiles of

cytoplasmic us.

ton_ uet a Cs 2

cyolsi grnue (lrearw ipa lcrnluetaes ae2 350

tl470 d'";

Fig 5 Abnormal cell (large arrow) within cytoplasm ofmacrophage; exceptfor presence ofcytoplasmic granules,abnormal cell resembles intermediate erythroblast. Three organisms (curved arrows) are seen inside macrophage.Case 2. x 11850.

273leishmaniasis



j.com/

J Clin P


ownloaded from

http://jcp.bmj.com/

274

Discussion

In these three patients with kala-azar the distributionof intracellular organisms within the bone marrowwas similar to that reported previously in a single pa-tient studied three days after treatment with antimonyhad been started.9 Most of the parasites were situatedwithin macrophages, and a few were found insideneutrophil and eosinophil granulocytes. Most of theneutrophil and eosinophil granulocytes containingthe organisms were observed near disrupted infectedmacrophages, suggesting that the organisms locatedwithin granulocytes were derived from damaged mac-rophages. The damage to macrophages had probablyoccurred in vivo as a consequence of their invasion.The possibility cannot be excluded, however, that atleast some of the intragranulocytic parasites had beenreleased from macrophages which had been damagedduring the marrow aspiration and had entered thegranulocytes during the short period (less than 30minutes) between marrow aspiration and the fixationof marrow fragments for electron microscopy. Theultrastructural features of the organisms within boththe bone marrow macrophages and the bone marrowgranulocytes of our patients were similar to those de-scribed previously for organisms found within thesplenic macrophages of hamsters in vivo14 and laryn-geal macrophages of a patient from central Italy. 1 5

The physiological reserve of normal bone marrowis such that this tissue can increase the rate of deliveryof new red cells into the circulation (effective erythro-poiesis) by a factor of 6-8. If the absolute reticulocytecount is taken as a rough index of effective erythro-poiesis the three patients with kala-azar we studiedshowed only two to five-fold increases in the rate ofeffective erythropoiesis and thus seemed to havemounted a suboptimal marrow response despite per-sistent moderately severe anaemia. As the three pa-tients showed severe erythroid hyperplasia in themarrow and a substantial increase in the prevalenceof morphologically abnormal erythroblasts, it seemslikely that the suboptimal marrow response resultedfrom increased dyserythropoiesis and ineffectiveerythropoiesis. The possibility that erythropoiesis wasineffective was supported by the finding that occa-sional intermediate and late erythroblasts seemed tohave been phagocytosed by macrophages.Thus thehaematological data in our patients support the pre-viously published finding that although the anaemiaof kala-azar may be primarily a consequence ofsplenomegaly, dyserythropoiesis and ineffectiveerythropoiesis also play a part in its pathogenesis.6 Asthere was no evidence that the dyserythropoiesis wasrelated to iron, vitamin B12 or folate deficiency (table1) it seems likely that the abnormality of erythro-poiesis was some consequence of the infection. This


view is supported by our finding of improved erythro-blast morphology two weeks after antimony treat-ment had been started. Macrophages are in intimatecontact with haemopoietic cells and seem to par-ticipate in the generation of various substances whichinfluence haemopoiesis, such as colony simulatingfactors, erythropoietin, prostaglandin E, interferonsand interleukin I, either by producing them or by re-leasing factors which promote their production byother cells.6 -21 Infection of bone marrow macro-phages in visceral leishmaniasis may interfere with thegeneration of such substances and thereby impairhaemopoiesis. In addition, infected macrophages mayrelease cytotoxic factors, such as superoxide or para-site products, which may adversely effect hae-mopoietic cells. In view of the present light andelectron microscopic evidence, indicative of dys-erythropoiesis and increased ineffective erythro-poiesis in kala-azar, it would be of interest to make aproper quantitative assessment of total, effective, andineffective erythropoiesis in this disorder from de-tailed ferrokinetic and erythrokinetic studies. Itshould also be pointed out that absolute reticulocytecounts might provide an underestimate of the rate ofeffective erythropoiesis in our patients because of apreferential sequestration of reticulocytes within theenlarged spleen.An important finding of this study was that

11-23% of the nucleated marrow cells consisted of anabnormal cell type. The profiles of the abnormal cellswere often large, usually included immature lookingnuclei, and contained between one and 20 large elec-tron dense cytoplasmic granules of variable appear-ance. On morphological grounds we consider it to bevery unlikely that these large granular cells were natu-ral killer cells; T suppressor cells; or abnormally de-veloping early precursors of basophils, mast cells, orneutrophils. The abnormal cells found in our patientsdiffered from circulating human natural killer and Tsuppressor cells in that they were larger and had muchless heterochromatin, less rough endoplasmic reticu-lum, and a larger number of cytoplasmic granules.22The abnormal cells also differed from normal neu-trophil promyelocytes and early basophil precursorsin having little rough endoplasmic reticulum andshowing considerable rhopheocytotic activity.

Several of our findings strongly suggest that the ab-normal cells represented immature erythroblasts withgiant lysosomes. These findings are: (i) extensive rho-pheocytotic activity at the surface of the abnormalcells; (ii) the presence of ferritin molecules withinsome of the intracytoplasmic granules; (iii) the occur-rence of siderosomes in occasional abnormal cells;and (iv) the presence of large granules of the type seenin the abnormal cells in several cells that otherwiseresembled intermediate or late erythroblasts. The for-



j.com/

J Clin P


ownloaded from

http://jcp.bmj.com/

Ultrastructure of bone marrow in patients with visceral leishmaniasis 275mation of giant lysosomes may be a manifestation ofdamage to early erythroblasts, possibly caused by alocally generated or circulating parasite product ormacrophage product, and could well be associatedwith a high mortality among the affected cells. In thisrespect it is noteworthy that the prevalence of abnor-mal cells with large granules was greatest in case 1,with the highest percentage of mature erythroblastswith dyserythropoietic features, and that occasionalabnormal cells seemed to have been phagocytosed bymacrophages. Electron microscopic studies usinggold labelled cell lineage specific monoclonal anti-bodies and electron microscopic autoradiographicstudies of iron ("Fe) uptake would be required for adefinitive biochemically based identification of theabnormal cells.

This project was supported by a grant from the Rock-efeller Foundation. We thank Professor MutumaMugambi, director of the Kenya Medical ResearchInstitute, for giving us the opportunity to investigatepatients attending the Clinical Research Centre,Nairobi. We are grateful to Ms Madeleine Hughes forher invaluable assistance with the electron micro-scopic studies.References

1 Cartwright GE, Chung H-L, Chang A. Studies on the pan-cytopenia of kala-azar. Blood 1948;3:249-75.

2 Knight R, Woodruff AW, Pettitt LE. The mechanisms of anae-mia in kala-azar. A study of two patients. Trans R Soc TropMed Hyg 1967;61:701-5.

3 Woodruff AV, Topley E, Knight R, Downie Campbell GB. Theanaemia of kala-azar. Br J Haematol 1972;22:319-29.

4 Musumeci S, D'Agata A, Panebianco MG. Platelet andfibrinogen survival in kala-azar. Trans R Soc Trop Med Hyg1974;68:360-7.

5 Musumeci S, D'Agata A, Schiliro G, Fischer A. Studies of theneutropenia in kala-azar: results in two patients. Trans R SocTrop Med Hyg 1976;70:500-3.

6 Kasili EG. Haematological abnormalities in visceral leish-maniasis. East Afr Med J 1980;57:634-40.

7 Aikat BK, Mohanty D, Pathaniya AGS, et al. Haematological

investigations in kala-azar patients in Bihar. Indian J Med Res1979;70:571-82.

8 Abdalla SH, Kasili EG, Weatherall DJ. The Coombs direct anti-globulin test in Kenyans. Trans R Soc Trop Med Hyg 1983;77:99-102.

9 Inferrera C, Princi P. Leishmaniosi viscerale: reperti ultra-strutturali nelle cellule midollari. Arch De Vecchi Anal Patol1 967;49:757-75.

10 Abdalla SH, Weatherall DJ. The direct antiglobulin test in Pfalciparum malaria. Br J Haematol 1982;51:415-25.

11 Wickramasinghe SN, Hughes M, Hollan SR, Horanyi M,Szelenyi J. Electron microscope and high resolution auto-radiographic studies of the erythroblasts in haemoglobin Hdisease. Br J Haematol 1980;45:401-4.

12 Wickramasinghe SN. Blood and bone marrow. Systemic Pathol-ogy. 3rd ed. Vol 2. Edinburgh: Churchill Livingstone, 1986.

13 Abdalla SH, Wickramasinghe SN, Weatherall DJ. The deoxy-urine suppression test in severe anaemia following Plasmodiumfalciparum malaria. Trans R Soc Trop Med Hyg 1984;78:60-3.

14 Rudzinska MA, D'Alesandro PA, Trager W. The fine structureof Leishmania donovani and the role of the kinetoplast in theLeishmania-Leptomonad transformation. J Protozool 1964;11:166-91.

15 Garcovich A, Ormea F, Fabrizi G. Osservazioni ultrastrutturalisull'amastigote di Leishmania donovani in un caso di leish-maniosi laringea. Minerva Med 1975;66:2883-98.

16 Pelus LM, Broxmeyer HF, Kurland JI, Moore MAS. Regulationof macrophage and granulocyte proliferation. Specificities ofprostaglandin E and lactoferrin. J Exp Med 1979;150:277-92.

17 Rich IN, Heit W, Kubanek B. Extrarenal erythropoietin pro-duction by macrophages. Blood 1982;60:1007-18.

18 Kampschmidt RF. The numerous postulated biological mani-festations of interleukin-1. J Leukocyte Biol 1984;36:341-55.

19 Moore RN, Larsen HS, Horohov DW, Rouse BT. Endogenousregulation of macrophage proliferative expansion by colony-stimulating factor-induced interferon. Science 1984;223:178-81.

20 Nicola NA, Vadas M. Hemopoietic colony-stimulating factors.Immunology Today 1984;5:76-80.

21 Zoumbos NC, Djeu JY, Young NS. Interferon is the suppressorofhematopoiesis generated by stimulated lymphocytes in vitro.J Immunol 1984;133:769-74.

22 Grossi CE, Cadoni A, Zicca A, Leprini A, Ferrarini M. Largegranular lymphocytes in human peripheral blood: ultra-structural and cytochemical characterisation of the granules.Blood 1982;59:277-83.

Requests for reprints to: Professor SN Wickramasinghe,Department of Haematology, St Mary's Hospital MedicalSchool, London W2 IPG, England. on M

ay 13, 2020 by guest. Protected by copyright.

http://jcp.bmj.com

/J C

lin Pathol: first published as 10.1136/jcp.40.3.267 on 1 M

arch 1987. Dow

nloaded from

http://jcp.bmj.com/

Ultrastructure marrow leishmaniasis · with visceral leishmaniasis. The patients were moderately anaemic but showed a suboptimal in-crease in the absolute reticulocyte count. Serumandred

Documents