4
J Clin Pathol 1985;38:985-988 Splenic pathology in immune thrombocytopenia MM HAYES,* P JACOBS,J LUCILLE WOOD,t DM DENTf From the Departments of *Anatomical Pathology, tHaematology, and tSurgery, University of Cape Town and Groote Schuur Hospital, Cape Town, South Africa SUMMARY Splenic pathology was analysed in 73 patients with immune thrombocytopenic pur- pura who underwent splenectomy for bleeding that had been resistant to adrenocorticosteroids. The mean splenic weight was 100 g. The only notable macroscopic feature was the prominence of Malpighian corpuscles in 15 cases. Microscopic examination showed formation of germinal centres in the lymphoid tissue of the white pulp in 40 cases, prominence of the histiocytes in the red pulp in 18 cases, and infiltration with neutrophils in the same area in 49 cases. Myeloid metaplasia throughout the splenic tissue was minimal in 58 cases, moderate in 15, and extreme in two. No distinguishing features were found in the spleen from patients who had not received previous immunosuppressive treatment (n = 3), those treated with prednisone (1 mg/kg/day) for a median of 14 days (n = 62), or those who had received the same dose of prednisone and additional azathioprine or cyclophosphamide (2 mg/kg/day) for a median of four weeks (n = 8). No correlation could be shown between histological features and the age of the patient or titre of antiplatelet antibodies. Similarly, no distinguishing features were found in patients with associ- ated systemic lupus erythematosus (n = 8), hyperthyroidism (n = 6), immune haemolysis (n = 3), or recent viral illness (n = 3). The typical histological features of spleens removed from patients with idiopathic immune throm- bocytopenia are reported to be lymphoid hyperp- lasia with formation of germinal centres, infiltration of the splenic cords with granulocytes, and variable proliferation of histiocytes in the red pulp.'-4 In addition, splenic haematopoiesis of minimal degree has been described, although it is unusual for this to be a prominent feature; in rare cases it may reach a degree comparable with that seen in the myelo- proliferative syndrome.5 This study re-examined splenic pathology in patients with immune throm- bocytopenia, focusing on the range of myeloid metaplasia that may occur and seeking histological correlations with the previous administration of adrenocorticosteroids or other immunosuppressive drugs that are potential modifiers of splenic histol- ogy,6 the titre of antiplatelet antibody, the age of the patient, and the presence of underlying systemic disorders that may be associated with secondary immune thrombocytopenia. Material and methods The spleens from 73 patients diagnosed as having Accepted for publication 13 May 1985 immune thrombocytopenia were reviewed. All sub- jects fulfilled the criteria for this diagnosis7 and were managed according to a standard protocol whereby splenectomy was undertaken for symptomatic refractory thrombocytopenia after treatment with adrenocorticosteroids alone for a median of two weeks (group 1, n = 62); after treatment with adrenocorticosteroids combined with azathioprine for a median of four weeks (group 2, n = 8); or after all previous treatment had been refused (group 3, n = 3). Patients with associated lymphoproliferative disorders were excluded. Included were patients with systemic lupus erythematosus (n = 8), hyper- thyroidism (n = 6), immune haemolytic anaemia (n = 3), and recent viral illnesses (n = 3). Table 1 shows the distribution of the patients by age and sex. The clinical details and response to treatment of the 148 patients from whom the sub- group undergoing splenectomy were drawn are available from PJ.8 All spleens were weighed and examined macroscopically immediately after surgery and samples selected for histological exami- nation after fixation in buffered 10% formol saline. Sections were stained with haemotoxylin and eosin and, using a simple scoring system, were assessed for the degree of lymphoid hyperplasia, infiltration by granulocytes, histocytic proliferation, and splenic 985 copyright. on February 18, 2022 by guest. Protected by http://jcp.bmj.com/ J Clin Pathol: first published as 10.1136/jcp.38.9.985 on 1 September 1985. Downloaded from

Splenicpathology in immune thrombocytopenia

  • Upload
    others

  • View
    3

  • Download
    0

Embed Size (px)

Citation preview

J Clin Pathol 1985;38:985-988

Splenic pathology in immune thrombocytopeniaMM HAYES,* P JACOBS,J LUCILLE WOOD,t DM DENTf

From the Departments of *Anatomical Pathology, tHaematology, and tSurgery, University of Cape Townand Groote Schuur Hospital, Cape Town, South Africa

SUMMARY Splenic pathology was analysed in 73 patients with immune thrombocytopenic pur-pura who underwent splenectomy for bleeding that had been resistant to adrenocorticosteroids.The mean splenic weight was 100 g. The only notable macroscopic feature was the prominence ofMalpighian corpuscles in 15 cases. Microscopic examination showed formation of germinalcentres in the lymphoid tissue of the white pulp in 40 cases, prominence of the histiocytes in thered pulp in 18 cases, and infiltration with neutrophils in the same area in 49 cases. Myeloidmetaplasia throughout the splenic tissue was minimal in 58 cases, moderate in 15, and extreme intwo. No distinguishing features were found in the spleen from patients who had not receivedprevious immunosuppressive treatment (n = 3), those treated with prednisone (1 mg/kg/day) fora median of 14 days (n = 62), or those who had received the same dose of prednisone andadditional azathioprine or cyclophosphamide (2 mg/kg/day) for a median of four weeks (n = 8).No correlation could be shown between histological features and the age of the patient or titre ofantiplatelet antibodies. Similarly, no distinguishing features were found in patients with associ-ated systemic lupus erythematosus (n = 8), hyperthyroidism (n = 6), immune haemolysis(n = 3), or recent viral illness (n = 3).

The typical histological features of spleens removedfrom patients with idiopathic immune throm-bocytopenia are reported to be lymphoid hyperp-lasia with formation of germinal centres, infiltrationof the splenic cords with granulocytes, and variableproliferation of histiocytes in the red pulp.'-4 Inaddition, splenic haematopoiesis of minimal degreehas been described, although it is unusual for this tobe a prominent feature; in rare cases it may reach adegree comparable with that seen in the myelo-proliferative syndrome.5 This study re-examinedsplenic pathology in patients with immune throm-bocytopenia, focusing on the range of myeloidmetaplasia that may occur and seeking histologicalcorrelations with the previous administration ofadrenocorticosteroids or other immunosuppressivedrugs that are potential modifiers of splenic histol-ogy,6 the titre of antiplatelet antibody, the age of thepatient, and the presence of underlying systemicdisorders that may be associated with secondaryimmune thrombocytopenia.

Material and methods

The spleens from 73 patients diagnosed as having

Accepted for publication 13 May 1985

immune thrombocytopenia were reviewed. All sub-jects fulfilled the criteria for this diagnosis7 and weremanaged according to a standard protocol wherebysplenectomy was undertaken for symptomaticrefractory thrombocytopenia after treatment withadrenocorticosteroids alone for a median of twoweeks (group 1, n = 62); after treatment withadrenocorticosteroids combined with azathioprinefor a median of four weeks (group 2, n = 8); or afterall previous treatment had been refused (group 3, n= 3). Patients with associated lymphoproliferativedisorders were excluded. Included were patientswith systemic lupus erythematosus (n = 8), hyper-thyroidism (n = 6), immune haemolytic anaemia (n= 3), and recent viral illnesses (n = 3).Table 1 shows the distribution of the patients by

age and sex. The clinical details and response totreatment of the 148 patients from whom the sub-group undergoing splenectomy were drawn areavailable from PJ.8 All spleens were weighed andexamined macroscopically immediately aftersurgery and samples selected for histological exami-nation after fixation in buffered 10% formol saline.Sections were stained with haemotoxylin and eosinand, using a simple scoring system, were assessed forthe degree of lymphoid hyperplasia, infiltration bygranulocytes, histocytic proliferation, and splenic

985

copyright. on F

ebruary 18, 2022 by guest. Protected by

http://jcp.bmj.com

/J C

lin Pathol: first published as 10.1136/jcp.38.9.985 on 1 S

eptember 1985. D

ownloaded from

986

Table 1 Details ofpatients and methods oftreatment

Group No in Median age and sexgroup

MIF (years)

Group 1 (splenectomy alone) 3 0/3 33Group 2 (prednisone and

splenectomy) 62 15/47 34Group 3 (prednisone withimmunosuppressives andsplenectomy) 8 1/7 35

extramedullary haematopoiesis or myeloid metap-lasia.The presence of systemic lupus erythematosus

was confirmed serologically,89 hyperthyroidism aspreviously described,'0 and membrane associatedimmunoglobulin by established techniques."

Results

MACROSCOPIC APPEARANCEThe splenic weights varied from 5 to 470 g, with amean of 100 SD (40) g. The gross appearance wasnormal in 48 cases. Sixteen cases showed promi-nence of the Malpighian corpuscles, four were con-gested, and three had a "beefy" cut surface. No dif-ferences were apparent between the three treatmentgroups or in the spleens removed from patients withidiopathic, as opposed to secondary immune,thrombocytopenia.

HISTOLOGICAL FEATURESLymphoid tissue varied considerably in quantity, asdid the number of germinal centres in the whitepulp. The germinal centres were absent in 33

i. ' A

Fig. 1 Well delineated reactive germinal centre present insplenic white pulp from patient with immunethrombocytopenia. (Haematoxylin and eosin.) xI00.

Hayes, Jacobs, Wood, Dent

spleens, and common (17) to very common (one) inonly 17 cases (Fig. 1).Histiocytes were normally distributed in 18 spleens.Thirty seven cases showed aggregates of foamyhistiocytes, resembling lipogranulomas in or adja-cent to the white pulp (Fig. 2): these are a commonincidental finding in normal spleens. A moderate(13) to high (5) diffuse proliferation of histiocytes inthe red pulp was noted in 18 cases (Fig. 3).Neutrophils infiltrated the red pulp variably. Noincrease in numbers was noted in 24 cases. In 30cases granulocytes were seen to be concentrated inthe red pulp immediately surrounding the lymphoidtissue. 19 cases exhibited a more diffuse infiltrationthroughout the cords of the red pulp.Haematopoiesis was either absent (19 cases) orlimited to an occasional megakaryocyte or earlymyeloid cell in 39 cases. More extensive extramedul-lary haematopoiesis was present in 15 spleens (13with moderate evidence, two high), and in two thisreached the degree usually encountered in themyeloproliferative syndrome,'2 in which groups oferythroblasts, myeloid precursors, and megakaryo-cytes were present in the sinusoids (Fig. 4). Occa-sional haematopoietic cells were also noted in thesplenic cords.

CORRELATION WITH TREATMENTIn none of the three treatment groups were distinc-tive histological features present. Specifically,neither the two week course of adrenocorticos-teroids that preceded splenectomy (n = 62) nor thefour week course of adrenocorticosteroids combinedwith immunosuppressive agents (n = 8) produced

Fig. 2 Striking lipogranuloma located adjacent to splenicwhite pulp. (Haematoxylin and eosin.) x400.

copyright. on F

ebruary 18, 2022 by guest. Protected by

http://jcp.bmj.com

/J C

lin Pathol: first published as 10.1136/jcp.38.9.985 on 1 S

eptember 1985. D

ownloaded from

Splenic pathology in immune thrombocytopeniadiscernible differences in splenic pathology(Table 1).

EFFECT OF AGEThe distribution of age was not strikingly differentbetween the three groups, and no correlation wasrecognisable with any of the histological features.

INFLUENCE OF ANTIBODY TITREWhen measured, concentrations of antiplateletmembrane associated immunoglobulin were consis-tently raised in patients with immune throm-bocytopenia who failed to respond to adrenocor-ticosteroids with or without additional immunosup-pressive agents. Falling titres of this antiplateletantibody were associated with remission in clinicalbleeding and increased platelet counts, so that thesepatients would not have undergone splenectomy.Nevertheless, no correlation between splenic his-topathology and antibody titre could be found.

SECONDARY IMMUNE THROMBOCYTOPENIASpleens removed from patients with systemiclupus erythematosus, hyperthyroidism, associatedimmune haemolytic anaemia, or viral diseaseshowed histological features that were indistinguish-able from those of the spleens removed frompatients with idiopathic immune thrombocytopenia.

Discussion

Immune thrombocytopenia, whether idiopathic orsecondary to underlying disease but specifically

987

excluding lymphoproliferative disorders, was tre-ated initially with adrenocorticosteroids7 and inrefractory patients by splenectomy. In this series of73 patients spleens that had been removed becauseof immune thrombocytopenia resistant to drugtreatment showed considerable variability in his-tological change.Three points are germane to this discussion.

Firstly, the classically described features of prolifer-ation of lymphoid germinal centres and neutrophilicinfiltration in the spleens of patients with immunethrombocytopenia'l4 could not be shown consis-tently in the present study. Secondly, although dif-fuse proliferation of foamy histiocytes has beenreported to be a feature associated with a systemicresponse to treatment,'3 many of the spleens frompatients in the present study failed to show this fea-ture. The most likely explanation for this discordantobservation is the short time during which ourpatients received medical management withadrenocorticosteroids before splenectomy.

Finally, although the presence of extramedullaryhaematopoiesis in the spleen has been previouslyreported'4 the present survey indicates that suchmyeloid metaplasia is usually limited to the presenceof occasional megakaryocytes or metamyelocytes inthe sinusoids. The prevalence of moderate to highamounts of haematopoietic tissue in the spleen wasfar lower than has been reported before.3s In addi-tion, this tissue was situated mainly in the sinusoids,with only occasional cells visible in the splenic cords.Such a pattern could not be distinguished from thatwhich occurs in the myeloproliferative syndrome.

Fig. 4 Illustration ofextensive extramedullaryFig. 3 Photomicrograph illustrating diffuse proliferation haematopoiesis with prominent megakaryocyte andoffoamy histiocytes in spleen. (Haematoxylin and eosin.) erythroblasts present within sinusoids ofspleen in immunexboo. thrombocytopenia. (Haematoxylin and eosin.) x400.

copyright. on F

ebruary 18, 2022 by guest. Protected by

http://jcp.bmj.com

/J C

lin Pathol: first published as 10.1136/jcp.38.9.985 on 1 S

eptember 1985. D

ownloaded from

988

Our observation contradicts previous reports show-ing the haematopoietic tissue to be limited to theextravascular space of the splenic cords in patientswith immune thrombocytopenia, whereas inmyelofibrosis the haematopoietic cells were distri-buted equally between the cords and sinuses of thered pulp.5

This experience does not support the existence ofreliable diagnostic criteria for distinguishing the pat-tern of haematopoiesis in spleens removed frompatients with immune thrombocytopenia, myelo-proliferative syndrome, or other infiltrative lesionsof the bone marrow. Thus in view of the rarity ofextensive splenic haematopoiesis in immune throm-bocytopenia we recommend that such patients beinvestigated carefully to exclude the possibility ofcoincidental myeloproliferative syndrome or otherunderlying disorders.These findings confirm the presence of generally

minimal extramedullary haematopoiesis in thespleen of patients with immune thrombocytopeniaand emphasise the rarity with which this reaches a

degree comparable with that found in the myelop-roliferative syndrome. Furthermore, our observa-tion that splenic histopathology in patients withimmune thrombocytopenia, whether idiopathic or

secondary, is not modified by previous adrenocor-ticosteroid or immunosuppressive treatment is atvariance with earlier reports but may in part beexplained by the short period of medical manage-ment that our patients received before undergoingsplenectomy.

Finally, in patients with immune throm-bocytopenia no distinguishing features were presenton histopathological criteria that could be correlatedwith antiplatelet antibody titres or with systemiclupus erythematosus, hyperthyroidism, immunehaemolysis, or viral disease. In particular, theperivascular onion skin fibrosis and degenerativechanges in the capsular collagen described in sys-temic lupus erythematosus were not shown.'5

This work was financially assisted by the Universityof Cape Town Leukaemia Centre and StaffResearch Fund, the National Cancer Association,and the Medical Research Council.

Hayes, Jacobs, Wood, Dent

We thank Jackie Davies for typing the manuscriptand the medical superintendent of Groote SchuurHospital for permission to publish our findings.

References

Burke JS. Surgical pathology of the spleen: an approach to thedifferential diagnosis of splenic lymphomas and leukemias.Part II. Diseases of the splenic red pulp. Am J Surg Pathol1981;5:681-94.

2 Tavassoli M, McMillan R. Structure of the spleen in idiopathicthrombocytopenic purpura. Am J Clin Pathol 1975;64: 180-91.

Nickerson DA, Sunderland DA. The histopathology ofidiopathic thrombocytopenic purpura hemorrhagica. Am JPathol 1937; 13:463-90.

Enriquez P, Neiman RS. The pathology of the spleen. A func-tional approach. Chicago: American Society of ClinicalPathologists, 1976.

Yam LT, McMillan R, Tavassoli M, Crosby WH. Splenichemopoiesis in idiopathic thrombocytopenic purpura. Am JClin Pathol 1974;62:830-7.

6 Bowman HE, Pettit VD, Caldwell FT, Smith ED. Morphology ofthe spleen in idiopathic thrombocytopenic purpura. Lab Invest1955;4:206-16.

7 McMillan R. Chronic idiopathic thrombocytopenic purpura. NEnglJ Med 1981;304:1135-47.

NJacobs P, Wood I, Denf DN. Results of treatment in immunethrombocytopenia. J Med (In press).

Gewurz H, Suyehira LA. Complement. In: Rose NR, FriedmanH, eds. Manual of clinical immunology. Washington, DC:American Society for Microbiology, 1976:36-47.

'° Talal N, Pillarisetty R. Radioimmunoassay for antibodies todeoxyribonucleic acid. In: Rose NR, Friedman H, eds. Manualof clinical immunology. Washington, DC: American Societyfor Microbiology, 1976:652-9.

"Jacobs P, Majoos F, Perrotta A. Hyperthyroidism and immunethrombocytopenia. Postgrad Med J 1984; 60:657-61.

2 McMillan R, Smith RS, Longmire RL, Yelenosky R, Reid RT,Craddock CG. Immunoglobulins associated with humanplatelets. Blood 1971;37:316-22.

3 Pitcock JA, Reinhard EH, Justus BW, Mendelsohn RS. A clini-cal and pathological study of seventy cases of myelofibrosis.Ann Intern Med 1962;57:73-84.

'4 Saltzstein SL. Phospholipid accumulation in histiocytes of splenicpulp associated with thrombocytopenic purpura. Blood1961; 18: 73-88.

Soderstrom N, Bandmann U, Lundh B. Patho-anatomicalfeatures of the spleen and liver. In: Videbaek A, ed.Polycythaemia and myelofibrosis. London: WB Saunders,1975:309-29.

16 Anderson WAD, Kissane JM. Spleen in auto-immune diseases.Pathology, 7th ed. St Louis: CV Mosby, 1977:1504-6.

Requests for reprints to: Professor P Jacobs, Departmentof Haematology Research Centre, University of CapeTown Medical School, Anzio Road, Observatory 7925,Cape Town, South Africa.

copyright. on F

ebruary 18, 2022 by guest. Protected by

http://jcp.bmj.com

/J C

lin Pathol: first published as 10.1136/jcp.38.9.985 on 1 S

eptember 1985. D

ownloaded from