Cytologic evaluation of primary and secondary myelodysplastic syndromes in the dog

Myelodysplastic syndromes are a heterogeneous group of acquired primary and secondary alterations of hematopoietic stem cells that result in cytopenias in blood and cytologic features of dysplasia in blood and/or bone marrow. To better understand the cytologic features that would permit differentiation of primary and secondary forms of myelodysplasia, we reviewed 267 consecutive bone marrow reports from dogs. These reports indicated that 34 dogs (12.7%) had dysgranulopoiesis, dyserythropoiesis, and/or dysthrombopoiesis in >10% of granulopoietic cells, erythroid cells, and/or megakaryocytes, respectively. Thirteen dogs had primary myelodysplastic syndromes, and 21 had secondary myelodysplastic syndromes. Of the 13 dogs with primary myelodysplasia, 4 were subclassified as myelodysplastic syndrome with refractory anemia (MDS-RA), and 9 were subclassified as myelodysplastic syndrome with excess blasts (MDS-EB). Secondary conditions associated with dysplasia in the bone marrow included malignant lymphoma (n = 5), myelofibrosis (n = 3), immune-mediated thrombocytopenia (n = 4), immune-mediated hemolytic anemia (n = 5), multiple myeloma with melphalan administration (n = 1), pyometra with estrogen administration (n = 1), polycythemia vera (n = 1), and thrombopathia (n = 1). MDS-RA was characterized by <5% myeloblasts in bone marrow, normal granulocyte maturation ratio, increased erythroid maturation ratio, and dysplastic changes in >15% of erythroid cells. MSD-EB was characterized by >/=5% myeloblasts in bone marrow, high granulocyte maturation and erythroid maturation ratios, >/=32% dysplastic granulocytes, and the presence of small atypical immature myeloid cells. Secondary myelodysplastic syndromes were characterized by <5% myeloblasts in bone marrow, variable granulocyte maturation and erythroid maturation ratios, and variable dysplastic features. These results indicate that morphology alone cannot be used to distinguish primary and secondary myelodysplastic syndromes in dogs.     

Recognition and classification of dysmyelopoiesis in the dog: a review

Dysmyelopoiesis is defined as a hematologic disorder characterized by the presence of cytopenias in the blood and dysplastic cells in one or more hematologic cell lines in the blood or bone marrow. The causes of dysmyelopoiesis include acquired mutations in hematopoietic stem cells (i.e., myelodysplastic syndromes [MDSs]), congenital defects in hematopoiesis, and dysmyelopoietic conditions associated with various disease processes, drug treatments, or toxin exposure. Two major subtypes of MDSs (i.e., MDS with refractory cytopenias and MDS with excess myeloblasts) have been described that differ in clinical presentation, response to treatment, and survival time. The most frequently occurring causes of secondary dysmyelopoiesis include immune-mediated hematologic diseases, lymphoid malignancies, and exposure to chemotherapeutic drugs. Differentiation of the various causes of dysmyelopoiesis is essential for establishing an appropriate therapeutic plan and for determining prognosis.     

Blood smear from a cat: features to "dys"cover

A 4-year-old, spayed female, domestic shorthair cat was presented for lethargy, nonregenerative anemia, and inappetence. Results of a CBC included macrocytic, normochromic, nonregenerative anemia and a glucocorticoid-associated leukogram. On blood smear examination, neutrophils had abnormal features including hyposegmentation and a diffuse chromatin pattern with nuclear filament formation and nuclear blebbing. Microscopic examination of a roll preparation of bone marrow revealed hypolobulated megakaryocytes with asynchronous maturation of nuclei. The granulocytic to erythrocyte (G:E) ratio was 76. Segmented neutrophils had asynchronous maturation and dysplastic features. The entire erythroid lineage was markedly decreased for the degree of anemia and rare dysplastic features were noted in erythroid precursor cells. The interpretation of bone marrow findings was erythroid hypoplasia, megakaryocytic dysplasia, and granulocytic hyperplasia with dysplasia. Histopathologic examination of a bone marrow core sample also revealed myeloid hyperplasia and erythroid hypoplasia. The result of a direct immunofluorescence assay for FeLV performed on the bone marrow roll preparation was positive. A diagnosis of dysmyelopoiesis associated with FeLV infection was made. This case was unique in that the dysplastic changes occurred in cell lines that did not have associated cytopenias. The dysmyelopoiesis most closely resembled myelodysplastic syndrome with refractory cytopenia (MDS-RC); however, secondary dysmyelopoiesis could not be ruled out.     

Primary myelodysplastic syndromes of dogs: a report of 12 cases

Clinical bone marrow specimens submitted to the University of Minnesota's Veterinary Teaching Hospital Cytology Service over a 3-year period were evaluated for the presence of myelodysplastic features. Of 220 bone marrow specimens examined, 30 contained dysplastic features. Twenty-seven of these dogs were evaluated further. Twelve were categorized as primary myelodysplastic syndromes, and 15 were categorized as secondary myelodysplastic syndromes. Of the primary myelodysplastic syndromes, 4 were subcategorized as refractory anemia and 8 were categorized as myelodysplasia. Primary refractory anemia was characterized by nonregenerative anemia without leukopenia or thrombocytopenia and with prolonged survival. Primary myelodysplasia was characterized by pancytopenia, greater than 5% myeloblasts in bone marrow, dysplastic features in all bone marrow cell lines, and short survival time. Results of this study indicate that differentiating primary refractory anemia from primary myelodysplasia has both therapeutic and prognostic significance. Dogs with primary refractory anemia tend to have prolonged survival and respond to erythropoietin treatment, whereas dogs with primary myelodysplasia have short survival and do not respond to standard treatments.     

A hematological study on thirteen cats with myelodysplastic syndrome

Hematological abnormalities were investigated in 13 cats with myelodysplastic syndrome (MDS). Examination of the peripheral blood samples from the 13 cats revealed anemia in 11 cats, leukopenia in 9 cats, and thrombocytopenia in 9 cats. Four cats had pancytopenia (30.8%) and 9 cats had bicytopenia (69.2%). Dysplastic changes of erythrocytes, neutrophils, and platelets in the peripheral blood were found in 5, 10 and 8 cats, respectively. Bone marrow examination of the 13 cats revealed that ratios of blast cells to all nucleated cells (ANC) ranged from 0 to 20%. Ratios of erythroid progenitor cells to ANC were more than 50% in 3 cats and less than 50% in 10 cats. Eosinophils accounted for more than 5% of non-erythroid cells in 10 cats. Dysplastic changes in the granurocytic, erythrocytic, and megakaryocytic cells in the bone marrow were found in 11, 7 and 5 cats, respectively. Dysplastic changes in these cats included giant neutrophils, ring-nucleated neutrophils, binuclear myelocytes, hypersegmented and hyposegmented neutrophils, megaloblastoid erythroblasts, multinucleated erythroblasts, micromegakaryocytes, and segmented multinucleated megakaryocytes. Virological examination indicated the presence of feline leukemia virus antigen in the peripheral blood from all of the 13 cats with MDS. The peripheral blood cytopenias and dysplastic changes in each blood cell lineage in the bone marrow were shown to be important for the diagnosis of MDS in cats.     

Long-term case study of myelodysplastic syndrome in a dog

A 10-year-old, female shih tzu was diagnosed as having myelodysplastic syndrome (MDS) based on the presence of a nonregenerative anemia, dysplastic changes in the three hematopoietic cell lines, a normal to hypercellular bone marrow, and less than 30% blast cells of all nucleated cells in the bone marrow. Low-dose aclarubicin, a differentiation-induction therapy for MDS and atypical leukemias in humans, was administered. Hematological improvement was observed, and the dog lived for 809 days after the first presentation.     

Differentiating benign and malignant causes of lymphocytosis in feline bone marrow

Differentiation of benign and malignant causes of lymphocytosis in blood or bone marrow can be problematic. In the present study, reports of examinations of bone marrow from cats, submitted over an 8-year period, were reviewed to identify cats with increased numbers of small lymphocytes. Of 203 reports reviewed, 12 (5.9%) indicated increased numbers of small lymphocytes. Diagnoses for these cats included chronic lymphocytic leukemia (CLL; n = 2), pure red cell aplasia (PRCA; n = 4), immune-mediated hemolytic anemia (IMHA; n = 3), thymoma (n = 1), cholangiohepatitis (n = 1), and fever of unknown origin (n = 1). Several factors were identified that could be used to differentiate reactive lymphocytosis from CLL. Cats with CLL tended to be older, and lymphocytes were slightly larger and had cleaved or lobulated nuclei. Reactive lymphocytosis was associated with immune-mediated anemias and inflammatory diseases. In reactive lymphocytosis, the proliferating lymphocytes were organized into lymphoid aggregates in bone marrow and were predominately B cells. Alternatively, in CLL and thymoma, the proliferating lymphocytes were diffusely distributed and were predominately T cells. Therefore, differentiation of the causes of lymphocytosis should include evaluation of signalment, concurrent disease conditions, lymphocyte morphology, lymphocyte distribution in bone marrow, and immunophenotype. Cat age, presence of severe anemia, and evidence of inflammatory disease also should be considered.     

Dysmyelopoiesis in the cat: a hematological disorder resembling refractory anemia with excess blasts in man.

Clinical and pathological findings in three cats affected with a myelodysplastic disorder are presented. This hematological disorder resembles that of refractory anemia with excess of blasts as seen in man. The hematological profile in man is one of peripheral cytopenia in one or all of the marrow cell lines which occurs despite a normal to hypercellular bone marrow. Quantitatively, the marrow has a preponderance of blasts (up to 20%). Qualitative abnormalities consist of dysthrombopoiesis and/or dyserythropoiesis and/or dysgranulopoiesis. Myelodysplastic disorders in the cat are a form of marrow failure often associated with infection with feline leukemia virus. The use of the term refractory anemia with excess of blasts appears to be applicable to the cat and should be considered in evaluating dysplastic disorders of the feline bone marrow.     

A retrospective study of the incidence and the classification of bone marrow disorders in the dog at a veterinary teaching hospital (1996-2004)

BACKGROUND: An 8-year retrospective study was conducted to evaluate the prevalence and the classification of canine bone marrow disorders in a clinical pathology service at a university referral hospital. ANIMALS: Dogs evaluated for bone marrow disorders at a veterinary teaching hospital. HYPOTHESIS: A better understanding of the spectrum and the prevalence of canine bone marrow disorders can be achieved with a multiyear retrospective study. METHODS: Bone marrow aspirate smears, core biopsy specimens, and case records from 717 dogs were reviewed. RESULTS: Bone marrow specimens were first categorized based on the presence or the absence of a primary bone marrow disorder. Nondysplastic and nonmalignant pathologic changes were placed into 14 subcategories. Frequently observed pathologic disorders included nonregenerative immune-mediated anemia, pure red cell aplasia, bone marrow necrosis, myelofibrosis, and hemophagocytic syndrome. Dysmyelopoiesis (n = 61) was subcategorized into myelodysplastic syndromes (n = 27), and congenital (n = 1) and secondary (n = 33) dysmyelopoiesis. One hundred twenty-six cases of neoplasia were divided into acute leukemia (n = 46), chronic leukemia (n = 7), stage 5 malignant lymphoma (n = 28), multiple myeloma (n = 25), malignant histiocytosis (n = 11), metastatic mast-cell tumor (n = 3), sarcoma (n = 5), and carcinoma (n = 1). CONCLUSIONS AND CLINICAL IMPORTANCE: This study provides a general indication of the spectrum and the prevalence of canine bone marrow disorders at a referral center in North America.     

Prognostic markers for myeloid neoplasms: a comparative review of the literature and goals for future investigation

Myeloid neoplasms include cancers associated with both rapid (acute myeloid leukemias) and gradual (myelodysplastic syndromes and myeloproliferative neoplasms) disease progression. Percentage of blast cells in marrow is used to separate acute (rapid) from chronic (gradual) and is the most consistently applied prognostic marker in veterinary medicine. However, since there is marked variation in tumor progression within groups, there is a need for more complex schemes to stratify animals into specific risk groups. In people with acute myeloid leukemia (AML), pretreatment karyotyping and molecular genetic analysis have greater utility as prognostic markers than morphologic and immunologic phenotypes. Karyotyping is not available as a prognostic marker for AML in dogs and cats, but progress in molecular genetics has created optimism about the eventual ability of veterinarians to discern conditions potentially responsive to medical intervention. In people with myelodysplastic syndromes (MDS), detailed prognostic scoring systems have been devised that use various combinations of blast cell percentage, hematocrit, platelet counts, unilineal versus multilineal cytopenias and dysplasia, karyotype, gender, age, immunophenotype, transfusion dependence, and colony-forming assays. Predictors of outcome for animals with MDS have been limited to blast cell percentage, anemia versus multilineal cytopenias, and morphologic phenotype. Prognostic markers for myeloproliferative neoplasms (eg, polycythemia vera, essential thrombocythemia) include clinical and hematological factors and in people also include cytogenetics and molecular genetics. Validation of prognostic markers for myeloid neoplasms in animals has been thwarted by the lack of a large case series that requires cooperation across institutions and veterinary specialties. Future progress requires overcoming these barriers.     

A dog with myelodysplastic syndrome: chronic myelomonocytic leukemia

An eleven-year-old female pug was referred to Yamaguchi University Animal Hospital for evaluation of anemia and thrombocytopenia. The cytological examination of the peripheral blood showed some giant monocytic lineage blast cells. A few granulocytes and platelets had dysplastic features. On day 7, in addition to increasing the monocytic lineage cells, the dysplastic features of the blood had also increased compared to the initial examination. We performed bone marrow aspiration upon her death. The bone marrow revealed dysplastic features in all three hematopoietic cell lines, and an increase in the monocytic cell line. Based on the features of the bone marrow and the peripheral blood, this case was confirmed to be myelodysplastic syndrome--Chronic myelomonocytic leukaemia (MDS-CMML).     

Clinical and lymphohematologic responses after bone marrow transplantation in sibling and unrelated donor-recipient pairs of cats

Conditions necessary for establishment of a graft, posttransplant supportive care and complications, and lymphohematopoietic reconstitution after bone marrow transplantation were evaluated in 7 cats. Donor-recipient pairs were selected on the basis of low mutual reactivity in one-way mixed lymphocyte reactions. Before transplantation, cats were given marrow ablative (7 Gray) total-body gamma irradiation. Cyclosporine A was administered to cat 7, which was given marrow from an unrelated donor. Rapid hematologic recovery was attained in 5 of 5 (cats 1 to 5) sibling bone marrow recipients and 1 (cat 7; cyclosporine A-treated) of 2 recipients from unrelated donors. Lymphocyte recovery was prolonged, requiring up to 100 days to attain reference concentrations. Lymphocyte blastogenic responses were below reference range in 2 of 3 cats (cats 1 and 3) examined approximately 1 to 3 months after transplantation. Serum IgG concentrations determined 1 to 6 months after transplantation were within reference range in cats 1 to 5 which were given sibling bone marrow. Fatal infections did not develop in cats that had established grafts. Antimicrobial-responsive fevers did develop, but were generally detected only when granulocyte counts were low (less than 1 x 10(9) cells/L). Clinical signs of disease in the immediate posttransplant period consisted of hepatic lipidosis (fatal) in cat 4, hepatitis (mild graft-vs-host disease) in cat 3, and immune-mediated hemolytic anemia and thrombocytopenia in cat 7. Cats with hepatitis and immune-mediated disease responded to immunosuppressive therapy.     

Non-lymphoid hematopoietic neoplasia in cats: a retrospective study of 60 cases

Sixty cats with hematologic abnormalities indicative of non-lymphoid hematopoietic neoplasia were classified into two groups, myelodysplastic syndromes (MDS) and acute myelogenous leukemias (AML), using criteria developed for human patients with similar diseases. Cats with myeloblast counts in bone marrow of less than 30% were classed as MDS and cats with myeloblast counts of 30% or greater were classed as AML. The clinical, laboratory, and postmortem findings in each group were described and compared. Clinical signs of disease were similar in both groups, the most common being inappetance, lethargy, and weakness. Non-regenerative anemia, macrocytosis, neutropenia, and thrombocytopenia were frequent hemogram abnormalities in both groups. Diagnostically useful differences in physical and peripheral blood findings were a higher prevalence of splenomegaly and/or hepatomegaly, thrombocytopenia, and severe anemia in the AML group. Circulating myeloblasts were found only in cats in the AML group. Outcome of disease was similar in both groups; 85% of the cats in each group died or were euthanatized within one week of diagnosis. In cats that were necropsied, extramedullary leukemic infiltrates were found in all cats in the AML group and in none of the cats in the MDS group.     

Coombs’, haemoplasma and retrovirus testing in feline anaemia

Objective : To investigate the associations between Coombs’ testing, haemoplasma and retroviral infections, and feline anaemia. Methods : Haematology, Coombs’ testing (including assessment of persistent autoagglutination) and selected infection testing (haemoplasma, feline leukaemia virus/feline immunodeficiency virus provirus) were performed in blood samples collected from 60 anaemic and 60 non-anaemic cats. Results : No association between infection and anaemia or Coombs’ positivity existed. Anaemic cats (21.7%) were significantly more likely than non-anaemic cats (0%) to have cold autoagglutination (P<0.0001), but significance (set at ≤0.0025 due to multiple testing) was not quite reached when Coombs’ positivity was compared between anaemic (40.4% and 21.7% positive at 4°C and 37°C, respectively) and non-anaemic (20% and 3.3% positive, P=0.021 and P=0.004, at 4°C and 37°C, respectively) cats. Cats with immune-mediated haemolytic anaemia were significantly more likely to have persistent cold autoagglutination (P<0.0001) and be Coombs’ positive at 37°C with polyvalent (P<0.0001), immunoglobulin (Ig)G (P<0.0001) or any antiserum (P<0.0001). Haemoplasmas and retroviruses were uncommonly detected. Clinical Significance : Cats suspected of having immune-mediated haemolytic anaemia should be evaluated for persistent autoagglutination at 4°C as well as performing Coombs’ testing at 37°C, but positive results may occur in with other forms of anaemia. Testing for erythrocyte-bound antibodies should always be interpreted in parallel with documentation of haemolysis in anaemic cats.     

Role of Latent Feline Leukemia Virus Infection in Nonregenerative Cytopenias of Cats

Background: Nonregenerative cytopenias such as nonregenerative anemia, neutropenia, and thrombocytopenia in cats with feline leukemia virus (FeLV) antigen are assumed to be caused by the underlying FeLV infection. In addition, cats with negative FeLV antigen-test results that have cytopenias of unknown etiology often are suspected to suffer from latent FeLV infection that is responsible for the nonregenerative cytopenias.
Objective: The purpose of this study was to assess the role of latent FeLV infection by polymerase chain reaction (PCR) in bone marrow of cats with nonregenerative cytopenias that had negative FeLV antigen test results in blood.
Animals: Thirty-seven cats were included in the patient group. Inclusion criteria were (1) nonregenerative cytopenia of unknown origin and (2) negative FeLV antigen test result. Antigenemia was determined by detection of free FeLV p27 antigen by ELISA in serum. Furthermore, 7 cats with positive antigen test results with nonregenerative cytopenia were included as control group I, and 30 cats with negative antigen test results without nonregenerative cytopenia were included as control group II.
Methods: Whole blood and bone marrow samples were tested by 2 different PCR assays detecting sequences of the envelope or long terminal repeat genes. FeLV immunohistochemistry was performed in bone marrow samples.
Results: Two of the 37 cats (5.4%) in the patient group were positive on the bone marrow PCR results and thus were latently infected with FeLV.
Conclusions and Clinical Importance: The findings of this study suggest that FeLV latency is rare in cats with nonregenerative cytopenias.     

Suppression of feline bone marrow fibroblast colony-forming units by feline leukemia virus

Bone marrow fibroblast colony-forming units (CFU-F) were evaluated in cats experimentally infected with feline leukemia virus (FeLV). Cats that developed persistent viral infection and anemia (progressor cats) had a progressive decrease in the number of CFU-F at 2, 4, 6, 8, and 10 weeks after inoculation with FeLV. This suppression of CFU-F number in progressor cats ranged from 16 to 44% of the preinoculation CFU-F value. Cats that did not develop persistent viral infection or anemia (regressor cats) had decreased numbers of CFU-F (24% of the preinoculation CFU-F value) at 2 weeks after inoculation, but normal CFU-F numbers at 4, 6, 8, and 10 weeks after inoculation. In vitro incubation of bone marrow mononuclear cells from healthy cats with the 15,000-dalton envelope protein of FeLV resulted in decreased number of CFU-F (21% of that of untreated cultures). The number of CFU-F from bone marrow mononuclear cells incubated with the 27,000-dalton core protein of FeLV was similar to that from untreated cultures.     

Chronic myelomonocytic leukemia in a cat

A 1-year-old spayed domestic short-haired cat was referred with anorexia and weight loss. Hematologic findings indicated nonregenerative anemia, severe neutropenia and monocytosis. The feline leukemia virus (FeLV) antigen test was positive reaction by enzyme-linked immunosorbent assay. Dysgranulopoiesis with slight increase in blast cells were observed in bone marrow smears. On the basis of blood and bone marrow findings, the cat was diagnosed as chronic myelomonocytic leukemia (CMMoL), which possibly corresponds to a kind of the subtypes in human myelodysplastic syndrome (MDS).     

Hemophagocytosis and Histoplasma‐like fungal infection in 32 cats

Histoplasmosis is one of the most common systemic fungal diseases in cats from the United States. It commonly causes respiratory or disseminated disease and is often associated with one or more cytopenias. Here, we describe 32 cats in which a Histoplasma‐like fungal infection was associated with concurrent hemophagia in at least one sample site, commonly spleen, bone marrow, liver, and/or lymph node. The degree of hemophagia was characterized as moderate or marked in the majority of cases, and in all cases, there was a predominance of phagocytized mature erythrocytes. A few cases also had macrophages with phagocytized erythroid precursors, platelets, and/or neutrophils. Complete blood count results were available for 25 cats, and cytopenias were common (20/25), including solitary anemia (10), anemia and thrombocytopenia (5), solitary neutropenia (2), pancytopenia (2), and anemia and neutropenia (1). Bone marrow samples were only available in a small subset of cases, preventing the further assessment of the causes of the cytopenias. Hemophagocytosis has been previously reported in cats with neoplastic diseases and a cat with calicivirus infection, and likely occurs with other conditions as well, such as hemorrhage or hemolysis. Results of this report suggest that systemic fungal disease is an additional differential to consider when there is hemophagia in a feline cytology sample.     

New insights into the physiology and treatment of acquired myelodysplastic syndromes and aplastic pancytopenia.

MDS are a diverse group of primary and secondary bone marrow disorders that are characterized by cytopenias in blood, prominent dysplastic features in blood or bone marrow, and normal or hypercellular bone marrow. MDS in cats are typically associated with FeLV infection. Dogs with MDS-RC and MDS-Er seem to respond to erythropoietin administration and have prolonged survival. Dogs with MDS-EB respond poorly to present treatments, and survival is short. Prognosis and probability of progression to acute myelogenous leukemia can be predicted based on the percentage of myeloblasts in bone marrow. Several experimental therapeutic modalities in human beings have been described that may be useful in treating MDS-EB in dogs and cats. Aplastic pancytopenia is a relatively rare disorder in dogs and cats. Causes include Ehrlichia spp, Parvovirus, and FeLV infections; sepsis; chronic renal failure; drug and toxin exposure; and idiopathic causes. Diagnosis is based on identification of multiple cytopenias in the blood and hypoplastic/aplastic bone marrow, with the marrow space replaced by adipose tissue. Treatment and outcome are dependent on determining the underlying cause of the bone marrow failure.     

Cytologic evaluation of benign and malignant hemophagocytic disorders in canine bone marrow

Abstract: Canine hemophagocytic disorders were studied to better understand the cytologic features that differentiate benign and malignant disease. Of 286 canine clinical bone marrow reports evaluated retrospectively, 13 (4.5%) noted at least 3% hemophagocytic macrophages. Macrophages comprised between 6% and 44% of nucleated bone marrow cells. Clinical diagnoses for dogs with hemophagocytic disorders included malignant histiocytosis (n = 2), myelodysplastic syndromes (n = 4), round cell neoplasia (n = 2), immune-mediated disorders (n = 2), and idiopathic hemophagocytic syndrome (n = 3). Differentiation of benign and malignant forms of histiocytosis was problematic. Two dogs with a diagnosis of hemophagocytic syndrome had macrophages with atypical features similar to those described for malignant histiocytosis. Furthermore, only 2 of 11 dogs with presumably benign hemophagocytic disorders had exclusively mature macrophages in bone marrow. Other dogs had variable numbers of large reticular-type cells characterized by lacy chromatin, anisocytosis, anisokaryosis, and prominent and/or multiple nucleoli. On the basis of these results, cytomorphologic evaluation of bone marrow alone may not be adequate to consistently differentiate benign and malignant forms of hemophagocytic disorders.