Castleman's disease is a heterogeneous lymphoproliferative
disease group, the cause of which remains unknown.
Hypothesis for the causes of CD have included
autoimmunity, disorder cytokine production and infection
with the HHV-8. HHV-8 has been associated with the
multicentric form of CD in 25% of cases. As it occurs
rarely, our information on this disease is mostly based
upon retrospective studies with a limited number of patients
and case presentations.
CD is usually of unicentric hyaline vascular type. Keller
et al. described 81 cases. Of these, 91% was of
hyaline vascular variant and the disease was localized.
Localized CD is a self-limited process, curable with local
Eighty to ninety percent of the cases of multicentric
CD belong to plasma cell variant. Of 21 cases reported
by Chronowski et al. 9 were multicentric and among
them only one was histologically hyaline vascular type.
Herreda et al. reported 15 patients and among them 8
cases were with multicentric development and only one of
them was of hyaline vascular type. In our case, the most
striking laboratory finding was that he had pronounced
lymphocytosis (absolute lymphocyte number 20x109/l).
Incorporation of immunophenotypic features into the
diagnostic criteria is helpful in differentiating CLL
from other lymphoproliferative disorders. Lymphocytes
in B-CLL coexpress CD19, CD20, CD23, CD5, and a single immunoglobulin light chain, kappa or lambda.
CD10 (CALLA) expression is usually absent. Mantle cell
lymphoma is distinguished from CLL by absent or very
dim expression of CD23 (10). In our case lymphocytes
coexpress CD11c, CD19, CD22, CD23, FMC7, HLA DR,
kappa, and CD 20. Flow cytometric study did not support
CLL and other lymphoproliferative disorders.
In literature review, slight lymphocytosis was reported. Nevertheless, a CD case presenting with marked lymphocytosis
was not encountered.
Leukopenia and thrombocytopenia may be observed
in multicentric CD[3,7]. In bone marrow examination,
slight increase in plasma cells may be observed[3,4,11].
In our case, increase in plasma cells and lymphocytes were
not seen in bone marrow aspiration and biopsy evaluation.
Interleukin 6 (IL-6) has been implicated in the pathophysiology
of CD. The multicentric variant of CD is as-sociated with HHV-8 in many cases[12-14]. This virus
encodes a functional analogue of IL-6, providing further
evidence that this cytokine has a pivotal role in the disease. It causes B-cell proliferation resulting in hyperplastic
follicles and hence the enlarged lymph nodes. IL-6 also
increases secretion of vascular endothelial growth factor
(VEGF), causing angiogenesis and capillary proliferation
with endothelial hyperplasia. IL-6 is also responsible for
polarization of T lymphocytes to a Type 2 cytokine profile
leading to autoimmune phenomena including autoimmune
hemolytic anemia, antinuclear antibody (ANA) positivity
and elevation of IgE. IL-6 induces an acute phase reaction
comprising increases in ESR, C-reactive protein (CRP),
IgGs, serum fibrinogen, and serum Amyloid A Protein
(SAA). Increased SAA levels may result in AA Amyloidosis,
whilst hyperfibrinogenemia may play a role in
venous thrombosis and thrombotic. Finally, B-type symptomatology
is virtually always associated with increased
IL-6 levels[16,17]. It is suggested that high IL-6 levels
may account for some pathological characteristics of the
disease and clinical and laboratory findings[3,6,18,19].
Administration of recombinant IL-6 at pharmacological
doses may lead to slight lymphocytosis[20,21]. However, it is difficult to describe the marked lymphocytosis seen
in our case by the effect of IL-6 since the lymphocytosis
associated with IL-6 is mild. Lymphocyte infiltration not
having been detected in bone marrow may be explained
by focal involvement.
In the study of Bacon et al. bone marrow was
evaluated in 12 HHV-8 positive CD cases. In three cases,
lymphoid follicles similar to the lymph nodes in multicentric
CD were detected. They suggested that this is characteristic
for HHV-8 associated multicentric CD. Our case
was also HHV-8 positive. However, in the study of Bacon
et al. all cases were also HIV-positive and none of
the cases was lymphocytosis unlike our case. The cause
of marked lymphocytosis seen in our case could not be
In conclusion, Castleman's disease is a heterogeneous
disease in terms of histological, clinical and laboratory
findings. It should be borne in mind that patients may
have marked lymphocytosis as well as anemia, neutropenia
and thrombocytopenia. These cases can simulate
chronic lymphocytic leukemia. However, these cases may
be distinguished from CLL or other B cell lymphoproliferative