Enigmatic Kikuchi-Fujimoto Disease: A Comprehensive Review

Xavier Bosch, MD; Antonio Guilabert, MD; Rosa Miquel, MD; Elias Campo, MD


Am J Clin Pathol. 2004;122(1) 

In This Article

Cause and Pathogenesis

There is much speculation about the cause of KFD; a viral or autoimmune cause has been suggested. Some initial reports hinted at Yersinia enterocolitica and Toxoplasma gondii as possible causative agents of KFD, mainly on the basis of positive serologic test results. But subsequent studies failed to support these hypotheses. In addition, the histologic features of lymphadenitis associated with these microorganisms clearly differ from those of KFD23-25 ( Table 1 ).

The role of Epstein-Barr virus (EBV), as well as other viruses, in the pathogenesis of KFD remains controversial. Serologic tests including antibodies to EBV, cytomegalovirus, and a host of other viruses have consistently proven noncontributory.[1] A viral infection is, nevertheless, possible by virtue of clinical manifestations, as described by Unger and colleagues[2] (upper respiratory prodrome, atypical lymphocytosis, and lack of response to antibiotic therapy), and certain histopathologic features (ie, proliferation of immunoblasts, presence of necrotic zones localized to T-cell areas, expansion of the paracortex, and predominance of T cells as revealed by immunologic marker studies). However, no viral particles have been identified ultrastructurally. Histologic, ultrastructural, and immunohistochemical findings might support a hyperimmune reaction, perhaps to several organisms. It is possible that KFD might represent an exuberant T cell-mediated immune response in genetically susceptible people to a variety of nonspecific stimuli. Some HLA class II genes are more frequent in patients with KFD. In particular, the incidence of DPA1*01 and DPB1*0202 alleles is significantly higher in patients with KFD than in healthy control subjects. These genes are extremely rare or absent among Caucasians but relatively common among Asiatic people (eg, French, 0.4%; Italian, 0.8%; Korean, 9.9%; and Japanese, 4.5%). This might provide an admissible explanation about the aforementioned epidemiologic pattern.[37]

Among others, EBV and herpesviruses 6 and 8 have been suggested as potential causative agents of KFD. With regard to EBV, there are 2 studies-including 11 cases of KFD-that detected EBV by means of in situ hybridization for EBV-encoded RNA expression[19,26] and polymerase chain reaction-based methods (Epstein-Barr nuclear antigen-1 DNA).[19] However, there have been several other studies that, by using the same and other molecular pathology procedures (eg, Southern blot analysis) to localize the virus genome, have concluded that neither EBV nor herpesvirus 6 or herpesvirus 8 has a putative role in the pathogenesis of KFD. This conclusion is based on the facts that most cases were negative and that, if positive results were observed, the percentage of viral detection in control subjects also was augmented.[27,28,29,30,31,32,33]

Electron microscopic studies have identified tubular reticular structures in the cytoplasm of stimulated lymphocytes and histiocytes in patients with KFD.[1] Because these structures also have been noted within endothelial cells and lymphocytes of patients with systemic lupus erythematosus (SLE) and other autoimmune disorders, Imamura and coworkers[38] hypothesized that KFD might reflect a self-limited SLE-like autoimmune condition induced by virus-infected transformed lymphocytes. Yet the results of serologic studies testing antinuclear antibodies, rheumatoid factor, and other immunologic parameters consistently have been negative in these patients,[1] providing no support for an autoimmune nature of the disease. Nevertheless, as we will comment, the association between KFD and SLE has been reported with a frequency probably greater than that expected by chance alone.

Some physicochemical factors have been pointed out anecdotally as triggers that might lead to KFD. One case of a patient who underwent a pacemaker implantation 6 weeks before the onset of KFD has recently been described.[39] In addition, the simultaneous occurrence of KFD and silicone lymphadenopathy in an axillary lymph node of a patient with a leaking silicone breast implant was reported in 1996.[40] In this patient, a lymph node biopsy revealed silicone lymphadenopathy along with the classic morphologic and immunophenotypic features of KFD.[40]

Although the mechanism of cell death involved in KFD has not been studied extensively, some works have proposed that it is characterized by apoptosis.[41,42,43,44,45] The finding of nuclear debris, which is one of the characteristic features of KFD, might indicate cell death by apoptosis. By means of an in situ end-labeling procedure, Felgar et al[41] noticed that the lymphocytes within and surrounding the typical areas of necrosis in affected lymph nodes from patients with KFD had nuclear fragmentation, a typical feature of early apoptosis. Furthermore, it has been reported that CD8+ T lymphocytes seem to be the lymphocytes that undergo apoptosis.[41,43,44,45] T-cell-restricted intracellular antigen-1 (TIA-1) cytotoxic granules were detected within the cytoplasm of apoptotic bodies in KFD necrotizing lesions.[41] In double stainings, TIA-1+ lymphocytes were found to be CD8+ rather than CD4+.[43]

On the other hand, several immunohistochemical studies have established that the predominant proliferating cell in KFD lymph nodes is the CD8+ T lymphocyte.[16,17,18,46,47] To elucidate the apoptotic mechanism related to this condition and the type of cells involved, Ohshima and coworkers[45] studied perforin and Fas pathways as CD8+ T-cell cytotoxic mechanisms that could induce apoptosis in target cells in patients with KFD. According to their observations, proliferating CD8+ T cells might act as "killers" and "victims" in the apoptotic process via the Fas and perforin pathways. The high Fas/FasL (ligand) frequency among CD8+ T cells rather than CD4+ T cells would support this hypothesis. Interestingly, these authors found a high frequency of Fas/FasL in histiocytes. They also suggested that there is an activated CD8+ T-cell proliferating/dying functional balance that might be beneficial to eradicate the responsible agent.[45]

CD123bright plasmacytoid predendritic cells-the so-called plasmacytoid monocytes or plasmacytoid T cells-have been reported to be a striking histopathologic finding of KFD. Plasmacytoid cells represent a population composed mainly of lymphoid (and also myeloid) cells that at an immature stage produce type I interferon in response to viral infection and, depending on microenvironmental stimuli, differentiate into potent antigen-presenting cells. These cells would have a role in the pathogenesis of KFD via their migration from bone marrow to affected lymph nodes, where they might produce large amounts of type I interferon, thus promoting a T-helper 1 T-cell response and the aforementioned cytotoxic immune reaction.[20,48,49,50,51]

Monocyte and macrophage lineage cells have been proposed as enhancers of the apoptotic event. The existence of a Fas/FasL interaction in these cells might imply a histiocyte-dependent death of CD8+ T cells, which could amplify the background process. Abe and coworkers[52] recently demonstrated the presence of receptor-binding cancer antigen expressed on SISo cells (RCAS1) in the macrophages from lymph nodes of patients with KFD. RCAS1 seems to be an apoptosis-associated protein that induces apoptosis in activated T cells and erythroid progenitor cells. The authors suggested that the high presence of RCAS1 on macrophages might contribute to the aforementioned histiocyte-dependent CD8+ T-cell death.[52]

Serum concentrations of some inflammation mediators such as interferon (IFN)-γ, FasL, and interleukin-6 have been reported to be increased during the acute phase of KFD, returning to normal levels during the convalescent phase, thus raising the possibility that these cytokines could have a role in the pathogenesis of this condition.[53,54] Ohshima et al[55] noticed an abundance of cytokines and chemokines such as IFN-γ, interleukin-18, monokine induced by IFN-γ, and Cys-X-Cys chemokine IFN-γ-inducible protein-10 (which are related to the aforementioned perforin and Fas pathways) in lymphocytes and histiocytes in KFD lymph nodes. Nonspecific lymphadenitis cases were used as control cases. These immunologic pathways might, therefore, be closely involved in the apoptotic process of the disease.[55]


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: