to the editor: The case of human herpesvirus 8 (HHV-8) reported by Dagna et al. (July 14 issue)1 is considered to be distinct from HHV-8–associated multicentric Castleman’s disease owing to the absence of typical histologic findings and elevated plasma levels of interleukin-6. The histopathology of HHV-8–associated multicentric Castleman’s disease is variable, as compared with the classic descriptions of HHV-8–negative Castleman’s disease2-4 and would encompass the changes described in the report, especially in the context of cytotoxic treatment before biopsy. In HHV-8–associated multicentric Castleman’s disease, the HHV-8– infected B-lineage cells can be seen either within the B-cell follicles or in the interstitium of many tissues, including bone marrow4,5 and lymph nodes, as seen in this case. Phenotypically, these cells resemble the plasmablasts in the follicles and the plasma cells in the interstitium. The infected cells secrete viral interleukin-6, which triggers an autocrine feedback loop involving human interleukin6,3 as reflected by the elevated plasma levels of interleukin-6, particularly at the time of the biopsy. The downstream effects of interleukin-6 are probably the cause of the clinical syndrome, which is indistinguishable from HHV-8–associated multicentric Castleman’s disease. We feel that this case report represents an HHV-8–associated multicentric Castleman’s disease, rather than a newly recognized HHV-8–associated inflammatory syndrome.