Search tips
Search criteria 


Logo of jclinpathJournal of Clinical PathologyVisit this articleSubmit a manuscriptReceive email alertsContact usBMJ
J Clin Pathol. 2007 July; 60(7): 839–840.
PMCID: PMC1995774

“In situ” detection of human cytomegalovirus infection of bone marrow in a patient previously treated for B‐prolymphocytic leukaemia

Infection of immunosuppressed patients by human cytomegalovirus (CMV) may be responsible for a wide range of symptoms, depending on the affected organ. Iatrogenic immunosuppression is an important contributing factor secondary to organ transplantation or to treatment for haematological malignancies. In these conditions appropriate diagnosis of infection by CMV has practical relevance for the management of the patients.1 Detection of CMV in bone marrow by the pathologist has been reported as a rare event, although pancytopenia may be indicative of such infection.2,3,4,5 The purpose of this report is to present a case of a patient who had been treated for a B‐cell prolymphocytic leukaemia five years before the onset of a relapse of the disease accompanied by CMV infection of bone marrow. It is intended to stress the possibility of this rare feature that can be missed by the histopathologist at routine diagnosis.

Case report

We report the case of a woman in her 70s in whom a B‐cell prolymphocytic leukaemia had been diagnosed. She was successfully treated with fludarabine and Endoxan (cyclophosphamide). In 2005, five years after initial diagnosis, she was referred to our hospital to investigate a recent weight loss, accompanied by fever and pancytopenia. The possibility of Richter's syndrome was suspected, but a bone marrow trephine biopsy was performed to rule out the diagnosis of tuberculosis, for which the patient had been treated recently.

Materials and methods

The bone marrow specimen was fixed in Duboscq‐Brazil (alcoholic based Bouin), decalcified in 10% EDTA, and embedded in paraffin. Histological sections stained with H&E were reviewed. Immunohistochemical staining was performed using a Ventana immunostainer (Ventana‐Biotech, Tucson, Arizona, USA). The panel included antibodies to CD20, CD79a, CD3, CD5, CD10, CD21, CD23, cyclin D1, and to an immediate‐early (IE‐1) CMV antigen. The marker for CMV was included because morphology showed suspicious atypical cells (see Discussion). A streptavidin–biotin–peroxidase detection system and staining with diaminobenzidine was used.


At morphological evaluation, the bone marrow was moderately infiltrated by lymphoma cells (fig 11).). A few large atypical cells suspicious of viral infection could also be seen, but no typical cytological feature of viral infection was evidenced. Immunophenotyping on paraffin sections confirmed the diagnosis of bone marrow infiltration by the same lymphoid neoplasia, most cells being positive for CD20 and CD79a; reactive T‐lymphocytes expressing CD3 and CD5 were sparse. Neoplastic cells did not express CD5, CD10, CD21, CD23 or cyclin D1. CMV infection was confirmed by immunoreactivity in rare atypical cells in areas not infiltrated by the neoplasm (fig 22).). Furthermore, CMV infection was confirmed by serology and PCR. During investigation, hepatopathy was diagnosed, interpreted as superimposition of CMV and hepatitis C virus (HCV) infection. The patient was treated with antiviral therapy, foscarnet and ganciclovir, but the clinical course was unfavourable: she developed encephalopathy, pneumopathy and renal failure, dying few weeks after onset of therapy.

figure cp33936.f1
Figure 1 Hypercellular area of bone marrow, infiltrated by atypical small lymphoid cells, diagnosed as B‐prolymphocytic leukaemia (H&E, 400×).
figure cp33936.f2
Figure 2 Area with normal cellularity, where an infected CMV‐positive cell can be seen (immunoperoxidase, anti‐CMV, 400×; inset 1600×).


Immunophenotyping of bone marrow allowed the confirmation of both the diagnosis of relapse of the B‐cell prolymphocytic leukaemia and of CMV infection. The marker for CMV was included because it is known that: CMV infection may be also responsible for fever and pancytopenia; this infection must always be suspected in immunocompromised patients; and the virus may be detected in the analysis of precursor blood cells. This diagnostic procedure providing evidence for a CMV infection, was important for adequate treatment planning.2 Careful analysis of blood films are also reported to allow the recognition of infected cells, which may be confused with neoplastic cells, as often happens in histopathology.4 In addition, examination of the nature of lymphoid infiltrates by immunohistochemistry may provide an important tool for the differentiation between immunological reaction to CMV infection and infiltration of bone marrow by malignant lymphoma, as has already been discussed elsewhere.6 In the present case, the neoplastic nature of the lymphoid infiltrate of bone marrow was given first morphologically, by the density and irregularity of lymphoid nodules, and the polymorphism of the nuclei. The neoplastic nature of this infiltrate was confirmed by the striking predominance of CD20 and CD79a expression by the lymphoma cells.

The present case illustrates a rare immunoexpression of CMV antigen in atypical cells infiltrating the bone marrow along with lymphoma infiltration, emphasising the role of the pathologist in this diagnosis. It also stresses the importance of an early diagnosis of this viral infection, for establishment of proper treatment, especially in a patient who is debilitated because of the underlying lymphoproliferative neoplasm. The fact that therapy has not been successful does not diminish the importance for the pathologist to establish the correct diagnosis as early as possible. The difficulty in treating CMV infection in such patients is recognised.1 However, a simple and rapid method, such as immunohistochemistry, may be useful in diagnosing CMV infection in addition to neoplastic infiltration using the same specimen, allowing prompt initiation of antiviral therapy, which may increase the chance of a successful outcome.


JV is a researcher at the Conselho Nacional de Pesquisa Cientifica (CNPq, Brazil).


Funding: Financial support was provided by Fundação de Amparo à Pesquisa de São Paulo (FAPESP, Brazil), Fundo de Apoio ao Ensino e Pesquisa (FAEP‐UNICAMP, Brazil) and the University of Toulouse (France)

Competing interests: None declared.


1. Gandhi M K, Khanna R. Human cytomegalovirus: clinical aspects, immune regulation, and emerging treatments. Lancet Infect Dis 2004. 4725–738.738 [PubMed]
2. Penchansky L, Krause J R. Identification of cytomegalovirus in bone marrow biopsy. South Med J 1979. 72500–501.501 [PubMed]
3. Abramowitz A, Livni N, Morag A. et al An immunoperoxidase study of cytomegalovirus mononucleosis. Arch Pathol Lab Med 1982. 106115–118.118 [PubMed]
4. Pooley R J, Jr, Peterson L, Finn W G. et al Cytomegalovirus‐infected cells in routinely prepared peripheral blood films of immunosuppressed patients. Am J Clin Pathol 1999. 112108–112.112 [PubMed]
5. Raddehase M J, Dreher‐Stumpp L, Angele P. et al Hematopoietic stem cell deficiency resulting from cytomegalovirus infection of bone marrow stroma. Ann Hematol 1992. 64(Suppl)A125–A127.A127 [PubMed]
6. Magalhães S M, Duarte F B, Vassallo J. et al Multiple lymphoid nodules in bone marrow biopsy in immunocompetent patient with cytomegalovirus infection: an immunohistochemical analysis. Rev Soc Bras Med Trop 2001. 34365–368.368 [PubMed]

Articles from Journal of Clinical Pathology are provided here courtesy of BMJ Publishing Group