Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.
Multicentric Castleman’s disease (CD) is an unusual B cell lymphoproliferative disorder in HIV infected individuals,1 and is caused by human herpesvirus 8.
A 41 year old white homosexual man presented in December 2000 with a 3 month history of weight loss, fevers, slowly progressive cervical lymphadenopathy, and oro-cutaneous Kaposi’s sarcoma. At this time he was found to be HIV-1 antibody positive; baseline investigations showed CD4 count of 120 cells × 106/l, HIV viral load 130 000 copies/ml, Hb = 8.4 g/dl, white blood count = 2.8 × 109/l, and platelet count 118 × 109/l. A bone marrow aspirate showed a hypercellular marrow with marked dysplasia of all three cell lines. These changes were ascribed to HIV infection. A chest radiograph was normal. Antiretroviral therapy with stavudine, lamivudine, and efavirenz, and anti-pneumocystis prophylaxis with co-trimoxazole was commenced.
Two weeks later, in January 2001, an excision biopsy of a cervical lymph node was performed, the histology of which showed Kaposi’s sarcoma and plasma cell variant CD. At this time the patient reported 2 weeks of slowly progressive exertional dyspnoea, intermittent fever, sweats and rigors, a cough productive of mucoid sputum, and a sensation of epigastric fullness/distension after eating. Examination revealed multiple rubbery lymph nodes (each ≤ 0.5 cm) in the submandibular, cervical chain, axillary and inguinal regions, and multiple lesions of Kaposi’s sarcoma on the trunk, limbs, and in the mouth sublingually. In the chest, there were bibasal end inspiratory crackles; Sao2 (on room air) = 98%. In the abdomen there was a 1 cm firm hepatosplenomegaly. Investigations showed persistent pancytopenia; urea and electrolytes and liver function tests were normal. Culture of blood, stool, and urine were negative for bacteria, mycobacteria, and fungi. A chest radiograph showed bilateral hilar lymphadenopathy and diffuse interstitial parenchymal changes. Abdominal ultrasound showed homogeneous hepatosplenomegaly. At bronchoscopy the tracheobronchial tree was normal; staining and culture of bronchoalveolar lavage fluid was negative for Pneumocystis carinii and other fungi, bacteria, mycobacteria, and viruses. A thoracic computed tomograph (CT) scan showed patchy ground glass shadowing and nodular thickening of the interlobular septae (fig 1).
Gastroscopy demonstrated Kaposi’s sarcoma in the lower oesophagus and duodenum. Open lung biopsy (right upper and lower lobes) showed CD only; staining and culture were negative for bacteria, mycobacteria, fungi, and viruses. The patient was slow to recover postoperatively. Persistent pancytopenia required blood transfusion. Increasing dyspnoea preceded a fatal hypoxic arrest, 15 days after operation. Necropsy was not performed.
Open lung biopsy was performed, given the discrepancy between the CT scan appearances of ground glass shadowing, suggesting P carinii or other opportunistic infection,2 and nodular thickening of interlobular septae, suggesting malignant infiltration with Kaposi’s sarcoma or lymphoma,3 and the bronchoscopic findings.
In contrast with this case, CD in non-HIV infected patients invariably shows hilar/mediastinal lymphadenopathy and centrilobular nodules on CT4,5; additional parenchymal abnormalities are rare. This case underscores the value of open lung biopsy in investigation of HIV infected patients with respiratory episodes and negative results from bronchoscopy.6 It also suggests that CD should be added to the differential diagnosis of CT appearances suggestive of malignant infiltration.
JB abstracted the patient’s clinical and pathological records and co-wrote the first draft of the manuscript; PJS reported the patient’s CT scan, wrote the figure legend, and commented on drafts of the manuscript; JDC commented on drafts of the manuscript; RFM proposed the project, co-wrote the first draft, incorporated comments from co-authors on drafts, and wrote the final version of the manuscript.
Source of funding: nil.
Conflict of interest: nil.