PNEUMOCYSTIS-CARINII PNEUMONIA IN NON HIV PATIENTS

OBJECTIVES: The diagnosis of Pneumocystis carinii pneumonia is made mo re and more frequently in patients free of human immunodeficiency viru s (HIV) infection. Mortality remains high despite diagnostic and the r apeutic advances. Risk factors, phophylaxis and factors of severity ar e less well known than in HIV-infected patients. PATIENTS AND METHODS: We studied retrospectively the medical files of 31 adult patients fre e of HIV-infection who were treated for Fl carinii pneumonia between J anuary 1991 and July 1997 in the Nantes University Hospital. RESULTS: All of the patients had at least one immunodepression factor: Malignan t hematology disease predominated (n=17). Other conditions included or gan grafts (n=6), solid cancers (n=5), and meningioma, disseminated lu pus erythematosus and Still's disease (n=1 for each). Ten patients had a history of lung disease. All patients were taking an immunosuppress ive treatment: 30/31 were on corticosteroids and 28/31 had corticoster oid therapy associated with another immunosuppressive treatment Three patients had corticosteroids alone and one patient had an immunosuppre ssive treatment without corticosteroids. Thirteen patients had radioth erapy (6 thoracic, 4 cerebral, 2 total body and 1 abdominal). Twenty-f our patients had lymphopenia, 17 had hypoalbuminemia, and 10 had renal failure. Hypogammaglobulinemia was found in 10/16 tested patients. El even of the patients died during the first month of the disease. No re currences were observed with or without prophylaxis. DISCUSSION: Excep ting certain conditions such as acute lympho-blastic leukemia, bone ma rrow graft or myeloma treated with high-dose dexamethasone where the f requency of Fl carinii pneumonia requires systematic prophylaxis, the risk factors remain poorly identified. Lymphopenia, hypoalbuminemia, r enal failure, hypogammaglobulinemia, radio-therapy, and prior lung dis ease would appear to aggravate immunodepression factors due to the ini tial disease and immunosuppressive therapy. A risk score could be esta blished using these factors to guide prophylaxis. (C) 1998, Masson, Pa ris.