CLINICAL IMPLICATIONS OF AUTOANTIBODIES IN HIV-INFECTION

Objective: To study the frequency and specificity of autoantibodies in HIV-infected subjects and their association with rheumatic manifestat ions, immunodeficiency, and prognosis. Design: Prospective study of se quentially selected HIV-infected patients. Indirect immunofluorescence reading was performed by two independent observers blinded for the pa tient diagnosis. Enzyme-linked immunosorbent assay (ELISA) was perform ed using coded serum samples. Setting: The study was performed at the Infectious Disease and Rheumatology Divisions of a tertiary care unive rsity hospital. Patients: One hundred sequentially selected HIV-infect ed patients formed group A. Controls included 80 non-HIV-infected high -risk individuals (group B), 20 herpesvirus-infected patients (group C ), and 30 healthy blood donors (group D). Main outcome measures: Patie nts were followed for 2 years and evaluated for the presence of immuno deficiency, rheumatic manifestations, circulating autoantibodies and t otal CD4+ cell count. Indirect immunofluorescence was used to investig ate antinuclear antibodies, antibodies to native DNA, smooth muscle, p arietal cell, glomeruli, thyroid, and neutrophil cytoplasm. Agglutinat ion was used to detect antibodies to erythrocytes and rheumatoid facto r. ELISA was used to determine antibodies to cardiolipin and denatured DNA. CD4+ lymphocytes were counted by flow cytometry. Immunoglobulin (Ig) G, IgM and IgA serum levels were determined by radial immunodiffu sion. Results: HIV-infected patients presented higher overall frequenc y of autoantibodies than the other groups. No difference was observed between immunodeficient and asymptomatic HIV-infected patients. The mo st frequent specificities were antibodies to cardiolipin and to denatu red DNA. Ig serum levels did not correlate with the occurrence of auto antibodies. The presence of autoantibodies was associated with lower C D4+ cell counts and with higher mortality within 2 years. Rheumatic ma nifestations were observed in 35 HIV-infected patients and were not as sociated with the occurrence of autoantibodies or the presence of immu nodeficiency. Conclusions: HIV infection is associated with an increas ed incidence of autoantibodies. Although not related to the occurrence of rheumatic manifestations, the presence of autoantibodies was signi ficantly associated with lower CD4+ lymphocyte counts and increased mo rtality, which implies prognostic significance to this phenomenon in t he context of HIV infection.