OBJECTIVE: We evaluated assays for the measurement of acute phase protein levels in plasma for their usefulness to identify sensitively an inflammatory response to active cytomegalovirus CMV infection in HIV-infected patients. METHODS: Plasma samples were collected from 28 CMV-seropositive patients with advanced HIV-infection (CD4-cell count <200/microl) before commencement of antiretroviral therapy. Sensitivity, specificity, and area under receiver operating characteristic curve for the selected acute phase protein assays (haptoglobin, fibronectin, high-sensitivity C-reactive protein (hs-CRP), human interleukin-6, serum amyloid A (SAA), and human lipopolysacharide binding protein) were compared with results of a CMV-specific PCR assay. RESULTS: CMV viremia was detectable in 8/28 patients. Levels of SAA correlated well with those of hs-CRP (r' = 0.439, P = 0.019 (Spearman rank correlation)). Levels of SAA >3 mg/L discriminated with 100% sensitivity and 40% specificity between HIV-infected patients with and without active CMV infection. Sensitivity of fibronectin was 100% and specificity 15% at a threshold-value corresponding with the lower limit of normal values as defined by the manufacturer of the assay (>29 mg/dL). Levels of the other acute phase proteins evaluated did not correlate with detection of CMV-DNA in plasma. CONCLUSION: Increased levels of SAA indicate sensitively an inflammatory response to active CMV infection. Use of a CMV-specific virological assay is required to confirm the specificity of a high SAA-level but may be limited to samples with high SAA-levels. Hence, screening for increased levels of SAA in patients with advanced HIV-infection may allow early identification of active CMV infection.
OBJECTIVE: We evaluated assays for the measurement of acute phase protein levels in plasma for their usefulness to identify sensitively an inflammatory response to active cytomegalovirus CMV infection in HIV-infectedpatients. METHODS: Plasma samples were collected from 28 CMV-seropositive patients with advanced HIV-infection (CD4-cell count <200/microl) before commencement of antiretroviral therapy. Sensitivity, specificity, and area under receiver operating characteristic curve for the selected acute phase protein assays (haptoglobin, fibronectin, high-sensitivity C-reactive protein (hs-CRP), humaninterleukin-6, serum amyloid A (SAA), and human lipopolysacharide binding protein) were compared with results of a CMV-specific PCR assay. RESULTS:CMV viremia was detectable in 8/28 patients. Levels of SAA correlated well with those of hs-CRP (r' = 0.439, P = 0.019 (Spearman rank correlation)). Levels of SAA >3 mg/L discriminated with 100% sensitivity and 40% specificity between HIV-infectedpatients with and without active CMV infection. Sensitivity of fibronectin was 100% and specificity 15% at a threshold-value corresponding with the lower limit of normal values as defined by the manufacturer of the assay (>29 mg/dL). Levels of the other acute phase proteins evaluated did not correlate with detection of CMV-DNA in plasma. CONCLUSION: Increased levels of SAA indicate sensitively an inflammatory response to active CMV infection. Use of a CMV-specific virological assay is required to confirm the specificity of a high SAA-level but may be limited to samples with high SAA-levels. Hence, screening for increased levels of SAA in patients with advanced HIV-infection may allow early identification of active CMV infection.