Kaplan–Meier survival curves showing the relationship between a positive CMV DNA value MK-1775 price in plasma at baseline and the different endpoints are shown in Figure 3. The HRs (with 95% CIs) associated

with each factor in the univariate and multivariate analyses are shown in Table 2. Age at baseline and CMV DNA were significantly associated with the development of CMV end-organ disease. Patients with a positive CMV DNA value (above 80 copies/mL) were 13 times more likely to develop the disease (HR 13.0). In the univariate analysis, IDU, age at baseline, CD4 cell count, use of HAART and CMV DNA were correlated with mortality. In the multivariate analysis, use of HAART was significantly associated with a decreased risk of death (HR 0.1), whereas, as expected, the risk of mortality increased with age (HR 1.4 per 10 years). Detectable CMV DNA at baseline was significantly associated with an increased risk of dying during the following year (HR 1.9). Only CMV DNA was significantly associated with the development of other ODs. The risk doubled in the BAY 57-1293 ic50 case of a positive value (HR 2.6). Use of HAART, in contrast, significantly decreased this risk (HR 0.4). Not only was the detection of CMV DNA at baseline significantly associated with the three endpoints, but there was a significant relationship between the CMV DNA value and the risk of CMV end-organ disease and death. The

higher the viral load, the greater the risk of CMV end-organ disease, and the risk was especially high for values of CMV DNA above

1000 copies/mL (HR 17.1; 95% CI 6.8–49.0; P<0.01). In the multivariate analysis, patients with CMV DNA values above 1000 copies/mL were 15 times more likely to develop CMV end-organ disease (HR 15.3; 95% CI 5.6–42.0; P<0.01). The risk of dying increased significantly above 1000 copies/mL (HR 2.5; 95% CI 1.3–4.8; P<0.01) and was associated, in the multivariate analysis, with a fourfold increase in risk (HR 3.9; 95% CI 1.9–8.0; P<0.01). We calculated the positive and negative predictive values at 6 months of a single measurement of CMV DNA. The negative predictive values for CMV end-organ disease enough and death, were excellent regardless of the viral load (99.5; 95% CI 99.0–99.9 and 96.8; 95% CI 95.5–98.0, respectively). The positive predictive values were low (5.9; 95% CI 2.4–9.8 and 8.5; 95% CI 4.2–12.3, respectively), but increased for viral loads above 1000 copies/mL (11.5; 95% CI 3.6–20.8 and 14.7; 95% CI 4.8–21.6, respectively). The objective of our study was to evaluate the clinical relevance of a detectable CMV DNA in the plasma of immunosuppressed HIV-infected patients, using an ultrasensitive PCR, in the HAART era. Our study shows that a single positive measurement of low CMV viraemia (using DNA PCR) is significantly associated not only with the development of CMV end-organ disease but also with other ODs and death.