We aimed to assess whether oxidative stress is a predictor of mortality in HIV-infected patients. We conducted a nested case-control study in CoRIS, a contemporary, multicentre cohort of HIV-infected patients, antiretroviral-naïve at entry, launched in 2004. Cases were patients who died with available stored plasma samples collected. Two age and sex-matched controls for each case were selected. We measured F2-isoprostanes (F2-IsoPs) and malondialdehyde (MDA) plasma levels in the first blood sample obtained after cohort engagement. 54 cases and 93 controls were included. Median F2-IsoPs and MDA levels were significantly higher in cases than in controls. When adjustment was performed for age, HIV-transmission category, CD4 cell count and HIV viral load at cohort entry, and subclinical inflammation measured with highly-sensitive C-reactive protein (hsCRP), the association of F2-IsoPs with mortality remained significant (adjusted OR per 1 log10 increase, 2.34 1.23-4.47, P = 0.009). The association of MDA with mortality was attenuated after adjustment: adjusted OR (95% CI) per 1 log10 increase, 2.05 0.91-4.59, P = 0.080. Median hsCRP was also higher in cases, and it also proved to be an independent predictor of mortality in the adjusted analysis: OR (95% CI) per 1 log10 increase, 1.39 (1.01-1.91), P = 0.043; and OR (95% CI) per 1 log10 increase, 1.46 (1.07-1.99), P = 0.014, respectively, when adjustment included F2-IsoPs and MDA. Oxidative stress is a predictor of all-cause mortality in HIV-infected patients. For plasma F2-IsoPs, this association is independent of HIV-related factors and subclinical inflammation.