ABSTRACT
The impact of proviral DNA reverse transcriptase mutations on virological failure was evaluated in 50 HIV-1 HAART-treated patients switching from a protease inhibitor to a non-nucleoside reverse transcriptase inhibitor. Neither the M184I/V mutation detected in 12 patients nor stop codons at tryptophane positions detected in 13 patients were associated with virological failure. Stop codons appeared under successful therapy in 12 patients. Their presence should be assessed in studies with higher statistical power.
Subject(s)
Codon, Terminator , HIV Infections/virology , HIV Reverse Transcriptase/genetics , HIV-1/genetics , Proviruses/genetics , Reverse Transcriptase Inhibitors/therapeutic use , Antiretroviral Therapy, Highly Active , CD4 Lymphocyte Count , Cohort Studies , DNA, Viral/blood , HIV Infections/drug therapy , HIV Infections/immunology , Humans , Mutation , Treatment OutcomeABSTRACT
PURPOSE: To describe information bias due to missing data for hepatitis C (HCV) status in the analysis of factors associated with mortality in HIV-infected patients. METHOD: The prospective APROCO cohort enrolled 1,151 HIV-infected adults at the first initiation of highly active antiretroviral treatment in 1997-1998. Conversely to other characteristics, hepatitis B and C serologic status were recorded retrospectively. RESULTS: In a first dataset, HCV status was missing in 29%. HCV infection was associated with a higher hazard of death (Cox model, hazard ratio [HR]=4.1; 95% confidence interval [95% CI], 1.5-11.3). After more efforts to actively document HCV status, the information remained missing in only 10%. All deceased patients who were secondarily documented were recorded as being HCV negative. In fact, before systematic collection of HCV status, nonstructured additional documentation for all deaths led to spontaneous notification of HCV-positive serology at death and not HCV negative. HCV was no longer associated with the hazard of death (HR=1.2; 95% CI, 0.6-2.7). CONCLUSION: These results underline the need to minimize missing data and to investigate the impact of missing data on the results, although the mechanism of bias is difficult to identify. In addition, these results might shed light on the current debate about the association between HCV and progression of HIV infection.