Bayoumi AM, Goia C, Gardner S, Major C, Millson M, Robinson G, Remis RS, Rachlis A; International Conference on AIDS.
Int Conf AIDS. 2002 Jul 7-12; 14: abstract no. TuPeC4829.
Ontario HIV Treatment Network, Toronto, Canada
BACKGROUND: HIV-specific interventions are sometimes less effective in practice than in clinical trials. We examined whether a matched-cohort study could be used to evaluate the usefulness of genotypic resistance testing (GRT) in a non-experimental setting. METHODS: We selected cases from the HIV Ontario Observational Database, a voluntary longitudinal cohort study. Cases had complete antiretroviral histories on file and had received GRT from Ontario's Public Health Laboratory (n=180). We selected controls by matching cases' cumulative number of previously used antiretroviral medications (within 1), date of viral load (VL) at time of GRT (within 3 months), and VL level (within 1 log10 copy/mL). We evaluated VL outcomes 3 to 9 months after GRT using McNemar's test and conditional logistic regression. RESULTS: We found 176 (98%) matched controls. Although VL level was a matching variable, cases had slightly higher VL levels at the time of GRT (median difference 0.23 log10 copies/mL, p=0.004). Cases and controls had similar CD4 counts (median difference -21, p=0.49) and prior use of protease inhibitor containing regimens at baseline (93% vs. 88%, p=0.18). Follow-up VL was available for only 142 pairs (81%) and was undetectable for 20% of cases and 24% of controls, yielding an odds ratio (OR) of 0.83 (95% confidence interval [CI] 0.46 to 1.47). VL decreased by at least 1 log10 copy/mL in 28% of cases and 30% of controls (OR=0.94, 95% CI 0.55 to 1.59). After adjusting for CD4 count and prior use of protease inhibitors, the OR of attaining undetectable VL was not significantly different for cases and controls, but the confidence intervals were wide (0.65, 95% CI 0.30 to 1.45). CONCLUSIONS: A matched-cohort study matched most cases but residual confounding and sample size issues may have limited our analysis. Our results indicate that GRT may be less effective in practice than in clinical trials, but this finding requires validation in other observational settings.
Publication Types:
Keywords:
- Acquired Immunodeficiency Syndrome
- Anti-HIV Agents
- CD4 Lymphocyte Count
- Case-Control Studies
- Cohort Studies
- HIV
- HIV Infections
- HIV Protease Inhibitors
- HIV Seropositivity
- Longitudinal Studies
- Ontario
- Research Design
- Viral Load
Other ID:
UI: 102254215
From Meeting Abstracts