Objectives: To estimate the prevalence of transmitted antiretroviral (ARV) drug resistance, and to assess whether resistance testing influences first-line ARV regimen selection.
Methods: Data on patients' characteristics were collected through questionnaires. ARV drug resistance was tested by genotypic methods and defined by Quest-Stanford classification rule. Physicians reported the intended and actual treatments and the factors considered in treatment selection.
Results: Two hundred and twenty-eight patients were included. The prevalence of ARV drug resistance was 12.1%, with 9.8% for non-nucleoside reverse transcriptase inhibitors (NNRTIs), 4.5% for nucleoside reverse transcriptase inhibitors and 1.8% for protease inhibitors (PIs). Pill burdens, dosing frequency and physicians' experience with regimens were the major factors considered in treatment selection. The intended and actual treatment differed for 73 and 44% of the patients with and without ARV drug resistance, respectively [odds ratio (95% confidence interval, CI)=3.6 (1.5-9.0), P=0.006]. NNRTI-based regimens were intended for 10 patients with resistance to NNRTIs; these patients were prescribed PI-based regimens after genotypic testing.
Conclusions: Transmitted ARV drug resistance was detected in 12.1% of treatment-naïve patients, with resistance to NNRTIs the most common. Resistance-testing results played a partial role in first-line treatment selection. However, resistance to NNRTIs pre-empted NNRTI use.