BACKGROUND: HIV drug resistance surveillance is limited by recruitment and selection bias and by limited information regarding HIV incidence rates, secondary resistance, and treatment prevalence. METHODS: A second-generation HIV sentinel surveillance among men who have sex with men (MSM), regardless of prior history of HIV screening, serostatus, or treatment, was conducted in Peru in 2002. Recent HIV infection was estimated using sensitive/less sensitive enzyme immunoassay testing. Genotypic resistance testing was performed. RESULTS: HIV prevalence was 13.9% (456 HIV positive of 3280 participants). HIV incidence was estimated to be 5.1 per 100 person-years (95% confidence interval: 3.1-8.3). Among 143 MSM who were aware of their HIV infection before testing, only 20 (14.0%) were receiving antiretrovirals (ARV). Mutations conferring ARV resistance were found in 12 (3.3%) of 359 treatment-naive and 5 (31.3%) of 16 treatment-experienced participants with successful genotyping. One recently infected man from Lima demonstrated 3-class multidrug resistance. The most frequently observed mutations in treatment-naive, chronically infected persons from Lima were M184V (1.7%), D30N (1.3%), L90M (1.3%), and L10I (1.3%). CONCLUSIONS: The prevalence of ARV resistance among treatment-naive MSM in Peru is low, reflecting limited access to treatment before 2004, and contrasts with the history of ARV treatment in developed countries, where high levels of nucleoside reverse transcriptase inhibitor resistance occurred before introduction of highly active antiretroviral therapy. Linking ARV resistance and HIV sentinel surveillance in developing settings is feasible and should be considered in third-generation HIV sentinel surveillance programs. Copyright © 2006 by Lippincott Williams & Wilkins.