Continuous surveillance of HIV primary resistance mutations is highly important due to their potential clinical impact. All patients naïve to antiretrovirals who had > or =1 genotypic resistance testing at the Institute of Infectious Diseases (Brescia, Northern Italy) between 2001 and 2006 were analyzed. Primary resistance mutations were defined using epidemiological and clinical criteria. Mutations were interpreted using the Stanford University Algorithm. Logistic regression analysis was used to assess possible predictors of primary resistance mutations. Among 569 patients, 11% presented > or =1 mutation. Prevalence of primary resistance mutations to nucleoside reverse-transcriptase inhibitors (NRTI), non-nucleoside reverse-transcriptase inhibitors (NNRTI), and protease inhibitors (PI) was 6.3%, 6%, and 1.6%, respectively. The most frequent mutations to NRTI were substitutions at position 215 (215Y in 3 patients, and 215 revertants in 16), 41L (13), 219Q (12), and 210W (10). Among mutations to NNRTI, 103N was found in 21 patients, while 181C, 188L, and 190A/S in 8, 3, and 4 patients, respectively. Fifty-one patients (9%) had high-to-intermediate resistance to > or =1 antiretroviral drug before starting the treatment. Regarding the new generation drugs, nine patients had intermediate resistance to etravirine, five patients had intermediate resistance to tipranavir, while five, one, and seven patients had low resistance to etravirine, tipranavir, and darunavir. Homosexuals were more likely to harbor a virus with primary resistance mutations (OR:2.68; 95% CI:1.44-5.00; P = 0.002) while non-Italian nationality was protective (OR:0.38; 95% CI:0.17-0.86; P = 0.020). Prevalence of primary resistance mutations suggests that genotypic resistance testing should be performed before starting treatment in naïve patients in Italy, particularly when NNRTI are prescribed.