Witness for the prosecution: Structured treatment interruption (STI) has been considered among the possible therapeutic options for multidrug experienced HIV-infected individuals facing virological failure, since early studies suggested the possibility of viral reversion from drug-resistant mutated species to drug-sensitive wild-type virus following the release of drug pressure. However, the persistence of mutated viral populations during STI has been observed. The lack of detection of resistance mutations in plasma is mainly due to the low sensitivity of the currently available genotypic resistance testing, which is unable to detect resistant less fit virus masked by overwhelming re-emerged wild-type quasi-species. Restarting antiretroviral therapy (ART) is frequently followed by the reappearance of resistant virus, as well as by disease progression. STI should be applied with extreme caution in multidrug experienced HIV-infected individuals with virological failure. Use of genotype resistance testing in this particular setting, although useful in controlled clinical trials, should be considered of little help in clinical practice. Witness for the defence: The rationale for salvage STI has been elucidated. Data from observational and randomized studies have been revised, and the following conclusions in defence of salvage STI are: in more than 60% of patients undergoing salvage STI for at least 3 months there is a reversion of drug-associated mutations. Concurrently, increases in HIV-RNA and decreases in CD4 cells occur. Few clinical events have been reported in patients with CD4 counts of < 200 cells/mm3. New resistance-driven salvage regimens, including at least 1 new drug, result in good virological outcome in a short-term follow-up. Warnings are related to possible clinical progression and to the course of the infection in reservoirs. Randomized studies are required to elucidate these findings.