Importantly, the definition of treatment failure (virological, immunological

or clinical) was the strongest predictor of resistance. Immunological and clinical treatment failure were categorised as inclusion criteria because access to plasma HIV-1 RNA and CD4 quantification was irregular during the study period. The finding that immunological and clinical criteria were poor predictors of treatment failure attributable to resistance is important and has relevance for other resource-poor settings where access to VL testing may be limited. Our results are in agreement with recent Selleck MLN0128 data which also show that CD4 cell counts and clinical criteria do not accurately identify patients with virological treatment failure [17–19]. In this study, we estimated that the overall prevalence of resistance-associated mutations was 81% among the investigated Honduran patients,

who were selected on the basis of signs and symptoms of treatment failure. This finding agrees well with results from the United Kingdom (80% resistance) [10], United States (76% resistance) [11], and France (88% resistance) [12], in spite of the fact that the cART conditions in these countries are very different from those in Honduras. It is somewhat surprising that 19% of the 138 studied patients did not appear to harbour a resistant virus even though they had been selected as experiencing treatment failure. The absence of resistance in 19% of the patients indicates that adherence in these patients may have been AZD8055 ic50 too low to drive the development of resistance. However, and as discussed above, the significant association of resistance

with type of failure (virological, immunological or clinical) also demonstrates that it is difficult DNA Damage inhibitor to monitor HIV therapy without regular access to plasma HIV-1 quantification. Thus, patients with adequate adherence and low plasma HIV RNA levels may incorrectly have been perceived to have immunological or clinical treatment failure. We observed that the presence of genotypic resistance was positively correlated with years on therapy and the number of ART regimens used. When this study was carried out, access in Honduras to boosted PIs was very limited. The broad resistance to NRTIs, NNRTIs and unboosted PIs indicates that many of our study subjects were in need of salvage therapy with boosted PIs as well as entry and integrase inhibitors [3]. Improved access to these drugs is urgently needed for these and other heavily treatment-experienced Honduran patients. M184V and K103N were the most prevalent NRTI and NNRTI mutations in our study, at 62% and 30%, respectively. M184V causes high-level (>100-fold) resistance to lamivudine/emtricitabine and emerges rapidly in patients who receive lamivudine monotherapy [20]. It is also the first mutation to develop in patients receiving partially suppressive triple combination therapy including lamivudine [21–23].