The pol gene sequences and recommended subtype reference
sequences (http://www.hiv.lanl.gov) were also used to construct neighbour-joining phylogenetic trees using the mega 4 software . Phylogenetic tree analysis was used to determine the subtype of the pol gene sequence and to facilitate detection of MAPK inhibitor possible PCR contamination and sample mix-up. The prevalence of drug resistance mutations was calculated with a 95% confidence interval (CI) based on the binomial distribution. Univariable and multivariable logistic regression analyses were used to estimate odds ratios (ORs) with 95% CIs for the association between resistance status and various factors. Statistical analyses were carried out using statistica version 8.0 (StatSoft Inc., Tulsa, OK, selleck chemicals USA) and stata version 8.2 (StataCorp LP, College Station, TX, USA). A total of 138 HIV-1-infected individuals with treatment failure were included in the study (Table 2); 97 were adults and 41 were children under 18 years of age. A little more than half of the study subjects were male (57%). The median age was 38 years for the adults
and 10 years for the children. All individuals were native Hondurans; the most frequent route of transmission was heterosexual transmission (66%), followed by mother-to-child transmission (28%), blood products (3%) and homosexual transmission (3%). The severity of disease according to the Florfenicol Centers for Disease Control and Prevention (CDC) Classification System  was stage A for 12 patients, stage B for 60 patients and stage C for 66 patients. Adherence was scored as good in 99 patients (72%), intermediate in 26 patients (19%) and poor in 13 patients (9%). The median time on antiretroviral therapy was 3.2 years (range 1–12 years). The median CD4 count was 185 cells/μL
and the median VL was 4.5 log10 copies/mL (Table 2). The median time span between sampling for the resistance test and the last available CD4 cell count was 5 months (range 0–36 months); one patient had never undergone CD4 cell count measurements. The median time span between the last VL measurement and sampling for resistance was 5 months (range 0–25 months); seven patients had never undergone VL testing. Only 37 patients (28%) had CD4 cell counts and VL determined simultaneously with the resistance test. The patients were recruited using three different criteria for treatment failure (virological, immunological and clinical) because access to plasma HIV-1 RNA and CD4 quantification was irregular during the study period. Table 2 shows that 51% of the treatment failures were identified virologically, 21% immunologically and 28% clinically.