Incidência e fatores associados às modificações da Terapia Antirretroviral em pessoas com HIV/AIDS

Abstract

Background: Antiretroviral therapy (ART) has led to increased life expectancy and has given to HIV infection chronic characteristics, bringing up new challenges for long-term ART use. Over the years, the number of antiretroviral drugs at the pharmaceutical market has increased, allowing modifications in the ART. However, multiple antiretroviral drug modifications increase probability of HIV in developing mutations. HIV could get resistance to antiretroviral drugs from the same therapeutic class, and salvage therapy could be necessary. This study aimed at investigate ART modifications in HIV-infected adults, initiating ART between 2001 and 2005 in three HIV/AIDS referral centers in Belo Horizonte (MG). Method: Non-concurrent cohort study (historical cohort) from three HIV/AIDS referral centers in Belo Horizonte (MG), Brazil. It included treatment nave HIV-infected adults who had their first ART prescription between 2001 and 2005, followed up for at least one year and for a maximum of five years. The main endpoint was ART modifications defined as at least one drug alteration in the triple combined ART or addition to monotherapy or dual therapy with zidovudine (ZDV). Modifications in first ART were evaluated in the first year of treatment. Also, first, second and third ART modifications were assessed in a 5-year follow-up, between 2001 and 2010. For the analysis of first ART modification, all patients from initial cohort were considered (Population 1, n=247). For the analysis of second ART modification, all patients from Population 1, who had at least one ART modification were considered (Population 2, n=119). For the analysis of third ART modification, all patients from Population 2, who had at least a second ART modification were considered (Population 3, n=51). Exposure variables were sociodemographic characteristics, clinical, related to ART and to health care. Descriptive analysis was performed to characterize studied population and ART modifications profile. Pearson's qui-square test was performed to compare frequencies of ART modification in different population categories. The outcome was also describe by incidence of patients who had ART modified expressed in 100 persons-month. Kaplan-Meier curves were employed to describe ART durability in population categories. Multivariate logistic regression analysis was used to evaluate factors associated to first ART modification in the first year of treatment. Cox proportional hazard model was performed to evaluate the hazard ratio of first, second and third ART modifications during five years of follow-up. Results: 247 patients enrolled at initial cohort. 119 (48.2%) patients modified ART at least once and 51 (25.6%) modified at least two times. One patient modified ART five times. The vast majority of modifications occurred in the first year of treatment (53.8%), in protease inhibitor (PI)-based regimens, and the main reason for modification was adverse drug reactions. Among patients who modified ART, median time for first, second and third modifications were 10.7, 13.4 and 5.1 months, respectively. Non-nucleoside reverse transcriptase inhibitors (NNRTI)-based regimens had higher durability when used as first, second and third ART option. Patients who started ART with PI-based regimens had 3.46 higher chance of having at least one ART modification than those who started 10 ART with NNRTI-regimens (odds ratio [OR]=3.46; IC95%=2.02 - 5.92), in the first year of treatment. Over five years, female sex, have AIDS-defining illness before ART and start ART with PI-based regimens or monotherapy or dual therapy with ZDV were risk factors for ART modification (hazard ratio [HR]=1.67; CI95%=1.08 - 2.57; HR=1.59; CI95%=1.01 - 2.51; HR=2.50; CI95%=1.54 - 4.04 e HR=11.45; IC95%=4.28 - 30.63, respectively]. It was not verified any variable statistically associated to second ART modification. Patients who had more than four medical visits per year were at higher risk of third ART modification (HR=7.12; CI95%=1.95 - 26.07), as well as those who start third ART with PI-based regimens (HR=4.84; CI95%=1.32 - 17.83). Conclusions: PI-based regimens are not well tolerated and are more modified due to adverse drug reactions. Delay in ART start makes treatment more complex and it is associated to ART modifications in the first year of treatment. Monitoring the adverse reactions to antiretroviral drugs at any stage of treatment is essential to avoid modifications in ART, contributing to ART effectiveness.