Background Current guidelines recommend starting antiretroviral treatment (ART) in all HIV-infected patients irrespective of the CD4 count.¹ Some studies have described that more than 40% of patients switch their initial ART.²

Purpose To describe initial ART in naïve patients, its persistence and the reasons leading to an ART switch.

Material and methods Retrospective observational study including all ART-naïve adult patients from January 2012 to August 2017 from our cohort of 2,060 HIV-infected patients. Patients restarting ART were excluded.

Data collected: demographic, HIV viral load (VL) and CD4⁺ count at baseline, initial ART and persistence.

Reasons for switching were classified as schedule optimisation, adverse events, toxicity prevention, drug-drug interactions, low-level viraemia, drug resistance and others.

Categorical variables, n (%); quantitative variables, mean ±SD.

The probability of switching the initial ART over time was calculated by Kaplan–Meier curves and log-rank test. Relative hazards of switching ART-naïve were calculated by Cox regression (adjusted for age, sex and CD4⁺ count).

Results During this period, 448 naïve-patients began ART: 202 (45.1%) INSTI, 137 (30.6%) PI and 109 (24.3%) NNRTI. ART-naïve was switched in 252 patients (56.3): 215 (85.3%) males, age: 39.3±10.0 years, VL≥100,000: 110 (43.8%), CD4 <200: 86 (34.4%). No differences in sex, age, baseline VL and CD4⁺ count were observed between patients with and without switching.

Kaplan–Meier showed differences in the persistence between different ART being the shortest time with the PI (p<0.001). There were statistically significant differences between ART-naïve (Hazard Ratio=2.7, p<0.001, 95% CI: 1.9 to 3.9).

Conclusion During the study period, more than 50% of patients switched their initial ART.

Differences in the persistence were observed between different ART, having the PI the shortest time.

The most common reasons for switching IP, INSTI and NNRTI were schedule optimisation, the presence of adverse events and toxicity prevention, respectively.

References and/or acknowledgements 1. http://www.eacsociety.org/files/guidelines_9.0-english.pdf%0Ahttp://www.eacsociety.org/guidelines/eacs-guidelines/eacs-guidelines.html

2. http://www.ncbi.nlm.nih.gov/pubmed/11546952

No conflict of interest.