Background and Importance Clinical practice guidelines recommend initiation of antiretroviral therapy (ART) as soon as possible after diagnosis of HIV infection with a combination of nucleoside reverse transcriptase inhibitors (NRTI) with integrase inhibitors (INSTI), non-nucleoside NNRTI or protease inhibitors pharmacologically boosted (PI/b).

Aim and Objectives Compare the potency of different combinations of NRTI with NNRTIs, INSTIs or PI/bs.

Material and Methods Retrospective observational study of naive patients diagnosed between January-2012 and June-2022. Variables analysed were age, sex, route of infection, ART, AIDS, viral load (VL) and time to reach undetectable VL (<50 copies/ml).

Data were collected from the electronic medical records (Mambrino XXI®) and outpatient dispensing software DPE Farmatools®.

Statistical analysis was performed using a linear regression method (dependent variable: potency of the combination characterised as the reduction in VL corrected for the time (months) in which undetectable VL is achieved and an analysis of variance (ANOVA) using SPSS® v.15.

Results Ninety-six people were diagnosed with HIV infection. Median age: 34 years (RIC 30-43), 78% male. AIDS stage was present in 34%. The most common route of transmission was men sex men (MSM) 53%.

Initiation of ART NRTI combined with INSTI was 73%, NNRTI 7% and IP/b 20%. The mean log VL baseline was 4.63 (SD: 0.93).

The mean VL reduction per month of treatment in patients treated with NRTI + INSTI was 2.45 copies/ml/month, NRTI+ PI/b was 1.72 copies/ml/month and NRTI + NNRTI was 1.63 copies/ml/month. The significance of the analysis of variances of the means obtained was 0.112.

Conclusion and Relevance INSTIs potency was higher than the other TAR combinations, although the differences were not significant.

Study heterogeneity in the follow-up times between diagnosis and the date of VL analysis as well as the number of patients treated with NNRTIs, and PI/bs was lower than the INSTIs group may explain the non-significant results.

It would be interesting to extend the sample with a multicentre study to validate the results obtained.

Conflict of Interest No conflict of interest