Background The DHHS (USA Department of Health and Human Services) Adult Antiretroviral Treatment Guidelines and Portuguese Government HIV Guidelines were updated in 2012/2013. These recommend a non-nucleoside reverse transcriptase inhibitor (NNRTI) as initial treatment combined antiretroviral treatment and a ritonavir-boosted protease inhibitor (PI/r) as an alternative regimen for infections that have a higher genetic barrier. Regimens based on thymidine analogues such as zidovudine (AZT) are considered not recommended regimens based on tolerability and adverse reactions, however cost issues were taken into account in our hospital and we chose a protocol with AZT/lamivudine (3TC)/LPV/r (lopinavir/ritonavir).

Purpose To evaluate the percentage of patients treated according to the guidelines, assess the impact on mortality, virological and immunological response with both regimens (recommended and non recommended) and direct associated costs. Patient compliance was also checked by analysing pharmacy electronic files.

Materials and methods An observational retrospective study was conducted (medical records) including 72 HIV-positive treatment-naïve subjects aged over 18 years who started antiretroviral treatment from January to December 2011. We checked patient electronic clinical files for one year. First prescribed treatment, further treatment changes, virological and immunological status before and after 6 months treatment were assessed. Treatments were classified as recommended, alternative or not recommended according to guidelines. Virological response was defined as undetectable viral load after 6 months of treatment.

Results The mean age was 42 years old and 48 patients (66.7%) belonged to the male gender.

At enrolment, 9 patients (12.5%) didn’t have a drug resistance test recorded in their clinical charts and 5 (7%) had mutated resistant virus strains. The mean CD4 count was 231 cells/µL and mean viral load was 323115 copies/mL

Of the 72 patients, 21 (29%), 14(19%) and 37(51%) started a recommended, alternative and not recommended treatment respectively.

None of the patients who were given the recommended regimen had to change treatment or had virological failure.

The most commonly prescribed regimen was LPV/r + 3TC + AZT 59.7% (43 patients) and was not associated with higher mortality or poor compliance. This treatment saved 3000 € /patient /year with a potent virological efficacy (100%) and only 19% of adverse effects due to AZT (8 patients). Mean compliance with treatment was 93%.

Conclusions Not adherence to guidelines didn’t have a negative impact on virological and clinical outcomes.

Given the severe budget restraints we had decided to maintain the existing protocol in our hospital.

No conflict of interest.