Methods: The Cochrane HIV/AIDS Review Group Trials register, the Cochrane library, PubMed, EMBASE and CENTRAL were searched in February 2015. Titles, abstracts, and full-text articles were screened using a priori criteria. Study-level effects were synthesized by meta-analysis.
Results: Of 910 records identified in the published literature since 2012, no additional trials were found. Three additional trials were included from conference proceedings and one additional trial was included from the pre-2012 literature. Study authors were contacted for unpublished data, as needed. In children aged >3 years, there was no difference in the hazard for treatment failure between NNRTI (non-nucleoside reverse transcriptase inhibitor) and PI (protease inhibitor)–based regimens (HR 0.99, 95% CI 0.36–2.73). In addition, there was no difference in efficacy and toxicity between abacavir (ABC) and zidovudine (AZT) among ART-naïve children. Children initially virologically suppressed on a ritonavir-boosted lopinavir (LPV/r)-based regimen who were switched to an NNRTI—either nevirapine (NVP) or efavirenz EFV)—were less likely to develop virological failure (>50 copies/mL), compared with children staying on LPV/r (NVP: HR 0.62, 95% CI 0.41–0.92; EFV: HR 0.58, 95% CI 0.36–0.95).
Conclusion: Among children aged >3 years, PIs appear comparable to NNRTIs in children in terms of efficacy. There was no difference in efficacy or toxicity between ABC or AZT. Substitution of LPV/r with either NVP or EFV remains an alternative to continuing LPV/r among young children. These findings support existing WHO recommendations from 2013.
N. Sugandhi, None
A. Prendergast, None
M. Penazzato, None