STUDY QUESTION Are antiretroviral therapies associated with semen alterations in HIV-infected men? SUMMARY ANSWER Antiretroviral regimens that included the non-nucleosidic reverse transcriptase inhibitor efavirenz were associated with a significant impairment of sperm motility, whereas regimens without efavirenz were not associated with significant semen changes. WHAT IS KNOWN ALREADY Semen alterations including decreased ejaculate volume and sperm motility have been reported in HIV-infected men. The hypothesis ascribing reduced sperm motility to damages induced in sperm mitochondria by nucleosidic (or nucleotidic) reverse transcriptase inhibitors (NRTIs) has not been confirmed in HIV-infected patients and the effects of antiretroviral treatments on semen parameters remain unclear. STUDY DESIGN, SIZE, DURATION This case–control study compared semen characteristics across 378 HIV-1 infected patients receiving different antiretroviral regimens or never treated by antiretroviral drugs, in whom an initial semen analysis was done between 2001 and 2007. PARTICIPANTS/MATERIALS, SETTING, METHODS The patients were partners from serodiscordant couples requesting medical assistance to procreate safely. Their status with regard to antiretroviral therapy at the time of semen analysis was categorized as follows: 1/ never treated patients (n = 66); 2/ patients receiving NRTIs only (n = 49); 3/ patients receiving a NRTIs + protease inhibitor (PI) regimen (n = 144); 4/ patients receiving a NRTIs + non-nucleosidic reverse transcriptase inhibitor (NNRTI) regimen (n = 119). Semen parameters were assessed through standard semen analysis. Additional analyses included measurement of sperm motion parameters using computer-assisted semen analysis, seminal bacteriological analysis, seminal biochemical markers and testosterone plasmatic levels. All analyses were performed in the Cochin academic hospital. The data were analyzed through multivariate analysis. MAIN RESULTS AND THE ROLE OF CHANCE Sperm motility was the only semen parameter which significantly varied according to treatment status. The median percentage of rapid spermatozoa was 5% in the group of patients receiving a regimen including efavirenz versus 20% in the other groups (P < 0.0001). Accordingly, sperm velocity was reduced by about 30% in this group (P < 0.0001). The role of chance was minimized by the strict definition and the size of the study population, which included a large enough group of never treated patients, the controlled conditions of semen collection and analysis, the multivariate analysis, the specificity and the high significance level of the observed differences. LIMITATIONS, REASONS FOR CAUTION The design of the study did not allow demonstrating a causal link between exposure to efavirenz and sperm motility. WIDER IMPLICATIONS OF THE FINDINGS As efavirenz is widely used in current antiretroviral therapy, these findings may concern many HIV-infected men wishing to have children. This justifies further assessment of the consequences on fertility of the exposure to efavirenz. Moreover, the possibility of common cellular impacts underlying adverse effects of efavirenz in sperm cells and neurons deserved investigation. STUDY FUNDING/COMPETING INTEREST(S) No external funding was used for this study. None of the authors has any conflict of interest to declare.