The difficulty in accessing hard-to-reach populations as men who have sex with men presents a dilemma for HIV surveillance as their omission from surveillance systems leaves significant gaps in our understanding of HIV/AIDS epidemics. Several methods for recruiting difficult-to-access populations and collecting data on trends of HIV prevalence and behavioural factors for surveillance and research purposes have emerged. This paper aims to critically review different sampling approaches, from chain-referral and venue-based to respondent-driven, time-location and internet sampling methods, focusing on its main advantages and challenges for conducting HIV research among key populations, such as men who have sex with men. The benefits of using these approaches to recruit participants must be weighed against privacy concerns inherent in any social situation or health condition. Nevertheless, the methods discussed in this paper represent some of the best efforts to effectively reach most-at-risk subgroups of men who have sex with men, contributing to obtain unbiased trends of HIV prevalence and HIV-related risk behaviours among this population group.