BACKGROUND Many studies have compared cement-retained and screw-retained implant prostheses. One disadvantage of cement-retained crowns is the lack of predictable retrievability, which can be overcome by creating a screw access channel on the occlusal surface. OBJECTIVE The purpose of this study was to evaluate the influence of a screw access channel on the retention of cement-retained implant prostheses. METHODS Sixteen cast metal crowns were fabricated and divided into 2 groups of 8 specimens each: a control group (CG) comprising 8 cement-retained prostheses and an experimental group (EG) comprising 8 cement-retained prostheses with a screw access channel. Castings were cemented to abutments with RelyX U100, and the opening screw access channels of EG were filled with photopolymerized composite resin (Filtek Supreme XT). The tensile force required to separate the cemented castings from the abutments was measured after 24 hours of cementation with a universal load-testing machine (EMIC DL 2000). A significance level of 5% (α=.05) was considered statistically significant (Statistical analysis was performed by Kolmogorov-Smirnov non-parametric test and the Student t test). RESULTS The mean force required to achieve the separation was 191 N for the control group (CG) and 161 N for the experimental group (EG). As shown by the nonparametric Kolmogorov-Smirnov test, the dependent variable followed a normal distribution (P=.923). The Student t test found no statistically significant difference (P=.353) between the groups. CONCLUSIONS Fabricating cement-retained implant prostheses with screw access does not compromise or reduce the retention of the crowns.