Purpose: The purpose of this study was to evaluate the effectiveness of the platform-switching technique to prevent crestal bone loss following the restoration of dental implants. Materials and Methods: This randomized prospective multicenter trial analyzed 60 partially edentulous adults recruited at 12 professional dental centers. Subjects were randomly selected to receive either platform-enlarged or control cylindric implants in three different surgical procedures: conventional nonsubmerged, submerged, and submerged with a reduced abutment. The primary outcome measure was the change in crestal bone level assessed radiographically 12 and 24 months following placement. Nonparametric analysis of variance for repeated measures (the Friedman test) was used to assess the overall significance over time of the differences among implants in changes in crestal bone levels. Comparisons among and between groups of implants were performed by the nonparametric Friedman and Wilcoxon tests, respectively. In all the analyses an a = .05 was considered significant. Results: A total of 360 implants were placed (60 for each group). Three control implants failed during the 2nd year following placement. All submerged and 92% of nonsubmerged platform-enlarged implants exhibited no bone loss. Control implants with an abutment as large as the implant platform exhibited more bone loss than their platform-enlarged counterparts (P .001) or control implants with a reduced abutment (P .001). Submerged implants with an enlarged platform showed better crestal bone preservation than submerged control implants with a reduced abutment (P = .06). Conclusions: The findings of the current trial indicated that the use of implants with an enlarged platform can result in better preservation of crestal bone as compared with conventional cylindric implants when a reduced abutment is mounted. Keywords: bone loss, crestal bone, dental implants, platform-switching technique