To assess the transfer accuracy of three dimensional-printed trays (3D-printed trays) and to compare the accuracy of this approach with the other methods used for the same purpose.

An electronic search was performed in PubMed, Scopus, Web of science, Google Scholar and ProQuest (dissertation and theses). Studies were eligible for inclusion if they involved the assessment of transfer errors measured by the superimposition of the original set-up of brackets and the final positions after transfer. Risk of bias was assessed using Cochrane's tools.

Seven studies were included. The pooled estimate of the transfer error in the linear dimensions showed an error of 0.095mm in the mesiodistal direction (95%CI: 0.035, 0.155), 0.114mm in the buccolingual direction (95%CI: 0.067, 0.160) and 0.111mm in the vertical direction (95%CI: -0.033, 0.255). Concerning the angular dimensions, the pooled estimate showed a tipping error of 1.342⁰ (95%CI: 0.444, 2.240), rotational error of 0.998⁰ (95%CI: 0.323, 1.672), and torque error of 1.913⁰ (95%CI: 0.922, 2.903). Thus, the maximum transfer error was in the bucco-lingual and torque directions. 3D-printed trays had better control in mesiodistal and vertical directions than vacuum-formed trays. Polyvinyl siloxane trays were more accurate in the vertical direction than 3D-printed trays. The scarcity of randomized clinical trials is the main risk of bias of the included studies.

According to the available evidence, 3D-printed trays have an acceptable transfer accuracy, based on the American Board of Orthodontics Objective Grading System (ABO OGS), with maximum linear transfer error in the buccolingual direction and maximum angulation error in the torque. Conducting randomized clinical trials on this topic is highly recommended.