Aim: The purpose of this in vitro study is to assess the passive fit of prosthetic metal frameworks obtained through a novel digital impression system, for full-arch rehabilitations on multiple implants. Materials and methods: Five master casts, reproducing edentulous jaws with 4 tilted implants, were poured (Groups: MC #1, MC #2, MC #3, MC #4, MC #5). An intraoral scanner system [True Definition Scanner, 3M ESPE, St. Paul, MN, USA] was used to perform five digital impressions (DI) of each master cast (n=25). The implant position was detected with 4 special scan bodies [Toothless, Simbiosi srl, Empoli Firenze, Italy]. A single DI, presenting mean values compared to the others, was selected from each group in order to fabricate a metal framework with CAD-CAM technology (n=5). Passive fit was assessed with the Sheffield Test, screwing each framework on the corresponding master cast. A stereomicroscope [Wild M3Z, Wild Heerbrugg, Heerbrugg, Switzerland] (40x magnification) was used to record maximum gap values at the framework-implant analog interface. Results: Sheffield test reported the following mean values of gap: MC #1 = 0.024 ± 0.019 mm (range: 0.003-0.044 mm), MC #2 = 0.022 ± 0.014 mm (range: 0.003-0.047 mm), MC #3 = 0.027 ± 0.015 mm (range: 0.003-0.045 mm), MC #4 = 0.021 ± 0.012 mm (range: 0.003-0.037 mm), MC #5 = 0.021 ± 0.016 mm (range: 0.002-0.046 mm). No statistically significant differences were found among the groups (p>0.05). Conclusion: Within the limits of this study, digital impression represents a reliable method to fabricate full-arch implant frameworks provided with passive fit. Keywords: intraoral scanner, digital impression, impression accuracy, implant, passive fit, prosthetic framework