Although sports mouthguards provide protection against trauma, dentoalveolar injuries can still occur with the mouthguard in place. This study examined the effect of mouthguard protection in an in vitro model. A simulated maxilla, out of a PMMA arch containing replaceable resin teeth (Frasaco, Tettnang, Germany) was used to assess the performance of different mouthguard designs. 'Boil and bite' and custom-fitted mouthguards [ethylene vinyl acetate (EVA)] laminated with hard [polyvinylchloride (PVC)] or soft labial intermediate EVA layers were fabricated according to manufacturers' instructions. A steel ram was dropped onto the mouthguards at the maxillary incisor (21) region from a height of 25 cm (184 N i. e. fracture toughness of the resin tooth). The mV-peaks transmitted through the sample were measured at the back of the upper incisor (21) with a strain gauge. Ten readings were noted for each type of mouthguard. Data were analyzed by ANOVA at a significance level of 0.05. Bilaminated mouthguards with hard intermediate PVC layers of 0.8 mm, 1.5 mm or 2 mm thickness showed no significant differences to those of 1.5 mm thick soft intermediate EVA layers (peaks: -0.380 to -0.403 mV). Absorption rate amounted to 33 % compared with the unprotected tooth. 'Boil and bite' and mouthguards layered with silicon or with small hard PVC layers of 1.5 mm thickness demonstrated less absorption (peaks: -0.445 to -0.9 mV) and differed significantly from the other mouthguard systems (p 0.05). These in vitro tests demonstrated that mouthguards laminated with soft EVA (1.5 mm thickness) or hard labial PVC (0.8, 1.5 or 2 mm thickness) intermediate layers protected artifical teeth more than those with small hard PVC intermediate layers or with silicon layers. Keywords: mouthguard, protection, trauma, absorption