Objective: To compare the residual cement between computer aided design/computer aided manufacturing customized abutments (CCA) and stock abutments (SA), and to evaluate the feasibility of digital measurement for residual cement volume by three-dimensional scanning.Methods: Twenty master models needed in this study were all taken from one 47-year-old patient with arrested periodontitis, who had already had an implant placed at his right upper central incisor site in the Department of Periodonto-logy, Peking University School and Hospital of Stomatology. After 4 weeks of soft tissue conditioning by means of customized healing abutment, the height of peri-implant soft tissue was measured, from the implant platform to mucosal margin, as 5 mm. Using customized impression coping, the impression was taken and twenty models were fabricated and allocated to 4 groups according to the type of abutments: CCA1 (5 mm transmucosal height CCA, with margin at tissue level), CCA2 (4 mm transmucosal height CCA, with 1 mm submucosal margin), SA1 (3 mm transmucosal height SA, with 2 mm submucosal margin) and SA2 (1 mm transmucosal height SA, with 4 mm submucosal margin). Crowns were cemented to the abutments, which were seated on the working models. Excess cement was removed by a prosthodontic specialist. Thereafter, the volume of residual cement was evaluated by using three-dimensional scanning technique. The area proportion of residual cement was calculated on photographs taken by a single lens reflex camera. The weight of residual cement was weighed by an analytical balance. And the correlation of residual cement volume data with residual cement area proportion or weight of residual cement acquired by traditional methods was analyzed.Results: Residual cement was observed on all the experiment samples. The residual cement volume of CCA was significantly less than that of SA [(0.635 3±0.535 4) mm³ vs. (2.293 8±0.943 8) mm ³, P<0.001]. Consistently, CCA had less residual cement area proportion and weight than those of SA [area proportion: 7.57%±2.99% vs. 22.68%±10.06%,P<0.001; weight: (0.001 5±0.001 0) g vs. (0.003 7±0.001 4) g, P<0.001]. The residual cement volume was strongly correlated with the residual cement area proportion and residual cement weight (r>0.75, P<0.001).Conclusion: These in vitro results suggest that CCA minimized the residual cement more effectively than SA. The method to digitally evaluate the residual cement volume is feasible, but its validity and reliability need to be further studied.