PURPOSE To test whether or not one of two membranes is superior for peri-implant-guided bone regeneration in terms of clinical and histologic outcomes. MATERIAL AND METHODS In 27 patients, 27 two-piece dental implants were placed in single-tooth gaps in the esthetic area. Buccal dehiscence and/or fenestration-type defects were regenerated using demineralized bovine bone mineral and randomly covered with either a resorbable membrane (RES) or a titanium-reinforced non-resorbable membrane (N-RES). Clinical measurements included vertical defect resolution and the horizontal thickness of regenerated bone at implant placement and at 6 months. Statistics were performed by means of nonparametric testing. RESULTS The remaining mean vertical defect measured 4 mm (±2.07) (RES) and 2.36 mm (±2.09) (N-RES) (P = 0.044) at baseline and 0.77 mm (±0.85) (RES) and 0.21 mm (±0.80) (N-RES) (P = 0.021) at re-entry. This translated into a defect resolution of 85% (RES) and 90.7% (N-RES) (P = 0.10). The horizontal thickness after augmentation measured 3.46 mm (±0.52) (RES) and 2.82 mm (±0.50) (N-RES) (P = 0.004). The mean loss in horizontal thickness from baseline to re-entry measured 2.23 mm (SD ±1.21) (RES) and 0.14 mm (±0.79) (N-RES) (P = 0.017). The horizontal changes in thickness at the implant shoulder level were statistically significant between the groups (P = 0.0001). CONCLUSIONS Both treatment modalities were clinically effective in regenerating bone as demonstrated by a similar horizontal thickness and vertical defect fill at 6 months. The N-RES group exhibited significantly less horizontal bone thickness reduction from baseline to follow-up.