Aim: To assess whether the treatment outcome differed between surgical debridement, with or without a bone substitute. Materials and Methods: Forty-one adults with three- or four-wall peri-implant bone defects were enrolled in a 1-year RCT. Surgical debridement (control group), or in combination with a bone substitute (Endobon®) (test group) was performed. Results: Radiographic evidence of defect fill (primary outcome) was only significant in the test group (P = 0.004). At year 1, no bleeding on probing (BOP) in the control and test groups were 7/20 (35%) and 10/21 (47.6%), respectively (χ2= 0.67, P = 0.41). Plaque scores did not differ by study group at baseline (P = 0.31), or at year 1 (P = 0.08). Mid-buccal soft tissue recession changes did not differ by groups (P = 0.76). Successful treatment outcome (defect fill ≥1.0 mm, PPD values at implant ≤5 mm, no BOP, and no suppuration was identified in 1/20 (5.0%) control, and 9/21 (42.9%) test individuals (F = 7, 9, P < 0.01). Number needed to treat analysis identified an absolute risk reduction of 32.8% in benefit of the test procedure. (F = 7, 9, P < 0.01). Conclusions: Successful treatment outcome using a bone substitute was more predictable when a composite therapeutic endpoint was considered. © 2018 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd