Purpose: The detrimental effect of cigarette smoking on implant survival has been previously demonstrated. The purpose of this study was to retrospectively investigate the effect of smoking on marginal bone loss around endosseous dental implants. Materials and Methods: The sample consisted of 767 Brånemark implants placed in 235 patients between 1979 and 1999. Bone level changes were determined using periapical radiographs taken at annual recall visits for 1 to 20 years following prosthesis insertion. Nonparametric tests and multiple linear regression were used to determine the influence of various factors on peri-implant bone loss during the first year of clinical loading and for all subsequent years. Results: The mean annual bone loss was 0.178 mm ± 0.401 during the first year of clinical loading and 0.066 mm ± 0.227 per year thereafter. A positive smoking history was associated with a higher rate of peri-implant bone loss, and the majority of implant failures were observed in this group of patients. Smoking at the time of stage 1 surgery did not appear to predispose implants to more marginal bone loss. Conclusion: Cigarette smoking should not be an absolute contraindication for implant therapy; rather, long-term heavy smokers must be informed that they are at a slightly higher risk of late implant failure and are susceptible to more marginal bone loss over the long-term, irrespective of their smoking status at the time of implant placement.