Smoking affects the survival of turned titanium implants. Although smoking has less impact on the failure rate of rough surface implants, the effect on bone loss on rough surface implants has not been studied yet and may be an important factor in biological stability. To determine the effect of smoking on early implant failures and bone remodeling around moderately rough implants (Southern Implants®, Southern Implants, Irene, South Africa). Three hundred twenty-nine patient records, containing information on 712 installed implants, were scrutinized retrospectively and periapical radiographs were analyzed for interproximal bone level. Mann-Whitney U-test and Fisher's exact test were performed to compare bone level and implant survival in smokers and nonsmokers. Only implants with at least 6 months of function time were analyzed for bone level changes. The overall survival rate was 98.3%. Implants in smokers had a threefold higher failure rate compared with nonsmokers (5/104 = 4.8% vs 7/608 = 1.2%). This was statistically significant on implant level (p = .007) but not on patient level (1/41 vs 7/288, p = .997). Readable radiographs from 363 implants in 169 patients were available with a mean follow-up of 12 months (SD 5.11; range 6-28). The mean interproximal bone level was 1.36 mm (n = 363; SD 0.41; range 0.48-3.70). Bone levels were independent of jaw location. Sixty implants from 21 smokers lost statistically significantly (p = .001) more bone (mean 1.56; SD 0.53; range 0.75-3.22) than the 303 implants in 148 nonsmokers (mean 1.32 mm; SD 0.38; range 0.48-3.7). The maxilla is especially prone to bone loss compared with the mandible (1.70 mm vs 1.26 mm, p < .001). The Southern Implants® system demonstrated a high absolute survival rate. Although smokers are not more prone to implant loss, more pronounced peri-implant bone loss was observed, especially in the maxilla. Whether this affects future biological complications remains to be investigated in prospective long-term studies.