This registry study was set up to evaluate the effect of smoking on the pre-operative status, intraoperative findings, and post-operative status after rotator cuff reconstruction. Five hundred seventy-six consecutive shoulders with primarily arthroscopically repaired penetrating rotator cuff tear were followed up. Tobacco consumption was recorded as pack-years. Age-adjusted Constant score was used as an outcome measure. Five hundred sixty-four patients were available for 1-year follow-up (dropout rate 2%). One hundred fourteen (20%) and 450 (80%) patients were pre-operatively recorded to be smokers and non-smokers, respectively. The gender distribution did not differ between the groups (P = 0.286). The mean age of all patients was 55 years in smokers (SD 9.1) and 61 years in non-smokers (SD 9.4) (P < 0.001). There was no statistically significant difference in pre-operative Constant score (P = 0.075) or mean size of intraoperatively measured tendon tear (P = 0.290) between the groups. At final follow-up, there was a statistically significant difference in Constant scores between smokers [71 (SE 1.4)] and non-smokers [75 (SE 0.7)] (P = 0.017). The pack-years of smoking correlated with neither the Constant score (P = 0.815) nor the size of the tear (P = 0.786). We conclude that operatively treated rotator cuff tear patients who smoked were significantly younger than non-smokers, and that smoking was associated with lower post-operative Constant score.