Background The notion that smoking cannabis may damage the respiratory tract has been introduced in recent years but there is still a paucity of studies on this subject. The aim of this study was to investigate the relationship between cannabis smoking, pneumothorax and bullous lung disease in a population of operated patients. Methods and findings We performed a retrospective study on patients operated on for spontaneous pneumothorax. Patients were divided into three groups according to their smoking habit: cannabis smokers, only-tobacco smokers and nonsmokers. Cannabis lifetime exposure was expressed in dose-years (1d/y = 1 gram of cannabis/week for one year). Clinical, radiological and perioperative variables were collected. The variables were analyzed to find associations with smoking habit. The impact of the amount of cannabis consumption was also investigated by ROC curves analysis. Of 112 patients, 39 smoked cannabis, 23 smoked only tobacco and 50 were nonsmokers. Median cannabis consumption was 28 dose/years, median tobacco consumption was 6 pack/ years. Cannabis smokers presented with more severe chronic respiratory symptoms and bullous lung disease and with a higher incidence of tension pneumothorax than both tobacco smokers and nonsmokers. Cannabis smokers also developed a larger pneumothorax, experienced prolonged postoperative stay and demonstrated a higher incidence of pneumothorax recurrence after the operation than nonsmokers did. The risk of occurrence of chronic respiratory symptoms and bullous lung disease in cannabis smokers was dose-related. Conclusions Cannabis smoking seems to increase the risk of suffering from respiratory complaints and can have detrimental effects on lung parenchyma, in a dose-dependent manner. Cannabis smoking also negatively affected the outcome of patients operated for spontaneous