Medical ozone is used to treat various diseases, including numerous pathologies associated with chronic pain. Chronic pain may be treated by systemic administration of ozone, with ozonated autohemotherapy (OAH) being the commonly used method. In the clinic, intravenous infusion of ozonized saline has been used to treat various diseases. Compared with OAH, ozonized saline infusion is less technically demanding and causes minimal damage to veins. However, it has been indicated that ozone may oxidize saline and generate toxic substances, and therefore, the safety of ozone treatment has been questioned. In the present study, the potential chemical compounds produced from ozone and saline, including chlorite, chlorate and perchlorate, were examined at various time-points with ion chromatography-mass spectrometry (IC-MS). A control group (pure oxygen group) and an ozone group were included in the present study. Two subgroups were included within each group: A saline bottle (made from polypropylene) subgroup and an ozone-resistant blood transfusion bag [made from medical polyvinyl chloride, di(2-ethyl) hexyl phthalate plasticized] subgroup. For the ozone group, 100 ml saline and 100 ml medical ozone at various concentrations (20, 40 or 60 µg/ml in pure oxygen) were injected into the saline bottle or blood bag, and for the control group, 100 ml of pure oxygen was injected into the saline bottle or blood bag. The presence and the content of chlorite, chlorate and perchlorate were determined at different time-points (3, 6 and 15 days after mixing) by IC-MS. Chlorate was detected in the ozone groups at three time-points and its content increased as the ozone concentration and the reaction time increased. Under the same conditions (the same ozone concentration and the same incubation time), the chlorate content (0.90±0.14-7.69±0.48 µg/l) in the blood bag subgroup was significantly lower than that in the saline bottle subgroup (45.23±6.14-207.6±15.63 µg/l). However, chlorite and perchlorate were not detected at any time-point in the two groups. In addition, in the control group (pure oxygen group), chlorite, chlorate and perchlorate were not detected at any time-point. These results indicate that ozone reacts with saline to produce chlorate. Ozone may also react with the polypropylene saline bottle to increase the chlorate content in the bottled solution. Due to a lack of toxicology studies of chlorate in blood, it remains elusive whether ozonated saline and chlorate at the range of 0.90±0.14-7.69±0.48 µg/l has any toxic effects. The potential toxicity of chlorate should be considered when ozonated saline is used for clinical infusions.