Patients undergoing cancer chemotherapy frequently suffer from mucositis, particularly if they become leukopenic (leucocytes <1000/microL). To identify a possible benefit from antiseptic rinsing of the oral cavity, 47 patients were randomized to rinse either with a chlorhexidine-based product (chlorhexidine concentration 0.3%; N=24) or with an amine-stannous fluoride combination (control group; N=23). Patients were asked to rinse three times a day for 30s from the beginning of chemotherapy until the end of leukopenia. Before rinsing, as well as during and after leukopenia, aerobic and anaerobic bacteria in the oral cavity were counted. At the same time, the patients were assessed for mucositis. In the chlorhexidine-based group, a significant decrease of the aerobic (P=0.042) and anaerobic (P=0.008) bacterial flora was identified. In the control group, the numbers of aerobic and anaerobic bacteria remained unchanged (P>0.05). Fifteen patients in the chlorhexidine-based group had a C-reactive protein (CRP) increase >50mg/L, compared with only eight patients in the control group [odds ratio: 3.13, confidence interval (CI) 0.82-12.39]. Nine patients in the chlorhexidine-based group but only two patients in the control group developed severe mucositis. This difference was statistically significant with an odds ratio of 6.30 (CI: 1.02-49.67). As not all of the 47 patients developed severe leukopenia, a separate analysis was carried out for patients with <1000 leucocytes/microL for a minimum of three days. The results of the microbial counts were very similar, with a clear reduction in the chlorhexidine group and no major alterations in the control group. Twelve of 15 patients in the chlorhexidine-based group had a CRP >50mg/L whereas only eight of 15 patients did so in the control group, which can be regarded as a slightly elevated risk for a CRP increase in the former group. Seven of 15 patients developed severe mucositis in the chlorhexidine-based group, but only two of 15 patients in the control group. These differences were not significant, but patients treated with chlorhexidine-based product seemed to have more problems with inflammation of the oral mucous membranes, resulting in an elevated mucositis score and a CRP increase. Other parameters such as body temperature or application of antibiotics did not differ between the two groups. We conclude that treatment with the chlorhexidine-based product did not provide a clinical benefit for cancer chemotherapy patients. On the contrary, the risk of mucositis and clinical sequelae seems to be enhanced, although the counts of micro-organisms on the oral mucous membranes are significantly reduced.