The clinical efficacy and safety of HBK were compaired with those of amikacin (AMK) in 230 patients with respiratory infections (pneumonia or lung abscess, chronic respiratory tract infections) by means of a double-blind study at 50 institutions in Japan. Patients over 15 years old with apparent clinical signs and symptoms were administered HBK or AMK intramuscularly b. i. d. for 14 days with a daily dose of 200 mg (pot.) of HBK or 400 mg (pot.) of AMK. The parameters assessed were clinical efficacy, rate of improvement of clinical signs and symptoms, bacteriological response, appearance of side effect, abnormal laboratory findings and clinical usefulness.<BR>Clinical efficacy was analysed statistically in 199 patients (100 administered HBK, 99 administered AMK), but 31 patients out of a total of 230 patients were excluded. Side effects were also anlaysed in 213 patients (HBK: 108, AMK: 105) in whom judgement was possible.<BR>1. On the basis of committee judgement, the overall clinical efficacy rate for pneumonia or lungabscess was 63.2% for HBK and 69.4% for AMK. Those for chronic respiratory tract infections were 46.5% for HBK and 60.0% for AMK. No significant difference was observed between the two drug groups.<BR>In the evaluation of clinical efficacy by the doctor, the overall efficacy ratesinpneumonia or lung abscess cases were 70.4% for HBK and 61.2% for AMK. Those for chronic respiratory tract infections were 55.8% for HBK and 62.0% for AMK. No significant difference was observed between the two drug groups.<BR>2. With respect to the rate of improvement of clinical signs and symptoms, the HBKgroup was superior to the AMK group in the improvement of sputum volume five days after treatment (p<0.05) in patients with pneumonia or lung abscess. In patients with chronic respiratory tract infections, the AMK group was superior to the HBK group in the improvement of body temperature and CRPseven days after treatment (p<0.05).<BR>3. The bacteriological eradication rate of causative organisms was 38.3% out of 51 patients treated with HBK and 37.5% out of 50 treated with AMK. There was no significant differencebetween the two groups.<BR>4. Some side effects were observed in 10 patients out of 108 in the HBK group, butnone were observed in the AMK group. The HBK group displayed a significantly higher rate of side effects than the AMK group. Abnormal laboratory findings were observed at the rate of 23.4% forHBK and 20.0% for AMK, without significant differences between the two groups.<BR>5. Regarding usefulness as judged by committee members, the respective rate for the HBK group and AMK group were 53.8% and 64.6%. Usefulness, as judged by the doctor in charge, was 63.0% for the HBK group and 63.3% for the AMK group. These assessments showed no significantdifferences between the two groups.<BR>From the above results, it was concluded that HBK is as equally useful as AMK for the drug treatment of respiratory infections.