Cefamandole sodium (CMD), a new cephalosporin-derivative, was synthesized in the Laboratory of Eli-Lilly Co. Ltd. U.S.A. in 1972. CMD, which is several times more active than cefmetazole (CMZ, a cephamycin antibiotic) against Gram-positive cocci, is only as active as the latter antibiotic against Gram-negative bacilli. Against Haemophilus influenzae, CMD exhibits an antimicrobial activity which is as strong as that of ampicillin sodium. Our previous comparative tests on efficacy and safety of CMD versus cefazolin (CEZ) demonstrated that CMD was as effective and safe as CEZ in the treatment of respiratory tract infections. In the present clinical trial, the efficacy and safety of CMD are evaluated by a comparative double blind method using CMZ, a more recently synthesized cephamycin antibiotic, as a reference drug. For this purpose, a comparative double blind study was carried out in 50 institutions and clinics in Tohoku and Hokkaido districts in Japan. A total of 272 inpatients, who was aged over 16 years and was diagnosed as having pneumonia, lung abscess or acute infectious exacerbation of chronic obstructive pulmonary diseases, was included in this trial. They received 2 g of CMD or CMZ twice a day by intravenous drip infusions, as a rule, for 14 days. Of these patients, 264 (133 received CMD and 131 CMZ) were available for the evaluation of safety and usefulness. Two hundred and thirty-eight patients (122 received CMD and 116 CMZ) were adopted for the evaluation of efficacy. Prior to the treatment, there was no significant difference with respect to age, sex, severity of infection and underlying diseases between subjects in 2 treatment groups. An excellent or good clinical response was obtained in 82% of the patients treated with CMD, and in 81% of those treated with CMZ. Thus, there was no significant difference in cure rate between 2 treatment groups. However, an excellent clinical response was found in 12.3% of the patients treated with CMD, whereas only in 4.3% of those treated with CMZ. This difference in percentage of excellent clinical response between 2 treatment groups was statistically significant (P less than 0.05). Of the 87 patients with moderate to severe infection who were treated with CMD, 13 showed an excellent response. Only 4 of 90 patients treated with CMZ showed an excellent response. Statistically the difference in the rate of excellent response between these 2 groups was significant (P less than 0.05).(ABSTRACT TRUNCATED AT 400 WORDS)