Failure of single-dose metronidazole prophylaxis in colorectal surgery. No benefit from high dosage or combination with nalidixic acid

Abstract

In an open prospective study of postoperative infectious complications, patients undergoing elective colorectal surgery were randomly allocated to one of three groups receiving parenteral single-dose antimicrobial prophylaxis (1 g or 3 g metronidazole or 1 g metronidazole + 3 g nalidixic acid). Because of an unacceptably high rate of surgical infection in all three groups (36%, 29% and 39%, p greater than 0.1) among the first 103 evaluable patients, the study was discontinued. Nalidixic acid--though effective in in vitro tests of bacterial susceptibility--thus was found to be of little or no value as prophylaxis against Gram-negative infection. As the observed infection rate when metronidazole was given at the start of operation was seven-fold that previously found in the same department when 1 g metronidazole was administered 3-4 hours preoperatively (28/103 vs. 2/50, p less than 0.01), the timing of metronidazole prophylaxis was assumed to be potentially important for its ability to protect also against aerobic postoperative infection.