Prevention of postoperative infections by administration of antimicrobial agents immediately before surgery for patients with gastrointestinal cancers.

BACKGROUND/AIMS: This study was conducted to judge the effect of intravenous administration of antimicrobial agents immediately before surgery for patients with gastrointestinal cancers to prevent postoperative surgical site infections (SSIs) and remote site infections.

METHODOLOGY

A total of 3437 patients with gastrointestinal cancers underwent standby operations in Wakayama Medical University Hospital between 1987 and 2002. Of these, 1483 were treated between 1987 and 1995, and intravenous antimicrobial agents were used only postoperatively for 2 to 5 days (no AMP group). In addition to the postoperative administration, antimicrobial agents were injected immediately before surgery in 1954 patients (AMP group). If the operation continues more than 3 hours, antimicrobial agents were injected every 3 hours during operation. A comparison was made between the no AMP group and AMP group concerning the bacteria detected and the incidence of SSIs and remote site infections.

RESULTS

The incidence of superficial or deep incisional SSI after surgery was higher in esophageal cancer (17.2%) throughout the whole period than in gastric cancer (4.2%), colon cancer (5.2%) and hepatic/biliary/pancreatic cancers (4.9%) (p < 0.00001). On the other hand, the incidence of space/organ SSI after surgery was higher in hepatic/biliary/pancreatic cancers (14.7%) than esophageal cancer (8.4%; p = 0.02), gastric cancer (7.9%), and colon cancer (8.1%; p < 0.00001). The overall incidence of superficial or deep incisional SSI after surgery for gastrointestinal cancers was 7.2% in the no AMP group, and 4.1% in the AMP group (p = 0.00006). However, in the overall incidence of space/organ SSI, no significant difference was observed between the no AMP group (10.3%) and the AMP group (8.8%). In addition, the incidence of remote site infections also showed no significant difference between the two groups. Regarding bacterial isolates detected after surgery, the prevalence of methicillin-resistant Staphylococcus aureus (MRSA) decreased from 10.8% to 6.2% among the bacterial strains detected (p = 0.00009), and Pseudomonas sp. also decreased from 13.5% to 10.2% (p = 0.002), but Enterococcus sp. increased from 12.1% to 20.4% (p < 0.00001).

CONCLUSIONS

Preoperative AMP was useful to suppress postoperative superficial or deep incisional SSI, but was unlikely to suppress organ/space SSI or remote site infections. In addition, due to preoperative AMP, MRSA and Pseudomonas sp., which showed SSIs, were decreased in detection rates, whereas the detection rate of Enterococcus sp., which is resistant to cephems, was increased.