Effects of commonly used bowel preparations on the large bowel mucosal-associated and luminal microflora in the rat model.

Studies of colonic microflora have indicated there are two distinct populations, one intraluminal and one mucosal surface-associated. This investigation further characterizes the mucosal surface microflora and assesses the effects of common preoperative bowel preparations on both microflora. Quantitative and qualitative bacterial cultures and scanning electron microscopy were used to study the microflora in five groups of seven rats each: control; intraoperative colonic instillation of ten percent povidone-iodine for 20 minutes; mechanical preparation with magnesium citrate; mechanical preparation followed by intramuscular cefoxitin (30 milligrams per kilogram) one hour preoperatively; and mechanical preparation followed by oral neomycin sulfate and erythromycin base (15 milligrams/kilogram each) given by gavage tube 18, 14, and 4 hours preoperatively. Microflora on the mucosal surface was visualized by scanning electron microscopy in all groups except the neomycin/erythromycin group. Results showed fewer bacterial isolates recovered from the mucosal surface compared with the lumen, as well as several log10 units lower for each bacterial classification. The greatest suppression of both microflora was seen in the neomycin/erythromycin group. Total aerobic and anaerobic luminal counts decreased by 3.7 (P less than 0.009) and 6.3 (P less than 0.009) log10 units, while total aerobic and anaerobic wall counts decreased by 2.3 (P less than 0.009) and 2.8 (not significantly) log10 units, respectively. Lesser reductions were noted in the povidone-iodine group (P less than 0.009, P less than 0.009, P less than 0.009, and P less than 0.048, respectively). There were no statistically significant reductions in either total aerobic or anaerobic counts in the mechanical preparation or cefoxitin groups. These results indicate that neomycin/erythromycin is the most effective regimen in reducing both microflora.