Timing of antibiotic administration and infectious morbidity following cesarean delivery: incorporating policy change into workflow.

PURPOSE

To evaluate the success of a multidisciplinary approach to policy change regarding timing of antibiotic administration for the prevention of surgical-site infection after cesarean delivery.

METHODS

After review of the evidence, our multidisciplinary Obstetrics Leadership Committee decided to change policy on the timing of antibiotic prophylaxis for cesarean delivery. Using a combination of meetings, email communications, and local champions, 100% compliance with the new policy was achieved in 5 weeks. The effect of this policy change was investigated through a prospective cohort study of consecutive patients undergoing cesarean delivery at one institution from January 2009 through May 2009. Approximately halfway through the study period our department implemented a practice change that required antibiotic administration before skin incision rather than after clamping the umbilical cord. We compared the incidence of surgical-site infection, including endometritis, cellulitis, and total infectious morbidity, among women who received antibiotics before skin incision to those who received antibiotics after umbilical cord clamp.

RESULTS

There were 533 consecutive women who underwent cesarean delivery during the study period. Two hundred forty (45.0%) women received antibiotics after cord clamping, and 285 (53.5%) women received antibiotics before skin incision; timing could not be determined for 8 (1.5%) women. Within 5 weeks of the policy change, 100% of the women undergoing cesarean delivery received perioperative prophylactic antibiotics before skin incision. The incidence of infectious morbidity fell from 5.4 to 2.5% when antibiotics were given before skin incision. Compared to the administration of antibiotics before skin incision, receiving antibiotics after cord clamp yielded a crude relative risk (RR) of 2.21 (95% CI 0.89-5.44) for total infectious morbidity and 3.56 (95% CI 0.73-17.49) for endometritis. Although not statistically significant, there was an increased risk of cellulitis (RR 1.66; 95% CI 0.53-5.17) when antibiotics were administered after cord clamping.

CONCLUSIONS

A multidisciplinary approach was successful in achieving 100% adherence to our institution's policy change regarding timing of prophylactic antibiotics. This approach was necessary in order to incorporate this type of change into the labor and delivery workflow and may serve as a paradigm for success in implementing labor and delivery quality improvement projects. In addition, administration of prophylactic antibiotics before skin incision resulted in fewer surgical-site infections following cesarean delivery. As the clinical and economic impact of surgical-site infections is considerable, the once common practice of administering antibiotics after cord clamping should be avoided.