Timing of antibiotic prophylaxis in cesarean section: retrospective, difference-in-differences estimation of the effect on surgical-site-infection

Background: Cesarean section (CS) is one of the most common surgical procedures performed worldwide. Surgical-site-infection (SSI) occurs in approximately 5-10% of CS. The benefit of prophylactic antibiotics for prevention of SSI has been demonstrated in the literature. The optimal timing of antibiotic prophylaxis (prior to surgical incision versus after cord clamping) was investigated in recent studies. In January 2014, the Israeli Ministry of Health introduced a national quality measure which monitors the administration of prophylactic antibiotics in CS. The custom clinical practice in our medical center was to administer prophylactic antibiotics immediately after cord clamping. Upon introduction of the national quality measurement program, the practice was changed to administration of antibiotics prior to surgical incision. Our objective was to examine the effect of timing of prophylactic antibiotics administration on the incidence of SSI following CS, in a single medical center that performs a large volume of deliveries, with a low rate of CS.

Material and methods: Taking advantage of a discrete change in clinical practice, we used retrospective data and applied difference-in-differences design to estimate the effect of the timing of prophylactic antibiotics administration on SSI rates. The analysis included all CSs performed during 2012-2015 and all hysterectomies conducted during the study period.

Results: The coverage rates of prophylactic antibiotics in CS before and after the policy change were 99.10% and 99.03%, respectively. The rates of SSI following CS, before and after the policy change, were 2.63% (n = 2499) and 2.32% (n = 3840), respectively. The rates of SSI following hysterectomy, before and after the policy, change were 6.82% (n = 396) and 7.09% (n = 437), respectively. Difference-in-differences (DID) estimates of the effect of policy change on the incidence of SSI in linear and logistic regression models were not significant (B = -0.6%, p = .64; odds ratio = 0.84, p = .58, respectively).

Conclusions: We found no effect of the timing of prophylactic antibiotic administration (prior to surgical incision versus after cord clamping) on SSI rates following CS.