It is estimated that 2 to 5 percent of extra-abdominal surgeries and as many as 20 percent of intra-abdominal surgeries are complicated by a surgical site infection. The Centers for Medicare and Medicaid Services and the Centers for Disease Control and Prevention convened the Surgical Infection Prevention (SIP) project to promote the proper selection and timing of prophylactic antibiotics used during surgery. Bratzler and Houck summarize the consensus guideline on surgical antibiotic prophylaxis developed by the SIP.
Authors from several specialty colleges and most of the major national guidelines published on surgical antibiotic prophylaxis were brought together for the SIP. The consensus guideline was designed to highlight areas of agreement among the various guidelines and investigate areas where recommendations were inconsistent.
One of the chief issues on which there was wide agreement was the timing of the first prophylactic antibiotic dose. Ideally, this dose should provide a sufficient antibiotic serum level throughout the surgery to combat organisms most likely to cause a site infection. The SIP authors recommend that the first dose be timed to occur within 60 minutes before the surgical incision is made. If a fluoroquinolone or vancomycin is chosen for prophylaxis, the first dose should be administered within 120 minutes of the start of surgery. If the surgery involves the use of a tourniquet (e.g., hip or knee arthroplasty), the antibiotic infusion should be completed before inflation of the tourniquet. For most surgeries, the SIP recommends that use of prophylactic antibiotics end within 24 hours after surgery.
In surgical centers with a higher rate of methicillin-resistant Staphylococcus aureus (MRSA) infections, the preferred prophylactic antibiotic would be vancomycin. Use of intranasal mupirocin has been shown to decrease nasal carriage with S. aureus, but did not improve surgical infection rates.
The guideline also addressed the specific choice of antibiotic for prophylaxis at different surgical sites (see accompanying table). Cefazolin or cefuroxime are suggested for cardiothoracic surgery, with one guideline recommending extension of prophylactic antibiotics up to 72 hours to avoid deep sternal infections. The authors note, however, that there is no evidence that prolonged use decreases infection rates, although it does increase the incidence of resistant bacteria when infection occurs.
|Cardiothoracic||Cefazolin or cefuroxime; if beta lactam allergy, vancomycin or clindamycin||72-hour duration advocated by some, but 24 hours is likely to be adequate|
|Vascular||Cefazolin or cefuroxime; if beta lactam allergy, vancomycin with or without gentamicin, or clindamycin||—|
|Colon||Oral: neomycin, with erythromycin base or metronidazole||Combination of oral and parenteral prophylaxis may decrease infection rates|
|Parenteral: cefoxitin or cefotetan, or cefazolin plus metronidazole|
|Hip or knee arthroplasty||Cefazolin or cefuroxime; if beta lactam allergy, vancomycin or clindamycin||Infuse antibiotic before tourniquet is inflated|
|Hysterectomy||Cefazolin, cefotetan, cefoxitin, or cefuroxime||Similar recommendations apply for cesarean delivery|
With colorectal surgery, oral or parenteral prophylactic antibiotics have been suggested. Combination oral and parenteral therapy has been shown in some studies to decrease infection rates and is often the method chosen by U.S. doctors.
At times, antibiotic therapy is continued until all drains and catheters are removed after arthroplasty surgery, but the SIP authors noted that there was no evidence that this step decreased infection rates. They further noted that use of drains remains controversial because, over time, the tip can become colonized with bacteria.
The preferred antibiotic agents for use in hysterectomies (cefazolin, cefotetan, cefoxitin, or cefuroxime) also may be used in cesarean deliveries, according to the guideline. The first dose typically is given after the baby has been delivered and the umbilical cord has been clamped, to prevent masking signs of possible infection in the infant.
In conclusion, the guideline authors note that many of the studies that provide the evidence for these recommendations are becoming dated. Antibiotic sensitivity patterns change over time and vary among institutions, thus requiring individualization of general recommendations.