PERIOPERATIVE ANTIMICROBIAL PROPHYLAXIS
Classification of Surgical Procedures
Biliary Tract Surgery
Obstetric and Gynecologic Surgery
Infections following surgical procedures are a major cause of morbidity and occasional mortality. Currently, postoperative wound infections represent up to 24% of all nosocomial infections, prolong hospital stay by 7 to 14 days, and cost in excess of $1 billion annually. Another way of viewing such data is that 2% to 3% of all surgical procedures are complicated by infection, resulting in more than 3,000 deaths annually. The majority of postoperative wounds harbor bacteria in low counts. The role of antibiotics is to reduce the likelihood of bacterial proliferation. These agents have been demonstrated to decrease the likelihood of infection following selected types of surgery; however, their indiscriminate use may result in emergence of resistant pathogens, increased costs, adverse reactions, and a false sense of security.
Asepsis and good surgical technique still remain cornerstones in the prevention of wound infection. The cleanliness of the operating room environment, adequacy of preoperative skin preparation, judicious shaving, level of activity in the operating room, and duration of both preoperative hospitalization and surgical procedure are other potential factors. Patient factors such as poor nutrition, diabetes mellitus, old age, remote foci of infection, and lack of immunologic integrity can contribute greatly to the risk for wound infection. As an example, the presence of a remote focus of infection may increase the risk of postoperative infection fourfold.
Classification of Surgical Procedures
Surgical procedures have historically been classified as clean (class 1), clean/contaminated (class 2), contaminated (class 3), and dirty/infected (class 4). Table 77-1 defines and gives examples of procedures within each class. In general, a major break in asepsis moves a procedure into the next higher class. Such a classification of procedures has been useful to help define anticipated postoperative infection rates. Clean operations are generally associated with infection rates below 2%. Clean/contaminated procedures performed without prophylaxis have historically been associated with wound infection rates higher than 8%, and often much higher. Factors within class 1 and 2 procedures that are associated with a higher likelihood of infection include abdominal surgery, surgery lasting more than 2 hours, and the presence of more than three underlying diseases. The other classes are associated with infection rates above 30%; in these situations, the administration of antimicrobial agents should be considered therapeutic. Factored into this classification should be categorization based on three independent factors established by the Centers for Disease Control and Prevention as important in the likelihood of postoperative wound infection. These are as follows: American Society of Anesthesiology (ASA) score of at least 3, duration of surgery beyond a given length of time for each surgical procedure, and class 3 or 4 surgery. Based on these criteria (scores of 0 to 3), the risk for infection increases from less than 2% to 13%. Thus, a patient with multiple comorbidities (ASA class 3 or greater) for whom prolonged surgery is anticipated may benefit from antibiotic prophylaxis even if undergoing class 1 surgery.
Table 77-1. Classification of surgical procedures
Parenteral antimicrobials have been used for perioperative prophylaxis for several decades; however, only since the early 1980s have well-designed and controlled studies been performed that document efficacy in selected situations. Factors to be considered in prophylaxis include timing, duration, choice of agent, type of surgical procedure, and cost.
Animal studies performed by Burke in 1961 and subsequently corroborated by human investigations demonstrate the importance of the timing of the initial dose. Antimicrobial agents administered postoperatively (i.e., after contamination has occurred) or more than 2 hours preoperatively give no greater protection than placebo. Most experts now recommend that therapy be initiated 30 minutes before skin incision so that adequate tissue levels have been achieved when tissues are incised. This can be conveniently performed by anesthesiology within the operating suite during other surgical preparations. Levels should be maintained throughout the procedure, but prolonging them beyond this has not been proved necessary; it may increase costs, encourage the emergence of resistant organisms, and increase adverse reactions without decreasing morbidity. Recent studies show the benefit of single-dose perioperative antimicrobial prophylaxis for many surgical procedures, including biliary tract and vaginal hysterectomy. No well-performed investigations have demonstrated enhanced efficacy with multiple-dose regimens.
The agent chosen for prophylaxis should be safe and effective against most anticipated pathogens. To utilize agents most effectively, clinicians must know about the likely postoperative pathogens in their institution. Of especial importance are data regarding the risk for infection with methicillin-resistant Staphylococcus aureus or methicillin-resistant Staphylococcus epidermidis, as a risk for these pathogens requires increased prophylactic use of vancomycin (with the attendant risks for immediate complications and potential for emergence of vancomycin-resistant Enterococcus faecium). The pharmacokinetic properties of the agent employed should allow availability throughout the procedure. Cefazolin, the agent most commonly used, can be given as a single preoperative 1-g dose for procedures lasting under 4 hours. For prolonged procedures, a second dose may be given at about 4 hours. Alternatively, some experts recommend use of an agent with an extended half-life (e.g., cefotetan, ceftriaxone) when surgery is anticipated to extend beyond this length of time. Some authorities also recommend that antimicrobials used for treatment of established infections not be routinely used for prophylaxis.
The role of prophylaxis for clean surgical procedures is controversial. With most of these, infection rates are low, and large numbers of patients would have to receive antimicrobials to prevent small numbers of infections. However, when large numbers of patients have been studied, benefits have been shown for several apparently low-risk procedures (e.g., herniorrhaphy), and patients undergoing some types of breast surgery. Patients with implantation of hardware or grafts and cardiac surgery patients should receive prophylaxis. Additional clean procedures in which prophylaxis should be employed include those lasting more than 4 hours, procedures in patients with distant foci of infection, and procedures in insulin-dependent diabetics. In the presence of these risks, infection rates approach 8% approximately 300% more than in patients undergoing similar procedures without such risks.
Clean/contaminated surgical procedures benefit the most from antimicrobial prophylaxis, and in general perioperative prophylaxis represents a standard of care. Risks for infection from selected procedures may decline from above 15% to well below 8%.
Biliary Tract Surgery
Cholelithiasis is associated with the presence of bacteria in bile in more than 50% of patients. The organisms most commonly identified are Escherichia coli, Klebsiella and Proteus species, and Enterococcus faecalis. Risk factors for postoperative infection are age over 70 years, cholelithiasis, jaundice, and positive bile cultures. In general, agents active against the common gram-negative enteric bacteria have proved effective. These include trimethoprim-sulfamethoxazole and cefazolin. For high-risk patients, 1 g of cefazolin is recommended. Trimethoprim-sulfamethoxazole is a reasonable alternative.
Antimicrobial prophylaxis should always be used for this surgery. Patients given oral antimicrobials as well as cleansing enemas had decreased rates of postoperative wound infection when compared with controls given only vigorous purgation; however, recent limitations on length of preoperative stay may impede the ability to perform satisfactory bowel cleansing. Oral neomycin plus erythromycin decreased wound infection rates from 43% to 9% in comparison studies. Similar results have been achieved with metronidazole and doxycycline. A standardized regimen for oral prophylaxis is given in Table 77-2.
Table 77-2. Bowel preparation for elective colorectal operations
The value of additional parenteral antimicrobials remains controversial. One study demonstrated that cefoxitin decreased infection rates from 18.3% to 6.6%. Many clinicians now add either a single preoperative dose of cefotetan or cefoxitin to a bowel preparation including nonabsorbable antimicrobials.
When emergency colorectal operations must be performed on an unprepared bowel, the likelihood of fecal soilage is great. In this event, parenteral antimicrobial prophylaxis should be initiated. Cefoxitin or cefotetan are reasonable choices. Follow-up treatment depends on surgical findings. Gross spillage necessitates continuing antimicrobials for therapeutic reasons, but if no evidence of bowel perforation is noted, these agents can be discontinued.
Obstetric and Gynecologic Surgery
Patients undergoing vaginal hysterectomy benefit most from parenteral antimicrobial prophylaxis. Surgical wound infection rates decline from between 20% and 40% to between 4% and 8%. Cefazolin has been the most frequently studied antimicrobial and is considered the agent of choice. Other agents with proven effectiveness are ampicillin and cefoxitin. Single-dose prophylaxis is sufficient.
Data for abdominal hysterectomy and cesarean section are less conclusive. For the former, antimicrobials as diverse as cefazolin and metronidazole have demonstrated statistically significant decreases in pelvic and wound infections. However, mechanical methods, such as closed suction drainage, may be equally effective. Patients undergoing cesarean sections should be given prophylaxis if the procedures are unscheduled or associated with prolonged rupture of membranes. The timing of prophylaxis in cesarean sections remains controversial. Some clinicians administer the drug after cord clamping to reduce the possibility of antimicrobial transfer to the neonate.
Cost-benefit analyses of the use of cefazolin in both vaginal and abdominal hysterectomies demonstrate the importance of antimicrobial prophylaxis in both of these procedures. Costs were reduced by $1,777 and $716, respectively, and for vaginal and abdominal procedures, wound infection rates declined from 21.4% to 2.3% and from 21.1% to 14.1%, respectively. The authors report that savings would be substantially higher if more expensive drugs or prolonged prophylaxis had been utilized. A single IV dose of 1 g of cefazolin, therefore, appears to be a prudent choice for antimicrobial prophylaxis in most obstetric and gynecologic operations. (R.B.B.)
Classen DC, et al. The timing of prophylactic administration of antibiotics and the risk of surgical-wound infection. N Engl J Med 1992;326:281–286.
Approximately 2,800 patients were prospectively monitored to determine the relationship of postoperative wound infection to the timing of antimicrobial prophylaxis. Rates of infection were statistically higher when prophylaxis was initiated either more than 2 hours preoperatively or postoperatively. The authors could not differentiate between groups that were given a first dose just before surgery or intraoperatively. This large clinical study confirms many animal data.
Cruse PJE, Foord R. The epidemiology of wound infections: a 10-year prospective study of 62,939 wounds. Surg Clin North Am 1980;60:27–40.
The authors prospectively followed more than 60,000 surgical wounds during a 10-year period. The survey included telephone follow-up with patients at 28 days. There was a direct relationship between surgical wound class and risk for postoperative infection. The rate for clean procedures was 1.5%, and for class 4 procedures, 40%. Infection prolonged hospital stay by an average of 10 days. The authors conclude that wound infections can be decreased by such factors as shortened hospital stay, preoperative hexachlorophene shower, minimal shaving, excellent surgical technique, and expeditious surgery.
DiPiro JT, et al. Single-dose systemic antibiotic prophylaxis of surgical wound infections. Am J Surg 1986;152:552–559.
The authors critically evaluate the available literature on single-dose antimicrobial prophylaxis in surgery. Data are broken out by type of study and surgical procedure. Approximately 40 studies were identified, and in no instance was single-dose prophylaxis demonstrated to be inferior to a multiple-dose regimen.
Haley RW, et al. Identifying patients at high risk of surgical wound infection. Am J Epidemiol 1985;121:206–215.
Information concerning more than 59,000 surgical patients was analyzed to establish risks for postoperative infections. Factors associated with postoperative infection included abdominal operation, surgical procedure longer than 2 hours, class 3 or 4 surgery, and diagnosis of more than three underlying conditions. Addition of factors other than surgical class identified at least twice as many infections.
Hirschmann JV, Inui TS. Antimicrobial prophylaxis: a critique of recent trials. Rev Infect Dis 1980;2:1–23.
Although somewhat dated and not limited to prophylaxis in surgery, this excellent article critically assesses published literature concerning antimicrobial prophylaxis. It depicts problems in study design that accompany many investigations and reviews data based on individual types of surgical procedures. It is one of the first reviews that demonstrate the value of short-course (often single-dose) prophylaxis.
Kaiser AB. Antimicrobial prophylaxis in surgery. N Engl J Med 1986;315:1129–1137.
This review article presents historical data and criteria for study design and assessment and recommendations for antimicrobial prophylaxis by type of surgery in tabular form.
Martin CE, the French Study Group on Antimicrobial Prophylaxis in Surgery, the French Society of Anesthesia and Intensive Care. Antimicrobial prophylaxis in surgery: general concepts and clinical guidelines. Infect Control Hosp Epidemiol 1994;15:463–471.
The authors review basic principles of antibiotic prophylaxis in surgery and present recommendations for numerous specific procedures. They spend much time describing the ideal antibiotic, which of course does not exist.
The Medical Letter. Antimicrobial prophylaxis in surgery. Med Lett Drugs Ther 1997; 39:97–102.
The latest in a series of recommendations from this group of experts seeking to update us on the most recent recommendations. They continue to conclude that cefazolin remains the agent of choice for most procedures. Cefoxitin or cefotetan is sensible for colorectal or appendiceal procedures. Routine vancomycin use should be discouraged. Recommendations for numerous procedures are provided.
Nichols RL. Prophylaxis for surgical infections. In: Gorbach SL, Bartlett JG, Blacklow NR, eds. Infectious diseases, 2nd ed. Philadelphia: WB Saunders, 1998:470–480.
Dr. Nichols presents an excellent overview of the rationale, indications, and antibiotic choices for perioperative prophylaxis. Cephalosporins remain the mainstay of antibiotic prophylaxis, unless resistance issues in particular hospitals or specific operations are recognized.
Nichols RL, et al. Current practices of preoperative bowel preparation among North American colorectal surgeons. Clin Infect Dis 1997;24:609–619.
The authors sent questionnaires to more than 800 colorectal surgeons and received responses from 58%. All utilized some form of mechanical bowel preparation. Almost 90% also utilized antibiotics (the majority both oral and parenteral). The most common regimen was a combination of oral neomycin plus either metronidazole or eryth-romycin plus a parenteral antibiotic.
Platt R, et al. Perioperative antibiotic prophylaxis for herniorrhaphy and breast surgery. N Engl J Med 1990;322:153–160.
This important investigation validates the concept that patients with “clean” surgery, unassociated with implants or other major risks, may benefit from antimicrobial prophylaxis. In this study, ceforanide was employed versus placebo in more than 1,200 patients undergoing herniorrhaphy or breast surgery. For both surgical procedures, postoperative wound infection was less likely to develop in patients who received antimicrobial prophylaxis. It remains uncertain whether all patients undergoing clean surgical procedures would benefit similarly.
Shapiro M, et al. Risk factors for infection at the operative site after abdominal or vaginal hysterectomy. N Engl J Med 1982;307:1661–1666.
Three hundred women undergoing vaginal hysterectomy and 1,125 women undergoing abdominal hysterectomy were prospectively followed to determine factors associated with postoperative surgical wound infections. Longer duration of surgery, young age, lack of antimicrobial prophylaxis, and abdominal approach were associated with infection. The effect of antimicrobial prophylaxis diminished significantly with prolonged surgery, being unmeasurable after 3 hours. This is one of the few studies demonstrating the importance of antimicrobial availability throughout the surgical procedure, which has important implications for other forms of surgery.
Ulualp K, Condon RE. Antibiotic prophylaxis for scheduled operative procedures. Infect Dis Clin North Am 1992;6:613–625.
A contemporary review that includes information on normal flora of various body areas and antimicrobial pharmacokinetics. Recommends specific antimicrobials for various types of procedures.