1 Introduction

Despite applying aseptic principles to the practice of surgery, surgical site infections still occur, not only after open fractures but also after clean surgical procedures such as internal fixation of closed fractures or joint replacement surgery.

Antimicrobial prophylaxis is mainly indicated in procedures associated with a high rate of infection, such as clean contaminated or contaminated operations [1].

The criteria for a clean surgical wound include [2]:

• 
  

  Elective procedure (ie, not emergency) with primary wound closure

  
  
• 
  

  Absence of acute inflammation

  
  
• 
  

  No break in aseptic technique

  
  
• 
  

  Absence of transection of colonized surfaces (ie, respiratory, alimentary, or genitourinary tracts)

Clean surgery is associated with an infection rate of 1.5% (in a series of 47,000 procedures) [3]. In such surgery, antibiotic prophylaxis is generally not indicated because the benefits of antibiotics are outweighed by the risks in terms of allergy, antibiotic-associated diarrhea, Clostridium difficile infection, increases in antibiotic resistance, selection for multiresistant organisms, and lack of cost-effectiveness.

Dirty operations are characterized by pus, previously perforated hollow viscus, or open injuries more than 4 hours old. These, by definition, are infected and hence require treatment.

Perioperative antibiotic prophylaxis has become the standard practice in surgery using implants [4].

Elek and Conen [5] first demonstrated the role of foreign bodies in increasing the incidence of wound infection. In the presence of suture material, a 10,000-fold lower inoculum of Staphylococcus aureus still led to skin abscesses in human volunteers. Using an animal model, Zimmerli et al [6] demonstrated that implant material increased the risk of infection. They suggested this was due to locally acquired defects in granulocyte function.

Prophylaxis of surgical site infection not only depends on the appropriate antibiotic but also on operating room discipline, appropriate surgical technique, and proper awareness, and avoidance of the risk factors summarized in Table 4.5-1 [7] with a comprehensive review and recommendations available online from the World Health Organization [8].

Antibiotic prophylaxis should not distract surgeons from careful aseptic surgery, avoidance of risk factors, adequate operating room ventilation, operating room discipline, and meticulous surgical technique.

An expert group [9] has published the quality standard for antimicrobial prophylaxis in surgical procedures. Parenteral antimicrobial prophylaxis should be administered in orthopedic procedures with implantation of prosthetic material.

The risk of infection varies according to the type of fracture and the surgical procedure:

• 
  

  Patients undergoing joint replacement, or with closed fractures, have an infection rate of 0–5% [10].

  
  
• 
  

  In patients with open fractures, the incidence of wound infection correlates directly with the extent of soft-tissue damage. Fractures graded by the Gustilo classification, have infection rates of:

  
  
  
  
  • 
    

    – 0–2% for type I

    
    
  • 
    

    – 2–7% for type II

    
    
  • 
    

    – 7% for type IIIA

    
    
  • 
    

    – 10–50% for type IIIB

    
    
  • 
    

    – 25–50% for type IIIC fractures [11]

In Gustilo type III open fractures, there is extensive soft-tissue damage. In this situation, surgery often takes place in a heavily contaminated field. Therefore, short-term empirical treatment, and not only prophylaxis, should be administered.

2 Microbiology of bone-implant infections

The microbiology of prosthetic joint infections is well known. As in other types of implant-related infections, staphylococci are the predominant infecting agents. This is mainly due to:

• 
  

  Presence of staphylococci within lower layers of skin, which are not reached by antiseptic skin preparation

  
  
• 
  

  Staphylococci possessing virulence factors that bind to host proteins, such as fibrin and fibronectin, and facilitate adherence of staphylococci to implants [12]

In patients undergoing joint replacement surgery, the most common infecting agents are coagulase-negative staphylococci (30–41%) and S aureus (12–39%) [13]. No organism is detected in 5–12% of infections [14].

In fracture surgery, S aureus largely predominates in comparison with other microorganisms. Coagulase-negative staphylococci infections are less prevalent than in joint replacement procedures.

Boxma et al [15] performed a randomized trial of single-dose antibiotic prophylaxis versus placebo in surgery for closed fractures. In the placebo group, the infecting microorganisms were:

• 
  

  S aureus 64%

  
  
• 
  

  Coagulase-negative staphylococci 3%

  
  
• 
  

  Streptococci 8%

  
  
• 
  

  Mixed Gram-positive cocci 5%

  
  
• 
  

  Gram-negative bacilli 6%

  
  
• 
  

  Mixed Gram-positive/Gram-negative microorganisms 8%

  
  
• 
  

  Mixed aerobic/anaerobic bacteria 5%

Patients with open fracture are exposed to a wide range of environmental organisms covering the spectrum of Gram-positive, Gram-negative, and anaerobic bacteria including Clostridium tetani. Dirty wounds are often exposed to what is termed fecal or farmyard flora. Furthermore, wound infections among combat casualties with tibial fractures have been associated with drug-resistant aerobic Gram-negative organisms, including Acinetobacter baumannii, Pseudomonas aeruginosa, and fungi [16].

3 Antimicrobial drugs for prophylaxis

Various antimicrobial substances have been shown to be effective in perioperative prophylaxis.

The following aspects of antimicrobials should be considered when choosing an antibiotic prophylaxis regimen [17]:

• 
  

  The drug(s) should be active against the most common infecting microorganisms involved in implant-associated infection.

  
  
• 
  

  The susceptibility pattern of infecting microorganisms differs in various hospitals. Each hospital needs up-to-date analysis of the resistance pattern of surgical site isolates and prophylactic antimicrobial agents tailored accordingly.

  
  
• 
  

  The risk of causing allergic reactions or adverse effects should be minimal.

  
  
• 
  

  Antimicrobial substances with a high potency to produce resistant strains should be avoided, for example, strong β-lactamase inducers like cefoxitin or ceftazidime.

  
  
• 
  

  The potency of different antimicrobials to precipitate symptomatic Clostridium difficile infection need be reviewed.

  
  
• 
  

  Speed of administration. For example, while vancomycin requires at least 1 hour to infuse (otherwise “red-man syndrome” occurs), the alternative glycopeptide agent, teicoplanin, can be given as a bolus, allowing operating time to be optimized.

  
  
• 
  

  Fracture classification will alter the choice of antibiotic regimen.

  
  
• 
  

  If drugs have similar efficacy, cost should be considered.

The Hospital Infection Control Practices Advisory Committee (HICPAC) [18] recommendations for preventing the spread of vancomycin resistance clearly discourage the use of glycopeptides in routine surgical prophylaxis. Importantly, a metaanalysis of six randomized controlled trials [19], involving 2,886 patients, evaluated the effectiveness of teicoplanin compared with first- or second-generation cephalosporins for perioperative antiinfective prophylaxis. No differences were found between teicoplanin and cephalosporins with respect to the development of infection at the site of surgery. This data support the recommendation to reserve glycopeptide use for patients who are allergic to β-lactam antibiotics or who are colonized with methicillin-resistant S aureus (MRSA).

In closed and open Gustilo types I and II fracture surgery, first- or second-generation cephalosporins, such as cefazolin, cefamandole, or cefuroxime, are a rational choice. If a hospital has concerns about Clostridium difficile infection, then flucloxacillin in conjunction with gentamicin is a preferable combination. If the patient is allergic to β-lactam antibiotics (ie, penicillins, cephalosporins, and carbapenems) or known to be colonized with MRSA, the glycopeptide antibiotics vancomycin or teicoplanin are alternative options.

In Gustilo type III fracture surgery, there may be gross contamination of the wound. In a prospective study [20] of 227 patients with open fractures, prophylaxis with clindamycin was compared with cloxacillin. Unacceptably high rates of infection were reported in grade III fractures for both clindamycin (29.0%) and cloxacillin (51.8%), demonstrating the need for additional Gram-negative coverage in higher Gustilo type fractures [20]. Another randomized prospective study [21] compared the efficacy of intravenous ciprofloxacin with ceftazidime/gentamicin. This study enrolled 163 patients with 171 open fractures (type I [65], type II [54], and type III [52]). In types I and II fractures, the infection rate for the ciprofloxacin group and the ceftazidime/gentamicin group was 5.8% and 6.0%, respectively. While ceftazidime/gentamicin also demonstrated good efficacy among those with type III fractures (7.7% infection rate), ciprofloxacin was associated with an unacceptably high infection rate (31%) [21]. Furthermore, animal model studies [22] have demonstrated a delay in fracture union with the use of ciprofloxacin prophylaxis, hence this antibiotic is not recommended for routine use. In conclusion, prophylactic antibiotics used in Gustilo type III fractures need to be broad spectrum covering all likely pathogens [23].

4 Timing of prophylaxis

Timing of prophylaxis varies depending on the classification of the fracture. Patients with open fractures should be evaluated for their need of tetanus immunization [24]. For all open fractures, administration of parenteral antibiotics is warranted as soon as possible (and ideally within 3 hours of trauma) to reduce the risk of soft-tissue infection or osteomyelitis. Antibiotic efficacy administered either before or at the time of primary treatment was demonstrated in a metaanalysis including 1,106 patients with open fractures [25]. Use of prophylactic antibiotics was associated with an absolute risk reduction in infection of 0.07 (95% confidence interval [CI]: 0.03–0.10), when combined with wound irrigation, surgical debridement, and fracture stabilization.

For both open and closed fractures requiring insertion of metalwork, perioperative antibiotic prophylaxis is essential at the time of wound debridement. Optimal antibiotic efficacy is achieved by providing inhibitory antimicrobial tissue levels at the time of incision and during the entire surgical procedure.

In a seminal animal study, Burke [26] observed that the effective window for antibiotic prophylaxis may be just 3 hours. Even a 1-hour delay markedly decreased the efficacy of the single-dose prophylaxis dose. These conclusions from animal data have been confirmed with a large retrospective clinical study [27] of 2,847 wounds. The risk of surgical site infection increased sixfold when prophylaxis was given either too early (> 2 hours before surgery) or too late (> 3 hours after surgery).

Parenteral perioperative prophylaxis should be administered intravenously in a period beginning 60 minutes before incision [9, 17].

Antibiotic tissue levels are insufficient if the drug is administered less than 5–10 minutes before inflation of a tourniquet [28].

For adequate prophylaxis, antibiotics should be given at least 10 minutes before the tourniquet is inflated.