TEXT

Top Abstract Text References

A wound is the result of physical disruption of the skin, one of the major obstacles to the establishment of infections by bacterial pathogens in internal tissues. When bacteria breach this barrier, infection can result (1, 7). The most common underlying event for all wounds is trauma. Trauma may be accidental or intentionally induced. The latter category includes hospital-acquired wounds, which can be grouped according to how they are acquired, such as surgically and by use of intravenous medical devices. Although not intentionally induced, hospital-acquired wounds can be the pressure sores caused by local ischemia, too. They are also referred as decubitus ulcers, and when such wounds become infected, they are often colonized by multiple bacterial species (7). Most wound infections can be classified into two major categories: skin and soft tissue infections, although they often overlap as a consequence of disease progression (5, 7, 8, 13). Infections of hospital-acquired wounds are among the leading nosocomial causes of morbidity and increasing medical expense. Routine surveillance for hospital-acquired wound infections is recommended by both the Centers for Disease Control and Prevention (6) and the Surgical Infection Society (2). The objectives of the present study were to identify the etiologies of surgical wound infections over the course of 6 years and characterize the antimicrobial susceptibilities of the pathogen isolates.

This retrospective study included 676 patients who underwent surgical treatment (abdominal, vascular, orthopedic, and reparative surgery) during the 6-year period from May 1993 to April 1999. All patients presented signs and symptoms indicative of surgical wound infections. A definite case of surgical wound infection was defined as one in which there was any skin eruption or drainage at the surgical site that was positive for bacteria by culture within 60 days of a surgical procedure. On the other hand, a presumptive case was one in which there was any skin eruption or drainage at the surgical site that was either culture negative or unresponsive to appropriate antibiotic therapy for organisms obtained on culture.

Semiquantitative aerobic and anaerobic cultures were taken routinely before (time zero [T₀]), during (T₁), and at the end of (T₂) antibiotic therapy. For the isolation of anaerobes, specimens were inoculated onto Columbia blood agar plates enriched with hemin and menadione, incubated in an anaerobic chamber at 37°C, and examined at 48 and 96 h. Contemporaneously, specimens were Gram stained for direct examination. Altogether, 963 pre-antibiotic treatment specimens from 676 individuals were examined. One-thousand sixty bacterial strains were isolated from 614 individuals. Particularly, a single agent was identified in 271 patients, multiple agents were found in 343 patients, and no agent was identified in 62 patients. A high preponderance of aerobic bacteria was observed. Among the common pathogens were Staphylococcus aureus (191 patients, 28.2%), Pseudomonas aeruginosa (170 patients, 25.2%), Escherichia coli (53 patients, 7.8%), Staphylococcus epidermidis (48 patients, 7.1%), and Enterococcus faecalis (38 patients, 5.6%). Pure cultures most commonly yielded S. aureus (98 strains), P. aeruginosa (82 strains), and Enterobacteriaceae (102 strains). Polymicrobial infections involved a similar spectrum of pathogens and frequently involved gram-positive and gram-negative organisms, especially S. aureus together with P. aeruginosa (54 cases). Interestingly, the association between S. aureus and P. aeruginosa became increasingly more frequent, with 31 cases in the 2-year period from May 1997 to April 1999. Methicillin resistance was documented in 23 (74.2%) out of these 31 S. aureus isolates. MICs of several antimicrobial agents were determined by the broth microdilution method according to the procedures outlined by the National Committee for Clinical Laboratory Standards (9, 10). Results from cultures and susceptibility tests performed before antibiotic therapy are summarized in Table 1.

Independently of culture results, antibiotic treatment was started for all patients. During treatment, 681 T₁ control specimens were obtained from 582 (95.0%) of the above-mentioned 614 culture-positive (C⁺) individuals, while 71 specimens were obtained from 55 (88.7%) of the 62 culture-negative (C) patients. Overall, bacterial pathogens were isolated from 131 (21.3%) C⁺ patients, while the C patients remained culture negative, with the exception of two patients positive for the presence of P. aeruginosa and Stenotrophomonas maltophilia. Finally, successive control specimens were obtained at the end of antibiotic treatment from all the 131 patients with T₁ control specimens positive for bacterial pathogens. Nineteen individuals out of these 131 patients had persistently positive culture results in spite of specific antibiotic treatment.

Overall, on the basis of clinical and microbiological data, 595 (96.9%) out of 614 C⁺ individuals were classified as having definite cases of surgical wound infection, while the above-mentioned 62 C patients and 19 (3.1%) out of 614 C⁺ patients were classified as having presumptive cases of surgical wound infection.

The susceptibility patterns of the 1,060 bacterial strains, divided into three 2-year periods, to several antimicrobial agents are summarized in Table 2. Some consequential observations arose from the data in Table 2. More than 50% of the Enterobacteriaceae tested were resistant to ampicillin, while only a few (<20%) were resistant to the combination of amoxicillin and clavulanate. This finding suggests that the resistance observed was due mainly to the production of -lactamase by the organisms. In addition, most isolates were susceptible to ceftriaxone but more than 50% were resistant to cefazolin.

Most P. aeruginosa isolates were susceptible to piperacillin, ceftazidime, and imipenem, although a gradual emergence of resistance to these -lactams has been observed. In addition, only a few isolates were resistant to netilmicin, while a severe decrease in ciprofloxacin activity has been noted in the last few years.

In this study S. aureus was the most common cause of surgical wound infections. Methicillin resistance was documented in 104 (54.4%) of 191 S. aureus isolates. Although amoxicillin-clavulanate, cefazolin, and imipenem were shown to be active in vitro against more than 60% of the isolates, according to National Committee for Clinical Laboratory Standards recommendations, the methicillin-resistant staphylococci were considered resistant to all -lactams, including penicillins, cephalosporins, -lactam--lactamase inhibitor combinations, and carbapenems, since these agents may be clinically ineffective against such organisms.

Enterococci, a frequent cause of infection in surgical wounds, were isolated from 48 patients. Nearly all of the 38 Enterococcus faecalis isolates were susceptible in vitro to glycopeptides (Table 1) and gentamicin (data not shown). In contrast, most of the strains were resistant to cefazolin. Finally, good in vitro activities were shown by amoxicillin-clavulanate and imipenem.

Anaerobic species (36 strains) were isolated from 21 distinct patients. Overall, the anaerobic gram-positive cocci (27 isolates) were susceptible to all the drugs tested, while the gram-negative isolates (nine Bacteroides spp. strains) were shown to be resistant to ampicillin and cefazolin.

Epidemiological data about the emergence of antibiotic resistance were drawn by dividing the susceptibility patterns of the T₀ isolates on the basis of the microbiological results obtained during three 2-year periods (Table 2). The susceptibility data collected in this study suggest that some antibiotics would have very limited usefulness for the prophylaxis or the empirical treatment of wound infections. For instance, most of the gram-negative isolates were found to be resistant to ampicillin and cefazolin while the majority of staphylococcal strains were resistant to methicillin. These are remarkable data, since virtually all the patients received first- or second-generation cephalosporins as antibiotic prophylaxis. Overall, a progressive variation in causative pathogens and resistance patterns has been observed throughout the study. In fact, the susceptibility to antibiotics constantly decreased while multiresistant Pseudomonas and staphylococcal strains were isolated with increasing frequency. According to literature data, perioperative prophylaxis can decrease the incidence of wound infection (2, 3, 6, 7, 10-12, 14, 16). Cefazolin is the most used agent for surgical prophylaxis in our hospitals but can be ineffective against the increasingly common wound pathogens methicillin-resistant S. aureus, methicillin-resistant coagulase-negative staphylococci, P. aeruginosa, and other species of gram-negative rods. The inappropriate usage of antimicrobials in surgical perioperative prophylaxis is still a problem, and a close collaboration between surgeons and microbiologists is needed (4, 15). On the basis of our results, antimicrobial agents or drug combinations with wider spectra of activity and stronger resistance to enzymatic degradation are desirable for perioperative prophylaxis or treatment of surgical infection.