ORIGINAL ARTICLE
Current Bacterial Speciation and Antibiotic Resistance in Deep Infections After Operative Fixation of Fractures Jesse T. Torbert, MD,* Manjari Joshi, MBBS,† Adrienne Moraff, BS,* Paul E. Matuszewski, MD,* Amanda Holmes, MS,* Andrew N. Pollak, MD,* and Robert V. O’Toole, MD*
Objectives: Infection after fracture fixation is a major source of
treatment of closed fractures of the pelvis, acetabulum, and proximal femur.
morbidity. Information regarding bacterial speciation and antibiotic resistance is lacking. We attempted to determine the speciation and drug resistance profiles associated with fracture fixation infections.
Key Words: bacterial speciation, antibiotic resistance, surgical site, infection, deep infection, fracture, trauma
Design: Retrospective study.
Level of Evidence: Prognostic Level IV. See Instructions for Authors for a complete description of levels of evidence.
Setting: Level I trauma center.
(J Orthop Trauma 2015;29:7–17)
Patients: Two hundred eleven patients with 214 infections underwent surgery for postoperative infection from December 2006 to December 2010. Deep postoperative infections within 12 months of fixation were included.
Intervention: None. Main Outcome Measurements: Incidence of each bacterial species and rate of clinically relevant resistance in Staphylococcus aureus, gram-negative rod (GNR), and Enterococcus species. The effect of timing of infection presentation and location of fracture on bacterial speciation was also investigated. Results: Fifty-six percent of infections had S. aureus present, with 58% of those (32% of all infections) being methicillin-resistant S. aureus. Thirty-two percent of infections had at least one GNR present, with only 4% of those being multidrug resistant. We found a marked increase in the rate of GNR infections of the pelvis, acetabulum, and proximal femur (63%) compared with other locations (27%), which was statistically significant (P = 0.0002).
Conclusions: At our center, S. aureus and GNR are most often found in deep postoperative infections after fixation. Methicillinresistant S. aureus is common in this population. Our GNR rate is high, but resistance in this group was low. The proportion of GNR infections in the pelvis, acetabulum, and proximal femur was high even in closed fractures. These data provide a modern snapshot of orthopaedic infections after fracture fixation and might be useful in designing future studies and protocols for antibiotic prophylactic treatment. We are considering the use of aminoglycosides in the Accepted for publication May 8, 2014. From the *Department of Orthopaedics and †Division of Infectious Disease, R Adams Cowley Shock Trauma Center, University of Maryland School of Medicine, Baltimore, MD. The authors report no conflict of interest. Reprints: Robert V. O’Toole, MD, Department of Orthopaedics, R Adams Cowley Shock Trauma Center, University of Maryland School of Medicine, 22 S. Greene St, T1R77, Baltimore, MD 21201 (e-mail: rvo3@ yahoo.com). Copyright © 2014 by Lippincott Williams & Wilkins
J Orthop Trauma Volume 29, Number 1, January 2015
INTRODUCTION
Infection after internal fixation of fractures is not infrequent and poses significant problems. Published infection rates after internal fixation range from 1% for closed femoral fractures to 67% for open pilon fractures.1 In the setting of a fracture that has not yet healed, occurrence of infection poses the dilemma of removing the fixation device or retaining it while attempting to eliminate the infection. Infection after internal fixation increases the patient’s morbidity and chance for limb loss in addition to increasing the use of resources, cost of care, and cost to society. Patzakis et al,2 with a prospective, randomized controlled study, showed that the infection rate associated with open fractures treated with cephalothin was 2.3% compared with 13.9% in the control group. This study established the use of cephalosporins for open fractures, which was supported by subsequent work.3 Many data and recommendations regarding prophylactic antibiotic treatment for closed and open fractures are more than 25 years old.2–7 More recent studies8,9 have focused on treatment strategies for infections after internal fixation, but data on bacterial species and susceptibility are lacking. To our knowledge, no current clinical studies in the English language literature are devoted to both the speciation and the antibiotic resistance in infections after internal fixation of fractures. Our study was designed to determine the current bacterial speciation and the presence of drug resistance in deep infections occurring secondary to internal fixation of the extremities, pelvis, and acetabulum.
PATIENTS AND METHODS After receiving institutional review board approval of the study protocol, we conducted a retrospective analysis of all orthopaedic surgeries performed at our institution from December 1, 2006, to December 1, 2010, by screening for Current Procedural Terminology (CPT)10 and International www.jorthotrauma.com |
7
Torbert et al
Classification of Diseases, Ninth Revision (ICD-9)11 codes suspicious for postoperative infection (Table 1). Nine hundred thirty-eight patients had at least one of the described CPT or ICD-9 codes. Inclusion criteria were treatment of extremity, pelvic, or acetabular fracture with open reduction and internal fixation or intramedullary nailing, deep surgical site infection requiring surgical intervention with verification of deep surgical site infection (deep to the subcutaneous layer) during surgical debridement, surgical debridement performed at our institution within 12 months of internal fixation, and the availability of intraoperative culture data. External fixation pin site infections were not considered deep surgical site infections and were excluded. This yielded 211 patients (152 males, 59 females; average age, 45 years; age range, 16– 95 years) with 214 separate deep infections. Infections with recurrences were not counted more than once; however, multiple infections in a single patient in anatomically distinct sites were counted separately. The location of fracture, presence and type of open fracture, timing of infection presentation, species of bacteria, and presence of clinically relevant antibiotic resistance were recorded for each patient. Clinically relevant antibiotic resistance was defined for each bacterial group (Table 2). Resistance of Staphylococcus aureus is clinically relevant in the setting of methicillin-resistant S. aureus (MRSA). Antibiotic resistance does occur in coagulase-negative staphylococci. However, it is already well established and therefore not clinically relevant. More than 80% of coagulase-negative staphylococcal isolates are resistant to methicillin.12 Additionally, coagulase-negative staphylococci often develop resistance to methicillin in vivo despite culture data indicating that it is sensitive.13 Therefore, serious infections with both methicillin-sensitive and methicillin-resistant coagulase-negative staphylococci are often treated with vancomycin, making methicillin resistance in coagulase-negative staphylococci clinically irrelevant. Evaluation of resistance in streptococcal species is also irrelevant because they have remained susceptible to most antibiotics. Reproducible standards to interpret susceptibility results for gram-positive rods and anaerobes are not available. Therefore, it is hard to establish resistance in these organisms. Resistance in gram-negative rods (GNRs) is increasing, and antimicrobial choices vary based on susceptibility data. Our definition of multidrug-resistant GNR was based on susceptibility to fewer than 3 of the following antimicrobial classes: cephalosporins, penicillins, quinolones, aminoglycosides, and carbapenems (Table 2). The percentage of postoperative infections containing each bacterial species was determined. The percentage of clinically relevant antibiotic resistance was determined for each species of bacteria. Patients with open fractures treated initially at our center received intravenously administered antibiotics as soon as possible during the resuscitation process and regularly until taken to the operating room for debridement. Preoperative antibiotics within 1 hour of the procedure and 24 hours of postoperative antibiotics were administered to all patients (open and closed fractures) undergoing open reduction and internal fixation. Intravenously administered cefazolin was used unless the patient was allergic to penicillins or cephalosporins. Our normal algorithm for treating type III
8
| www.jorthotrauma.com
J Orthop Trauma Volume 29, Number 1, January 2015
TABLE 1. CPT and ICD-9 Codes Searched CPT Codes 27360
27640 27641 28120
28122
11011
11012
11043
11044
10180 ICD-9 Codes 996.67
998.59 730.20 730.25 730.26 730.27
Procedure Partial excision (craterization, saucerization, or diaphysectomy), bone, femur, proximal tibia, and/or fibula (eg, osteomyelitis or bone abscess) Partial excision (craterization, saucerization, or diaphysectomy), bone (eg, osteomyelitis); tibia Partial excision (craterization, saucerization, or diaphysectomy), bone (eg, osteomyelitis); fibula Partial excision (craterization, saucerization, sequestrectomy, or diaphysectomy), bone (eg, osteomyelitis or bossing); talus or calcaneus Partial excision (craterization, saucerization, sequestrectomy, or diaphysectomy), bone (eg, osteomyelitis or bossing); tarsal or metatarsal bone, except talus or calcaneus Debridement including removal of foreign material at the site of an open fracture and/or an open dislocation (eg, excisional debridement); skin, subcutaneous tissue, muscle fascia, and muscle Debridement including removal of foreign material at the site of an open fracture and/or an open dislocation (eg, excisional debridement); skin, subcutaneous tissue, muscle fascia, muscle, and bone Debridement, muscle and/or fascia (includes epidermis, dermis, and subcutaneous tissue, if performed); first 20 cm2 or less Debridement, bone (includes epidermis, dermis, subcutaneous tissue, muscle, and/or fascia, if performed); first 20 cm2 or less Incision and drainage, complex, postoperative wound infection Diagnosis Infection and inflammatory reaction because of other internal orthopaedic device, implant, and graft: bone growth stimulator (electrode), internal fixation device (pin, rod, screw) Other postoperative infection: postoperative abscess or septicemia Unspecified osteomyelitis, site unspecified Unspecified osteomyelitis, pelvic region and thigh Unspecified osteomyelitis, lower leg Unspecified osteomyelitis, ankle and foot
open fractures did not include an aminoglycoside. Penicillin was added for barnyard-type injuries. At the time of presentation with deep infection, the treating surgeon’s preference determined whether the patient received preoperative antibiotics. As a general rule, we attempt to hold antibiotics before surgical debridement during which cultures will be obtained unless the patient is in septic shock or will otherwise be harmed by holding antibiotics. We often obtain multiple aerobic and anaerobic surgical cultures as soon as infection is visible, before performing debridement and irrigation. The culture is obtained with a swab, and a tissue sample is added if possible. Adequate debridement was performed, and the decision to retain, remove, or exchange the fixation device was made by the treating surgeon. Standard microbiological procedures14 were used for processing all specimens. For all swabs and tissue samples, Ó 2014 Lippincott Williams & Wilkins
J Orthop Trauma Volume 29, Number 1, January 2015
TABLE 2. Our Definitions of Clinically Relevant Resistance for Each Group Bacterial Species
Clinically Relevant Resistance
Staphylococcus aureus Coagulase-negative Staphylococcus Streptococcus Enterococcus GNRs
Gram-positive rods Anaerobes
MRSA No clinically relevant resistance No clinically relevant resistance Vancomycin-resistant Enterococcus Multidrug resistant Susceptible to fewer than 3 of the following: cephalosporins, penicillins, quinolones, aminoglycosides, carbapenems No clinically relevant resistance No clinically relevant resistance
a Gram stain and aerobic and anaerobic cultures were performed. The Gram stain was performed to identify organisms. The aerobic cultures were plated on blood, chocolate, MacConkey, and phenylethyl alcohol agars. Plates were examined each day for 5 days. The MacConkey plate can be discarded if no growth is evident on the plate after 48 hours of incubation. If growth occurs on the plates, the organisms are identified. Susceptibility testing is conducted as appropriate on each organism isolated. For anaerobic cultures, Brucella, Bacteroides Bile Esculin Agar, and Brucella Laked Blood Agar with Kanamycin and Vancomycin, thioglycollate broth, and phenylethyl alcohol blood agars were used. All anaerobic specimen plates were placed into an anaerobic jar or bag within 20 minutes after inoculation. Anaerobic cultures were routinely kept for 5 days (longer if endocarditis or certain organisms were suspected). Susceptibility tests were not routinely conducted on anaerobic organisms. A 2-tailed Fisher exact text was conducted for statistical analysis when determining the association of 2 variables. P values #0.05 were considered to be statistically significant. When appropriate, post hoc analyses were conducted with a “test comparing 2 proportions,” using a power of 0.80 and a statistical significance of P = 0.05. Anatomic locations of infections are presented in Table 3. One hundred thirteen infections occurred in closed fractures and 101 in open fractures (4 type I, 13 type II, and 84 type III).
RESULTS Of the 214 infections, 198 (93%) grew at least one species of bacteria in culture, with an average of 1.5 species per infection (range, 1–5 species). Sixteen (7%) had negative cultures despite having purulence found at the time of surgical debridement. The number of infections with each type of bacteria and the rate of clinically relevant resistance within each type are presented in Figure 1 and Table 4. The number and genus of gram-positive cocci, GNR, gram-positive rods, and anaerobes that were cultured are listed in Table 5. Of the 249 cultures that were positive for S. aureus, GNR, or Enterococcus, 79 (32%) had clinically relevant resistance. Overall, 76 patients (36% of 211 patients) had at least one clinically Ó 2014 Lippincott Williams & Wilkins
Bacterial Speciation and Antibiotic Resistance
relevant resistant bacterium; 2 patients each had 2 clinically relevant resistant bacteria. The rates of GNR infection in open (32%) and closed (33%) fractures were similar (Table 6), with no statistically significant difference (P = 0.885) assessed by Fisher exact test. The rate of GNR infection was 63% in pelvic, acetabular, and proximal femoral fractures, compared with 27% in all other fractures (Table 7). This difference was statistically significant (P = 0.0002). The types of bacteria found in the pelvic, acetabular, and proximal femoral infections are presented in Table 8. Postoperative infections occurred an average of 11 weeks after surgery (range, 3 days to 51 weeks) and were more prevalent early on. Infections in cases of closed fractures seemed to present sooner than those in cases of open fractures, as indicated by the graph shown in Figure 2. However, 24% of closed fractures presented within 2 weeks, compared with 23% of open fractures. Statistical analysis with Fisher exact test showed no statistically significant difference (P = 0.87). Twenty-three percent of all infections presented by the end of week 2, and 50% of all infections presented by the end of week 5. All infections were grouped and displayed over time (Fig. 3). Sixty-two percent of infections with GNR presented within 2 weeks of open reduction and internal fixation, compared with 47% of all other infections (P = 0.0001) (Fig. 4). This difference remained statistically significant when data were stratified to open (P = 0.002) and closed (P = 0.05) fractures. Infections with GNR were also more likely to present earlier than S. aureus, with 62% presenting within 2 weeks compared with 25% of S. aureus (P = 0.0001) (Fig. 5). Infections with S. aureus were more likely to present later than all other infections, with 75% presenting after 2 weeks, compared with 53% of all other
TABLE 3. Number of Infections at Each Fracture Site Location Diaphyseal tibia Distal tibia Proximal tibia Ankle Acetabulum Distal femur Diaphyseal femur Pelvis Calcaneus Proximal humerus Proximal femur Talus Wrist Distal humerus Elbow Forearm Foot Patella Hand Diaphyseal humerus Total
Number of Infections 40 32 31 22 17 11 10 8 6 6 5 5 5 4 3 2 2 2 2 1 214
www.jorthotrauma.com |
9
J Orthop Trauma Volume 29, Number 1, January 2015
Torbert et al
FIGURE 1. Bar graph shows number of infections with each type of bacteria and number of infections with and without clinically relevant resistance within each type. CNS, coagulase-negative Staphylococcus; GPR, gram-positive rods.
infections (P = 0.0001) (Fig. 6). This difference remained statistically significant when data were stratified to open (P = 0.0005) and closed (P = 0.0076) fractures. Three patients had infections in more than one location. The first patient had ipsilateral distal humeral and radial shaft fractures with MRSA. The onsets of the infection occurred 6 months apart. The second patient had ipsilateral acetabular and ankle fractures. The acetabular Klebsiella infection was diagnosed first, and 1 month later, the coagulase-negative staphylococcus infection in the ankle was found. The third patient had bilateral fractures of the distal radii. The closed fracture of the distal radius was diagnosed with MRSA, and 2 months later, the open fracture of the distal radius was diagnosed with infection. However, the results of the cultures were negative, likely because of ongoing antibiotic treatment at the time of culture.
DISCUSSION To our knowledge, this is the largest series of postoperative infections after internal fixation to be presented in the English language literature and the only study that focuses on bacterial speciation and antibiotic resistance in infections after internal fixation. Our goal was to determine the current bacterial speciation and the presence of antibiotic resistance in cases of deep infection occurring secondary to internal fixation of the extremities, pelvis, and acetabulum. We chose to study clinically relevant resistance, which we defined as the presence of antibiotic resistance that necessitated the use of an antibiotic other than the standard antibiotic typically used for that species. This was defined for each group as described previously. Our rationale was that if a bacterial species was resistant to a particular antibiotic that would typically be administered as part of the treatment, it was clinically relevant and should be considered a clinically relevant resistance. Therefore, we limited our study of drug resistance to MRSA, vancomycin-resistant Enterococcus, and multidrug-resistant GNR because these are the main groups that pose clinically
10
| www.jorthotrauma.com
problematic drug resistance. Other types of bacteria do develop drug resistance. However, such resistance is either well established or not well defined because it does not have reproducible standards. Therefore, the choice of antimicrobial therapy is not dictated by the susceptibility data. In addition, defining the microbiology of infections after internal fixation of fractures in the present times has great impact on the choice of initial empirical therapy for such infections. For other infections, such as ventilator-associated pneumonia and septic shock, it has been shown that inappropriate initial antibiotic therapy results in higher morbidity and mortality,15 although this has not been clearly established for orthopaedic infections. We found that 36% of patients with deep infections occurring secondary to fracture fixation had at least one type of clinically relevant resistant bacteria. Fifty-six percent of infections had S. aureus present, with 58% of those (32% of all infections) being MRSA. Berkes et al8 found similar data in a recent study focusing on the retention of fixation devices in the setting of deep infection. They found Staphylococcus in 59% of infections, with 55% of those being MRSA. In a similar study, Rightmire et al9 found that 72% of cultures were S. aureus positive, with 77% of those being MRSA. However, the data on bacterial species and resistance reported by Rightmire et al9 were limited because not all patients had culture data available. All recent data, including our own, show that MRSA is a significant concern in cases of deep infection after internal fixation. Because the standard prophylactic preoperative antibiotic, cefazolin, does not provide adequate coverage for MRSA (MRSA, by definition, is resistant to cefazolin), the use of an alternative antibiotic, such as vancomycin, might seem intuitive. However, vancomycin is less bactericidal toward Staphylococcus than cefazolin and might therefore be less efficacious clinically.16,17 Any increase in effectiveness against MRSA might be further offset by a higher methicillinsensitive S. aureus (MSSA) infection rate. The prophylactic Ó 2014 Lippincott Williams & Wilkins
J Orthop Trauma Volume 29, Number 1, January 2015
Bacterial Speciation and Antibiotic Resistance
TABLE 4. Clinically Relevant Resistance by Bacteria Bacteria
Fracture Type
Infections With Specific Bacteria* (N = 214), n (%)
Infections With Clinically Relevant Resistance, n (%)
119 (56) 64 (30) 2 (1) 7 (3) 46 (21) 69† (32) 37 (17) 0 (0) 3 (1) 29 (14) 30 (14) 13 (6) 1 (0) 3 (1) 13 (6) 29 (13) 15 (7) 0 (0) 0 (0) 14 (7) 25 (12) 12 (6) 0 (0) 0 (0) 13 (6) 13 (6) 4 (2) 0 (0) 0 (0) 9 (4) 11 (5) 5 (2) 0 (0) 1 (0) 5 (2) 16 (7) 6 (3) 1 (0) 2 (1) 7 (3)
69 of 119 were MRSA (58)
Staphylococcus aureus Closed Type I open Type II open Type III open GNRs Closed Type I open Type II open Type III open Enterococcus Closed Type I open Type II open Type III open Anaerobes Closed Type I open Type II open Type III open Coagulase-negative Staphylococcus Closed Type I open Type II open Type III open Streptococcus Closed Type I open Type II open Type III open Gram-positive rods Closed Type I open Type II open Type III open Negative cultures Closed Type I open Type II open Type III open
3 of 69 were MDR (4)‡
6 of 30 were VRE (20)
NA
NA
NA
NA
NA
*The sum of the percentages of infections with specific bacteria does not equal 100% because many infections had more than 1 species of bacteria. †Sixty-nine infections had at least one type of GNR present. One hundred GNR types were isolated from the 69 infections. ‡Three of the 100 isolated GNR types of bacteria were multidrug resistant. MDR, multidrug resistant; NA, not applicable; VRE, vancomycin-resistant Enterococcus.
use of vancomycin in addition to cefazolin might reduce the MRSA rate while rendering the increase in MSSA infections unlikely. A prospective trial would be useful to determine the efficacy of prophylactic treatment using cefazolin and vancomycin in preventing both MSSA and MRSA infections. Interestingly, we found that 32% of infections had at least one GNR present, compared with 16.5% of infections in the study by Rightmire et al.9 Berkes et al8 included data on some specific GNR infections, but not all, rendering Ó 2014 Lippincott Williams & Wilkins
comparisons with our data impractical. In a recent update on health care–associated infections reported to the Centers for Disease Control and Prevention,18 12% of all cultures of the surgical site infections occurring secondary to open reduction and internal fixation, knee prostheses, and hip prostheses had GNR present. To explain why our GNR rate was higher than expected, we analyzed the GNR infections more closely. As noted, our normal algorithm for treating type III open fractures did not include an aminoglycoside. The www.jorthotrauma.com |
11
J Orthop Trauma Volume 29, Number 1, January 2015
Torbert et al
TABLE 5. Causative Organisms Isolated From Cultures Listed by Species
TABLE 6. Open and Closed Fractures With GNR Infection
Genus
Fractures
Gram-positive cocci Staphylococcus aureus Coagulase-negative Staphylococcus Beta hemolytic streptococci group A Beta hemolytic streptococci group B Beta hemolytic streptococci group G Beta hemolytic streptococci unspecified Enterococcus spp Total GNRs Acinetobacter spp Aeromonas spp Citrobacter spp Escherichia coli Enterobacter spp Hafnia spp Klebsiella spp Morganella spp Proteus spp Providencia spp Pseudomonas spp Serratia spp Unspecified Total Gram-positive rods Corynebacterium spp Unspecified Total Anaerobes Bacteroides spp Clostridium spp Lactobacillus spp Peptostreptococcus spp Propionibacterium spp Veillonella spp Total
Number 119 25 6
37 (33) 32 (32) 0 (0) 3 (23) 3 (18) 29 (35)
76 69 4 10 14 55
2 1 30 187 2 1 4 16 33 1 6 3 8 1 17 7 1 100* 7 4 11 4 5 1 14 4 1 29
recommended use of aminoglycosides in cases of type III open fractures5,6,19 has been presented in the literature over several decades. However, after their systematic review of the literature, the Council of the Surgical Infection Society found insufficient evidence to support the use of antibiotic coverage extending to gram-negative bacteria for open fractures.20 Aminoglycosides have been shown to cause nephrotoxicity21 and are more nephrotoxic if administered more than once daily22 or in patients with hypotension or shock.21,23 However, prophylactic administration in patients with open | www.jorthotrauma.com
Infections Without GNR
4
*Sixty-nine infections had at least one type of GNR present. One hundred GNR types were isolated from the 69 infections.
12
Closed Open Type I Type II Types I and II combined Type III
GNR Infections, n (%)
fractures has been found to be safe.24 Because we do not routinely use aminoglycosides in type III open fractures because of nephrotoxicity concerns, this was a likely source of increased GNR infections. However, as shown in Table 6, the rates of GNR infections in open (32%) and closed (33%) fractures were similar, with no statistically significant difference (P = 0.885) when compared by conducting Fisher exact test. Therefore, the risk of increased GNR infections cannot be completely explained on the basis of withholding an aminoglycoside-based prophylactic for type III open fractures. We also compared the rates of GNR infections of type III open fractures (35%) with those of less severe (types I and II combined) open fractures (18%) and found no statistically significant difference (P = 0.254). However, our number of types I and II open fractures with infections was low (17 fractures). Post hoc analysis with a test comparing 2 proportions, using a power of 0.80, revealed that we would need a 53% larger study size (327 infections) to obtain a statistically significant (P = 0.05) difference in type III open fractures and types I and II open fractures combined. Therefore, we do not disagree with the findings presented by Gustilo et al5 that the rate of GNR infection in type III open fractures is high and treatment with aminoglycosides in this group is reasonable. During our additional analysis, we found a particularly high rate of GNR infection in fractures involving the pelvis, acetabulum, and proximal femur. In addition, 90% of the infections in these locations occurred in closed fractures. We therefore compared this group of fractures with all other fractures and found that infections after internal fixation of pelvic, acetabular, and proximal femoral fractures are more likely (63% vs. 27%; P = 0.0002) to have at least one GNR present (Table 7). Some of the bacteria found in these infections (Table 8) are normal flora of the intestine or urinary tract and can exist as opportunistic infections in compromised patients. Several hypotheses exist to explain why pelvic, acetabular, and proximal femoral fractures are highly susceptible
TABLE 7. Pelvic, Acetabular, and Proximal Femoral Fractures With GNR Infections Compared With All Other Fractures Fractures Pelvic, acetabular, and proximal femoral All others
GNR Infections, n (%)
Infections Without GNR
19 (63)
11
50 (27)
134
Ó 2014 Lippincott Williams & Wilkins
J Orthop Trauma Volume 29, Number 1, January 2015
Bacterial Speciation and Antibiotic Resistance
TABLE 8. Fracture Type, Bacterial Species, and Antibiotic Resistance of Pelvic, Acetabular, and Proximal Femoral Fracture Infections Anatomic Location Pelvis
Acetabulum
Proximal femur
Patient Number
Fracture Type
1 2 3 4
Open Open Closed Closed
5
Closed
6
Closed
7
Closed
8 1 2 3 4
Closed Closed Closed Closed Closed
5 6 7 8 9
Closed Closed Closed Closed Open
10 11 12
Closed Closed Closed
13 14
Closed Closed
15 16
Closed Closed
17 1
Closed Closed
2 3
Closed Closed
4
Closed
5
Closed
Bacteria
Antibiotic Resistance
Escherichia coli E. coli Enterobacter cloacae Acinetobacter baumannii Coagulase-negative staphylococci E. coli Pseudomonas Enterococcus E. coli Bacteroides E. coli Enterococcus Staphylococcus aureus Klebsiella pneumoniae S. aureus S. aureus S. aureus Lactobacillus E. cloacae Coagulase-negative staphylococci S. aureus E. cloacae S. aureus K. pneumoniae Proteus E. coli K. pneumoniae Coagulase-negative staphylococci S. aureus GNR Group G streptococci Corynebacterium Enterobacter S. aureus Enterococcus Coagulase-negative staphylococci Candida Corynebacterium E. coli K. pneumoniae S. aureus Enterococcus Pseudomonas Proteus Negative E. coli K. pneumoniae Morganella Enterococcus Pseudomonas Enterococcus Coagulase-negative staphylococci E. coli S. aureus
No No No MDR No No No No No No No No MRSA No MRSA MRSA MRSA No No No MRSA No MRSA No No No No No No No No No No MSSA VRE No No No No No MSSA VRE No No NA No No No No No No No No MRSA
MDR, multidrug resistant; VRE, vancomycin-resistant Enterococcus; NA, not applicable.
Ó 2014 Lippincott Williams & Wilkins
www.jorthotrauma.com |
13
Torbert et al
J Orthop Trauma Volume 29, Number 1, January 2015
FIGURE 2. Time from open reduction and internal fixation until infection in closed and open fractures. Line graph shows that infections occurred earlier in closed fractures (solid line) compared with open fractures (dashed line). All cases of infection were diagnosed by 52 weeks because of the inclusion criteria of the study: deep surgical site infection treated at our institution within 12 months of internal fixation.
to GNR infections. The locations of these fractures and the resulting surgical incisions are in close proximity to the perineum and bowel flora. This could increase the chance of wound contamination by stool and urine, both intraoperatively and postoperatively. Also, pelvic and acetabular fractures could be contaminated through intra-abdominal injuries, either diagnosed or undiagnosed. In this retrospective study, we did not find intra-abdominal injuries in our patients with acetabular, pelvic, or proximal femoral fractures. Arciola et al25 studied bacterial cultures from orthopaedic revision procedures. They found a higher prevalence (44%) of Enterobacter, a GNR, in surgical incisions near the perineum; however, the number of surgical cases near the perineum was low (9 of 749 cases). They suggested that the different
pattern of bacteria found in the perineal region indicates the influence and contamination risks linked to specific anatomic locations and the related resident flora. In the current study, we found a high rate of GNR in pelvic, acetabular, and proximal femoral infections. The exact etiology is unknown and requires further investigation. We think this increased rate of GNR warrants consideration when choosing prophylactic (intraoperative and postoperative) antibiotic treatment and empiric treatment (before obtaining culture data) of infections occurring after open reduction and internal fixation of pelvic, acetabular, and proximal femoral fractures. Because of our high rate of GNR infections in type III open fractures and in high-risk closed fractures of the pelvis, acetabulum, and proximal femur, we currently are
FIGURE 3. Time line of presentation of all infections.
14
| www.jorthotrauma.com
Ó 2014 Lippincott Williams & Wilkins
J Orthop Trauma Volume 29, Number 1, January 2015
Bacterial Speciation and Antibiotic Resistance
FIGURE 4. Time line of presentation of GNR (dashed line) compared with all other infections (solid line).
considering adjusting our algorithm to add gram-negative coverage for these select fractures. Coagulase-negative staphylococci were found in 12% of the infections in the current study. This is a relatively low rate compared with that of S. aureus in this study (56%). We reviewed the existing literature and found a similar trend in infections after internal fixation. Rightmire et al9 found no coagulase-negative staphylococci in 56 closed and 23 open fracture infections. The authors of that study did find that approximately 73% of the cultures contained S. aureus. Arciola et al25 found that 30% of infections after internal fixation of fractures were caused by coagulase-negative staphylococci and that 43% were caused by S. aureus. The authors of that study found that among hip and knee arthroplasty infections, a higher percentage had coagulase-negative staphylococci (60%) compared with S. aureus (25%). Similar
findings have been reported by Fitzgerald.26 The difference in the rates might be explained by the differences in elective surgery versus emergent surgery and the soft tissue injury present in trauma patients. Regarding the timing of infection presentation, we found it surprising that S. aureus infections presented late compared with GNR and all other infections. Several hypothetical explanations might explain this finding. Trauma patients often are simultaneously injured in other anatomic sites and might require intense critical care support. Severe injuries and the critical nature of their illnesses predispose them to various infections and multiple surgical procedures, both requiring administration of antibiotics. Susceptible GNR, anaerobes, and streptococci are relatively easily killed by antibiotics. However, S. aureus, specifically MRSA, requires targeted therapy and might not be killed by standard antimicrobial
FIGURE 5. Time line of presentation of S. aureus (solid line) compared with GNR (dashed line). Ó 2014 Lippincott Williams & Wilkins
www.jorthotrauma.com |
15
Torbert et al
J Orthop Trauma Volume 29, Number 1, January 2015
FIGURE 6. Time line of presentation of S. aureus (solid line) compared with all other infections (dashed line).
prophylaxis or treatment. Moreover, GNR, anaerobes, enterococci, and streptococci are acquired into the surgical wound at the time of injury from either the patient’s own gut or the soil. They are not part of the normal skin flora, unlike the Staphylococcus sp. Therefore, after initial contamination, no ongoing source of these organisms is present, again unlike Staphylococcus sp. with which skin and wounds could be providing ongoing contamination. Staphylococcus sp. is known for forming biofilms and for the ability to survive in the presence of foreign bodies. Unfortunately, we did not specifically review all other injuries, presence of other infections, and use of antibiotics for other indications besides orthopaedic prophylaxis or infections. We were unable to find previously published data on timing and speciation of infections after open reduction and internal fixation. We thought that more virulent organisms, such as S. aureus, would present rapidly, whereas indolent organisms, such as GNR, would present later. We hypothesized that open infections might be more likely to have GNR and that those infections might present earlier because of poor open wound healing and soft tissue concerns. However, stratification into closed, open, and type III open fractures still showed that S. aureus presented later in both open and closed groups. Moreover, we found an unexpected trend for earlier presentation of infections in cases of closed fractures compared with those in cases of open fractures, although the differences were not statistically significant. This finding seems counterintuitive, and we are unable to explain it. Weaknesses of our study include the retrospective nature. All data were obtained through CPT and ICD-9 searches, which can result in missed cases. The retrospective design did not allow for a strict set of treatment guidelines to be followed. In addition, some patients were initially transferred from outside hospitals where treatment had already commenced and not all details regarding that treatment were available. In addition, our diagnosis of infection obtained from surgical cultures of patients referred from outside hospitals specifically for infections might have been delayed. The
16
| www.jorthotrauma.com
strengths of the study are the large number of patients and the integrated team of orthopaedic surgery and infectious disease professionals who managed the orthopaedic infections and provided careful scrutiny of all infections.
CONCLUSIONS These data provide a modern snapshot of bacterial species and resistance in orthopaedic infections after fracture fixation and might be useful in designing both prophylactic and treatment antibiotic protocols and future studies. Staphylococcus aureus and GNR are most often found in cases of deep postoperative infection after fixation. Of the cultures that were positive for S. aureus, GNR, or Enterococcus, 32% had clinically relevant resistance. Overall, 36% of patients had at least one clinically relevant resistant bacterium. As is reflected in national trends for non-orthopaedic infections, our percentage of MRSA is greater than that of MSSA. Similar to other fracture fixation studies, we found a lower rate of coagulase-negative staphylococci than of S. aureus. We found it surprising that S. aureus infections presented late compared with GNR and all other infections. In addition, we found an unexpected trend toward earlier presentation of infections in cases of closed fractures compared with open fractures, although the differences were not statistically significant. Our overall rate of GNR infection was high compared with previously reported rates, but the rate of GNR multidrug resistance was low. We found statistically significant increases in the rate of GNR infection of the pelvis, acetabulum, and proximal femur compared with other fracture sites. We also observed a high rate of GNR infection in our type III open fractures because we have not included aminoglycoside coverage because of nephrotoxicity concerns. We currently are considering adjusting our algorithm to add gram-negative coverage for these type III open fractures and fractures of the pelvis, acetabulum, and proximal femur. Ó 2014 Lippincott Williams & Wilkins
J Orthop Trauma Volume 29, Number 1, January 2015
ACKNOWLEDGMENTS The authors thank Senior Editor and Writer Dori Kelly, MA, University of Maryland School of Medicine, for valuable assistance with the manuscript. REFERENCES 1. Bucholz RW, Heckman JD, Court-Brown C, et al. Rockwood and Green’s Fractures in Adults. 6th ed. Philadelphia, PA: Lippincott Williams & Wilkins; 2005. 2. Patzakis MJ, Harvey JP Jr, Ivler D. The role of antibiotics in the management of open fractures. J Bone Joint Surg Am. 1974;56:532–541. 3. Gustilo RB, Anderson JT. Prevention of infection in the treatment of one thousand and twenty-five open fractures of long bones: retrospective and prospective analyses. J Bone Joint Surg Am. 1976;58:453–458. 4. Patzakis MJ, Wilkins J, Moore TM. Use of antibiotics in open tibial fractures. Clin Orthop Relat Res. 1983;178:31–35. 5. Gustilo RB, Mendoza RM, Williams DN. Problems in the management of type III (severe) open fractures: a new classification of type III open fractures. J Trauma. 1984;24:742–746. 6. Patzakis MJ, Bains RS, Lee J, et al. Prospective, randomized, doubleblind study comparing single-agent antibiotic therapy, ciprofloxacin, to combination antibiotic therapy in open fracture wounds. J Orthop Trauma. 2000;14:529–533. 7. Dzwonkowska J, Kurlenda J, Baczkowski B, et al. The effect of antibiotic therapy on the incidence of Staphylococcus aureus infections in orthopaedic patients. Ortop Traumatol Rehabil. 2007;9:532–547. 8. Berkes M, Obremskey WT, Scannell B, et al. Maintenance of hardware after early postoperative infection following fracture internal fixation. J Bone Joint Surg Am. 2010;92:823–828. 9. Rightmire E, Zurakowski D, Vrahas M. Acute infections after fracture repair: management with hardware in place. Clin Orthop Relat Res. 2008; 466:466–472. 10. Thorwarth W Jr ed. Current Procedural Terminology 2012. 4th ed. Chicago, IL: American Medical Association; 2011. 11. Ingenix. ICD-9-CM 2012 Expert for Physicians. Vol 1 and 2. Eden Prairie, MN: Ingenix; 2011. 12. Diekema DJ, Pfaller MA, Schmitz FJ, et al. Survey of infections due to Staphylococcus species: frequency of occurrence and antimicrobial susceptibility of isolates collected in the United States, Canada, Latin America, Europe, and the Western Pacific region for the SENTRY antimicrobial surveillance program, 1997-1999. Clin Infect Dis. 2001; 32(suppl 2):S114–S132.
Ó 2014 Lippincott Williams & Wilkins
Bacterial Speciation and Antibiotic Resistance 13. Archer GL, Climo MW. Antimicrobial susceptibility of coagulase-negative staphylococci. Antimicrob Agents Chemother. 1994;38:2231–2237. 14. University of Maryland Medical Center Microbiology Laboratory. Transudates and Exudates Procedure Manual. Baltimore, MD: University of Maryland; 2012. 15. Kollef MH, Sherman G, Ward S, et al. Inadequate antimicrobial treatment of infections: a risk factor for hospital mortality among critically ill patients. Chest. 1999;115:462–474. 16. Kapusnik-Uner JE, Sande MA, Chambers HF. Antimicrobial agents: tetracyclines, chloramphenicol, erythromycin, and miscellaneous antibacterial agents. In: Hardman JG, Limbird LE, Molinoff PB, et al, eds. Goodman & Gilman’s The Pharmacological Basis of Therapeutics. 9th ed. New York, NY: McGraw-Hill; 1996:1146. 17. Tice AD, Hoaglund PA, Shoultz DA. Risk factors and treatment outcomes in osteomyelitis. J Antimicrob Chemother. 2003;51:1261–1268. 18. Hidron AI, Edwards JR, Patel J, et al. NHSN annual update: antimicrobialresistant pathogens associated with healthcare-associated infections: annual summary of data reported to the National Healthcare Safety Network at the Centers for Disease Control and Prevention, 20062007. Infect Control Hosp Epidemiol. 2008;29:996–1011. 19. Russell GV Jr, King C, May CG, et al. Once daily high-dose gentamicin to prevent infection in open fractures of the tibial shaft: a preliminary investigation. South Med J. 2001;94:1185–1191. 20. Hauser CJ, Adams CA Jr, Eachempati SR; Council of the Surgical Infection Society. Surgical Infection Society guideline: prophylactic antibiotic use in open fractures: an evidence-based guideline. Surg Infect (Larchmt). 2006;7:379–405. 21. Oliveira JF, Silva CA, Barbieri CD, et al. Prevalence and risk factors for aminoglycoside nephrotoxicity in intensive care units. Antimicrob Agents Chemother. 2009;53:2887–2891. 22. Murry KR, McKinnon PS, Mitrzyk B, et al. Pharmacodynamic characterization of nephrotoxicity associated with once-daily aminoglycoside. Pharmacotherapy. 1999;19:1252–1260. 23. Boyer A, Gruson D, Bouchet S, et al. Aminoglycosides in septic shock: an overview, with specific consideration given to their nephrotoxic risk. Drug Saf. 2013;36:217–230. 24. Sorger JI, Kirk PG, Ruhnke CJ, et al. Once daily, high dose versus divided, low dose gentamicin for open fractures. Clin Orthop Relat Res. 1999;366:197–204. 25. Arciola CR, An YH, Campoccia D, et al. Etiology of implant orthopedic infections: a survey on 1027 clinical isolates. Int J Artif Organs. 2005;28: 1091–1100. 26. Fitzgerald RH Jr. Total hip arthroplasty sepsis: prevention and diagnosis. Orthop Clin North Am. 1992;23:259–264.
www.jorthotrauma.com |
17
Current Bacterial Speciation and Antibiotic Resistance in Deep Infections After Operative Fixation of Fractures Jesse T. Torbert, MD,* Manjari Joshi, MBBS,† Adrienne Moraff, BS,* Paul E. Matuszewski, MD,* Amanda Holmes, MS,* Andrew N. Pollak, MD,* and Robert V. O’Toole, MD*
Objectives: Infection after fracture fixation is a major source of
treatment of closed fractures of the pelvis, acetabulum, and proximal femur.
morbidity. Information regarding bacterial speciation and antibiotic resistance is lacking. We attempted to determine the speciation and drug resistance profiles associated with fracture fixation infections.
Key Words: bacterial speciation, antibiotic resistance, surgical site, infection, deep infection, fracture, trauma
Design: Retrospective study.
Level of Evidence: Prognostic Level IV. See Instructions for Authors for a complete description of levels of evidence.
Setting: Level I trauma center.
(J Orthop Trauma 2015;29:7–17)
Patients: Two hundred eleven patients with 214 infections underwent surgery for postoperative infection from December 2006 to December 2010. Deep postoperative infections within 12 months of fixation were included.
Intervention: None. Main Outcome Measurements: Incidence of each bacterial species and rate of clinically relevant resistance in Staphylococcus aureus, gram-negative rod (GNR), and Enterococcus species. The effect of timing of infection presentation and location of fracture on bacterial speciation was also investigated. Results: Fifty-six percent of infections had S. aureus present, with 58% of those (32% of all infections) being methicillin-resistant S. aureus. Thirty-two percent of infections had at least one GNR present, with only 4% of those being multidrug resistant. We found a marked increase in the rate of GNR infections of the pelvis, acetabulum, and proximal femur (63%) compared with other locations (27%), which was statistically significant (P = 0.0002).
Conclusions: At our center, S. aureus and GNR are most often found in deep postoperative infections after fixation. Methicillinresistant S. aureus is common in this population. Our GNR rate is high, but resistance in this group was low. The proportion of GNR infections in the pelvis, acetabulum, and proximal femur was high even in closed fractures. These data provide a modern snapshot of orthopaedic infections after fracture fixation and might be useful in designing future studies and protocols for antibiotic prophylactic treatment. We are considering the use of aminoglycosides in the Accepted for publication May 8, 2014. From the *Department of Orthopaedics and †Division of Infectious Disease, R Adams Cowley Shock Trauma Center, University of Maryland School of Medicine, Baltimore, MD. The authors report no conflict of interest. Reprints: Robert V. O’Toole, MD, Department of Orthopaedics, R Adams Cowley Shock Trauma Center, University of Maryland School of Medicine, 22 S. Greene St, T1R77, Baltimore, MD 21201 (e-mail: rvo3@ yahoo.com). Copyright © 2014 by Lippincott Williams & Wilkins
J Orthop Trauma Volume 29, Number 1, January 2015
INTRODUCTION
Infection after internal fixation of fractures is not infrequent and poses significant problems. Published infection rates after internal fixation range from 1% for closed femoral fractures to 67% for open pilon fractures.1 In the setting of a fracture that has not yet healed, occurrence of infection poses the dilemma of removing the fixation device or retaining it while attempting to eliminate the infection. Infection after internal fixation increases the patient’s morbidity and chance for limb loss in addition to increasing the use of resources, cost of care, and cost to society. Patzakis et al,2 with a prospective, randomized controlled study, showed that the infection rate associated with open fractures treated with cephalothin was 2.3% compared with 13.9% in the control group. This study established the use of cephalosporins for open fractures, which was supported by subsequent work.3 Many data and recommendations regarding prophylactic antibiotic treatment for closed and open fractures are more than 25 years old.2–7 More recent studies8,9 have focused on treatment strategies for infections after internal fixation, but data on bacterial species and susceptibility are lacking. To our knowledge, no current clinical studies in the English language literature are devoted to both the speciation and the antibiotic resistance in infections after internal fixation of fractures. Our study was designed to determine the current bacterial speciation and the presence of drug resistance in deep infections occurring secondary to internal fixation of the extremities, pelvis, and acetabulum.
PATIENTS AND METHODS After receiving institutional review board approval of the study protocol, we conducted a retrospective analysis of all orthopaedic surgeries performed at our institution from December 1, 2006, to December 1, 2010, by screening for Current Procedural Terminology (CPT)10 and International www.jorthotrauma.com |
7
Torbert et al
Classification of Diseases, Ninth Revision (ICD-9)11 codes suspicious for postoperative infection (Table 1). Nine hundred thirty-eight patients had at least one of the described CPT or ICD-9 codes. Inclusion criteria were treatment of extremity, pelvic, or acetabular fracture with open reduction and internal fixation or intramedullary nailing, deep surgical site infection requiring surgical intervention with verification of deep surgical site infection (deep to the subcutaneous layer) during surgical debridement, surgical debridement performed at our institution within 12 months of internal fixation, and the availability of intraoperative culture data. External fixation pin site infections were not considered deep surgical site infections and were excluded. This yielded 211 patients (152 males, 59 females; average age, 45 years; age range, 16– 95 years) with 214 separate deep infections. Infections with recurrences were not counted more than once; however, multiple infections in a single patient in anatomically distinct sites were counted separately. The location of fracture, presence and type of open fracture, timing of infection presentation, species of bacteria, and presence of clinically relevant antibiotic resistance were recorded for each patient. Clinically relevant antibiotic resistance was defined for each bacterial group (Table 2). Resistance of Staphylococcus aureus is clinically relevant in the setting of methicillin-resistant S. aureus (MRSA). Antibiotic resistance does occur in coagulase-negative staphylococci. However, it is already well established and therefore not clinically relevant. More than 80% of coagulase-negative staphylococcal isolates are resistant to methicillin.12 Additionally, coagulase-negative staphylococci often develop resistance to methicillin in vivo despite culture data indicating that it is sensitive.13 Therefore, serious infections with both methicillin-sensitive and methicillin-resistant coagulase-negative staphylococci are often treated with vancomycin, making methicillin resistance in coagulase-negative staphylococci clinically irrelevant. Evaluation of resistance in streptococcal species is also irrelevant because they have remained susceptible to most antibiotics. Reproducible standards to interpret susceptibility results for gram-positive rods and anaerobes are not available. Therefore, it is hard to establish resistance in these organisms. Resistance in gram-negative rods (GNRs) is increasing, and antimicrobial choices vary based on susceptibility data. Our definition of multidrug-resistant GNR was based on susceptibility to fewer than 3 of the following antimicrobial classes: cephalosporins, penicillins, quinolones, aminoglycosides, and carbapenems (Table 2). The percentage of postoperative infections containing each bacterial species was determined. The percentage of clinically relevant antibiotic resistance was determined for each species of bacteria. Patients with open fractures treated initially at our center received intravenously administered antibiotics as soon as possible during the resuscitation process and regularly until taken to the operating room for debridement. Preoperative antibiotics within 1 hour of the procedure and 24 hours of postoperative antibiotics were administered to all patients (open and closed fractures) undergoing open reduction and internal fixation. Intravenously administered cefazolin was used unless the patient was allergic to penicillins or cephalosporins. Our normal algorithm for treating type III
8
| www.jorthotrauma.com
J Orthop Trauma Volume 29, Number 1, January 2015
TABLE 1. CPT and ICD-9 Codes Searched CPT Codes 27360
27640 27641 28120
28122
11011
11012
11043
11044
10180 ICD-9 Codes 996.67
998.59 730.20 730.25 730.26 730.27
Procedure Partial excision (craterization, saucerization, or diaphysectomy), bone, femur, proximal tibia, and/or fibula (eg, osteomyelitis or bone abscess) Partial excision (craterization, saucerization, or diaphysectomy), bone (eg, osteomyelitis); tibia Partial excision (craterization, saucerization, or diaphysectomy), bone (eg, osteomyelitis); fibula Partial excision (craterization, saucerization, sequestrectomy, or diaphysectomy), bone (eg, osteomyelitis or bossing); talus or calcaneus Partial excision (craterization, saucerization, sequestrectomy, or diaphysectomy), bone (eg, osteomyelitis or bossing); tarsal or metatarsal bone, except talus or calcaneus Debridement including removal of foreign material at the site of an open fracture and/or an open dislocation (eg, excisional debridement); skin, subcutaneous tissue, muscle fascia, and muscle Debridement including removal of foreign material at the site of an open fracture and/or an open dislocation (eg, excisional debridement); skin, subcutaneous tissue, muscle fascia, muscle, and bone Debridement, muscle and/or fascia (includes epidermis, dermis, and subcutaneous tissue, if performed); first 20 cm2 or less Debridement, bone (includes epidermis, dermis, subcutaneous tissue, muscle, and/or fascia, if performed); first 20 cm2 or less Incision and drainage, complex, postoperative wound infection Diagnosis Infection and inflammatory reaction because of other internal orthopaedic device, implant, and graft: bone growth stimulator (electrode), internal fixation device (pin, rod, screw) Other postoperative infection: postoperative abscess or septicemia Unspecified osteomyelitis, site unspecified Unspecified osteomyelitis, pelvic region and thigh Unspecified osteomyelitis, lower leg Unspecified osteomyelitis, ankle and foot
open fractures did not include an aminoglycoside. Penicillin was added for barnyard-type injuries. At the time of presentation with deep infection, the treating surgeon’s preference determined whether the patient received preoperative antibiotics. As a general rule, we attempt to hold antibiotics before surgical debridement during which cultures will be obtained unless the patient is in septic shock or will otherwise be harmed by holding antibiotics. We often obtain multiple aerobic and anaerobic surgical cultures as soon as infection is visible, before performing debridement and irrigation. The culture is obtained with a swab, and a tissue sample is added if possible. Adequate debridement was performed, and the decision to retain, remove, or exchange the fixation device was made by the treating surgeon. Standard microbiological procedures14 were used for processing all specimens. For all swabs and tissue samples, Ó 2014 Lippincott Williams & Wilkins
J Orthop Trauma Volume 29, Number 1, January 2015
TABLE 2. Our Definitions of Clinically Relevant Resistance for Each Group Bacterial Species
Clinically Relevant Resistance
Staphylococcus aureus Coagulase-negative Staphylococcus Streptococcus Enterococcus GNRs
Gram-positive rods Anaerobes
MRSA No clinically relevant resistance No clinically relevant resistance Vancomycin-resistant Enterococcus Multidrug resistant Susceptible to fewer than 3 of the following: cephalosporins, penicillins, quinolones, aminoglycosides, carbapenems No clinically relevant resistance No clinically relevant resistance
a Gram stain and aerobic and anaerobic cultures were performed. The Gram stain was performed to identify organisms. The aerobic cultures were plated on blood, chocolate, MacConkey, and phenylethyl alcohol agars. Plates were examined each day for 5 days. The MacConkey plate can be discarded if no growth is evident on the plate after 48 hours of incubation. If growth occurs on the plates, the organisms are identified. Susceptibility testing is conducted as appropriate on each organism isolated. For anaerobic cultures, Brucella, Bacteroides Bile Esculin Agar, and Brucella Laked Blood Agar with Kanamycin and Vancomycin, thioglycollate broth, and phenylethyl alcohol blood agars were used. All anaerobic specimen plates were placed into an anaerobic jar or bag within 20 minutes after inoculation. Anaerobic cultures were routinely kept for 5 days (longer if endocarditis or certain organisms were suspected). Susceptibility tests were not routinely conducted on anaerobic organisms. A 2-tailed Fisher exact text was conducted for statistical analysis when determining the association of 2 variables. P values #0.05 were considered to be statistically significant. When appropriate, post hoc analyses were conducted with a “test comparing 2 proportions,” using a power of 0.80 and a statistical significance of P = 0.05. Anatomic locations of infections are presented in Table 3. One hundred thirteen infections occurred in closed fractures and 101 in open fractures (4 type I, 13 type II, and 84 type III).
RESULTS Of the 214 infections, 198 (93%) grew at least one species of bacteria in culture, with an average of 1.5 species per infection (range, 1–5 species). Sixteen (7%) had negative cultures despite having purulence found at the time of surgical debridement. The number of infections with each type of bacteria and the rate of clinically relevant resistance within each type are presented in Figure 1 and Table 4. The number and genus of gram-positive cocci, GNR, gram-positive rods, and anaerobes that were cultured are listed in Table 5. Of the 249 cultures that were positive for S. aureus, GNR, or Enterococcus, 79 (32%) had clinically relevant resistance. Overall, 76 patients (36% of 211 patients) had at least one clinically Ó 2014 Lippincott Williams & Wilkins
Bacterial Speciation and Antibiotic Resistance
relevant resistant bacterium; 2 patients each had 2 clinically relevant resistant bacteria. The rates of GNR infection in open (32%) and closed (33%) fractures were similar (Table 6), with no statistically significant difference (P = 0.885) assessed by Fisher exact test. The rate of GNR infection was 63% in pelvic, acetabular, and proximal femoral fractures, compared with 27% in all other fractures (Table 7). This difference was statistically significant (P = 0.0002). The types of bacteria found in the pelvic, acetabular, and proximal femoral infections are presented in Table 8. Postoperative infections occurred an average of 11 weeks after surgery (range, 3 days to 51 weeks) and were more prevalent early on. Infections in cases of closed fractures seemed to present sooner than those in cases of open fractures, as indicated by the graph shown in Figure 2. However, 24% of closed fractures presented within 2 weeks, compared with 23% of open fractures. Statistical analysis with Fisher exact test showed no statistically significant difference (P = 0.87). Twenty-three percent of all infections presented by the end of week 2, and 50% of all infections presented by the end of week 5. All infections were grouped and displayed over time (Fig. 3). Sixty-two percent of infections with GNR presented within 2 weeks of open reduction and internal fixation, compared with 47% of all other infections (P = 0.0001) (Fig. 4). This difference remained statistically significant when data were stratified to open (P = 0.002) and closed (P = 0.05) fractures. Infections with GNR were also more likely to present earlier than S. aureus, with 62% presenting within 2 weeks compared with 25% of S. aureus (P = 0.0001) (Fig. 5). Infections with S. aureus were more likely to present later than all other infections, with 75% presenting after 2 weeks, compared with 53% of all other
TABLE 3. Number of Infections at Each Fracture Site Location Diaphyseal tibia Distal tibia Proximal tibia Ankle Acetabulum Distal femur Diaphyseal femur Pelvis Calcaneus Proximal humerus Proximal femur Talus Wrist Distal humerus Elbow Forearm Foot Patella Hand Diaphyseal humerus Total
Number of Infections 40 32 31 22 17 11 10 8 6 6 5 5 5 4 3 2 2 2 2 1 214
www.jorthotrauma.com |
9
J Orthop Trauma Volume 29, Number 1, January 2015
Torbert et al
FIGURE 1. Bar graph shows number of infections with each type of bacteria and number of infections with and without clinically relevant resistance within each type. CNS, coagulase-negative Staphylococcus; GPR, gram-positive rods.
infections (P = 0.0001) (Fig. 6). This difference remained statistically significant when data were stratified to open (P = 0.0005) and closed (P = 0.0076) fractures. Three patients had infections in more than one location. The first patient had ipsilateral distal humeral and radial shaft fractures with MRSA. The onsets of the infection occurred 6 months apart. The second patient had ipsilateral acetabular and ankle fractures. The acetabular Klebsiella infection was diagnosed first, and 1 month later, the coagulase-negative staphylococcus infection in the ankle was found. The third patient had bilateral fractures of the distal radii. The closed fracture of the distal radius was diagnosed with MRSA, and 2 months later, the open fracture of the distal radius was diagnosed with infection. However, the results of the cultures were negative, likely because of ongoing antibiotic treatment at the time of culture.
DISCUSSION To our knowledge, this is the largest series of postoperative infections after internal fixation to be presented in the English language literature and the only study that focuses on bacterial speciation and antibiotic resistance in infections after internal fixation. Our goal was to determine the current bacterial speciation and the presence of antibiotic resistance in cases of deep infection occurring secondary to internal fixation of the extremities, pelvis, and acetabulum. We chose to study clinically relevant resistance, which we defined as the presence of antibiotic resistance that necessitated the use of an antibiotic other than the standard antibiotic typically used for that species. This was defined for each group as described previously. Our rationale was that if a bacterial species was resistant to a particular antibiotic that would typically be administered as part of the treatment, it was clinically relevant and should be considered a clinically relevant resistance. Therefore, we limited our study of drug resistance to MRSA, vancomycin-resistant Enterococcus, and multidrug-resistant GNR because these are the main groups that pose clinically
10
| www.jorthotrauma.com
problematic drug resistance. Other types of bacteria do develop drug resistance. However, such resistance is either well established or not well defined because it does not have reproducible standards. Therefore, the choice of antimicrobial therapy is not dictated by the susceptibility data. In addition, defining the microbiology of infections after internal fixation of fractures in the present times has great impact on the choice of initial empirical therapy for such infections. For other infections, such as ventilator-associated pneumonia and septic shock, it has been shown that inappropriate initial antibiotic therapy results in higher morbidity and mortality,15 although this has not been clearly established for orthopaedic infections. We found that 36% of patients with deep infections occurring secondary to fracture fixation had at least one type of clinically relevant resistant bacteria. Fifty-six percent of infections had S. aureus present, with 58% of those (32% of all infections) being MRSA. Berkes et al8 found similar data in a recent study focusing on the retention of fixation devices in the setting of deep infection. They found Staphylococcus in 59% of infections, with 55% of those being MRSA. In a similar study, Rightmire et al9 found that 72% of cultures were S. aureus positive, with 77% of those being MRSA. However, the data on bacterial species and resistance reported by Rightmire et al9 were limited because not all patients had culture data available. All recent data, including our own, show that MRSA is a significant concern in cases of deep infection after internal fixation. Because the standard prophylactic preoperative antibiotic, cefazolin, does not provide adequate coverage for MRSA (MRSA, by definition, is resistant to cefazolin), the use of an alternative antibiotic, such as vancomycin, might seem intuitive. However, vancomycin is less bactericidal toward Staphylococcus than cefazolin and might therefore be less efficacious clinically.16,17 Any increase in effectiveness against MRSA might be further offset by a higher methicillinsensitive S. aureus (MSSA) infection rate. The prophylactic Ó 2014 Lippincott Williams & Wilkins
J Orthop Trauma Volume 29, Number 1, January 2015
Bacterial Speciation and Antibiotic Resistance
TABLE 4. Clinically Relevant Resistance by Bacteria Bacteria
Fracture Type
Infections With Specific Bacteria* (N = 214), n (%)
Infections With Clinically Relevant Resistance, n (%)
119 (56) 64 (30) 2 (1) 7 (3) 46 (21) 69† (32) 37 (17) 0 (0) 3 (1) 29 (14) 30 (14) 13 (6) 1 (0) 3 (1) 13 (6) 29 (13) 15 (7) 0 (0) 0 (0) 14 (7) 25 (12) 12 (6) 0 (0) 0 (0) 13 (6) 13 (6) 4 (2) 0 (0) 0 (0) 9 (4) 11 (5) 5 (2) 0 (0) 1 (0) 5 (2) 16 (7) 6 (3) 1 (0) 2 (1) 7 (3)
69 of 119 were MRSA (58)
Staphylococcus aureus Closed Type I open Type II open Type III open GNRs Closed Type I open Type II open Type III open Enterococcus Closed Type I open Type II open Type III open Anaerobes Closed Type I open Type II open Type III open Coagulase-negative Staphylococcus Closed Type I open Type II open Type III open Streptococcus Closed Type I open Type II open Type III open Gram-positive rods Closed Type I open Type II open Type III open Negative cultures Closed Type I open Type II open Type III open
3 of 69 were MDR (4)‡
6 of 30 were VRE (20)
NA
NA
NA
NA
NA
*The sum of the percentages of infections with specific bacteria does not equal 100% because many infections had more than 1 species of bacteria. †Sixty-nine infections had at least one type of GNR present. One hundred GNR types were isolated from the 69 infections. ‡Three of the 100 isolated GNR types of bacteria were multidrug resistant. MDR, multidrug resistant; NA, not applicable; VRE, vancomycin-resistant Enterococcus.
use of vancomycin in addition to cefazolin might reduce the MRSA rate while rendering the increase in MSSA infections unlikely. A prospective trial would be useful to determine the efficacy of prophylactic treatment using cefazolin and vancomycin in preventing both MSSA and MRSA infections. Interestingly, we found that 32% of infections had at least one GNR present, compared with 16.5% of infections in the study by Rightmire et al.9 Berkes et al8 included data on some specific GNR infections, but not all, rendering Ó 2014 Lippincott Williams & Wilkins
comparisons with our data impractical. In a recent update on health care–associated infections reported to the Centers for Disease Control and Prevention,18 12% of all cultures of the surgical site infections occurring secondary to open reduction and internal fixation, knee prostheses, and hip prostheses had GNR present. To explain why our GNR rate was higher than expected, we analyzed the GNR infections more closely. As noted, our normal algorithm for treating type III open fractures did not include an aminoglycoside. The www.jorthotrauma.com |
11
J Orthop Trauma Volume 29, Number 1, January 2015
Torbert et al
TABLE 5. Causative Organisms Isolated From Cultures Listed by Species
TABLE 6. Open and Closed Fractures With GNR Infection
Genus
Fractures
Gram-positive cocci Staphylococcus aureus Coagulase-negative Staphylococcus Beta hemolytic streptococci group A Beta hemolytic streptococci group B Beta hemolytic streptococci group G Beta hemolytic streptococci unspecified Enterococcus spp Total GNRs Acinetobacter spp Aeromonas spp Citrobacter spp Escherichia coli Enterobacter spp Hafnia spp Klebsiella spp Morganella spp Proteus spp Providencia spp Pseudomonas spp Serratia spp Unspecified Total Gram-positive rods Corynebacterium spp Unspecified Total Anaerobes Bacteroides spp Clostridium spp Lactobacillus spp Peptostreptococcus spp Propionibacterium spp Veillonella spp Total
Number 119 25 6
37 (33) 32 (32) 0 (0) 3 (23) 3 (18) 29 (35)
76 69 4 10 14 55
2 1 30 187 2 1 4 16 33 1 6 3 8 1 17 7 1 100* 7 4 11 4 5 1 14 4 1 29
recommended use of aminoglycosides in cases of type III open fractures5,6,19 has been presented in the literature over several decades. However, after their systematic review of the literature, the Council of the Surgical Infection Society found insufficient evidence to support the use of antibiotic coverage extending to gram-negative bacteria for open fractures.20 Aminoglycosides have been shown to cause nephrotoxicity21 and are more nephrotoxic if administered more than once daily22 or in patients with hypotension or shock.21,23 However, prophylactic administration in patients with open | www.jorthotrauma.com
Infections Without GNR
4
*Sixty-nine infections had at least one type of GNR present. One hundred GNR types were isolated from the 69 infections.
12
Closed Open Type I Type II Types I and II combined Type III
GNR Infections, n (%)
fractures has been found to be safe.24 Because we do not routinely use aminoglycosides in type III open fractures because of nephrotoxicity concerns, this was a likely source of increased GNR infections. However, as shown in Table 6, the rates of GNR infections in open (32%) and closed (33%) fractures were similar, with no statistically significant difference (P = 0.885) when compared by conducting Fisher exact test. Therefore, the risk of increased GNR infections cannot be completely explained on the basis of withholding an aminoglycoside-based prophylactic for type III open fractures. We also compared the rates of GNR infections of type III open fractures (35%) with those of less severe (types I and II combined) open fractures (18%) and found no statistically significant difference (P = 0.254). However, our number of types I and II open fractures with infections was low (17 fractures). Post hoc analysis with a test comparing 2 proportions, using a power of 0.80, revealed that we would need a 53% larger study size (327 infections) to obtain a statistically significant (P = 0.05) difference in type III open fractures and types I and II open fractures combined. Therefore, we do not disagree with the findings presented by Gustilo et al5 that the rate of GNR infection in type III open fractures is high and treatment with aminoglycosides in this group is reasonable. During our additional analysis, we found a particularly high rate of GNR infection in fractures involving the pelvis, acetabulum, and proximal femur. In addition, 90% of the infections in these locations occurred in closed fractures. We therefore compared this group of fractures with all other fractures and found that infections after internal fixation of pelvic, acetabular, and proximal femoral fractures are more likely (63% vs. 27%; P = 0.0002) to have at least one GNR present (Table 7). Some of the bacteria found in these infections (Table 8) are normal flora of the intestine or urinary tract and can exist as opportunistic infections in compromised patients. Several hypotheses exist to explain why pelvic, acetabular, and proximal femoral fractures are highly susceptible
TABLE 7. Pelvic, Acetabular, and Proximal Femoral Fractures With GNR Infections Compared With All Other Fractures Fractures Pelvic, acetabular, and proximal femoral All others
GNR Infections, n (%)
Infections Without GNR
19 (63)
11
50 (27)
134
Ó 2014 Lippincott Williams & Wilkins
J Orthop Trauma Volume 29, Number 1, January 2015
Bacterial Speciation and Antibiotic Resistance
TABLE 8. Fracture Type, Bacterial Species, and Antibiotic Resistance of Pelvic, Acetabular, and Proximal Femoral Fracture Infections Anatomic Location Pelvis
Acetabulum
Proximal femur
Patient Number
Fracture Type
1 2 3 4
Open Open Closed Closed
5
Closed
6
Closed
7
Closed
8 1 2 3 4
Closed Closed Closed Closed Closed
5 6 7 8 9
Closed Closed Closed Closed Open
10 11 12
Closed Closed Closed
13 14
Closed Closed
15 16
Closed Closed
17 1
Closed Closed
2 3
Closed Closed
4
Closed
5
Closed
Bacteria
Antibiotic Resistance
Escherichia coli E. coli Enterobacter cloacae Acinetobacter baumannii Coagulase-negative staphylococci E. coli Pseudomonas Enterococcus E. coli Bacteroides E. coli Enterococcus Staphylococcus aureus Klebsiella pneumoniae S. aureus S. aureus S. aureus Lactobacillus E. cloacae Coagulase-negative staphylococci S. aureus E. cloacae S. aureus K. pneumoniae Proteus E. coli K. pneumoniae Coagulase-negative staphylococci S. aureus GNR Group G streptococci Corynebacterium Enterobacter S. aureus Enterococcus Coagulase-negative staphylococci Candida Corynebacterium E. coli K. pneumoniae S. aureus Enterococcus Pseudomonas Proteus Negative E. coli K. pneumoniae Morganella Enterococcus Pseudomonas Enterococcus Coagulase-negative staphylococci E. coli S. aureus
No No No MDR No No No No No No No No MRSA No MRSA MRSA MRSA No No No MRSA No MRSA No No No No No No No No No No MSSA VRE No No No No No MSSA VRE No No NA No No No No No No No No MRSA
MDR, multidrug resistant; VRE, vancomycin-resistant Enterococcus; NA, not applicable.
Ó 2014 Lippincott Williams & Wilkins
www.jorthotrauma.com |
13
Torbert et al
J Orthop Trauma Volume 29, Number 1, January 2015
FIGURE 2. Time from open reduction and internal fixation until infection in closed and open fractures. Line graph shows that infections occurred earlier in closed fractures (solid line) compared with open fractures (dashed line). All cases of infection were diagnosed by 52 weeks because of the inclusion criteria of the study: deep surgical site infection treated at our institution within 12 months of internal fixation.
to GNR infections. The locations of these fractures and the resulting surgical incisions are in close proximity to the perineum and bowel flora. This could increase the chance of wound contamination by stool and urine, both intraoperatively and postoperatively. Also, pelvic and acetabular fractures could be contaminated through intra-abdominal injuries, either diagnosed or undiagnosed. In this retrospective study, we did not find intra-abdominal injuries in our patients with acetabular, pelvic, or proximal femoral fractures. Arciola et al25 studied bacterial cultures from orthopaedic revision procedures. They found a higher prevalence (44%) of Enterobacter, a GNR, in surgical incisions near the perineum; however, the number of surgical cases near the perineum was low (9 of 749 cases). They suggested that the different
pattern of bacteria found in the perineal region indicates the influence and contamination risks linked to specific anatomic locations and the related resident flora. In the current study, we found a high rate of GNR in pelvic, acetabular, and proximal femoral infections. The exact etiology is unknown and requires further investigation. We think this increased rate of GNR warrants consideration when choosing prophylactic (intraoperative and postoperative) antibiotic treatment and empiric treatment (before obtaining culture data) of infections occurring after open reduction and internal fixation of pelvic, acetabular, and proximal femoral fractures. Because of our high rate of GNR infections in type III open fractures and in high-risk closed fractures of the pelvis, acetabulum, and proximal femur, we currently are
FIGURE 3. Time line of presentation of all infections.
14
| www.jorthotrauma.com
Ó 2014 Lippincott Williams & Wilkins
J Orthop Trauma Volume 29, Number 1, January 2015
Bacterial Speciation and Antibiotic Resistance
FIGURE 4. Time line of presentation of GNR (dashed line) compared with all other infections (solid line).
considering adjusting our algorithm to add gram-negative coverage for these select fractures. Coagulase-negative staphylococci were found in 12% of the infections in the current study. This is a relatively low rate compared with that of S. aureus in this study (56%). We reviewed the existing literature and found a similar trend in infections after internal fixation. Rightmire et al9 found no coagulase-negative staphylococci in 56 closed and 23 open fracture infections. The authors of that study did find that approximately 73% of the cultures contained S. aureus. Arciola et al25 found that 30% of infections after internal fixation of fractures were caused by coagulase-negative staphylococci and that 43% were caused by S. aureus. The authors of that study found that among hip and knee arthroplasty infections, a higher percentage had coagulase-negative staphylococci (60%) compared with S. aureus (25%). Similar
findings have been reported by Fitzgerald.26 The difference in the rates might be explained by the differences in elective surgery versus emergent surgery and the soft tissue injury present in trauma patients. Regarding the timing of infection presentation, we found it surprising that S. aureus infections presented late compared with GNR and all other infections. Several hypothetical explanations might explain this finding. Trauma patients often are simultaneously injured in other anatomic sites and might require intense critical care support. Severe injuries and the critical nature of their illnesses predispose them to various infections and multiple surgical procedures, both requiring administration of antibiotics. Susceptible GNR, anaerobes, and streptococci are relatively easily killed by antibiotics. However, S. aureus, specifically MRSA, requires targeted therapy and might not be killed by standard antimicrobial
FIGURE 5. Time line of presentation of S. aureus (solid line) compared with GNR (dashed line). Ó 2014 Lippincott Williams & Wilkins
www.jorthotrauma.com |
15
Torbert et al
J Orthop Trauma Volume 29, Number 1, January 2015
FIGURE 6. Time line of presentation of S. aureus (solid line) compared with all other infections (dashed line).
prophylaxis or treatment. Moreover, GNR, anaerobes, enterococci, and streptococci are acquired into the surgical wound at the time of injury from either the patient’s own gut or the soil. They are not part of the normal skin flora, unlike the Staphylococcus sp. Therefore, after initial contamination, no ongoing source of these organisms is present, again unlike Staphylococcus sp. with which skin and wounds could be providing ongoing contamination. Staphylococcus sp. is known for forming biofilms and for the ability to survive in the presence of foreign bodies. Unfortunately, we did not specifically review all other injuries, presence of other infections, and use of antibiotics for other indications besides orthopaedic prophylaxis or infections. We were unable to find previously published data on timing and speciation of infections after open reduction and internal fixation. We thought that more virulent organisms, such as S. aureus, would present rapidly, whereas indolent organisms, such as GNR, would present later. We hypothesized that open infections might be more likely to have GNR and that those infections might present earlier because of poor open wound healing and soft tissue concerns. However, stratification into closed, open, and type III open fractures still showed that S. aureus presented later in both open and closed groups. Moreover, we found an unexpected trend for earlier presentation of infections in cases of closed fractures compared with those in cases of open fractures, although the differences were not statistically significant. This finding seems counterintuitive, and we are unable to explain it. Weaknesses of our study include the retrospective nature. All data were obtained through CPT and ICD-9 searches, which can result in missed cases. The retrospective design did not allow for a strict set of treatment guidelines to be followed. In addition, some patients were initially transferred from outside hospitals where treatment had already commenced and not all details regarding that treatment were available. In addition, our diagnosis of infection obtained from surgical cultures of patients referred from outside hospitals specifically for infections might have been delayed. The
16
| www.jorthotrauma.com
strengths of the study are the large number of patients and the integrated team of orthopaedic surgery and infectious disease professionals who managed the orthopaedic infections and provided careful scrutiny of all infections.
CONCLUSIONS These data provide a modern snapshot of bacterial species and resistance in orthopaedic infections after fracture fixation and might be useful in designing both prophylactic and treatment antibiotic protocols and future studies. Staphylococcus aureus and GNR are most often found in cases of deep postoperative infection after fixation. Of the cultures that were positive for S. aureus, GNR, or Enterococcus, 32% had clinically relevant resistance. Overall, 36% of patients had at least one clinically relevant resistant bacterium. As is reflected in national trends for non-orthopaedic infections, our percentage of MRSA is greater than that of MSSA. Similar to other fracture fixation studies, we found a lower rate of coagulase-negative staphylococci than of S. aureus. We found it surprising that S. aureus infections presented late compared with GNR and all other infections. In addition, we found an unexpected trend toward earlier presentation of infections in cases of closed fractures compared with open fractures, although the differences were not statistically significant. Our overall rate of GNR infection was high compared with previously reported rates, but the rate of GNR multidrug resistance was low. We found statistically significant increases in the rate of GNR infection of the pelvis, acetabulum, and proximal femur compared with other fracture sites. We also observed a high rate of GNR infection in our type III open fractures because we have not included aminoglycoside coverage because of nephrotoxicity concerns. We currently are considering adjusting our algorithm to add gram-negative coverage for these type III open fractures and fractures of the pelvis, acetabulum, and proximal femur. Ó 2014 Lippincott Williams & Wilkins
J Orthop Trauma Volume 29, Number 1, January 2015
ACKNOWLEDGMENTS The authors thank Senior Editor and Writer Dori Kelly, MA, University of Maryland School of Medicine, for valuable assistance with the manuscript. REFERENCES 1. Bucholz RW, Heckman JD, Court-Brown C, et al. Rockwood and Green’s Fractures in Adults. 6th ed. Philadelphia, PA: Lippincott Williams & Wilkins; 2005. 2. Patzakis MJ, Harvey JP Jr, Ivler D. The role of antibiotics in the management of open fractures. J Bone Joint Surg Am. 1974;56:532–541. 3. Gustilo RB, Anderson JT. Prevention of infection in the treatment of one thousand and twenty-five open fractures of long bones: retrospective and prospective analyses. J Bone Joint Surg Am. 1976;58:453–458. 4. Patzakis MJ, Wilkins J, Moore TM. Use of antibiotics in open tibial fractures. Clin Orthop Relat Res. 1983;178:31–35. 5. Gustilo RB, Mendoza RM, Williams DN. Problems in the management of type III (severe) open fractures: a new classification of type III open fractures. J Trauma. 1984;24:742–746. 6. Patzakis MJ, Bains RS, Lee J, et al. Prospective, randomized, doubleblind study comparing single-agent antibiotic therapy, ciprofloxacin, to combination antibiotic therapy in open fracture wounds. J Orthop Trauma. 2000;14:529–533. 7. Dzwonkowska J, Kurlenda J, Baczkowski B, et al. The effect of antibiotic therapy on the incidence of Staphylococcus aureus infections in orthopaedic patients. Ortop Traumatol Rehabil. 2007;9:532–547. 8. Berkes M, Obremskey WT, Scannell B, et al. Maintenance of hardware after early postoperative infection following fracture internal fixation. J Bone Joint Surg Am. 2010;92:823–828. 9. Rightmire E, Zurakowski D, Vrahas M. Acute infections after fracture repair: management with hardware in place. Clin Orthop Relat Res. 2008; 466:466–472. 10. Thorwarth W Jr ed. Current Procedural Terminology 2012. 4th ed. Chicago, IL: American Medical Association; 2011. 11. Ingenix. ICD-9-CM 2012 Expert for Physicians. Vol 1 and 2. Eden Prairie, MN: Ingenix; 2011. 12. Diekema DJ, Pfaller MA, Schmitz FJ, et al. Survey of infections due to Staphylococcus species: frequency of occurrence and antimicrobial susceptibility of isolates collected in the United States, Canada, Latin America, Europe, and the Western Pacific region for the SENTRY antimicrobial surveillance program, 1997-1999. Clin Infect Dis. 2001; 32(suppl 2):S114–S132.
Ó 2014 Lippincott Williams & Wilkins
Bacterial Speciation and Antibiotic Resistance 13. Archer GL, Climo MW. Antimicrobial susceptibility of coagulase-negative staphylococci. Antimicrob Agents Chemother. 1994;38:2231–2237. 14. University of Maryland Medical Center Microbiology Laboratory. Transudates and Exudates Procedure Manual. Baltimore, MD: University of Maryland; 2012. 15. Kollef MH, Sherman G, Ward S, et al. Inadequate antimicrobial treatment of infections: a risk factor for hospital mortality among critically ill patients. Chest. 1999;115:462–474. 16. Kapusnik-Uner JE, Sande MA, Chambers HF. Antimicrobial agents: tetracyclines, chloramphenicol, erythromycin, and miscellaneous antibacterial agents. In: Hardman JG, Limbird LE, Molinoff PB, et al, eds. Goodman & Gilman’s The Pharmacological Basis of Therapeutics. 9th ed. New York, NY: McGraw-Hill; 1996:1146. 17. Tice AD, Hoaglund PA, Shoultz DA. Risk factors and treatment outcomes in osteomyelitis. J Antimicrob Chemother. 2003;51:1261–1268. 18. Hidron AI, Edwards JR, Patel J, et al. NHSN annual update: antimicrobialresistant pathogens associated with healthcare-associated infections: annual summary of data reported to the National Healthcare Safety Network at the Centers for Disease Control and Prevention, 20062007. Infect Control Hosp Epidemiol. 2008;29:996–1011. 19. Russell GV Jr, King C, May CG, et al. Once daily high-dose gentamicin to prevent infection in open fractures of the tibial shaft: a preliminary investigation. South Med J. 2001;94:1185–1191. 20. Hauser CJ, Adams CA Jr, Eachempati SR; Council of the Surgical Infection Society. Surgical Infection Society guideline: prophylactic antibiotic use in open fractures: an evidence-based guideline. Surg Infect (Larchmt). 2006;7:379–405. 21. Oliveira JF, Silva CA, Barbieri CD, et al. Prevalence and risk factors for aminoglycoside nephrotoxicity in intensive care units. Antimicrob Agents Chemother. 2009;53:2887–2891. 22. Murry KR, McKinnon PS, Mitrzyk B, et al. Pharmacodynamic characterization of nephrotoxicity associated with once-daily aminoglycoside. Pharmacotherapy. 1999;19:1252–1260. 23. Boyer A, Gruson D, Bouchet S, et al. Aminoglycosides in septic shock: an overview, with specific consideration given to their nephrotoxic risk. Drug Saf. 2013;36:217–230. 24. Sorger JI, Kirk PG, Ruhnke CJ, et al. Once daily, high dose versus divided, low dose gentamicin for open fractures. Clin Orthop Relat Res. 1999;366:197–204. 25. Arciola CR, An YH, Campoccia D, et al. Etiology of implant orthopedic infections: a survey on 1027 clinical isolates. Int J Artif Organs. 2005;28: 1091–1100. 26. Fitzgerald RH Jr. Total hip arthroplasty sepsis: prevention and diagnosis. Orthop Clin North Am. 1992;23:259–264.
www.jorthotrauma.com |
17
