Research Article Volume 14 Issue 2
1Department of Orthopaedic Surgery, University of Pretoria, South Africa
2Department of Statistics, Faculty of Natural and Agricultural Sciences, University of Pretoria, South Africa
Correspondence: Dr Jaco Viljoen, MBChB, MMed Orth (UP), Department of Orthopaedic Surgery, University of Pretoria, Steve Biko Academic Hospital, South Africa, Level 7, Bridge A, Steve Biko Academic Hospital University of Pretoria
Received: December 24, 2021 | Published: March 2, 2022
Citation: Landman I, Nolte S, Botha T, et al. Prevalence and antibiogram study of bacterial isolates in infection after fracture fixation at a tertiary hospital. MOJ Orthop Rheumatol. 2022;14(2):29-33. DOI: 10.15406/mojor.2022.14.00573
Background: Current treatment protocols for Infection After Fracture Fixation (IAFF) prescribes Vancomycin and Meropenem. This study assesses the appropriateness of empirical antibiotic regimens in the treatment of IAFF and the feasibility of reserving Vancomycin and Carbapenem antibiotics for selected cases. To achieve this aim, we determined the causative bacteria and antibiogram of orthopaedic implant infections and susceptibility to commonly used antibiotics at our institution.
Methods: A retrospective cross-sectional descriptive study of patients who were treated for IAFF between June 2016 and February 2020. The outcome measures include microscopy, culture, and sensitivity (MCS) of samples obtained intraoperatively under sterile conditions.
Results: 178 cases were collected with 65 cases excluded based on the exclusion criteria resulting in 113 cases and 136 cultures. Of the 113 cases investigated in the study, 58% (n=66) were males and 42% (n=47) were females with ages ranging from 12 to 90 years with a mean age of 41 years. 26 cultured no bacteria (77% culture-positive rate). Forty six percent (46%) of patients cultured a single organism, 54% were mixed bacterial cultures of which 37% tested positive for two organisms and 17% more than two organisms. 110 cultures were included in the antibiogram analysis. The most frequently cultured bacteria was Staphylococcus aureus 29% (n=32), followed by Enterobacter spp. 16 (n=16), Acinetobacter Baumannii Complex 9% (n=10), Pseudomonas aeruginosa 7% (n=8), Proteus 7% (8) and Klebsiella 7% (n=8). Bacteriological cultures revealed a Gram-Negative predominance (60%) including Enterobacter, Pseudomonas, Proteus and Acinetobacter. Among the Gram-Positive bacterial isolates recovered (40%), S. aureus was the most common, with only 4 Methicillin-resistant Staphylococcus Aureus (MRSA) cultures.
Sensitivity testing of the Gram-Positive bacteria revealed 84% sensitivity to Cloxacillin, with the remaining 16% sensitive to Vancomycin. Staphylococcus aureus infections were commonly methicillin sensitive. Of the Gram-Negative bacteria, 68% were sensitive to Cefepime. All Gram Negatives were resistant to Penicillin, Ampicillin, Co-Amoxiclav and Cloxacillin.
Conclusions: The results of this study show that IAFF may be safely treated at our institution with a combination of Cloxacillin and Cefepime, adjusting antibiotic cover once culture results are available.
Keywords: infection after fracture fixation, fracture related infection, antibiogram, empirical treatment, implant infection
Infection after fracture fixation (IAFF) is a feared and devastating complication with significant morbidity and, in some instances, mortality.1,2 This frequently encountered complication poses a unique challenge where both fracture union and control of sepsis are the ultimate goals.3,4
The current treatment protocol prescribes Vancomycin and Meropenem to cover respectively Methicillin-resistant Staphylococcus aureus (MRSA) and Cephalosporin resistant Gram-Negative microbes with the addition of Rifampicin to prevent biofilm formation.3,5–8 Much of the current treatment protocols for IAFF has been extrapolated from research on prosthetic joint infections (PJI) and data focusing on IAFF is scarce.9–16
Due to an emerging trend in the sensitivity of Gram-Negative bacteria to a 4th generation cephalosporin at our institution Steve Biko Academic Hospital (SBAH), we propose altering our institutional empirical antibiotic protocol by replacing the Carbapenems with a 4th generation cephalosporin. With this new protocol, we intend to reserve the Carbapenems and Vancomycin thereby limiting antibiotic resistance. We propose the hypothesis that IAFF may be treated safely with a less aggressive antibiotic choice, reserving the Carbapenems and Vancomycin for selected cases. It is difficult to substantiate such a claim due to the limited research on the microbiology of IAFF. The purpose of this cross-sectional descriptive study is to determine the most appropriate empirical antibiotic regime by identifying the common microbes and their antibiotic sensitivity.
Microscopy, culture, and sensitivity results of all patients who underwent debridement for IAFF at SBAH from June 2016 to February 2020 were included in the study. The sampling frame consisted of the theatre register as well as patient records, X-rays and microbiology results on LabTrack kept at SBAH. The theatre record book of SBAH was used to identify patients that underwent debridement of septic wounds. The patient’s name, hospital number, and the date of the procedure were recorded.
The patient file was requested from records and examined for the indication for debridement, the date of the original injury and original fixation. X-ray records were accessed via the Pax system to cross-reference to the patient file. LabTract was used to trace the MC&S results of the qualifying
A total number of 178 cases were collected with 65 cases excluded based on the exclusion criteria resulting in 113 cases and 136 cultures. From the 113 cases investigated in the study, 58% (n=66) were males and 42% (n=47) were females with ages ranging from 12 to 90 years with a mean age of 41 years (Table 1). Thirteen patients sustained open fractures and the remaining 100 patients closed injuries. Upper limb injuries accounted for 22% (n=25) of the cases and 78% (n=88) was due to injuries of the lower limbs. Details of anatomical distribution are listed in Table 2.
n=113 |
|
Mean Age (years) |
41 years (12-90) |
Sex (% male) |
58% (66) |
Table 1 Age and Sex
Upper limb 22% (n=25) |
|
Clavicle |
2 |
Humerus |
10 |
Forearm |
12 |
Lower limb 78% (n88) |
|
Femur |
32 |
Patella |
5 |
Tibia |
20 |
Foot and Ankle |
31 |
Table 2 Anatomical distribution
Of the total 113 patients, 26 cultured no bacteria (77% culture-positive rate). Forty six percent (46%) of patients cultured a single organism, 54% were mixed bacterial cultures of which 37% accounted for two organisms and 17% more than two organisms, resulting in 110 cultures being included in the antibiogram analysis.
The most frequently cultured bacteria was Staphylococcus aureus 29% (n=32), followed by Enterobacter spp. 14.5% (n=16), Acinetobacter Baumannii complex 9% (n=10), Pseudomonas aeruginosa 7% (n=8), Proteus 7% (8), and Klebsiella 7% (n=8). Table 3 depicts bacterial cultures from intraoperative samples in patients with IAFF.
Total n=110 |
|
Staphylococci |
|
Staphylococcus aureus |
32 |
Staphylococcus epidermidis |
1 |
Streptococci |
|
Streptococcus group a |
3 |
Streptococcus group b |
1 |
Enterococci |
|
Enterococcus faecalis |
4 |
Enterococcus faecium |
2 |
Enterobacterales |
|
Proteus mirabilis |
7 |
Proteus vulgaris |
1 |
Enterobacter cloacae complex |
16 |
Burkholderia Cepacia |
1 |
Klebsiella pneumoniae |
6 |
Klebsiella oxytoca |
2 |
Serratia marcescens |
6 |
Escherichia coli |
4 |
Morganella morganii subsp morganii |
2 |
Non-fermenting Gram-Negative bacilli |
|
Pseudomonas aeruginosa |
8 |
Acinetobacter baumannii |
10 |
Sphingomonadales |
|
Sphingomonas paucimobilis |
2 |
Burkholderiales |
|
Achromobacter denitrificans |
1 |
Alcaligenes faecalis subsp faecalis |
1 |
Table 3 Bacterial cultures from intraoperative samples in patients with IAFF
Bacteriological cultures revealed a 60% (n=66) Gram-Negative predominance including Enterobacter, Pseudomonas, Proteus, and Acinetobacter. Among the Gram-Positive bacterial isolates recovered (n=44), S. aureus was the most common (n=32), with only 4 MRSA cultures. Sensitivity testing of the Gram-Positive bacteria revealed an 84% sensitivity to Cloxacillin, with the remaining 16% sensitive to Vancomycin. Staphylococcus aureus infections were commonly Methicillin-sensitive, with only 4 of the 32 S. aureus cultures being MRSA. The susceptibility of the bacterial isolates to commonly used antibiotic regimes are demonstrated in Table 4. Of the Gram-Negative bacteria, 68% were sensitive to Cefepime. All Gram-Negatives were resistant to Penicillin, Ampicillin, Co-Amoxiclav, and Cloxacillin. The susceptibility of isolates to tested antibiotics are detailed in Table 5.
Antibiotic regimen |
Patients tested (n) |
Cover n (%) |
Cloxacillin |
32 |
28 (88%) |
Vancomycin |
4 |
4 (100%) |
Penicillin + Ampicillin |
29 |
4 (14%) |
Ceftazidime |
31 |
14(45%) |
Amoxicillin-Clavulanic acid |
28 |
5(18%) |
Piperacillin-Tazobactam |
30 |
15(50%) |
Cefepime |
47 |
32(68%) |
Cloxacillin + Cefepime |
79 |
60 (76%) |
Table 4 Susceptibility of bacterial isolates to commonly used antibiotic regimes
Antibiotic regimen |
Number of Gram-Positive isolates tested (n = 44) |
Susceptible Gram-Positive isolates. n, (%) |
Number of Gram-Negative isolates tested (n = 66) |
Susceptible Gram-Negative isolates. n, (%) |
Amoxicillin-clavulanic |
- |
- |
28 |
5 (18%) |
Penicillin/Ampicillin |
29 |
4 (14%) |
||
Cloxacillin |
32 |
28 (88%) |
1 |
0 (0%) |
Vancomycin |
5 |
5 (100%) |
1 |
0 (0%) |
Ceftazidime |
- |
- |
31 |
14 (45%) |
Cefepime |
- |
- |
47 |
32 (68%) |
Piperacillin/Tazobactam |
- |
- |
30 |
15 (50%) |
Ertapenem |
- |
- |
13 |
12 (92%) |
Imipenem |
- |
- |
18 |
10 (56%) |
Meropenem |
- |
- |
- |
- |
Table 5 Susceptibility of isolates to tested antibiotics
Infection after fracture fixation is a major cause of morbidity in trauma patients and an accurate microbiological and susceptibility profile of causative organisms in a given hospital is important to determine appropriate empirical antibiotic therapy.17–19
Negative cultures
Positive cultures were obtained in 75% of the studied patients, comparing well with the review by Venter et al.13 and Pincher et al.20 in which 79% and 81% had positive cultures respectively. This is noticeably lower than in comparable studies by Zimmeli et al.9 89%, Tuon et al.17 93%, and Khosravi et al.21 93.9%.9,17,21 We postulate that our lower positive culture rate may be due to the lack of routinely extended cultures, lack of sonification of explanted hardware, and routine administration of intraoperative prophylactic antibiotics at our institution.22–24
Gram stain
In the current study we demonstrated a Gram-Negative predominance as is common in developing countries, particularly Enterobacteriacea, P. aeruginosa, and Acinetobacter baumannii,17,25,26 this corresponds well with the study by Tuon et al.17 In contrast, Venter et al.13 demonstrated a 60% Gram-Positive predominance. This may be attributed to the fact that the study by Venter et al.13 included all patients with chronic osteomyelitis and not only patients with IAFF.
Staphylococcus aureus
Staphylococcus aureus was the most frequent culture in our study and agreed with the study by Arciola et al.,11 Toun et al.17 and Khosravi et al.21 which reported Staphylococci as the most prevalent organism.11,17,21 We demonstrated an MRSA rate of 3.6% that is significantly lower than in studies by Tuon et al.17 who reported a rate of 12% and Venter et al.21 demonstrating an MRSA rate of 15%. This finding supports our hypothesis that Vancomycin may be reserved for culture-proven resistant infections at our institution.
Sensitivity
The antimicrobial susceptibility test further revealed that:
Changing empirical therapy
There is no generally accepted consensus on how resistance data should be applied to treatment decisions about empirical antibiotic regimens.27 The decision to alter empirical antimicrobial regime is a complicated one, and cannot be made based on one single parameter. Considerations include cost, consequences of failed treatment, the correlation between laboratory resistance and clinical response, and overuse of reserve agents.28,29
In addition, there is no universally accepted resistance threshold at which to change empirical antibiotic therapy, neither can a universal threshold be applied to all diseases. As an example, the World Health Organisation (WHO) recommend an arbitrary resistance threshold of 5% for treating gonorrhoea.30 In the treatment of Malaria, the WHO recommends a change in first-line therapy when an arbitrary figure of 25% of patients treated relapse.31
The decision to change empirical antibiotic regimes should be made by a multidisciplinary team taking into consideration the severity of the condition and the potential consequences of ineffective treatment.19,27,32 It is, therefore, reasonable to accept relatively high rates of resistance to first-line agents in the management of non-life-threatening infections in otherwise healthy patients.
The results of this study suggest that Cloxacillin and Cefepime may safely be used as empirical treatment for IAFF in healthy hosts at our institution.
This study contributes to the existing literature on the management of IAFF and antibiotic stewardship, reserving the Carbapenems and Vancomycin for selected resistant cases.
The limitations of this study are typical of a retrospective study where the strength of the study is dependent on the quality of the medical records. A retrospective study design was chosen to utilise the vast amount of available data and to avoid the high cost associated with a prospective study. The sample size was small but similar to other studies focused on IAFF. The study was performed at a single tertiary centre treating complex trauma and limb reconstruction cases. This may limit the external validity of the study when applied to smaller centres. The high rate of negative cultures may predispose the study to bias.
The authors declare that this submission is in accordance with the principles laid down by the Responsible Research Publication Position Statements as developed at the Second World Conference on Research Integrity in Singapore, 2010. Before commencement of the study, ethical approval was obtained from the following ethical review board: University of Pretoria MMed committee, as well as the University of Pretoria Faculty of Health Sciences Ethics Committee; Ethics reference number: 524/2019. Patients’ records were anonymised by allocating a study number to each patient. Ethical principles as outlined by the World Medical Association Declaration of Helsinki were adhered to.
For this study formal consent was not required.
No funding was received for this study.
The authors declare authorship of this article and that they have followed sound scientific research practice. This research is original and does not transgress plagiarism policies.
IL: Data capture, data analysis, draft preparation, manuscript preparation
SN: Data capture, manuscript revision
TB: Statistical Analysis
JV: Study conceptualisation, data capture, data analysis, manuscript revision.
None.
The authors declare that there are no conflicts of interest.
©2022 Landman, et al. This is an open access article distributed under the terms of the, which permits unrestricted use, distribution, and build upon your work non-commercially.