Outcomes of surgical site infections in orthopedic trauma surgeries in a tertiary care centre in IndiaN Rajkumari1, AK Gupta2, P Mathur1, V Trikha3, V Sharma3, K Farooque3, MC Misra4
1 Department of Laboratory Medicine (Microbiology Division), Jai Prakash Narayan Apex Trauma Centre, All India Institute of Medical Sciences, New Delhi, India
2 Hospital Infection Control Unit, Jai Prakash Narayan Apex Trauma Centre, All India Institute of Medical Sciences, New Delhi, India
3 Department of Orthopedics, Jai Prakash Narayan Apex Trauma Centre, All India Institute of Medical Sciences, New Delhi, India
4 Department of Surgery, Jai Prakash Narayan Apex Trauma Centre, All India Institute of Medical Sciences, New Delhi, India
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0022-3859.138731
Source of Support: None, Conflict of Interest: None
Background: Surgical site infections (SSIs) still cause significant morbidity and mortality despite advances in trauma care. We have studied in this paper the rate of SSIs, their outcomes in patients undergoing interventions for trauma and SSI trends in developing countries. Materials and Methods: A 16-month study (May, 2011- August, 2012) was carried out. Patients undergoing interventions for orthopedic trauma were followed and assessed for SSIs and their outcomes and antimicrobial sensitivity patterns of the micro-organisms isolated were noted and correlated. Results: A total of 40 (4.4%) confirmed cases of SSIs were identified among 852 patients of orthopedic trauma. Based on the new CDC criteria, after ruling out cellulitis, only 24 (2.6%) were found to have SSIs. A total of 12.5% of the SSIs were detected during follow-up. Acinetobacter baumannii was the predominant organism as also Staphylococcus aureus. Outcomes observed included changes in antibiotic regime, revision surgery, readmission to hospital and deaths. Conclusion: SSI is prevalent in orthopaedic trauma patients and an active surveillance program will help in early management and prevention.
Keywords: Antibiotic resistance, orthopedic surgery, surgical site infections, surveillance, trauma
Surgical site infections (SSIs) account for approximately 20% of all hospital-associated infections (HAIs) and represent the second-most common type of HAIs in United States.  From 1986 to 1996, hospitals conducting SSI surveillance in the National Nosocomial Infections Surveillance (NNIS) system reported that there were 15,523 SSIs following 593,344 operations (CDC, unpublished data).  A study showed that SSIs were the most common nosocomial infection among surgical patients, responsible for 38% of all such infections.  These cause enormous morbidity, disabilities, adverse surgical outcomes, increased health care costs and sometimes lead to fatal outcomes. Besides this, the stress and anxiety suffered by the patient and their families cannot be quantified. ,, The majority of SSIs are caused by the patient's own colonizing or endogenous flora. 
The causes and risk factors for SSIs are varied. Trauma patients are a unique cohort of predominantly young males, with no underlying illnesses (usually) and in the prime of their economically productive age. Vehicular accidents in developing countries are on the rise. Most of these admitted trauma victims require surgical interventions at some point putting them at risk for SSIs.
SSIs reported in different types of surgeries vary widely. One possible factor accounting for such wide disparity in SSI rates is the use of different case definitions of SSI. Before 2010, cases of cellulitis could have been considered SSI cases according to the 4 th Centers for Disease Control and Prevention (CDC) criterion: A physician diagnosis of infection. However, SSI now has a new definition from CDC.  The impact of excluding cellulitis was assessed and later excluded from this definition. The definition specifies that infections should develop within 30 days of surgery (if no implant was left in place or up to 1 year with an implant) for it to be called as SSI.  However, this update specifically excludes cellulitis as a superficial incisional SSI, making the definition of SSI more stringent and restrictive.
Complications due to orthopedic SSIs range from superficial infections to deep-seated organ space infections. The most recent National Healthcare Safety Network (NHSN) report, which includes data from 2006 to 2008, showed that post knee replacement infection rates range from 0.68% to 1.60%, depending on patient risk, and hip replacement infection rates range from 0.67% to 2.4%.  The overall rate of surgical site infection after open reduction and internal fixation of tibial plateau fractures during the 7 years of this study was 7.8% (20 of 256).  Orthopedic surgery frequently involves the placement of an implant and hence its related infections.  In view of the serious adverse implications of orthopedic SSIs, prevention and regular surveillance need to be augmented which is especially needed for developing countries like India. Many of the cases of SSIs are lost to follow-up on discharge. Post-discharge follow-up of SSIs is important to ascertain the magnitude of the problem in orthopedic trauma. This study describes the trend of SSIs in orthopedic trauma at a level-1 Trauma Centre of India as also its trend in developing countries
Settings and participants
This was a prospective observational study. All post-operative patients over a 16 month period were monitored by the surgeons, microbiologists and hospital infection control nurses for development of SSI (May 2011- Aug 2012). Also, the type of injury and surgical intervention done were recorded and followed. A total of 2,287 patients admitted during this period for orthopedic trauma were screened and 852 were found to be possible candidates for the study. Institutional review board approval for the study was obtained and written, informed consent was taken from these patients.
A case was defined as any patient who had been admitted for any orthopedic trauma (CDC criteria)  and had undergone manipulation or surgery related to it in this center, who later developed purulent discharge at the operation site or purulent discharge with mocrobiologically positive cultures or signs and symptoms of fever along with a significant rise in total leucocyte counts after the surgery along with culture positivity or any sign of inflammation with gaping of incisional site, failed graft/implant with discharge but absence of negative culture and clinical/surgeon's diagnosis of SSI.
For those patients falling under clinical/surgeon's diagnosis of SSIs, careful consideration was done to rule out pure cellulitis as such patients were excluded from the study and were treated with antibiotics only. Patients were included as per the case definition. Those who had been treated or operated outside the center, but later referred here or who were on external fixators were excluded as also those who developed cellulitis. Co-morbidities such as diabetes mellitus, hypertension, renal or liver diseases were also exclusions.
All these patients were carefully followed with regard to their duration of hospital stay, mode of treatment, surgical interventions done, type of implants, any change in their treatment regime, wound condition, etc. The antimicrobial therapy given pre-operatively and other management procedures were noted.
Follow up and sample collection
Appropriate samples were taken from the wound and microbiological cultures were performed as per the standard protocols, based on the clinical judgement of the treating physicians. Aseptically collected samples from the wound were processed for microbiological culture and sensitivity. Repeat samples were taken after a week. Sampling was also repeated depending on the wound condition and purulent discharge. The change in the wound microbiological flora was noted with its sensitivity pattern. The sample processing for diagnosis of bacterial and fungal pathogens was done by standard methods. , The bacterial isolates were identified by the VITEK 2 ® compact system (BioMιrieux, Durham, NC, USA). The antimicrobial susceptibility testing was done by the disc diffusion method, according to the CLSI guidelines , and the VITEK 2 system. A definite set of antimicrobials for both Gram-negative and Gram-positive bacteria was tested. Initially, all first line antimicrobials were tested and if an isolate was found to be resistant to them, then second line antibiotics including colistin and tigecycline were tested. Multi-drug resistance was defined according to O'Fallon et al.  An empiric regime was started on such patients as trauma patients are considered to be potentially infected and are later shifted to a more specific regime depending on the micro-organism isolated and its sensitivity report. Based on the hospital antibiotic policy, empiric therapy was usually started with parenteral Augmentin (amoxicillin/clavulinic acid) along with aminoglycosides and/or metronidazole and later shifted to more specific antimicrobials depending on the sensitivity report. However, aminoglycosides were used cautiously or avoided in very severe trauma patients due to its potential nephrotoxicity in a hypovolemic patient at risk of renal insufficiency. The treating orthopedic team was intimated regarding microbiological and antimicrobial sensitivity pattern to guide in the further management. All patients were closely monitored and the condition of their wounds and the development of SSIs was discussed and noted.
Changes in treatment protocols like revision surgery or change in antimicrobial therapy or removal of implants were noted in patients who had been diagnosed with SSI during the hospital stay. The patients' status and wound conditions were noted at the time of discharge. All such patients were called regularly for follow-up and their conditions noted. Telephonic contact was maintained with patients living outside the city and who could not follow up. They were made aware of their culture results and advised to attend the clinic for further follow-up and care. Due care was taken during inclusion to prevent duplication of cases where a discharged patient had to be readmitted. During the follow-up of such patients, various parameters were recorded such as condition of the wound, discharge from the wound, failure of implant or gaping of the incision site, etc. This was done to note how many patients had healthy wounds at discharge but later developed SSIs related to their orthopedic interventions. Any change in the antimicrobial therapy was also noted. All data was collated, compared and analysed using descriptive statistics.
A total of 852 orthopedic trauma patients were seen during this time period, of which 596 (70%) were males and remaining 256 (30%) were females (2.1:1). The mean age was 29.5 (±10.5) years. Of the total, 84% were pure orthopedic trauma patients where interventions were performed and the rest (16%) had other surgical or neurosurgical procedures besides orthopedic surgery.
Orthopedic SSI rates
Including definite cases of cellulitis, a total of 40 cases were identified in 914 operated sites among 852 patients. Thus, the rate of orthopedic SSI inclusive of cellulitis was 4.4% per operated site. When cellulitis was excluded from the study, only 22 cases meeting at least one of categories of CDC (1-3) were found. Among them, 14 patients had purulent discharge (CDC category 1), five patients had positive cultures (CDC category 2) and three patients had signs of inflammation and gaping of the incision site. Thus the rate of SSI according to the revised definition of CDC was 2.6%.
Impact of excluding cellulitis on defined SSI rates
Out of the 40 SSIs, 18 (45 %) were classified as category 4 (surgeon's diagnosis) based on the old CDC criteria. Of these 18 patients, 16 of them had cellulitis without purulent drainage or opening of the wound and resolved with antibiotic treatment, although hospital admission with intravenous antibiotics was required in six patients. The remaining two patients had cellulitis with opening of the wound, and cultures obtained were negative but had been preceded by antibiotic treatment and hence were categorized as physician diagnosis both by the old and new CDC criteria. A two fold reduction in orthopedic SSIs from 4.4% to 2.6% was observed if the surgeon's diagnosis of definite cellulitis was calculated [Table 1].
Duration of hospitalization
On an average, the median duration of hospital stay among those patients who did not develop SSI was 15 days (5-25 days). [Table 2] shows the duration of hospital stay and interventions done on the 24 patients who later developed SSIs. The shortest stay of 20 days was seen among those with pure orthopedic upper limb trauma. It was seen that 14 (58%) developed purulent discharge, 6 (25%) had redness and erythrema of the site, 4 (17%) of them had gaping of the incision site and had to undergo revision surgery. A total of 21 (87.5%) complained of severe pain at the operated site and needed treatment review.
Of the 24 SSIs patients, 16 patients (%) developed SSI related to orthopedic intervention within 30 days of the surgery; the remaining eight developed SSI after 30 days but within 1 year of surgery [Figure 1]. The median time to SSI diagnosis was 14 days (range of 7- 64 days) with 12 (50%) patients developing SSI within the first 14 days of surgery. However, there was no significant difference in time to infection in those patients presenting with pure cellulitis versus the other categories of SSI [Figure 1]. It was seen that the rate of infections was more during July- September, which usually have the highest numbers of admissions also [Figure 2].
Nature of isolates
A total of 46 isolates were obtained. Multidrug resistance (MDR) among Gram-negative bacteria was defined as resistance to three or more antimicrobials or antimicrobial groups including extended-spectrum penicillins (ampicillin/sulbactam or piperacillin/tazobactam), cephalosporins (cefazolin or ceftriaxone), gentamicin, ciprofloxacin, and trimethoprim-sulfamethoxazole (TMP/SMX).  [Table 3] shows the details of the microbiological cultures. So, it was observed that majority of the organisms were MDR except Enterobacter spp. and Streptococcus pyogenes.
Of the 40 SSI patients (inclusive of cellulitis), all were treated with antibiotics; 22 (55%) of whom were treated with oral antibiotics only, 16 (40%) with both oral and intravenous antibiotics and two (5%) with intravenous antibiotics only. Out of the 40 patients, 32 (80%) underwent debridement, 12 (30%) had to undergo drainage in the operating room or in the OPD and 15 (37.5%) patients had to undergo revision surgery or removal of the implant. A total of 5 (12.5%) patients were diagnosed after they came back with discharge, gaping wounds or failure of implants. Of the total 40 patients, two died during treatment due to complications of SSI (both male). One of them was operated for multiple fractures of both the upper limb and lower limb and had an associated pelvic fracture; the second patient had undergone hip arthoplasty and later developed SSI [Table 4].
SSIs in orthopedic surgery are not unknown. Our study found male predominance (3:1) in SSI. Our finding was in concordance with another study which also showed male predominance.  However, this may be due to higher admission rates of males to trauma care facilities. In our study, the total SSI rate was 2.6% based on the new definition of SSI. This finding was almost at par with SSIs seen in orthopedic surgeries in developed countries ,, but lower than the rates seen in some developing countries. , This may reflect the standard of care at our hospital, and the combined effect of availability of ultra-clean ventilated operation rooms and the institution of advanced trauma life support (ATLS) protocols at our center. Not surprisingly, in a study from Serbia, the incidence rate of SSI was 22.7% as they had included pure cellulitis patients.  The authors in that study had also stratified the patients according to wound condition on admission in which more than 70% of the wounds were either contaminated or dirty/infected wounds.  In our study, 80% of the wounds were either contaminated or dirty.
Most of the SSI falls under superficial incisional SSI which includes the pure cellulitis cases (30%). About 32.5% fall under deep incisional SSI and remaining in organ/space SSI.
As a consequence of an increasing trend of shorter hospital stay, majority of SSIs occur after discharge from the hospital.  It was also seen in our study that 12.5% of the SSI diagnosed were picked up during follow-up. This also shows that a good SSI prevention team is required to actively keep such patients under surveillance and diagnose and treat all new cases of SSI.
Gram-negative bacteria like Acinetobacter baumannii, Klebsiella pneumonia and Pseudomonas aeruginosa re found to be the predominant organisms responsible. Of Gram-positive bacteria, Staphylococcus aureus was maximally responsible for causing SSIs in our study in contrast to other studies , Although nasal carriage of S. aureus can be effectively eradicated by mupirocin, our study has shown that it may not significantly reduce the overall rate of SSIs,  as Gram negatives also form a major chunk of organisms responsible.
Our study shows that infections can prolong the hospital stay and increase the morbidity and mortality. ,, Also, we found that the rate of infections was higher during July to September, which usually have the highest numbers of admissions. The higher rates of SSIs during these months may be due to the fact that more surgeries are performed under emergency conditions to meet the load and the generally humid and hot conditions prevailing at that time.
A dedicated infection control unit which can aggressively monitor such conditions is the need of the hour. Proper identification and risk assessment of orthopedic patients needs to be done along with pre-operative management of the risks factors involved. A major limitation of this study is that we did not study the variables affecting the development of SSI.
The authors would like to thank the All India Institute of Medical Sciences, India for funding this study. The funding source did not have any role in the study design; in the collection, analysis and interpretation of data; in the writing of the manuscript; or in the decision to submit the manuscript for publication. Also, the authors would like to thank Mrs. Kumkum Sharma and Mrs. Jacinta Gunjiyal for their contribution towards patient follow ups and their technical help.
[Figure 1], [Figure 2]
[Table 1], [Table 2], [Table 3], [Table 4]