Risk factors analysis: screening of extended-spectrum-β-lactamase producing Gram-negative isolates of burn infections from tertiary care hospital Lahore, Pakistan

Burnt patients are highly susceptible to invasion of multidrug resistant strains after the skin damage. The main objective of this study was to estimate the frequency of ESBLs producing Gram-negative in post-burn infections and its correlation with different risk factors (age, gender, socio-economic status, burns etiology etc.) The descriptive case-series study was conducted at Jinnah Hospital Lahore’s Burn and Reconstructive Surgery Centre (JB & RSC) and the Department of Microbiology and Molecular Genetics (MMG), University of the Punjab, Lahore. The clinical specimens of wound swabs, biopsy, and blood were collected from 300 patients during 12th August, 2017 to 12th August 2018. The cephalosporins resistant Gram-negative isolates were further analyzed. The clinical history of burnt patients was recorded which included the age, gender, socio-economic status, residence, occupation, hospital stay, wards, burn degree, total burnt surface area (TBSA%), etiology of burn and other factors. CLSI guidelines 2017 were followed for the antimicrobial susceptibility testing (AST) and ESBLs screening.


Conclusion
It has been concluded that Gram-negative MDR strains are rapidly causing infections in burnt patients and need to be properly alleviated. The high frequency of multidrug resistant Pseudomonas spp., was associated with the burn infections. The patients belonging to young age were significantly found to be infected by MDR strain after burns. Background Burn injuries are caused by chemical, electrical, and thermal exposures to the skin and underlying epidermis, dermis and tissues [1]. Burns are classified by calculating the percentage of total burnt area of the body [2]. The extent of skin damage determines the first, second, and third degree burn [3]. First degree burns affect the surface; second degree burns involve broken skin layers; and third degree burns deeply affect underlying tissues [4]. Burn cases annually account for 0.3 million demises worldwide [5]. The frequency is in USA the 0.2%, a little high in Turkey 2.3%, and in India 3.6% cases per thousand people annually [6]. Post-burn infections caused by multidrug resistant (MDR) bacterial strains account for 75% of mortality rate [7]. Bacterial pathogens responsible for post-burn infections include multidrug resistant (MDR) strains of Gram-negative bacteria (GNB) [8].These include mostly the Acinetobacter baumannii, Escherichia coli, Klebsiella pneumoniae, Proteus spp., Pseudomonas aeruginosa, Staphylococcus aureus, and Streptococcus pyogenes [9]. Sepsis is the ultimate consequence of burnt skin invasion by MDR bacterial strains which begin to proliferate within 48 hours [4]. It has been estimated that 73% of post-burn mortalities happen due to septic shock [10].
Community acquired and nosocomial infectious agents are transmitted to burnt patients by different sources [4]. Operative procedures include catheterization, dialysis, mechanical ventilation and surgery. Health care workers (HCWs) are also found to be associated with transmission of pathogens by fomites or hands [11]. The demographic parameters found to be associated with the multidrug resistance development include age, gender, residence (rural/urban), occupation, literacy, socioeconomic status, and comorbidities. Empirical or home based antibiotics intake, ineffective diagnosis, and inappropriate antibiotic prescription are also the major contributing factors [12,13].
Duration of therapy and stay at hospital also increase the risk of acquiring infection by multidrug resistant (MDR) Gram negative ESBLs producing bacteria [14]. Horizontal gene transfer leads to the accumulation of antimicrobial resistance genes resulting into the emergence of MDRs [15]. Multidrug resistant strains of A. baumannii are involved in adjournment of wound healing process. These strains get invaded into the blood circulatory system and cause transplant damage and septic shock [16].
Survival of the fittest is the likely phenomenon which eliminates the susceptible bacteria from the pool and allows flourishing the MDRs [17]. Department of Centres for Disease Control and Prevention has reported that annually around 23000 patients in the United States die out of two million people got infected by MDRs [18,19]. There is an estimated loss of 1.5 billion euros annually in the Europe which is an outcome of antimicrobial resistance [20]. It has been estimated that these outcomes may lead to worldwide 10 billion deaths annually by 2050. National Institute of Health in Pakistan has implemented the World Health Organization (WHO) recommendations of antimicrobial surveillance 2015 as the National Action Plan to overcome the resistance [21]. Pakistan Institute of Medical Sciences (PIMS), Islamabad harbors more proportions of Gram-negative bacteria in surgical wards and burn unit [22]. According to the best of our knowledge, we are lacking the understanding of risk factors associated with post-burn infections in our society, therefore this study was initiated.There is urgent need to assemble the countrywide statistics on antimicrobial resistance and to initiate the infection control and antibiotic stewardship plans. The main objective of this study is to obtain a clear picture of antimicrobial susceptibility and resistance in bacterial pathogens and to categorize the risk factors associated with the emergence of MDRs isolated from burnt patients.

Study design
The study has been designed according to the prospective investigation principles to

Sampling and data collection
Purposive non-probability method was for the sampling from indoor/outdoor burn wards and intensive care unit (ICU). Accordingly, 300 clinical specimens including wound swabs, biopsy, and blood were collected within duration of one year from 12 th August, 2017 to 12 th August 2018. The demographic information regarding the age, gender, socioeconomic status, residence, occupation, hospital stay, wards, burn degree, total burnt surface area (TBSA%), etiology of burns and other factors have been collected by a designing a data form.

Bacterial isolation and identification
The specimens were immediately moved to the Microbiology Laboratory (JB&RSC) for clinical processing. The aerobic culturing by differential and selective media was performed for the isolation of bacteria. The specimen enrichment was performed in trypticase soy broth (TSB) and after overnight incubation at 37°C the Blood agar and MacConkey's agar were inoculated. Microscopy, colony morphology, and biochemical characterization was performed by API-20E kit system (Bio-Mer ieux, France) [23].

Antibiotics Susceptibility Testing (AST)
Antibiotic susceptibility testing of bacterial isolates was performed on Mueller Hinton agar by Kirby Bauer's disc diffusion method according to Clinical Laboratory Standards Institute (CLSI) guidelines 2017 [9]. Antibiotics tested were including (OXOID) penicillins

Phenotypic screening of extended spectrum β-lactamases (ESBLs)
Double disk synergism test (DDST) as a gold-standard phenotypic method and combination disc test (CDT) as confirmatory test for ESBLs. The tests were performed on Mueller-Hinton's (MH) agar by Kirby Bauer's disk diffusion method as previously reported by Chen et al., [13].

Statistical analysis
The

Frequency distribution of burnt patients and risk factors
A total of 300 clinical specimens were collected from burnt patients during 12 months duration from August 2017 to August 2018. No growth specimens, colonization, cephalosporins sensitive and Gram-positive strains were excluded from the analysis. The remaining specimens included wound swabs 92.00 % (n=138), blood 6.70 % (n=10), and tissue biopsy 1.30 % (n=2) for further analysis. These specimens were taken from 72.00 % of the males (n=108) and 28.00 % of females (n=42). The age of the patients ranged from 4 to 85 years (Mean of age=28 years, ±15.65 SD). Burn injuries were more frequent in the young age groups having the age of ≤ 29 years. Patients belonging to the urban areas 62.00 % (n=93) and the low socio-economic background 80.00% (n=120) were found to be the victims of burn injuries. There was lower proportion of patients 32.70 % (n=49) having the smoking history. females. The indoor patients were found to be more exposed to the MDR strains 84.10 (n=90). Second degree burn patients were frequently associated with ESBLs 86.70 % (n=65) as compared to first and third degree burns (Table-1).

Phenotypic screening of ESBLs
Screening by DDST was effective only to detect 16.00 % (n=24) ESBLs producers. The confirmation was made by CDT and only 14 % (n=21) ESBLs strains were detected.
Pseudomonas spp. was the mainly detected as ESBL producer in 6.67 % (n=10) patients detected by DDST. This number was further reduced to 3.33 % (n=5) when confirmed by CDT (Table-2).

Discussion
Post-burn infections are a major cause of mortality due to severe types of infections. The compromised immune system of burnt patients makes them susceptible to microbial infections by bacterial or fungal pathogens [15,25]. Here, burnt patients were observed with a mean age of 28 ± 15.65 years while Melake et al., reported 15 ± 17.60 years mean age [4]. Males (72.00 %) were found to be more exposed to burns than females (28.00 %).
Saaiq et al., also reported the high frequency of post-burn infections in in males (53.68%) than females (46.31%) [22]. The higher frequency 62.00 % of burnt patients belonged to urban areas, while Anvarinejad et al., reported 74.50 % cases from rural areas [26]. 80.00 % of the patients belonged to low socio-economic status. Chamania et al., also reported a significant association between the low socio-economic status of the patients with MDR infections [27]. The negative correlation of smoking with MDR infections coincides with Melake et al., findings [4]. The reason behind the significant correlation of burn injuries with higher frequency in males is that they are more exposed to the environment that harbors burn sources [28]. People with low-socio-economic status are observed working in industries located in urban areas [29].
Higher frequency of burnt patients was observed with 50.00 % second and 22.00 % third degree burns. Vural et al., also reported higher number of patients with the second degree burns [30]. The reason behind the low frequency of first degree burns is the home-based treatment by commercially available tonics and topical ointments. Second and third degree burns involve broken skin layers and underlying tissues, therefore their treatment is very difficult at home [4]. Burn etiology also varies in different studies. Flames were the most common cause of burns in 52.70 % patients, as was reported by Saaiq et al., where 76.00 % patients were exposed to flames [22]. Anvarinejad et al., reported 93.00 % of burn injuries caused by chemical exposure [26].
The extent of skin damage (TBSA %) was less than 20% in majority of our patients (66.00 %) revealing non-significant association with MDR infections. Findings reported by Fadeyibi et al., differ from our study where a strong correlation was found between the extent of skin damage and MDR infections [31]. The occupation of the patients was also not significantly associated with the MDR infections, as non-occupational incidents were reported in 71.30 % patients. Melake et al., also reported that there was not any significant association between the occupational injuries and MDR infections [4].
Although a large number of patients 71.30 % were admitted as indoor but the hospital stay of 76.00 % patients was less than one week, revealing non-significant correlation between longer hospital and MDR infections. Fadeyibi et al., have reported strong association between MDR infections and longer hospital stay where 50.00 % burnt patients were monitored for more than three weeks. It means that longer hospitalization is significantly associated with the MDRs infections [31]. Here, 26.00% burnt patients were admitted in the ICU for the treatment revealing non-significant association with MDRs.
Leseva et al., reported a positive correlation between the MDR infections and the number of patients admitted in intensive care units (ICUs) [8].
Mono-bacterial Gram-negative cultures were analyzed only and the isolates with more than one bacterial species were excluded from further analysis. Saaiq et al., also reported 93% mono-bacterial infections frequently caused by Gram-negative bacteria [22]. The reason behind analyzing the single species isolates is to lessen the probability of contamination as a single bacterium in-vitro is enough to cause infection [32]. Two decades ago Pruitt et al., reported that post-burn infections are mostly categorized as nosocomial infections as they are frequently caused by hospital-acquired pathogens [33]. Leseva et al., also reported only the nosocomial infections in 10.6% per thousand burnt patients during the year 2011, but the community-acquired infections were not assessed [8]. Our findings differ from these studies as the community-acquired infections (50.70 %) were more frequent than the nosocomial infections (49.30 %) in burnt patients. The reason behind the increase in community-acquired infections is the empirical use of antibiotics and previous exposures of longer antibiotic-therapy [34].
Gram-negative bacterial pathogens are ubiquitous and also known for their abundance in hospitals and surgical settings [35]. Most of the Gram-negative bacteria adhere to the surgical instruments and make sure resistance against desiccation [36]. The association between the surgical procedures and MDR infections was not statistically significant while Vinodkumar et al., reported positive correlation [37]. Comorbidities were already present in only 39.33 % patients, although Melake et al., reported a strong correlation of MDR infections with 28.60 % patients having comorbidities [4].
Pseudomonasaeruginosa has been frequently isolated from the burnt patients in the recent studies [22,38,39]. We also observed thatPseudomonas spp., was the most abundantly isolated bacterial pathogen from 49.33 % burnt patients. The frequency of MDR infections was also higher including 83.33 % of patients. There were 68.70 % isolates with antimicrobial resistance against the imipenem (IMI). Polymyxin E (colistin) was effective against MDR strains with 55.30 % sensitivity. Mohamed has also reported similar results where Pseudomonas spp., was the most prevalent pathogen among all of the clinical isolates (49%). The prevalence of MDR Gram negative isolates was 60% where most of the isolates were resistant to imipenem (65%). The colistin was the most promising antibiotic with the 84% sensitivity [40]. 60% of the isolates were resistant to imipenem (IMI), and ertapenem (ERT) in a study reported from Riyadh, Saudi Arabia [41].
We observed that only 16.00 % of isolates were detected as ESBLs producers by double disk synergy test (DDST) and only 14.00 % were confirmed by combination disk test (CDT).
These findings differ from the study reported by Hassan et al., where 99.00 % of the isolates were detected as ESBLs producers by CDT and 68.70 % by DDST [42]. Therefore it is highly recommended that the ESBLs detection should be accomplished by highly advanced molecular techniques.
The infections in burnt patients emerge due to invasion of pathogens or by colonization of the opportunistic bacterial strains [31]. Decolonization strategies may prove helpful in minimizing the infections by highly pathogenic MDR bacterial strains [43].

Availability of data and materials
The data sets analyzed during the current study are available from the corresponding author.

Competing interests
Authors declare that they have no competing interests.

Funding
No funding was received for this study.    Figure 1 Distribution of burnt patients exposed to different types of burn sources. Distribution of ESBLs positive isolates and comparison of DDST and CDT efficacy.