Skin, the largest organ in the human body, plays crucial roles in sustaining life, including water regulation, thermoregulation and most importantly, as the first line of defense against external agents such as micro-organisms.1 Exposure of the subcutaneous tissue following loss of skin integrity provides an environment that is moist, warm, and nutritive, all of which are conducive for microbial colonization and proliferation.2
Wound injuries are the most common and serious types of trauma and represent a major public health concern. There are many potential factors (age, sex, diabetes, stress, nutrition, oxygenation) involved in the complex wound healing process that can delay healing.3 Wound infections are mainly caused due to proliferation of microorganisms that enter the wound site once the skin is injured. Because of localized inflammation, there is pus formation which consists of white blood cells, damaged cells and dead tissue.4 Factors such as age, malnutrition, obesity, endocrine or metabolic disorders, microbial load and host defense mechanisms influence the development of wound infection.5 In 2015, the overall incidence of wound sepsis in India was 10-33%.6
Apart from trauma, other causes for wound infections include surgical site infections (SSI) and diabetic ulcers. Incidence of SSI is a major concern in hospitals, as it increases the length of hospital stay, treatment cost and in a few cases, causes significant morbidity and mortality.7
Wound infections are typically polymicrobial, and harbor bacteria, fungi, parasites and viruses that can be aerobic/anaerobic/facultative in nature.2 The most common causative agents include Staphylococcus aureus, accounting for 20-40% of the infection, methicillin-resistant S. aureus, followed by Pseudomonas aeruginosa (5-15%), Escherichia coli, Enterococcus sp., Proteus sp. and Klebsiella sp.8, 9
Appropriate selection of antibiotics depends on the causative agent, the pathophysiology, along with the pharmacokinetics and pharmacodynamics of the drug. The emerging issue of increased antibiotic resistance has escalated the level of difficulty with respect to optimum treatment protocols, especially concerning Gram negative organisms.10, 11
Antibiotic resistance being a concern in current times, the protocol of testing of micro-organisms in terms of culturing and sensitivity at the initial stage is vital to provide the appropriate treatment and prevent further complications.12 Therefore, this study was conducted to study the type of wound infections, the aerobic bacteria associated, along with their antibiograms at our tertiary care centre.
Materials and Methods
This prospective study was conducted at a tertiary care center in Bengaluru, Karnataka, India, from January 2014 to December 2015. Approval from the Institutional Ethics Committee (STD-1/EC/13-14) and written informed consent from participants were acquired before the commencement of the study.
The study included pus and tissue samples from 160 patients from outpatient department (OPD)/wards, both surgical and non-surgical, who were suspected to have wound infection (clinical and microbiological). Patients who were very/terminally ill, had bed sores and those who were on antibiotic therapy for more than two weeks at the time of study in case of non-surgical wounds were excluded from the study.
The sample size for the study was calculated based on the formula:
Where, p is the proportion or prevalence, E is the relative error, Z is the value corresponding to level of confidence required.
From the study conducted by Setty et al., the prevalence of SSI patients who underwent various surgeries in the General Surgery department was 21.66%.13 Considering the relative error as 25% of prevalence and 90% confidence level, the sample size considered was 160 for this study.
Demographic data (age and gender), department, date of admission, comorbidities, date of SSI event, duration, antibiotic therapy of every patient was recorded. Patients were monitored from the time of inclusion in the study to the date of discharge from hospital.
Pus or tissue samples collected from patients were subjected to microbiological processing in the laboratory.
Collection of Pus from swab
Open wounds which had superficial debris was removed by thorough irrigation and cleansing with sterile saline. The swab was gently rolled over the surface of the wound approximately 5 times focusing on area where there was evidence of pus or inflamed tissue. For dry wounds, 2 sterile cotton swabs were used after moistening with sterile saline. Swabs were carried in aerobic transporter tube to the laboratory.
Collection of aspirates
The skin of affected site was thoroughly cleaned with 70% alcohol, followed with betadine solution and again by 70% alcohol. The surface was allowed to dry and then the pus was aspirated from the deepest portion of the lesion by using a sterile syringe and needle. The aspirate was collected into a sterile wide mouth container and transported immediately to the lab.
Collection of tissue
The surface of the wound was cleaned through irrigation and cleansing with sterile saline. The tissue samples were collected from areas within and around the infection using sterile scalpels/ blades and were sent to the laboratory in a sterile wide mouth container immediately.
All samples were transported to the laboratory within 30 minutes of collection and processed within an hour of collection. All the samples were processed for Gram stain and Ziehl- Neelsen [ZN] stain (only if the sample was an aspirate) to look for the presence of Gram positive/Gram negative and acid-fast bacteria respectively.14 Pus cells were graded as occasional (<1/oil immersion field [OIF]), few [1-5/OIF], moderate [5-10/OIF] and numerous [>10/OIF]. Organisms were graded in a similar way.
Simultaneously, qualitative culturing was performed on 5% sheep blood agar (SBA) and MacConkey’s agar (MA) using standard techniques.14 Small quantity of the sample was inoculated, using a sterile 4 mm nichrome wire loop (Hi-media), in the thioglycolate broth as a back-up. Samples were incubated at 37°C in an aerobic atmosphere for 24 h. Negative culture plates were discarded and the thioglycolate broth was retained for 72 h, to check for turbidity. In case of turbidity, samples were sub-cultured on SBA and MA plates. The isolate/isolates obtained were identified by Gram stain and standard biochemical reactions.
The isolated colonies, depending on their Gram reaction were then subjected to biochemical tests (catalase, oxidase, coagulase, nitrate reduction, indole, methyl red, Voges Proskauer, citrate utilization, urease, mannitol, motility and triple sugar iron agar) for identification.15
Antibiotic susceptibility testing
The susceptibility test was performed on Mueller-Hinton agar (MHA) by Kirby Bauer disk diffusion method, according to Clinical Laboratory Standards Institute (CLSI) M2-A9 2013 guidelines.16 The antibiotics (HiMedia Laboratories Pvt Ltd) used for the test were ampicillin (10 μg), amikacin (30 μg), aztreonam (30 μg), amoxicillin-clavulanic acid (20/10 μg), cefotaxime (10 μg), ceftriaxone (10 μg), ceftazidime (30 μg), cefoxitin (30 μg), cefipime (30 μg), cefixime (5μg), ciprofloxacin (1 μg), clindamycin (2 μg), doxycycline (30 μg), erythromycin (5 μg), gentamicin (10 μg), imipenem (10 μg), meropenem (10 μg), netilmicin (30 μg), ofloxacin (5 μg), oxacillin (5 μg), penicillin g (10 units), piperacillin (100 μg), piperacillin-tazobactam (100/10μg), tigecycline (15 μg), tobramycin (10 μg), trimethoprim/sulphamethazole (1.2 μg/23.8 μg), high level gentamicin (120 μg), vancomycin (30 μg) and teicoplanin (30 μg). The diameter of the zone of inhibition was measured (measuring scale) and interpreted according to the CLSI M2-A9 2013 guidelines.
Extended spectrum ß-lactamase (ESBL) detection method
The Gram negative bacilli (Enterobacteriaceae family) showing reduced zone of inhibition against ceftazidime, cefotaxime and ceftriaxone, when compared with CLSI M2-A9 2013 guidelines were suspected as ESBL producing organism and were confirmed by combination disc method.
Combination disk test
The antibiotics used for this test were cefotaxime (30 µg), cefotaxime/clavulanic acid (30/10 µg), ceftazidime (30 µg) and ceftazidime/clavulanic acid (30/10 µg). Four to five colonies of similar morphology were inoculated to 5 ml peptone water and incubated at 37°C for 4-6 h until turbidity matched to that of McFarland 0.5 turbidity standard (1.5 x 108 CFU/ml). Lawn culture method was adapted on MHA plates and cephalosporin along with cephalosporin with clavulanic acid disks were placed 50 mm apart from each other.17
The zone of inhibition was measured around the disk. An increase of 5 mm in zone of inhibition in a disk containing clavulanic acid compared to the drug alone was considered as ESBL producer.
Modified Hodge Test for Carbapenemase detection
Two to three identical colonies of Escherichia coli (ATCC 25922) were inoculated into saline and incubated at 37° C for 4 to 6 h, until the optical density matched to that of 0.5 McFarland turbidity standard. This suspension was then diluted 1:10, by adding 0.5 ml of 0.5 McFarland to 4.5 ml of the test suspension. A lawn culture of 1:10 diluted Escherichia coli on to the MHA plates with a sterile cotton swab was produced. The plate was left undisturbed for 5 mins at room temperature. An imipenem (10 μg) disc was placed at the centre and the test organism was streaked in a straight line from the edge of the disc to the edge of the plate. The plate was incubated overnight at 35°C in ambient air. The presence of distorted zone of inhibition or clover leaf type of indentation at the intersection of the test organism and E. coli, within the zone of inhibition of the imipenem susceptibility disc was interpreted as positive result.18
AmpC β-lactamase enzyme detection (AmpC disk test)
Isolates showing reduced susceptibility to ceftazidime, cefotaxime and cefoxitin were considered as “screen positive” and selected for detection of AmpC β-lactamases by AmpC disk test.19
AmpC Disk Test
A lawn culture of Escherichia coli (ATCC 25922) was produced on MHA plate. Sterile disks (6 mm) were moistened with sterile saline (20 μl) and inoculated with a minimum of 3-4 colonies of test organism. The inoculated disks were then placed close to a cefoxitin disk on the inoculated plate. The plates were incubated overnight at 35°C. Appearance of a flattening or indentation in the cefoxitin inhibition zone in the vicinity of the test disk was considered positive. A negative test had an undistorted zone.
All patients with positive cultures from pus samples were treated as per the antibiogram reports obtained. After the modification of antibiotic drug regimen, these patients were followed up to their discharge and the outcome was noted.
Data were analyzed using statistical software R version 3.6.3 and MS Excel. Categorical variables are represented by number (%). They were compared using chi-square test/Cochran Armitage test. Continuous variables are represented by mean ± standard deviation form. Level of significance was set at p≤0.05.
The mean age of participants in this study was 46.47±18.09 years, including 108 (67.5%) males and 52 (32.5%) females. Majority of participants belonged to the age group of 41-60 years (n=73, 45.63%). Many patients were admitted to the general surgery department (n=65, 40.63%) (Table 1).
Majority of the patients exhibited SSI (n=146, 91.25%), early infection (n=133, 91.1%), with most being categorized as a clean wound (n=98, 67.12%). The majority were subjected to surgical prophylaxis (n=136, 93.15%) (Table 2).
With respect to surgical prophylaxis for SSI patients (n=146), 11 (6.88%) received cefazolin-metrogyl, 54 (33.75%) received ceftriaxone, 14 (8.75%) received ceftrioxone-metrogyl (Table 3).
With respect to Gram staining, of the 128 (80%) samples that were positive on culture, 111 (86.71%) showed the presence of organisms on direct smear while 17 (13.28%) samples were smear negative. Gram staining of 14 (10.93%) samples showed the presence of organisms on direct smear but culture yielded no growth. None of the samples collected by aspiration (n=13) showed the presence of acid-fast bacilli. Of the 128 samples that tested culture positive, 107 (83.59%) samples had growth of a single organism, 20 (15.62%) had growth of two organisms and 1 (0.01%) sample had growth of three organisms.
Amongst the positive culture samples (n=128), Staphylococcus aureus was identified in majority of the samples (n=40, 31.25%) followed by Escherichia coli in 30 (23.43%) and Pseudomonas aeruginosa in 16 (12.5%) samples (Table 4).
Majority of the cases in the study with SSI reported the growth of Staphylococcus aureus and minimal growth of micro-organisms were observed in cases with diabetic ulcer and traumatic type of wounds (Figure 1).
Antibiotic susceptibility patterns
Amongst the Gram positive isolates (excluding Enterococcus) (n=54), 40 (74.07%) were Staphylococcus aureus and 13 (24.07%) were Coagulase negative Staphylococci (CoNS) and 1 (1.85%) isolate was Streptococcus pyogenes. Out of the 40 S. aureus isolates, 33 (82.5%) were methicillin sensitive and 7 (17.5%) were methicillin resistant. Overall, 47 (88.68%) Gram positive isolates were resistant to penicillin, 41 (77.35%) to augmentin and 35 (66.03%) to ciprofloxacin, whereas, 52 (98.11%) isolates were sensitive to vancomycin and 51 (96.22%) to teicoplanin and 49 (92.45%) to linezolid
Amongst the Gram negative isolates (excluding nonfermentors) (n=58), 55 (94.82%) isolates were identified as ESBL producers, 50 (86.2%) as AmpC producers and 12 (20.69%) isolates as Carbapenemase producers. Overall, 49 (84.48%) Gram negative isolates were sensitive to imepenem, 44 (75.86%) to meropenem and 43 (74.13%) to amikacin.
Overall, 47 Gram positive and 60 Gram negative isolates were multidrug resistant in this study.
Overall, 47 Gram positive and 60 Gram negative isolates were multidrug resistant in this study.
Amongst the β-lactamase producing organisms, majority of Escherichia coli isolates were ESBL (n=28) and AmpC (n=25) producers. A greater number of Pseudomonas aeruginosa isolates were AmpC (n=15) producers (Figure 2).
Amongst the nonfermentors (n=23) i.e. 16 (69.56%) Pseudomonas aeruginosa and 7 (30.43%) Acinetobacter baumanii isolates, 17 (73.91%) were resistant to both aztreonam and ciprofloxacin, 14 (60.86%) to ceftazidime, 13 (56.52%) to both piperacillin and tobramycin along with 12 (52.17%) isolates to amikacin. All isolates of Acinetobacter baumanii were sensitive to tigecycline.
Among the 13 isolates belonging to genus Enterococcus, 9 (69.23%) were resistant to high level gentamicin (HLG), 10 (76.92%) were resistant to ciprofloxacin and 1 (7.69%) to vancomycin and all the were sensitive to linezolid and teicoplanin.
It was observed that the incidence of clinically suspected SSI was 91.25% (n=146), whereas the incidence of culture confirmed SSI was 72.5% (n=116) in this study. The study variables such as age, gender, type of wound along with co-morbidities such as diabetes mellitus and hypertension were insignificantly associated with the incidence of SSI (p>0.05) (Table 5).
Overall, 128 (80%) samples showed growth of pathogenic organisms and were treated according to the antibiotic susceptibility report. Patients were followed up regularly, where 121 (75.62%) patients recovered following the wound infection treatment and 7 (4.37%) patients expired.
The management of infection (mono or polymicrobial) is an important aspect of wound care. Understanding the microbial nature is an imperative part of an efficient treatment strategy. Antibiotic agents are one of the wonder discoveries of the 20th century.20 Therefore, in this study, the bacteriological profile and the antibiotic sensitivity patterns of patients having wound infections were analysed.
Most samples were from the general surgery and orthopaedics department, as many surgeries are undertaken by these departments. The nature of the wound for majority of the patients in this study was SSI. The SSIs are now considered the most common nosocomial infections, representing a major clinical problem with respect to morbidity, mortality, length of hospital stay and the overall costs. They are also considered a global priority because 20-35% of these infections are caused by antibiotic resistant strains.21 The proportion of patients with early infection i.e. presenting within 30 days of surgery with SSI was more when compared to intermediate and late infections in this study.
The rate of infection in patients developing SSIs was observed to be higher (67.12%) in clean wounds unlike in other studies where the incidence of infection rates due to clean wounds was very less (3-6%).22, 23 This finding probably suggests that the factor mainly contributing to the development of infection was in the post-operative period. Surgical prophylaxis was administered to 93.15% patients and the rest who had not received surgical prophylaxis probably underwent emergency surgery as the relevant information was not recorded.
The most common antibiotic given as surgical prophylaxis was ceftriaxone followed by combination of ceftriaxone-metronidazole and cefazolin-metronidazole. As observed in the study, there is no uniform guidelines in the hospital for the usage of surgical prophylaxis antibiotics since there were more than twenty different combinations which were given. The exception was in the Obstetrics and Gynaecology (OBGYN) department were the surgical prophylaxis practice was uniform and all patients received cefazolin-metronidazole as prophylaxis. Though not statistically significant, this could have been the probable reason for lesser number of wound infections in OBGYN department. The use of cephalosporins along with metronidazole is preferred for surgical prophylaxis in hospitals globally because of its effectiveness in alleviating the infection rates.24
In this study, majority of patients had Diabetes mellitus (DM), hypertension (HTN) or DM/HTN as a comorbidity. High blood sugar can increase infection rate and impair wound healing. Poorly controlled diabetes adversely affects the ability of leukocytes to destroy invading bacteria and prevent the harmful proliferation of usually benign bacteria present in the healthy body. Hypertension is a worldwide epidemic and is also common among patients with diabetes.25 However, the association between comorbidities such as DM and HTN with SSI in this study was statistically insignificant (p>0.05).
Gram staining yielded 80% positive cultures. As most samples collected were swabs, there could have been the possibility of the organism load being lesser, and hence in case of smear, results were negative for 13.28% of the positive cultures. In a few cases, the culture yielded no growth despite being positive for the presence of organisms on direct smear. There is a chance of them being anaerobic organisms, the isolation of which was not included in the study.
In this study, majority of the isolates were Staphylococcus aureus, followed by Escherichia coli and Pseudomonas aeruginosa. The least common organism to be isolated was Streptococcus pyogenes and non-fermenting Gram negative bacilli (NFGNB). The organisms isolated depend on the site which is opened – either skin incision or opening of the gastrointestinal tract. As the number of superficial incisional SSIs and orthopaedic infections were greater in this study as compared to other type of wound infections, Staphylococcus aureus can be substantiated for being isolated as the most common pathogen as most of the times the source of infection is patients own endogenous flora.26 Sawdekar et al. also reported Staphylococcus aureus being the predominant organism isolated from the infected wounds.27
The antibiotic susceptibility patterns of all isolates were studied. The antibiotic panel for Gram positive cocci, Gram negative bacilli, Enterococcus, Pseudomonas/Acinetobacter sp. were selected according to CLSI M2-A9 2013 guidelines. The difference in number of isolates obtained in this study with respect to the organisms being Gram positive or negative in nature could be due to the variations in common nosocomial pathogens inhabiting the hospital setup.
It was observed that of all the S. aureus isolates (n=40), 82.5% were methicillin sensitive and 10% were methicillin resistant. A study conducted by Sudhaharan et al. reported 51.9% of S. aureus isolates being methicillin sensitive.28 However, the Gram positive cocci were resistant to commonly used antibiotics like amoxicillin with clavulanic acid, ciprofloxacin and pencillin. This difference could be due to better hospital infection control practices followed in our hospital.
Most of the Gram negative bacilli were multidrug resistant with organisms being ESBL, AmpC and carbapenamase producers. This finding alarms the clinicians to be aware of emerging multidrug resistance among Gram negative isolates as the only options for treatment include amikacin and carbapenems. The usage of amikacin in renally compromised patients is questionable and usage of carbapenems for the treatment of wound infections has to be looked upon due to cost factor as most of the patients attending our hospital cannot afford them.29 However, carbapenems are known to possess broad spectrum antibacterial activity, having a unique structure that is defined by a carbapenem coupled to a β-lactam ring that confers protection against most β lactamases such as metallo-β-lactamase (MBL) as well as ESBL and AmpC producers.30 Consequently, carbapenems are considered one of the most reliable drugs for treating bacterial infections and the emergence and spread of resistance to these antibiotics constitute a major public health concern.31
Majority of P. aeruginosa isolates showed reduced susceptibility to antibiotics like ciprofloxacin, ofloxacin, and meropenem. Decreased susceptibility of Pseudomonal strains to commonly used antibiotics is a serious threat to the community and may be attributed to inappropriate use of antipseudomonal drugs by the clinicians. Relaxation of general hygenic measures, mass production of low quality antiseptic solutions and medicinal solutions, difficulties in the proper definition of responsibilities among the hospital staff all contribute to development of resistance in Pseudomonas sp.32
Majority of the Enterococcus isolates were resistant to high level gentamycin and ciprofloxacin and only one isolate was vancomycin resistant, while all isolates were sensitive to linezolid and teicoplanin. This finding restricted the use of flouroquinlones, aminoglycosides and the combination of ampicillin and aminoglycoside synergy as a drug of choice for the treatment of wound infections.33 Isolation of vancomycin resistant Enterococci is a warning sign for both clinicians and microbiologists for strict surveillance and more stringent hospital infection control measures. Vancomycin resistance was checked by Kirby Bauer disk diffusion method for Enterococcus, approved by CLSI M2-A9 2013 guidelines.
The overall rate of bacterial isolation of wound samples in this study was 80%, comparable to observations reported by Shimekaw et al. (72.6%) and Wadekar et al. (85.5%). 3, 34 On the contrary, studies conducted by Abraham et al. (41%) and Biadglegne et al. (53%) reported lower isolation rates.35, 36 The difference in these rates could be attributed to the facilities of the hospital management and implementation of infection prevention and control program. The total number of patients in the hospital along with the environmental conditions could also be contributing factors. All patients were followed up from the date of reporting of wound infection to their discharge/death. Patients who recovered had a longer duration of hospital stay, until their wounds healed, and antibiotic course was completed. Seven patients in our study succumbed to death, where 5 of them expired due to underlying comorbidities such as diabetes and hypertension, along with increased age as an additional risk factor.
Limitations of this study include the small sample size along with the fact that microbiological examination for anaerobes and fungi were not included as a part of this study.
The most common pathogen associated with wound infection in our hospital was methicillin sensitive Staphylococcus aureus with SSI being the most common type of wound infection. Majority of the SSIs occurred in early postoperative phase and in patients having clean wound at the time of surgery. The most common comorbidity associated with wound infections was diabetes and hypertension. Multidrug resistance pattern exhibited by organisms cause difficulty with respect to selecting appropriate antibiotics for treatment. High number of ESBLs, AmpC and Carbapenamse producing organisms emphasize implementing highly stringent infection control measures. Regular surveillance with Root Cause Analysis (RCA) of all wound infections must be done to check the rate of the same. Strict adherence to the hospital’s antibiotic policy by the prescribing clinicians must be worked upon. The formulation of infection control measures and appropriate use of antibiotics must be considered compulsory to alleviate wound infection rates.