Bacteriological Profile of Pathogens in Burns Unit of a Tertiary Care Center: A Retrospective Observational Study

Objective. This retrospective observational study analyzes the bacteriological profile of pathogens causing burn wound infections in a tertiary care center. Materials and Methods. This study was conducted at Father Muller Medical College Hospital, Karnataka, India, from January 2014 through December 2016. The specimens (ie, pus or a wound swab) were collected from patients with suspected of infection and processed as per standard microbiological techniques. The antibiotic sensitivity testing was performed by the Kirby Bauer’s disk diffusion test on Mueller-Hinton agar as per Clinical and Laboratory Standards Institute guidelines. Results. During the study period, a total of 124 eligible patient samples were collected; 22 samples were excluded as there was no significant growth/colonization. Among the 102 patients included in the study, 56 (54.9%) were females and the majority (33, 32.35%) of the patients were between 18 to 30 years. Acinetobacter species and Pseudomonas aeruginosa (26.56% each) were the most common pathogen among gram-negative bacteria and Staphylococcus aureus (36, 11.25%) was the most common gram-positive bacteria. Methicillin resistance was 30.5% among the Staphylococcus aureus isolates. Most of Acinetobacter species isolates were resistant to piperacillin tazobactum (84.71%), meropenem (80%), and amikacin (87.06%). Other gram-negative bacteria were also emerging with multidrug resistance. Conclusions. The current study revealed the non-fermenting Gram-negative bacteria as the leading cause of burn wound infection and are highly resistant to available high-level antibacterial agents.

About 6 million to 7 million burn cases are reported in India annually.¹ Around 10% of these need hospital care, and there is a 50% mortality rate in these admitted patients. Infection is one of the main reasons for mortality in patients with severe burns.² A patient with a burn loses the protective skin barrier and both cellular and humoral immunity is lowered. Release of inflammatory mediators leads to systemic inflammatory response syndrome contributing to death of these patients.³ The infection in admitted cases is often caused by Gram negative bacteria (GNB) like Pseudomonas aeruginosa (71%), and Acinetobacter species (29%) and Gram-positive bacteria like Staphylococcus aureus.⁴ These pathogens are usually hospital acquired and multidrug resistant (MDR), further narrowing the antibacterial agents against them, leading to the death of the patients within a few weeks or a month of their admission.⁵ Mortality of patients with burns infected with MDR pathogens is triple that of the patients with burns without infection.⁶ In order to deliver effective, empirically proven antibacterial therapy and reduce the mortality and morbidity of the patients with burns and infection, the bacteriological profile with the antibiotic resistance pattern of the cases admitted in the burns unit is essential. This study aimed to determine the bacteriological profile and antibiotic resistance patterns from a burn unit.

This retrospective observational analytic study was conducted in the Department of Microbiology and Burns at Father Muller Medical College Hospital, Karnataka, India, after obtaining ethical clearance from the Institutional Ethics Committee. All patients with greater than or equal to 20% of their total body surface area affected by burns, who were admitted in burn unit/intensive care unit (BICU) from January 2014 through December 2016, were eligible for study inclusion.

Specimens such as pus or wound swabs were collected from patients with signs of infection (pain, redness, swelling, and oozing/purulent discharge from the wound). Wounds were cleaned using sterile normal saline to remove residual topical antibacterial agents and prevent bacterial colonization of the wound. Sterile gauze was used to blot excess normal saline from the wound. A sterile swab was moistened with sterile normal saline and the cotton tip was rolled over the wound by rotating the swab stick. The swab was taken from the edge of the wound, which appeared deep with pus discharge, and where the pathogens were invading the healthy tissue. The swab was immediately transported to the microbiology laboratory in a sterile, leak-proof container.

All specimens were inoculated in 5% sheep blood agar and MacConkey agar and were incubated aerobically overnight at 37°C. The Gram staining of the specimens was designed to look for significance of the bacteria. The colonizer was ruled out if organisms had epithelial cells and no pus cells. Any significant bacterial growth was identified by standard microbiological techniques.

The antibiotic sensitivity testing was performed by the Kirby-Bauer disk diffusion method on Mueller-Hinton agar. The isolates were tested against ampicillin (10 µg), amoxyclav (20/10 µg), cephazolin (30 µg), cefuroxime (30 µg), ceftriaxone/ceftazidime (30 µg), cotrimoxazole (1.25/23.75 µg), gentamicin (10 µg), amikacin (30 µg), levofloxacin (5 µg) erythromycin (15 µg), clindamycin (2 µg), piperacillin/tazobactum (100/10 µg), and meropenem (10 µg) according to Clinical and Laboratory Standards Institute guidelines.⁷Escherichia coli strain ATCC 25922, Staphylococcus aureus ATCC 25923, and Pseudomonas aeruginosa ATCC 27853 were used as control strains. The patients’ demographic details, the isolated pathogen, and its antibiogram were collected, entered in Microsoft Office Excel 2007 software and analyzed by percentage.

A total of 124 patients, who were admitted for more than 48 hours in the BICU and were suspected of having an infection were eligible for study inclusion (Tables 1-5). Among the 124 pus and wound samples received in the lab, 22 samples were excluded as there was no significant growth or bacterial colonization. Among the 102 patients included in the study, 56 (54.9%) were female (Table 1). The majority (33, 32.35%) of the patients were between 18 and 30 years of age, followed by less than 18 years (29, 28.43%) (Table 1). When the details of the burns were investigated, only 92 patients’ details were available. Among 92 patients, 48.91% sustained second degree burns and 86 (93.48%) had thermal burns. Accidental burns were reported in 94.57% of the patients. In about 50% of the patients, 30% to 60% of total body surface area was affected. The trunk (62, 67.39%) was the most commonly affected body area followed by the upper limb (58, 63.04%). More than half of the patients (58.7%) were admitted for 1 week to 1 month. Forty-eight patients (52.17%) were treated successfully and discharged, whereas 35.87% succumbed to complications from their burn injuries (Table 2).

From a microbiological perspective, a total of 320 pathogens were isolated from the 102 patients included in the study. Among the pathogens isolated, 256 (80%) accounted for gram-negative and 64 (20%) gram-positive (Table 3). Among the gram-negative isolates, most common was Acinetobacter species and P aeruginosa (26.56% each), followed by Klebsiella species (17.19%). Staphylococcus aureus (36, 11.25%) was the most frequent isolate among the Gram positive, followed by Enterococcus species (15, 4.7%). The methicillin-resistant S aureus (MRSA) infection was observed in 11 patients, 3.44 % of total patients. However, the MRSA isolation rate was as high as 30.5% (11/36 Staphylococcus aureus).

There was high level of sensitivity for third generation cephalosporin, aminoglycosides, and clindamycin among the gram-positive pathogens. Among MRSA isolates, high level of resistance was observed for clindamycin and levofloxacin. Among Enterococcus isolates, resistance for ampicillin and high-level gentamicin was noted to levels of 53.33% and 66.67%, respectively (Table 4). Among the S aureus isolated, no resistance was observed for high end antibacterial like vancomycin, linezolid, and teicoplanin. The most common gram-negative pathogens, Acinetobacter species and P aeruginosa, were highly resistant to β lactam antimicrobials like ampicillin, third generation cephalosporins, aminoglycosides, and fluoroquinolones (Table 5). The high end anti-microbial agents like β lactam-β lactamase inhibitors and carbapenems did not appear to be promising in the BICU patients infected with gram-negative pathogens. See Figure 1 and Figure 2.

Burns wound infections (BWI) occur due the loss of the skin, which makes the host more susceptible to infection by the normal flora of the patient. Patients with burn wounds need prolonged lengths of stay in ICU, which is the predisposing factor for acquiring infection during ICU care. Patients with infection require additional care with appropriate antibacterial medications. The bacteriological profile of a burn wound infection is a prerequisite for an effective antimicrobial stewardship program. In this study, the authors identified the profile of the pathogens causing infection in patients with burn wounds and observed the emergence of MDR pathogens in these patients such as MRSA and carbapenem-resistant non-fermenters (ie, Acinetobacter species and P aeruginosa).

The researchers observed that the incidence of burns was higher in females and subsequently had higher chances of infection. This sex predominance for infection in burns has also been observed in various other populations.^8-11 The majority of the patients who acquired burns were 18 to 30 years old (33, 32.35%), which is mainly the employed class and similar to observations of other studies.^8-11 Epidemiological details found in the current study, such as female predominance and accidental thermal burns with 30% to 60% total body surface area involvement, were also observed.¹² Sepsis is common cause of death in burn victims who survive for more than a week.¹³

The GNB were more in number than the other pathogens. In developing countries, these GNB are the most prominent bacteria responsible for the mortality and morbidity in BWI. The flora causing BWI can differ among geographical areas, hospitals, and even ICUs within a hospital. The present authors isolated Acinetobacter species and Pseudomonas aeruginosa as the most common pathogen-causing infection among GNB in the authors’ ICU. However, most of the India-based studies the authors evaluation reported showed P aeruginosa as the most common isolate,^14-16and gram-positive S aureus has also been demonstrated as a common pathogen.^9,11 Among 86 patients with infected burn wounds in Ghana, P aeruginosa accounted for 30.2%, followed by Escherichia coli (19.8%) and Klebsiella species (14%).¹⁷ Similarly, gram-negative pathogens are known to cause infection in different parts of the globe.^18-20 In this study, S aureus was the fourth most common pathogen with an isolation rate of approximately 10%. This difference in the flora is multifactorial and may be attributed to the type, degree, and percentage of total body surface area involved in burns, personal hygiene of the patients, and hospital environment.

Among the gram-positive pathogens, both of which are methicillin resistant, S aureus and coagulase-negative staphylococci were reported to 30.55% (11/ 36) and 53.85% (7/13), respectively. The isolation of MRSA in patients with burn wounds varies from 18.09% to 94%in India.^15,21 The resistance pattern of Acinetobacter species isolated are having very high resistance to piperacillin/tazoabactum (84.71%), meropenem (80%), and amikacin (87.06%), which is contrary to another study.⁹ Such a high level of antimicrobial resistance presents a major challenge to treating BWI. P aeruginosa and Klebsiella species are also emerging with around 50% of the isolates resistant to these high-level antimicrobials.

Emergence of the MDR strains in the hospital environment is attributed to over-use and misuse of the high-level antimicrobials, especially carbapenems in the ICU.²² When MDR pathogens cause infection in patients with burn wounds, it can lead to failure in antimicrobial therapy and increases the rate of mortality in these patients. Strict aseptic precautions with an effective antimicrobial stewardship program need to be implemented to prevent these infections in physiologically and immunologically compromised patients. The present study has helped identify the flora causing BWI in the authors’ hospital, and to start effective, appropriate, and empirically proven antimicrobial therapy to reduce the failure in treatment and reduce mortality.

This was a single-center, retrospective study conducted in a tertiary care center. The authors were unable to study the outcome of the treatment in relation with duration, degree, type of burns intervention, and antimicrobial treatment. There is a need for multicenter studies with larger study populations for the formulation of a comprehensive empirical antimicrobial treatment for patients with infections resulting from burn wounds.

The current study has shown that GNB such as Acinetobacter species and P aeruginosa are the most common pathogens causing BWI, which are significantly resistant to carbapenem. This is a worrisome issue, mainly for plastic surgeons and clinical microbiologists who are treating this patient population and warrants close monitoring to avoid overuse or misuse of carbapenems, aminoglycosides, and fluroquinolones, as well as effective implementation of infection control practices and antimicrobial stewardship program.

Authors: Sanjana Joy, MBBS¹; Reshmina C. D’souza, MS²; Shreevidya K, MD³; Rekha Boloor, MD³; Vidya Rao Surlu, MHA⁴; Sucharitha Suresh, PhD⁵; Ramakrishna Pai Jakribettu, MD³; and Manjeshwar Shrinath Baliga, PhD⁶

Affiliations:¹MBBS Intern, Father Muller Medical College, Mangalore, Karnataka, India; ²Department of Surgery, Father Muller Medical College; ³Department of Microbiology/Infection Control, Father Muller Medical College; ⁴Quality Manager/Infection Control, Father Muller Medical College; ⁵Department of Community Medicine, Father Muller Medical College; and ⁶Senior Scientist, Father Muller Research Centre, Mangalore, Karnataka, India

Correspondence: Dr. Manjeshwar Shrinath Baliga, In Charge Head of Research, Mangalore Institute of Oncology, Mangalore, Karnataka, India-575002; msbaliga@gmail.com
or
Dr Ramakrishna Pai Jakribettu, Department of Microbiology/Infection Control, Father Muller Medical College, Mangalore, Karnataka, India-575002; ramakrishna.paij@gmail.com

Disclosure: The authors disclose no financial or other conflicts of interest.