Profile and Antibiotic Susceptibility of Bacteria Isolates in Burn Patients Hospitalized in a Moroccan Hospital: A Cross-sectional Study

Objective. The aim of this cross-sectional study is to assess the epidemiological profile and antibiotic susceptibility of bacteria isolates in burn patients hospitalized from 2007 to 2015 in a Moroccan hospital. Materials and Methods. Of 627 samples, 451 wounds, 126 blood cultures, and 50 catheter samples were analyzed. A total of 610 microorganisms were isolated from a total of 108 patients. Results. The most prevalent isolates were Pseudomonas aeruginosa (18.5%), Acinetobacter baumannii (14.7%), and Staphylococcus aureus (14.5%); a high prevalence (31.1%) of methicillin-resistant S aureus also was found. Imipenem followed by piperacillin/tazobactam performed best against P aeruginosa. A baumannii isolates showed a very high resistance to imipenem and all beta-lactams antibiotics but were sensitive to colistin. Conclusions. The profile of this bacterial ecology shows there are resistant species requiring frequent use of second-line antibiotics and strengthening of aseptic procedures in burn treatments. 

After the initial shock period burn patients experience, infections are the most observed complication. Subsequently, bacterial colonization and invasive bacterial infection remain the major causes of morbidity and mortality.¹ This complication contributes to higher costs of health care in burn treatment, extends the duration of hospitalization and treatment, and increases mortality rates of up to 75%.² The epidemiological monitoring of bacterial species in burn patients and the study of antibiotic sensitivity profiles are essential to establishing a local antibiotic resistance policy and therapeutic strategy.³ Wound swab samples, blood cultures, and venous catheter samples remain the simplest methods for bacteriological monitoring in the burn care unit. This monitoring is helpful in choosing the best antibiotic protocol for improving burn treatments.

The aim of this study is to assess the ecology and sensitivity of bacteria isolated from hospitalized burn patients at the Army Teaching Hospital Mohammed V of Rabat (ATHMV; Rabat, Morocco).

Study design

This cross-sectional study was carried out from April 2007 to December 2015. Inclusion criteria consisted of systematic samples collected from all patients admitted to the Department of Plastic Surgery and Burns (DPSB) at ATHMV that showed local (wound site) or systemic infection symptoms. Exclusion criteria included samples from the same patients (20) with the same isolate in different sites and the same antimicrobial susceptibility in order to avoid bias and violation of independency.

Samples included swabs from burn wounds, a serial of consecutive 3 blood cultures and venous catheters. Wound swabs were seeded on blood agar and chocolate agar, and the catheters were plated on blood agar according to the Brun-Buisson technique.⁴ After 18 to 48 hours of incubation at 37°C, isolates were identified using standard microbiological techniques. The isolates’ susceptibility patterns to antibiotics were determined by the method of Bauer et al⁵ according to recommendations of the Antibiogram Committee of French Microbiology Society.⁶ Blood cultures were incubated in the BD Bactec 9240 automatic system (Becton, Dickinson and Company, Franklin Lakes, NJ). A similar method was used for positive blood cultures, which were first transferred to brain heart infusion agar for aerobic bottles and on Schaedler agar and blood agar for anaerobic bottles. Clinical information was extracted from the laboratory information system. 

Statistical analyses

Statistical analyses were performed using SPSS 18.0 software (SPSS Inc, Chicago, IL). Categorical variables were expressed as numbers and percentages and were compared using a chi-squared test. Statistical significance was considered at P > .05.

Ethics

The study was approved by the ethics committee of the Medical School of Rabat (Morocco). Patients or their legal representative provided written informed consent upon admission to the hospital.

A total of 627 samples were collected from 108 patients with 451 wound samples (72%), 126 blood cultures (20%), and 50 venous catheters (8%). Mean age was 41.19 ± 22.15 years with 17 pediatric patients and a sex ratio (M/F) of 1.13. The distribution of sampling sites is presented in Table 1. A total of 610 strains were isolated. 

The prevalence of gram-positive and gram-negative bacteria was 32.3% and 65.2%, respectively. Table 2 shows the isolate distributions by family and species. The proportion of nonfermenting gram-negative bacilli was 35.08%, Enterobacteriaceae was 31.14%, and staphylococcal isolates was 23.44%. Pseudomonas aeruginosa (18.5%), Acinetobacter baumannii (14.8%), and Staphylococcus aureus (14.6%) were the most common species of the isolates, followed by Klebsiella pneumonia (9.0%), Enterobacter cloacae (8.5%), and Proteus mirabilis (5.9%). This distribution did not show any significant change in trend during the study period with P = .92 (Figure).

Most of the isolates showed resistance to classical antimicrobial agents. P aeruginosa was almost equally resistant to ceftazidime (58.7%) and imipenem (60%). More than 90% of A baumannii isolates were resistant to ceftazidime, imipenem, and gentamicin. S aureus isolates showed resistance to fluoroquinolones (67.0%) and cefoxitin (31.1%), but S aureus remained completely sensitive to glycopeptide antibiotics (Table 3). 

Of the 610 strains isolated in this study, 32.3% were gram-positive bacteria and 65.2% were gram-negative bacteria, which are in line with other series, including the Chinese study of 2212 strains.⁷ Gram- negative bacteria were most prevalent in venous catheter culture (91.7%) and equally prevalent in blood (73.6%) and wound secretions (75.3%) culture, while Staphylococcus was prevalent in wounds samples (42%).

The most frequently isolated pathogens in wounds were P aeruginosa, S aureus, and A baumannii. The latter is most isolated in blood cultures and venous catheters followed by K pneumonia and P aeruginosa.

Irrespective of the specimen sites, P aeruginosa was the most frequent species isolated (18.5%). This finding was consistent with other European and US studies^8,9 but less than others from developing countries where the prevalence rates of P aeruginosa were up to 54%.^10-13

P aeruginosa is characterized by a high resistance to antibiotics and therefore is difficult to control. It is responsible for serious infections in burn patients and has a mortality rate of up to 40% to 50%.^2,13-15 In the present study (similar to other publications^16-19), the investigators found P aeruginosa and S aureus to be the most frequent isolates in burn injuries. The prevalence of P aeruginosa and A baumannii isolates increased in the present study of up to 38.4% in 2010, followed by a stabilization of this prevalence in the years following. Indeed, the emergence of A baumannii was reported by others in a 5-year time period^20,21; in the present study, the incidence increased from 0.83% in 2007 to 14.5% in 2015. This shows that colonization and infection in patients with burns start from the environmental flora where A baumannii is able to resist for several months,^22,23 although all aseptic measures are applied. Different susceptibility patterns of an A baumannii clone can coexist in the same hospital depending on the selective pressure associated with antibiotic exposure,²⁴ and some can be attributed to epidemic nidus.²⁵ 

In this study, more than 90% of isolated A baumannii were multiresistant except for colistin (100% susceptible), similar to De Vos et al.²⁶ Colonization and/or infection by A baumannii are additional risk factors of mortality for burn patients.²⁷ To counter this, De Vos et al²⁶ recommended early use of colistin or rifampicin; however, in Morocco (setting of this study), where tuberculosis is still an endemic, the investigators strongly recommend the use of colistin rather than rifampicin as an effective antibiotic against A baumannii.

The high frequency of Staphylococcus isolates reported by Guggenheim et al in his 20-year study²⁸ was not observed in the present study. According to Reig et al,²⁹ it seems the prevalence of such microorganisms in the areas of the body that sustained burns depends on the extent of these burned areas and the date of sampling. During the first days, the coagulase-negative Staphylococcus species prevailed. The frequency of S aureus and gram-negative bacilli, especially A baumannii and P aeruginosa, increases with duration of hospitalization.¹⁹

S aureus was the second most prevalent isolate (equal to A baumannii) with a rate of 14.9%, of which 31.1% were methicillin- resistant S aureus (MRSA). These findings are less than other studies that recorded prevalence as high as 18.6% to 30% of S aureus and up to 40% to 60% of MRSA.^10,12,30,31 The difference in prevalence can be explained to some extent by the Moroccan cultural daily ablution rituals the people partake in, which include cleansing the nose and hand washing 3 times per ablution 3 to 5 times daily. This ritual may contribute to less asymptomatic portage of Staphylococcus in patients and health care workers. Indeed, most recommend screening for S aureus for nasal and/or groin carriage during the hot and humid season for all patients to prevent occurrence of Staphylococcus infections, especially MRSA.^32,33 Resistance associated with MRSA to other groups of antibiotics (such as aminoglycosides, quinolones, and macrolides) makes treatment more difficult.³⁴ Thus, glycopeptide antibiotics are the treatment of choice.³⁵ In the present study, no S aureus isolate showed a resistance to glycopeptides.

Despite an increasing evolution of resistance to imipenem during the last few years in the ATHMV, this molecule remains most effective against Enterobacteria and P aeruginosa.³⁶ The high resistance to imipenem of A baumannii (94.4%) is a serious problem in Morocco and can be related to noncontrolled prescriptions of imipenem. The associated resistance to other antibiotics confirms the multiresistant character and supports the nosocomial origin of A baumannii. Other factors associated with multiresistant isolates include antibiotic-selected cross transmission to burn patients and highly populated health care facilities.^37,38 Netilmicin and amikacin molecules remain the most potentially active aminoglycosides and hence can be used as synergistic drugs.

During the 9 years of this study, the distribution of microorganisms isolated did not change, as the nonfermenting gram-negative bacilli, Enterobacteria, and Staphylococcus remained the most common isolates. These results are in line with other reports,^1,39 as there has not yet been a significant change in the statistics of bacterial burn wound infection in the last 20 years, except for the virulence of MRSA, extended spectrum beta-lactamases, and emergence of multidrug-resistant P aeruginosa and A baumanii.⁴⁰ 

Due to the length of the study period (9 years), data may have been lost. The investigators could not record all clinical and demographic information. Specimens were collected when patients showed clinical or biological markers of infection, so they could not confirm the nosocomial origin of the infections. Moreover, some antibiotics have not yet been tested for all species. 

In the Moroccan hospital setting in which this study took place, P aeruginosa, A baumannii, and S aureus are the most frequently isolated species from burn patients with a high prevalence of multidrug-resistant bacteria. In light of the sensitivity results, the investigators recommend colistin as the most effective antimicrobial drug for A baumannii and glycopeptides for S aureus, whereas P aeruginosa and Enterobacteria need a combination of imipenem and amikacin. Moreover, in order to reduce the prevalence and the drug resistance, especially for A baumannii, strong efforts should be taken to avoid introduction of these microorganisms in such critical care units by strict isolation of colonized or infected burn patients, disinfection and antisepsis procedures, restriction of cross-transmission, and controlled antibiotic prescriptions. 

Affiliations: Department of Bacteriology, Army Teaching Hospital Mohammed V, Rabat, Morocco; Faculty of Medicine and Pharmacy of Rabat, University Mohammed V, Rabat, Morocco; Group of Research and Study of Antibiotic Resistance and Bacterial Infections, University Mohammed V; Department of Hygiene and biostatistics, Army Teaching Hospital Mohammed V; and Faculty of Medicine, University Mohammed First, Oujda, Morocco

Correspondence: Mohammed Frikh, MD, Department of Bacteriology, Army Teaching Hospital Mohammed V of Rabat, Avenue des FAR, Hay Riad, 10100, Rabat, Morocco; frikmed@yahoo.fr

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